diff options
author | Trond Myklebust <Trond.Myklebust@netapp.com> | 2006-12-07 21:35:17 (GMT) |
---|---|---|
committer | Trond Myklebust <Trond.Myklebust@netapp.com> | 2006-12-07 21:35:17 (GMT) |
commit | 21b4e736922f546e0f1aa7b9d6c442f309a2444a (patch) | |
tree | e1be8645297f8ebe87445251743ebcc52081a20d /fs | |
parent | 34161db6b14d984fb9b06c735b7b42f8803f6851 (diff) | |
parent | 68380b581383c028830f79ec2670f4a193854aa6 (diff) | |
download | linux-fsl-qoriq-21b4e736922f546e0f1aa7b9d6c442f309a2444a.tar.xz |
Merge branch 'master' of /home/trondmy/kernel/linux-2.6/ into merge_linus
Diffstat (limited to 'fs')
219 files changed, 3451 insertions, 2232 deletions
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 5241c60..18f26cd 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -256,7 +256,7 @@ static int v9fs_create(struct v9fs_session_info *v9ses, u32 pfid, char *name, u32 perm, u8 mode, char *extension, u32 *fidp, struct v9fs_qid *qid, u32 *iounit) { - u32 fid; + int fid; int err; struct v9fs_fcall *fcall; @@ -310,7 +310,7 @@ static struct v9fs_fid* v9fs_clone_walk(struct v9fs_session_info *v9ses, u32 fid, struct dentry *dentry) { int err; - u32 nfid; + int nfid; struct v9fs_fid *ret; struct v9fs_fcall *fcall; diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 9ade139..5023351 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -36,7 +36,7 @@ void __adfs_error(struct super_block *sb, const char *function, const char *fmt, va_list args; va_start(args, fmt); - vsprintf(error_buf, fmt, args); + vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); printk(KERN_CRIT "ADFS-fs error (device %s)%s%s: %s\n", @@ -212,12 +212,12 @@ static int adfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t *adfs_inode_cachep; +static struct kmem_cache *adfs_inode_cachep; static struct inode *adfs_alloc_inode(struct super_block *sb) { struct adfs_inode_info *ei; - ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, SLAB_KERNEL); + ei = (struct adfs_inode_info *)kmem_cache_alloc(adfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -228,7 +228,7 @@ static void adfs_destroy_inode(struct inode *inode) kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c index ccd624e..f4de4b9 100644 --- a/fs/affs/amigaffs.c +++ b/fs/affs/amigaffs.c @@ -445,7 +445,7 @@ affs_error(struct super_block *sb, const char *function, const char *fmt, ...) va_list args; va_start(args,fmt); - vsprintf(ErrorBuffer,fmt,args); + vsnprintf(ErrorBuffer,sizeof(ErrorBuffer),fmt,args); va_end(args); printk(KERN_CRIT "AFFS error (device %s): %s(): %s\n", sb->s_id, @@ -461,7 +461,7 @@ affs_warning(struct super_block *sb, const char *function, const char *fmt, ...) va_list args; va_start(args,fmt); - vsprintf(ErrorBuffer,fmt,args); + vsnprintf(ErrorBuffer,sizeof(ErrorBuffer),fmt,args); va_end(args); printk(KERN_WARNING "AFFS warning (device %s): %s(): %s\n", sb->s_id, diff --git a/fs/affs/bitmap.c b/fs/affs/bitmap.c index b0b9536..b330009 100644 --- a/fs/affs/bitmap.c +++ b/fs/affs/bitmap.c @@ -289,12 +289,11 @@ int affs_init_bitmap(struct super_block *sb, int *flags) sbi->s_bmap_count = (sbi->s_partition_size - sbi->s_reserved + sbi->s_bmap_bits - 1) / sbi->s_bmap_bits; size = sbi->s_bmap_count * sizeof(*bm); - bm = sbi->s_bitmap = kmalloc(size, GFP_KERNEL); + bm = sbi->s_bitmap = kzalloc(size, GFP_KERNEL); if (!sbi->s_bitmap) { printk(KERN_ERR "AFFS: Bitmap allocation failed\n"); return -ENOMEM; } - memset(sbi->s_bitmap, 0, size); bmap_blk = (__be32 *)sbi->s_root_bh->b_data; blk = sb->s_blocksize / 4 - 49; diff --git a/fs/affs/super.c b/fs/affs/super.c index 5ea72c3..3de93e7 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -66,12 +66,12 @@ affs_write_super(struct super_block *sb) pr_debug("AFFS: write_super() at %lu, clean=%d\n", get_seconds(), clean); } -static kmem_cache_t * affs_inode_cachep; +static struct kmem_cache * affs_inode_cachep; static struct inode *affs_alloc_inode(struct super_block *sb) { struct affs_inode_info *ei; - ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, SLAB_KERNEL); + ei = (struct affs_inode_info *)kmem_cache_alloc(affs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; @@ -83,7 +83,7 @@ static void affs_destroy_inode(struct inode *inode) kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; diff --git a/fs/afs/kafsasyncd.c b/fs/afs/kafsasyncd.c index f09a794..615df24 100644 --- a/fs/afs/kafsasyncd.c +++ b/fs/afs/kafsasyncd.c @@ -20,6 +20,7 @@ #include <linux/init.h> #include <linux/sched.h> #include <linux/completion.h> +#include <linux/freezer.h> #include "cell.h" #include "server.h" #include "volume.h" diff --git a/fs/afs/kafstimod.c b/fs/afs/kafstimod.c index 65bc05a..694344e 100644 --- a/fs/afs/kafstimod.c +++ b/fs/afs/kafstimod.c @@ -13,6 +13,7 @@ #include <linux/init.h> #include <linux/sched.h> #include <linux/completion.h> +#include <linux/freezer.h> #include "cell.h" #include "volume.h" #include "kafstimod.h" diff --git a/fs/afs/server.c b/fs/afs/server.c index 22afaae..44aff81 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -55,13 +55,12 @@ int afs_server_lookup(struct afs_cell *cell, const struct in_addr *addr, _enter("%p,%08x,", cell, ntohl(addr->s_addr)); /* allocate and initialise a server record */ - server = kmalloc(sizeof(struct afs_server), GFP_KERNEL); + server = kzalloc(sizeof(struct afs_server), GFP_KERNEL); if (!server) { _leave(" = -ENOMEM"); return -ENOMEM; } - memset(server, 0, sizeof(struct afs_server)); atomic_set(&server->usage, 1); INIT_LIST_HEAD(&server->link); diff --git a/fs/afs/super.c b/fs/afs/super.c index 67d1f5c..18d9b77 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -35,7 +35,7 @@ struct afs_mount_params { struct afs_volume *volume; }; -static void afs_i_init_once(void *foo, kmem_cache_t *cachep, +static void afs_i_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags); static int afs_get_sb(struct file_system_type *fs_type, @@ -65,7 +65,7 @@ static struct super_operations afs_super_ops = { .put_super = afs_put_super, }; -static kmem_cache_t *afs_inode_cachep; +static struct kmem_cache *afs_inode_cachep; static atomic_t afs_count_active_inodes; /*****************************************************************************/ @@ -242,14 +242,12 @@ static int afs_fill_super(struct super_block *sb, void *data, int silent) kenter(""); /* allocate a superblock info record */ - as = kmalloc(sizeof(struct afs_super_info), GFP_KERNEL); + as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); if (!as) { _leave(" = -ENOMEM"); return -ENOMEM; } - memset(as, 0, sizeof(struct afs_super_info)); - afs_get_volume(params->volume); as->volume = params->volume; @@ -384,7 +382,7 @@ static void afs_put_super(struct super_block *sb) /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(void *_vnode, kmem_cache_t *cachep, +static void afs_i_init_once(void *_vnode, struct kmem_cache *cachep, unsigned long flags) { struct afs_vnode *vnode = (struct afs_vnode *) _vnode; @@ -412,7 +410,7 @@ static struct inode *afs_alloc_inode(struct super_block *sb) struct afs_vnode *vnode; vnode = (struct afs_vnode *) - kmem_cache_alloc(afs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(afs_inode_cachep, GFP_KERNEL); if (!vnode) return NULL; @@ -47,8 +47,8 @@ unsigned long aio_nr; /* current system wide number of aio requests */ unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ /*----end sysctl variables---*/ -static kmem_cache_t *kiocb_cachep; -static kmem_cache_t *kioctx_cachep; +static struct kmem_cache *kiocb_cachep; +static struct kmem_cache *kioctx_cachep; static struct workqueue_struct *aio_wq; @@ -666,17 +666,6 @@ static ssize_t aio_run_iocb(struct kiocb *iocb) ssize_t (*retry)(struct kiocb *); ssize_t ret; - if (iocb->ki_retried++ > 1024*1024) { - printk("Maximal retry count. Bytes done %Zd\n", - iocb->ki_nbytes - iocb->ki_left); - return -EAGAIN; - } - - if (!(iocb->ki_retried & 0xff)) { - pr_debug("%ld retry: %zd of %zd\n", iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); - } - if (!(retry = iocb->ki_retry)) { printk("aio_run_iocb: iocb->ki_retry = NULL\n"); return 0; @@ -1005,9 +994,6 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) kunmap_atomic(ring, KM_IRQ1); pr_debug("added to ring %p at [%lu]\n", iocb, tail); - - pr_debug("%ld retries: %zd of %zd\n", iocb->ki_retried, - iocb->ki_nbytes - iocb->ki_left, iocb->ki_nbytes); put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); @@ -1413,7 +1399,6 @@ static ssize_t aio_setup_single_vector(struct kiocb *kiocb) kiocb->ki_iovec->iov_len = kiocb->ki_left; kiocb->ki_nr_segs = 1; kiocb->ki_cur_seg = 0; - kiocb->ki_nbytes = kiocb->ki_left; return 0; } @@ -1591,7 +1576,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, req->ki_opcode = iocb->aio_lio_opcode; init_waitqueue_func_entry(&req->ki_wait, aio_wake_function); INIT_LIST_HEAD(&req->ki_wait.task_list); - req->ki_retried = 0; ret = aio_setup_iocb(req); diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index 38ede5c..f968d13 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -28,10 +28,11 @@ void autofs_kill_sb(struct super_block *sb) /* * In the event of a failure in get_sb_nodev the superblock * info is not present so nothing else has been setup, so - * just exit when we are called from deactivate_super. + * just call kill_anon_super when we are called from + * deactivate_super. */ if (!sbi) - return; + goto out_kill_sb; if ( !sbi->catatonic ) autofs_catatonic_mode(sbi); /* Free wait queues, close pipe */ @@ -44,6 +45,7 @@ void autofs_kill_sb(struct super_block *sb) kfree(sb->s_fs_info); +out_kill_sb: DPRINTK(("autofs: shutting down\n")); kill_anon_super(sb); } @@ -209,7 +211,6 @@ fail_iput: fail_free: kfree(sbi); s->s_fs_info = NULL; - kill_anon_super(s); fail_unlock: return -EINVAL; } diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index ce7c0f1..9c48250 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -152,10 +152,11 @@ void autofs4_kill_sb(struct super_block *sb) /* * In the event of a failure in get_sb_nodev the superblock * info is not present so nothing else has been setup, so - * just exit when we are called from deactivate_super. + * just call kill_anon_super when we are called from + * deactivate_super. */ if (!sbi) - return; + goto out_kill_sb; sb->s_fs_info = NULL; @@ -167,6 +168,7 @@ void autofs4_kill_sb(struct super_block *sb) kfree(sbi); +out_kill_sb: DPRINTK("shutting down"); kill_anon_super(sb); } @@ -426,7 +428,6 @@ fail_ino: fail_free: kfree(sbi); s->s_fs_info = NULL; - kill_anon_super(s); fail_unlock: return -EINVAL; } diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 07f7144..bce402e 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -61,7 +61,7 @@ static const struct super_operations befs_sops = { }; /* slab cache for befs_inode_info objects */ -static kmem_cache_t *befs_inode_cachep; +static struct kmem_cache *befs_inode_cachep; static const struct file_operations befs_dir_operations = { .read = generic_read_dir, @@ -277,7 +277,7 @@ befs_alloc_inode(struct super_block *sb) { struct befs_inode_info *bi; bi = (struct befs_inode_info *)kmem_cache_alloc(befs_inode_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; @@ -289,7 +289,7 @@ befs_destroy_inode(struct inode *inode) kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index ed27ffb..eac175e 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -228,12 +228,12 @@ static void bfs_write_super(struct super_block *s) unlock_kernel(); } -static kmem_cache_t * bfs_inode_cachep; +static struct kmem_cache * bfs_inode_cachep; static struct inode *bfs_alloc_inode(struct super_block *sb) { struct bfs_inode_info *bi; - bi = kmem_cache_alloc(bfs_inode_cachep, SLAB_KERNEL); + bi = kmem_cache_alloc(bfs_inode_cachep, GFP_KERNEL); if (!bi) return NULL; return &bi->vfs_inode; @@ -244,7 +244,7 @@ static void bfs_destroy_inode(struct inode *inode) kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bfs_inode_info *bi = foo; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index cc72bb4..be5869d 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -47,10 +47,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs); static int load_elf_library(struct file *); static unsigned long elf_map (struct file *, unsigned long, struct elf_phdr *, int, int); -#ifndef elf_addr_t -#define elf_addr_t unsigned long -#endif - /* * If we don't support core dumping, then supply a NULL so we * don't even try. @@ -243,8 +239,9 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, if (interp_aout) { argv = sp + 2; envp = argv + argc + 1; - __put_user((elf_addr_t)(unsigned long)argv, sp++); - __put_user((elf_addr_t)(unsigned long)envp, sp++); + if (__put_user((elf_addr_t)(unsigned long)argv, sp++) || + __put_user((elf_addr_t)(unsigned long)envp, sp++)) + return -EFAULT; } else { argv = sp; envp = argv + argc + 1; @@ -254,7 +251,8 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, p = current->mm->arg_end = current->mm->arg_start; while (argc-- > 0) { size_t len; - __put_user((elf_addr_t)p, argv++); + if (__put_user((elf_addr_t)p, argv++)) + return -EFAULT; len = strnlen_user((void __user *)p, PAGE_SIZE*MAX_ARG_PAGES); if (!len || len > PAGE_SIZE*MAX_ARG_PAGES) return 0; @@ -265,7 +263,8 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec, current->mm->arg_end = current->mm->env_start = p; while (envc-- > 0) { size_t len; - __put_user((elf_addr_t)p, envp++); + if (__put_user((elf_addr_t)p, envp++)) + return -EFAULT; len = strnlen_user((void __user *)p, PAGE_SIZE*MAX_ARG_PAGES); if (!len || len > PAGE_SIZE*MAX_ARG_PAGES) return 0; @@ -545,7 +544,7 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) unsigned long reloc_func_desc = 0; char passed_fileno[6]; struct files_struct *files; - int have_pt_gnu_stack, executable_stack = EXSTACK_DEFAULT; + int executable_stack = EXSTACK_DEFAULT; unsigned long def_flags = 0; struct { struct elfhdr elf_ex; @@ -708,7 +707,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) executable_stack = EXSTACK_DISABLE_X; break; } - have_pt_gnu_stack = (i < loc->elf_ex.e_phnum); /* Some simple consistency checks for the interpreter */ if (elf_interpreter) { @@ -856,7 +854,13 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs) * default mmap base, as well as whatever program they * might try to exec. This is because the brk will * follow the loader, and is not movable. */ - load_bias = ELF_PAGESTART(ELF_ET_DYN_BASE - vaddr); + if (current->flags & PF_RANDOMIZE) + load_bias = randomize_range(0x10000, + ELF_ET_DYN_BASE, + 0); + else + load_bias = ELF_ET_DYN_BASE; + load_bias = ELF_PAGESTART(load_bias - vaddr); } error = elf_map(bprm->file, load_bias + vaddr, elf_ppnt, diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index f86d5c9..ed9a61c 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -40,9 +40,6 @@ #include <asm/pgalloc.h> typedef char *elf_caddr_t; -#ifndef elf_addr_t -#define elf_addr_t unsigned long -#endif #if 0 #define kdebug(fmt, ...) printk("FDPIC "fmt"\n" ,##__VA_ARGS__ ) @@ -30,7 +30,7 @@ #define BIO_POOL_SIZE 256 -static kmem_cache_t *bio_slab __read_mostly; +static struct kmem_cache *bio_slab __read_mostly; #define BIOVEC_NR_POOLS 6 @@ -44,7 +44,7 @@ mempool_t *bio_split_pool __read_mostly; struct biovec_slab { int nr_vecs; char *name; - kmem_cache_t *slab; + struct kmem_cache *slab; }; /* diff --git a/fs/block_dev.c b/fs/block_dev.c index 36c0e7a..13816b4d 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -235,11 +235,11 @@ static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(bdev_lock); -static kmem_cache_t * bdev_cachep __read_mostly; +static struct kmem_cache * bdev_cachep __read_mostly; static struct inode *bdev_alloc_inode(struct super_block *sb) { - struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, SLAB_KERNEL); + struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -253,7 +253,7 @@ static void bdev_destroy_inode(struct inode *inode) kmem_cache_free(bdev_cachep, bdi); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct bdev_inode *ei = (struct bdev_inode *) foo; struct block_device *bdev = &ei->bdev; diff --git a/fs/buffer.c b/fs/buffer.c index 35527dc..517860f 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2908,7 +2908,7 @@ asmlinkage long sys_bdflush(int func, long data) /* * Buffer-head allocation */ -static kmem_cache_t *bh_cachep; +static struct kmem_cache *bh_cachep; /* * Once the number of bh's in the machine exceeds this level, we start @@ -2961,7 +2961,7 @@ void free_buffer_head(struct buffer_head *bh) EXPORT_SYMBOL(free_buffer_head); static void -init_buffer_head(void *data, kmem_cache_t *cachep, unsigned long flags) +init_buffer_head(void *data, struct kmem_cache *cachep, unsigned long flags) { if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { @@ -2972,7 +2972,6 @@ init_buffer_head(void *data, kmem_cache_t *cachep, unsigned long flags) } } -#ifdef CONFIG_HOTPLUG_CPU static void buffer_exit_cpu(int cpu) { int i; @@ -2994,7 +2993,6 @@ static int buffer_cpu_notify(struct notifier_block *self, buffer_exit_cpu((unsigned long)hcpu); return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ void __init buffer_init(void) { diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 84976cd..71bc87a 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -34,6 +34,7 @@ #include <linux/mempool.h> #include <linux/delay.h> #include <linux/kthread.h> +#include <linux/freezer.h> #include "cifsfs.h" #include "cifspdu.h" #define DECLARE_GLOBALS_HERE @@ -81,7 +82,7 @@ extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static int cifs_read_super(struct super_block *sb, void *data, @@ -232,11 +233,11 @@ static int cifs_permission(struct inode * inode, int mask, struct nameidata *nd) return generic_permission(inode, mask, NULL); } -static kmem_cache_t *cifs_inode_cachep; -static kmem_cache_t *cifs_req_cachep; -static kmem_cache_t *cifs_mid_cachep; -kmem_cache_t *cifs_oplock_cachep; -static kmem_cache_t *cifs_sm_req_cachep; +static struct kmem_cache *cifs_inode_cachep; +static struct kmem_cache *cifs_req_cachep; +static struct kmem_cache *cifs_mid_cachep; +struct kmem_cache *cifs_oplock_cachep; +static struct kmem_cache *cifs_sm_req_cachep; mempool_t *cifs_sm_req_poolp; mempool_t *cifs_req_poolp; mempool_t *cifs_mid_poolp; @@ -245,7 +246,7 @@ static struct inode * cifs_alloc_inode(struct super_block *sb) { struct cifsInodeInfo *cifs_inode; - cifs_inode = kmem_cache_alloc(cifs_inode_cachep, SLAB_KERNEL); + cifs_inode = kmem_cache_alloc(cifs_inode_cachep, GFP_KERNEL); if (!cifs_inode) return NULL; cifs_inode->cifsAttrs = 0x20; /* default */ @@ -668,7 +669,7 @@ const struct file_operations cifs_dir_ops = { }; static void -cifs_init_once(void *inode, kmem_cache_t * cachep, unsigned long flags) +cifs_init_once(void *inode, struct kmem_cache * cachep, unsigned long flags) { struct cifsInodeInfo *cifsi = inode; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 71f7791..2caca06 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -31,6 +31,7 @@ #include <linux/delay.h> #include <linux/completion.h> #include <linux/pagevec.h> +#include <linux/freezer.h> #include <asm/uaccess.h> #include <asm/processor.h> #include "cifspdu.h" diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index bbc9cd3..aedf683 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -153,7 +153,7 @@ cifs_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_req_poolp, SLAB_KERNEL | SLAB_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_req_poolp, GFP_KERNEL | GFP_NOFS); /* clear the first few header bytes */ /* for most paths, more is cleared in header_assemble */ @@ -192,7 +192,7 @@ cifs_small_buf_get(void) albeit slightly larger than necessary and maxbuffersize defaults to this and can not be bigger */ ret_buf = - (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, SLAB_KERNEL | SLAB_NOFS); + (struct smb_hdr *) mempool_alloc(cifs_sm_req_poolp, GFP_KERNEL | GFP_NOFS); if (ret_buf) { /* No need to clear memory here, cleared in header assemble */ /* memset(ret_buf, 0, sizeof(struct smb_hdr) + 27);*/ diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 48d47b4..f80007e 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -34,7 +34,7 @@ #include "cifs_debug.h" extern mempool_t *cifs_mid_poolp; -extern kmem_cache_t *cifs_oplock_cachep; +extern struct kmem_cache *cifs_oplock_cachep; static struct mid_q_entry * AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) @@ -51,7 +51,7 @@ AllocMidQEntry(const struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) } temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, - SLAB_KERNEL | SLAB_NOFS); + GFP_KERNEL | GFP_NOFS); if (temp == NULL) return temp; else { @@ -118,7 +118,7 @@ AllocOplockQEntry(struct inode * pinode, __u16 fid, struct cifsTconInfo * tcon) return NULL; } temp = (struct oplock_q_entry *) kmem_cache_alloc(cifs_oplock_cachep, - SLAB_KERNEL); + GFP_KERNEL); if (temp == NULL) return temp; else { diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 88d1233..b64659f 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -38,12 +38,12 @@ static void coda_clear_inode(struct inode *); static void coda_put_super(struct super_block *); static int coda_statfs(struct dentry *dentry, struct kstatfs *buf); -static kmem_cache_t * coda_inode_cachep; +static struct kmem_cache * coda_inode_cachep; static struct inode *coda_alloc_inode(struct super_block *sb) { struct coda_inode_info *ei; - ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, SLAB_KERNEL); + ei = (struct coda_inode_info *)kmem_cache_alloc(coda_inode_cachep, GFP_KERNEL); if (!ei) return NULL; memset(&ei->c_fid, 0, sizeof(struct CodaFid)); @@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode) kmem_cache_free(coda_inode_cachep, ITOC(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/compat.c b/fs/compat.c index 06dad66..a7e3f16 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -871,7 +871,7 @@ asmlinkage long compat_sys_mount(char __user * dev_name, char __user * dir_name, retval = -EINVAL; - if (type_page) { + if (type_page && data_page) { if (!strcmp((char *)type_page, SMBFS_NAME)) { do_smb_super_data_conv((void *)data_page); } else if (!strcmp((char *)type_page, NCPFS_NAME)) { @@ -1144,7 +1144,9 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, lastdirent = buf.previous; if (lastdirent) { typeof(lastdirent->d_off) d_off = file->f_pos; - __put_user_unaligned(d_off, &lastdirent->d_off); + error = -EFAULT; + if (__put_user_unaligned(d_off, &lastdirent->d_off)) + goto out_putf; error = count - buf.count; } @@ -1611,14 +1613,14 @@ int compat_get_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, nr &= ~1UL; while (nr) { unsigned long h, l; - __get_user(l, ufdset); - __get_user(h, ufdset+1); + if (__get_user(l, ufdset) || __get_user(h, ufdset+1)) + return -EFAULT; ufdset += 2; *fdset++ = h << 32 | l; nr -= 2; } - if (odd) - __get_user(*fdset, ufdset); + if (odd && __get_user(*fdset, ufdset)) + return -EFAULT; } else { /* Tricky, must clear full unsigned long in the * kernel fdset at the end, this makes sure that @@ -1630,14 +1632,14 @@ int compat_get_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, } static -void compat_set_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, - unsigned long *fdset) +int compat_set_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, + unsigned long *fdset) { unsigned long odd; nr = ROUND_UP(nr, __COMPAT_NFDBITS); if (!ufdset) - return; + return 0; odd = nr & 1UL; nr &= ~1UL; @@ -1645,13 +1647,14 @@ void compat_set_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, unsigned long h, l; l = *fdset++; h = l >> 32; - __put_user(l, ufdset); - __put_user(h, ufdset+1); + if (__put_user(l, ufdset) || __put_user(h, ufdset+1)) + return -EFAULT; ufdset += 2; nr -= 2; } - if (odd) - __put_user(*fdset, ufdset); + if (odd && __put_user(*fdset, ufdset)) + return -EFAULT; + return 0; } @@ -1726,10 +1729,10 @@ int compat_core_sys_select(int n, compat_ulong_t __user *inp, ret = 0; } - compat_set_fd_set(n, inp, fds.res_in); - compat_set_fd_set(n, outp, fds.res_out); - compat_set_fd_set(n, exp, fds.res_ex); - + if (compat_set_fd_set(n, inp, fds.res_in) || + compat_set_fd_set(n, outp, fds.res_out) || + compat_set_fd_set(n, exp, fds.res_ex)) + ret = -EFAULT; out: kfree(bits); out_nofds: diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index a91f262..bcc3caf 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -211,8 +211,10 @@ static int do_video_stillpicture(unsigned int fd, unsigned int cmd, unsigned lon up_native = compat_alloc_user_space(sizeof(struct video_still_picture)); - put_user(compat_ptr(fp), &up_native->iFrame); - put_user(size, &up_native->size); + err = put_user(compat_ptr(fp), &up_native->iFrame); + err |= put_user(size, &up_native->size); + if (err) + return -EFAULT; err = sys_ioctl(fd, cmd, (unsigned long) up_native); @@ -236,8 +238,10 @@ static int do_video_set_spu_palette(unsigned int fd, unsigned int cmd, unsigned err |= get_user(length, &up->length); up_native = compat_alloc_user_space(sizeof(struct video_spu_palette)); - put_user(compat_ptr(palp), &up_native->palette); - put_user(length, &up_native->length); + err = put_user(compat_ptr(palp), &up_native->palette); + err |= put_user(length, &up_native->length); + if (err) + return -EFAULT; err = sys_ioctl(fd, cmd, (unsigned long) up_native); @@ -2043,16 +2047,19 @@ static int serial_struct_ioctl(unsigned fd, unsigned cmd, unsigned long arg) struct serial_struct ss; mm_segment_t oldseg = get_fs(); __u32 udata; + unsigned int base; if (cmd == TIOCSSERIAL) { if (!access_ok(VERIFY_READ, ss32, sizeof(SS32))) return -EFAULT; if (__copy_from_user(&ss, ss32, offsetof(SS32, iomem_base))) return -EFAULT; - __get_user(udata, &ss32->iomem_base); + if (__get_user(udata, &ss32->iomem_base)) + return -EFAULT; ss.iomem_base = compat_ptr(udata); - __get_user(ss.iomem_reg_shift, &ss32->iomem_reg_shift); - __get_user(ss.port_high, &ss32->port_high); + if (__get_user(ss.iomem_reg_shift, &ss32->iomem_reg_shift) || + __get_user(ss.port_high, &ss32->port_high)) + return -EFAULT; ss.iomap_base = 0UL; } set_fs(KERNEL_DS); @@ -2063,12 +2070,12 @@ static int serial_struct_ioctl(unsigned fd, unsigned cmd, unsigned long arg) return -EFAULT; if (__copy_to_user(ss32,&ss,offsetof(SS32,iomem_base))) return -EFAULT; - __put_user((unsigned long)ss.iomem_base >> 32 ? - 0xffffffff : (unsigned)(unsigned long)ss.iomem_base, - &ss32->iomem_base); - __put_user(ss.iomem_reg_shift, &ss32->iomem_reg_shift); - __put_user(ss.port_high, &ss32->port_high); - + base = (unsigned long)ss.iomem_base >> 32 ? + 0xffffffff : (unsigned)(unsigned long)ss.iomem_base; + if (__put_user(base, &ss32->iomem_base) || + __put_user(ss.iomem_reg_shift, &ss32->iomem_reg_shift) || + __put_user(ss.port_high, &ss32->port_high)) + return -EFAULT; } return err; } @@ -2397,6 +2404,7 @@ HANDLE_IOCTL(SIOCGIFMAP, dev_ifsioc) HANDLE_IOCTL(SIOCSIFMAP, dev_ifsioc) HANDLE_IOCTL(SIOCGIFADDR, dev_ifsioc) HANDLE_IOCTL(SIOCSIFADDR, dev_ifsioc) +HANDLE_IOCTL(SIOCSIFHWBROADCAST, dev_ifsioc) /* ioctls used by appletalk ddp.c */ HANDLE_IOCTL(SIOCATALKDIFADDR, dev_ifsioc) diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index 3f4ff7a..f92cd30 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -49,7 +49,7 @@ struct configfs_dirent { #define CONFIGFS_NOT_PINNED (CONFIGFS_ITEM_ATTR) extern struct vfsmount * configfs_mount; -extern kmem_cache_t *configfs_dir_cachep; +extern struct kmem_cache *configfs_dir_cachep; extern int configfs_is_root(struct config_item *item); diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c index 68bd5c9..ed67852 100644 --- a/fs/configfs/mount.c +++ b/fs/configfs/mount.c @@ -38,7 +38,7 @@ struct vfsmount * configfs_mount = NULL; struct super_block * configfs_sb = NULL; -kmem_cache_t *configfs_dir_cachep; +struct kmem_cache *configfs_dir_cachep; static int configfs_mnt_count = 0; static struct super_operations configfs_ops = { diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index a624c3e..0509ced 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -481,6 +481,8 @@ static int cramfs_readpage(struct file *file, struct page * page) pgdata = kmap(page); if (compr_len == 0) ; /* hole */ + else if (compr_len > (PAGE_CACHE_SIZE << 1)) + printk(KERN_ERR "cramfs: bad compressed blocksize %u\n", compr_len); else { mutex_lock(&read_mutex); bytes_filled = cramfs_uncompress_block(pgdata, diff --git a/fs/dcache.c b/fs/dcache.c index fd4a428..d68631f 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -43,7 +43,7 @@ static __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); EXPORT_SYMBOL(dcache_lock); -static kmem_cache_t *dentry_cache __read_mostly; +static struct kmem_cache *dentry_cache __read_mostly; #define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname)) @@ -68,15 +68,19 @@ struct dentry_stat_t dentry_stat = { .age_limit = 45, }; -static void d_callback(struct rcu_head *head) +static void __d_free(struct dentry *dentry) { - struct dentry * dentry = container_of(head, struct dentry, d_u.d_rcu); - if (dname_external(dentry)) kfree(dentry->d_name.name); kmem_cache_free(dentry_cache, dentry); } +static void d_callback(struct rcu_head *head) +{ + struct dentry * dentry = container_of(head, struct dentry, d_u.d_rcu); + __d_free(dentry); +} + /* * no dcache_lock, please. The caller must decrement dentry_stat.nr_dentry * inside dcache_lock. @@ -85,7 +89,11 @@ static void d_free(struct dentry *dentry) { if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); - call_rcu(&dentry->d_u.d_rcu, d_callback); + /* if dentry was never inserted into hash, immediate free is OK */ + if (dentry->d_hash.pprev == NULL) + __d_free(dentry); + else + call_rcu(&dentry->d_u.d_rcu, d_callback); } /* @@ -2072,10 +2080,10 @@ static void __init dcache_init(unsigned long mempages) } /* SLAB cache for __getname() consumers */ -kmem_cache_t *names_cachep __read_mostly; +struct kmem_cache *names_cachep __read_mostly; /* SLAB cache for file structures */ -kmem_cache_t *filp_cachep __read_mostly; +struct kmem_cache *filp_cachep __read_mostly; EXPORT_SYMBOL(d_genocide); diff --git a/fs/dcookies.c b/fs/dcookies.c index 0c4b067..21af162 100644 --- a/fs/dcookies.c +++ b/fs/dcookies.c @@ -37,7 +37,7 @@ struct dcookie_struct { static LIST_HEAD(dcookie_users); static DEFINE_MUTEX(dcookie_mutex); -static kmem_cache_t *dcookie_cache __read_mostly; +static struct kmem_cache *dcookie_cache __read_mostly; static struct list_head *dcookie_hashtable __read_mostly; static size_t hash_size __read_mostly; diff --git a/fs/dlm/Kconfig b/fs/dlm/Kconfig index 81b2c64..b5654a2 100644 --- a/fs/dlm/Kconfig +++ b/fs/dlm/Kconfig @@ -1,14 +1,32 @@ menu "Distributed Lock Manager" - depends on INET && IP_SCTP && EXPERIMENTAL + depends on EXPERIMENTAL && INET config DLM tristate "Distributed Lock Manager (DLM)" depends on IPV6 || IPV6=n select CONFIGFS_FS + select IP_SCTP if DLM_SCTP help A general purpose distributed lock manager for kernel or userspace applications. +choice + prompt "Select DLM communications protocol" + depends on DLM + default DLM_TCP + help + The DLM Can use TCP or SCTP for it's network communications. + SCTP supports multi-homed operations whereas TCP doesn't. + However, SCTP seems to have stability problems at the moment. + +config DLM_TCP + bool "TCP/IP" + +config DLM_SCTP + bool "SCTP" + +endchoice + config DLM_DEBUG bool "DLM debugging" depends on DLM diff --git a/fs/dlm/Makefile b/fs/dlm/Makefile index 1832e02..6538894 100644 --- a/fs/dlm/Makefile +++ b/fs/dlm/Makefile @@ -4,7 +4,6 @@ dlm-y := ast.o \ dir.o \ lock.o \ lockspace.o \ - lowcomms.o \ main.o \ member.o \ memory.o \ @@ -17,3 +16,6 @@ dlm-y := ast.o \ util.o dlm-$(CONFIG_DLM_DEBUG) += debug_fs.o +dlm-$(CONFIG_DLM_TCP) += lowcomms-tcp.o + +dlm-$(CONFIG_DLM_SCTP) += lowcomms-sctp.o
\ No newline at end of file diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 1e5cd67..1ee8195 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -471,6 +471,7 @@ struct dlm_ls { char *ls_recover_buf; int ls_recover_nodeid; /* for debugging */ uint64_t ls_rcom_seq; + spinlock_t ls_rcom_spin; struct list_head ls_recover_list; spinlock_t ls_recover_list_lock; int ls_recover_list_count; @@ -488,7 +489,8 @@ struct dlm_ls { #define LSFL_RUNNING 1 #define LSFL_RECOVERY_STOP 2 #define LSFL_RCOM_READY 3 -#define LSFL_UEVENT_WAIT 4 +#define LSFL_RCOM_WAIT 4 +#define LSFL_UEVENT_WAIT 5 /* much of this is just saving user space pointers associated with the lock that we pass back to the user lib with an ast */ diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 3f2befa..30878de 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -2372,6 +2372,7 @@ static int send_lookup_reply(struct dlm_ls *ls, struct dlm_message *ms_in, static void receive_flags(struct dlm_lkb *lkb, struct dlm_message *ms) { lkb->lkb_exflags = ms->m_exflags; + lkb->lkb_sbflags = ms->m_sbflags; lkb->lkb_flags = (lkb->lkb_flags & 0xFFFF0000) | (ms->m_flags & 0x0000FFFF); } @@ -3028,10 +3029,17 @@ int dlm_receive_message(struct dlm_header *hd, int nodeid, int recovery) while (1) { if (dlm_locking_stopped(ls)) { - if (!recovery) - dlm_add_requestqueue(ls, nodeid, hd); - error = -EINTR; - goto out; + if (recovery) { + error = -EINTR; + goto out; + } + error = dlm_add_requestqueue(ls, nodeid, hd); + if (error == -EAGAIN) + continue; + else { + error = -EINTR; + goto out; + } } if (lock_recovery_try(ls)) diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index f8842ca..59012b0 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -22,6 +22,7 @@ #include "memory.h" #include "lock.h" #include "recover.h" +#include "requestqueue.h" #ifdef CONFIG_DLM_DEBUG int dlm_create_debug_file(struct dlm_ls *ls); @@ -478,6 +479,8 @@ static int new_lockspace(char *name, int namelen, void **lockspace, ls->ls_recoverd_task = NULL; mutex_init(&ls->ls_recoverd_active); spin_lock_init(&ls->ls_recover_lock); + spin_lock_init(&ls->ls_rcom_spin); + get_random_bytes(&ls->ls_rcom_seq, sizeof(uint64_t)); ls->ls_recover_status = 0; ls->ls_recover_seq = 0; ls->ls_recover_args = NULL; @@ -684,6 +687,7 @@ static int release_lockspace(struct dlm_ls *ls, int force) * Free structures on any other lists */ + dlm_purge_requestqueue(ls); kfree(ls->ls_recover_args); dlm_clear_free_entries(ls); dlm_clear_members(ls); diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms-sctp.c index 6da6b14..fe158d7 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms-sctp.c @@ -2,7 +2,7 @@ ******************************************************************************* ** ** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. -** Copyright (C) 2004-2005 Red Hat, Inc. All rights reserved. +** Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. ** ** This copyrighted material is made available to anyone wishing to use, ** modify, copy, or redistribute it subject to the terms and conditions @@ -75,13 +75,13 @@ struct nodeinfo { }; static DEFINE_IDR(nodeinfo_idr); -static struct rw_semaphore nodeinfo_lock; -static int max_nodeid; +static DECLARE_RWSEM(nodeinfo_lock); +static int max_nodeid; struct cbuf { - unsigned base; - unsigned len; - unsigned mask; + unsigned int base; + unsigned int len; + unsigned int mask; }; /* Just the one of these, now. But this struct keeps @@ -90,9 +90,9 @@ struct cbuf { #define CF_READ_PENDING 1 struct connection { - struct socket *sock; + struct socket *sock; unsigned long flags; - struct page *rx_page; + struct page *rx_page; atomic_t waiting_requests; struct cbuf cb; int eagain_flag; @@ -102,36 +102,40 @@ struct connection { struct writequeue_entry { struct list_head list; - struct page *page; + struct page *page; int offset; int len; int end; int users; - struct nodeinfo *ni; + struct nodeinfo *ni; }; -#define CBUF_ADD(cb, n) do { (cb)->len += n; } while(0) -#define CBUF_EMPTY(cb) ((cb)->len == 0) -#define CBUF_MAY_ADD(cb, n) (((cb)->len + (n)) < ((cb)->mask + 1)) -#define CBUF_DATA(cb) (((cb)->base + (cb)->len) & (cb)->mask) +static void cbuf_add(struct cbuf *cb, int n) +{ + cb->len += n; +} -#define CBUF_INIT(cb, size) \ -do { \ - (cb)->base = (cb)->len = 0; \ - (cb)->mask = ((size)-1); \ -} while(0) +static int cbuf_data(struct cbuf *cb) +{ + return ((cb->base + cb->len) & cb->mask); +} -#define CBUF_EAT(cb, n) \ -do { \ - (cb)->len -= (n); \ - (cb)->base += (n); \ - (cb)->base &= (cb)->mask; \ -} while(0) +static void cbuf_init(struct cbuf *cb, int size) +{ + cb->base = cb->len = 0; + cb->mask = size-1; +} +static void cbuf_eat(struct cbuf *cb, int n) +{ + cb->len -= n; + cb->base += n; + cb->base &= cb->mask; +} /* List of nodes which have writes pending */ -static struct list_head write_nodes; -static spinlock_t write_nodes_lock; +static LIST_HEAD(write_nodes); +static DEFINE_SPINLOCK(write_nodes_lock); /* Maximum number of incoming messages to process before * doing a schedule() @@ -141,8 +145,7 @@ static spinlock_t write_nodes_lock; /* Manage daemons */ static struct task_struct *recv_task; static struct task_struct *send_task; -static wait_queue_head_t lowcomms_recv_wait; -static atomic_t accepting; +static DECLARE_WAIT_QUEUE_HEAD(lowcomms_recv_wait); /* The SCTP connection */ static struct connection sctp_con; @@ -161,11 +164,11 @@ static int nodeid_to_addr(int nodeid, struct sockaddr *retaddr) return error; if (dlm_local_addr[0]->ss_family == AF_INET) { - struct sockaddr_in *in4 = (struct sockaddr_in *) &addr; + struct sockaddr_in *in4 = (struct sockaddr_in *) &addr; struct sockaddr_in *ret4 = (struct sockaddr_in *) retaddr; ret4->sin_addr.s_addr = in4->sin_addr.s_addr; } else { - struct sockaddr_in6 *in6 = (struct sockaddr_in6 *) &addr; + struct sockaddr_in6 *in6 = (struct sockaddr_in6 *) &addr; struct sockaddr_in6 *ret6 = (struct sockaddr_in6 *) retaddr; memcpy(&ret6->sin6_addr, &in6->sin6_addr, sizeof(in6->sin6_addr)); @@ -174,6 +177,8 @@ static int nodeid_to_addr(int nodeid, struct sockaddr *retaddr) return 0; } +/* If alloc is 0 here we will not attempt to allocate a new + nodeinfo struct */ static struct nodeinfo *nodeid2nodeinfo(int nodeid, gfp_t alloc) { struct nodeinfo *ni; @@ -184,44 +189,45 @@ static struct nodeinfo *nodeid2nodeinfo(int nodeid, gfp_t alloc) ni = idr_find(&nodeinfo_idr, nodeid); up_read(&nodeinfo_lock); - if (!ni && alloc) { - down_write(&nodeinfo_lock); + if (ni || !alloc) + return ni; - ni = idr_find(&nodeinfo_idr, nodeid); - if (ni) - goto out_up; + down_write(&nodeinfo_lock); - r = idr_pre_get(&nodeinfo_idr, alloc); - if (!r) - goto out_up; + ni = idr_find(&nodeinfo_idr, nodeid); + if (ni) + goto out_up; - ni = kmalloc(sizeof(struct nodeinfo), alloc); - if (!ni) - goto out_up; + r = idr_pre_get(&nodeinfo_idr, alloc); + if (!r) + goto out_up; - r = idr_get_new_above(&nodeinfo_idr, ni, nodeid, &n); - if (r) { - kfree(ni); - ni = NULL; - goto out_up; - } - if (n != nodeid) { - idr_remove(&nodeinfo_idr, n); - kfree(ni); - ni = NULL; - goto out_up; - } - memset(ni, 0, sizeof(struct nodeinfo)); - spin_lock_init(&ni->lock); - INIT_LIST_HEAD(&ni->writequeue); - spin_lock_init(&ni->writequeue_lock); - ni->nodeid = nodeid; - - if (nodeid > max_nodeid) - max_nodeid = nodeid; - out_up: - up_write(&nodeinfo_lock); + ni = kmalloc(sizeof(struct nodeinfo), alloc); + if (!ni) + goto out_up; + + r = idr_get_new_above(&nodeinfo_idr, ni, nodeid, &n); + if (r) { + kfree(ni); + ni = NULL; + goto out_up; } + if (n != nodeid) { + idr_remove(&nodeinfo_idr, n); + kfree(ni); + ni = NULL; + goto out_up; + } + memset(ni, 0, sizeof(struct nodeinfo)); + spin_lock_init(&ni->lock); + INIT_LIST_HEAD(&ni->writequeue); + spin_lock_init(&ni->writequeue_lock); + ni->nodeid = nodeid; + + if (nodeid > max_nodeid) + max_nodeid = nodeid; +out_up: + up_write(&nodeinfo_lock); return ni; } @@ -279,13 +285,13 @@ static void make_sockaddr(struct sockaddr_storage *saddr, uint16_t port, in4_addr->sin_port = cpu_to_be16(port); memset(&in4_addr->sin_zero, 0, sizeof(in4_addr->sin_zero)); memset(in4_addr+1, 0, sizeof(struct sockaddr_storage) - - sizeof(struct sockaddr_in)); + sizeof(struct sockaddr_in)); *addr_len = sizeof(struct sockaddr_in); } else { struct sockaddr_in6 *in6_addr = (struct sockaddr_in6 *)saddr; in6_addr->sin6_port = cpu_to_be16(port); memset(in6_addr+1, 0, sizeof(struct sockaddr_storage) - - sizeof(struct sockaddr_in6)); + sizeof(struct sockaddr_in6)); *addr_len = sizeof(struct sockaddr_in6); } } @@ -324,7 +330,7 @@ static void send_shutdown(sctp_assoc_t associd) cmsg->cmsg_type = SCTP_SNDRCV; cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo)); outmessage.msg_controllen = cmsg->cmsg_len; - sinfo = (struct sctp_sndrcvinfo *)CMSG_DATA(cmsg); + sinfo = CMSG_DATA(cmsg); memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo)); sinfo->sinfo_flags |= MSG_EOF; @@ -387,7 +393,7 @@ static void process_sctp_notification(struct msghdr *msg, char *buf) if ((int)sn->sn_assoc_change.sac_assoc_id <= 0) { log_print("COMM_UP for invalid assoc ID %d", - (int)sn->sn_assoc_change.sac_assoc_id); + (int)sn->sn_assoc_change.sac_assoc_id); init_failed(); return; } @@ -398,15 +404,18 @@ static void process_sctp_notification(struct msghdr *msg, char *buf) fs = get_fs(); set_fs(get_ds()); ret = sctp_con.sock->ops->getsockopt(sctp_con.sock, - IPPROTO_SCTP, SCTP_PRIMARY_ADDR, - (char*)&prim, &prim_len); + IPPROTO_SCTP, + SCTP_PRIMARY_ADDR, + (char*)&prim, + &prim_len); set_fs(fs); if (ret < 0) { struct nodeinfo *ni; log_print("getsockopt/sctp_primary_addr on " "new assoc %d failed : %d", - (int)sn->sn_assoc_change.sac_assoc_id, ret); + (int)sn->sn_assoc_change.sac_assoc_id, + ret); /* Retry INIT later */ ni = assoc2nodeinfo(sn->sn_assoc_change.sac_assoc_id); @@ -426,12 +435,10 @@ static void process_sctp_notification(struct msghdr *msg, char *buf) return; /* Save the assoc ID */ - spin_lock(&ni->lock); ni->assoc_id = sn->sn_assoc_change.sac_assoc_id; - spin_unlock(&ni->lock); log_print("got new/restarted association %d nodeid %d", - (int)sn->sn_assoc_change.sac_assoc_id, nodeid); + (int)sn->sn_assoc_change.sac_assoc_id, nodeid); /* Send any pending writes */ clear_bit(NI_INIT_PENDING, &ni->flags); @@ -507,13 +514,12 @@ static int receive_from_sock(void) sctp_con.rx_page = alloc_page(GFP_ATOMIC); if (sctp_con.rx_page == NULL) goto out_resched; - CBUF_INIT(&sctp_con.cb, PAGE_CACHE_SIZE); + cbuf_init(&sctp_con.cb, PAGE_CACHE_SIZE); } memset(&incmsg, 0, sizeof(incmsg)); memset(&msgname, 0, sizeof(msgname)); - memset(incmsg, 0, sizeof(incmsg)); msg.msg_name = &msgname; msg.msg_namelen = sizeof(msgname); msg.msg_flags = 0; @@ -532,17 +538,17 @@ static int receive_from_sock(void) * iov[0] is the bit of the circular buffer between the current end * point (cb.base + cb.len) and the end of the buffer. */ - iov[0].iov_len = sctp_con.cb.base - CBUF_DATA(&sctp_con.cb); + iov[0].iov_len = sctp_con.cb.base - cbuf_data(&sctp_con.cb); iov[0].iov_base = page_address(sctp_con.rx_page) + - CBUF_DATA(&sctp_con.cb); + cbuf_data(&sctp_con.cb); iov[1].iov_len = 0; /* * iov[1] is the bit of the circular buffer between the start of the * buffer and the start of the currently used section (cb.base) */ - if (CBUF_DATA(&sctp_con.cb) >= sctp_con.cb.base) { - iov[0].iov_len = PAGE_CACHE_SIZE - CBUF_DATA(&sctp_con.cb); + if (cbuf_data(&sctp_con.cb) >= sctp_con.cb.base) { + iov[0].iov_len = PAGE_CACHE_SIZE - cbuf_data(&sctp_con.cb); iov[1].iov_len = sctp_con.cb.base; iov[1].iov_base = page_address(sctp_con.rx_page); msg.msg_iovlen = 2; @@ -557,7 +563,7 @@ static int receive_from_sock(void) msg.msg_control = incmsg; msg.msg_controllen = sizeof(incmsg); cmsg = CMSG_FIRSTHDR(&msg); - sinfo = (struct sctp_sndrcvinfo *)CMSG_DATA(cmsg); + sinfo = CMSG_DATA(cmsg); if (msg.msg_flags & MSG_NOTIFICATION) { process_sctp_notification(&msg, page_address(sctp_con.rx_page)); @@ -583,29 +589,29 @@ static int receive_from_sock(void) if (r == 1) return 0; - CBUF_ADD(&sctp_con.cb, ret); + cbuf_add(&sctp_con.cb, ret); ret = dlm_process_incoming_buffer(cpu_to_le32(sinfo->sinfo_ppid), page_address(sctp_con.rx_page), sctp_con.cb.base, sctp_con.cb.len, PAGE_CACHE_SIZE); if (ret < 0) goto out_close; - CBUF_EAT(&sctp_con.cb, ret); + cbuf_eat(&sctp_con.cb, ret); - out: +out: ret = 0; goto out_ret; - out_resched: +out_resched: lowcomms_data_ready(sctp_con.sock->sk, 0); ret = 0; - schedule(); + cond_resched(); goto out_ret; - out_close: +out_close: if (ret != -EAGAIN) log_print("error reading from sctp socket: %d", ret); - out_ret: +out_ret: return ret; } @@ -619,10 +625,12 @@ static int add_bind_addr(struct sockaddr_storage *addr, int addr_len, int num) set_fs(get_ds()); if (num == 1) result = sctp_con.sock->ops->bind(sctp_con.sock, - (struct sockaddr *) addr, addr_len); + (struct sockaddr *) addr, + addr_len); else result = sctp_con.sock->ops->setsockopt(sctp_con.sock, SOL_SCTP, - SCTP_SOCKOPT_BINDX_ADD, (char *)addr, addr_len); + SCTP_SOCKOPT_BINDX_ADD, + (char *)addr, addr_len); set_fs(fs); if (result < 0) @@ -719,10 +727,10 @@ static int init_sock(void) return 0; - create_delsock: +create_delsock: sock_release(sock); sctp_con.sock = NULL; - out: +out: return result; } @@ -756,16 +764,13 @@ void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc) int users = 0; struct nodeinfo *ni; - if (!atomic_read(&accepting)) - return NULL; - ni = nodeid2nodeinfo(nodeid, allocation); if (!ni) return NULL; spin_lock(&ni->writequeue_lock); e = list_entry(ni->writequeue.prev, struct writequeue_entry, list); - if (((struct list_head *) e == &ni->writequeue) || + if ((&e->list == &ni->writequeue) || (PAGE_CACHE_SIZE - e->end < len)) { e = NULL; } else { @@ -776,7 +781,7 @@ void *dlm_lowcomms_get_buffer(int nodeid, int len, gfp_t allocation, char **ppc) spin_unlock(&ni->writequeue_lock); if (e) { - got_one: + got_one: if (users == 0) kmap(e->page); *ppc = page_address(e->page) + offset; @@ -803,9 +808,6 @@ void dlm_lowcomms_commit_buffer(void *arg) int users; struct nodeinfo *ni = e->ni; - if (!atomic_read(&accepting)) - return; - spin_lock(&ni->writequeue_lock); users = --e->users; if (users) @@ -822,7 +824,7 @@ void dlm_lowcomms_commit_buffer(void *arg) } return; - out: +out: spin_unlock(&ni->writequeue_lock); return; } @@ -878,7 +880,7 @@ static void initiate_association(int nodeid) cmsg->cmsg_level = IPPROTO_SCTP; cmsg->cmsg_type = SCTP_SNDRCV; cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo)); - sinfo = (struct sctp_sndrcvinfo *)CMSG_DATA(cmsg); + sinfo = CMSG_DATA(cmsg); memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo)); sinfo->sinfo_ppid = cpu_to_le32(dlm_local_nodeid); @@ -892,7 +894,7 @@ static void initiate_association(int nodeid) } /* Send a message */ -static int send_to_sock(struct nodeinfo *ni) +static void send_to_sock(struct nodeinfo *ni) { int ret = 0; struct writequeue_entry *e; @@ -903,13 +905,13 @@ static int send_to_sock(struct nodeinfo *ni) struct sctp_sndrcvinfo *sinfo; struct kvec iov; - /* See if we need to init an association before we start + /* See if we need to init an association before we start sending precious messages */ spin_lock(&ni->lock); if (!ni->assoc_id && !test_and_set_bit(NI_INIT_PENDING, &ni->flags)) { spin_unlock(&ni->lock); initiate_association(ni->nodeid); - return 0; + return; } spin_unlock(&ni->lock); @@ -923,7 +925,7 @@ static int send_to_sock(struct nodeinfo *ni) cmsg->cmsg_level = IPPROTO_SCTP; cmsg->cmsg_type = SCTP_SNDRCV; cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo)); - sinfo = (struct sctp_sndrcvinfo *)CMSG_DATA(cmsg); + sinfo = CMSG_DATA(cmsg); memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo)); sinfo->sinfo_ppid = cpu_to_le32(dlm_local_nodeid); sinfo->sinfo_assoc_id = ni->assoc_id; @@ -955,7 +957,7 @@ static int send_to_sock(struct nodeinfo *ni) goto send_error; } else { /* Don't starve people filling buffers */ - schedule(); + cond_resched(); } spin_lock(&ni->writequeue_lock); @@ -964,15 +966,16 @@ static int send_to_sock(struct nodeinfo *ni) if (e->len == 0 && e->users == 0) { list_del(&e->list); + kunmap(e->page); free_entry(e); continue; } } spin_unlock(&ni->writequeue_lock); - out: - return ret; +out: + return; - send_error: +send_error: log_print("Error sending to node %d %d", ni->nodeid, ret); spin_lock(&ni->lock); if (!test_and_set_bit(NI_INIT_PENDING, &ni->flags)) { @@ -982,7 +985,7 @@ static int send_to_sock(struct nodeinfo *ni) } else spin_unlock(&ni->lock); - return ret; + return; } /* Try to send any messages that are pending */ @@ -994,7 +997,7 @@ static void process_output_queue(void) spin_lock_bh(&write_nodes_lock); list_for_each_safe(list, temp, &write_nodes) { struct nodeinfo *ni = - list_entry(list, struct nodeinfo, write_list); + list_entry(list, struct nodeinfo, write_list); clear_bit(NI_WRITE_PENDING, &ni->flags); list_del(&ni->write_list); @@ -1106,7 +1109,7 @@ static int dlm_recvd(void *data) set_current_state(TASK_INTERRUPTIBLE); add_wait_queue(&lowcomms_recv_wait, &wait); if (!test_bit(CF_READ_PENDING, &sctp_con.flags)) - schedule(); + cond_resched(); remove_wait_queue(&lowcomms_recv_wait, &wait); set_current_state(TASK_RUNNING); @@ -1118,12 +1121,12 @@ static int dlm_recvd(void *data) /* Don't starve out everyone else */ if (++count >= MAX_RX_MSG_COUNT) { - schedule(); + cond_resched(); count = 0; } } while (!kthread_should_stop() && ret >=0); } - schedule(); + cond_resched(); } return 0; @@ -1138,7 +1141,7 @@ static int dlm_sendd(void *data) while (!kthread_should_stop()) { set_current_state(TASK_INTERRUPTIBLE); if (write_list_empty()) - schedule(); + cond_resched(); set_current_state(TASK_RUNNING); if (sctp_con.eagain_flag) { @@ -1166,7 +1169,7 @@ static int daemons_start(void) p = kthread_run(dlm_recvd, NULL, "dlm_recvd"); error = IS_ERR(p); - if (error) { + if (error) { log_print("can't start dlm_recvd %d", error); return error; } @@ -1174,7 +1177,7 @@ static int daemons_start(void) p = kthread_run(dlm_sendd, NULL, "dlm_sendd"); error = IS_ERR(p); - if (error) { + if (error) { log_print("can't start dlm_sendd %d", error); kthread_stop(recv_task); return error; @@ -1197,43 +1200,28 @@ int dlm_lowcomms_start(void) error = daemons_start(); if (error) goto fail_sock; - atomic_set(&accepting, 1); return 0; - fail_sock: +fail_sock: close_connection(); return error; } -/* Set all the activity flags to prevent any socket activity. */ - void dlm_lowcomms_stop(void) { - atomic_set(&accepting, 0); + int i; + sctp_con.flags = 0x7; daemons_stop(); clean_writequeues(); close_connection(); dealloc_nodeinfo(); max_nodeid = 0; -} -int dlm_lowcomms_init(void) -{ - init_waitqueue_head(&lowcomms_recv_wait); - spin_lock_init(&write_nodes_lock); - INIT_LIST_HEAD(&write_nodes); - init_rwsem(&nodeinfo_lock); - return 0; -} - -void dlm_lowcomms_exit(void) -{ - int i; + dlm_local_count = 0; + dlm_local_nodeid = 0; for (i = 0; i < dlm_local_count; i++) kfree(dlm_local_addr[i]); - dlm_local_count = 0; - dlm_local_nodeid = 0; } diff --git a/fs/dlm/lowcomms-tcp.c b/fs/dlm/lowcomms-tcp.c new file mode 100644 index 0000000..8f2791f --- /dev/null +++ b/fs/dlm/lowcomms-tcp.c @@ -0,0 +1,1189 @@ +/****************************************************************************** +******************************************************************************* +** +** Copyright (C) Sistina Software, Inc. 1997-2003 All rights reserved. +** Copyright (C) 2004-2006 Red Hat, Inc. All rights reserved. +** +** This copyrighted material is made available to anyone wishing to use, +** modify, copy, or redistribute it subject to the terms and conditions +** of the GNU General Public License v.2. +** +******************************************************************************* +******************************************************************************/ + +/* + * lowcomms.c + * + * This is the "low-level" comms layer. + * + * It is responsible for sending/receiving messages + * from other nodes in the cluster. + * + * Cluster nodes are referred to by their nodeids. nodeids are + * simply 32 bit numbers to the locking module - if they need to + * be expanded for the cluster infrastructure then that is it's + * responsibility. It is this layer's + * responsibility to resolve these into IP address or + * whatever it needs for inter-node communication. + * + * The comms level is two kernel threads that deal mainly with + * the receiving of messages from other nodes and passing them + * up to the mid-level comms layer (which understands the + * message format) for execution by the locking core, and + * a send thread which does all the setting up of connections + * to remote nodes and the sending of data. Threads are not allowed + * to send their own data because it may cause them to wait in times + * of high load. Also, this way, the sending thread can collect together + * messages bound for one node and send them in one block. + * + * I don't see any problem with the recv thread executing the locking + * code on behalf of remote processes as the locking code is + * short, efficient and never waits. + * + */ + + +#include <asm/ioctls.h> +#include <net/sock.h> +#include <net/tcp.h> +#include <linux/pagemap.h> + +#include "dlm_internal.h" +#include "lowcomms.h" +#include "midcomms.h" +#include "config.h" + +struct cbuf { + unsigned int base; + unsigned int len; + unsigned int mask; +}; + +#define NODE_INCREMENT 32 +static void cbuf_add(struct cbuf *cb, int n) +{ + cb->len += n; +} + +static int cbuf_data(struct cbuf *cb) +{ + return ((cb->base + cb->len) & cb->mask); +} + +static void cbuf_init(struct cbuf *cb, int size) +{ + cb->base = cb->len = 0; + cb->mask = size-1; +} + +static void cbuf_eat(struct cbuf *cb, int n) +{ + cb->len -= n; + cb->base += n; + cb->base &= cb->mask; +} + +static bool cbuf_empty(struct cbuf *cb) +{ + return cb->len == 0; +} + +/* Maximum number of incoming messages to process before + doing a cond_resched() +*/ +#define MAX_RX_MSG_COUNT 25 + +struct connection { + struct socket *sock; /* NULL if not connected */ + uint32_t nodeid; /* So we know who we are in the list */ + struct rw_semaphore sock_sem; /* Stop connect races */ + struct list_head read_list; /* On this list when ready for reading */ + struct list_head write_list; /* On this list when ready for writing */ + struct list_head state_list; /* On this list when ready to connect */ + unsigned long flags; /* bit 1,2 = We are on the read/write lists */ +#define CF_READ_PENDING 1 +#define CF_WRITE_PENDING 2 +#define CF_CONNECT_PENDING 3 +#define CF_IS_OTHERCON 4 + struct list_head writequeue; /* List of outgoing writequeue_entries */ + struct list_head listenlist; /* List of allocated listening sockets */ + spinlock_t writequeue_lock; + int (*rx_action) (struct connection *); /* What to do when active */ + struct page *rx_page; + struct cbuf cb; + int retries; + atomic_t waiting_requests; +#define MAX_CONNECT_RETRIES 3 + struct connection *othercon; +}; +#define sock2con(x) ((struct connection *)(x)->sk_user_data) + +/* An entry waiting to be sent */ +struct writequeue_entry { + struct list_head list; + struct page *page; + int offset; + int len; + int end; + int users; + struct connection *con; +}; + +static struct sockaddr_storage dlm_local_addr; + +/* Manage daemons */ +static struct task_struct *recv_task; +static struct task_struct *send_task; + +static wait_queue_t lowcomms_send_waitq_head; +static DECLARE_WAIT_QUEUE_HEAD(lowcomms_send_waitq); +static wait_queue_t lowcomms_recv_waitq_head; +static DECLARE_WAIT_QUEUE_HEAD(lowcomms_recv_waitq); + +/* An array of pointers to connections, indexed by NODEID */ +static struct connection **connections; +static DECLARE_MUTEX(connections_lock); +static kmem_cache_t *con_cache; +static int conn_array_size; + +/* List of sockets that have reads pending */ +static LIST_HEAD(read_sockets); +static DEFINE_SPINLOCK(read_sockets_lock); + +/* List of sockets which have writes pending */ +static LIST_HEAD(write_sockets); +static DEFINE_SPINLOCK(write_sockets_lock); + +/* List of sockets which have connects pending */ +static LIST_HEAD(state_sockets); +static DEFINE_SPINLOCK(state_sockets_lock); + +static struct connection *nodeid2con(int nodeid, gfp_t allocation) +{ + struct connection *con = NULL; + + down(&connections_lock); + if (nodeid >= conn_array_size) { + int new_size = nodeid + NODE_INCREMENT; + struct connection **new_conns; + + new_conns = kzalloc(sizeof(struct connection *) * + new_size, allocation); + if (!new_conns) + goto finish; + + memcpy(new_conns, connections, sizeof(struct connection *) * conn_array_size); + conn_array_size = new_size; + kfree(connections); + connections = new_conns; + + } + + con = connections[nodeid]; + if (con == NULL && allocation) { + con = kmem_cache_zalloc(con_cache, allocation); + if (!con) + goto finish; + + con->nodeid = nodeid; + init_rwsem(&con->sock_sem); + INIT_LIST_HEAD(&con->writequeue); + spin_lock_init(&con->writequeue_lock); + + connections[nodeid] = con; + } + +finish: + up(&connections_lock); + return con; +} + +/* Data available on socket or listen socket received a connect */ +static void lowcomms_data_ready(struct sock *sk, int count_unused) +{ + struct connection *con = sock2con(sk); + + atomic_inc(&con->waiting_requests); + if (test_and_set_bit(CF_READ_PENDING, &con->flags)) + return; + + spin_lock_bh(&read_sockets_lock); + list_add_tail(&con->read_list, &read_sockets); + spin_unlock_bh(&read_sockets_lock); + + wake_up_interruptible(&lowcomms_recv_waitq); +} + +static void lowcomms_write_space(struct sock *sk) +{ + struct connection *con = sock2con(sk); + + if (test_and_set_bit(CF_WRITE_PENDING, &con->flags)) + return; + + spin_lock_bh(&write_sockets_lock); + list_add_tail(&con->write_list, &write_sockets); + spin_unlock_bh(&write_sockets_lock); + + wake_up_interruptible(&lowcomms_send_waitq); +} + +static inline void lowcomms_connect_sock(struct connection *con) +{ + if (test_and_set_bit(CF_CONNECT_PENDING, &con->flags)) + return; + + spin_lock_bh(&state_sockets_lock); + list_add_tail(&con->state_list, &state_sockets); + spin_unlock_bh(&state_sockets_lock); + + wake_up_interruptible(&lowcomms_send_waitq); +} + +static void lowcomms_state_change(struct sock *sk) +{ + if (sk->sk_state == TCP_ESTABLISHED) + lowcomms_write_space(sk); +} + +/* Make a socket active */ +static int add_sock(struct socket *sock, struct connection *con) +{ + con->sock = sock; + + /* Install a data_ready callback */ + con->sock->sk->sk_data_ready = lowcomms_data_ready; + con->sock->sk->sk_write_space = lowcomms_write_space; + con->sock->sk->sk_state_change = lowcomms_state_change; + + return 0; +} + +/* Add the port number to an IP6 or 4 sockaddr and return the address + length */ +static void make_sockaddr(struct sockaddr_storage *saddr, uint16_t port, + int *addr_len) +{ + saddr->ss_family = dlm_local_addr.ss_family; + if (saddr->ss_family == AF_INET) { + struct sockaddr_in *in4_addr = (struct sockaddr_in *)saddr; + in4_addr->sin_port = cpu_to_be16(port); + *addr_len = sizeof(struct sockaddr_in); + } else { + struct sockaddr_in6 *in6_addr = (struct sockaddr_in6 *)saddr; + in6_addr->sin6_port = cpu_to_be16(port); + *addr_len = sizeof(struct sockaddr_in6); + } +} + +/* Close a remote connection and tidy up */ +static void close_connection(struct connection *con, bool and_other) +{ + down_write(&con->sock_sem); + + if (con->sock) { + sock_release(con->sock); + con->sock = NULL; + } + if (con->othercon && and_other) { + /* Will only re-enter once. */ + close_connection(con->othercon, false); + } + if (con->rx_page) { + __free_page(con->rx_page); + con->rx_page = NULL; + } + con->retries = 0; + up_write(&con->sock_sem); +} + +/* Data received from remote end */ +static int receive_from_sock(struct connection *con) +{ + int ret = 0; + struct msghdr msg; + struct iovec iov[2]; + mm_segment_t fs; + unsigned len; + int r; + int call_again_soon = 0; + + down_read(&con->sock_sem); + + if (con->sock == NULL) + goto out; + if (con->rx_page == NULL) { + /* + * This doesn't need to be atomic, but I think it should + * improve performance if it is. + */ + con->rx_page = alloc_page(GFP_ATOMIC); + if (con->rx_page == NULL) + goto out_resched; + cbuf_init(&con->cb, PAGE_CACHE_SIZE); + } + + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_iovlen = 1; + msg.msg_iov = iov; + msg.msg_name = NULL; + msg.msg_namelen = 0; + msg.msg_flags = 0; + + /* + * iov[0] is the bit of the circular buffer between the current end + * point (cb.base + cb.len) and the end of the buffer. + */ + iov[0].iov_len = con->cb.base - cbuf_data(&con->cb); + iov[0].iov_base = page_address(con->rx_page) + cbuf_data(&con->cb); + iov[1].iov_len = 0; + + /* + * iov[1] is the bit of the circular buffer between the start of the + * buffer and the start of the currently used section (cb.base) + */ + if (cbuf_data(&con->cb) >= con->cb.base) { + iov[0].iov_len = PAGE_CACHE_SIZE - cbuf_data(&con->cb); + iov[1].iov_len = con->cb.base; + iov[1].iov_base = page_address(con->rx_page); + msg.msg_iovlen = 2; + } + len = iov[0].iov_len + iov[1].iov_len; + + fs = get_fs(); + set_fs(get_ds()); + r = ret = sock_recvmsg(con->sock, &msg, len, + MSG_DONTWAIT | MSG_NOSIGNAL); + set_fs(fs); + + if (ret <= 0) + goto out_close; + if (ret == len) + call_again_soon = 1; + cbuf_add(&con->cb, ret); + ret = dlm_process_incoming_buffer(con->nodeid, + page_address(con->rx_page), + con->cb.base, con->cb.len, + PAGE_CACHE_SIZE); + if (ret == -EBADMSG) { + printk(KERN_INFO "dlm: lowcomms: addr=%p, base=%u, len=%u, " + "iov_len=%u, iov_base[0]=%p, read=%d\n", + page_address(con->rx_page), con->cb.base, con->cb.len, + len, iov[0].iov_base, r); + } + if (ret < 0) + goto out_close; + cbuf_eat(&con->cb, ret); + + if (cbuf_empty(&con->cb) && !call_again_soon) { + __free_page(con->rx_page); + con->rx_page = NULL; + } + +out: + if (call_again_soon) + goto out_resched; + up_read(&con->sock_sem); + return 0; + +out_resched: + lowcomms_data_ready(con->sock->sk, 0); + up_read(&con->sock_sem); + cond_resched(); + return 0; + +out_close: + up_read(&con->sock_sem); + if (ret != -EAGAIN && !test_bit(CF_IS_OTHERCON, &con->flags)) { + close_connection(con, false); + /* Reconnect when there is something to send */ + } + + return ret; +} + +/* Listening socket is busy, accept a connection */ +static int accept_from_sock(struct connection *con) +{ + int result; + struct sockaddr_storage peeraddr; + struct socket *newsock; + int len; + int nodeid; + struct connection *newcon; + + memset(&peeraddr, 0, sizeof(peeraddr)); + result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, + IPPROTO_TCP, &newsock); + if (result < 0) + return -ENOMEM; + + down_read(&con->sock_sem); + + result = -ENOTCONN; + if (con->sock == NULL) + goto accept_err; + + newsock->type = con->sock->type; + newsock->ops = con->sock->ops; + + result = con->sock->ops->accept(con->sock, newsock, O_NONBLOCK); + if (result < 0) + goto accept_err; + + /* Get the connected socket's peer */ + memset(&peeraddr, 0, sizeof(peeraddr)); + if (newsock->ops->getname(newsock, (struct sockaddr *)&peeraddr, + &len, 2)) { + result = -ECONNABORTED; + goto accept_err; + } + + /* Get the new node's NODEID */ + make_sockaddr(&peeraddr, 0, &len); + if (dlm_addr_to_nodeid(&peeraddr, &nodeid)) { + printk("dlm: connect from non cluster node\n"); + sock_release(newsock); + up_read(&con->sock_sem); + return -1; + } + + log_print("got connection from %d", nodeid); + + /* Check to see if we already have a connection to this node. This + * could happen if the two nodes initiate a connection at roughly + * the same time and the connections cross on the wire. + * TEMPORARY FIX: + * In this case we store the incoming one in "othercon" + */ + newcon = nodeid2con(nodeid, GFP_KERNEL); + if (!newcon) { + result = -ENOMEM; + goto accept_err; + } + down_write(&newcon->sock_sem); + if (newcon->sock) { + struct connection *othercon = newcon->othercon; + + if (!othercon) { + othercon = kmem_cache_zalloc(con_cache, GFP_KERNEL); + if (!othercon) { + printk("dlm: failed to allocate incoming socket\n"); + up_write(&newcon->sock_sem); + result = -ENOMEM; + goto accept_err; + } + othercon->nodeid = nodeid; + othercon->rx_action = receive_from_sock; + init_rwsem(&othercon->sock_sem); + set_bit(CF_IS_OTHERCON, &othercon->flags); + newcon->othercon = othercon; + } + othercon->sock = newsock; + newsock->sk->sk_user_data = othercon; + add_sock(newsock, othercon); + } + else { + newsock->sk->sk_user_data = newcon; + newcon->rx_action = receive_from_sock; + add_sock(newsock, newcon); + + } + + up_write(&newcon->sock_sem); + + /* + * Add it to the active queue in case we got data + * beween processing the accept adding the socket + * to the read_sockets list + */ + lowcomms_data_ready(newsock->sk, 0); + up_read(&con->sock_sem); + + return 0; + +accept_err: + up_read(&con->sock_sem); + sock_release(newsock); + + if (result != -EAGAIN) + printk("dlm: error accepting connection from node: %d\n", result); + return result; +} + +/* Connect a new socket to its peer */ +static void connect_to_sock(struct connection *con) +{ + int result = -EHOSTUNREACH; + struct sockaddr_storage saddr; + int addr_len; + struct socket *sock; + + if (con->nodeid == 0) { + log_print("attempt to connect sock 0 foiled"); + return; + } + + down_write(&con->sock_sem); + if (con->retries++ > MAX_CONNECT_RETRIES) + goto out; + + /* Some odd races can cause double-connects, ignore them */ + if (con->sock) { + result = 0; + goto out; + } + + /* Create a socket to communicate with */ + result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, + IPPROTO_TCP, &sock); + if (result < 0) + goto out_err; + + memset(&saddr, 0, sizeof(saddr)); + if (dlm_nodeid_to_addr(con->nodeid, &saddr)) + goto out_err; + + sock->sk->sk_user_data = con; + con->rx_action = receive_from_sock; + + make_sockaddr(&saddr, dlm_config.tcp_port, &addr_len); + + add_sock(sock, con); + + log_print("connecting to %d", con->nodeid); + result = + sock->ops->connect(sock, (struct sockaddr *)&saddr, addr_len, + O_NONBLOCK); + if (result == -EINPROGRESS) + result = 0; + if (result == 0) + goto out; + +out_err: + if (con->sock) { + sock_release(con->sock); + con->sock = NULL; + } + /* + * Some errors are fatal and this list might need adjusting. For other + * errors we try again until the max number of retries is reached. + */ + if (result != -EHOSTUNREACH && result != -ENETUNREACH && + result != -ENETDOWN && result != EINVAL + && result != -EPROTONOSUPPORT) { + lowcomms_connect_sock(con); + result = 0; + } +out: + up_write(&con->sock_sem); + return; +} + +static struct socket *create_listen_sock(struct connection *con, + struct sockaddr_storage *saddr) +{ + struct socket *sock = NULL; + mm_segment_t fs; + int result = 0; + int one = 1; + int addr_len; + + if (dlm_local_addr.ss_family == AF_INET) + addr_len = sizeof(struct sockaddr_in); + else + addr_len = sizeof(struct sockaddr_in6); + + /* Create a socket to communicate with */ + result = sock_create_kern(dlm_local_addr.ss_family, SOCK_STREAM, IPPROTO_TCP, &sock); + if (result < 0) { + printk("dlm: Can't create listening comms socket\n"); + goto create_out; + } + + fs = get_fs(); + set_fs(get_ds()); + result = sock_setsockopt(sock, SOL_SOCKET, SO_REUSEADDR, + (char *)&one, sizeof(one)); + set_fs(fs); + if (result < 0) { + printk("dlm: Failed to set SO_REUSEADDR on socket: result=%d\n", + result); + } + sock->sk->sk_user_data = con; + con->rx_action = accept_from_sock; + con->sock = sock; + + /* Bind to our port */ + make_sockaddr(saddr, dlm_config.tcp_port, &addr_len); + result = sock->ops->bind(sock, (struct sockaddr *) saddr, addr_len); + if (result < 0) { + printk("dlm: Can't bind to port %d\n", dlm_config.tcp_port); + sock_release(sock); + sock = NULL; + con->sock = NULL; + goto create_out; + } + + fs = get_fs(); + set_fs(get_ds()); + + result = sock_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE, + (char *)&one, sizeof(one)); + set_fs(fs); + if (result < 0) { + printk("dlm: Set keepalive failed: %d\n", result); + } + + result = sock->ops->listen(sock, 5); + if (result < 0) { + printk("dlm: Can't listen on port %d\n", dlm_config.tcp_port); + sock_release(sock); + sock = NULL; + goto create_out; + } + +create_out: + return sock; +} + + +/* Listen on all interfaces */ +static int listen_for_all(void) +{ + struct socket *sock = NULL; + struct connection *con = nodeid2con(0, GFP_KERNEL); + int result = -EINVAL; + + /* We don't support multi-homed hosts */ + set_bit(CF_IS_OTHERCON, &con->flags); + + sock = create_listen_sock(con, &dlm_local_addr); + if (sock) { + add_sock(sock, con); + result = 0; + } + else { + result = -EADDRINUSE; + } + + return result; +} + + + +static struct writequeue_entry *new_writequeue_entry(struct connection *con, + gfp_t allocation) +{ + struct writequeue_entry *entry; + + entry = kmalloc(sizeof(struct writequeue_entry), allocation); + if (!entry) + return NULL; + + entry->page = alloc_page(allocation); + if (!entry->page) { + kfree(entry); + return NULL; + } + + entry->offset = 0; + entry->len = 0; + entry->end = 0; + entry->users = 0; + entry->con = con; + + return entry; +} + +void *dlm_lowcomms_get_buffer(int nodeid, int len, + gfp_t allocation, char **ppc) +{ + struct connection *con; + struct writequeue_entry *e; + int offset = 0; + int users = 0; + + con = nodeid2con(nodeid, allocation); + if (!con) + return NULL; + + e = list_entry(con->writequeue.prev, struct writequeue_entry, list); + if ((&e->list == &con->writequeue) || + (PAGE_CACHE_SIZE - e->end < len)) { + e = NULL; + } else { + offset = e->end; + e->end += len; + users = e->users++; + } + spin_unlock(&con->writequeue_lock); + + if (e) { + got_one: + if (users == 0) + kmap(e->page); + *ppc = page_address(e->page) + offset; + return e; + } + + e = new_writequeue_entry(con, allocation); + if (e) { + spin_lock(&con->writequeue_lock); + offset = e->end; + e->end += len; + users = e->users++; + list_add_tail(&e->list, &con->writequeue); + spin_unlock(&con->writequeue_lock); + goto got_one; + } + return NULL; +} + +void dlm_lowcomms_commit_buffer(void *mh) +{ + struct writequeue_entry *e = (struct writequeue_entry *)mh; + struct connection *con = e->con; + int users; + + users = --e->users; + if (users) + goto out; + e->len = e->end - e->offset; + kunmap(e->page); + spin_unlock(&con->writequeue_lock); + + if (test_and_set_bit(CF_WRITE_PENDING, &con->flags) == 0) { + spin_lock_bh(&write_sockets_lock); + list_add_tail(&con->write_list, &write_sockets); + spin_unlock_bh(&write_sockets_lock); + + wake_up_interruptible(&lowcomms_send_waitq); + } + return; + +out: + spin_unlock(&con->writequeue_lock); + return; +} + +static void free_entry(struct writequeue_entry *e) +{ + __free_page(e->page); + kfree(e); +} + +/* Send a message */ +static void send_to_sock(struct connection *con) +{ + int ret = 0; + ssize_t(*sendpage) (struct socket *, struct page *, int, size_t, int); + const int msg_flags = MSG_DONTWAIT | MSG_NOSIGNAL; + struct writequeue_entry *e; + int len, offset; + + down_read(&con->sock_sem); + if (con->sock == NULL) + goto out_connect; + + sendpage = con->sock->ops->sendpage; + + spin_lock(&con->writequeue_lock); + for (;;) { + e = list_entry(con->writequeue.next, struct writequeue_entry, + list); + if ((struct list_head *) e == &con->writequeue) + break; + + len = e->len; + offset = e->offset; + BUG_ON(len == 0 && e->users == 0); + spin_unlock(&con->writequeue_lock); + + ret = 0; + if (len) { + ret = sendpage(con->sock, e->page, offset, len, + msg_flags); + if (ret == -EAGAIN || ret == 0) + goto out; + if (ret <= 0) + goto send_error; + } + else { + /* Don't starve people filling buffers */ + cond_resched(); + } + + spin_lock(&con->writequeue_lock); + e->offset += ret; + e->len -= ret; + + if (e->len == 0 && e->users == 0) { + list_del(&e->list); + kunmap(e->page); + free_entry(e); + continue; + } + } + spin_unlock(&con->writequeue_lock); +out: + up_read(&con->sock_sem); + return; + +send_error: + up_read(&con->sock_sem); + close_connection(con, false); + lowcomms_connect_sock(con); + return; + +out_connect: + up_read(&con->sock_sem); + lowcomms_connect_sock(con); + return; +} + +static void clean_one_writequeue(struct connection *con) +{ + struct list_head *list; + struct list_head *temp; + + spin_lock(&con->writequeue_lock); + list_for_each_safe(list, temp, &con->writequeue) { + struct writequeue_entry *e = + list_entry(list, struct writequeue_entry, list); + list_del(&e->list); + free_entry(e); + } + spin_unlock(&con->writequeue_lock); +} + +/* Called from recovery when it knows that a node has + left the cluster */ +int dlm_lowcomms_close(int nodeid) +{ + struct connection *con; + + if (!connections) + goto out; + + log_print("closing connection to node %d", nodeid); + con = nodeid2con(nodeid, 0); + if (con) { + clean_one_writequeue(con); + close_connection(con, true); + atomic_set(&con->waiting_requests, 0); + } + return 0; + +out: + return -1; +} + +/* API send message call, may queue the request */ +/* N.B. This is the old interface - use the new one for new calls */ +int lowcomms_send_message(int nodeid, char *buf, int len, gfp_t allocation) +{ + struct writequeue_entry *e; + char *b; + + e = dlm_lowcomms_get_buffer(nodeid, len, allocation, &b); + if (e) { + memcpy(b, buf, len); + dlm_lowcomms_commit_buffer(e); + return 0; + } + return -ENOBUFS; +} + +/* Look for activity on active sockets */ +static void process_sockets(void) +{ + struct list_head *list; + struct list_head *temp; + int count = 0; + + spin_lock_bh(&read_sockets_lock); + list_for_each_safe(list, temp, &read_sockets) { + + struct connection *con = + list_entry(list, struct connection, read_list); + list_del(&con->read_list); + clear_bit(CF_READ_PENDING, &con->flags); + + spin_unlock_bh(&read_sockets_lock); + + /* This can reach zero if we are processing requests + * as they come in. + */ + if (atomic_read(&con->waiting_requests) == 0) { + spin_lock_bh(&read_sockets_lock); + continue; + } + + do { + con->rx_action(con); + + /* Don't starve out everyone else */ + if (++count >= MAX_RX_MSG_COUNT) { + cond_resched(); + count = 0; + } + + } while (!atomic_dec_and_test(&con->waiting_requests) && + !kthread_should_stop()); + + spin_lock_bh(&read_sockets_lock); + } + spin_unlock_bh(&read_sockets_lock); +} + +/* Try to send any messages that are pending + */ +static void process_output_queue(void) +{ + struct list_head *list; + struct list_head *temp; + + spin_lock_bh(&write_sockets_lock); + list_for_each_safe(list, temp, &write_sockets) { + struct connection *con = + list_entry(list, struct connection, write_list); + clear_bit(CF_WRITE_PENDING, &con->flags); + list_del(&con->write_list); + + spin_unlock_bh(&write_sockets_lock); + send_to_sock(con); + spin_lock_bh(&write_sockets_lock); + } + spin_unlock_bh(&write_sockets_lock); +} + +static void process_state_queue(void) +{ + struct list_head *list; + struct list_head *temp; + + spin_lock_bh(&state_sockets_lock); + list_for_each_safe(list, temp, &state_sockets) { + struct connection *con = + list_entry(list, struct connection, state_list); + list_del(&con->state_list); + clear_bit(CF_CONNECT_PENDING, &con->flags); + spin_unlock_bh(&state_sockets_lock); + + connect_to_sock(con); + spin_lock_bh(&state_sockets_lock); + } + spin_unlock_bh(&state_sockets_lock); +} + + +/* Discard all entries on the write queues */ +static void clean_writequeues(void) +{ + int nodeid; + + for (nodeid = 1; nodeid < conn_array_size; nodeid++) { + struct connection *con = nodeid2con(nodeid, 0); + + if (con) + clean_one_writequeue(con); + } +} + +static int read_list_empty(void) +{ + int status; + + spin_lock_bh(&read_sockets_lock); + status = list_empty(&read_sockets); + spin_unlock_bh(&read_sockets_lock); + + return status; +} + +/* DLM Transport comms receive daemon */ +static int dlm_recvd(void *data) +{ + init_waitqueue_entry(&lowcomms_recv_waitq_head, current); + add_wait_queue(&lowcomms_recv_waitq, &lowcomms_recv_waitq_head); + + while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); + if (read_list_empty()) + cond_resched(); + set_current_state(TASK_RUNNING); + + process_sockets(); + } + + return 0; +} + +static int write_and_state_lists_empty(void) +{ + int status; + + spin_lock_bh(&write_sockets_lock); + status = list_empty(&write_sockets); + spin_unlock_bh(&write_sockets_lock); + + spin_lock_bh(&state_sockets_lock); + if (list_empty(&state_sockets) == 0) + status = 0; + spin_unlock_bh(&state_sockets_lock); + + return status; +} + +/* DLM Transport send daemon */ +static int dlm_sendd(void *data) +{ + init_waitqueue_entry(&lowcomms_send_waitq_head, current); + add_wait_queue(&lowcomms_send_waitq, &lowcomms_send_waitq_head); + + while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); + if (write_and_state_lists_empty()) + cond_resched(); + set_current_state(TASK_RUNNING); + + process_state_queue(); + process_output_queue(); + } + + return 0; +} + +static void daemons_stop(void) +{ + kthread_stop(recv_task); + kthread_stop(send_task); +} + +static int daemons_start(void) +{ + struct task_struct *p; + int error; + + p = kthread_run(dlm_recvd, NULL, "dlm_recvd"); + error = IS_ERR(p); + if (error) { + log_print("can't start dlm_recvd %d", error); + return error; + } + recv_task = p; + + p = kthread_run(dlm_sendd, NULL, "dlm_sendd"); + error = IS_ERR(p); + if (error) { + log_print("can't start dlm_sendd %d", error); + kthread_stop(recv_task); + return error; + } + send_task = p; + + return 0; +} + +/* + * Return the largest buffer size we can cope with. + */ +int lowcomms_max_buffer_size(void) +{ + return PAGE_CACHE_SIZE; +} + +void dlm_lowcomms_stop(void) +{ + int i; + + /* Set all the flags to prevent any + socket activity. + */ + for (i = 0; i < conn_array_size; i++) { + if (connections[i]) + connections[i]->flags |= 0xFF; + } + + daemons_stop(); + clean_writequeues(); + + for (i = 0; i < conn_array_size; i++) { + if (connections[i]) { + close_connection(connections[i], true); + if (connections[i]->othercon) + kmem_cache_free(con_cache, connections[i]->othercon); + kmem_cache_free(con_cache, connections[i]); + } + } + + kfree(connections); + connections = NULL; + + kmem_cache_destroy(con_cache); +} + +/* This is quite likely to sleep... */ +int dlm_lowcomms_start(void) +{ + int error = 0; + + error = -ENOMEM; + connections = kzalloc(sizeof(struct connection *) * + NODE_INCREMENT, GFP_KERNEL); + if (!connections) + goto out; + + conn_array_size = NODE_INCREMENT; + + if (dlm_our_addr(&dlm_local_addr, 0)) { + log_print("no local IP address has been set"); + goto fail_free_conn; + } + if (!dlm_our_addr(&dlm_local_addr, 1)) { + log_print("This dlm comms module does not support multi-homed clustering"); + goto fail_free_conn; + } + + con_cache = kmem_cache_create("dlm_conn", sizeof(struct connection), + __alignof__(struct connection), 0, + NULL, NULL); + if (!con_cache) + goto fail_free_conn; + + + /* Start listening */ + error = listen_for_all(); + if (error) + goto fail_unlisten; + + error = daemons_start(); + if (error) + goto fail_unlisten; + + return 0; + +fail_unlisten: + close_connection(connections[0], false); + kmem_cache_free(con_cache, connections[0]); + kmem_cache_destroy(con_cache); + +fail_free_conn: + kfree(connections); + +out: + return error; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/fs/dlm/lowcomms.h b/fs/dlm/lowcomms.h index 2d045e0..a9a9618 100644 --- a/fs/dlm/lowcomms.h +++ b/fs/dlm/lowcomms.h @@ -14,8 +14,6 @@ #ifndef __LOWCOMMS_DOT_H__ #define __LOWCOMMS_DOT_H__ -int dlm_lowcomms_init(void); -void dlm_lowcomms_exit(void); int dlm_lowcomms_start(void); void dlm_lowcomms_stop(void); int dlm_lowcomms_close(int nodeid); diff --git a/fs/dlm/main.c b/fs/dlm/main.c index a8da8dc..162fbae 100644 --- a/fs/dlm/main.c +++ b/fs/dlm/main.c @@ -16,7 +16,6 @@ #include "lock.h" #include "user.h" #include "memory.h" -#include "lowcomms.h" #include "config.h" #ifdef CONFIG_DLM_DEBUG @@ -47,20 +46,14 @@ static int __init init_dlm(void) if (error) goto out_config; - error = dlm_lowcomms_init(); - if (error) - goto out_debug; - error = dlm_user_init(); if (error) - goto out_lowcomms; + goto out_debug; printk("DLM (built %s %s) installed\n", __DATE__, __TIME__); return 0; - out_lowcomms: - dlm_lowcomms_exit(); out_debug: dlm_unregister_debugfs(); out_config: @@ -76,7 +69,6 @@ static int __init init_dlm(void) static void __exit exit_dlm(void) { dlm_user_exit(); - dlm_lowcomms_exit(); dlm_config_exit(); dlm_memory_exit(); dlm_lockspace_exit(); diff --git a/fs/dlm/member.c b/fs/dlm/member.c index a3f7de7..85e2897 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -186,6 +186,14 @@ int dlm_recover_members(struct dlm_ls *ls, struct dlm_recover *rv, int *neg_out) struct dlm_member *memb, *safe; int i, error, found, pos = 0, neg = 0, low = -1; + /* previously removed members that we've not finished removing need to + count as a negative change so the "neg" recovery steps will happen */ + + list_for_each_entry(memb, &ls->ls_nodes_gone, list) { + log_debug(ls, "prev removed member %d", memb->nodeid); + neg++; + } + /* move departed members from ls_nodes to ls_nodes_gone */ list_for_each_entry_safe(memb, safe, &ls->ls_nodes, list) { diff --git a/fs/dlm/memory.c b/fs/dlm/memory.c index 989b608..5352b03 100644 --- a/fs/dlm/memory.c +++ b/fs/dlm/memory.c @@ -15,7 +15,7 @@ #include "config.h" #include "memory.h" -static kmem_cache_t *lkb_cache; +static struct kmem_cache *lkb_cache; int dlm_memory_init(void) diff --git a/fs/dlm/rcom.c b/fs/dlm/rcom.c index 518239a..4cc31be 100644 --- a/fs/dlm/rcom.c +++ b/fs/dlm/rcom.c @@ -90,13 +90,28 @@ static int check_config(struct dlm_ls *ls, struct rcom_config *rf, int nodeid) return 0; } +static void allow_sync_reply(struct dlm_ls *ls, uint64_t *new_seq) +{ + spin_lock(&ls->ls_rcom_spin); + *new_seq = ++ls->ls_rcom_seq; + set_bit(LSFL_RCOM_WAIT, &ls->ls_flags); + spin_unlock(&ls->ls_rcom_spin); +} + +static void disallow_sync_reply(struct dlm_ls *ls) +{ + spin_lock(&ls->ls_rcom_spin); + clear_bit(LSFL_RCOM_WAIT, &ls->ls_flags); + clear_bit(LSFL_RCOM_READY, &ls->ls_flags); + spin_unlock(&ls->ls_rcom_spin); +} + int dlm_rcom_status(struct dlm_ls *ls, int nodeid) { struct dlm_rcom *rc; struct dlm_mhandle *mh; int error = 0; - memset(ls->ls_recover_buf, 0, dlm_config.buffer_size); ls->ls_recover_nodeid = nodeid; if (nodeid == dlm_our_nodeid()) { @@ -108,12 +123,14 @@ int dlm_rcom_status(struct dlm_ls *ls, int nodeid) error = create_rcom(ls, nodeid, DLM_RCOM_STATUS, 0, &rc, &mh); if (error) goto out; - rc->rc_id = ++ls->ls_rcom_seq; + + allow_sync_reply(ls, &rc->rc_id); + memset(ls->ls_recover_buf, 0, dlm_config.buffer_size); send_rcom(ls, mh, rc); error = dlm_wait_function(ls, &rcom_response); - clear_bit(LSFL_RCOM_READY, &ls->ls_flags); + disallow_sync_reply(ls); if (error) goto out; @@ -150,14 +167,21 @@ static void receive_rcom_status(struct dlm_ls *ls, struct dlm_rcom *rc_in) static void receive_sync_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) { - if (rc_in->rc_id != ls->ls_rcom_seq) { - log_debug(ls, "reject old reply %d got %llx wanted %llx", - rc_in->rc_type, rc_in->rc_id, ls->ls_rcom_seq); - return; + spin_lock(&ls->ls_rcom_spin); + if (!test_bit(LSFL_RCOM_WAIT, &ls->ls_flags) || + rc_in->rc_id != ls->ls_rcom_seq) { + log_debug(ls, "reject reply %d from %d seq %llx expect %llx", + rc_in->rc_type, rc_in->rc_header.h_nodeid, + (unsigned long long)rc_in->rc_id, + (unsigned long long)ls->ls_rcom_seq); + goto out; } memcpy(ls->ls_recover_buf, rc_in, rc_in->rc_header.h_length); set_bit(LSFL_RCOM_READY, &ls->ls_flags); + clear_bit(LSFL_RCOM_WAIT, &ls->ls_flags); wake_up(&ls->ls_wait_general); + out: + spin_unlock(&ls->ls_rcom_spin); } static void receive_rcom_status_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) @@ -171,7 +195,6 @@ int dlm_rcom_names(struct dlm_ls *ls, int nodeid, char *last_name, int last_len) struct dlm_mhandle *mh; int error = 0, len = sizeof(struct dlm_rcom); - memset(ls->ls_recover_buf, 0, dlm_config.buffer_size); ls->ls_recover_nodeid = nodeid; if (nodeid == dlm_our_nodeid()) { @@ -185,12 +208,14 @@ int dlm_rcom_names(struct dlm_ls *ls, int nodeid, char *last_name, int last_len) if (error) goto out; memcpy(rc->rc_buf, last_name, last_len); - rc->rc_id = ++ls->ls_rcom_seq; + + allow_sync_reply(ls, &rc->rc_id); + memset(ls->ls_recover_buf, 0, dlm_config.buffer_size); send_rcom(ls, mh, rc); error = dlm_wait_function(ls, &rcom_response); - clear_bit(LSFL_RCOM_READY, &ls->ls_flags); + disallow_sync_reply(ls); out: return error; } @@ -370,9 +395,10 @@ static void receive_rcom_lock_reply(struct dlm_ls *ls, struct dlm_rcom *rc_in) static int send_ls_not_ready(int nodeid, struct dlm_rcom *rc_in) { struct dlm_rcom *rc; + struct rcom_config *rf; struct dlm_mhandle *mh; char *mb; - int mb_len = sizeof(struct dlm_rcom); + int mb_len = sizeof(struct dlm_rcom) + sizeof(struct rcom_config); mh = dlm_lowcomms_get_buffer(nodeid, mb_len, GFP_KERNEL, &mb); if (!mh) @@ -391,6 +417,9 @@ static int send_ls_not_ready(int nodeid, struct dlm_rcom *rc_in) rc->rc_id = rc_in->rc_id; rc->rc_result = -ESRCH; + rf = (struct rcom_config *) rc->rc_buf; + rf->rf_lvblen = -1; + dlm_rcom_out(rc); dlm_lowcomms_commit_buffer(mh); @@ -412,9 +441,10 @@ void dlm_receive_rcom(struct dlm_header *hd, int nodeid) ls = dlm_find_lockspace_global(hd->h_lockspace); if (!ls) { - log_print("lockspace %x from %d not found", - hd->h_lockspace, nodeid); - send_ls_not_ready(nodeid, rc); + log_print("lockspace %x from %d type %x not found", + hd->h_lockspace, nodeid, rc->rc_type); + if (rc->rc_type == DLM_RCOM_STATUS) + send_ls_not_ready(nodeid, rc); return; } diff --git a/fs/dlm/recover.c b/fs/dlm/recover.c index a5e6d18..cf9f683 100644 --- a/fs/dlm/recover.c +++ b/fs/dlm/recover.c @@ -252,6 +252,7 @@ static void recover_list_clear(struct dlm_ls *ls) spin_lock(&ls->ls_recover_list_lock); list_for_each_entry_safe(r, s, &ls->ls_recover_list, res_recover_list) { list_del_init(&r->res_recover_list); + r->res_recover_locks_count = 0; dlm_put_rsb(r); ls->ls_recover_list_count--; } diff --git a/fs/dlm/recoverd.c b/fs/dlm/recoverd.c index 362e3ef..650536aa 100644 --- a/fs/dlm/recoverd.c +++ b/fs/dlm/recoverd.c @@ -45,7 +45,7 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) unsigned long start; int error, neg = 0; - log_debug(ls, "recover %llx", rv->seq); + log_debug(ls, "recover %llx", (unsigned long long)rv->seq); mutex_lock(&ls->ls_recoverd_active); @@ -94,14 +94,6 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) } /* - * Purge directory-related requests that are saved in requestqueue. - * All dir requests from before recovery are invalid now due to the dir - * rebuild and will be resent by the requesting nodes. - */ - - dlm_purge_requestqueue(ls); - - /* * Wait for all nodes to complete directory rebuild. */ @@ -164,10 +156,31 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) */ dlm_recover_rsbs(ls); + } else { + /* + * Other lockspace members may be going through the "neg" steps + * while also adding us to the lockspace, in which case they'll + * be doing the recover_locks (RS_LOCKS) barrier. + */ + dlm_set_recover_status(ls, DLM_RS_LOCKS); + + error = dlm_recover_locks_wait(ls); + if (error) { + log_error(ls, "recover_locks_wait failed %d", error); + goto fail; + } } dlm_release_root_list(ls); + /* + * Purge directory-related requests that are saved in requestqueue. + * All dir requests from before recovery are invalid now due to the dir + * rebuild and will be resent by the requesting nodes. + */ + + dlm_purge_requestqueue(ls); + dlm_set_recover_status(ls, DLM_RS_DONE); error = dlm_recover_done_wait(ls); if (error) { @@ -199,7 +212,8 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) dlm_astd_wake(); - log_debug(ls, "recover %llx done: %u ms", rv->seq, + log_debug(ls, "recover %llx done: %u ms", + (unsigned long long)rv->seq, jiffies_to_msecs(jiffies - start)); mutex_unlock(&ls->ls_recoverd_active); @@ -207,11 +221,16 @@ static int ls_recover(struct dlm_ls *ls, struct dlm_recover *rv) fail: dlm_release_root_list(ls); - log_debug(ls, "recover %llx error %d", rv->seq, error); + log_debug(ls, "recover %llx error %d", + (unsigned long long)rv->seq, error); mutex_unlock(&ls->ls_recoverd_active); return error; } +/* The dlm_ls_start() that created the rv we take here may already have been + stopped via dlm_ls_stop(); in that case we need to leave the RECOVERY_STOP + flag set. */ + static void do_ls_recovery(struct dlm_ls *ls) { struct dlm_recover *rv = NULL; @@ -219,7 +238,8 @@ static void do_ls_recovery(struct dlm_ls *ls) spin_lock(&ls->ls_recover_lock); rv = ls->ls_recover_args; ls->ls_recover_args = NULL; - clear_bit(LSFL_RECOVERY_STOP, &ls->ls_flags); + if (rv && ls->ls_recover_seq == rv->seq) + clear_bit(LSFL_RECOVERY_STOP, &ls->ls_flags); spin_unlock(&ls->ls_recover_lock); if (rv) { diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index 7b2b089..65008d7 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -30,26 +30,36 @@ struct rq_entry { * lockspace is enabled on some while still suspended on others. */ -void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd) +int dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd) { struct rq_entry *e; int length = hd->h_length; - - if (dlm_is_removed(ls, nodeid)) - return; + int rv = 0; e = kmalloc(sizeof(struct rq_entry) + length, GFP_KERNEL); if (!e) { log_print("dlm_add_requestqueue: out of memory\n"); - return; + return 0; } e->nodeid = nodeid; memcpy(e->request, hd, length); + /* We need to check dlm_locking_stopped() after taking the mutex to + avoid a race where dlm_recoverd enables locking and runs + process_requestqueue between our earlier dlm_locking_stopped check + and this addition to the requestqueue. */ + mutex_lock(&ls->ls_requestqueue_mutex); - list_add_tail(&e->list, &ls->ls_requestqueue); + if (dlm_locking_stopped(ls)) + list_add_tail(&e->list, &ls->ls_requestqueue); + else { + log_debug(ls, "dlm_add_requestqueue skip from %d", nodeid); + kfree(e); + rv = -EAGAIN; + } mutex_unlock(&ls->ls_requestqueue_mutex); + return rv; } int dlm_process_requestqueue(struct dlm_ls *ls) @@ -120,6 +130,10 @@ static int purge_request(struct dlm_ls *ls, struct dlm_message *ms, int nodeid) { uint32_t type = ms->m_type; + /* the ls is being cleaned up and freed by release_lockspace */ + if (!ls->ls_count) + return 1; + if (dlm_is_removed(ls, nodeid)) return 1; diff --git a/fs/dlm/requestqueue.h b/fs/dlm/requestqueue.h index 349f0d2..6a53ea0 100644 --- a/fs/dlm/requestqueue.h +++ b/fs/dlm/requestqueue.h @@ -13,7 +13,7 @@ #ifndef __REQUESTQUEUE_DOT_H__ #define __REQUESTQUEUE_DOT_H__ -void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd); +int dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, struct dlm_header *hd); int dlm_process_requestqueue(struct dlm_ls *ls); void dlm_wait_requestqueue(struct dlm_ls *ls); void dlm_purge_requestqueue(struct dlm_ls *ls); diff --git a/fs/dnotify.c b/fs/dnotify.c index 2b0442d..1f26a2b 100644 --- a/fs/dnotify.c +++ b/fs/dnotify.c @@ -23,7 +23,7 @@ int dir_notify_enable __read_mostly = 1; -static kmem_cache_t *dn_cache __read_mostly; +static struct kmem_cache *dn_cache __read_mostly; static void redo_inode_mask(struct inode *inode) { @@ -77,7 +77,7 @@ int fcntl_dirnotify(int fd, struct file *filp, unsigned long arg) inode = filp->f_dentry->d_inode; if (!S_ISDIR(inode->i_mode)) return -ENOTDIR; - dn = kmem_cache_alloc(dn_cache, SLAB_KERNEL); + dn = kmem_cache_alloc(dn_cache, GFP_KERNEL); if (dn == NULL) return -ENOMEM; spin_lock(&inode->i_lock); @@ -131,7 +131,7 @@ static struct quota_format_type *quota_formats; /* List of registered formats */ static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES; /* SLAB cache for dquot structures */ -static kmem_cache_t *dquot_cachep; +static struct kmem_cache *dquot_cachep; int register_quota_format(struct quota_format_type *fmt) { @@ -600,7 +600,7 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type) { struct dquot *dquot; - dquot = kmem_cache_alloc(dquot_cachep, SLAB_NOFS); + dquot = kmem_cache_alloc(dquot_cachep, GFP_NOFS); if(!dquot) return NODQUOT; diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f63a775..7196f50 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -628,7 +628,7 @@ int ecryptfs_decrypt_page(struct file *file, struct page *page) num_extents_per_page = PAGE_CACHE_SIZE / crypt_stat->extent_size; base_extent = (page->index * num_extents_per_page); lower_page_virt = kmem_cache_alloc(ecryptfs_lower_page_cache, - SLAB_KERNEL); + GFP_KERNEL); if (!lower_page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Error getting page for encrypted " @@ -1334,7 +1334,7 @@ int ecryptfs_write_headers(struct dentry *ecryptfs_dentry, goto out; } /* Released in this function */ - page_virt = kmem_cache_alloc(ecryptfs_header_cache_0, SLAB_USER); + page_virt = kmem_cache_alloc(ecryptfs_header_cache_0, GFP_USER); if (!page_virt) { ecryptfs_printk(KERN_ERR, "Out of memory\n"); rc = -ENOMEM; @@ -1493,7 +1493,7 @@ int ecryptfs_read_headers(struct dentry *ecryptfs_dentry, &ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->crypt_stat; /* Read the first page from the underlying file */ - page_virt = kmem_cache_alloc(ecryptfs_header_cache_1, SLAB_USER); + page_virt = kmem_cache_alloc(ecryptfs_header_cache_1, GFP_USER); if (!page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Unable to allocate page_virt\n"); diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index a92ef05..42099e7 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -250,7 +250,7 @@ static int ecryptfs_open(struct inode *inode, struct file *file) int lower_flags; /* Released in ecryptfs_release or end of function if failure */ - file_info = kmem_cache_alloc(ecryptfs_file_info_cache, SLAB_KERNEL); + file_info = kmem_cache_alloc(ecryptfs_file_info_cache, GFP_KERNEL); ecryptfs_set_file_private(file, file_info); if (!file_info) { ecryptfs_printk(KERN_ERR, diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index dfcc684..8a1945a 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -369,7 +369,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, BUG_ON(!atomic_read(&lower_dentry->d_count)); ecryptfs_set_dentry_private(dentry, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(dentry)) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, "Out of memory whilst attempting " @@ -404,7 +404,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, /* Released in this function */ page_virt = (char *)kmem_cache_alloc(ecryptfs_header_cache_2, - SLAB_USER); + GFP_USER); if (!page_virt) { rc = -ENOMEM; ecryptfs_printk(KERN_ERR, @@ -795,7 +795,7 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) /* Released at out_free: label */ ecryptfs_set_file_private(&fake_ecryptfs_file, kmem_cache_alloc(ecryptfs_file_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (unlikely(!ecryptfs_file_to_private(&fake_ecryptfs_file))) { rc = -ENOMEM; goto out; diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index c3746f5..745c0f1 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -207,7 +207,7 @@ parse_tag_3_packet(struct ecryptfs_crypt_stat *crypt_stat, /* Released: wipe_auth_tok_list called in ecryptfs_parse_packet_set or * at end of function upon failure */ auth_tok_list_item = - kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, SLAB_KERNEL); + kmem_cache_alloc(ecryptfs_auth_tok_list_item_cache, GFP_KERNEL); if (!auth_tok_list_item) { ecryptfs_printk(KERN_ERR, "Unable to allocate memory\n"); rc = -ENOMEM; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a78d87d..3ede12b 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -378,7 +378,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* Released in ecryptfs_put_super() */ ecryptfs_set_superblock_private(sb, kmem_cache_alloc(ecryptfs_sb_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_superblock_to_private(sb)) { ecryptfs_printk(KERN_WARNING, "Out of memory\n"); rc = -ENOMEM; @@ -402,7 +402,7 @@ ecryptfs_fill_super(struct super_block *sb, void *raw_data, int silent) /* through deactivate_super(sb) from get_sb_nodev() */ ecryptfs_set_dentry_private(sb->s_root, kmem_cache_alloc(ecryptfs_dentry_info_cache, - SLAB_KERNEL)); + GFP_KERNEL)); if (!ecryptfs_dentry_to_private(sb->s_root)) { ecryptfs_printk(KERN_ERR, "dentry_info_cache alloc failed\n"); @@ -546,7 +546,7 @@ inode_info_init_once(void *vptr, struct kmem_cache *cachep, unsigned long flags) } static struct ecryptfs_cache_info { - kmem_cache_t **cache; + struct kmem_cache **cache; const char *name; size_t size; void (*ctor)(void*, struct kmem_cache *, unsigned long); @@ -691,7 +691,7 @@ static ssize_t version_show(struct ecryptfs_obj *obj, char *buff) static struct ecryptfs_attribute sysfs_attr_version = __ATTR_RO(version); -struct ecryptfs_version_str_map_elem { +static struct ecryptfs_version_str_map_elem { u32 flag; char *str; } ecryptfs_version_str_map[] = { diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 825757a..eaa5daa 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -50,7 +50,7 @@ static struct inode *ecryptfs_alloc_inode(struct super_block *sb) struct inode *inode = NULL; ecryptfs_inode = kmem_cache_alloc(ecryptfs_inode_info_cache, - SLAB_KERNEL); + GFP_KERNEL); if (unlikely(!ecryptfs_inode)) goto out; ecryptfs_init_crypt_stat(&ecryptfs_inode->crypt_stat); diff --git a/fs/efs/super.c b/fs/efs/super.c index b3f5065..dfebf21 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -52,12 +52,12 @@ static struct pt_types sgi_pt_types[] = { }; -static kmem_cache_t * efs_inode_cachep; +static struct kmem_cache * efs_inode_cachep; static struct inode *efs_alloc_inode(struct super_block *sb) { struct efs_inode_info *ei; - ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, SLAB_KERNEL); + ei = (struct efs_inode_info *)kmem_cache_alloc(efs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -68,7 +68,7 @@ static void efs_destroy_inode(struct inode *inode) kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index ae228ec..88a6f8d 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -283,10 +283,10 @@ static struct mutex epmutex; static struct poll_safewake psw; /* Slab cache used to allocate "struct epitem" */ -static kmem_cache_t *epi_cache __read_mostly; +static struct kmem_cache *epi_cache __read_mostly; /* Slab cache used to allocate "struct eppoll_entry" */ -static kmem_cache_t *pwq_cache __read_mostly; +static struct kmem_cache *pwq_cache __read_mostly; /* Virtual fs used to allocate inodes for eventpoll files */ static struct vfsmount *eventpoll_mnt __read_mostly; @@ -961,7 +961,7 @@ static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead, struct epitem *epi = ep_item_from_epqueue(pt); struct eppoll_entry *pwq; - if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, SLAB_KERNEL))) { + if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) { init_waitqueue_func_entry(&pwq->wait, ep_poll_callback); pwq->whead = whead; pwq->base = epi; @@ -1004,7 +1004,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, struct ep_pqueue epq; error = -ENOMEM; - if (!(epi = kmem_cache_alloc(epi_cache, SLAB_KERNEL))) + if (!(epi = kmem_cache_alloc(epi_cache, GFP_KERNEL))) goto eexit_1; /* Item initialization follow here ... */ @@ -404,7 +404,7 @@ int setup_arg_pages(struct linux_binprm *bprm, bprm->loader += stack_base; bprm->exec += stack_base; - mpnt = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + mpnt = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!mpnt) return -ENOMEM; @@ -1515,7 +1515,8 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) ispipe = 1; } else file = filp_open(corename, - O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE, 0600); + O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, + 0600); if (IS_ERR(file)) goto fail_unlock; inode = file->f_dentry->d_inode; diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c index 1dfba77..e3cf8c8 100644 --- a/fs/ext2/ioctl.c +++ b/fs/ext2/ioctl.c @@ -44,6 +44,7 @@ int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, if (!S_ISDIR(inode->i_mode)) flags &= ~EXT2_DIRSYNC_FL; + mutex_lock(&inode->i_mutex); oldflags = ei->i_flags; /* @@ -53,13 +54,16 @@ int ext2_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, * This test looks nicer. Thanks to Pauline Middelink */ if ((flags ^ oldflags) & (EXT2_APPEND_FL | EXT2_IMMUTABLE_FL)) { - if (!capable(CAP_LINUX_IMMUTABLE)) + if (!capable(CAP_LINUX_IMMUTABLE)) { + mutex_unlock(&inode->i_mutex); return -EPERM; + } } flags = flags & EXT2_FL_USER_MODIFIABLE; flags |= oldflags & ~EXT2_FL_USER_MODIFIABLE; ei->i_flags = flags; + mutex_unlock(&inode->i_mutex); ext2_set_inode_flags(inode); inode->i_ctime = CURRENT_TIME_SEC; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index d8b9abd..255cef5 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -135,12 +135,12 @@ static void ext2_put_super (struct super_block * sb) return; } -static kmem_cache_t * ext2_inode_cachep; +static struct kmem_cache * ext2_inode_cachep; static struct inode *ext2_alloc_inode(struct super_block *sb) { struct ext2_inode_info *ei; - ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, SLAB_KERNEL); + ei = (struct ext2_inode_info *)kmem_cache_alloc(ext2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; #ifdef CONFIG_EXT2_FS_POSIX_ACL @@ -156,7 +156,7 @@ static void ext2_destroy_inode(struct inode *inode) kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; @@ -1090,8 +1090,10 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) { struct super_block *sb = dentry->d_sb; struct ext2_sb_info *sbi = EXT2_SB(sb); + struct ext2_super_block *es = sbi->s_es; unsigned long overhead; int i; + u64 fsid; if (test_opt (sb, MINIX_DF)) overhead = 0; @@ -1104,7 +1106,7 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) * All of the blocks before first_data_block are * overhead */ - overhead = le32_to_cpu(sbi->s_es->s_first_data_block); + overhead = le32_to_cpu(es->s_first_data_block); /* * Add the overhead attributed to the superblock and @@ -1125,14 +1127,18 @@ static int ext2_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_type = EXT2_SUPER_MAGIC; buf->f_bsize = sb->s_blocksize; - buf->f_blocks = le32_to_cpu(sbi->s_es->s_blocks_count) - overhead; + buf->f_blocks = le32_to_cpu(es->s_blocks_count) - overhead; buf->f_bfree = ext2_count_free_blocks(sb); - buf->f_bavail = buf->f_bfree - le32_to_cpu(sbi->s_es->s_r_blocks_count); - if (buf->f_bfree < le32_to_cpu(sbi->s_es->s_r_blocks_count)) + buf->f_bavail = buf->f_bfree - le32_to_cpu(es->s_r_blocks_count); + if (buf->f_bfree < le32_to_cpu(es->s_r_blocks_count)) buf->f_bavail = 0; - buf->f_files = le32_to_cpu(sbi->s_es->s_inodes_count); - buf->f_ffree = ext2_count_free_inodes (sb); + buf->f_files = le32_to_cpu(es->s_inodes_count); + buf->f_ffree = ext2_count_free_inodes(sb); buf->f_namelen = EXT2_NAME_LEN; + fsid = le64_to_cpup((void *)es->s_uuid) ^ + le64_to_cpup((void *)es->s_uuid + sizeof(u64)); + buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL; + buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL; return 0; } diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index af52a7f..247efd0 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -342,12 +342,9 @@ static void ext2_xattr_update_super_block(struct super_block *sb) if (EXT2_HAS_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR)) return; - lock_super(sb); - EXT2_SB(sb)->s_es->s_feature_compat |= - cpu_to_le32(EXT2_FEATURE_COMPAT_EXT_ATTR); + EXT2_SET_COMPAT_FEATURE(sb, EXT2_FEATURE_COMPAT_EXT_ATTR); sb->s_dirt = 1; mark_buffer_dirty(EXT2_SB(sb)->s_sbh); - unlock_super(sb); } /* diff --git a/fs/ext3/Makefile b/fs/ext3/Makefile index 704cd44..e77766a 100644 --- a/fs/ext3/Makefile +++ b/fs/ext3/Makefile @@ -5,7 +5,7 @@ obj-$(CONFIG_EXT3_FS) += ext3.o ext3-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ - ioctl.o namei.o super.o symlink.o hash.o resize.o + ioctl.o namei.o super.o symlink.o hash.o resize.o ext3_jbd.o ext3-$(CONFIG_EXT3_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext3-$(CONFIG_EXT3_FS_POSIX_ACL) += acl.o diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index b41a7d7..2216174 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -144,7 +144,7 @@ restart: printk("Block Allocation Reservation Windows Map (%s):\n", fn); while (n) { - rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node); + rsv = rb_entry(n, struct ext3_reserve_window_node, rsv_node); if (verbose) printk("reservation window 0x%p " "start: %lu, end: %lu\n", @@ -730,7 +730,7 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh, here = 0; p = ((char *)bh->b_data) + (here >> 3); - r = memscan(p, 0, (maxblocks - here + 7) >> 3); + r = memscan(p, 0, ((maxblocks + 7) >> 3) - (here >> 3)); next = (r - ((char *)bh->b_data)) << 3; if (next < maxblocks && next >= start && ext3_test_allocatable(next, bh)) @@ -949,7 +949,7 @@ static int find_next_reservable_window( prev = rsv; next = rb_next(&rsv->rsv_node); - rsv = list_entry(next,struct ext3_reserve_window_node,rsv_node); + rsv = rb_entry(next,struct ext3_reserve_window_node,rsv_node); /* * Reached the last reservation, we can just append to the @@ -1148,7 +1148,7 @@ retry: * check if the first free block is within the * free space we just reserved */ - if (start_block >= my_rsv->rsv_start && start_block < my_rsv->rsv_end) + if (start_block >= my_rsv->rsv_start && start_block <= my_rsv->rsv_end) return 0; /* success */ /* * if the first free bit we found is out of the reservable space @@ -1193,7 +1193,7 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv, if (!next) my_rsv->rsv_end += size; else { - next_rsv = list_entry(next, struct ext3_reserve_window_node, rsv_node); + next_rsv = rb_entry(next, struct ext3_reserve_window_node, rsv_node); if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) my_rsv->rsv_end += size; @@ -1271,7 +1271,7 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, } /* * grp_goal is a group relative block number (if there is a goal) - * 0 < grp_goal < EXT3_BLOCKS_PER_GROUP(sb) + * 0 <= grp_goal < EXT3_BLOCKS_PER_GROUP(sb) * first block is a filesystem wide block number * first block is the block number of the first block in this group */ @@ -1307,10 +1307,14 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, if (!goal_in_my_reservation(&my_rsv->rsv_window, grp_goal, group, sb)) grp_goal = -1; - } else if (grp_goal > 0 && - (my_rsv->rsv_end-grp_goal+1) < *count) - try_to_extend_reservation(my_rsv, sb, - *count-my_rsv->rsv_end + grp_goal - 1); + } else if (grp_goal >= 0) { + int curr = my_rsv->rsv_end - + (grp_goal + group_first_block) + 1; + + if (curr < *count) + try_to_extend_reservation(my_rsv, sb, + *count - curr); + } if ((my_rsv->rsv_start > group_last_block) || (my_rsv->rsv_end < group_first_block)) { @@ -1511,10 +1515,8 @@ retry_alloc: if (group_no >= ngroups) group_no = 0; gdp = ext3_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) { - *errp = -EIO; - goto out; - } + if (!gdp) + goto io_error; free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); /* * skip this group if the number of @@ -1548,6 +1550,7 @@ retry_alloc: */ if (my_rsv) { my_rsv = NULL; + windowsz = 0; group_no = goal_group; goto retry_alloc; } diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c index d0b54f3..5a9313e 100644 --- a/fs/ext3/dir.c +++ b/fs/ext3/dir.c @@ -154,6 +154,9 @@ static int ext3_readdir(struct file * filp, ext3_error (sb, "ext3_readdir", "directory #%lu contains a hole at offset %lu", inode->i_ino, (unsigned long)filp->f_pos); + /* corrupt size? Maybe no more blocks to read */ + if (filp->f_pos > inode->i_blocks << 9) + break; filp->f_pos += sb->s_blocksize - offset; continue; } diff --git a/fs/ext3/ext3_jbd.c b/fs/ext3/ext3_jbd.c new file mode 100644 index 0000000..e1f91fd --- /dev/null +++ b/fs/ext3/ext3_jbd.c @@ -0,0 +1,59 @@ +/* + * Interface between ext3 and JBD + */ + +#include <linux/ext3_jbd.h> + +int __ext3_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_undo_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_get_write_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = journal_forget(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_revoke(const char *where, handle_t *handle, + unsigned long blocknr, struct buffer_head *bh) +{ + int err = journal_revoke(handle, blocknr, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_get_create_access(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext3_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = journal_dirty_metadata(handle, bh); + if (err) + ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 03ba5bc..beaf25f 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1148,37 +1148,102 @@ static int do_journal_get_write_access(handle_t *handle, return ext3_journal_get_write_access(handle, bh); } +/* + * The idea of this helper function is following: + * if prepare_write has allocated some blocks, but not all of them, the + * transaction must include the content of the newly allocated blocks. + * This content is expected to be set to zeroes by block_prepare_write(). + * 2006/10/14 SAW + */ +static int ext3_prepare_failure(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + struct address_space *mapping; + struct buffer_head *bh, *head, *next; + unsigned block_start, block_end; + unsigned blocksize; + int ret; + handle_t *handle = ext3_journal_current_handle(); + + mapping = page->mapping; + if (ext3_should_writeback_data(mapping->host)) { + /* optimization: no constraints about data */ +skip: + return ext3_journal_stop(handle); + } + + head = page_buffers(page); + blocksize = head->b_size; + for ( bh = head, block_start = 0; + bh != head || !block_start; + block_start = block_end, bh = next) + { + next = bh->b_this_page; + block_end = block_start + blocksize; + if (block_end <= from) + continue; + if (block_start >= to) { + block_start = to; + break; + } + if (!buffer_mapped(bh)) + /* prepare_write failed on this bh */ + break; + if (ext3_should_journal_data(mapping->host)) { + ret = do_journal_get_write_access(handle, bh); + if (ret) { + ext3_journal_stop(handle); + return ret; + } + } + /* + * block_start here becomes the first block where the current iteration + * of prepare_write failed. + */ + } + if (block_start <= from) + goto skip; + + /* commit allocated and zeroed buffers */ + return mapping->a_ops->commit_write(file, page, from, block_start); +} + static int ext3_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { struct inode *inode = page->mapping->host; - int ret, needed_blocks = ext3_writepage_trans_blocks(inode); + int ret, ret2; + int needed_blocks = ext3_writepage_trans_blocks(inode); handle_t *handle; int retries = 0; retry: handle = ext3_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; - } + if (IS_ERR(handle)) + return PTR_ERR(handle); if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_prepare_write(page, from, to, ext3_get_block); else ret = block_prepare_write(page, from, to, ext3_get_block); if (ret) - goto prepare_write_failed; + goto failure; if (ext3_should_journal_data(inode)) { ret = walk_page_buffers(handle, page_buffers(page), from, to, NULL, do_journal_get_write_access); + if (ret) + /* fatal error, just put the handle and return */ + journal_stop(handle); } -prepare_write_failed: - if (ret) - ext3_journal_stop(handle); + return ret; + +failure: + ret2 = ext3_prepare_failure(file, page, from, to); + if (ret2 < 0) + return ret2; if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries)) goto retry; -out: + /* retry number exceeded, or other error like -EDQUOT */ return ret; } diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 906731a..60d2f9d 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -552,6 +552,15 @@ static int htree_dirblock_to_tree(struct file *dir_file, dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(0)); for (; de < top; de = ext3_next_entry(de)) { + if (!ext3_check_dir_entry("htree_dirblock_to_tree", dir, de, bh, + (block<<EXT3_BLOCK_SIZE_BITS(dir->i_sb)) + +((char *)de - bh->b_data))) { + /* On error, skip the f_pos to the next block. */ + dir_file->f_pos = (dir_file->f_pos | + (dir->i_sb->s_blocksize - 1)) + 1; + brelse (bh); + return count; + } ext3fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || ((hinfo->hash == start_hash) && diff --git a/fs/ext3/super.c b/fs/ext3/super.c index afc2d4f..580b8a6 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -436,7 +436,7 @@ static void ext3_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext3_inode_cachep; +static struct kmem_cache *ext3_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -445,7 +445,7 @@ static struct inode *ext3_alloc_inode(struct super_block *sb) { struct ext3_inode_info *ei; - ei = kmem_cache_alloc(ext3_inode_cachep, SLAB_NOFS); + ei = kmem_cache_alloc(ext3_inode_cachep, GFP_NOFS); if (!ei) return NULL; #ifdef CONFIG_EXT3_FS_POSIX_ACL @@ -462,7 +462,7 @@ static void ext3_destroy_inode(struct inode *inode) kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext3_inode_info *ei = (struct ext3_inode_info *) foo; @@ -1264,6 +1264,12 @@ static void ext3_orphan_cleanup (struct super_block * sb, return; } + if (bdev_read_only(sb->s_bdev)) { + printk(KERN_ERR "EXT3-fs: write access " + "unavailable, skipping orphan cleanup.\n"); + return; + } + if (EXT3_SB(sb)->s_mount_state & EXT3_ERROR_FS) { if (es->s_last_orphan) jbd_debug(1, "Errors on filesystem, " @@ -2387,6 +2393,7 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) struct ext3_super_block *es = sbi->s_es; ext3_fsblk_t overhead; int i; + u64 fsid; if (test_opt (sb, MINIX_DF)) overhead = 0; @@ -2433,6 +2440,10 @@ static int ext3_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_files = le32_to_cpu(es->s_inodes_count); buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); buf->f_namelen = EXT3_NAME_LEN; + fsid = le64_to_cpup((void *)es->s_uuid) ^ + le64_to_cpup((void *)es->s_uuid + sizeof(u64)); + buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL; + buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL; return 0; } diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index f86f248..99857a40 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -459,14 +459,11 @@ static void ext3_xattr_update_super_block(handle_t *handle, if (EXT3_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_EXT_ATTR)) return; - lock_super(sb); if (ext3_journal_get_write_access(handle, EXT3_SB(sb)->s_sbh) == 0) { - EXT3_SB(sb)->s_es->s_feature_compat |= - cpu_to_le32(EXT3_FEATURE_COMPAT_EXT_ATTR); + EXT3_SET_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_EXT_ATTR); sb->s_dirt = 1; ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); } - unlock_super(sb); } /* diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile index a6acb96..ae6e7e5 100644 --- a/fs/ext4/Makefile +++ b/fs/ext4/Makefile @@ -5,7 +5,8 @@ obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ - ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o + ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \ + ext4_jbd2.o ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 5d45582..c4dd110 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -165,7 +165,7 @@ restart: printk("Block Allocation Reservation Windows Map (%s):\n", fn); while (n) { - rsv = list_entry(n, struct ext4_reserve_window_node, rsv_node); + rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); if (verbose) printk("reservation window 0x%p " "start: %llu, end: %llu\n", @@ -747,7 +747,7 @@ find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, here = 0; p = ((char *)bh->b_data) + (here >> 3); - r = memscan(p, 0, (maxblocks - here + 7) >> 3); + r = memscan(p, 0, ((maxblocks + 7) >> 3) - (here >> 3)); next = (r - ((char *)bh->b_data)) << 3; if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh)) @@ -966,7 +966,7 @@ static int find_next_reservable_window( prev = rsv; next = rb_next(&rsv->rsv_node); - rsv = list_entry(next,struct ext4_reserve_window_node,rsv_node); + rsv = rb_entry(next,struct ext4_reserve_window_node,rsv_node); /* * Reached the last reservation, we can just append to the @@ -1165,7 +1165,7 @@ retry: * check if the first free block is within the * free space we just reserved */ - if (start_block >= my_rsv->rsv_start && start_block < my_rsv->rsv_end) + if (start_block >= my_rsv->rsv_start && start_block <= my_rsv->rsv_end) return 0; /* success */ /* * if the first free bit we found is out of the reservable space @@ -1210,7 +1210,7 @@ static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, if (!next) my_rsv->rsv_end += size; else { - next_rsv = list_entry(next, struct ext4_reserve_window_node, rsv_node); + next_rsv = rb_entry(next, struct ext4_reserve_window_node, rsv_node); if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) my_rsv->rsv_end += size; @@ -1288,7 +1288,7 @@ ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, } /* * grp_goal is a group relative block number (if there is a goal) - * 0 < grp_goal < EXT4_BLOCKS_PER_GROUP(sb) + * 0 <= grp_goal < EXT4_BLOCKS_PER_GROUP(sb) * first block is a filesystem wide block number * first block is the block number of the first block in this group */ @@ -1324,10 +1324,14 @@ ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, if (!goal_in_my_reservation(&my_rsv->rsv_window, grp_goal, group, sb)) grp_goal = -1; - } else if (grp_goal > 0 && - (my_rsv->rsv_end-grp_goal+1) < *count) - try_to_extend_reservation(my_rsv, sb, - *count-my_rsv->rsv_end + grp_goal - 1); + } else if (grp_goal >= 0) { + int curr = my_rsv->rsv_end - + (grp_goal + group_first_block) + 1; + + if (curr < *count) + try_to_extend_reservation(my_rsv, sb, + *count - curr); + } if ((my_rsv->rsv_start > group_last_block) || (my_rsv->rsv_end < group_first_block)) { @@ -1525,10 +1529,8 @@ retry_alloc: if (group_no >= ngroups) group_no = 0; gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); - if (!gdp) { - *errp = -EIO; - goto out; - } + if (!gdp) + goto io_error; free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); /* * skip this group if the number of @@ -1562,6 +1564,7 @@ retry_alloc: */ if (my_rsv) { my_rsv = NULL; + windowsz = 0; group_no = goal_group; goto retry_alloc; } diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c index f859578..f2ed3e7 100644 --- a/fs/ext4/dir.c +++ b/fs/ext4/dir.c @@ -153,6 +153,9 @@ static int ext4_readdir(struct file * filp, ext4_error (sb, "ext4_readdir", "directory #%lu contains a hole at offset %lu", inode->i_ino, (unsigned long)filp->f_pos); + /* corrupt size? Maybe no more blocks to read */ + if (filp->f_pos > inode->i_blocks << 9) + break; filp->f_pos += sb->s_blocksize - offset; continue; } diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c new file mode 100644 index 0000000..d6afe4e --- /dev/null +++ b/fs/ext4/ext4_jbd2.c @@ -0,0 +1,59 @@ +/* + * Interface between ext4 and JBD + */ + +#include <linux/ext4_jbd2.h> + +int __ext4_journal_get_undo_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_get_undo_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_get_write_access(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_get_write_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_forget(const char *where, handle_t *handle, + struct buffer_head *bh) +{ + int err = jbd2_journal_forget(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_revoke(const char *where, handle_t *handle, + ext4_fsblk_t blocknr, struct buffer_head *bh) +{ + int err = jbd2_journal_revoke(handle, blocknr, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_get_create_access(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = jbd2_journal_get_create_access(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} + +int __ext4_journal_dirty_metadata(const char *where, + handle_t *handle, struct buffer_head *bh) +{ + int err = jbd2_journal_dirty_metadata(handle, bh); + if (err) + ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err); + return err; +} diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 2608dce..dc2724f 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -48,7 +48,7 @@ * ext_pblock: * combine low and high parts of physical block number into ext4_fsblk_t */ -static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) +static ext4_fsblk_t ext_pblock(struct ext4_extent *ex) { ext4_fsblk_t block; @@ -61,7 +61,7 @@ static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) * idx_pblock: * combine low and high parts of a leaf physical block number into ext4_fsblk_t */ -static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) +static ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) { ext4_fsblk_t block; @@ -75,7 +75,7 @@ static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) * stores a large physical block number into an extent struct, * breaking it into parts */ -static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) +static void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) { ex->ee_start = cpu_to_le32((unsigned long) (pb & 0xffffffff)); ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); @@ -86,7 +86,7 @@ static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb * stores a large physical block number into an index struct, * breaking it into parts */ -static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) +static void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) { ix->ei_leaf = cpu_to_le32((unsigned long) (pb & 0xffffffff)); ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); @@ -186,7 +186,8 @@ static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, depth = path->p_depth; /* try to predict block placement */ - if ((ex = path[depth].p_ext)) + ex = path[depth].p_ext; + if (ex) return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block)); /* it looks like index is empty; @@ -215,7 +216,7 @@ ext4_ext_new_block(handle_t *handle, struct inode *inode, return newblock; } -static inline int ext4_ext_space_block(struct inode *inode) +static int ext4_ext_space_block(struct inode *inode) { int size; @@ -228,7 +229,7 @@ static inline int ext4_ext_space_block(struct inode *inode) return size; } -static inline int ext4_ext_space_block_idx(struct inode *inode) +static int ext4_ext_space_block_idx(struct inode *inode) { int size; @@ -241,7 +242,7 @@ static inline int ext4_ext_space_block_idx(struct inode *inode) return size; } -static inline int ext4_ext_space_root(struct inode *inode) +static int ext4_ext_space_root(struct inode *inode) { int size; @@ -255,7 +256,7 @@ static inline int ext4_ext_space_root(struct inode *inode) return size; } -static inline int ext4_ext_space_root_idx(struct inode *inode) +static int ext4_ext_space_root_idx(struct inode *inode) { int size; @@ -476,13 +477,12 @@ ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path) /* account possible depth increase */ if (!path) { - path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 2), + path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 2), GFP_NOFS); if (!path) return ERR_PTR(-ENOMEM); alloc = 1; } - memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); path[0].p_hdr = eh; /* walk through the tree */ @@ -543,7 +543,8 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, struct ext4_extent_idx *ix; int len, err; - if ((err = ext4_ext_get_access(handle, inode, curp))) + err = ext4_ext_get_access(handle, inode, curp); + if (err) return err; BUG_ON(logical == le32_to_cpu(curp->p_idx->ei_block)); @@ -641,10 +642,9 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, * We need this to handle errors and free blocks * upon them. */ - ablocks = kmalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); + ablocks = kzalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); if (!ablocks) return -ENOMEM; - memset(ablocks, 0, sizeof(ext4_fsblk_t) * depth); /* allocate all needed blocks */ ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); @@ -665,7 +665,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, } lock_buffer(bh); - if ((err = ext4_journal_get_create_access(handle, bh))) + err = ext4_journal_get_create_access(handle, bh); + if (err) goto cleanup; neh = ext_block_hdr(bh); @@ -702,18 +703,21 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, set_buffer_uptodate(bh); unlock_buffer(bh); - if ((err = ext4_journal_dirty_metadata(handle, bh))) + err = ext4_journal_dirty_metadata(handle, bh); + if (err) goto cleanup; brelse(bh); bh = NULL; /* correct old leaf */ if (m) { - if ((err = ext4_ext_get_access(handle, inode, path + depth))) + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) goto cleanup; path[depth].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m); - if ((err = ext4_ext_dirty(handle, inode, path + depth))) + err = ext4_ext_dirty(handle, inode, path + depth); + if (err) goto cleanup; } @@ -736,7 +740,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, } lock_buffer(bh); - if ((err = ext4_journal_get_create_access(handle, bh))) + err = ext4_journal_get_create_access(handle, bh); + if (err) goto cleanup; neh = ext_block_hdr(bh); @@ -780,7 +785,8 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, set_buffer_uptodate(bh); unlock_buffer(bh); - if ((err = ext4_journal_dirty_metadata(handle, bh))) + err = ext4_journal_dirty_metadata(handle, bh); + if (err) goto cleanup; brelse(bh); bh = NULL; @@ -800,9 +806,6 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode, } /* insert new index */ - if (err) - goto cleanup; - err = ext4_ext_insert_index(handle, inode, path + at, le32_to_cpu(border), newblock); @@ -857,7 +860,8 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, } lock_buffer(bh); - if ((err = ext4_journal_get_create_access(handle, bh))) { + err = ext4_journal_get_create_access(handle, bh); + if (err) { unlock_buffer(bh); goto out; } @@ -877,11 +881,13 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, set_buffer_uptodate(bh); unlock_buffer(bh); - if ((err = ext4_journal_dirty_metadata(handle, bh))) + err = ext4_journal_dirty_metadata(handle, bh); + if (err) goto out; /* create index in new top-level index: num,max,pointer */ - if ((err = ext4_ext_get_access(handle, inode, curp))) + err = ext4_ext_get_access(handle, inode, curp); + if (err) goto out; curp->p_hdr->eh_magic = EXT4_EXT_MAGIC; @@ -1073,27 +1079,31 @@ int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, */ k = depth - 1; border = path[depth].p_ext->ee_block; - if ((err = ext4_ext_get_access(handle, inode, path + k))) + err = ext4_ext_get_access(handle, inode, path + k); + if (err) return err; path[k].p_idx->ei_block = border; - if ((err = ext4_ext_dirty(handle, inode, path + k))) + err = ext4_ext_dirty(handle, inode, path + k); + if (err) return err; while (k--) { /* change all left-side indexes */ if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr)) break; - if ((err = ext4_ext_get_access(handle, inode, path + k))) + err = ext4_ext_get_access(handle, inode, path + k); + if (err) break; path[k].p_idx->ei_block = border; - if ((err = ext4_ext_dirty(handle, inode, path + k))) + err = ext4_ext_dirty(handle, inode, path + k); + if (err) break; } return err; } -static int inline +static int ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, struct ext4_extent *ex2) { @@ -1145,7 +1155,8 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, le16_to_cpu(newext->ee_len), le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len), ext_pblock(ex)); - if ((err = ext4_ext_get_access(handle, inode, path + depth))) + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) return err; ex->ee_len = cpu_to_le16(le16_to_cpu(ex->ee_len) + le16_to_cpu(newext->ee_len)); @@ -1195,7 +1206,8 @@ repeat: has_space: nearex = path[depth].p_ext; - if ((err = ext4_ext_get_access(handle, inode, path + depth))) + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) goto cleanup; if (!nearex) { @@ -1383,7 +1395,7 @@ int ext4_ext_walk_space(struct inode *inode, unsigned long block, return err; } -static inline void +static void ext4_ext_put_in_cache(struct inode *inode, __u32 block, __u32 len, __u32 start, int type) { @@ -1401,7 +1413,7 @@ ext4_ext_put_in_cache(struct inode *inode, __u32 block, * calculate boundaries of the gap that the requested block fits into * and cache this gap */ -static inline void +static void ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, unsigned long block) { @@ -1442,7 +1454,7 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP); } -static inline int +static int ext4_ext_in_cache(struct inode *inode, unsigned long block, struct ext4_extent *ex) { @@ -1489,10 +1501,12 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, path--; leaf = idx_pblock(path->p_idx); BUG_ON(path->p_hdr->eh_entries == 0); - if ((err = ext4_ext_get_access(handle, inode, path))) + err = ext4_ext_get_access(handle, inode, path); + if (err) return err; path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); - if ((err = ext4_ext_dirty(handle, inode, path))) + err = ext4_ext_dirty(handle, inode, path); + if (err) return err; ext_debug("index is empty, remove it, free block %llu\n", leaf); bh = sb_find_get_block(inode->i_sb, leaf); @@ -1509,7 +1523,7 @@ int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, * the caller should calculate credits under truncate_mutex and * pass the actual path. */ -int inline ext4_ext_calc_credits_for_insert(struct inode *inode, +int ext4_ext_calc_credits_for_insert(struct inode *inode, struct ext4_ext_path *path) { int depth, needed; @@ -1534,16 +1548,17 @@ int inline ext4_ext_calc_credits_for_insert(struct inode *inode, /* * tree can be full, so it would need to grow in depth: - * allocation + old root + new root + * we need one credit to modify old root, credits for + * new root will be added in split accounting */ - needed += 2 + 1 + 1; + needed += 1; /* * Index split can happen, we would need: * allocate intermediate indexes (bitmap + group) * + change two blocks at each level, but root (already included) */ - needed = (depth * 2) + (depth * 2); + needed += (depth * 2) + (depth * 2); /* any allocation modifies superblock */ needed += 1; @@ -1718,7 +1733,7 @@ out: * ext4_ext_more_to_rm: * returns 1 if current index has to be freed (even partial) */ -static int inline +static int ext4_ext_more_to_rm(struct ext4_ext_path *path) { BUG_ON(path->p_idx == NULL); @@ -1756,12 +1771,11 @@ int ext4_ext_remove_space(struct inode *inode, unsigned long start) * We start scanning from right side, freeing all the blocks * after i_size and walking into the tree depth-wise. */ - path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); + path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); if (path == NULL) { ext4_journal_stop(handle); return -ENOMEM; } - memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); path[0].p_hdr = ext_inode_hdr(inode); if (ext4_ext_check_header(__FUNCTION__, inode, path[0].p_hdr)) { err = -EIO; @@ -1932,7 +1946,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, mutex_lock(&EXT4_I(inode)->truncate_mutex); /* check in cache */ - if ((goal = ext4_ext_in_cache(inode, iblock, &newex))) { + goal = ext4_ext_in_cache(inode, iblock, &newex); + if (goal) { if (goal == EXT4_EXT_CACHE_GAP) { if (!create) { /* block isn't allocated yet and @@ -1971,7 +1986,8 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, */ BUG_ON(path[depth].p_ext == NULL && depth != 0); - if ((ex = path[depth].p_ext)) { + ex = path[depth].p_ext; + if (ex) { unsigned long ee_block = le32_to_cpu(ex->ee_block); ext4_fsblk_t ee_start = ext_pblock(ex); unsigned short ee_len = le16_to_cpu(ex->ee_len); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 0a60ec5..1d85d4e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1147,37 +1147,102 @@ static int do_journal_get_write_access(handle_t *handle, return ext4_journal_get_write_access(handle, bh); } +/* + * The idea of this helper function is following: + * if prepare_write has allocated some blocks, but not all of them, the + * transaction must include the content of the newly allocated blocks. + * This content is expected to be set to zeroes by block_prepare_write(). + * 2006/10/14 SAW + */ +static int ext4_prepare_failure(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + struct address_space *mapping; + struct buffer_head *bh, *head, *next; + unsigned block_start, block_end; + unsigned blocksize; + int ret; + handle_t *handle = ext4_journal_current_handle(); + + mapping = page->mapping; + if (ext4_should_writeback_data(mapping->host)) { + /* optimization: no constraints about data */ +skip: + return ext4_journal_stop(handle); + } + + head = page_buffers(page); + blocksize = head->b_size; + for ( bh = head, block_start = 0; + bh != head || !block_start; + block_start = block_end, bh = next) + { + next = bh->b_this_page; + block_end = block_start + blocksize; + if (block_end <= from) + continue; + if (block_start >= to) { + block_start = to; + break; + } + if (!buffer_mapped(bh)) + /* prepare_write failed on this bh */ + break; + if (ext4_should_journal_data(mapping->host)) { + ret = do_journal_get_write_access(handle, bh); + if (ret) { + ext4_journal_stop(handle); + return ret; + } + } + /* + * block_start here becomes the first block where the current iteration + * of prepare_write failed. + */ + } + if (block_start <= from) + goto skip; + + /* commit allocated and zeroed buffers */ + return mapping->a_ops->commit_write(file, page, from, block_start); +} + static int ext4_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { struct inode *inode = page->mapping->host; - int ret, needed_blocks = ext4_writepage_trans_blocks(inode); + int ret, ret2; + int needed_blocks = ext4_writepage_trans_blocks(inode); handle_t *handle; int retries = 0; retry: handle = ext4_journal_start(inode, needed_blocks); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - goto out; - } + if (IS_ERR(handle)) + return PTR_ERR(handle); if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) ret = nobh_prepare_write(page, from, to, ext4_get_block); else ret = block_prepare_write(page, from, to, ext4_get_block); if (ret) - goto prepare_write_failed; + goto failure; if (ext4_should_journal_data(inode)) { ret = walk_page_buffers(handle, page_buffers(page), from, to, NULL, do_journal_get_write_access); + if (ret) + /* fatal error, just put the handle and return */ + journal_stop(handle); } -prepare_write_failed: - if (ret) - ext4_journal_stop(handle); + return ret; + +failure: + ret2 = ext4_prepare_failure(file, page, from, to); + if (ret2 < 0) + return ret2; if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; -out: + /* retry number exceeded, or other error like -EDQUOT */ return ret; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 8b1bd03..859990e 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -552,6 +552,15 @@ static int htree_dirblock_to_tree(struct file *dir_file, dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0)); for (; de < top; de = ext4_next_entry(de)) { + if (!ext4_check_dir_entry("htree_dirblock_to_tree", dir, de, bh, + (block<<EXT4_BLOCK_SIZE_BITS(dir->i_sb)) + +((char *)de - bh->b_data))) { + /* On error, skip the f_pos to the next block. */ + dir_file->f_pos = (dir_file->f_pos | + (dir->i_sb->s_blocksize - 1)) + 1; + brelse (bh); + return count; + } ext4fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || ((hinfo->hash == start_hash) && diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b4b022a..486a641 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -486,7 +486,7 @@ static void ext4_put_super (struct super_block * sb) return; } -static kmem_cache_t *ext4_inode_cachep; +static struct kmem_cache *ext4_inode_cachep; /* * Called inside transaction, so use GFP_NOFS @@ -495,7 +495,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) { struct ext4_inode_info *ei; - ei = kmem_cache_alloc(ext4_inode_cachep, SLAB_NOFS); + ei = kmem_cache_alloc(ext4_inode_cachep, GFP_NOFS); if (!ei) return NULL; #ifdef CONFIG_EXT4DEV_FS_POSIX_ACL @@ -513,7 +513,7 @@ static void ext4_destroy_inode(struct inode *inode) kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; @@ -1321,6 +1321,12 @@ static void ext4_orphan_cleanup (struct super_block * sb, return; } + if (bdev_read_only(sb->s_bdev)) { + printk(KERN_ERR "EXT4-fs: write access " + "unavailable, skipping orphan cleanup.\n"); + return; + } + if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) { if (es->s_last_orphan) jbd_debug(1, "Errors on filesystem, " @@ -2460,6 +2466,7 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) struct ext4_super_block *es = sbi->s_es; ext4_fsblk_t overhead; int i; + u64 fsid; if (test_opt (sb, MINIX_DF)) overhead = 0; @@ -2506,6 +2513,10 @@ static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) buf->f_files = le32_to_cpu(es->s_inodes_count); buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); buf->f_namelen = EXT4_NAME_LEN; + fsid = le64_to_cpup((void *)es->s_uuid) ^ + le64_to_cpup((void *)es->s_uuid + sizeof(u64)); + buf->f_fsid.val[0] = fsid & 0xFFFFFFFFUL; + buf->f_fsid.val[1] = (fsid >> 32) & 0xFFFFFFFFUL; return 0; } diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 63233cd..dc969c3 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -459,14 +459,11 @@ static void ext4_xattr_update_super_block(handle_t *handle, if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_EXT_ATTR)) return; - lock_super(sb); if (ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh) == 0) { - EXT4_SB(sb)->s_es->s_feature_compat |= - cpu_to_le32(EXT4_FEATURE_COMPAT_EXT_ATTR); + EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_EXT_ATTR); sb->s_dirt = 1; ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); } - unlock_super(sb); } /* diff --git a/fs/fat/cache.c b/fs/fat/cache.c index 82cc4f59..05c2941 100644 --- a/fs/fat/cache.c +++ b/fs/fat/cache.c @@ -34,9 +34,9 @@ static inline int fat_max_cache(struct inode *inode) return FAT_MAX_CACHE; } -static kmem_cache_t *fat_cache_cachep; +static struct kmem_cache *fat_cache_cachep; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct fat_cache *cache = (struct fat_cache *)foo; @@ -63,7 +63,7 @@ void fat_cache_destroy(void) static inline struct fat_cache *fat_cache_alloc(struct inode *inode) { - return kmem_cache_alloc(fat_cache_cachep, SLAB_KERNEL); + return kmem_cache_alloc(fat_cache_cachep, GFP_KERNEL); } static inline void fat_cache_free(struct fat_cache *cache) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 78945b5..a9e4688 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -477,12 +477,12 @@ static void fat_put_super(struct super_block *sb) kfree(sbi); } -static kmem_cache_t *fat_inode_cachep; +static struct kmem_cache *fat_inode_cachep; static struct inode *fat_alloc_inode(struct super_block *sb) { struct msdos_inode_info *ei; - ei = kmem_cache_alloc(fat_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(fat_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -493,7 +493,7 @@ static void fat_destroy_inode(struct inode *inode) kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; @@ -553,7 +553,7 @@ int send_sigurg(struct fown_struct *fown) } static DEFINE_RWLOCK(fasync_lock); -static kmem_cache_t *fasync_cache __read_mostly; +static struct kmem_cache *fasync_cache __read_mostly; /* * fasync_helper() is used by some character device drivers (mainly mice) @@ -567,7 +567,7 @@ int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fap int result = 0; if (on) { - new = kmem_cache_alloc(fasync_cache, SLAB_KERNEL); + new = kmem_cache_alloc(fasync_cache, GFP_KERNEL); if (!new) return -ENOMEM; } @@ -21,7 +21,6 @@ struct fdtable_defer { spinlock_t lock; struct work_struct wq; - struct timer_list timer; struct fdtable *next; }; @@ -75,22 +74,6 @@ static void __free_fdtable(struct fdtable *fdt) kfree(fdt); } -static void fdtable_timer(unsigned long data) -{ - struct fdtable_defer *fddef = (struct fdtable_defer *)data; - - spin_lock(&fddef->lock); - /* - * If someone already emptied the queue return. - */ - if (!fddef->next) - goto out; - if (!schedule_work(&fddef->wq)) - mod_timer(&fddef->timer, 5); -out: - spin_unlock(&fddef->lock); -} - static void free_fdtable_work(struct work_struct *work) { struct fdtable_defer *f = @@ -144,13 +127,8 @@ static void free_fdtable_rcu(struct rcu_head *rcu) spin_lock(&fddef->lock); fdt->next = fddef->next; fddef->next = fdt; - /* - * vmallocs are handled from the workqueue context. - * If the per-cpu workqueue is running, then we - * defer work scheduling through a timer. - */ - if (!schedule_work(&fddef->wq)) - mod_timer(&fddef->timer, 5); + /* vmallocs are handled from the workqueue context */ + schedule_work(&fddef->wq); spin_unlock(&fddef->lock); put_cpu_var(fdtable_defer_list); } @@ -354,9 +332,6 @@ static void __devinit fdtable_defer_list_init(int cpu) struct fdtable_defer *fddef = &per_cpu(fdtable_defer_list, cpu); spin_lock_init(&fddef->lock); INIT_WORK(&fddef->wq, free_fdtable_work); - init_timer(&fddef->timer); - fddef->timer.data = (unsigned long)fddef; - fddef->timer.function = fdtable_timer; fddef->next = NULL; } diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 4786d51..0b7ae89 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -46,7 +46,7 @@ extern const struct address_space_operations vxfs_immed_aops; extern struct inode_operations vxfs_immed_symlink_iops; -kmem_cache_t *vxfs_inode_cachep; +struct kmem_cache *vxfs_inode_cachep; #ifdef DIAGNOSTIC @@ -103,7 +103,7 @@ vxfs_blkiget(struct super_block *sbp, u_long extent, ino_t ino) struct vxfs_inode_info *vip; struct vxfs_dinode *dip; - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(bp->b_data + offset); memcpy(vip, dip, sizeof(*vip)); @@ -145,7 +145,7 @@ __vxfs_iget(ino_t ino, struct inode *ilistp) struct vxfs_dinode *dip; caddr_t kaddr = (char *)page_address(pp); - if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, SLAB_KERNEL))) + if (!(vip = kmem_cache_alloc(vxfs_inode_cachep, GFP_KERNEL))) goto fail; dip = (struct vxfs_dinode *)(kaddr + offset); memcpy(vip, dip, sizeof(*vip)); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 66571ea..357764d 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -19,7 +19,7 @@ MODULE_ALIAS_MISCDEV(FUSE_MINOR); -static kmem_cache_t *fuse_req_cachep; +static struct kmem_cache *fuse_req_cachep; static struct fuse_conn *fuse_get_conn(struct file *file) { @@ -41,7 +41,7 @@ static void fuse_request_init(struct fuse_req *req) struct fuse_req *fuse_request_alloc(void) { - struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, SLAB_KERNEL); + struct fuse_req *req = kmem_cache_alloc(fuse_req_cachep, GFP_KERNEL); if (req) fuse_request_init(req); return req; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index c71a6c0..1cabdb2 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -141,9 +141,6 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) struct fuse_req *forget_req; struct dentry *parent; - /* Doesn't hurt to "reset" the validity timeout */ - fuse_invalidate_entry_cache(entry); - /* For negative dentries, always do a fresh lookup */ if (!inode) return 0; @@ -1027,6 +1024,8 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) if (attr->ia_valid & ATTR_SIZE) { unsigned long limit; is_truncate = 1; + if (IS_SWAPFILE(inode)) + return -ETXTBSY; limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur; if (limit != RLIM_INFINITY && attr->ia_size > (loff_t) limit) { send_sig(SIGXFSZ, current, 0); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 763a50d..128f79c 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -754,6 +754,42 @@ static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl) return err; } +static sector_t fuse_bmap(struct address_space *mapping, sector_t block) +{ + struct inode *inode = mapping->host; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_req *req; + struct fuse_bmap_in inarg; + struct fuse_bmap_out outarg; + int err; + + if (!inode->i_sb->s_bdev || fc->no_bmap) + return 0; + + req = fuse_get_req(fc); + if (IS_ERR(req)) + return 0; + + memset(&inarg, 0, sizeof(inarg)); + inarg.block = block; + inarg.blocksize = inode->i_sb->s_blocksize; + req->in.h.opcode = FUSE_BMAP; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + req->out.numargs = 1; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + request_send(fc, req); + err = req->out.h.error; + fuse_put_request(fc, req); + if (err == -ENOSYS) + fc->no_bmap = 1; + + return err ? 0 : outarg.block; +} + static const struct file_operations fuse_file_operations = { .llseek = generic_file_llseek, .read = do_sync_read, @@ -787,6 +823,7 @@ static const struct address_space_operations fuse_file_aops = { .commit_write = fuse_commit_write, .readpages = fuse_readpages, .set_page_dirty = fuse_set_page_dirty, + .bmap = fuse_bmap, }; void fuse_init_file_inode(struct inode *inode) diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 91edb89..b98b20d 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -298,6 +298,9 @@ struct fuse_conn { reply, before any other request, and never cleared */ unsigned conn_error : 1; + /** Connection successful. Only set in INIT */ + unsigned conn_init : 1; + /** Do readpages asynchronously? Only set in INIT */ unsigned async_read : 1; @@ -339,6 +342,9 @@ struct fuse_conn { /** Is interrupt not implemented by fs? */ unsigned no_interrupt : 1; + /** Is bmap not implemented by fs? */ + unsigned no_bmap : 1; + /** The number of requests waiting for completion */ atomic_t num_waiting; @@ -365,6 +371,9 @@ struct fuse_conn { /** Key for lock owner ID scrambling */ u32 scramble_key[4]; + + /** Reserved request for the DESTROY message */ + struct fuse_req *destroy_req; }; static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb) diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index fc42035..12450d2 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -22,7 +22,7 @@ MODULE_AUTHOR("Miklos Szeredi <miklos@szeredi.hu>"); MODULE_DESCRIPTION("Filesystem in Userspace"); MODULE_LICENSE("GPL"); -static kmem_cache_t *fuse_inode_cachep; +static struct kmem_cache *fuse_inode_cachep; struct list_head fuse_conn_list; DEFINE_MUTEX(fuse_mutex); @@ -39,6 +39,7 @@ struct fuse_mount_data { unsigned group_id_present : 1; unsigned flags; unsigned max_read; + unsigned blksize; }; static struct inode *fuse_alloc_inode(struct super_block *sb) @@ -46,7 +47,7 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) struct inode *inode; struct fuse_inode *fi; - inode = kmem_cache_alloc(fuse_inode_cachep, SLAB_KERNEL); + inode = kmem_cache_alloc(fuse_inode_cachep, GFP_KERNEL); if (!inode) return NULL; @@ -205,10 +206,23 @@ static void fuse_umount_begin(struct vfsmount *vfsmnt, int flags) fuse_abort_conn(get_fuse_conn_super(vfsmnt->mnt_sb)); } +static void fuse_send_destroy(struct fuse_conn *fc) +{ + struct fuse_req *req = fc->destroy_req; + if (req && fc->conn_init) { + fc->destroy_req = NULL; + req->in.h.opcode = FUSE_DESTROY; + req->force = 1; + request_send(fc, req); + fuse_put_request(fc, req); + } +} + static void fuse_put_super(struct super_block *sb) { struct fuse_conn *fc = get_fuse_conn_super(sb); + fuse_send_destroy(fc); spin_lock(&fc->lock); fc->connected = 0; fc->blocked = 0; @@ -274,6 +288,7 @@ enum { OPT_DEFAULT_PERMISSIONS, OPT_ALLOW_OTHER, OPT_MAX_READ, + OPT_BLKSIZE, OPT_ERR }; @@ -285,14 +300,16 @@ static match_table_t tokens = { {OPT_DEFAULT_PERMISSIONS, "default_permissions"}, {OPT_ALLOW_OTHER, "allow_other"}, {OPT_MAX_READ, "max_read=%u"}, + {OPT_BLKSIZE, "blksize=%u"}, {OPT_ERR, NULL} }; -static int parse_fuse_opt(char *opt, struct fuse_mount_data *d) +static int parse_fuse_opt(char *opt, struct fuse_mount_data *d, int is_bdev) { char *p; memset(d, 0, sizeof(struct fuse_mount_data)); d->max_read = ~0; + d->blksize = 512; while ((p = strsep(&opt, ",")) != NULL) { int token; @@ -345,6 +362,12 @@ static int parse_fuse_opt(char *opt, struct fuse_mount_data *d) d->max_read = value; break; + case OPT_BLKSIZE: + if (!is_bdev || match_int(&args[0], &value)) + return 0; + d->blksize = value; + break; + default: return 0; } @@ -400,6 +423,8 @@ static struct fuse_conn *new_conn(void) void fuse_conn_put(struct fuse_conn *fc) { if (atomic_dec_and_test(&fc->count)) { + if (fc->destroy_req) + fuse_request_free(fc->destroy_req); mutex_destroy(&fc->inst_mutex); kfree(fc); } @@ -456,6 +481,7 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->bdi.ra_pages = min(fc->bdi.ra_pages, ra_pages); fc->minor = arg->minor; fc->max_write = arg->minor < 5 ? 4096 : arg->max_write; + fc->conn_init = 1; } fuse_put_request(fc, req); fc->blocked = 0; @@ -500,15 +526,23 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) struct dentry *root_dentry; struct fuse_req *init_req; int err; + int is_bdev = sb->s_bdev != NULL; if (sb->s_flags & MS_MANDLOCK) return -EINVAL; - if (!parse_fuse_opt((char *) data, &d)) + if (!parse_fuse_opt((char *) data, &d, is_bdev)) return -EINVAL; - sb->s_blocksize = PAGE_CACHE_SIZE; - sb->s_blocksize_bits = PAGE_CACHE_SHIFT; + if (is_bdev) { +#ifdef CONFIG_BLOCK + if (!sb_set_blocksize(sb, d.blksize)) + return -EINVAL; +#endif + } else { + sb->s_blocksize = PAGE_CACHE_SIZE; + sb->s_blocksize_bits = PAGE_CACHE_SHIFT; + } sb->s_magic = FUSE_SUPER_MAGIC; sb->s_op = &fuse_super_operations; sb->s_maxbytes = MAX_LFS_FILESIZE; @@ -547,6 +581,12 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (!init_req) goto err_put_root; + if (is_bdev) { + fc->destroy_req = fuse_request_alloc(); + if (!fc->destroy_req) + goto err_put_root; + } + mutex_lock(&fuse_mutex); err = -EINVAL; if (file->private_data) @@ -598,10 +638,47 @@ static struct file_system_type fuse_fs_type = { .kill_sb = kill_anon_super, }; +#ifdef CONFIG_BLOCK +static int fuse_get_sb_blk(struct file_system_type *fs_type, + int flags, const char *dev_name, + void *raw_data, struct vfsmount *mnt) +{ + return get_sb_bdev(fs_type, flags, dev_name, raw_data, fuse_fill_super, + mnt); +} + +static struct file_system_type fuseblk_fs_type = { + .owner = THIS_MODULE, + .name = "fuseblk", + .get_sb = fuse_get_sb_blk, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +static inline int register_fuseblk(void) +{ + return register_filesystem(&fuseblk_fs_type); +} + +static inline void unregister_fuseblk(void) +{ + unregister_filesystem(&fuseblk_fs_type); +} +#else +static inline int register_fuseblk(void) +{ + return 0; +} + +static inline void unregister_fuseblk(void) +{ +} +#endif + static decl_subsys(fuse, NULL, NULL); static decl_subsys(connections, NULL, NULL); -static void fuse_inode_init_once(void *foo, kmem_cache_t *cachep, +static void fuse_inode_init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct inode * inode = foo; @@ -617,24 +694,34 @@ static int __init fuse_fs_init(void) err = register_filesystem(&fuse_fs_type); if (err) - printk("fuse: failed to register filesystem\n"); - else { - fuse_inode_cachep = kmem_cache_create("fuse_inode", - sizeof(struct fuse_inode), - 0, SLAB_HWCACHE_ALIGN, - fuse_inode_init_once, NULL); - if (!fuse_inode_cachep) { - unregister_filesystem(&fuse_fs_type); - err = -ENOMEM; - } - } + goto out; + err = register_fuseblk(); + if (err) + goto out_unreg; + + fuse_inode_cachep = kmem_cache_create("fuse_inode", + sizeof(struct fuse_inode), + 0, SLAB_HWCACHE_ALIGN, + fuse_inode_init_once, NULL); + err = -ENOMEM; + if (!fuse_inode_cachep) + goto out_unreg2; + + return 0; + + out_unreg2: + unregister_fuseblk(); + out_unreg: + unregister_filesystem(&fuse_fs_type); + out: return err; } static void fuse_fs_cleanup(void) { unregister_filesystem(&fuse_fs_type); + unregister_fuseblk(); kmem_cache_destroy(fuse_inode_cachep); } diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig index 8c27de8..c0791cb 100644 --- a/fs/gfs2/Kconfig +++ b/fs/gfs2/Kconfig @@ -2,6 +2,7 @@ config GFS2_FS tristate "GFS2 file system support" depends on EXPERIMENTAL select FS_POSIX_ACL + select CRC32 help A cluster filesystem. diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 5f959b8..6e80844 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -74,11 +74,11 @@ int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access) { if (!GFS2_SB(&ip->i_inode)->sd_args.ar_posix_acl) return -EOPNOTSUPP; - if (current->fsuid != ip->i_di.di_uid && !capable(CAP_FOWNER)) + if (current->fsuid != ip->i_inode.i_uid && !capable(CAP_FOWNER)) return -EPERM; - if (S_ISLNK(ip->i_di.di_mode)) + if (S_ISLNK(ip->i_inode.i_mode)) return -EOPNOTSUPP; - if (!access && !S_ISDIR(ip->i_di.di_mode)) + if (!access && !S_ISDIR(ip->i_inode.i_mode)) return -EACCES; return 0; @@ -145,14 +145,14 @@ out: } /** - * gfs2_check_acl_locked - Check an ACL to see if we're allowed to do something + * gfs2_check_acl - Check an ACL to see if we're allowed to do something * @inode: the file we want to do something to * @mask: what we want to do * * Returns: errno */ -int gfs2_check_acl_locked(struct inode *inode, int mask) +int gfs2_check_acl(struct inode *inode, int mask) { struct posix_acl *acl = NULL; int error; @@ -170,21 +170,6 @@ int gfs2_check_acl_locked(struct inode *inode, int mask) return -EAGAIN; } -int gfs2_check_acl(struct inode *inode, int mask) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder i_gh; - int error; - - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); - if (!error) { - error = gfs2_check_acl_locked(inode, mask); - gfs2_glock_dq_uninit(&i_gh); - } - - return error; -} - static int munge_mode(struct gfs2_inode *ip, mode_t mode) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); @@ -198,10 +183,10 @@ static int munge_mode(struct gfs2_inode *ip, mode_t mode) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_assert_withdraw(sdp, - (ip->i_di.di_mode & S_IFMT) == (mode & S_IFMT)); - ip->i_di.di_mode = mode; + (ip->i_inode.i_mode & S_IFMT) == (mode & S_IFMT)); + ip->i_inode.i_mode = mode; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -215,12 +200,12 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct posix_acl *acl = NULL, *clone; struct gfs2_ea_request er; - mode_t mode = ip->i_di.di_mode; + mode_t mode = ip->i_inode.i_mode; int error; if (!sdp->sd_args.ar_posix_acl) return 0; - if (S_ISLNK(ip->i_di.di_mode)) + if (S_ISLNK(ip->i_inode.i_mode)) return 0; memset(&er, 0, sizeof(struct gfs2_ea_request)); @@ -232,7 +217,7 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) return error; if (!acl) { mode &= ~current->fs->umask; - if (mode != ip->i_di.di_mode) + if (mode != ip->i_inode.i_mode) error = munge_mode(ip, mode); return error; } @@ -244,7 +229,7 @@ int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip) posix_acl_release(acl); acl = clone; - if (S_ISDIR(ip->i_di.di_mode)) { + if (S_ISDIR(ip->i_inode.i_mode)) { er.er_name = GFS2_POSIX_ACL_DEFAULT; er.er_name_len = GFS2_POSIX_ACL_DEFAULT_LEN; error = gfs2_system_eaops.eo_set(ip, &er); diff --git a/fs/gfs2/acl.h b/fs/gfs2/acl.h index 05c294f..6751930 100644 --- a/fs/gfs2/acl.h +++ b/fs/gfs2/acl.h @@ -31,7 +31,6 @@ int gfs2_acl_validate_set(struct gfs2_inode *ip, int access, struct gfs2_ea_request *er, int *remove, mode_t *mode); int gfs2_acl_validate_remove(struct gfs2_inode *ip, int access); -int gfs2_check_acl_locked(struct inode *inode, int mask); int gfs2_check_acl(struct inode *inode, int mask); int gfs2_acl_create(struct gfs2_inode *dip, struct gfs2_inode *ip); int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr); diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 06e9a8c..8240c1f 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -38,8 +38,8 @@ struct metapath { }; typedef int (*block_call_t) (struct gfs2_inode *ip, struct buffer_head *dibh, - struct buffer_head *bh, u64 *top, - u64 *bottom, unsigned int height, + struct buffer_head *bh, __be64 *top, + __be64 *bottom, unsigned int height, void *data); struct strip_mine { @@ -163,6 +163,7 @@ int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page) if (ip->i_di.di_size) { *(__be64 *)(di + 1) = cpu_to_be64(block); ip->i_di.di_blocks++; + gfs2_set_inode_blocks(&ip->i_inode); di->di_blocks = cpu_to_be64(ip->i_di.di_blocks); } @@ -230,7 +231,7 @@ static int build_height(struct inode *inode, unsigned height) struct buffer_head *blocks[GFS2_MAX_META_HEIGHT]; struct gfs2_dinode *di; int error; - u64 *bp; + __be64 *bp; u64 bn; unsigned n; @@ -255,7 +256,7 @@ static int build_height(struct inode *inode, unsigned height) GFS2_FORMAT_IN); gfs2_buffer_clear_tail(blocks[n], sizeof(struct gfs2_meta_header)); - bp = (u64 *)(blocks[n]->b_data + + bp = (__be64 *)(blocks[n]->b_data + sizeof(struct gfs2_meta_header)); *bp = cpu_to_be64(blocks[n+1]->b_blocknr); brelse(blocks[n]); @@ -272,6 +273,7 @@ static int build_height(struct inode *inode, unsigned height) *(__be64 *)(di + 1) = cpu_to_be64(bn); ip->i_di.di_height += new_height; ip->i_di.di_blocks += new_height; + gfs2_set_inode_blocks(&ip->i_inode); di->di_height = cpu_to_be16(ip->i_di.di_height); di->di_blocks = cpu_to_be64(ip->i_di.di_blocks); brelse(dibh); @@ -360,15 +362,15 @@ static void find_metapath(struct gfs2_inode *ip, u64 block, * metadata tree. */ -static inline u64 *metapointer(struct buffer_head *bh, int *boundary, +static inline __be64 *metapointer(struct buffer_head *bh, int *boundary, unsigned int height, const struct metapath *mp) { unsigned int head_size = (height > 0) ? sizeof(struct gfs2_meta_header) : sizeof(struct gfs2_dinode); - u64 *ptr; + __be64 *ptr; *boundary = 0; - ptr = ((u64 *)(bh->b_data + head_size)) + mp->mp_list[height]; - if (ptr + 1 == (u64 *)(bh->b_data + bh->b_size)) + ptr = ((__be64 *)(bh->b_data + head_size)) + mp->mp_list[height]; + if (ptr + 1 == (__be64 *)(bh->b_data + bh->b_size)) *boundary = 1; return ptr; } @@ -394,7 +396,7 @@ static int lookup_block(struct gfs2_inode *ip, struct buffer_head *bh, int *new, u64 *block) { int boundary; - u64 *ptr = metapointer(bh, &boundary, height, mp); + __be64 *ptr = metapointer(bh, &boundary, height, mp); if (*ptr) { *block = be64_to_cpu(*ptr); @@ -415,17 +417,35 @@ static int lookup_block(struct gfs2_inode *ip, struct buffer_head *bh, *ptr = cpu_to_be64(*block); ip->i_di.di_blocks++; + gfs2_set_inode_blocks(&ip->i_inode); *new = 1; return 0; } +static inline void bmap_lock(struct inode *inode, int create) +{ + struct gfs2_inode *ip = GFS2_I(inode); + if (create) + down_write(&ip->i_rw_mutex); + else + down_read(&ip->i_rw_mutex); +} + +static inline void bmap_unlock(struct inode *inode, int create) +{ + struct gfs2_inode *ip = GFS2_I(inode); + if (create) + up_write(&ip->i_rw_mutex); + else + up_read(&ip->i_rw_mutex); +} + /** - * gfs2_block_pointers - Map a block from an inode to a disk block + * gfs2_block_map - Map a block from an inode to a disk block * @inode: The inode * @lblock: The logical block number - * @map_bh: The bh to be mapped - * @mp: metapath to use + * @bh_map: The bh to be mapped * * Find the block number on the current device which corresponds to an * inode's block. If the block had to be created, "new" will be set. @@ -433,8 +453,8 @@ static int lookup_block(struct gfs2_inode *ip, struct buffer_head *bh, * Returns: errno */ -static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh_map, struct metapath *mp) +int gfs2_block_map(struct inode *inode, u64 lblock, int create, + struct buffer_head *bh_map) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -448,57 +468,61 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, u64 dblock = 0; int boundary; unsigned int maxlen = bh_map->b_size >> inode->i_blkbits; + struct metapath mp; + u64 size; BUG_ON(maxlen == 0); if (gfs2_assert_warn(sdp, !gfs2_is_stuffed(ip))) return 0; + bmap_lock(inode, create); + clear_buffer_mapped(bh_map); + clear_buffer_new(bh_map); + clear_buffer_boundary(bh_map); bsize = gfs2_is_dir(ip) ? sdp->sd_jbsize : sdp->sd_sb.sb_bsize; - - height = calc_tree_height(ip, (lblock + 1) * bsize); - if (ip->i_di.di_height < height) { - if (!create) - return 0; - - error = build_height(inode, height); - if (error) - return error; + size = (lblock + 1) * bsize; + + if (size > ip->i_di.di_size) { + height = calc_tree_height(ip, size); + if (ip->i_di.di_height < height) { + if (!create) + goto out_ok; + + error = build_height(inode, height); + if (error) + goto out_fail; + } } - find_metapath(ip, lblock, mp); + find_metapath(ip, lblock, &mp); end_of_metadata = ip->i_di.di_height - 1; - error = gfs2_meta_inode_buffer(ip, &bh); if (error) - return error; + goto out_fail; for (x = 0; x < end_of_metadata; x++) { - lookup_block(ip, bh, x, mp, create, &new, &dblock); + lookup_block(ip, bh, x, &mp, create, &new, &dblock); brelse(bh); if (!dblock) - return 0; + goto out_ok; error = gfs2_meta_indirect_buffer(ip, x+1, dblock, new, &bh); if (error) - return error; + goto out_fail; } - boundary = lookup_block(ip, bh, end_of_metadata, mp, create, &new, &dblock); - clear_buffer_mapped(bh_map); - clear_buffer_new(bh_map); - clear_buffer_boundary(bh_map); - + boundary = lookup_block(ip, bh, end_of_metadata, &mp, create, &new, &dblock); if (dblock) { map_bh(bh_map, inode->i_sb, dblock); if (boundary) - set_buffer_boundary(bh); + set_buffer_boundary(bh_map); if (new) { struct buffer_head *dibh; error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } set_buffer_new(bh_map); @@ -507,8 +531,8 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, while(--maxlen && !buffer_boundary(bh_map)) { u64 eblock; - mp->mp_list[end_of_metadata]++; - boundary = lookup_block(ip, bh, end_of_metadata, mp, 0, &new, &eblock); + mp.mp_list[end_of_metadata]++; + boundary = lookup_block(ip, bh, end_of_metadata, &mp, 0, &new, &eblock); if (eblock != ++dblock) break; bh_map->b_size += (1 << inode->i_blkbits); @@ -518,43 +542,15 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, } out_brelse: brelse(bh); - return 0; -} - - -static inline void bmap_lock(struct inode *inode, int create) -{ - struct gfs2_inode *ip = GFS2_I(inode); - if (create) - down_write(&ip->i_rw_mutex); - else - down_read(&ip->i_rw_mutex); -} - -static inline void bmap_unlock(struct inode *inode, int create) -{ - struct gfs2_inode *ip = GFS2_I(inode); - if (create) - up_write(&ip->i_rw_mutex); - else - up_read(&ip->i_rw_mutex); -} - -int gfs2_block_map(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh) -{ - struct metapath mp; - int ret; - - bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, bh, &mp); +out_ok: + error = 0; +out_fail: bmap_unlock(inode, create); - return ret; + return error; } int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen) { - struct metapath mp; struct buffer_head bh = { .b_state = 0, .b_blocknr = 0 }; int ret; int create = *new; @@ -564,9 +560,7 @@ int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsi BUG_ON(!new); bh.b_size = 1 << (inode->i_blkbits + 5); - bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, &bh, &mp); - bmap_unlock(inode, create); + ret = gfs2_block_map(inode, lblock, create, &bh); *extlen = bh.b_size >> inode->i_blkbits; *dblock = bh.b_blocknr; if (buffer_new(&bh)) @@ -600,7 +594,7 @@ static int recursive_scan(struct gfs2_inode *ip, struct buffer_head *dibh, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *bh = NULL; - u64 *top, *bottom; + __be64 *top, *bottom; u64 bn; int error; int mh_size = sizeof(struct gfs2_meta_header); @@ -611,17 +605,17 @@ static int recursive_scan(struct gfs2_inode *ip, struct buffer_head *dibh, return error; dibh = bh; - top = (u64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + mp->mp_list[0]; - bottom = (u64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + sdp->sd_diptrs; + top = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + mp->mp_list[0]; + bottom = (__be64 *)(bh->b_data + sizeof(struct gfs2_dinode)) + sdp->sd_diptrs; } else { error = gfs2_meta_indirect_buffer(ip, height, block, 0, &bh); if (error) return error; - top = (u64 *)(bh->b_data + mh_size) + + top = (__be64 *)(bh->b_data + mh_size) + (first ? mp->mp_list[height] : 0); - bottom = (u64 *)(bh->b_data + mh_size) + sdp->sd_inptrs; + bottom = (__be64 *)(bh->b_data + mh_size) + sdp->sd_inptrs; } error = bc(ip, dibh, bh, top, bottom, height, data); @@ -660,7 +654,7 @@ out: */ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, - struct buffer_head *bh, u64 *top, u64 *bottom, + struct buffer_head *bh, __be64 *top, __be64 *bottom, unsigned int height, void *data) { struct strip_mine *sm = data; @@ -668,7 +662,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, struct gfs2_rgrp_list rlist; u64 bn, bstart; u32 blen; - u64 *p; + __be64 *p; unsigned int rg_blocks = 0; int metadata; unsigned int revokes = 0; @@ -770,6 +764,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, if (!ip->i_di.di_blocks) gfs2_consist_inode(ip); ip->i_di.di_blocks--; + gfs2_set_inode_blocks(&ip->i_inode); } if (bstart) { if (metadata) @@ -778,9 +773,9 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, gfs2_free_data(ip, bstart, blen); } - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); up_write(&ip->i_rw_mutex); @@ -819,7 +814,7 @@ static int do_grow(struct gfs2_inode *ip, u64 size) if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -853,14 +848,14 @@ static int do_grow(struct gfs2_inode *ip, u64 size) } ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto out_end_trans; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out_end_trans: @@ -968,9 +963,9 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) if (gfs2_is_stuffed(ip)) { ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode) + size); error = 1; @@ -980,10 +975,10 @@ static int trunc_start(struct gfs2_inode *ip, u64 size) if (!error) { ip->i_di.di_size = size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); ip->i_di.di_flags |= GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); } } @@ -1053,11 +1048,11 @@ static int trunc_end(struct gfs2_inode *ip) ip->i_num.no_addr; gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode)); } - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); ip->i_di.di_flags &= ~GFS2_DIF_TRUNC_IN_PROG; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out: @@ -1109,7 +1104,7 @@ int gfs2_truncatei(struct gfs2_inode *ip, u64 size) { int error; - if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_di.di_mode))) + if (gfs2_assert_warn(GFS2_SB(&ip->i_inode), S_ISREG(ip->i_inode.i_mode))) return -EINVAL; if (size > ip->i_di.di_size) diff --git a/fs/gfs2/daemon.c b/fs/gfs2/daemon.c index cab1f68..683cb5b 100644 --- a/fs/gfs2/daemon.c +++ b/fs/gfs2/daemon.c @@ -112,6 +112,7 @@ int gfs2_logd(void *data) struct gfs2_sbd *sdp = data; struct gfs2_holder ji_gh; unsigned long t; + int need_flush; while (!kthread_should_stop()) { /* Advance the log tail */ @@ -120,8 +121,10 @@ int gfs2_logd(void *data) gfs2_tune_get(sdp, gt_log_flush_secs) * HZ; gfs2_ail1_empty(sdp, DIO_ALL); - - if (time_after_eq(jiffies, t)) { + gfs2_log_lock(sdp); + need_flush = sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks); + gfs2_log_unlock(sdp); + if (need_flush || time_after_eq(jiffies, t)) { gfs2_log_flush(sdp, NULL); sdp->sd_log_flush_time = jiffies; } diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index e24af28b1..0fdcb77 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -131,8 +131,8 @@ static int gfs2_dir_write_stuffed(struct gfs2_inode *ip, const char *buf, memcpy(dibh->b_data + offset + sizeof(struct gfs2_dinode), buf, size); if (ip->i_di.di_size < offset + size) ip->i_di.di_size = offset + size; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); @@ -229,10 +229,10 @@ out: if (ip->i_di.di_size < offset + copied) ip->i_di.di_size = offset + copied; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); return copied; @@ -340,10 +340,15 @@ fail: return (copied) ? copied : error; } +static inline int gfs2_dirent_sentinel(const struct gfs2_dirent *dent) +{ + return dent->de_inum.no_addr == 0 || dent->de_inum.no_formal_ino == 0; +} + static inline int __gfs2_dirent_find(const struct gfs2_dirent *dent, const struct qstr *name, int ret) { - if (dent->de_inum.no_addr != 0 && + if (!gfs2_dirent_sentinel(dent) && be32_to_cpu(dent->de_hash) == name->hash && be16_to_cpu(dent->de_name_len) == name->len && memcmp(dent+1, name->name, name->len) == 0) @@ -388,7 +393,7 @@ static int gfs2_dirent_find_space(const struct gfs2_dirent *dent, unsigned actual = GFS2_DIRENT_SIZE(be16_to_cpu(dent->de_name_len)); unsigned totlen = be16_to_cpu(dent->de_rec_len); - if (!dent->de_inum.no_addr) + if (gfs2_dirent_sentinel(dent)) actual = GFS2_DIRENT_SIZE(0); if (totlen - actual >= required) return 1; @@ -405,7 +410,7 @@ static int gfs2_dirent_gather(const struct gfs2_dirent *dent, void *opaque) { struct dirent_gather *g = opaque; - if (dent->de_inum.no_addr) { + if (!gfs2_dirent_sentinel(dent)) { g->pdent[g->offset++] = dent; } return 0; @@ -433,10 +438,10 @@ static int gfs2_check_dirent(struct gfs2_dirent *dent, unsigned int offset, if (unlikely(offset + size > len)) goto error; msg = "zero inode number"; - if (unlikely(!first && !dent->de_inum.no_addr)) + if (unlikely(!first && gfs2_dirent_sentinel(dent))) goto error; msg = "name length is greater than space in dirent"; - if (dent->de_inum.no_addr && + if (!gfs2_dirent_sentinel(dent) && unlikely(sizeof(struct gfs2_dirent)+be16_to_cpu(dent->de_name_len) > size)) goto error; @@ -598,7 +603,7 @@ static int dirent_next(struct gfs2_inode *dip, struct buffer_head *bh, return ret; /* Only the first dent could ever have de_inum.no_addr == 0 */ - if (!tmp->de_inum.no_addr) { + if (gfs2_dirent_sentinel(tmp)) { gfs2_consist_inode(dip); return -EIO; } @@ -621,7 +626,7 @@ static void dirent_del(struct gfs2_inode *dip, struct buffer_head *bh, { u16 cur_rec_len, prev_rec_len; - if (!cur->de_inum.no_addr) { + if (gfs2_dirent_sentinel(cur)) { gfs2_consist_inode(dip); return; } @@ -633,7 +638,8 @@ static void dirent_del(struct gfs2_inode *dip, struct buffer_head *bh, out the inode number and return. */ if (!prev) { - cur->de_inum.no_addr = 0; /* No endianess worries */ + cur->de_inum.no_addr = 0; + cur->de_inum.no_formal_ino = 0; return; } @@ -664,7 +670,7 @@ static struct gfs2_dirent *gfs2_init_dirent(struct inode *inode, struct gfs2_dirent *ndent; unsigned offset = 0, totlen; - if (dent->de_inum.no_addr) + if (!gfs2_dirent_sentinel(dent)) offset = GFS2_DIRENT_SIZE(be16_to_cpu(dent->de_name_len)); totlen = be16_to_cpu(dent->de_rec_len); BUG_ON(offset + name->len > totlen); @@ -713,12 +719,12 @@ static int get_leaf(struct gfs2_inode *dip, u64 leaf_no, static int get_leaf_nr(struct gfs2_inode *dip, u32 index, u64 *leaf_out) { - u64 leaf_no; + __be64 leaf_no; int error; error = gfs2_dir_read_data(dip, (char *)&leaf_no, - index * sizeof(u64), - sizeof(u64), 0); + index * sizeof(__be64), + sizeof(__be64), 0); if (error != sizeof(u64)) return (error < 0) ? error : -EIO; @@ -837,7 +843,8 @@ static int dir_make_exhash(struct inode *inode) struct gfs2_leaf *leaf; int y; u32 x; - u64 *lp, bn; + __be64 *lp; + u64 bn; int error; error = gfs2_meta_inode_buffer(dip, &dibh); @@ -893,20 +900,20 @@ static int dir_make_exhash(struct inode *inode) gfs2_trans_add_bh(dip->i_gl, dibh, 1); gfs2_buffer_clear_tail(dibh, sizeof(struct gfs2_dinode)); - lp = (u64 *)(dibh->b_data + sizeof(struct gfs2_dinode)); + lp = (__be64 *)(dibh->b_data + sizeof(struct gfs2_dinode)); for (x = sdp->sd_hash_ptrs; x--; lp++) *lp = cpu_to_be64(bn); dip->i_di.di_size = sdp->sd_sb.sb_bsize / 2; dip->i_di.di_blocks++; + gfs2_set_inode_blocks(&dip->i_inode); dip->i_di.di_flags |= GFS2_DIF_EXHASH; - dip->i_di.di_payload_format = 0; for (x = sdp->sd_hash_ptrs, y = -1; x; x >>= 1, y++) ; dip->i_di.di_depth = y; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); @@ -929,7 +936,8 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name) struct gfs2_leaf *nleaf, *oleaf; struct gfs2_dirent *dent = NULL, *prev = NULL, *next = NULL, *new; u32 start, len, half_len, divider; - u64 bn, *lp, leaf_no; + u64 bn, leaf_no; + __be64 *lp; u32 index; int x, moved = 0; int error; @@ -974,7 +982,7 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name) /* Change the pointers. Don't bother distinguishing stuffed from non-stuffed. This code is complicated enough already. */ - lp = kmalloc(half_len * sizeof(u64), GFP_NOFS | __GFP_NOFAIL); + lp = kmalloc(half_len * sizeof(__be64), GFP_NOFS | __GFP_NOFAIL); /* Change the pointers */ for (x = 0; x < half_len; x++) lp[x] = cpu_to_be64(bn); @@ -1000,7 +1008,7 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name) if (dirent_next(dip, obh, &next)) next = NULL; - if (dent->de_inum.no_addr && + if (!gfs2_dirent_sentinel(dent) && be32_to_cpu(dent->de_hash) < divider) { struct qstr str; str.name = (char*)(dent+1); @@ -1037,7 +1045,8 @@ static int dir_split_leaf(struct inode *inode, const struct qstr *name) error = gfs2_meta_inode_buffer(dip, &dibh); if (!gfs2_assert_withdraw(GFS2_SB(&dip->i_inode), !error)) { dip->i_di.di_blocks++; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_set_inode_blocks(&dip->i_inode); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); } @@ -1117,7 +1126,7 @@ static int dir_double_exhash(struct gfs2_inode *dip) error = gfs2_meta_inode_buffer(dip, &dibh); if (!gfs2_assert_withdraw(sdp, !error)) { dip->i_di.di_depth++; - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); } @@ -1194,7 +1203,7 @@ static int do_filldir_main(struct gfs2_inode *dip, u64 *offset, int *copied) { const struct gfs2_dirent *dent, *dent_next; - struct gfs2_inum inum; + struct gfs2_inum_host inum; u64 off, off_next; unsigned int x, y; int run = 0; @@ -1341,7 +1350,7 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, u32 hsize, len = 0; u32 ht_offset, lp_offset, ht_offset_cur = -1; u32 hash, index; - u64 *lp; + __be64 *lp; int copied = 0; int error = 0; unsigned depth = 0; @@ -1365,7 +1374,7 @@ static int dir_e_read(struct inode *inode, u64 *offset, void *opaque, if (ht_offset_cur != ht_offset) { error = gfs2_dir_read_data(dip, (char *)lp, - ht_offset * sizeof(u64), + ht_offset * sizeof(__be64), sdp->sd_hash_bsize, 1); if (error != sdp->sd_hash_bsize) { if (error >= 0) @@ -1456,7 +1465,7 @@ out: */ int gfs2_dir_search(struct inode *dir, const struct qstr *name, - struct gfs2_inum *inum, unsigned int *type) + struct gfs2_inum_host *inum, unsigned int *type) { struct buffer_head *bh; struct gfs2_dirent *dent; @@ -1515,7 +1524,8 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) return error; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_blocks++; - gfs2_dinode_out(&ip->i_di, bh->b_data); + gfs2_set_inode_blocks(&ip->i_inode); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); return 0; } @@ -1531,7 +1541,7 @@ static int dir_new_leaf(struct inode *inode, const struct qstr *name) */ int gfs2_dir_add(struct inode *inode, const struct qstr *name, - const struct gfs2_inum *inum, unsigned type) + const struct gfs2_inum_host *inum, unsigned type) { struct gfs2_inode *ip = GFS2_I(inode); struct buffer_head *bh; @@ -1558,8 +1568,8 @@ int gfs2_dir_add(struct inode *inode, const struct qstr *name, break; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_entries++; - ip->i_di.di_mtime = ip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&ip->i_di, bh->b_data); + ip->i_inode.i_mtime.tv_sec = ip->i_inode.i_ctime.tv_sec = get_seconds(); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); error = 0; break; @@ -1644,8 +1654,8 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) gfs2_consist_inode(dip); gfs2_trans_add_bh(dip->i_gl, bh, 1); dip->i_di.di_entries--; - dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&dip->i_di, bh->b_data); + dip->i_inode.i_mtime.tv_sec = dip->i_inode.i_ctime.tv_sec = get_seconds(); + gfs2_dinode_out(dip, bh->b_data); brelse(bh); mark_inode_dirty(&dip->i_inode); @@ -1666,7 +1676,7 @@ int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *name) */ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, - struct gfs2_inum *inum, unsigned int new_type) + struct gfs2_inum_host *inum, unsigned int new_type) { struct buffer_head *bh; struct gfs2_dirent *dent; @@ -1692,8 +1702,8 @@ int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, gfs2_trans_add_bh(dip->i_gl, bh, 1); } - dip->i_di.di_mtime = dip->i_di.di_ctime = get_seconds(); - gfs2_dinode_out(&dip->i_di, bh->b_data); + dip->i_inode.i_mtime.tv_sec = dip->i_inode.i_ctime.tv_sec = get_seconds(); + gfs2_dinode_out(dip, bh->b_data); brelse(bh); return 0; } @@ -1715,7 +1725,7 @@ static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data) u32 hsize, len; u32 ht_offset, lp_offset, ht_offset_cur = -1; u32 index = 0; - u64 *lp; + __be64 *lp; u64 leaf_no; int error = 0; @@ -1735,7 +1745,7 @@ static int foreach_leaf(struct gfs2_inode *dip, leaf_call_t lc, void *data) if (ht_offset_cur != ht_offset) { error = gfs2_dir_read_data(dip, (char *)lp, - ht_offset * sizeof(u64), + ht_offset * sizeof(__be64), sdp->sd_hash_bsize, 1); if (error != sdp->sd_hash_bsize) { if (error >= 0) @@ -1859,6 +1869,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, if (!dip->i_di.di_blocks) gfs2_consist_inode(dip); dip->i_di.di_blocks--; + gfs2_set_inode_blocks(&dip->i_inode); } error = gfs2_dir_write_data(dip, ht, index * sizeof(u64), size); @@ -1873,7 +1884,7 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, goto out_end_trans; gfs2_trans_add_bh(dip->i_gl, dibh, 1); - gfs2_dinode_out(&dip->i_di, dibh->b_data); + gfs2_dinode_out(dip, dibh->b_data); brelse(dibh); out_end_trans: diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index 3712334..b21b336 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -31,17 +31,17 @@ struct gfs2_inum; typedef int (*gfs2_filldir_t) (void *opaque, const char *name, unsigned int length, u64 offset, - struct gfs2_inum *inum, unsigned int type); + struct gfs2_inum_host *inum, unsigned int type); int gfs2_dir_search(struct inode *dir, const struct qstr *filename, - struct gfs2_inum *inum, unsigned int *type); + struct gfs2_inum_host *inum, unsigned int *type); int gfs2_dir_add(struct inode *inode, const struct qstr *filename, - const struct gfs2_inum *inum, unsigned int type); + const struct gfs2_inum_host *inum, unsigned int type); int gfs2_dir_del(struct gfs2_inode *dip, const struct qstr *filename); int gfs2_dir_read(struct inode *inode, u64 * offset, void *opaque, gfs2_filldir_t filldir); int gfs2_dir_mvino(struct gfs2_inode *dip, const struct qstr *filename, - struct gfs2_inum *new_inum, unsigned int new_type); + struct gfs2_inum_host *new_inum, unsigned int new_type); int gfs2_dir_exhash_dealloc(struct gfs2_inode *dip); diff --git a/fs/gfs2/eaops.c b/fs/gfs2/eaops.c index 92c54e9..cd747c0 100644 --- a/fs/gfs2/eaops.c +++ b/fs/gfs2/eaops.c @@ -120,7 +120,7 @@ static int system_eo_set(struct gfs2_inode *ip, struct gfs2_ea_request *er) if (GFS2_ACL_IS_ACCESS(er->er_name, er->er_name_len)) { if (!(er->er_flags & GFS2_ERF_MODE)) { - er->er_mode = ip->i_di.di_mode; + er->er_mode = ip->i_inode.i_mode; er->er_flags |= GFS2_ERF_MODE; } error = gfs2_acl_validate_set(ip, 1, er, diff --git a/fs/gfs2/eattr.c b/fs/gfs2/eattr.c index a65a4cc..ebebbdc 100644 --- a/fs/gfs2/eattr.c +++ b/fs/gfs2/eattr.c @@ -112,7 +112,7 @@ fail: static int ea_foreach(struct gfs2_inode *ip, ea_call_t ea_call, void *data) { struct buffer_head *bh, *eabh; - u64 *eablk, *end; + __be64 *eablk, *end; int error; error = gfs2_meta_read(ip->i_gl, ip->i_di.di_eattr, DIO_WAIT, &bh); @@ -129,7 +129,7 @@ static int ea_foreach(struct gfs2_inode *ip, ea_call_t ea_call, void *data) goto out; } - eablk = (u64 *)(bh->b_data + sizeof(struct gfs2_meta_header)); + eablk = (__be64 *)(bh->b_data + sizeof(struct gfs2_meta_header)); end = eablk + GFS2_SB(&ip->i_inode)->sd_inptrs; for (; eablk < end; eablk++) { @@ -224,7 +224,8 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, struct gfs2_rgrpd *rgd; struct gfs2_holder rg_gh; struct buffer_head *dibh; - u64 *dataptrs, bn = 0; + __be64 *dataptrs; + u64 bn = 0; u64 bstart = 0; unsigned int blen = 0; unsigned int blks = 0; @@ -280,6 +281,7 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, if (!ip->i_di.di_blocks) gfs2_consist_inode(ip); ip->i_di.di_blocks--; + gfs2_set_inode_blocks(&ip->i_inode); } if (bstart) gfs2_free_meta(ip, bstart, blen); @@ -299,9 +301,9 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -444,7 +446,7 @@ static int ea_get_unstuffed(struct gfs2_inode *ip, struct gfs2_ea_header *ea, struct buffer_head **bh; unsigned int amount = GFS2_EA_DATA_LEN(ea); unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize); - u64 *dataptrs = GFS2_EA2DATAPTRS(ea); + __be64 *dataptrs = GFS2_EA2DATAPTRS(ea); unsigned int x; int error = 0; @@ -597,6 +599,7 @@ static int ea_alloc_blk(struct gfs2_inode *ip, struct buffer_head **bhp) ea->ea_num_ptrs = 0; ip->i_di.di_blocks++; + gfs2_set_inode_blocks(&ip->i_inode); return 0; } @@ -629,7 +632,7 @@ static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea, ea->ea_num_ptrs = 0; memcpy(GFS2_EA2DATA(ea), er->er_data, er->er_data_len); } else { - u64 *dataptr = GFS2_EA2DATAPTRS(ea); + __be64 *dataptr = GFS2_EA2DATAPTRS(ea); const char *data = er->er_data; unsigned int data_len = er->er_data_len; unsigned int copy; @@ -648,6 +651,7 @@ static int ea_write(struct gfs2_inode *ip, struct gfs2_ea_header *ea, gfs2_metatype_set(bh, GFS2_METATYPE_ED, GFS2_FORMAT_ED); ip->i_di.di_blocks++; + gfs2_set_inode_blocks(&ip->i_inode); copy = data_len > sdp->sd_jbsize ? sdp->sd_jbsize : data_len; @@ -686,7 +690,7 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -710,13 +714,13 @@ static int ea_alloc_skeleton(struct gfs2_inode *ip, struct gfs2_ea_request *er, if (!error) { if (er->er_flags & GFS2_ERF_MODE) { gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), - (ip->i_di.di_mode & S_IFMT) == + (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT)); - ip->i_di.di_mode = er->er_mode; + ip->i_inode.i_mode = er->er_mode; } - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -846,12 +850,12 @@ static int ea_set_simple_noalloc(struct gfs2_inode *ip, struct buffer_head *bh, if (er->er_flags & GFS2_ERF_MODE) { gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), - (ip->i_di.di_mode & S_IFMT) == (er->er_mode & S_IFMT)); - ip->i_di.di_mode = er->er_mode; + (ip->i_inode.i_mode & S_IFMT) == (er->er_mode & S_IFMT)); + ip->i_inode.i_mode = er->er_mode; } - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); out: gfs2_trans_end(GFS2_SB(&ip->i_inode)); @@ -931,12 +935,12 @@ static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *indbh, *newbh; - u64 *eablk; + __be64 *eablk; int error; int mh_size = sizeof(struct gfs2_meta_header); if (ip->i_di.di_flags & GFS2_DIF_EA_INDIRECT) { - u64 *end; + __be64 *end; error = gfs2_meta_read(ip->i_gl, ip->i_di.di_eattr, DIO_WAIT, &indbh); @@ -948,7 +952,7 @@ static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er, goto out; } - eablk = (u64 *)(indbh->b_data + mh_size); + eablk = (__be64 *)(indbh->b_data + mh_size); end = eablk + sdp->sd_inptrs; for (; eablk < end; eablk++) @@ -971,11 +975,12 @@ static int ea_set_block(struct gfs2_inode *ip, struct gfs2_ea_request *er, gfs2_metatype_set(indbh, GFS2_METATYPE_IN, GFS2_FORMAT_IN); gfs2_buffer_clear_tail(indbh, mh_size); - eablk = (u64 *)(indbh->b_data + mh_size); + eablk = (__be64 *)(indbh->b_data + mh_size); *eablk = cpu_to_be64(ip->i_di.di_eattr); ip->i_di.di_eattr = blk; ip->i_di.di_flags |= GFS2_DIF_EA_INDIRECT; ip->i_di.di_blocks++; + gfs2_set_inode_blocks(&ip->i_inode); eablk++; } @@ -1129,9 +1134,9 @@ static int ea_remove_stuffed(struct gfs2_inode *ip, struct gfs2_ea_location *el) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1202,7 +1207,7 @@ static int ea_acl_chmod_unstuffed(struct gfs2_inode *ip, struct buffer_head **bh; unsigned int amount = GFS2_EA_DATA_LEN(ea); unsigned int nptrs = DIV_ROUND_UP(amount, sdp->sd_jbsize); - u64 *dataptrs = GFS2_EA2DATAPTRS(ea); + __be64 *dataptrs = GFS2_EA2DATAPTRS(ea); unsigned int x; int error; @@ -1284,9 +1289,8 @@ int gfs2_ea_acl_chmod(struct gfs2_inode *ip, struct gfs2_ea_location *el, if (!error) { error = inode_setattr(&ip->i_inode, attr); gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error); - gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1300,7 +1304,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrp_list rlist; struct buffer_head *indbh, *dibh; - u64 *eablk, *end; + __be64 *eablk, *end; unsigned int rg_blocks = 0; u64 bstart = 0; unsigned int blen = 0; @@ -1319,7 +1323,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) goto out; } - eablk = (u64 *)(indbh->b_data + sizeof(struct gfs2_meta_header)); + eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header)); end = eablk + sdp->sd_inptrs; for (; eablk < end; eablk++) { @@ -1363,7 +1367,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) gfs2_trans_add_bh(ip->i_gl, indbh, 1); - eablk = (u64 *)(indbh->b_data + sizeof(struct gfs2_meta_header)); + eablk = (__be64 *)(indbh->b_data + sizeof(struct gfs2_meta_header)); bstart = 0; blen = 0; @@ -1387,6 +1391,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) if (!ip->i_di.di_blocks) gfs2_consist_inode(ip); ip->i_di.di_blocks--; + gfs2_set_inode_blocks(&ip->i_inode); } if (bstart) gfs2_free_meta(ip, bstart, blen); @@ -1396,7 +1401,7 @@ static int ea_dealloc_indirect(struct gfs2_inode *ip) error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -1441,11 +1446,12 @@ static int ea_dealloc_block(struct gfs2_inode *ip) if (!ip->i_di.di_blocks) gfs2_consist_inode(ip); ip->i_di.di_blocks--; + gfs2_set_inode_blocks(&ip->i_inode); error = gfs2_meta_inode_buffer(ip, &dibh); if (!error) { gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } diff --git a/fs/gfs2/eattr.h b/fs/gfs2/eattr.h index ffa6594..c82dbe0 100644 --- a/fs/gfs2/eattr.h +++ b/fs/gfs2/eattr.h @@ -19,7 +19,7 @@ struct iattr; #define GFS2_EA_SIZE(ea) \ ALIGN(sizeof(struct gfs2_ea_header) + (ea)->ea_name_len + \ ((GFS2_EA_IS_STUFFED(ea)) ? GFS2_EA_DATA_LEN(ea) : \ - (sizeof(u64) * (ea)->ea_num_ptrs)), 8) + (sizeof(__be64) * (ea)->ea_num_ptrs)), 8) #define GFS2_EA_IS_STUFFED(ea) (!(ea)->ea_num_ptrs) #define GFS2_EA_IS_LAST(ea) ((ea)->ea_flags & GFS2_EAFLAG_LAST) @@ -29,13 +29,13 @@ ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + (er)->er_data_len, 8) #define GFS2_EAREQ_SIZE_UNSTUFFED(sdp, er) \ ALIGN(sizeof(struct gfs2_ea_header) + (er)->er_name_len + \ - sizeof(u64) * DIV_ROUND_UP((er)->er_data_len, (sdp)->sd_jbsize), 8) + sizeof(__be64) * DIV_ROUND_UP((er)->er_data_len, (sdp)->sd_jbsize), 8) #define GFS2_EA2NAME(ea) ((char *)((struct gfs2_ea_header *)(ea) + 1)) #define GFS2_EA2DATA(ea) (GFS2_EA2NAME(ea) + (ea)->ea_name_len) #define GFS2_EA2DATAPTRS(ea) \ -((u64 *)(GFS2_EA2NAME(ea) + ALIGN((ea)->ea_name_len, 8))) +((__be64 *)(GFS2_EA2NAME(ea) + ALIGN((ea)->ea_name_len, 8))) #define GFS2_EA2NEXT(ea) \ ((struct gfs2_ea_header *)((char *)(ea) + GFS2_EA_REC_LEN(ea))) diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 55f5333..4381469 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -96,7 +96,7 @@ static inline rwlock_t *gl_lock_addr(unsigned int x) return &gl_hash_locks[x & (GL_HASH_LOCK_SZ-1)]; } #else /* not SMP, so no spinlocks required */ -static inline rwlock_t *gl_lock_addr(x) +static inline rwlock_t *gl_lock_addr(unsigned int x) { return NULL; } @@ -769,7 +769,7 @@ restart: } else { spin_unlock(&gl->gl_spin); - new_gh = gfs2_holder_get(gl, state, LM_FLAG_TRY, GFP_KERNEL); + new_gh = gfs2_holder_get(gl, state, LM_FLAG_TRY, GFP_NOFS); if (!new_gh) return; set_bit(HIF_DEMOTE, &new_gh->gh_iflags); @@ -785,21 +785,6 @@ out: gfs2_holder_put(new_gh); } -void gfs2_glock_inode_squish(struct inode *inode) -{ - struct gfs2_holder gh; - struct gfs2_glock *gl = GFS2_I(inode)->i_gl; - gfs2_holder_init(gl, LM_ST_UNLOCKED, 0, &gh); - set_bit(HIF_DEMOTE, &gh.gh_iflags); - spin_lock(&gl->gl_spin); - gfs2_assert(inode->i_sb->s_fs_info, list_empty(&gl->gl_holders)); - list_add_tail(&gh.gh_list, &gl->gl_waiters2); - run_queue(gl); - spin_unlock(&gl->gl_spin); - wait_for_completion(&gh.gh_wait); - gfs2_holder_uninit(&gh); -} - /** * state_change - record that the glock is now in a different state * @gl: the glock @@ -847,12 +832,12 @@ static void xmote_bh(struct gfs2_glock *gl, unsigned int ret) if (prev_state != LM_ST_UNLOCKED && !(ret & LM_OUT_CACHEABLE)) { if (glops->go_inval) - glops->go_inval(gl, DIO_METADATA | DIO_DATA); + glops->go_inval(gl, DIO_METADATA); } else if (gl->gl_state == LM_ST_DEFERRED) { /* We might not want to do this here. Look at moving to the inode glops. */ if (glops->go_inval) - glops->go_inval(gl, DIO_DATA); + glops->go_inval(gl, 0); } /* Deal with each possible exit condition */ @@ -954,7 +939,7 @@ void gfs2_glock_xmote_th(struct gfs2_glock *gl, unsigned int state, int flags) gfs2_assert_warn(sdp, state != gl->gl_state); if (gl->gl_state == LM_ST_EXCLUSIVE && glops->go_sync) - glops->go_sync(gl, DIO_METADATA | DIO_DATA | DIO_RELEASE); + glops->go_sync(gl); gfs2_glock_hold(gl); gl->gl_req_bh = xmote_bh; @@ -995,7 +980,7 @@ static void drop_bh(struct gfs2_glock *gl, unsigned int ret) state_change(gl, LM_ST_UNLOCKED); if (glops->go_inval) - glops->go_inval(gl, DIO_METADATA | DIO_DATA); + glops->go_inval(gl, DIO_METADATA); if (gh) { spin_lock(&gl->gl_spin); @@ -1041,7 +1026,7 @@ void gfs2_glock_drop_th(struct gfs2_glock *gl) gfs2_assert_warn(sdp, gl->gl_state != LM_ST_UNLOCKED); if (gl->gl_state == LM_ST_EXCLUSIVE && glops->go_sync) - glops->go_sync(gl, DIO_METADATA | DIO_DATA | DIO_RELEASE); + glops->go_sync(gl); gfs2_glock_hold(gl); gl->gl_req_bh = drop_bh; @@ -1244,9 +1229,6 @@ restart: clear_bit(GLF_PREFETCH, &gl->gl_flags); - if (error == GLR_TRYFAILED && (gh->gh_flags & GL_DUMP)) - dump_glock(gl); - return error; } @@ -1923,7 +1905,7 @@ out: static void scan_glock(struct gfs2_glock *gl) { - if (gl->gl_ops == &gfs2_inode_glops) + if (gl->gl_ops == &gfs2_inode_glops && gl->gl_object) return; if (gfs2_glmutex_trylock(gl)) { @@ -2078,7 +2060,7 @@ static int dump_inode(struct gfs2_inode *ip) printk(KERN_INFO " num = %llu %llu\n", (unsigned long long)ip->i_num.no_formal_ino, (unsigned long long)ip->i_num.no_addr); - printk(KERN_INFO " type = %u\n", IF2DT(ip->i_di.di_mode)); + printk(KERN_INFO " type = %u\n", IF2DT(ip->i_inode.i_mode)); printk(KERN_INFO " i_flags ="); for (x = 0; x < 32; x++) if (test_bit(x, &ip->i_flags)) diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 2b2a889..fb39108 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -27,8 +27,6 @@ #define GL_ATIME 0x00000200 #define GL_NOCACHE 0x00000400 #define GL_NOCANCEL 0x00001000 -#define GL_AOP 0x00004000 -#define GL_DUMP 0x00008000 #define GLR_TRYFAILED 13 #define GLR_CANCELED 14 @@ -108,7 +106,6 @@ void gfs2_glock_dq_uninit_m(unsigned int num_gh, struct gfs2_holder *ghs); void gfs2_glock_prefetch_num(struct gfs2_sbd *sdp, u64 number, const struct gfs2_glock_operations *glops, unsigned int state, int flags); -void gfs2_glock_inode_squish(struct inode *inode); /** * gfs2_glock_nq_init - intialize a holder and enqueue it on a glock diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 41a6b68..b068d10 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -92,7 +92,7 @@ static void gfs2_pte_inval(struct gfs2_glock *gl) ip = gl->gl_object; inode = &ip->i_inode; - if (!ip || !S_ISREG(ip->i_di.di_mode)) + if (!ip || !S_ISREG(inode->i_mode)) return; if (!test_bit(GIF_PAGED, &ip->i_flags)) @@ -107,89 +107,20 @@ static void gfs2_pte_inval(struct gfs2_glock *gl) } /** - * gfs2_page_inval - Invalidate all pages associated with a glock - * @gl: the glock - * - */ - -static void gfs2_page_inval(struct gfs2_glock *gl) -{ - struct gfs2_inode *ip; - struct inode *inode; - - ip = gl->gl_object; - inode = &ip->i_inode; - if (!ip || !S_ISREG(ip->i_di.di_mode)) - return; - - truncate_inode_pages(inode->i_mapping, 0); - gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), !inode->i_mapping->nrpages); - clear_bit(GIF_PAGED, &ip->i_flags); -} - -/** - * gfs2_page_wait - Wait for writeback of data - * @gl: the glock - * - * Syncs data (not metadata) for a regular file. - * No-op for all other types. - */ - -static void gfs2_page_wait(struct gfs2_glock *gl) -{ - struct gfs2_inode *ip = gl->gl_object; - struct inode *inode = &ip->i_inode; - struct address_space *mapping = inode->i_mapping; - int error; - - if (!S_ISREG(ip->i_di.di_mode)) - return; - - error = filemap_fdatawait(mapping); - - /* Put back any errors cleared by filemap_fdatawait() - so they can be caught by someone who can pass them - up to user space. */ - - if (error == -ENOSPC) - set_bit(AS_ENOSPC, &mapping->flags); - else if (error) - set_bit(AS_EIO, &mapping->flags); - -} - -static void gfs2_page_writeback(struct gfs2_glock *gl) -{ - struct gfs2_inode *ip = gl->gl_object; - struct inode *inode = &ip->i_inode; - struct address_space *mapping = inode->i_mapping; - - if (!S_ISREG(ip->i_di.di_mode)) - return; - - filemap_fdatawrite(mapping); -} - -/** * meta_go_sync - sync out the metadata for this glock * @gl: the glock - * @flags: DIO_* * * Called when demoting or unlocking an EX glock. We must flush * to disk all dirty buffers/pages relating to this glock, and must not * not return to caller to demote/unlock the glock until I/O is complete. */ -static void meta_go_sync(struct gfs2_glock *gl, int flags) +static void meta_go_sync(struct gfs2_glock *gl) { - if (!(flags & DIO_METADATA)) - return; - if (test_and_clear_bit(GLF_DIRTY, &gl->gl_flags)) { gfs2_log_flush(gl->gl_sbd, gl); gfs2_meta_sync(gl); - if (flags & DIO_RELEASE) - gfs2_ail_empty_gl(gl); + gfs2_ail_empty_gl(gl); } } @@ -264,31 +195,31 @@ static void inode_go_drop_th(struct gfs2_glock *gl) /** * inode_go_sync - Sync the dirty data and/or metadata for an inode glock * @gl: the glock protecting the inode - * @flags: * */ -static void inode_go_sync(struct gfs2_glock *gl, int flags) +static void inode_go_sync(struct gfs2_glock *gl) { - int meta = (flags & DIO_METADATA); - int data = (flags & DIO_DATA); + struct gfs2_inode *ip = gl->gl_object; + + if (ip && !S_ISREG(ip->i_inode.i_mode)) + ip = NULL; if (test_bit(GLF_DIRTY, &gl->gl_flags)) { - if (meta && data) { - gfs2_page_writeback(gl); - gfs2_log_flush(gl->gl_sbd, gl); - gfs2_meta_sync(gl); - gfs2_page_wait(gl); - clear_bit(GLF_DIRTY, &gl->gl_flags); - } else if (meta) { - gfs2_log_flush(gl->gl_sbd, gl); - gfs2_meta_sync(gl); - } else if (data) { - gfs2_page_writeback(gl); - gfs2_page_wait(gl); + gfs2_log_flush(gl->gl_sbd, gl); + if (ip) + filemap_fdatawrite(ip->i_inode.i_mapping); + gfs2_meta_sync(gl); + if (ip) { + struct address_space *mapping = ip->i_inode.i_mapping; + int error = filemap_fdatawait(mapping); + if (error == -ENOSPC) + set_bit(AS_ENOSPC, &mapping->flags); + else if (error) + set_bit(AS_EIO, &mapping->flags); } - if (flags & DIO_RELEASE) - gfs2_ail_empty_gl(gl); + clear_bit(GLF_DIRTY, &gl->gl_flags); + gfs2_ail_empty_gl(gl); } } @@ -301,15 +232,20 @@ static void inode_go_sync(struct gfs2_glock *gl, int flags) static void inode_go_inval(struct gfs2_glock *gl, int flags) { + struct gfs2_inode *ip = gl->gl_object; int meta = (flags & DIO_METADATA); - int data = (flags & DIO_DATA); if (meta) { gfs2_meta_inval(gl); - gl->gl_vn++; + if (ip) + set_bit(GIF_INVALID, &ip->i_flags); + } + + if (ip && S_ISREG(ip->i_inode.i_mode)) { + truncate_inode_pages(ip->i_inode.i_mapping, 0); + gfs2_assert_withdraw(GFS2_SB(&ip->i_inode), !ip->i_inode.i_mapping->nrpages); + clear_bit(GIF_PAGED, &ip->i_flags); } - if (data) - gfs2_page_inval(gl); } /** @@ -351,11 +287,10 @@ static int inode_go_lock(struct gfs2_holder *gh) if (!ip) return 0; - if (ip->i_vn != gl->gl_vn) { + if (test_bit(GIF_INVALID, &ip->i_flags)) { error = gfs2_inode_refresh(ip); if (error) return error; - gfs2_inode_attr_in(ip); } if ((ip->i_di.di_flags & GFS2_DIF_TRUNC_IN_PROG) && @@ -379,11 +314,8 @@ static void inode_go_unlock(struct gfs2_holder *gh) struct gfs2_glock *gl = gh->gh_gl; struct gfs2_inode *ip = gl->gl_object; - if (ip == NULL) - return; - if (test_bit(GLF_DIRTY, &gl->gl_flags)) - gfs2_inode_attr_in(ip); - gfs2_meta_cache_flush(ip); + if (ip) + gfs2_meta_cache_flush(ip); } /** @@ -491,13 +423,13 @@ static void trans_go_xmote_bh(struct gfs2_glock *gl) struct gfs2_sbd *sdp = gl->gl_sbd; struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode); struct gfs2_glock *j_gl = ip->i_gl; - struct gfs2_log_header head; + struct gfs2_log_header_host head; int error; if (gl->gl_state != LM_ST_UNLOCKED && test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { gfs2_meta_cache_flush(GFS2_I(sdp->sd_jdesc->jd_inode)); - j_gl->gl_ops->go_inval(j_gl, DIO_METADATA | DIO_DATA); + j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); error = gfs2_find_jhead(sdp->sd_jdesc, &head); if (error) diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 118dc69..734421e 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -14,8 +14,6 @@ #define DIO_WAIT 0x00000010 #define DIO_METADATA 0x00000020 -#define DIO_DATA 0x00000040 -#define DIO_RELEASE 0x00000080 #define DIO_ALL 0x00000100 struct gfs2_log_operations; @@ -41,7 +39,7 @@ struct gfs2_log_operations { void (*lo_before_commit) (struct gfs2_sbd *sdp); void (*lo_after_commit) (struct gfs2_sbd *sdp, struct gfs2_ail *ai); void (*lo_before_scan) (struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass); + struct gfs2_log_header_host *head, int pass); int (*lo_scan_elements) (struct gfs2_jdesc *jd, unsigned int start, struct gfs2_log_descriptor *ld, __be64 *ptr, int pass); @@ -67,8 +65,8 @@ struct gfs2_rgrpd { struct list_head rd_list_mru; struct list_head rd_recent; /* Recently used rgrps */ struct gfs2_glock *rd_gl; /* Glock for this rgrp */ - struct gfs2_rindex rd_ri; - struct gfs2_rgrp rd_rg; + struct gfs2_rindex_host rd_ri; + struct gfs2_rgrp_host rd_rg; u64 rd_rg_vn; struct gfs2_bitmap *rd_bits; unsigned int rd_bh_count; @@ -103,18 +101,17 @@ struct gfs2_bufdata { }; struct gfs2_glock_operations { - void (*go_xmote_th) (struct gfs2_glock * gl, unsigned int state, - int flags); - void (*go_xmote_bh) (struct gfs2_glock * gl); - void (*go_drop_th) (struct gfs2_glock * gl); - void (*go_drop_bh) (struct gfs2_glock * gl); - void (*go_sync) (struct gfs2_glock * gl, int flags); - void (*go_inval) (struct gfs2_glock * gl, int flags); - int (*go_demote_ok) (struct gfs2_glock * gl); - int (*go_lock) (struct gfs2_holder * gh); - void (*go_unlock) (struct gfs2_holder * gh); - void (*go_callback) (struct gfs2_glock * gl, unsigned int state); - void (*go_greedy) (struct gfs2_glock * gl); + void (*go_xmote_th) (struct gfs2_glock *gl, unsigned int state, int flags); + void (*go_xmote_bh) (struct gfs2_glock *gl); + void (*go_drop_th) (struct gfs2_glock *gl); + void (*go_drop_bh) (struct gfs2_glock *gl); + void (*go_sync) (struct gfs2_glock *gl); + void (*go_inval) (struct gfs2_glock *gl, int flags); + int (*go_demote_ok) (struct gfs2_glock *gl); + int (*go_lock) (struct gfs2_holder *gh); + void (*go_unlock) (struct gfs2_holder *gh); + void (*go_callback) (struct gfs2_glock *gl, unsigned int state); + void (*go_greedy) (struct gfs2_glock *gl); const int go_type; }; @@ -217,6 +214,7 @@ struct gfs2_alloc { }; enum { + GIF_INVALID = 0, GIF_QD_LOCKED = 1, GIF_PAGED = 2, GIF_SW_PAGED = 3, @@ -224,12 +222,11 @@ enum { struct gfs2_inode { struct inode i_inode; - struct gfs2_inum i_num; + struct gfs2_inum_host i_num; unsigned long i_flags; /* GIF_... */ - u64 i_vn; - struct gfs2_dinode i_di; /* To be replaced by ref to block */ + struct gfs2_dinode_host i_di; /* To be replaced by ref to block */ struct gfs2_glock *i_gl; /* Move into i_gh? */ struct gfs2_holder i_iopen_gh; @@ -450,7 +447,7 @@ struct gfs2_sbd { struct super_block *sd_vfs_meta; struct kobject sd_kobj; unsigned long sd_flags; /* SDF_... */ - struct gfs2_sb sd_sb; + struct gfs2_sb_host sd_sb; /* Constants computed on mount */ @@ -503,8 +500,8 @@ struct gfs2_sbd { spinlock_t sd_statfs_spin; struct mutex sd_statfs_mutex; - struct gfs2_statfs_change sd_statfs_master; - struct gfs2_statfs_change sd_statfs_local; + struct gfs2_statfs_change_host sd_statfs_master; + struct gfs2_statfs_change_host sd_statfs_local; unsigned long sd_statfs_sync_time; /* Resource group stuff */ diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index d470e52..d122074c 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -38,83 +38,12 @@ #include "trans.h" #include "util.h" -/** - * gfs2_inode_attr_in - Copy attributes from the dinode into the VFS inode - * @ip: The GFS2 inode (with embedded disk inode data) - * @inode: The Linux VFS inode - * - */ - -void gfs2_inode_attr_in(struct gfs2_inode *ip) -{ - struct inode *inode = &ip->i_inode; - struct gfs2_dinode *di = &ip->i_di; - - inode->i_ino = ip->i_num.no_addr; - - switch (di->di_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - inode->i_rdev = MKDEV(di->di_major, di->di_minor); - break; - default: - inode->i_rdev = 0; - break; - }; - - inode->i_mode = di->di_mode; - inode->i_nlink = di->di_nlink; - inode->i_uid = di->di_uid; - inode->i_gid = di->di_gid; - i_size_write(inode, di->di_size); - inode->i_atime.tv_sec = di->di_atime; - inode->i_mtime.tv_sec = di->di_mtime; - inode->i_ctime.tv_sec = di->di_ctime; - inode->i_atime.tv_nsec = 0; - inode->i_mtime.tv_nsec = 0; - inode->i_ctime.tv_nsec = 0; - inode->i_blocks = di->di_blocks << - (GFS2_SB(inode)->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT); - - if (di->di_flags & GFS2_DIF_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - - if (di->di_flags & GFS2_DIF_APPENDONLY) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; -} - -/** - * gfs2_inode_attr_out - Copy attributes from VFS inode into the dinode - * @ip: The GFS2 inode - * - * Only copy out the attributes that we want the VFS layer - * to be able to modify. - */ - -void gfs2_inode_attr_out(struct gfs2_inode *ip) -{ - struct inode *inode = &ip->i_inode; - struct gfs2_dinode *di = &ip->i_di; - gfs2_assert_withdraw(GFS2_SB(inode), - (di->di_mode & S_IFMT) == (inode->i_mode & S_IFMT)); - di->di_mode = inode->i_mode; - di->di_uid = inode->i_uid; - di->di_gid = inode->i_gid; - di->di_atime = inode->i_atime.tv_sec; - di->di_mtime = inode->i_mtime.tv_sec; - di->di_ctime = inode->i_ctime.tv_sec; -} - static int iget_test(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_inum *inum = opaque; + struct gfs2_inum_host *inum = opaque; - if (ip && ip->i_num.no_addr == inum->no_addr) + if (ip->i_num.no_addr == inum->no_addr) return 1; return 0; @@ -123,19 +52,20 @@ static int iget_test(struct inode *inode, void *opaque) static int iget_set(struct inode *inode, void *opaque) { struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_inum *inum = opaque; + struct gfs2_inum_host *inum = opaque; ip->i_num = *inum; + inode->i_ino = inum->no_addr; return 0; } -struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum *inum) +struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum_host *inum) { return ilookup5(sb, (unsigned long)inum->no_formal_ino, iget_test, inum); } -static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum *inum) +static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum_host *inum) { return iget5_locked(sb, (unsigned long)inum->no_formal_ino, iget_test, iget_set, inum); @@ -150,7 +80,7 @@ static struct inode *gfs2_iget(struct super_block *sb, struct gfs2_inum *inum) * Returns: A VFS inode, or an error */ -struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, unsigned int type) +struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum_host *inum, unsigned int type) { struct inode *inode = gfs2_iget(sb, inum); struct gfs2_inode *ip = GFS2_I(inode); @@ -188,7 +118,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, if (unlikely(error)) goto fail_put; - ip->i_vn = ip->i_gl->gl_vn - 1; + set_bit(GIF_INVALID, &ip->i_flags); error = gfs2_glock_nq_init(io_gl, LM_ST_SHARED, GL_EXACT, &ip->i_iopen_gh); if (unlikely(error)) goto fail_iopen; @@ -208,6 +138,63 @@ fail: return ERR_PTR(error); } +static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) +{ + struct gfs2_dinode_host *di = &ip->i_di; + const struct gfs2_dinode *str = buf; + + if (ip->i_num.no_addr != be64_to_cpu(str->di_num.no_addr)) { + if (gfs2_consist_inode(ip)) + gfs2_dinode_print(ip); + return -EIO; + } + if (ip->i_num.no_formal_ino != be64_to_cpu(str->di_num.no_formal_ino)) + return -ESTALE; + + ip->i_inode.i_mode = be32_to_cpu(str->di_mode); + ip->i_inode.i_rdev = 0; + switch (ip->i_inode.i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + ip->i_inode.i_rdev = MKDEV(be32_to_cpu(str->di_major), + be32_to_cpu(str->di_minor)); + break; + }; + + ip->i_inode.i_uid = be32_to_cpu(str->di_uid); + ip->i_inode.i_gid = be32_to_cpu(str->di_gid); + /* + * We will need to review setting the nlink count here in the + * light of the forthcoming ro bind mount work. This is a reminder + * to do that. + */ + ip->i_inode.i_nlink = be32_to_cpu(str->di_nlink); + di->di_size = be64_to_cpu(str->di_size); + i_size_write(&ip->i_inode, di->di_size); + di->di_blocks = be64_to_cpu(str->di_blocks); + gfs2_set_inode_blocks(&ip->i_inode); + ip->i_inode.i_atime.tv_sec = be64_to_cpu(str->di_atime); + ip->i_inode.i_atime.tv_nsec = 0; + ip->i_inode.i_mtime.tv_sec = be64_to_cpu(str->di_mtime); + ip->i_inode.i_mtime.tv_nsec = 0; + ip->i_inode.i_ctime.tv_sec = be64_to_cpu(str->di_ctime); + ip->i_inode.i_ctime.tv_nsec = 0; + + di->di_goal_meta = be64_to_cpu(str->di_goal_meta); + di->di_goal_data = be64_to_cpu(str->di_goal_data); + di->di_generation = be64_to_cpu(str->di_generation); + + di->di_flags = be32_to_cpu(str->di_flags); + gfs2_set_inode_flags(&ip->i_inode); + di->di_height = be16_to_cpu(str->di_height); + + di->di_depth = be16_to_cpu(str->di_depth); + di->di_entries = be32_to_cpu(str->di_entries); + + di->di_eattr = be64_to_cpu(str->di_eattr); + return 0; +} + /** * gfs2_inode_refresh - Refresh the incore copy of the dinode * @ip: The GFS2 inode @@ -229,21 +216,11 @@ int gfs2_inode_refresh(struct gfs2_inode *ip) return -EIO; } - gfs2_dinode_in(&ip->i_di, dibh->b_data); - + error = gfs2_dinode_in(ip, dibh->b_data); brelse(dibh); + clear_bit(GIF_INVALID, &ip->i_flags); - if (ip->i_num.no_addr != ip->i_di.di_num.no_addr) { - if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); - return -EIO; - } - if (ip->i_num.no_formal_ino != ip->i_di.di_num.no_formal_ino) - return -ESTALE; - - ip->i_vn = ip->i_gl->gl_vn; - - return 0; + return error; } int gfs2_dinode_dealloc(struct gfs2_inode *ip) @@ -255,7 +232,7 @@ int gfs2_dinode_dealloc(struct gfs2_inode *ip) if (ip->i_di.di_blocks != 1) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } @@ -318,14 +295,14 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) u32 nlink; int error; - BUG_ON(ip->i_di.di_nlink != ip->i_inode.i_nlink); - nlink = ip->i_di.di_nlink + diff; + BUG_ON(diff != 1 && diff != -1); + nlink = ip->i_inode.i_nlink + diff; /* If we are reducing the nlink count, but the new value ends up being bigger than the old one, we must have underflowed. */ - if (diff < 0 && nlink > ip->i_di.di_nlink) { + if (diff < 0 && nlink > ip->i_inode.i_nlink) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } @@ -333,16 +310,19 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) if (error) return error; - ip->i_di.di_nlink = nlink; - ip->i_di.di_ctime = get_seconds(); - ip->i_inode.i_nlink = nlink; + if (diff > 0) + inc_nlink(&ip->i_inode); + else + drop_nlink(&ip->i_inode); + + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); mark_inode_dirty(&ip->i_inode); - if (ip->i_di.di_nlink == 0) { + if (ip->i_inode.i_nlink == 0) { struct gfs2_rgrpd *rgd; struct gfs2_holder ri_gh, rg_gh; @@ -357,7 +337,6 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff) if (error) goto out_norgrp; - clear_nlink(&ip->i_inode); gfs2_unlink_di(&ip->i_inode); /* mark inode unlinked */ gfs2_glock_dq_uninit(&rg_gh); out_norgrp: @@ -394,7 +373,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, struct super_block *sb = dir->i_sb; struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_holder d_gh; - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error = 0; struct inode *inode = NULL; @@ -436,7 +415,7 @@ static int pick_formal_ino_1(struct gfs2_sbd *sdp, u64 *formal_ino) { struct gfs2_inode *ip = GFS2_I(sdp->sd_ir_inode); struct buffer_head *bh; - struct gfs2_inum_range ir; + struct gfs2_inum_range_host ir; int error; error = gfs2_trans_begin(sdp, RES_DINODE, 0); @@ -479,7 +458,7 @@ static int pick_formal_ino_2(struct gfs2_sbd *sdp, u64 *formal_ino) struct gfs2_inode *m_ip = GFS2_I(sdp->sd_inum_inode); struct gfs2_holder gh; struct buffer_head *bh; - struct gfs2_inum_range ir; + struct gfs2_inum_range_host ir; int error; error = gfs2_glock_nq_init(m_ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); @@ -500,21 +479,22 @@ static int pick_formal_ino_2(struct gfs2_sbd *sdp, u64 *formal_ino) if (!ir.ir_length) { struct buffer_head *m_bh; u64 x, y; + __be64 z; error = gfs2_meta_inode_buffer(m_ip, &m_bh); if (error) goto out_brelse; - x = *(u64 *)(m_bh->b_data + sizeof(struct gfs2_dinode)); - x = y = be64_to_cpu(x); + z = *(__be64 *)(m_bh->b_data + sizeof(struct gfs2_dinode)); + x = y = be64_to_cpu(z); ir.ir_start = x; ir.ir_length = GFS2_INUM_QUANTUM; x += GFS2_INUM_QUANTUM; if (x < y) gfs2_consist_inode(m_ip); - x = cpu_to_be64(x); + z = cpu_to_be64(x); gfs2_trans_add_bh(m_ip->i_gl, m_bh, 1); - *(u64 *)(m_bh->b_data + sizeof(struct gfs2_dinode)) = x; + *(__be64 *)(m_bh->b_data + sizeof(struct gfs2_dinode)) = z; brelse(m_bh); } @@ -567,7 +547,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, return error; /* Don't create entries in an unlinked directory */ - if (!dip->i_di.di_nlink) + if (!dip->i_inode.i_nlink) return -EPERM; error = gfs2_dir_search(&dip->i_inode, name, NULL, NULL); @@ -583,7 +563,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, if (dip->i_di.di_entries == (u32)-1) return -EFBIG; - if (S_ISDIR(mode) && dip->i_di.di_nlink == (u32)-1) + if (S_ISDIR(mode) && dip->i_inode.i_nlink == (u32)-1) return -EMLINK; return 0; @@ -593,24 +573,24 @@ static void munge_mode_uid_gid(struct gfs2_inode *dip, unsigned int *mode, unsigned int *uid, unsigned int *gid) { if (GFS2_SB(&dip->i_inode)->sd_args.ar_suiddir && - (dip->i_di.di_mode & S_ISUID) && dip->i_di.di_uid) { + (dip->i_inode.i_mode & S_ISUID) && dip->i_inode.i_uid) { if (S_ISDIR(*mode)) *mode |= S_ISUID; - else if (dip->i_di.di_uid != current->fsuid) + else if (dip->i_inode.i_uid != current->fsuid) *mode &= ~07111; - *uid = dip->i_di.di_uid; + *uid = dip->i_inode.i_uid; } else *uid = current->fsuid; - if (dip->i_di.di_mode & S_ISGID) { + if (dip->i_inode.i_mode & S_ISGID) { if (S_ISDIR(*mode)) *mode |= S_ISGID; - *gid = dip->i_di.di_gid; + *gid = dip->i_inode.i_gid; } else *gid = current->fsgid; } -static int alloc_dinode(struct gfs2_inode *dip, struct gfs2_inum *inum, +static int alloc_dinode(struct gfs2_inode *dip, struct gfs2_inum_host *inum, u64 *generation) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); @@ -650,9 +630,9 @@ out: */ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - const struct gfs2_inum *inum, unsigned int mode, + const struct gfs2_inum_host *inum, unsigned int mode, unsigned int uid, unsigned int gid, - const u64 *generation) + const u64 *generation, dev_t dev) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_dinode *di; @@ -669,14 +649,15 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, di->di_mode = cpu_to_be32(mode); di->di_uid = cpu_to_be32(uid); di->di_gid = cpu_to_be32(gid); - di->di_nlink = cpu_to_be32(0); - di->di_size = cpu_to_be64(0); + di->di_nlink = 0; + di->di_size = 0; di->di_blocks = cpu_to_be64(1); di->di_atime = di->di_mtime = di->di_ctime = cpu_to_be64(get_seconds()); - di->di_major = di->di_minor = cpu_to_be32(0); + di->di_major = cpu_to_be32(MAJOR(dev)); + di->di_minor = cpu_to_be32(MINOR(dev)); di->di_goal_meta = di->di_goal_data = cpu_to_be64(inum->no_addr); di->di_generation = cpu_to_be64(*generation); - di->di_flags = cpu_to_be32(0); + di->di_flags = 0; if (S_ISREG(mode)) { if ((dip->i_di.di_flags & GFS2_DIF_INHERIT_JDATA) || @@ -693,22 +674,22 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, } di->__pad1 = 0; - di->di_payload_format = cpu_to_be32(0); - di->di_height = cpu_to_be32(0); + di->di_payload_format = cpu_to_be32(S_ISDIR(mode) ? GFS2_FORMAT_DE : 0); + di->di_height = 0; di->__pad2 = 0; di->__pad3 = 0; - di->di_depth = cpu_to_be16(0); - di->di_entries = cpu_to_be32(0); + di->di_depth = 0; + di->di_entries = 0; memset(&di->__pad4, 0, sizeof(di->__pad4)); - di->di_eattr = cpu_to_be64(0); + di->di_eattr = 0; memset(&di->di_reserved, 0, sizeof(di->di_reserved)); brelse(dibh); } static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, - unsigned int mode, const struct gfs2_inum *inum, - const u64 *generation) + unsigned int mode, const struct gfs2_inum_host *inum, + const u64 *generation, dev_t dev) { struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); unsigned int uid, gid; @@ -729,7 +710,7 @@ static int make_dinode(struct gfs2_inode *dip, struct gfs2_glock *gl, if (error) goto out_quota; - init_dinode(dip, gl, inum, mode, uid, gid, generation); + init_dinode(dip, gl, inum, mode, uid, gid, generation, dev); gfs2_quota_change(dip, +1, uid, gid); gfs2_trans_end(sdp); @@ -759,8 +740,7 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, if (alloc_required < 0) goto fail; if (alloc_required) { - error = gfs2_quota_check(dip, dip->i_di.di_uid, - dip->i_di.di_gid); + error = gfs2_quota_check(dip, dip->i_inode.i_uid, dip->i_inode.i_gid); if (error) goto fail_quota_locks; @@ -782,16 +762,16 @@ static int link_dinode(struct gfs2_inode *dip, const struct qstr *name, goto fail_quota_locks; } - error = gfs2_dir_add(&dip->i_inode, name, &ip->i_num, IF2DT(ip->i_di.di_mode)); + error = gfs2_dir_add(&dip->i_inode, name, &ip->i_num, IF2DT(ip->i_inode.i_mode)); if (error) goto fail_end_trans; error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto fail_end_trans; - ip->i_di.di_nlink = 1; + ip->i_inode.i_nlink = 1; gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); return 0; @@ -860,13 +840,13 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip) */ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode) + unsigned int mode, dev_t dev) { struct inode *inode; struct gfs2_inode *dip = ghs->gh_gl->gl_object; struct inode *dir = &dip->i_inode; struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); - struct gfs2_inum inum; + struct gfs2_inum_host inum; int error; u64 generation; @@ -890,35 +870,12 @@ struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, if (error) goto fail_gunlock; - if (inum.no_addr < dip->i_num.no_addr) { - gfs2_glock_dq(ghs); - - error = gfs2_glock_nq_num(sdp, inum.no_addr, - &gfs2_inode_glops, LM_ST_EXCLUSIVE, - GL_SKIP, ghs + 1); - if (error) { - return ERR_PTR(error); - } - - gfs2_holder_reinit(LM_ST_EXCLUSIVE, 0, ghs); - error = gfs2_glock_nq(ghs); - if (error) { - gfs2_glock_dq_uninit(ghs + 1); - return ERR_PTR(error); - } - - error = create_ok(dip, name, mode); - if (error) - goto fail_gunlock2; - } else { - error = gfs2_glock_nq_num(sdp, inum.no_addr, - &gfs2_inode_glops, LM_ST_EXCLUSIVE, - GL_SKIP, ghs + 1); - if (error) - goto fail_gunlock; - } + error = gfs2_glock_nq_num(sdp, inum.no_addr, &gfs2_inode_glops, + LM_ST_EXCLUSIVE, GL_SKIP, ghs + 1); + if (error) + goto fail_gunlock; - error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation); + error = make_dinode(dip, ghs[1].gh_gl, mode, &inum, &generation, dev); if (error) goto fail_gunlock2; @@ -975,7 +932,7 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, if (ip->i_di.di_entries != 2) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); return -EIO; } @@ -997,7 +954,12 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, if (error) return error; - error = gfs2_change_nlink(ip, -2); + /* It looks odd, but it really should be done twice */ + error = gfs2_change_nlink(ip, -1); + if (error) + return error; + + error = gfs2_change_nlink(ip, -1); if (error) return error; @@ -1018,16 +980,16 @@ int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip) { - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error; if (IS_IMMUTABLE(&ip->i_inode) || IS_APPEND(&ip->i_inode)) return -EPERM; - if ((dip->i_di.di_mode & S_ISVTX) && - dip->i_di.di_uid != current->fsuid && - ip->i_di.di_uid != current->fsuid && !capable(CAP_FOWNER)) + if ((dip->i_inode.i_mode & S_ISVTX) && + dip->i_inode.i_uid != current->fsuid && + ip->i_inode.i_uid != current->fsuid && !capable(CAP_FOWNER)) return -EPERM; if (IS_APPEND(&dip->i_inode)) @@ -1044,7 +1006,7 @@ int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, if (!gfs2_inum_equal(&inum, &ip->i_num)) return -ENOENT; - if (IF2DT(ip->i_di.di_mode) != type) { + if (IF2DT(ip->i_inode.i_mode) != type) { gfs2_consist_inode(dip); return -EIO; } @@ -1194,7 +1156,7 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) return 0; curtime = get_seconds(); - if (curtime - ip->i_di.di_atime >= quantum) { + if (curtime - ip->i_inode.i_atime.tv_sec >= quantum) { gfs2_glock_dq(gh); gfs2_holder_reinit(LM_ST_EXCLUSIVE, gh->gh_flags & ~LM_FLAG_ANY, gh); @@ -1206,7 +1168,7 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) trying to get exclusive lock. */ curtime = get_seconds(); - if (curtime - ip->i_di.di_atime >= quantum) { + if (curtime - ip->i_inode.i_atime.tv_sec >= quantum) { struct buffer_head *dibh; struct gfs2_dinode *di; @@ -1220,11 +1182,11 @@ int gfs2_glock_nq_atime(struct gfs2_holder *gh) if (error) goto fail_end_trans; - ip->i_di.di_atime = curtime; + ip->i_inode.i_atime.tv_sec = curtime; gfs2_trans_add_bh(ip->i_gl, dibh, 1); di = (struct gfs2_dinode *)dibh->b_data; - di->di_atime = cpu_to_be64(ip->i_di.di_atime); + di->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); brelse(dibh); gfs2_trans_end(sdp); @@ -1249,92 +1211,6 @@ fail: return error; } -/** - * glock_compare_atime - Compare two struct gfs2_glock structures for sort - * @arg_a: the first structure - * @arg_b: the second structure - * - * Returns: 1 if A > B - * -1 if A < B - * 0 if A == B - */ - -static int glock_compare_atime(const void *arg_a, const void *arg_b) -{ - const struct gfs2_holder *gh_a = *(const struct gfs2_holder **)arg_a; - const struct gfs2_holder *gh_b = *(const struct gfs2_holder **)arg_b; - const struct lm_lockname *a = &gh_a->gh_gl->gl_name; - const struct lm_lockname *b = &gh_b->gh_gl->gl_name; - - if (a->ln_number > b->ln_number) - return 1; - if (a->ln_number < b->ln_number) - return -1; - if (gh_a->gh_state == LM_ST_SHARED && gh_b->gh_state == LM_ST_EXCLUSIVE) - return 1; - if (gh_a->gh_state == LM_ST_SHARED && (gh_b->gh_flags & GL_ATIME)) - return 1; - - return 0; -} - -/** - * gfs2_glock_nq_m_atime - acquire multiple glocks where one may need an - * atime update - * @num_gh: the number of structures - * @ghs: an array of struct gfs2_holder structures - * - * Returns: 0 on success (all glocks acquired), - * errno on failure (no glocks acquired) - */ - -int gfs2_glock_nq_m_atime(unsigned int num_gh, struct gfs2_holder *ghs) -{ - struct gfs2_holder **p; - unsigned int x; - int error = 0; - - if (!num_gh) - return 0; - - if (num_gh == 1) { - ghs->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC); - if (ghs->gh_flags & GL_ATIME) - error = gfs2_glock_nq_atime(ghs); - else - error = gfs2_glock_nq(ghs); - return error; - } - - p = kcalloc(num_gh, sizeof(struct gfs2_holder *), GFP_KERNEL); - if (!p) - return -ENOMEM; - - for (x = 0; x < num_gh; x++) - p[x] = &ghs[x]; - - sort(p, num_gh, sizeof(struct gfs2_holder *), glock_compare_atime,NULL); - - for (x = 0; x < num_gh; x++) { - p[x]->gh_flags &= ~(LM_FLAG_TRY | GL_ASYNC); - - if (p[x]->gh_flags & GL_ATIME) - error = gfs2_glock_nq_atime(p[x]); - else - error = gfs2_glock_nq(p[x]); - - if (error) { - while (x--) - gfs2_glock_dq(p[x]); - break; - } - } - - kfree(p); - return error; -} - - static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) { @@ -1345,10 +1221,8 @@ __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) if (!error) { error = inode_setattr(&ip->i_inode, attr); gfs2_assert_warn(GFS2_SB(&ip->i_inode), !error); - gfs2_inode_attr_out(ip); - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } return error; diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index f5d8617..b57f448 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -22,13 +22,19 @@ static inline int gfs2_is_jdata(struct gfs2_inode *ip) static inline int gfs2_is_dir(struct gfs2_inode *ip) { - return S_ISDIR(ip->i_di.di_mode); + return S_ISDIR(ip->i_inode.i_mode); +} + +static inline void gfs2_set_inode_blocks(struct inode *inode) +{ + struct gfs2_inode *ip = GFS2_I(inode); + inode->i_blocks = ip->i_di.di_blocks << + (GFS2_SB(inode)->sd_sb.sb_bsize_shift - GFS2_BASIC_BLOCK_SHIFT); } void gfs2_inode_attr_in(struct gfs2_inode *ip); -void gfs2_inode_attr_out(struct gfs2_inode *ip); -struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum *inum, unsigned type); -struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum *inum); +struct inode *gfs2_inode_lookup(struct super_block *sb, struct gfs2_inum_host *inum, unsigned type); +struct inode *gfs2_ilookup(struct super_block *sb, struct gfs2_inum_host *inum); int gfs2_inode_refresh(struct gfs2_inode *ip); @@ -37,19 +43,15 @@ int gfs2_change_nlink(struct gfs2_inode *ip, int diff); struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, int is_root, struct nameidata *nd); struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, - unsigned int mode); + unsigned int mode, dev_t dev); int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip); int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, struct gfs2_inode *ip); int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to); int gfs2_readlinki(struct gfs2_inode *ip, char **buf, unsigned int *len); - int gfs2_glock_nq_atime(struct gfs2_holder *gh); -int gfs2_glock_nq_m_atime(unsigned int num_gh, struct gfs2_holder *ghs); - int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); - struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); #endif /* __INODE_DOT_H__ */ diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 0cace3d..291415d 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -15,6 +15,7 @@ #include <linux/gfs2_ondisk.h> #include <linux/crc32.h> #include <linux/lm_interface.h> +#include <linux/delay.h> #include "gfs2.h" #include "incore.h" @@ -142,7 +143,7 @@ static int gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_ail *ai, int fl return list_empty(&ai->ai_ail1_list); } -void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags) +static void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags) { struct list_head *head = &sdp->sd_ail1_list; u64 sync_gen; @@ -261,6 +262,12 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail) * @sdp: The GFS2 superblock * @blks: The number of blocks to reserve * + * Note that we never give out the last 6 blocks of the journal. Thats + * due to the fact that there is are a small number of header blocks + * associated with each log flush. The exact number can't be known until + * flush time, so we ensure that we have just enough free blocks at all + * times to avoid running out during a log flush. + * * Returns: errno */ @@ -274,7 +281,7 @@ int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks) mutex_lock(&sdp->sd_log_reserve_mutex); gfs2_log_lock(sdp); - while(sdp->sd_log_blks_free <= blks) { + while(sdp->sd_log_blks_free <= (blks + 6)) { gfs2_log_unlock(sdp); gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL); @@ -319,7 +326,8 @@ static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn) bh_map.b_size = 1 << inode->i_blkbits; error = gfs2_block_map(inode, lbn, 0, &bh_map); if (error || !bh_map.b_blocknr) - printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, bh_map.b_blocknr, lbn); + printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, + (unsigned long long)bh_map.b_blocknr, lbn); gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr); return bh_map.b_blocknr; @@ -643,12 +651,9 @@ void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) up_read(&sdp->sd_log_flush_lock); gfs2_log_lock(sdp); - if (sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks)) { - gfs2_log_unlock(sdp); - gfs2_log_flush(sdp, NULL); - } else { - gfs2_log_unlock(sdp); - } + if (sdp->sd_log_num_buf > gfs2_tune_get(sdp, gt_incore_log_blocks)) + wake_up_process(sdp->sd_logd_process); + gfs2_log_unlock(sdp); } /** @@ -686,3 +691,21 @@ void gfs2_log_shutdown(struct gfs2_sbd *sdp) up_write(&sdp->sd_log_flush_lock); } + +/** + * gfs2_meta_syncfs - sync all the buffers in a filesystem + * @sdp: the filesystem + * + */ + +void gfs2_meta_syncfs(struct gfs2_sbd *sdp) +{ + gfs2_log_flush(sdp, NULL); + for (;;) { + gfs2_ail1_start(sdp, DIO_ALL); + if (gfs2_ail1_empty(sdp, DIO_ALL)) + break; + msleep(10); + } +} + diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 7f5737d..8e7aa0f 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -48,7 +48,6 @@ static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp, unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct, unsigned int ssize); -void gfs2_ail1_start(struct gfs2_sbd *sdp, int flags); int gfs2_ail1_empty(struct gfs2_sbd *sdp, int flags); int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks); @@ -61,5 +60,6 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl); void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans); void gfs2_log_shutdown(struct gfs2_sbd *sdp); +void gfs2_meta_syncfs(struct gfs2_sbd *sdp); #endif /* __LOG_DOT_H__ */ diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index ab6d111..4d7f94d 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -182,7 +182,7 @@ static void buf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) } static void buf_lo_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass) + struct gfs2_log_header_host *head, int pass) { struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); @@ -328,7 +328,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) } static void revoke_lo_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, int pass) + struct gfs2_log_header_host *head, int pass) { struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); @@ -509,7 +509,7 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) { LIST_HEAD(started); struct gfs2_bufdata *bd1 = NULL, *bd2, *bdt; - struct buffer_head *bh = NULL; + struct buffer_head *bh = NULL,*bh1 = NULL; unsigned int offset = sizeof(struct gfs2_log_descriptor); struct gfs2_log_descriptor *ld; unsigned int limit; @@ -537,8 +537,13 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) list_for_each_entry_safe_continue(bd1, bdt, &sdp->sd_log_le_databuf, bd_le.le_list) { + /* store off the buffer head in a local ptr since + * gfs2_bufdata might change when we drop the log lock + */ + bh1 = bd1->bd_bh; + /* An ordered write buffer */ - if (bd1->bd_bh && !buffer_pinned(bd1->bd_bh)) { + if (bh1 && !buffer_pinned(bh1)) { list_move(&bd1->bd_le.le_list, &started); if (bd1 == bd2) { bd2 = NULL; @@ -547,20 +552,21 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) bd_le.le_list); } total_dbuf--; - if (bd1->bd_bh) { - get_bh(bd1->bd_bh); - if (buffer_dirty(bd1->bd_bh)) { + if (bh1) { + if (buffer_dirty(bh1)) { + get_bh(bh1); + gfs2_log_unlock(sdp); - wait_on_buffer(bd1->bd_bh); - ll_rw_block(WRITE, 1, - &bd1->bd_bh); + + ll_rw_block(SWRITE, 1, &bh1); + brelse(bh1); + gfs2_log_lock(sdp); } - brelse(bd1->bd_bh); continue; } continue; - } else if (bd1->bd_bh) { /* A journaled buffer */ + } else if (bh1) { /* A journaled buffer */ int magic; gfs2_log_unlock(sdp); if (!bh) { @@ -582,16 +588,16 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) ld->ld_data2 = cpu_to_be32(0); memset(ld->ld_reserved, 0, sizeof(ld->ld_reserved)); } - magic = gfs2_check_magic(bd1->bd_bh); - *ptr++ = cpu_to_be64(bd1->bd_bh->b_blocknr); + magic = gfs2_check_magic(bh1); + *ptr++ = cpu_to_be64(bh1->b_blocknr); *ptr++ = cpu_to_be64((__u64)magic); - clear_buffer_escaped(bd1->bd_bh); + clear_buffer_escaped(bh1); if (unlikely(magic != 0)) - set_buffer_escaped(bd1->bd_bh); + set_buffer_escaped(bh1); gfs2_log_lock(sdp); if (n++ > num) break; - } else if (!bd1->bd_bh) { + } else if (!bh1) { total_dbuf--; sdp->sd_log_num_databuf--; list_del_init(&bd1->bd_le.le_list); diff --git a/fs/gfs2/lops.h b/fs/gfs2/lops.h index 5839c05..965bc65 100644 --- a/fs/gfs2/lops.h +++ b/fs/gfs2/lops.h @@ -60,7 +60,7 @@ static inline void lops_after_commit(struct gfs2_sbd *sdp, struct gfs2_ail *ai) } static inline void lops_before_scan(struct gfs2_jdesc *jd, - struct gfs2_log_header *head, + struct gfs2_log_header_host *head, unsigned int pass) { int x; diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index 9889c1e..7c1a9e2 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -25,7 +25,7 @@ #include "util.h" #include "glock.h" -static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_inode_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_inode *ip = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == @@ -37,7 +37,7 @@ static void gfs2_init_inode_once(void *foo, kmem_cache_t *cachep, unsigned long } } -static void gfs2_init_glock_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void gfs2_init_glock_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct gfs2_glock *gl = foo; if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 3912d6a..0e34d99 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -127,17 +127,17 @@ void gfs2_meta_sync(struct gfs2_glock *gl) /** * getbuf - Get a buffer with a given address space - * @sdp: the filesystem - * @aspace: the address space + * @gl: the glock * @blkno: the block number (filesystem scope) * @create: 1 if the buffer should be created * * Returns: the buffer */ -static struct buffer_head *getbuf(struct gfs2_sbd *sdp, struct inode *aspace, - u64 blkno, int create) +static struct buffer_head *getbuf(struct gfs2_glock *gl, u64 blkno, int create) { + struct address_space *mapping = gl->gl_aspace->i_mapping; + struct gfs2_sbd *sdp = gl->gl_sbd; struct page *page; struct buffer_head *bh; unsigned int shift; @@ -150,13 +150,13 @@ static struct buffer_head *getbuf(struct gfs2_sbd *sdp, struct inode *aspace, if (create) { for (;;) { - page = grab_cache_page(aspace->i_mapping, index); + page = grab_cache_page(mapping, index); if (page) break; yield(); } } else { - page = find_lock_page(aspace->i_mapping, index); + page = find_lock_page(mapping, index); if (!page) return NULL; } @@ -202,7 +202,7 @@ static void meta_prep_new(struct buffer_head *bh) struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno) { struct buffer_head *bh; - bh = getbuf(gl->gl_sbd, gl->gl_aspace, blkno, CREATE); + bh = getbuf(gl, blkno, CREATE); meta_prep_new(bh); return bh; } @@ -220,7 +220,7 @@ struct buffer_head *gfs2_meta_new(struct gfs2_glock *gl, u64 blkno) int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags, struct buffer_head **bhp) { - *bhp = getbuf(gl->gl_sbd, gl->gl_aspace, blkno, CREATE); + *bhp = getbuf(gl, blkno, CREATE); if (!buffer_uptodate(*bhp)) ll_rw_block(READ_META, 1, bhp); if (flags & DIO_WAIT) { @@ -379,11 +379,10 @@ void gfs2_unpin(struct gfs2_sbd *sdp, struct buffer_head *bh, void gfs2_meta_wipe(struct gfs2_inode *ip, u64 bstart, u32 blen) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct inode *aspace = ip->i_gl->gl_aspace; struct buffer_head *bh; while (blen) { - bh = getbuf(sdp, aspace, bstart, NO_CREATE); + bh = getbuf(ip->i_gl, bstart, NO_CREATE); if (bh) { struct gfs2_bufdata *bd = bh->b_private; @@ -472,6 +471,9 @@ int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, struct buffer_head *bh = NULL, **bh_slot = ip->i_cache + height; int in_cache = 0; + BUG_ON(!gl); + BUG_ON(!sdp); + spin_lock(&ip->i_spin); if (*bh_slot && (*bh_slot)->b_blocknr == num) { bh = *bh_slot; @@ -481,7 +483,7 @@ int gfs2_meta_indirect_buffer(struct gfs2_inode *ip, int height, u64 num, spin_unlock(&ip->i_spin); if (!bh) - bh = getbuf(gl->gl_sbd, gl->gl_aspace, num, CREATE); + bh = getbuf(gl, num, CREATE); if (!bh) return -ENOBUFS; @@ -532,7 +534,6 @@ err: struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) { struct gfs2_sbd *sdp = gl->gl_sbd; - struct inode *aspace = gl->gl_aspace; struct buffer_head *first_bh, *bh; u32 max_ra = gfs2_tune_get(sdp, gt_max_readahead) >> sdp->sd_sb.sb_bsize_shift; @@ -544,7 +545,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) if (extlen > max_ra) extlen = max_ra; - first_bh = getbuf(sdp, aspace, dblock, CREATE); + first_bh = getbuf(gl, dblock, CREATE); if (buffer_uptodate(first_bh)) goto out; @@ -555,7 +556,7 @@ struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen) extlen--; while (extlen) { - bh = getbuf(sdp, aspace, dblock, CREATE); + bh = getbuf(gl, dblock, CREATE); if (!buffer_uptodate(bh) && !buffer_locked(bh)) ll_rw_block(READA, 1, &bh); @@ -571,20 +572,3 @@ out: return first_bh; } -/** - * gfs2_meta_syncfs - sync all the buffers in a filesystem - * @sdp: the filesystem - * - */ - -void gfs2_meta_syncfs(struct gfs2_sbd *sdp) -{ - gfs2_log_flush(sdp, NULL); - for (;;) { - gfs2_ail1_start(sdp, DIO_ALL); - if (gfs2_ail1_empty(sdp, DIO_ALL)) - break; - msleep(10); - } -} - diff --git a/fs/gfs2/meta_io.h b/fs/gfs2/meta_io.h index 3ec939e..e037425 100644 --- a/fs/gfs2/meta_io.h +++ b/fs/gfs2/meta_io.h @@ -67,7 +67,6 @@ static inline int gfs2_meta_inode_buffer(struct gfs2_inode *ip, } struct buffer_head *gfs2_meta_ra(struct gfs2_glock *gl, u64 dblock, u32 extlen); -void gfs2_meta_syncfs(struct gfs2_sbd *sdp); #define buffer_busy(bh) \ ((bh)->b_state & ((1ul << BH_Dirty) | (1ul << BH_Lock) | (1ul << BH_Pinned))) diff --git a/fs/gfs2/ondisk.c b/fs/gfs2/ondisk.c index 1025960..f2495f1 100644 --- a/fs/gfs2/ondisk.c +++ b/fs/gfs2/ondisk.c @@ -15,6 +15,8 @@ #include "gfs2.h" #include <linux/gfs2_ondisk.h> +#include <linux/lm_interface.h> +#include "incore.h" #define pv(struct, member, fmt) printk(KERN_INFO " "#member" = "fmt"\n", \ struct->member); @@ -32,7 +34,7 @@ * first arg: the cpu-order structure */ -void gfs2_inum_in(struct gfs2_inum *no, const void *buf) +void gfs2_inum_in(struct gfs2_inum_host *no, const void *buf) { const struct gfs2_inum *str = buf; @@ -40,7 +42,7 @@ void gfs2_inum_in(struct gfs2_inum *no, const void *buf) no->no_addr = be64_to_cpu(str->no_addr); } -void gfs2_inum_out(const struct gfs2_inum *no, void *buf) +void gfs2_inum_out(const struct gfs2_inum_host *no, void *buf) { struct gfs2_inum *str = buf; @@ -48,13 +50,13 @@ void gfs2_inum_out(const struct gfs2_inum *no, void *buf) str->no_addr = cpu_to_be64(no->no_addr); } -static void gfs2_inum_print(const struct gfs2_inum *no) +static void gfs2_inum_print(const struct gfs2_inum_host *no) { printk(KERN_INFO " no_formal_ino = %llu\n", (unsigned long long)no->no_formal_ino); printk(KERN_INFO " no_addr = %llu\n", (unsigned long long)no->no_addr); } -static void gfs2_meta_header_in(struct gfs2_meta_header *mh, const void *buf) +static void gfs2_meta_header_in(struct gfs2_meta_header_host *mh, const void *buf) { const struct gfs2_meta_header *str = buf; @@ -63,23 +65,7 @@ static void gfs2_meta_header_in(struct gfs2_meta_header *mh, const void *buf) mh->mh_format = be32_to_cpu(str->mh_format); } -static void gfs2_meta_header_out(const struct gfs2_meta_header *mh, void *buf) -{ - struct gfs2_meta_header *str = buf; - - str->mh_magic = cpu_to_be32(mh->mh_magic); - str->mh_type = cpu_to_be32(mh->mh_type); - str->mh_format = cpu_to_be32(mh->mh_format); -} - -static void gfs2_meta_header_print(const struct gfs2_meta_header *mh) -{ - pv(mh, mh_magic, "0x%.8X"); - pv(mh, mh_type, "%u"); - pv(mh, mh_format, "%u"); -} - -void gfs2_sb_in(struct gfs2_sb *sb, const void *buf) +void gfs2_sb_in(struct gfs2_sb_host *sb, const void *buf) { const struct gfs2_sb *str = buf; @@ -97,7 +83,7 @@ void gfs2_sb_in(struct gfs2_sb *sb, const void *buf) memcpy(sb->sb_locktable, str->sb_locktable, GFS2_LOCKNAME_LEN); } -void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf) +void gfs2_rindex_in(struct gfs2_rindex_host *ri, const void *buf) { const struct gfs2_rindex *str = buf; @@ -109,7 +95,7 @@ void gfs2_rindex_in(struct gfs2_rindex *ri, const void *buf) } -void gfs2_rindex_print(const struct gfs2_rindex *ri) +void gfs2_rindex_print(const struct gfs2_rindex_host *ri) { printk(KERN_INFO " ri_addr = %llu\n", (unsigned long long)ri->ri_addr); pv(ri, ri_length, "%u"); @@ -120,22 +106,20 @@ void gfs2_rindex_print(const struct gfs2_rindex *ri) pv(ri, ri_bitbytes, "%u"); } -void gfs2_rgrp_in(struct gfs2_rgrp *rg, const void *buf) +void gfs2_rgrp_in(struct gfs2_rgrp_host *rg, const void *buf) { const struct gfs2_rgrp *str = buf; - gfs2_meta_header_in(&rg->rg_header, buf); rg->rg_flags = be32_to_cpu(str->rg_flags); rg->rg_free = be32_to_cpu(str->rg_free); rg->rg_dinodes = be32_to_cpu(str->rg_dinodes); rg->rg_igeneration = be64_to_cpu(str->rg_igeneration); } -void gfs2_rgrp_out(const struct gfs2_rgrp *rg, void *buf) +void gfs2_rgrp_out(const struct gfs2_rgrp_host *rg, void *buf) { struct gfs2_rgrp *str = buf; - gfs2_meta_header_out(&rg->rg_header, buf); str->rg_flags = cpu_to_be32(rg->rg_flags); str->rg_free = cpu_to_be32(rg->rg_free); str->rg_dinodes = cpu_to_be32(rg->rg_dinodes); @@ -144,7 +128,7 @@ void gfs2_rgrp_out(const struct gfs2_rgrp *rg, void *buf) memset(&str->rg_reserved, 0, sizeof(str->rg_reserved)); } -void gfs2_quota_in(struct gfs2_quota *qu, const void *buf) +void gfs2_quota_in(struct gfs2_quota_host *qu, const void *buf) { const struct gfs2_quota *str = buf; @@ -153,96 +137,56 @@ void gfs2_quota_in(struct gfs2_quota *qu, const void *buf) qu->qu_value = be64_to_cpu(str->qu_value); } -void gfs2_dinode_in(struct gfs2_dinode *di, const void *buf) -{ - const struct gfs2_dinode *str = buf; - - gfs2_meta_header_in(&di->di_header, buf); - gfs2_inum_in(&di->di_num, &str->di_num); - - di->di_mode = be32_to_cpu(str->di_mode); - di->di_uid = be32_to_cpu(str->di_uid); - di->di_gid = be32_to_cpu(str->di_gid); - di->di_nlink = be32_to_cpu(str->di_nlink); - di->di_size = be64_to_cpu(str->di_size); - di->di_blocks = be64_to_cpu(str->di_blocks); - di->di_atime = be64_to_cpu(str->di_atime); - di->di_mtime = be64_to_cpu(str->di_mtime); - di->di_ctime = be64_to_cpu(str->di_ctime); - di->di_major = be32_to_cpu(str->di_major); - di->di_minor = be32_to_cpu(str->di_minor); - - di->di_goal_meta = be64_to_cpu(str->di_goal_meta); - di->di_goal_data = be64_to_cpu(str->di_goal_data); - di->di_generation = be64_to_cpu(str->di_generation); - - di->di_flags = be32_to_cpu(str->di_flags); - di->di_payload_format = be32_to_cpu(str->di_payload_format); - di->di_height = be16_to_cpu(str->di_height); - - di->di_depth = be16_to_cpu(str->di_depth); - di->di_entries = be32_to_cpu(str->di_entries); - - di->di_eattr = be64_to_cpu(str->di_eattr); - -} - -void gfs2_dinode_out(const struct gfs2_dinode *di, void *buf) +void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf) { + const struct gfs2_dinode_host *di = &ip->i_di; struct gfs2_dinode *str = buf; - gfs2_meta_header_out(&di->di_header, buf); - gfs2_inum_out(&di->di_num, (char *)&str->di_num); + str->di_header.mh_magic = cpu_to_be32(GFS2_MAGIC); + str->di_header.mh_type = cpu_to_be32(GFS2_METATYPE_DI); + str->di_header.__pad0 = 0; + str->di_header.mh_format = cpu_to_be32(GFS2_FORMAT_DI); + str->di_header.__pad1 = 0; - str->di_mode = cpu_to_be32(di->di_mode); - str->di_uid = cpu_to_be32(di->di_uid); - str->di_gid = cpu_to_be32(di->di_gid); - str->di_nlink = cpu_to_be32(di->di_nlink); + gfs2_inum_out(&ip->i_num, &str->di_num); + + str->di_mode = cpu_to_be32(ip->i_inode.i_mode); + str->di_uid = cpu_to_be32(ip->i_inode.i_uid); + str->di_gid = cpu_to_be32(ip->i_inode.i_gid); + str->di_nlink = cpu_to_be32(ip->i_inode.i_nlink); str->di_size = cpu_to_be64(di->di_size); str->di_blocks = cpu_to_be64(di->di_blocks); - str->di_atime = cpu_to_be64(di->di_atime); - str->di_mtime = cpu_to_be64(di->di_mtime); - str->di_ctime = cpu_to_be64(di->di_ctime); - str->di_major = cpu_to_be32(di->di_major); - str->di_minor = cpu_to_be32(di->di_minor); + str->di_atime = cpu_to_be64(ip->i_inode.i_atime.tv_sec); + str->di_mtime = cpu_to_be64(ip->i_inode.i_mtime.tv_sec); + str->di_ctime = cpu_to_be64(ip->i_inode.i_ctime.tv_sec); str->di_goal_meta = cpu_to_be64(di->di_goal_meta); str->di_goal_data = cpu_to_be64(di->di_goal_data); str->di_generation = cpu_to_be64(di->di_generation); str->di_flags = cpu_to_be32(di->di_flags); - str->di_payload_format = cpu_to_be32(di->di_payload_format); str->di_height = cpu_to_be16(di->di_height); - + str->di_payload_format = cpu_to_be32(S_ISDIR(ip->i_inode.i_mode) && + !(ip->i_di.di_flags & GFS2_DIF_EXHASH) ? + GFS2_FORMAT_DE : 0); str->di_depth = cpu_to_be16(di->di_depth); str->di_entries = cpu_to_be32(di->di_entries); str->di_eattr = cpu_to_be64(di->di_eattr); - } -void gfs2_dinode_print(const struct gfs2_dinode *di) +void gfs2_dinode_print(const struct gfs2_inode *ip) { - gfs2_meta_header_print(&di->di_header); - gfs2_inum_print(&di->di_num); + const struct gfs2_dinode_host *di = &ip->i_di; + + gfs2_inum_print(&ip->i_num); - pv(di, di_mode, "0%o"); - pv(di, di_uid, "%u"); - pv(di, di_gid, "%u"); - pv(di, di_nlink, "%u"); printk(KERN_INFO " di_size = %llu\n", (unsigned long long)di->di_size); printk(KERN_INFO " di_blocks = %llu\n", (unsigned long long)di->di_blocks); - printk(KERN_INFO " di_atime = %lld\n", (long long)di->di_atime); - printk(KERN_INFO " di_mtime = %lld\n", (long long)di->di_mtime); - printk(KERN_INFO " di_ctime = %lld\n", (long long)di->di_ctime); - pv(di, di_major, "%u"); - pv(di, di_minor, "%u"); - printk(KERN_INFO " di_goal_meta = %llu\n", (unsigned long long)di->di_goal_meta); printk(KERN_INFO " di_goal_data = %llu\n", (unsigned long long)di->di_goal_data); pv(di, di_flags, "0x%.8X"); - pv(di, di_payload_format, "%u"); pv(di, di_height, "%u"); pv(di, di_depth, "%u"); @@ -251,7 +195,7 @@ void gfs2_dinode_print(const struct gfs2_dinode *di) printk(KERN_INFO " di_eattr = %llu\n", (unsigned long long)di->di_eattr); } -void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf) +void gfs2_log_header_in(struct gfs2_log_header_host *lh, const void *buf) { const struct gfs2_log_header *str = buf; @@ -263,7 +207,7 @@ void gfs2_log_header_in(struct gfs2_log_header *lh, const void *buf) lh->lh_hash = be32_to_cpu(str->lh_hash); } -void gfs2_inum_range_in(struct gfs2_inum_range *ir, const void *buf) +void gfs2_inum_range_in(struct gfs2_inum_range_host *ir, const void *buf) { const struct gfs2_inum_range *str = buf; @@ -271,7 +215,7 @@ void gfs2_inum_range_in(struct gfs2_inum_range *ir, const void *buf) ir->ir_length = be64_to_cpu(str->ir_length); } -void gfs2_inum_range_out(const struct gfs2_inum_range *ir, void *buf) +void gfs2_inum_range_out(const struct gfs2_inum_range_host *ir, void *buf) { struct gfs2_inum_range *str = buf; @@ -279,7 +223,7 @@ void gfs2_inum_range_out(const struct gfs2_inum_range *ir, void *buf) str->ir_length = cpu_to_be64(ir->ir_length); } -void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf) +void gfs2_statfs_change_in(struct gfs2_statfs_change_host *sc, const void *buf) { const struct gfs2_statfs_change *str = buf; @@ -288,7 +232,7 @@ void gfs2_statfs_change_in(struct gfs2_statfs_change *sc, const void *buf) sc->sc_dinodes = be64_to_cpu(str->sc_dinodes); } -void gfs2_statfs_change_out(const struct gfs2_statfs_change *sc, void *buf) +void gfs2_statfs_change_out(const struct gfs2_statfs_change_host *sc, void *buf) { struct gfs2_statfs_change *str = buf; @@ -297,7 +241,7 @@ void gfs2_statfs_change_out(const struct gfs2_statfs_change *sc, void *buf) str->sc_dinodes = cpu_to_be64(sc->sc_dinodes); } -void gfs2_quota_change_in(struct gfs2_quota_change *qc, const void *buf) +void gfs2_quota_change_in(struct gfs2_quota_change_host *qc, const void *buf) { const struct gfs2_quota_change *str = buf; diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 015640b..d8d69a7 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -156,19 +156,6 @@ out_ignore: return 0; } -static int zero_readpage(struct page *page) -{ - void *kaddr; - - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr, 0, PAGE_CACHE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - - SetPageUptodate(page); - - return 0; -} - /** * stuffed_readpage - Fill in a Linux page with stuffed file data * @ip: the inode @@ -183,9 +170,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) void *kaddr; int error; - /* Only the first page of a stuffed file might contain data */ - if (unlikely(page->index)) - return zero_readpage(page); + BUG_ON(page->index); error = gfs2_meta_inode_buffer(ip, &dibh); if (error) @@ -230,9 +215,9 @@ static int gfs2_readpage(struct file *file, struct page *page) /* gfs2_sharewrite_nopage has grabbed the ip->i_gl already */ goto skip_lock; } - gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|GL_AOP, &gh); + gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME|LM_FLAG_TRY_1CB, &gh); do_unlock = 1; - error = gfs2_glock_nq_m_atime(1, &gh); + error = gfs2_glock_nq_atime(&gh); if (unlikely(error)) goto out_unlock; } @@ -254,6 +239,8 @@ skip_lock: out: return error; out_unlock: + if (error == GLR_TRYFAILED) + error = AOP_TRUNCATED_PAGE; unlock_page(page); if (do_unlock) gfs2_holder_uninit(&gh); @@ -293,9 +280,9 @@ static int gfs2_readpages(struct file *file, struct address_space *mapping, goto skip_lock; } gfs2_holder_init(ip->i_gl, LM_ST_SHARED, - LM_FLAG_TRY_1CB|GL_ATIME|GL_AOP, &gh); + LM_FLAG_TRY_1CB|GL_ATIME, &gh); do_unlock = 1; - ret = gfs2_glock_nq_m_atime(1, &gh); + ret = gfs2_glock_nq_atime(&gh); if (ret == GLR_TRYFAILED) goto out_noerror; if (unlikely(ret)) @@ -366,10 +353,13 @@ static int gfs2_prepare_write(struct file *file, struct page *page, unsigned int write_len = to - from; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|GL_AOP, &ip->i_gh); - error = gfs2_glock_nq_m_atime(1, &ip->i_gh); - if (error) + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|LM_FLAG_TRY_1CB, &ip->i_gh); + error = gfs2_glock_nq_atime(&ip->i_gh); + if (unlikely(error)) { + if (error == GLR_TRYFAILED) + error = AOP_TRUNCATED_PAGE; goto out_uninit; + } gfs2_write_calc_reserv(ip, write_len, &data_blocks, &ind_blocks); @@ -386,7 +376,7 @@ static int gfs2_prepare_write(struct file *file, struct page *page, if (error) goto out_alloc_put; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_qunlock; @@ -482,8 +472,10 @@ static int gfs2_commit_write(struct file *file, struct page *page, SetPageUptodate(page); - if (inode->i_size < file_size) + if (inode->i_size < file_size) { i_size_write(inode, file_size); + mark_inode_dirty(inode); + } } else { if (sdp->sd_args.ar_data == GFS2_DATA_ORDERED || gfs2_is_jdata(ip)) @@ -498,11 +490,6 @@ static int gfs2_commit_write(struct file *file, struct page *page, di->di_size = cpu_to_be64(inode->i_size); } - di->di_mode = cpu_to_be32(inode->i_mode); - di->di_atime = cpu_to_be64(inode->i_atime.tv_sec); - di->di_mtime = cpu_to_be64(inode->i_mtime.tv_sec); - di->di_ctime = cpu_to_be64(inode->i_ctime.tv_sec); - brelse(dibh); gfs2_trans_end(sdp); if (al->al_requested) { @@ -624,7 +611,7 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, * on this path. All we need change is atime. */ gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - rv = gfs2_glock_nq_m_atime(1, &gh); + rv = gfs2_glock_nq_atime(&gh); if (rv) goto out; @@ -737,6 +724,9 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) if (!atomic_read(&aspace->i_writecount)) return 0; + if (!(gfp_mask & __GFP_WAIT)) + return 0; + if (time_after_eq(jiffies, t)) { stuck_releasepage(bh); /* should we withdraw here? */ diff --git a/fs/gfs2/ops_dentry.c b/fs/gfs2/ops_dentry.c index 00041b1..d355899 100644 --- a/fs/gfs2/ops_dentry.c +++ b/fs/gfs2/ops_dentry.c @@ -43,7 +43,7 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd) struct inode *inode = dentry->d_inode; struct gfs2_holder d_gh; struct gfs2_inode *ip; - struct gfs2_inum inum; + struct gfs2_inum_host inum; unsigned int type; int error; @@ -76,7 +76,7 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd) if (!gfs2_inum_equal(&ip->i_num, &inum)) goto invalid_gunlock; - if (IF2DT(ip->i_di.di_mode) != type) { + if (IF2DT(ip->i_inode.i_mode) != type) { gfs2_consist_inode(dip); goto fail_gunlock; } diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c index 86127d9..b4e7b87 100644 --- a/fs/gfs2/ops_export.c +++ b/fs/gfs2/ops_export.c @@ -27,15 +27,16 @@ #include "util.h" static struct dentry *gfs2_decode_fh(struct super_block *sb, - __u32 *fh, + __u32 *p, int fh_len, int fh_type, int (*acceptable)(void *context, struct dentry *dentry), void *context) { + __be32 *fh = (__force __be32 *)p; struct gfs2_fh_obj fh_obj; - struct gfs2_inum *this, parent; + struct gfs2_inum_host *this, parent; if (fh_type != fh_len) return NULL; @@ -65,9 +66,10 @@ static struct dentry *gfs2_decode_fh(struct super_block *sb, acceptable, context); } -static int gfs2_encode_fh(struct dentry *dentry, __u32 *fh, int *len, +static int gfs2_encode_fh(struct dentry *dentry, __u32 *p, int *len, int connectable) { + __be32 *fh = (__force __be32 *)p; struct inode *inode = dentry->d_inode; struct super_block *sb = inode->i_sb; struct gfs2_inode *ip = GFS2_I(inode); @@ -76,14 +78,10 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *fh, int *len, (connectable && *len < GFS2_LARGE_FH_SIZE)) return 255; - fh[0] = ip->i_num.no_formal_ino >> 32; - fh[0] = cpu_to_be32(fh[0]); - fh[1] = ip->i_num.no_formal_ino & 0xFFFFFFFF; - fh[1] = cpu_to_be32(fh[1]); - fh[2] = ip->i_num.no_addr >> 32; - fh[2] = cpu_to_be32(fh[2]); - fh[3] = ip->i_num.no_addr & 0xFFFFFFFF; - fh[3] = cpu_to_be32(fh[3]); + fh[0] = cpu_to_be32(ip->i_num.no_formal_ino >> 32); + fh[1] = cpu_to_be32(ip->i_num.no_formal_ino & 0xFFFFFFFF); + fh[2] = cpu_to_be32(ip->i_num.no_addr >> 32); + fh[3] = cpu_to_be32(ip->i_num.no_addr & 0xFFFFFFFF); *len = GFS2_SMALL_FH_SIZE; if (!connectable || inode == sb->s_root->d_inode) @@ -95,14 +93,10 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *fh, int *len, igrab(inode); spin_unlock(&dentry->d_lock); - fh[4] = ip->i_num.no_formal_ino >> 32; - fh[4] = cpu_to_be32(fh[4]); - fh[5] = ip->i_num.no_formal_ino & 0xFFFFFFFF; - fh[5] = cpu_to_be32(fh[5]); - fh[6] = ip->i_num.no_addr >> 32; - fh[6] = cpu_to_be32(fh[6]); - fh[7] = ip->i_num.no_addr & 0xFFFFFFFF; - fh[7] = cpu_to_be32(fh[7]); + fh[4] = cpu_to_be32(ip->i_num.no_formal_ino >> 32); + fh[5] = cpu_to_be32(ip->i_num.no_formal_ino & 0xFFFFFFFF); + fh[6] = cpu_to_be32(ip->i_num.no_addr >> 32); + fh[7] = cpu_to_be32(ip->i_num.no_addr & 0xFFFFFFFF); fh[8] = cpu_to_be32(inode->i_mode); fh[9] = 0; /* pad to double word */ @@ -114,12 +108,12 @@ static int gfs2_encode_fh(struct dentry *dentry, __u32 *fh, int *len, } struct get_name_filldir { - struct gfs2_inum inum; + struct gfs2_inum_host inum; char *name; }; static int get_name_filldir(void *opaque, const char *name, unsigned int length, - u64 offset, struct gfs2_inum *inum, + u64 offset, struct gfs2_inum_host *inum, unsigned int type) { struct get_name_filldir *gnfd = (struct get_name_filldir *)opaque; @@ -202,7 +196,7 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, void *inum_obj) { struct gfs2_sbd *sdp = sb->s_fs_info; struct gfs2_fh_obj *fh_obj = (struct gfs2_fh_obj *)inum_obj; - struct gfs2_inum *inum = &fh_obj->this; + struct gfs2_inum_host *inum = &fh_obj->this; struct gfs2_holder i_gh, ri_gh, rgd_gh; struct gfs2_rgrpd *rgd; struct inode *inode; diff --git a/fs/gfs2/ops_export.h b/fs/gfs2/ops_export.h index 09aca50..f925a95 100644 --- a/fs/gfs2/ops_export.h +++ b/fs/gfs2/ops_export.h @@ -15,7 +15,7 @@ extern struct export_operations gfs2_export_ops; struct gfs2_fh_obj { - struct gfs2_inum this; + struct gfs2_inum_host this; __u32 imode; }; diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 3064f13..b3f1e03 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -22,6 +22,7 @@ #include <linux/ext2_fs.h> #include <linux/crc32.h> #include <linux/lm_interface.h> +#include <linux/writeback.h> #include <asm/uaccess.h> #include "gfs2.h" @@ -71,7 +72,7 @@ static int gfs2_read_actor(read_descriptor_t *desc, struct page *page, size = count; kaddr = kmap(page); - memcpy(desc->arg.buf, kaddr + offset, size); + memcpy(desc->arg.data, kaddr + offset, size); kunmap(page); desc->count = count - size; @@ -86,7 +87,7 @@ int gfs2_internal_read(struct gfs2_inode *ip, struct file_ra_state *ra_state, struct inode *inode = &ip->i_inode; read_descriptor_t desc; desc.written = 0; - desc.arg.buf = buf; + desc.arg.data = buf; desc.count = size; desc.error = 0; do_generic_mapping_read(inode->i_mapping, ra_state, @@ -139,7 +140,7 @@ static loff_t gfs2_llseek(struct file *file, loff_t offset, int origin) */ static int filldir_func(void *opaque, const char *name, unsigned int length, - u64 offset, struct gfs2_inum *inum, + u64 offset, struct gfs2_inum_host *inum, unsigned int type) { struct filldir_reg *fdr = (struct filldir_reg *)opaque; @@ -253,7 +254,7 @@ static int gfs2_get_flags(struct file *filp, u32 __user *ptr) u32 fsflags; gfs2_holder_init(ip->i_gl, LM_ST_SHARED, GL_ATIME, &gh); - error = gfs2_glock_nq_m_atime(1, &gh); + error = gfs2_glock_nq_atime(&gh); if (error) return error; @@ -266,6 +267,24 @@ static int gfs2_get_flags(struct file *filp, u32 __user *ptr) return error; } +void gfs2_set_inode_flags(struct inode *inode) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_dinode_host *di = &ip->i_di; + unsigned int flags = inode->i_flags; + + flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC); + if (di->di_flags & GFS2_DIF_IMMUTABLE) + flags |= S_IMMUTABLE; + if (di->di_flags & GFS2_DIF_APPENDONLY) + flags |= S_APPEND; + if (di->di_flags & GFS2_DIF_NOATIME) + flags |= S_NOATIME; + if (di->di_flags & GFS2_DIF_SYNC) + flags |= S_SYNC; + inode->i_flags = flags; +} + /* Flags that can be set by user space */ #define GFS2_FLAGS_USER_SET (GFS2_DIF_JDATA| \ GFS2_DIF_DIRECTIO| \ @@ -336,8 +355,9 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) goto out_trans_end; gfs2_trans_add_bh(ip->i_gl, bh, 1); ip->i_di.di_flags = new_flags; - gfs2_dinode_out(&ip->i_di, bh->b_data); + gfs2_dinode_out(ip, bh->b_data); brelse(bh); + gfs2_set_inode_flags(inode); out_trans_end: gfs2_trans_end(sdp); out: @@ -425,7 +445,7 @@ static int gfs2_open(struct inode *inode, struct file *file) gfs2_assert_warn(GFS2_SB(inode), !file->private_data); file->private_data = fp; - if (S_ISREG(ip->i_di.di_mode)) { + if (S_ISREG(ip->i_inode.i_mode)) { error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); if (error) @@ -484,16 +504,40 @@ static int gfs2_close(struct inode *inode, struct file *file) * @file: the file that points to the dentry (we ignore this) * @dentry: the dentry that points to the inode to sync * + * The VFS will flush "normal" data for us. We only need to worry + * about metadata here. For journaled data, we just do a log flush + * as we can't avoid it. Otherwise we can just bale out if datasync + * is set. For stuffed inodes we must flush the log in order to + * ensure that all data is on disk. + * + * The call to write_inode_now() is there to write back metadata and + * the inode itself. It does also try and write the data, but thats + * (hopefully) a no-op due to the VFS having already called filemap_fdatawrite() + * for us. + * * Returns: errno */ static int gfs2_fsync(struct file *file, struct dentry *dentry, int datasync) { - struct gfs2_inode *ip = GFS2_I(dentry->d_inode); + struct inode *inode = dentry->d_inode; + int sync_state = inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC); + int ret = 0; - gfs2_log_flush(ip->i_gl->gl_sbd, ip->i_gl); + if (gfs2_is_jdata(GFS2_I(inode))) { + gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); + return 0; + } - return 0; + if (sync_state != 0) { + if (!datasync) + ret = write_inode_now(inode, 0); + + if (gfs2_is_stuffed(GFS2_I(inode))) + gfs2_log_flush(GFS2_SB(inode), GFS2_I(inode)->i_gl); + } + + return ret; } /** @@ -515,7 +559,7 @@ static int gfs2_lock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_POSIX)) return -ENOLCK; - if ((ip->i_di.di_mode & (S_ISGID | S_IXGRP)) == S_ISGID) + if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID) return -ENOLCK; if (sdp->sd_args.ar_localflocks) { @@ -617,7 +661,7 @@ static int gfs2_flock(struct file *file, int cmd, struct file_lock *fl) if (!(fl->fl_flags & FL_FLOCK)) return -ENOLCK; - if ((ip->i_di.di_mode & (S_ISGID | S_IXGRP)) == S_ISGID) + if ((ip->i_inode.i_mode & (S_ISGID | S_IXGRP)) == S_ISGID) return -ENOLCK; if (sdp->sd_args.ar_localflocks) diff --git a/fs/gfs2/ops_file.h b/fs/gfs2/ops_file.h index ce319f8..7e5d8ec 100644 --- a/fs/gfs2/ops_file.h +++ b/fs/gfs2/ops_file.h @@ -17,7 +17,7 @@ extern struct file gfs2_internal_file_sentinel; extern int gfs2_internal_read(struct gfs2_inode *ip, struct file_ra_state *ra_state, char *buf, loff_t *pos, unsigned size); - +extern void gfs2_set_inode_flags(struct inode *inode); extern const struct file_operations gfs2_file_fops; extern const struct file_operations gfs2_dir_fops; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 882873a..d14e139 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -237,7 +237,7 @@ fail: } static struct inode *gfs2_lookup_root(struct super_block *sb, - struct gfs2_inum *inum) + struct gfs2_inum_host *inum) { return gfs2_inode_lookup(sb, inum, DT_DIR); } @@ -246,7 +246,7 @@ static int init_sb(struct gfs2_sbd *sdp, int silent, int undo) { struct super_block *sb = sdp->sd_vfs; struct gfs2_holder sb_gh; - struct gfs2_inum *inum; + struct gfs2_inum_host *inum; struct inode *inode; int error = 0; diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index ef6e5ed..636dda4 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -59,7 +59,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, gfs2_holder_init(dip->i_gl, 0, 0, ghs); for (;;) { - inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFREG | mode, 0); if (!IS_ERR(inode)) { gfs2_trans_end(sdp); if (dip->i_alloc.al_rgd) @@ -144,7 +144,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, int alloc_required; int error; - if (S_ISDIR(ip->i_di.di_mode)) + if (S_ISDIR(inode->i_mode)) return -EPERM; gfs2_holder_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); @@ -169,7 +169,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, } error = -EINVAL; - if (!dip->i_di.di_nlink) + if (!dip->i_inode.i_nlink) goto out_gunlock; error = -EFBIG; if (dip->i_di.di_entries == (u32)-1) @@ -178,10 +178,10 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) goto out_gunlock; error = -EINVAL; - if (!ip->i_di.di_nlink) + if (!ip->i_inode.i_nlink) goto out_gunlock; error = -EMLINK; - if (ip->i_di.di_nlink == (u32)-1) + if (ip->i_inode.i_nlink == (u32)-1) goto out_gunlock; alloc_required = error = gfs2_diradd_alloc_required(dir, &dentry->d_name); @@ -196,8 +196,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (error) goto out_alloc; - error = gfs2_quota_check(dip, dip->i_di.di_uid, - dip->i_di.di_gid); + error = gfs2_quota_check(dip, dip->i_inode.i_uid, dip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -220,7 +219,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, } error = gfs2_dir_add(dir, &dentry->d_name, &ip->i_num, - IF2DT(ip->i_di.di_mode)); + IF2DT(inode->i_mode)); if (error) goto out_end_trans; @@ -326,7 +325,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFLNK | S_IRWXUGO, 0); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); @@ -339,7 +338,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, error = gfs2_meta_inode_buffer(ip, &dibh); if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); memcpy(dibh->b_data + sizeof(struct gfs2_dinode), symname, size); brelse(dibh); @@ -379,7 +378,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode); + inode = gfs2_createi(ghs, &dentry->d_name, S_IFDIR | mode, 0); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); @@ -387,10 +386,9 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) ip = ghs[1].gh_gl->gl_object; - ip->i_di.di_nlink = 2; + ip->i_inode.i_nlink = 2; ip->i_di.di_size = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode); ip->i_di.di_flags |= GFS2_DIF_JDATA; - ip->i_di.di_payload_format = GFS2_FORMAT_DE; ip->i_di.di_entries = 2; error = gfs2_meta_inode_buffer(ip, &dibh); @@ -414,7 +412,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) gfs2_inum_out(&dip->i_num, &dent->de_inum); dent->de_type = cpu_to_be16(DT_DIR); - gfs2_dinode_out(&ip->i_di, di); + gfs2_dinode_out(ip, di); brelse(dibh); } @@ -467,7 +465,7 @@ static int gfs2_rmdir(struct inode *dir, struct dentry *dentry) if (ip->i_di.di_entries < 2) { if (gfs2_consist_inode(ip)) - gfs2_dinode_print(&ip->i_di); + gfs2_dinode_print(ip); error = -EIO; goto out_gunlock; } @@ -504,47 +502,19 @@ out: static int gfs2_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev) { - struct gfs2_inode *dip = GFS2_I(dir), *ip; + struct gfs2_inode *dip = GFS2_I(dir); struct gfs2_sbd *sdp = GFS2_SB(dir); struct gfs2_holder ghs[2]; struct inode *inode; - struct buffer_head *dibh; - u32 major = 0, minor = 0; - int error; - - switch (mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - major = MAJOR(dev); - minor = MINOR(dev); - break; - case S_IFIFO: - case S_IFSOCK: - break; - default: - return -EOPNOTSUPP; - }; gfs2_holder_init(dip->i_gl, 0, 0, ghs); - inode = gfs2_createi(ghs, &dentry->d_name, mode); + inode = gfs2_createi(ghs, &dentry->d_name, mode, dev); if (IS_ERR(inode)) { gfs2_holder_uninit(ghs); return PTR_ERR(inode); } - ip = ghs[1].gh_gl->gl_object; - - ip->i_di.di_major = major; - ip->i_di.di_minor = minor; - - error = gfs2_meta_inode_buffer(ip, &dibh); - - if (!gfs2_assert_withdraw(sdp, !error)) { - gfs2_dinode_out(&ip->i_di, dibh->b_data); - brelse(dibh); - } - gfs2_trans_end(sdp); if (dip->i_alloc.al_rgd) gfs2_inplace_release(dip); @@ -592,11 +562,10 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Make sure we aren't trying to move a dirctory into it's subdir */ - if (S_ISDIR(ip->i_di.di_mode) && odip != ndip) { + if (S_ISDIR(ip->i_inode.i_mode) && odip != ndip) { dir_rename = 1; - error = gfs2_glock_nq_init(sdp->sd_rename_gl, - LM_ST_EXCLUSIVE, 0, + error = gfs2_glock_nq_init(sdp->sd_rename_gl, LM_ST_EXCLUSIVE, 0, &r_gh); if (error) goto out; @@ -637,10 +606,10 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) goto out_gunlock; - if (S_ISDIR(nip->i_di.di_mode)) { + if (S_ISDIR(nip->i_inode.i_mode)) { if (nip->i_di.di_entries < 2) { if (gfs2_consist_inode(nip)) - gfs2_dinode_print(&nip->i_di); + gfs2_dinode_print(nip); error = -EIO; goto out_gunlock; } @@ -666,7 +635,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, }; if (odip != ndip) { - if (!ndip->i_di.di_nlink) { + if (!ndip->i_inode.i_nlink) { error = -EINVAL; goto out_gunlock; } @@ -674,8 +643,8 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = -EFBIG; goto out_gunlock; } - if (S_ISDIR(ip->i_di.di_mode) && - ndip->i_di.di_nlink == (u32)-1) { + if (S_ISDIR(ip->i_inode.i_mode) && + ndip->i_inode.i_nlink == (u32)-1) { error = -EMLINK; goto out_gunlock; } @@ -702,8 +671,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) goto out_alloc; - error = gfs2_quota_check(ndip, ndip->i_di.di_uid, - ndip->i_di.di_gid); + error = gfs2_quota_check(ndip, ndip->i_inode.i_uid, ndip->i_inode.i_gid); if (error) goto out_gunlock_q; @@ -729,7 +697,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Remove the target file, if it exists */ if (nip) { - if (S_ISDIR(nip->i_di.di_mode)) + if (S_ISDIR(nip->i_inode.i_mode)) error = gfs2_rmdiri(ndip, &ndentry->d_name, nip); else { error = gfs2_dir_del(ndip, &ndentry->d_name); @@ -760,9 +728,9 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = gfs2_meta_inode_buffer(ip, &dibh); if (error) goto out_end_trans; - ip->i_di.di_ctime = get_seconds(); + ip->i_inode.i_ctime.tv_sec = get_seconds(); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); } @@ -771,7 +739,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, goto out_end_trans; error = gfs2_dir_add(ndir, &ndentry->d_name, &ip->i_num, - IF2DT(ip->i_di.di_mode)); + IF2DT(ip->i_inode.i_mode)); if (error) goto out_end_trans; @@ -867,6 +835,10 @@ static void *gfs2_follow_link(struct dentry *dentry, struct nameidata *nd) * @mask: * @nd: passed from Linux VFS, ignored by us * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. + * * Returns: errno */ @@ -875,15 +847,18 @@ static int gfs2_permission(struct inode *inode, int mask, struct nameidata *nd) struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_holder i_gh; int error; + int unlock = 0; - if (ip->i_vn == ip->i_gl->gl_vn) - return generic_permission(inode, mask, gfs2_check_acl); + if (gfs2_glock_is_locked_by_me(ip->i_gl) == 0) { + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); + if (error) + return error; + unlock = 1; + } - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); - if (!error) { - error = generic_permission(inode, mask, gfs2_check_acl_locked); + error = generic_permission(inode, mask, gfs2_check_acl); + if (unlock) gfs2_glock_dq_uninit(&i_gh); - } return error; } @@ -914,8 +889,8 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) u32 ouid, ogid, nuid, ngid; int error; - ouid = ip->i_di.di_uid; - ogid = ip->i_di.di_gid; + ouid = inode->i_uid; + ogid = inode->i_gid; nuid = attr->ia_uid; ngid = attr->ia_gid; @@ -946,10 +921,9 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) error = inode_setattr(inode, attr); gfs2_assert_warn(sdp, !error); - gfs2_inode_attr_out(ip); gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(&ip->i_di, dibh->b_data); + gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { @@ -1018,6 +992,12 @@ out: * @dentry: The dentry to stat * @stat: The inode's stats * + * This may be called from the VFS directly, or from within GFS2 with the + * inode locked, so we look to see if the glock is already locked and only + * lock the glock if its not already been done. Note that its the NFS + * readdirplus operation which causes this to be called (from filldir) + * with the glock already held. + * * Returns: errno */ @@ -1028,14 +1008,20 @@ static int gfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_holder gh; int error; + int unlock = 0; - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (!error) { - generic_fillattr(inode, stat); - gfs2_glock_dq_uninit(&gh); + if (gfs2_glock_is_locked_by_me(ip->i_gl) == 0) { + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) + return error; + unlock = 1; } - return error; + generic_fillattr(inode, stat); + if (unlock); + gfs2_glock_dq_uninit(&gh); + + return 0; } static int gfs2_setxattr(struct dentry *dentry, const char *name, diff --git a/fs/gfs2/ops_super.c b/fs/gfs2/ops_super.c index b47d959..7685b46 100644 --- a/fs/gfs2/ops_super.c +++ b/fs/gfs2/ops_super.c @@ -157,7 +157,8 @@ static void gfs2_write_super(struct super_block *sb) static int gfs2_sync_fs(struct super_block *sb, int wait) { sb->s_dirt = 0; - gfs2_log_flush(sb->s_fs_info, NULL); + if (wait) + gfs2_log_flush(sb->s_fs_info, NULL); return 0; } @@ -215,7 +216,7 @@ static int gfs2_statfs(struct dentry *dentry, struct kstatfs *buf) { struct super_block *sb = dentry->d_inode->i_sb; struct gfs2_sbd *sdp = sb->s_fs_info; - struct gfs2_statfs_change sc; + struct gfs2_statfs_change_host sc; int error; if (gfs2_tune_get(sdp, gt_statfs_slow)) @@ -293,8 +294,6 @@ static void gfs2_clear_inode(struct inode *inode) */ if (inode->i_private) { struct gfs2_inode *ip = GFS2_I(inode); - gfs2_glock_inode_squish(inode); - gfs2_assert(inode->i_sb->s_fs_info, ip->i_gl->gl_state == LM_ST_UNLOCKED); ip->i_gl->gl_object = NULL; gfs2_glock_schedule_for_reclaim(ip->i_gl); gfs2_glock_put(ip->i_gl); @@ -395,7 +394,7 @@ static void gfs2_delete_inode(struct inode *inode) if (!inode->i_private) goto out; - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | GL_NOCACHE, &gh); + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB, &gh); if (unlikely(error)) { gfs2_glock_dq_uninit(&ip->i_iopen_gh); goto out; @@ -407,7 +406,7 @@ static void gfs2_delete_inode(struct inode *inode) if (error) goto out_uninit; - if (S_ISDIR(ip->i_di.di_mode) && + if (S_ISDIR(inode->i_mode) && (ip->i_di.di_flags & GFS2_DIF_EXHASH)) { error = gfs2_dir_exhash_dealloc(ip); if (error) diff --git a/fs/gfs2/ops_vm.c b/fs/gfs2/ops_vm.c index 5453d29..45a5f11 100644 --- a/fs/gfs2/ops_vm.c +++ b/fs/gfs2/ops_vm.c @@ -76,7 +76,7 @@ static int alloc_page_backing(struct gfs2_inode *ip, struct page *page) if (error) goto out; - error = gfs2_quota_check(ip, ip->i_di.di_uid, ip->i_di.di_gid); + error = gfs2_quota_check(ip, ip->i_inode.i_uid, ip->i_inode.i_gid); if (error) goto out_gunlock_q; diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index a3deae7..d0db881 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -452,19 +452,19 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF) return 0; - error = qdsb_get(sdp, QUOTA_USER, ip->i_di.di_uid, CREATE, qd); + error = qdsb_get(sdp, QUOTA_USER, ip->i_inode.i_uid, CREATE, qd); if (error) goto out; al->al_qd_num++; qd++; - error = qdsb_get(sdp, QUOTA_GROUP, ip->i_di.di_gid, CREATE, qd); + error = qdsb_get(sdp, QUOTA_GROUP, ip->i_inode.i_gid, CREATE, qd); if (error) goto out; al->al_qd_num++; qd++; - if (uid != NO_QUOTA_CHANGE && uid != ip->i_di.di_uid) { + if (uid != NO_QUOTA_CHANGE && uid != ip->i_inode.i_uid) { error = qdsb_get(sdp, QUOTA_USER, uid, CREATE, qd); if (error) goto out; @@ -472,7 +472,7 @@ int gfs2_quota_hold(struct gfs2_inode *ip, u32 uid, u32 gid) qd++; } - if (gid != NO_QUOTA_CHANGE && gid != ip->i_di.di_gid) { + if (gid != NO_QUOTA_CHANGE && gid != ip->i_inode.i_gid) { error = qdsb_get(sdp, QUOTA_GROUP, gid, CREATE, qd); if (error) goto out; @@ -539,8 +539,7 @@ static void do_qc(struct gfs2_quota_data *qd, s64 change) qc->qc_id = cpu_to_be32(qd->qd_id); } - x = qc->qc_change; - x = be64_to_cpu(x) + change; + x = be64_to_cpu(qc->qc_change) + change; qc->qc_change = cpu_to_be64(x); spin_lock(&sdp->sd_quota_spin); @@ -743,7 +742,7 @@ static int do_glock(struct gfs2_quota_data *qd, int force_refresh, struct gfs2_sbd *sdp = qd->qd_gl->gl_sbd; struct gfs2_inode *ip = GFS2_I(sdp->sd_quota_inode); struct gfs2_holder i_gh; - struct gfs2_quota q; + struct gfs2_quota_host q; char buf[sizeof(struct gfs2_quota)]; struct file_ra_state ra_state; int error; @@ -1103,7 +1102,7 @@ int gfs2_quota_init(struct gfs2_sbd *sdp) for (y = 0; y < sdp->sd_qc_per_block && slot < sdp->sd_quota_slots; y++, slot++) { - struct gfs2_quota_change qc; + struct gfs2_quota_change_host qc; struct gfs2_quota_data *qd; gfs2_quota_change_in(&qc, bh->b_data + diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index 62cd223..d0c806b 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -132,10 +132,11 @@ void gfs2_revoke_clean(struct gfs2_sbd *sdp) */ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, - struct gfs2_log_header *head) + struct gfs2_log_header_host *head) { struct buffer_head *bh; - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; + const u32 nothing = 0; u32 hash; int error; @@ -143,11 +144,11 @@ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, if (error) return error; - memcpy(&lh, bh->b_data, sizeof(struct gfs2_log_header)); - lh.lh_hash = 0; - hash = gfs2_disk_hash((char *)&lh, sizeof(struct gfs2_log_header)); + hash = crc32_le((u32)~0, bh->b_data, sizeof(struct gfs2_log_header) - + sizeof(u32)); + hash = crc32_le(hash, (unsigned char const *)¬hing, sizeof(nothing)); + hash ^= (u32)~0; gfs2_log_header_in(&lh, bh->b_data); - brelse(bh); if (lh.lh_header.mh_magic != GFS2_MAGIC || @@ -174,7 +175,7 @@ static int get_log_header(struct gfs2_jdesc *jd, unsigned int blk, */ static int find_good_lh(struct gfs2_jdesc *jd, unsigned int *blk, - struct gfs2_log_header *head) + struct gfs2_log_header_host *head) { unsigned int orig_blk = *blk; int error; @@ -205,10 +206,10 @@ static int find_good_lh(struct gfs2_jdesc *jd, unsigned int *blk, * Returns: errno */ -static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { unsigned int blk = head->lh_blkno; - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; int error; for (;;) { @@ -245,9 +246,9 @@ static int jhead_scan(struct gfs2_jdesc *jd, struct gfs2_log_header *head) * Returns: errno */ -int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { - struct gfs2_log_header lh_1, lh_m; + struct gfs2_log_header_host lh_1, lh_m; u32 blk_1, blk_2, blk_m; int error; @@ -320,7 +321,7 @@ static int foreach_descriptor(struct gfs2_jdesc *jd, unsigned int start, length = be32_to_cpu(ld->ld_length); if (be32_to_cpu(ld->ld_header.mh_type) == GFS2_METATYPE_LH) { - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; error = get_log_header(jd, start, &lh); if (!error) { gfs2_replay_incr_blk(sdp, &start); @@ -363,7 +364,7 @@ static int foreach_descriptor(struct gfs2_jdesc *jd, unsigned int start, * Returns: errno */ -static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header *head) +static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head) { struct gfs2_inode *ip = GFS2_I(jd->jd_inode); struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); @@ -425,7 +426,7 @@ int gfs2_recover_journal(struct gfs2_jdesc *jd) { struct gfs2_inode *ip = GFS2_I(jd->jd_inode); struct gfs2_sbd *sdp = GFS2_SB(jd->jd_inode); - struct gfs2_log_header head; + struct gfs2_log_header_host head; struct gfs2_holder j_gh, ji_gh, t_gh; unsigned long t; int ro = 0; diff --git a/fs/gfs2/recovery.h b/fs/gfs2/recovery.h index 961feed..f7235e6 100644 --- a/fs/gfs2/recovery.h +++ b/fs/gfs2/recovery.h @@ -26,7 +26,7 @@ int gfs2_revoke_check(struct gfs2_sbd *sdp, u64 blkno, unsigned int where); void gfs2_revoke_clean(struct gfs2_sbd *sdp); int gfs2_find_jhead(struct gfs2_jdesc *jd, - struct gfs2_log_header *head); + struct gfs2_log_header_host *head); int gfs2_recover_journal(struct gfs2_jdesc *gfs2_jd); void gfs2_check_journals(struct gfs2_sbd *sdp); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index b261385..ff08465 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -253,7 +253,7 @@ void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd) } -static inline int rgrp_contains_block(struct gfs2_rindex *ri, u64 block) +static inline int rgrp_contains_block(struct gfs2_rindex_host *ri, u64 block) { u64 first = ri->ri_data0; u64 last = first + ri->ri_data; @@ -1217,7 +1217,7 @@ u64 gfs2_alloc_data(struct gfs2_inode *ip) al->al_alloced++; gfs2_statfs_change(sdp, 0, -1, 0); - gfs2_quota_change(ip, +1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid); spin_lock(&sdp->sd_rindex_spin); rgd->rd_free_clone--; @@ -1261,7 +1261,7 @@ u64 gfs2_alloc_meta(struct gfs2_inode *ip) al->al_alloced++; gfs2_statfs_change(sdp, 0, -1, 0); - gfs2_quota_change(ip, +1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, +1, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_trans_add_unrevoke(sdp, block); spin_lock(&sdp->sd_rindex_spin); @@ -1337,8 +1337,7 @@ void gfs2_free_data(struct gfs2_inode *ip, u64 bstart, u32 blen) gfs2_trans_add_rg(rgd); gfs2_statfs_change(sdp, 0, +blen, 0); - gfs2_quota_change(ip, -(s64)blen, - ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid); } /** @@ -1366,7 +1365,7 @@ void gfs2_free_meta(struct gfs2_inode *ip, u64 bstart, u32 blen) gfs2_trans_add_rg(rgd); gfs2_statfs_change(sdp, 0, +blen, 0); - gfs2_quota_change(ip, -(s64)blen, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -(s64)blen, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_meta_wipe(ip, bstart, blen); } @@ -1411,7 +1410,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno) void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) { gfs2_free_uninit_di(rgd, ip->i_num.no_addr); - gfs2_quota_change(ip, -1, ip->i_di.di_uid, ip->i_di.di_gid); + gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid); gfs2_meta_wipe(ip, ip->i_num.no_addr, 1); } diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 6a78b1b..43a24f2 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -97,7 +97,7 @@ void gfs2_tune_init(struct gfs2_tune *gt) * changed. */ -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb *sb, int silent) +int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent) { unsigned int x; @@ -180,6 +180,24 @@ static int end_bio_io_page(struct bio *bio, unsigned int bytes_done, int error) return 0; } +/** + * gfs2_read_super - Read the gfs2 super block from disk + * @sb: The VFS super block + * @sector: The location of the super block + * + * This uses the bio functions to read the super block from disk + * because we want to be 100% sure that we never read cached data. + * A super block is read twice only during each GFS2 mount and is + * never written to by the filesystem. The first time its read no + * locks are held, and the only details which are looked at are those + * relating to the locking protocol. Once locking is up and working, + * the sb is read again under the lock to establish the location of + * the master directory (contains pointers to journals etc) and the + * root directory. + * + * Returns: A page containing the sb or NULL + */ + struct page *gfs2_read_super(struct super_block *sb, sector_t sector) { struct page *page; @@ -199,7 +217,7 @@ struct page *gfs2_read_super(struct super_block *sb, sector_t sector) return NULL; } - bio->bi_sector = sector; + bio->bi_sector = sector * (sb->s_blocksize >> 9); bio->bi_bdev = sb->s_bdev; bio_add_page(bio, page, PAGE_SIZE, 0); @@ -508,7 +526,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode); struct gfs2_glock *j_gl = ip->i_gl; struct gfs2_holder t_gh; - struct gfs2_log_header head; + struct gfs2_log_header_host head; int error; error = gfs2_glock_nq_init(sdp->sd_trans_gl, LM_ST_SHARED, @@ -517,7 +535,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) return error; gfs2_meta_cache_flush(ip); - j_gl->gl_ops->go_inval(j_gl, DIO_METADATA | DIO_DATA); + j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); error = gfs2_find_jhead(sdp->sd_jdesc, &head); if (error) @@ -587,9 +605,9 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) int gfs2_statfs_init(struct gfs2_sbd *sdp) { struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct buffer_head *m_bh, *l_bh; struct gfs2_holder gh; int error; @@ -634,7 +652,7 @@ void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, s64 dinodes) { struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct buffer_head *l_bh; int error; @@ -660,8 +678,8 @@ int gfs2_statfs_sync(struct gfs2_sbd *sdp) { struct gfs2_inode *m_ip = GFS2_I(sdp->sd_statfs_inode); struct gfs2_inode *l_ip = GFS2_I(sdp->sd_sc_inode); - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; struct gfs2_holder gh; struct buffer_head *m_bh, *l_bh; int error; @@ -727,10 +745,10 @@ out: * Returns: errno */ -int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) +int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc) { - struct gfs2_statfs_change *m_sc = &sdp->sd_statfs_master; - struct gfs2_statfs_change *l_sc = &sdp->sd_statfs_local; + struct gfs2_statfs_change_host *m_sc = &sdp->sd_statfs_master; + struct gfs2_statfs_change_host *l_sc = &sdp->sd_statfs_local; spin_lock(&sdp->sd_statfs_spin); @@ -760,7 +778,7 @@ int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) */ static int statfs_slow_fill(struct gfs2_rgrpd *rgd, - struct gfs2_statfs_change *sc) + struct gfs2_statfs_change_host *sc) { gfs2_rgrp_verify(rgd); sc->sc_total += rgd->rd_ri.ri_data; @@ -782,7 +800,7 @@ static int statfs_slow_fill(struct gfs2_rgrpd *rgd, * Returns: errno */ -int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) +int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc) { struct gfs2_holder ri_gh; struct gfs2_rgrpd *rgd_next; @@ -792,7 +810,7 @@ int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc) int done; int error = 0, err; - memset(sc, 0, sizeof(struct gfs2_statfs_change)); + memset(sc, 0, sizeof(struct gfs2_statfs_change_host)); gha = kcalloc(slots, sizeof(struct gfs2_holder), GFP_KERNEL); if (!gha) return -ENOMEM; @@ -873,7 +891,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd; struct lfcc *lfcc; LIST_HEAD(list); - struct gfs2_log_header lh; + struct gfs2_log_header_host lh; int error; error = gfs2_jindex_hold(sdp, &ji_gh); diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index 5bb443a..e590b2d 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -14,7 +14,7 @@ void gfs2_tune_init(struct gfs2_tune *gt); -int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb *sb, int silent); +int gfs2_check_sb(struct gfs2_sbd *sdp, struct gfs2_sb_host *sb, int silent); int gfs2_read_sb(struct gfs2_sbd *sdp, struct gfs2_glock *gl, int silent); struct page *gfs2_read_super(struct super_block *sb, sector_t sector); @@ -45,8 +45,8 @@ int gfs2_statfs_init(struct gfs2_sbd *sdp); void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, s64 dinodes); int gfs2_statfs_sync(struct gfs2_sbd *sdp); -int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc); -int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change *sc); +int gfs2_statfs_i(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc); +int gfs2_statfs_slow(struct gfs2_sbd *sdp, struct gfs2_statfs_change_host *sc); int gfs2_freeze_fs(struct gfs2_sbd *sdp); void gfs2_unfreeze_fs(struct gfs2_sbd *sdp); diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 0e0ec98..983eaf1 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -426,9 +426,6 @@ static ssize_t name##_store(struct gfs2_sbd *sdp, const char *buf, size_t len)\ } \ TUNE_ATTR_2(name, name##_store) -TUNE_ATTR(ilimit, 0); -TUNE_ATTR(ilimit_tries, 0); -TUNE_ATTR(ilimit_min, 0); TUNE_ATTR(demote_secs, 0); TUNE_ATTR(incore_log_blocks, 0); TUNE_ATTR(log_flush_secs, 0); @@ -447,7 +444,6 @@ TUNE_ATTR(quota_simul_sync, 1); TUNE_ATTR(quota_cache_secs, 1); TUNE_ATTR(max_atomic_write, 1); TUNE_ATTR(stall_secs, 1); -TUNE_ATTR(entries_per_readdir, 1); TUNE_ATTR(greedy_default, 1); TUNE_ATTR(greedy_quantum, 1); TUNE_ATTR(greedy_max, 1); @@ -459,9 +455,6 @@ TUNE_ATTR_DAEMON(quotad_secs, quotad_process); TUNE_ATTR_3(quota_scale, quota_scale_show, quota_scale_store); static struct attribute *tune_attrs[] = { - &tune_attr_ilimit.attr, - &tune_attr_ilimit_tries.attr, - &tune_attr_ilimit_min.attr, &tune_attr_demote_secs.attr, &tune_attr_incore_log_blocks.attr, &tune_attr_log_flush_secs.attr, @@ -478,7 +471,6 @@ static struct attribute *tune_attrs[] = { &tune_attr_quota_cache_secs.attr, &tune_attr_max_atomic_write.attr, &tune_attr_stall_secs.attr, - &tune_attr_entries_per_readdir.attr, &tune_attr_greedy_default.attr, &tune_attr_greedy_quantum.attr, &tune_attr_greedy_max.attr, diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index 196c604..e5707a9 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -23,9 +23,9 @@ #include "lm.h" #include "util.h" -kmem_cache_t *gfs2_glock_cachep __read_mostly; -kmem_cache_t *gfs2_inode_cachep __read_mostly; -kmem_cache_t *gfs2_bufdata_cachep __read_mostly; +struct kmem_cache *gfs2_glock_cachep __read_mostly; +struct kmem_cache *gfs2_inode_cachep __read_mostly; +struct kmem_cache *gfs2_bufdata_cachep __read_mostly; void gfs2_assert_i(struct gfs2_sbd *sdp) { diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index 76a5089..28938a4 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -83,8 +83,7 @@ static inline int gfs2_meta_check_i(struct gfs2_sbd *sdp, char *file, unsigned int line) { struct gfs2_meta_header *mh = (struct gfs2_meta_header *)bh->b_data; - u32 magic = mh->mh_magic; - magic = be32_to_cpu(magic); + u32 magic = be32_to_cpu(mh->mh_magic); if (unlikely(magic != GFS2_MAGIC)) return gfs2_meta_check_ii(sdp, bh, "magic number", function, file, line); @@ -107,9 +106,8 @@ static inline int gfs2_metatype_check_i(struct gfs2_sbd *sdp, char *file, unsigned int line) { struct gfs2_meta_header *mh = (struct gfs2_meta_header *)bh->b_data; - u32 magic = mh->mh_magic; + u32 magic = be32_to_cpu(mh->mh_magic); u16 t = be32_to_cpu(mh->mh_type); - magic = be32_to_cpu(magic); if (unlikely(magic != GFS2_MAGIC)) return gfs2_meta_check_ii(sdp, bh, "magic number", function, file, line); @@ -146,9 +144,9 @@ int gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, gfs2_io_error_bh_i((sdp), (bh), __FUNCTION__, __FILE__, __LINE__); -extern kmem_cache_t *gfs2_glock_cachep; -extern kmem_cache_t *gfs2_inode_cachep; -extern kmem_cache_t *gfs2_bufdata_cachep; +extern struct kmem_cache *gfs2_glock_cachep; +extern struct kmem_cache *gfs2_inode_cachep; +extern struct kmem_cache *gfs2_bufdata_cachep; static inline unsigned int gfs2_tune_get_i(struct gfs2_tune *gt, unsigned int *p) diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 85b17b3..a369879 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -24,7 +24,7 @@ #include "hfs_fs.h" #include "btree.h" -static kmem_cache_t *hfs_inode_cachep; +static struct kmem_cache *hfs_inode_cachep; MODULE_LICENSE("GPL"); @@ -145,7 +145,7 @@ static struct inode *hfs_alloc_inode(struct super_block *sb) { struct hfs_inode_info *i; - i = kmem_cache_alloc(hfs_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfs_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } @@ -430,7 +430,7 @@ static struct file_system_type hfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfs_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfs_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfs_inode_info *i = p; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 194eede..0f513c6 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -434,13 +434,13 @@ MODULE_AUTHOR("Brad Boyer"); MODULE_DESCRIPTION("Extended Macintosh Filesystem"); MODULE_LICENSE("GPL"); -static kmem_cache_t *hfsplus_inode_cachep; +static struct kmem_cache *hfsplus_inode_cachep; static struct inode *hfsplus_alloc_inode(struct super_block *sb) { struct hfsplus_inode_info *i; - i = kmem_cache_alloc(hfsplus_inode_cachep, SLAB_KERNEL); + i = kmem_cache_alloc(hfsplus_inode_cachep, GFP_KERNEL); return i ? &i->vfs_inode : NULL; } @@ -467,7 +467,7 @@ static struct file_system_type hfsplus_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void hfsplus_init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void hfsplus_init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct hfsplus_inode_info *i = p; diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c index ecc9180..594f9c4 100644 --- a/fs/hpfs/dir.c +++ b/fs/hpfs/dir.c @@ -84,7 +84,8 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } if (!fno->dirflag) { e = 1; - hpfs_error(inode->i_sb, "not a directory, fnode %08x",inode->i_ino); + hpfs_error(inode->i_sb, "not a directory, fnode %08lx", + (unsigned long)inode->i_ino); } if (hpfs_inode->i_dno != fno->u.external[0].disk_secno) { e = 1; @@ -144,8 +145,11 @@ static int hpfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } if (de->first || de->last) { if (hpfs_sb(inode->i_sb)->sb_chk) { - if (de->first && !de->last && (de->namelen != 2 || de ->name[0] != 1 || de->name[1] != 1)) hpfs_error(inode->i_sb, "hpfs_readdir: bad ^A^A entry; pos = %08x", old_pos); - if (de->last && (de->namelen != 1 || de ->name[0] != 255)) hpfs_error(inode->i_sb, "hpfs_readdir: bad \\377 entry; pos = %08x", old_pos); + if (de->first && !de->last && (de->namelen != 2 + || de ->name[0] != 1 || de->name[1] != 1)) + hpfs_error(inode->i_sb, "hpfs_readdir: bad ^A^A entry; pos = %08lx", old_pos); + if (de->last && (de->namelen != 1 || de ->name[0] != 255)) + hpfs_error(inode->i_sb, "hpfs_readdir: bad \\377 entry; pos = %08lx", old_pos); } hpfs_brelse4(&qbh); goto again; diff --git a/fs/hpfs/dnode.c b/fs/hpfs/dnode.c index 229ff2f..fe83c2b 100644 --- a/fs/hpfs/dnode.c +++ b/fs/hpfs/dnode.c @@ -533,10 +533,13 @@ static void delete_empty_dnode(struct inode *i, dnode_secno dno) struct buffer_head *bh; struct dnode *d1; struct quad_buffer_head qbh1; - if (hpfs_sb(i->i_sb)->sb_chk) if (up != i->i_ino) { - hpfs_error(i->i_sb, "bad pointer to fnode, dnode %08x, pointing to %08x, should be %08x", dno, up, i->i_ino); + if (hpfs_sb(i->i_sb)->sb_chk) + if (up != i->i_ino) { + hpfs_error(i->i_sb, + "bad pointer to fnode, dnode %08x, pointing to %08x, should be %08lx", + dno, up, (unsigned long)i->i_ino); return; - } + } if ((d1 = hpfs_map_dnode(i->i_sb, down, &qbh1))) { d1->up = up; d1->root_dnode = 1; @@ -851,7 +854,9 @@ struct hpfs_dirent *map_pos_dirent(struct inode *inode, loff_t *posp, /* Going to the next dirent */ if ((d = de_next_de(de)) < dnode_end_de(dnode)) { if (!(++*posp & 077)) { - hpfs_error(inode->i_sb, "map_pos_dirent: pos crossed dnode boundary; pos = %08x", *posp); + hpfs_error(inode->i_sb, + "map_pos_dirent: pos crossed dnode boundary; pos = %08llx", + (unsigned long long)*posp); goto bail; } /* We're going down the tree */ diff --git a/fs/hpfs/ea.c b/fs/hpfs/ea.c index 66339dc..547a838 100644 --- a/fs/hpfs/ea.c +++ b/fs/hpfs/ea.c @@ -243,8 +243,9 @@ void hpfs_set_ea(struct inode *inode, struct fnode *fnode, char *key, char *data fnode->ea_offs = 0xc4; } if (fnode->ea_offs < 0xc4 || fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200) { - hpfs_error(s, "fnode %08x: ea_offs == %03x, ea_size_s == %03x", - inode->i_ino, fnode->ea_offs, fnode->ea_size_s); + hpfs_error(s, "fnode %08lx: ea_offs == %03x, ea_size_s == %03x", + (unsigned long)inode->i_ino, + fnode->ea_offs, fnode->ea_size_s); return; } if ((fnode->ea_size_s || !fnode->ea_size_l) && diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h index 32ab51e..1c07aa8 100644 --- a/fs/hpfs/hpfs_fn.h +++ b/fs/hpfs/hpfs_fn.h @@ -317,7 +317,8 @@ static inline struct hpfs_sb_info *hpfs_sb(struct super_block *sb) /* super.c */ -void hpfs_error(struct super_block *, char *, ...); +void hpfs_error(struct super_block *, const char *, ...) + __attribute__((format (printf, 2, 3))); int hpfs_stop_cycles(struct super_block *, int, int *, int *, char *); unsigned hpfs_count_one_bitmap(struct super_block *, secno); diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c index 7faef85..85d3e1d 100644 --- a/fs/hpfs/inode.c +++ b/fs/hpfs/inode.c @@ -251,7 +251,10 @@ void hpfs_write_inode_nolock(struct inode *i) de->file_size = 0; hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); - } else hpfs_error(i->i_sb, "directory %08x doesn't have '.' entry", i->i_ino); + } else + hpfs_error(i->i_sb, + "directory %08lx doesn't have '.' entry", + (unsigned long)i->i_ino); } mark_buffer_dirty(bh); brelse(bh); diff --git a/fs/hpfs/map.c b/fs/hpfs/map.c index 0fecdac..c472458 100644 --- a/fs/hpfs/map.c +++ b/fs/hpfs/map.c @@ -126,32 +126,40 @@ struct fnode *hpfs_map_fnode(struct super_block *s, ino_t ino, struct buffer_hea struct extended_attribute *ea; struct extended_attribute *ea_end; if (fnode->magic != FNODE_MAGIC) { - hpfs_error(s, "bad magic on fnode %08x", ino); + hpfs_error(s, "bad magic on fnode %08lx", + (unsigned long)ino); goto bail; } if (!fnode->dirflag) { if ((unsigned)fnode->btree.n_used_nodes + (unsigned)fnode->btree.n_free_nodes != (fnode->btree.internal ? 12 : 8)) { - hpfs_error(s, "bad number of nodes in fnode %08x", ino); + hpfs_error(s, + "bad number of nodes in fnode %08lx", + (unsigned long)ino); goto bail; } if (fnode->btree.first_free != 8 + fnode->btree.n_used_nodes * (fnode->btree.internal ? 8 : 12)) { - hpfs_error(s, "bad first_free pointer in fnode %08x", ino); + hpfs_error(s, + "bad first_free pointer in fnode %08lx", + (unsigned long)ino); goto bail; } } if (fnode->ea_size_s && ((signed int)fnode->ea_offs < 0xc4 || (signed int)fnode->ea_offs + fnode->acl_size_s + fnode->ea_size_s > 0x200)) { - hpfs_error(s, "bad EA info in fnode %08x: ea_offs == %04x ea_size_s == %04x", - ino, fnode->ea_offs, fnode->ea_size_s); + hpfs_error(s, + "bad EA info in fnode %08lx: ea_offs == %04x ea_size_s == %04x", + (unsigned long)ino, + fnode->ea_offs, fnode->ea_size_s); goto bail; } ea = fnode_ea(fnode); ea_end = fnode_end_ea(fnode); while (ea != ea_end) { if (ea > ea_end) { - hpfs_error(s, "bad EA in fnode %08x", ino); + hpfs_error(s, "bad EA in fnode %08lx", + (unsigned long)ino); goto bail; } ea = next_ea(ea); diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 450b5e0..d4abc1a 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -46,21 +46,17 @@ static void unmark_dirty(struct super_block *s) } /* Filesystem error... */ +static char err_buf[1024]; -#define ERR_BUF_SIZE 1024 - -void hpfs_error(struct super_block *s, char *m,...) +void hpfs_error(struct super_block *s, const char *fmt, ...) { - char *buf; - va_list l; - va_start(l, m); - if (!(buf = kmalloc(ERR_BUF_SIZE, GFP_KERNEL))) - printk("HPFS: No memory for error message '%s'\n",m); - else if (vsprintf(buf, m, l) >= ERR_BUF_SIZE) - printk("HPFS: Grrrr... Kernel memory corrupted ... going on, but it'll crash very soon :-(\n"); - printk("HPFS: filesystem error: "); - if (buf) printk("%s", buf); - else printk("%s\n",m); + va_list args; + + va_start(args, fmt); + vsnprintf(err_buf, sizeof(err_buf), fmt, args); + va_end(args); + + printk("HPFS: filesystem error: %s", err_buf); if (!hpfs_sb(s)->sb_was_error) { if (hpfs_sb(s)->sb_err == 2) { printk("; crashing the system because you wanted it\n"); @@ -76,7 +72,6 @@ void hpfs_error(struct super_block *s, char *m,...) } else if (s->s_flags & MS_RDONLY) printk("; going on - but anything won't be destroyed because it's read-only\n"); else printk("; corrupted filesystem mounted read/write - your computer will explode within 20 seconds ... but you wanted it so!\n"); } else printk("\n"); - kfree(buf); hpfs_sb(s)->sb_was_error = 1; } @@ -160,12 +155,12 @@ static int hpfs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * hpfs_inode_cachep; +static struct kmem_cache * hpfs_inode_cachep; static struct inode *hpfs_alloc_inode(struct super_block *sb) { struct hpfs_inode_info *ei; - ei = (struct hpfs_inode_info *)kmem_cache_alloc(hpfs_inode_cachep, SLAB_NOFS); + ei = (struct hpfs_inode_info *)kmem_cache_alloc(hpfs_inode_cachep, GFP_NOFS); if (!ei) return NULL; ei->vfs_inode.i_version = 1; @@ -177,7 +172,7 @@ static void hpfs_destroy_inode(struct inode *inode) kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 7f47569..0706f5a 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -513,7 +513,7 @@ static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo) } -static kmem_cache_t *hugetlbfs_inode_cachep; +static struct kmem_cache *hugetlbfs_inode_cachep; static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) { @@ -522,7 +522,7 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo))) return NULL; - p = kmem_cache_alloc(hugetlbfs_inode_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(hugetlbfs_inode_cachep, GFP_KERNEL); if (unlikely(!p)) { hugetlbfs_inc_free_inodes(sbinfo); return NULL; @@ -545,7 +545,7 @@ static const struct address_space_operations hugetlbfs_aops = { }; -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; @@ -97,7 +97,7 @@ static DEFINE_MUTEX(iprune_mutex); */ struct inodes_stat_t inodes_stat; -static kmem_cache_t * inode_cachep __read_mostly; +static struct kmem_cache * inode_cachep __read_mostly; static struct inode *alloc_inode(struct super_block *sb) { @@ -109,7 +109,7 @@ static struct inode *alloc_inode(struct super_block *sb) if (sb->s_op->alloc_inode) inode = sb->s_op->alloc_inode(sb); else - inode = (struct inode *) kmem_cache_alloc(inode_cachep, SLAB_KERNEL); + inode = (struct inode *) kmem_cache_alloc(inode_cachep, GFP_KERNEL); if (inode) { struct address_space * const mapping = &inode->i_data; @@ -209,7 +209,7 @@ void inode_init_once(struct inode *inode) EXPORT_SYMBOL(inode_init_once); -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct inode * inode = (struct inode *) foo; @@ -1242,9 +1242,6 @@ EXPORT_SYMBOL(inode_needs_sync); */ #ifdef CONFIG_QUOTA -/* Function back in dquot.c */ -int remove_inode_dquot_ref(struct inode *, int, struct list_head *); - void remove_dquot_ref(struct super_block *sb, int type, struct list_head *tofree_head) { diff --git a/fs/inotify_user.c b/fs/inotify_user.c index 017cb0f..e1956e6 100644 --- a/fs/inotify_user.c +++ b/fs/inotify_user.c @@ -34,8 +34,8 @@ #include <asm/ioctls.h> -static kmem_cache_t *watch_cachep __read_mostly; -static kmem_cache_t *event_cachep __read_mostly; +static struct kmem_cache *watch_cachep __read_mostly; +static struct kmem_cache *event_cachep __read_mostly; static struct vfsmount *inotify_mnt __read_mostly; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index c34b862..ea55b6c 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -57,12 +57,12 @@ static void isofs_put_super(struct super_block *sb) static void isofs_read_inode(struct inode *); static int isofs_statfs (struct dentry *, struct kstatfs *); -static kmem_cache_t *isofs_inode_cachep; +static struct kmem_cache *isofs_inode_cachep; static struct inode *isofs_alloc_inode(struct super_block *sb) { struct iso_inode_info *ei; - ei = kmem_cache_alloc(isofs_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(isofs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode) kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct iso_inode_info *ei = foo; diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index b85c686..10fff94 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -31,7 +31,7 @@ #include <linux/smp_lock.h> #include <linux/init.h> #include <linux/mm.h> -#include <linux/suspend.h> +#include <linux/freezer.h> #include <linux/pagemap.h> #include <linux/kthread.h> #include <linux/poison.h> @@ -1630,7 +1630,7 @@ void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd_1k", "jbd_2k", "jbd_4k", NULL, "jbd_8k" }; @@ -1693,7 +1693,7 @@ void jbd_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *journal_head_cache; +static struct kmem_cache *journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -1996,7 +1996,7 @@ static void __exit remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd_handle_cache; +struct kmem_cache *jbd_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index c532429..d204ab3 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -70,8 +70,8 @@ #include <linux/init.h> #endif -static kmem_cache_t *revoke_record_cache; -static kmem_cache_t *revoke_table_cache; +static struct kmem_cache *revoke_record_cache; +static struct kmem_cache *revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 4f82bcd6..d38e0d5 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -27,6 +27,8 @@ #include <linux/mm.h> #include <linux/highmem.h> +static void __journal_temp_unlink_buffer(struct journal_head *jh); + /* * get_transaction: obtain a new transaction_t object. * @@ -1499,7 +1501,7 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh) * * Called under j_list_lock. The journal may not be locked. */ -void __journal_temp_unlink_buffer(struct journal_head *jh) +static void __journal_temp_unlink_buffer(struct journal_head *jh) { struct journal_head **list = NULL; transaction_t *transaction; diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 70b2ae1..6bd8005 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -248,8 +248,12 @@ write_out_data: bufs = 0; goto write_out_data; } - } - else { + } else if (!locked && buffer_locked(bh)) { + __jbd2_journal_file_buffer(jh, commit_transaction, + BJ_Locked); + jbd_unlock_bh_state(bh); + put_bh(bh); + } else { BUFFER_TRACE(bh, "writeout complete: unfile"); __jbd2_journal_unfile_buffer(jh); jbd_unlock_bh_state(bh); diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c60f378..44fc32b 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -31,7 +31,7 @@ #include <linux/smp_lock.h> #include <linux/init.h> #include <linux/mm.h> -#include <linux/suspend.h> +#include <linux/freezer.h> #include <linux/pagemap.h> #include <linux/kthread.h> #include <linux/poison.h> @@ -1641,7 +1641,7 @@ void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) #define JBD_MAX_SLABS 5 #define JBD_SLAB_INDEX(size) (size >> 11) -static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static struct kmem_cache *jbd_slab[JBD_MAX_SLABS]; static const char *jbd_slab_names[JBD_MAX_SLABS] = { "jbd2_1k", "jbd2_2k", "jbd2_4k", NULL, "jbd2_8k" }; @@ -1704,7 +1704,7 @@ void jbd2_slab_free(void *ptr, size_t size) /* * Journal_head storage management */ -static kmem_cache_t *jbd2_journal_head_cache; +static struct kmem_cache *jbd2_journal_head_cache; #ifdef CONFIG_JBD_DEBUG static atomic_t nr_journal_heads = ATOMIC_INIT(0); #endif @@ -2007,7 +2007,7 @@ static void __exit jbd2_remove_jbd_proc_entry(void) #endif -kmem_cache_t *jbd2_handle_cache; +struct kmem_cache *jbd2_handle_cache; static int __init journal_init_handle_cache(void) { diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index 380d199..f506646 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -70,8 +70,8 @@ #include <linux/init.h> #endif -static kmem_cache_t *jbd2_revoke_record_cache; -static kmem_cache_t *jbd2_revoke_table_cache; +static struct kmem_cache *jbd2_revoke_record_cache; +static struct kmem_cache *jbd2_revoke_table_cache; /* Each revoke record represents one single revoked block. During journal replay, this involves recording the transaction ID of the diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index c051a94..3a87001 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -27,6 +27,8 @@ #include <linux/mm.h> #include <linux/highmem.h> +static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh); + /* * jbd2_get_transaction: obtain a new transaction_t object. * diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 3f7899e..9f15bce 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -61,8 +61,8 @@ static const struct file_operations jffs_dir_operations; static struct inode_operations jffs_dir_inode_operations; static const struct address_space_operations jffs_address_operations; -kmem_cache_t *node_cache = NULL; -kmem_cache_t *fm_cache = NULL; +struct kmem_cache *node_cache = NULL; +struct kmem_cache *fm_cache = NULL; /* Called by the VFS at mount time to initialize the whole file system. */ static int jffs_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c index 4a543e1..d0e783f 100644 --- a/fs/jffs/intrep.c +++ b/fs/jffs/intrep.c @@ -66,6 +66,7 @@ #include <linux/smp_lock.h> #include <linux/time.h> #include <linux/ctype.h> +#include <linux/freezer.h> #include "intrep.h" #include "jffs_fm.h" @@ -591,7 +592,7 @@ jffs_add_virtual_root(struct jffs_control *c) D2(printk("jffs_add_virtual_root(): " "Creating a virtual root directory.\n")); - if (!(root = kmalloc(sizeof(struct jffs_file), GFP_KERNEL))) { + if (!(root = kzalloc(sizeof(struct jffs_file), GFP_KERNEL))) { return -ENOMEM; } no_jffs_file++; @@ -603,7 +604,6 @@ jffs_add_virtual_root(struct jffs_control *c) DJM(no_jffs_node++); memset(node, 0, sizeof(struct jffs_node)); node->ino = JFFS_MIN_INO; - memset(root, 0, sizeof(struct jffs_file)); root->ino = JFFS_MIN_INO; root->mode = S_IFDIR | S_IRWXU | S_IRGRP | S_IXGRP | S_IROTH | S_IXOTH; diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c index 29b68d9..077258b 100644 --- a/fs/jffs/jffs_fm.c +++ b/fs/jffs/jffs_fm.c @@ -29,8 +29,8 @@ static int jffs_mark_obsolete(struct jffs_fmcontrol *fmc, __u32 fm_offset); static struct jffs_fm *jffs_alloc_fm(void); static void jffs_free_fm(struct jffs_fm *n); -extern kmem_cache_t *fm_cache; -extern kmem_cache_t *node_cache; +extern struct kmem_cache *fm_cache; +extern struct kmem_cache *node_cache; #if CONFIG_JFFS_FS_VERBOSE > 0 void diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index ff2a872..6eb3dae 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -16,6 +16,7 @@ #include <linux/mtd/mtd.h> #include <linux/completion.h> #include <linux/sched.h> +#include <linux/freezer.h> #include "nodelist.h" diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 33f2910..83f9881 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -19,16 +19,16 @@ /* These are initialised to NULL in the kernel startup code. If you're porting to other operating systems, beware */ -static kmem_cache_t *full_dnode_slab; -static kmem_cache_t *raw_dirent_slab; -static kmem_cache_t *raw_inode_slab; -static kmem_cache_t *tmp_dnode_info_slab; -static kmem_cache_t *raw_node_ref_slab; -static kmem_cache_t *node_frag_slab; -static kmem_cache_t *inode_cache_slab; +static struct kmem_cache *full_dnode_slab; +static struct kmem_cache *raw_dirent_slab; +static struct kmem_cache *raw_inode_slab; +static struct kmem_cache *tmp_dnode_info_slab; +static struct kmem_cache *raw_node_ref_slab; +static struct kmem_cache *node_frag_slab; +static struct kmem_cache *inode_cache_slab; #ifdef CONFIG_JFFS2_FS_XATTR -static kmem_cache_t *xattr_datum_cache; -static kmem_cache_t *xattr_ref_cache; +static struct kmem_cache *xattr_datum_cache; +static struct kmem_cache *xattr_ref_cache; #endif int __init jffs2_create_slab_caches(void) diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index bc4b810..7deb782 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -28,12 +28,12 @@ static void jffs2_put_super(struct super_block *); -static kmem_cache_t *jffs2_inode_cachep; +static struct kmem_cache *jffs2_inode_cachep; static struct inode *jffs2_alloc_inode(struct super_block *sb) { struct jffs2_inode_info *ei; - ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, SLAB_KERNEL); + ei = (struct jffs2_inode_info *)kmem_cache_alloc(jffs2_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode) kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } -static void jffs2_i_init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void jffs2_i_init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct jffs2_inode_info *ei = (struct jffs2_inode_info *) foo; diff --git a/fs/jfs/jfs_logmgr.c b/fs/jfs/jfs_logmgr.c index b89c9ab..5065baa 100644 --- a/fs/jfs/jfs_logmgr.c +++ b/fs/jfs/jfs_logmgr.c @@ -67,7 +67,7 @@ #include <linux/kthread.h> #include <linux/buffer_head.h> /* for sync_blockdev() */ #include <linux/bio.h> -#include <linux/suspend.h> +#include <linux/freezer.h> #include <linux/delay.h> #include <linux/mutex.h> #include "jfs_incore.h" diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 0cccd1c..b1a1c72 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -74,7 +74,7 @@ static inline void lock_metapage(struct metapage *mp) } #define METAPOOL_MIN_PAGES 32 -static kmem_cache_t *metapage_cache; +static struct kmem_cache *metapage_cache; static mempool_t *metapage_mempool; #define MPS_PER_PAGE (PAGE_CACHE_SIZE >> L2PSIZE) @@ -180,7 +180,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) #endif -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags) { struct metapage *mp = (struct metapage *)foo; diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index 81f6f04..d558e51 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -46,7 +46,7 @@ #include <linux/vmalloc.h> #include <linux/smp_lock.h> #include <linux/completion.h> -#include <linux/suspend.h> +#include <linux/freezer.h> #include <linux/module.h> #include <linux/moduleparam.h> #include <linux/kthread.h> diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 9c1c6e0..846ac8f 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -44,7 +44,7 @@ MODULE_DESCRIPTION("The Journaled Filesystem (JFS)"); MODULE_AUTHOR("Steve Best/Dave Kleikamp/Barry Arndt, IBM"); MODULE_LICENSE("GPL"); -static kmem_cache_t * jfs_inode_cachep; +static struct kmem_cache * jfs_inode_cachep; static struct super_operations jfs_super_operations; static struct export_operations jfs_export_operations; @@ -93,7 +93,7 @@ void jfs_error(struct super_block *sb, const char * function, ...) va_list args; va_start(args, function); - vsprintf(error_buf, function, args); + vsnprintf(error_buf, sizeof(error_buf), function, args); va_end(args); printk(KERN_ERR "ERROR: (device %s): %s\n", sb->s_id, error_buf); @@ -748,7 +748,7 @@ static struct file_system_type jfs_fs_type = { .fs_flags = FS_REQUIRES_DEV, }; -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo; diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 41c983a..497c3cd5 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -13,6 +13,7 @@ #include <linux/nfs_fs.h> #include <linux/utsname.h> #include <linux/smp_lock.h> +#include <linux/freezer.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> diff --git a/fs/lockd/host.c b/fs/lockd/host.c index fb24a97..3d4610c 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -36,34 +36,14 @@ static DEFINE_MUTEX(nlm_host_mutex); static void nlm_gc_hosts(void); static struct nsm_handle * __nsm_find(const struct sockaddr_in *, const char *, int, int); - -/* - * Find an NLM server handle in the cache. If there is none, create it. - */ -struct nlm_host * -nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, - const char *hostname, int hostname_len) -{ - return nlm_lookup_host(0, sin, proto, version, - hostname, hostname_len); -} - -/* - * Find an NLM client handle in the cache. If there is none, create it. - */ -struct nlm_host * -nlmsvc_lookup_host(struct svc_rqst *rqstp, - const char *hostname, int hostname_len) -{ - return nlm_lookup_host(1, &rqstp->rq_addr, - rqstp->rq_prot, rqstp->rq_vers, - hostname, hostname_len); -} +static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, + const char *hostname, + int hostname_len); /* * Common host lookup routine for server & client */ -struct nlm_host * +static struct nlm_host * nlm_lookup_host(int server, const struct sockaddr_in *sin, int proto, int version, const char *hostname, @@ -195,6 +175,29 @@ nlm_destroy_host(struct nlm_host *host) } /* + * Find an NLM server handle in the cache. If there is none, create it. + */ +struct nlm_host * +nlmclnt_lookup_host(const struct sockaddr_in *sin, int proto, int version, + const char *hostname, int hostname_len) +{ + return nlm_lookup_host(0, sin, proto, version, + hostname, hostname_len); +} + +/* + * Find an NLM client handle in the cache. If there is none, create it. + */ +struct nlm_host * +nlmsvc_lookup_host(struct svc_rqst *rqstp, + const char *hostname, int hostname_len) +{ + return nlm_lookup_host(1, &rqstp->rq_addr, + rqstp->rq_prot, rqstp->rq_vers, + hostname, hostname_len); +} + +/* * Create the NLM RPC client for an NLM peer */ struct rpc_clnt * @@ -495,7 +498,7 @@ out: return nsm; } -struct nsm_handle * +static struct nsm_handle * nsm_find(const struct sockaddr_in *sin, const char *hostname, int hostname_len) { return __nsm_find(sin, hostname, hostname_len, 1); @@ -142,12 +142,12 @@ int lease_break_time = 45; static LIST_HEAD(file_lock_list); static LIST_HEAD(blocked_list); -static kmem_cache_t *filelock_cache __read_mostly; +static struct kmem_cache *filelock_cache __read_mostly; /* Allocate an empty lock structure. */ static struct file_lock *locks_alloc_lock(void) { - return kmem_cache_alloc(filelock_cache, SLAB_KERNEL); + return kmem_cache_alloc(filelock_cache, GFP_KERNEL); } static void locks_release_private(struct file_lock *fl) @@ -199,7 +199,7 @@ EXPORT_SYMBOL(locks_init_lock); * Initialises the fields of the file lock which are invariant for * free file_locks. */ -static void init_once(void *foo, kmem_cache_t *cache, unsigned long flags) +static void init_once(void *foo, struct kmem_cache *cache, unsigned long flags) { struct file_lock *lock = (struct file_lock *) foo; diff --git a/fs/mbcache.c b/fs/mbcache.c index 0ff7125..deeb9dc 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -85,7 +85,7 @@ struct mb_cache { #ifndef MB_CACHE_INDEXES_COUNT int c_indexes_count; #endif - kmem_cache_t *c_entry_cache; + struct kmem_cache *c_entry_cache; struct list_head *c_block_hash; struct list_head *c_indexes_hash[0]; }; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 1e36bae..629e09b 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -51,12 +51,12 @@ static void minix_put_super(struct super_block *sb) return; } -static kmem_cache_t * minix_inode_cachep; +static struct kmem_cache * minix_inode_cachep; static struct inode *minix_alloc_inode(struct super_block *sb) { struct minix_inode_info *ei; - ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, SLAB_KERNEL); + ei = (struct minix_inode_info *)kmem_cache_alloc(minix_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -67,7 +67,7 @@ static void minix_destroy_inode(struct inode *inode) kmem_cache_free(minix_inode_cachep, minix_i(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; @@ -249,9 +249,11 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) /* * MAY_EXEC on regular files requires special handling: We override - * filesystem execute permissions if the mode bits aren't set. + * filesystem execute permissions if the mode bits aren't set or + * the fs is mounted with the "noexec" flag. */ - if ((mask & MAY_EXEC) && S_ISREG(mode) && !(mode & S_IXUGO)) + if ((mask & MAY_EXEC) && S_ISREG(mode) && (!(mode & S_IXUGO) || + (nd && nd->mnt && (nd->mnt->mnt_flags & MNT_NOEXEC)))) return -EACCES; /* Ordinary permission routines do not understand MAY_APPEND. */ @@ -1996,8 +1998,7 @@ asmlinkage long sys_mkdir(const char __user *pathname, int mode) void dentry_unhash(struct dentry *dentry) { dget(dentry); - if (atomic_read(&dentry->d_count)) - shrink_dcache_parent(dentry); + shrink_dcache_parent(dentry); spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); if (atomic_read(&dentry->d_count) == 2) diff --git a/fs/namespace.c b/fs/namespace.c index 55442a6..b00ac84 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -36,7 +36,7 @@ static int event; static struct list_head *mount_hashtable __read_mostly; static int hash_mask __read_mostly, hash_bits __read_mostly; -static kmem_cache_t *mnt_cache __read_mostly; +static struct kmem_cache *mnt_cache __read_mostly; static struct rw_semaphore namespace_sem; /* /sys/fs */ diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 72dad55..fae5324 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -40,12 +40,12 @@ static void ncp_delete_inode(struct inode *); static void ncp_put_super(struct super_block *); static int ncp_statfs(struct dentry *, struct kstatfs *); -static kmem_cache_t * ncp_inode_cachep; +static struct kmem_cache * ncp_inode_cachep; static struct inode *ncp_alloc_inode(struct super_block *sb) { struct ncp_inode_info *ei; - ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, SLAB_KERNEL); + ei = (struct ncp_inode_info *)kmem_cache_alloc(ncp_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -56,7 +56,7 @@ static void ncp_destroy_inode(struct inode *inode) kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 784bbb5..f9d678f 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -58,7 +58,7 @@ #define NFSDBG_FACILITY NFSDBG_VFS -static kmem_cache_t *nfs_direct_cachep; +static struct kmem_cache *nfs_direct_cachep; /* * This represents a set of asynchronous requests that we're waiting on @@ -143,7 +143,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) { struct nfs_direct_req *dreq; - dreq = kmem_cache_alloc(nfs_direct_cachep, SLAB_KERNEL); + dreq = kmem_cache_alloc(nfs_direct_cachep, GFP_KERNEL); if (!dreq) return NULL; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 7c32187..36680d1 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -55,7 +55,7 @@ static int nfs_update_inode(struct inode *, struct nfs_fattr *); static void nfs_zap_acl_cache(struct inode *); -static kmem_cache_t * nfs_inode_cachep; +static struct kmem_cache * nfs_inode_cachep; static inline unsigned long nfs_fattr_to_ino_t(struct nfs_fattr *fattr) @@ -1080,7 +1080,7 @@ void nfs4_clear_inode(struct inode *inode) struct inode *nfs_alloc_inode(struct super_block *sb) { struct nfs_inode *nfsi; - nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, SLAB_KERNEL); + nfsi = (struct nfs_inode *)kmem_cache_alloc(nfs_inode_cachep, GFP_KERNEL); if (!nfsi) return NULL; nfsi->flags = 0UL; @@ -1111,7 +1111,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi) #endif } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct nfs_inode *nfsi = (struct nfs_inode *) foo; diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index bc9fab6..ca4b1d4 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -21,13 +21,13 @@ #define NFS_PARANOIA 1 -static kmem_cache_t *nfs_page_cachep; +static struct kmem_cache *nfs_page_cachep; static inline struct nfs_page * nfs_page_alloc(void) { struct nfs_page *p; - p = kmem_cache_alloc(nfs_page_cachep, SLAB_KERNEL); + p = kmem_cache_alloc(nfs_page_cachep, GFP_KERNEL); if (p) { memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->wb_list); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 05cca66..a9c2652 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -39,7 +39,7 @@ static int nfs_pagein_one(struct list_head *, struct inode *); static const struct rpc_call_ops nfs_read_partial_ops; static const struct rpc_call_ops nfs_read_full_ops; -static kmem_cache_t *nfs_rdata_cachep; +static struct kmem_cache *nfs_rdata_cachep; static mempool_t *nfs_rdata_mempool; #define MIN_POOL_READ (32) @@ -47,7 +47,7 @@ static mempool_t *nfs_rdata_mempool; struct nfs_read_data *nfs_readdata_alloc(size_t len) { unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; - struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, SLAB_NOFS); + struct nfs_read_data *p = mempool_alloc(nfs_rdata_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 7f3844d..594eb16 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -85,7 +85,7 @@ static const struct rpc_call_ops nfs_write_partial_ops; static const struct rpc_call_ops nfs_write_full_ops; static const struct rpc_call_ops nfs_commit_ops; -static kmem_cache_t *nfs_wdata_cachep; +static struct kmem_cache *nfs_wdata_cachep; static mempool_t *nfs_wdata_mempool; static mempool_t *nfs_commit_mempool; @@ -93,7 +93,7 @@ static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion); struct nfs_write_data *nfs_commit_alloc(void) { - struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, SLAB_NOFS); + struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); @@ -118,7 +118,7 @@ void nfs_commit_free(struct nfs_write_data *wdata) struct nfs_write_data *nfs_writedata_alloc(size_t len) { unsigned int pagecount = (len + PAGE_SIZE - 1) >> PAGE_SHIFT; - struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, SLAB_NOFS); + struct nfs_write_data *p = mempool_alloc(nfs_wdata_mempool, GFP_NOFS); if (p) { memset(p, 0, sizeof(*p)); diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index b4baca3..277df40 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -24,10 +24,6 @@ #define NFSDDBG_FACILITY NFSDDBG_XDR -#ifdef NFSD_OPTIMIZE_SPACE -# define inline -#endif - /* * Mapping of S_IF* types to NFS file types @@ -42,14 +38,14 @@ static u32 nfs3_ftypes[] = { /* * XDR functions for basic NFS types */ -static inline __be32 * +static __be32 * encode_time3(__be32 *p, struct timespec *time) { *p++ = htonl((u32) time->tv_sec); *p++ = htonl(time->tv_nsec); return p; } -static inline __be32 * +static __be32 * decode_time3(__be32 *p, struct timespec *time) { time->tv_sec = ntohl(*p++); @@ -57,7 +53,7 @@ decode_time3(__be32 *p, struct timespec *time) return p; } -static inline __be32 * +static __be32 * decode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size; @@ -77,7 +73,7 @@ __be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp) return decode_fh(p, fhp); } -static inline __be32 * +static __be32 * encode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size = fhp->fh_handle.fh_size; @@ -91,7 +87,7 @@ encode_fh(__be32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline __be32 * +static __be32 * decode_filename(__be32 *p, char **namp, int *lenp) { char *name; @@ -107,7 +103,7 @@ decode_filename(__be32 *p, char **namp, int *lenp) return p; } -static inline __be32 * +static __be32 * decode_sattr3(__be32 *p, struct iattr *iap) { u32 tmp; @@ -153,7 +149,7 @@ decode_sattr3(__be32 *p, struct iattr *iap) return p; } -static inline __be32 * +static __be32 * encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { @@ -186,7 +182,7 @@ encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, return p; } -static inline __be32 * +static __be32 * encode_saved_post_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct inode *inode = fhp->fh_dentry->d_inode; @@ -776,7 +772,7 @@ nfs3svc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, return xdr_ressize_check(rqstp, p); } -static inline __be32 * +static __be32 * encode_entry_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen, ino_t ino) { @@ -790,7 +786,7 @@ encode_entry_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, return p; } -static inline __be32 * +static __be32 * encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, struct svc_fh *fhp) { diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index e431e93..640c92b 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -84,10 +84,10 @@ static void nfs4_set_recdir(char *recdir); */ static DEFINE_MUTEX(client_mutex); -static kmem_cache_t *stateowner_slab = NULL; -static kmem_cache_t *file_slab = NULL; -static kmem_cache_t *stateid_slab = NULL; -static kmem_cache_t *deleg_slab = NULL; +static struct kmem_cache *stateowner_slab = NULL; +static struct kmem_cache *file_slab = NULL; +static struct kmem_cache *stateid_slab = NULL; +static struct kmem_cache *deleg_slab = NULL; void nfs4_lock_state(void) @@ -1003,7 +1003,7 @@ alloc_init_file(struct inode *ino) } static void -nfsd4_free_slab(kmem_cache_t **slab) +nfsd4_free_slab(struct kmem_cache **slab) { if (*slab == NULL) return; diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 56ebb14..f5243f9 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -18,11 +18,6 @@ #define NFSDDBG_FACILITY NFSDDBG_XDR - -#ifdef NFSD_OPTIMIZE_SPACE -# define inline -#endif - /* * Mapping of S_IF* types to NFS file types */ @@ -55,7 +50,7 @@ __be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp) return decode_fh(p, fhp); } -static inline __be32 * +static __be32 * encode_fh(__be32 *p, struct svc_fh *fhp) { memcpy(p, &fhp->fh_handle.fh_base, NFS_FHSIZE); @@ -66,7 +61,7 @@ encode_fh(__be32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline __be32 * +static __be32 * decode_filename(__be32 *p, char **namp, int *lenp) { char *name; @@ -82,7 +77,7 @@ decode_filename(__be32 *p, char **namp, int *lenp) return p; } -static inline __be32 * +static __be32 * decode_pathname(__be32 *p, char **namp, int *lenp) { char *name; @@ -98,7 +93,7 @@ decode_pathname(__be32 *p, char **namp, int *lenp) return p; } -static inline __be32 * +static __be32 * decode_sattr(__be32 *p, struct iattr *iap) { u32 tmp, tmp1; diff --git a/fs/nls/nls_cp936.c b/fs/nls/nls_cp936.c index 046fde8..65e640c 100644 --- a/fs/nls/nls_cp936.c +++ b/fs/nls/nls_cp936.c @@ -4421,6 +4421,73 @@ static wchar_t *page_charset2uni[256] = { c2u_F8, c2u_F9, c2u_FA, c2u_FB, c2u_FC, c2u_FD, c2u_FE, NULL, }; +static unsigned char u2c_00[512] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x00-0x03 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x04-0x07 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x08-0x0B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x0C-0x0F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x10-0x13 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x14-0x17 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x18-0x1B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x1C-0x1F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x20-0x23 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x24-0x27 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x28-0x2B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x2C-0x2F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x30-0x33 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x34-0x37 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x38-0x3B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x3C-0x3F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x40-0x43 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x44-0x47 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x48-0x4B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x4C-0x4F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x50-0x53 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x54-0x57 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x58-0x5B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x5C-0x5F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x60-0x63 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x64-0x67 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x68-0x6B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x6C-0x6F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x70-0x73 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x74-0x77 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x78-0x7B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x7C-0x7F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x80-0x83 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x84-0x87 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x88-0x8B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x8C-0x8F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x90-0x93 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x94-0x97 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x98-0x9B */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x9C-0x9F */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xA0-0xA3 */ + 0xA1, 0xE8, 0x00, 0x00, 0x00, 0x00, 0xA1, 0xEC, /* 0xA4-0xA7 */ + 0xA1, 0xA7, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xA8-0xAB */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xAC-0xAF */ + 0xA1, 0xE3, 0xA1, 0xC0, 0x00, 0x00, 0x00, 0x00, /* 0xB0-0xB3 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xA1, 0xA4, /* 0xB4-0xB7 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xB8-0xBB */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xBC-0xBF */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xC0-0xC3 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xC4-0xC7 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xC8-0xCB */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xCC-0xCF */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xD0-0xD3 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xA1, 0xC1, /* 0xD4-0xD7 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xD8-0xDB */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xDC-0xDF */ + 0xA8, 0xA4, 0xA8, 0xA2, 0x00, 0x00, 0x00, 0x00, /* 0xE0-0xE3 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xE4-0xE7 */ + 0xA8, 0xA8, 0xA8, 0xA6, 0xA8, 0xBA, 0x00, 0x00, /* 0xE8-0xEB */ + 0xA8, 0xAC, 0xA8, 0xAA, 0x00, 0x00, 0x00, 0x00, /* 0xEC-0xEF */ + 0x00, 0x00, 0x00, 0x00, 0xA8, 0xB0, 0xA8, 0xAE, /* 0xF0-0xF3 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xA1, 0xC2, /* 0xF4-0xF7 */ + 0x00, 0x00, 0xA8, 0xB4, 0xA8, 0xB2, 0x00, 0x00, /* 0xF8-0xFB */ + 0xA8, 0xB9, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0xFC-0xFF */ +}; + static unsigned char u2c_01[512] = { 0xA8, 0xA1, 0xA8, 0xA1, 0x00, 0x00, 0x00, 0x00, /* 0x00-0x03 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x04-0x07 */ @@ -10825,7 +10892,7 @@ static unsigned char u2c_FF[512] = { }; static unsigned char *page_uni2charset[256] = { - NULL, u2c_01, u2c_02, u2c_03, u2c_04, NULL, NULL, NULL, + u2c_00, u2c_01, u2c_02, u2c_03, u2c_04, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, @@ -10936,11 +11003,34 @@ static int uni2char(const wchar_t uni, unsigned char *uni2charset; unsigned char cl = uni&0xFF; unsigned char ch = (uni>>8)&0xFF; - int n; + unsigned char out0,out1; if (boundlen <= 0) return -ENAMETOOLONG; + if (uni == 0x20ac) {/* Euro symbol.The only exception with a non-ascii unicode */ + out[0] = 0x80; + return 1; + } + + if (ch == 0) { /* handle the U00 plane*/ + /* if (cl == 0) return -EINVAL;*/ /*U0000 is legal in cp936*/ + out0 = u2c_00[cl*2]; + out1 = u2c_00[cl*2+1]; + if (out0 == 0x00 && out1 == 0x00) { + if (cl<0x80) { + out[0] = cl; + return 1; + } + return -EINVAL; + } else { + if (boundlen <= 1) + return -ENAMETOOLONG; + out[0] = out0; + out[1] = out1; + return 2; + } + } uni2charset = page_uni2charset[ch]; if (uni2charset) { @@ -10950,15 +11040,10 @@ static int uni2char(const wchar_t uni, out[1] = uni2charset[cl*2+1]; if (out[0] == 0x00 && out[1] == 0x00) return -EINVAL; - n = 2; - } else if (ch==0 && cl) { - out[0] = cl; - n = 1; + return 2; } else return -EINVAL; - - return n; } static int char2uni(const unsigned char *rawstring, int boundlen, @@ -10972,7 +11057,11 @@ static int char2uni(const unsigned char *rawstring, int boundlen, return -ENAMETOOLONG; if (boundlen == 1) { - *uni = rawstring[0]; + if (rawstring[0]==0x80) { /* Euro symbol.The only exception with a non-ascii unicode */ + *uni = 0x20ac; + } else { + *uni = rawstring[0]; + } return 1; } @@ -10986,7 +11075,11 @@ static int char2uni(const unsigned char *rawstring, int boundlen, return -EINVAL; n = 2; } else{ - *uni = ch; + if (ch==0x80) {/* Euro symbol.The only exception with a non-ascii unicode */ + *uni = 0x20ac; + } else { + *uni = ch; + } n = 1; } return n; diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index 9f08e85..c577d8e 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -1272,7 +1272,7 @@ ntfs_attr_search_ctx *ntfs_attr_get_search_ctx(ntfs_inode *ni, MFT_RECORD *mrec) { ntfs_attr_search_ctx *ctx; - ctx = kmem_cache_alloc(ntfs_attr_ctx_cache, SLAB_NOFS); + ctx = kmem_cache_alloc(ntfs_attr_ctx_cache, GFP_NOFS); if (ctx) ntfs_attr_init_search_ctx(ctx, ni, mrec); return ctx; diff --git a/fs/ntfs/index.c b/fs/ntfs/index.c index e32cde4..2194eff 100644 --- a/fs/ntfs/index.c +++ b/fs/ntfs/index.c @@ -38,7 +38,7 @@ ntfs_index_context *ntfs_index_ctx_get(ntfs_inode *idx_ni) { ntfs_index_context *ictx; - ictx = kmem_cache_alloc(ntfs_index_ctx_cache, SLAB_NOFS); + ictx = kmem_cache_alloc(ntfs_index_ctx_cache, GFP_NOFS); if (ictx) *ictx = (ntfs_index_context){ .idx_ni = idx_ni }; return ictx; diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 2d3de9c..2479898 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -324,7 +324,7 @@ struct inode *ntfs_alloc_big_inode(struct super_block *sb) ntfs_inode *ni; ntfs_debug("Entering."); - ni = kmem_cache_alloc(ntfs_big_inode_cache, SLAB_NOFS); + ni = kmem_cache_alloc(ntfs_big_inode_cache, GFP_NOFS); if (likely(ni != NULL)) { ni->state = 0; return VFS_I(ni); @@ -349,7 +349,7 @@ static inline ntfs_inode *ntfs_alloc_extent_inode(void) ntfs_inode *ni; ntfs_debug("Entering."); - ni = kmem_cache_alloc(ntfs_inode_cache, SLAB_NOFS); + ni = kmem_cache_alloc(ntfs_inode_cache, GFP_NOFS); if (likely(ni != NULL)) { ni->state = 0; return ni; diff --git a/fs/ntfs/unistr.c b/fs/ntfs/unistr.c index 6a495f7..005ca4b 100644 --- a/fs/ntfs/unistr.c +++ b/fs/ntfs/unistr.c @@ -266,7 +266,7 @@ int ntfs_nlstoucs(const ntfs_volume *vol, const char *ins, /* We do not trust outside sources. */ if (likely(ins)) { - ucs = kmem_cache_alloc(ntfs_name_cache, SLAB_NOFS); + ucs = kmem_cache_alloc(ntfs_name_cache, GFP_NOFS); if (likely(ucs)) { for (i = o = 0; i < ins_len; i += wc_len) { wc_len = nls->char2uni(ins + i, ins_len - i, diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 16b8d1b..941acf1 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -66,7 +66,7 @@ static struct file_operations dlmfs_file_operations; static struct inode_operations dlmfs_dir_inode_operations; static struct inode_operations dlmfs_root_inode_operations; static struct inode_operations dlmfs_file_inode_operations; -static kmem_cache_t *dlmfs_inode_cache; +static struct kmem_cache *dlmfs_inode_cache; struct workqueue_struct *user_dlm_worker; @@ -257,7 +257,7 @@ static ssize_t dlmfs_file_write(struct file *filp, } static void dlmfs_init_once(void *foo, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct dlmfs_inode_private *ip = @@ -276,7 +276,7 @@ static struct inode *dlmfs_alloc_inode(struct super_block *sb) { struct dlmfs_inode_private *ip; - ip = kmem_cache_alloc(dlmfs_inode_cache, SLAB_NOFS); + ip = kmem_cache_alloc(dlmfs_inode_cache, GFP_NOFS); if (!ip) return NULL; diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index f784177..856012b 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -221,7 +221,7 @@ EXPORT_SYMBOL_GPL(dlm_dump_all_mles); #endif /* 0 */ -static kmem_cache_t *dlm_mle_cache = NULL; +static struct kmem_cache *dlm_mle_cache = NULL; static void dlm_mle_release(struct kref *kref); diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index fcd4475..80ac69f 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -61,7 +61,7 @@ struct ocfs2_em_insert_context { struct ocfs2_extent_map_entry *right_ent; }; -static kmem_cache_t *ocfs2_em_ent_cachep = NULL; +static struct kmem_cache *ocfs2_em_ent_cachep = NULL; static struct ocfs2_extent_map_entry * diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 46a378f..1a7dd29 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -106,7 +106,7 @@ static inline struct ocfs2_inode_info *OCFS2_I(struct inode *inode) #define INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags & OCFS2_INODE_JOURNAL) #define SET_INODE_JOURNAL(i) (OCFS2_I(i)->ip_flags |= OCFS2_INODE_JOURNAL) -extern kmem_cache_t *ocfs2_inode_cache; +extern struct kmem_cache *ocfs2_inode_cache; extern const struct address_space_operations ocfs2_aops; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index d9b4214..4bf3954 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -68,7 +68,7 @@ #include "buffer_head_io.h" -static kmem_cache_t *ocfs2_inode_cachep = NULL; +static struct kmem_cache *ocfs2_inode_cachep = NULL; /* OCFS2 needs to schedule several differnt types of work which * require cluster locking, disk I/O, recovery waits, etc. Since these @@ -303,7 +303,7 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb) { struct ocfs2_inode_info *oi; - oi = kmem_cache_alloc(ocfs2_inode_cachep, SLAB_NOFS); + oi = kmem_cache_alloc(ocfs2_inode_cachep, GFP_NOFS); if (!oi) return NULL; @@ -914,7 +914,7 @@ bail: } static void ocfs2_inode_init_once(void *data, - kmem_cache_t *cachep, + struct kmem_cache *cachep, unsigned long flags) { struct ocfs2_inode_info *oi = data; @@ -1674,7 +1674,7 @@ void __ocfs2_error(struct super_block *sb, va_list args; va_start(args, fmt); - vsprintf(error_buf, fmt, args); + vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); /* Not using mlog here because we want to show the actual @@ -1695,7 +1695,7 @@ void __ocfs2_abort(struct super_block* sb, va_list args; va_start(args, fmt); - vsprintf(error_buf, fmt, args); + vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); printk(KERN_CRIT "OCFS2: abort (device %s): %s: %s\n", diff --git a/fs/ocfs2/uptodate.c b/fs/ocfs2/uptodate.c index 9707ed7..39814b9 100644 --- a/fs/ocfs2/uptodate.c +++ b/fs/ocfs2/uptodate.c @@ -69,7 +69,7 @@ struct ocfs2_meta_cache_item { sector_t c_block; }; -static kmem_cache_t *ocfs2_uptodate_cachep = NULL; +static struct kmem_cache *ocfs2_uptodate_cachep = NULL; void ocfs2_metadata_cache_init(struct inode *inode) { diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 592a640..26f44e0 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -330,13 +330,13 @@ out: return 0; } -static kmem_cache_t *op_inode_cachep; +static struct kmem_cache *op_inode_cachep; static struct inode *openprom_alloc_inode(struct super_block *sb) { struct op_inode_info *oi; - oi = kmem_cache_alloc(op_inode_cachep, SLAB_KERNEL); + oi = kmem_cache_alloc(op_inode_cachep, GFP_KERNEL); if (!oi) return NULL; @@ -415,7 +415,7 @@ static struct file_system_type openprom_fs_type = { .kill_sb = kill_anon_super, }; -static void op_inode_init_once(void *data, kmem_cache_t * cachep, unsigned long flags) +static void op_inode_init_once(void *data, struct kmem_cache * cachep, unsigned long flags) { struct op_inode_info *oi = (struct op_inode_info *) data; diff --git a/fs/partitions/amiga.c b/fs/partitions/amiga.c index 3068528..9917a8c 100644 --- a/fs/partitions/amiga.c +++ b/fs/partitions/amiga.c @@ -43,6 +43,7 @@ amiga_partition(struct parsed_partitions *state, struct block_device *bdev) if (warn_no_part) printk("Dev %s: unable to read RDB block %d\n", bdevname(bdev, b), blk); + res = -1; goto rdb_done; } if (*(__be32 *)data != cpu_to_be32(IDNAME_RIGIDDISK)) @@ -79,6 +80,7 @@ amiga_partition(struct parsed_partitions *state, struct block_device *bdev) if (warn_no_part) printk("Dev %s: unable to read partition block %d\n", bdevname(bdev, b), blk); + res = -1; goto rdb_done; } pb = (struct PartitionBlock *)data; diff --git a/fs/partitions/atari.c b/fs/partitions/atari.c index 192a6ad..1f3572d 100644 --- a/fs/partitions/atari.c +++ b/fs/partitions/atari.c @@ -88,7 +88,7 @@ int atari_partition(struct parsed_partitions *state, struct block_device *bdev) if (!xrs) { printk (" block %ld read failed\n", partsect); put_dev_sector(sect); - return 0; + return -1; } /* ++roman: sanity check: bit 0 of flg field must be set */ diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 6fb4b61..1901137 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -153,7 +153,7 @@ static struct parsed_partitions * check_partition(struct gendisk *hd, struct block_device *bdev) { struct parsed_partitions *state; - int i, res; + int i, res, err; state = kmalloc(sizeof(struct parsed_partitions), GFP_KERNEL); if (!state) @@ -165,19 +165,30 @@ check_partition(struct gendisk *hd, struct block_device *bdev) sprintf(state->name, "p"); state->limit = hd->minors; - i = res = 0; + i = res = err = 0; while (!res && check_part[i]) { memset(&state->parts, 0, sizeof(state->parts)); res = check_part[i++](state, bdev); + if (res < 0) { + /* We have hit an I/O error which we don't report now. + * But record it, and let the others do their job. + */ + err = res; + res = 0; + } + } if (res > 0) return state; + if (!err) + /* The partition is unrecognized. So report I/O errors if there were any */ + res = err; if (!res) printk(" unknown partition table\n"); else if (warn_no_part) printk(" unable to read partition table\n"); kfree(state); - return NULL; + return ERR_PTR(res); } /* @@ -494,6 +505,8 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev) disk->fops->revalidate_disk(disk); if (!get_capacity(disk) || !(state = check_partition(disk, bdev))) return 0; + if (IS_ERR(state)) /* I/O error reading the partition table */ + return PTR_ERR(state); for (p = 1; p < state->limit; p++) { sector_t size = state->parts[p].size; sector_t from = state->parts[p].from; diff --git a/fs/partitions/ibm.c b/fs/partitions/ibm.c index d352a73..9f7ad42 100644 --- a/fs/partitions/ibm.c +++ b/fs/partitions/ibm.c @@ -43,7 +43,7 @@ cchhb2blk (struct vtoc_cchhb *ptr, struct hd_geometry *geo) { int ibm_partition(struct parsed_partitions *state, struct block_device *bdev) { - int blocksize, offset, size; + int blocksize, offset, size,res; loff_t i_size; dasd_information_t *info; struct hd_geometry *geo; @@ -56,15 +56,16 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) unsigned char *data; Sector sect; + res = 0; blocksize = bdev_hardsect_size(bdev); if (blocksize <= 0) - return 0; + goto out_exit; i_size = i_size_read(bdev->bd_inode); if (i_size == 0) - return 0; + goto out_exit; if ((info = kmalloc(sizeof(dasd_information_t), GFP_KERNEL)) == NULL) - goto out_noinfo; + goto out_exit; if ((geo = kmalloc(sizeof(struct hd_geometry), GFP_KERNEL)) == NULL) goto out_nogeo; if ((label = kmalloc(sizeof(union label_t), GFP_KERNEL)) == NULL) @@ -72,7 +73,7 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) if (ioctl_by_bdev(bdev, BIODASDINFO, (unsigned long)info) != 0 || ioctl_by_bdev(bdev, HDIO_GETGEO, (unsigned long)geo) != 0) - goto out_noioctl; + goto out_freeall; /* * Get volume label, extract name and type. @@ -92,6 +93,8 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) EBCASC(type, 4); EBCASC(name, 6); + res = 1; + /* * Three different types: CMS1, VOL1 and LNX1/unlabeled */ @@ -156,6 +159,9 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) counter++; blk++; } + if (!data) + /* Are we not supposed to report this ? */ + goto out_readerr; } else { /* * Old style LNX1 or unlabeled disk @@ -171,18 +177,17 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) } printk("\n"); - kfree(label); - kfree(geo); - kfree(info); - return 1; + goto out_freeall; + out_readerr: -out_noioctl: + res = -1; +out_freeall: kfree(label); out_nolab: kfree(geo); out_nogeo: kfree(info); -out_noinfo: - return 0; +out_exit: + return res; } @@ -830,7 +830,14 @@ void free_pipe_info(struct inode *inode) static struct vfsmount *pipe_mnt __read_mostly; static int pipefs_delete_dentry(struct dentry *dentry) { - return 1; + /* + * At creation time, we pretended this dentry was hashed + * (by clearing DCACHE_UNHASHED bit in d_flags) + * At delete time, we restore the truth : not hashed. + * (so that dput() can proceed correctly) + */ + dentry->d_flags |= DCACHE_UNHASHED; + return 0; } static struct dentry_operations pipefs_dentry_operations = { @@ -891,17 +898,22 @@ struct file *create_write_pipe(void) if (!inode) goto err_file; - sprintf(name, "[%lu]", inode->i_ino); + this.len = sprintf(name, "[%lu]", inode->i_ino); this.name = name; - this.len = strlen(name); - this.hash = inode->i_ino; /* will go */ + this.hash = 0; err = -ENOMEM; dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &this); if (!dentry) goto err_inode; dentry->d_op = &pipefs_dentry_operations; - d_add(dentry, inode); + /* + * We dont want to publish this dentry into global dentry hash table. + * We pretend dentry is already hashed, by unsetting DCACHE_UNHASHED + * This permits a working /proc/$pid/fd/XXX on pipes + */ + dentry->d_flags &= ~DCACHE_UNHASHED; + d_instantiate(dentry, inode); f->f_vfsmnt = mntget(pipe_mnt); f->f_dentry = dentry; f->f_mapping = inode->i_mapping; diff --git a/fs/proc/Makefile b/fs/proc/Makefile index 7431d7b..f6c7762 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -8,8 +8,9 @@ proc-y := nommu.o task_nommu.o proc-$(CONFIG_MMU) := mmu.o task_mmu.o proc-y += inode.o root.o base.o generic.o array.o \ - kmsg.o proc_tty.o proc_misc.o + proc_tty.o proc_misc.o proc-$(CONFIG_PROC_KCORE) += kcore.o proc-$(CONFIG_PROC_VMCORE) += vmcore.o proc-$(CONFIG_PROC_DEVICETREE) += proc_devtree.o +proc-$(CONFIG_PRINTK) += kmsg.o diff --git a/fs/proc/base.c b/fs/proc/base.c index 795319c..b859fc7 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -683,8 +683,6 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, char buffer[PROC_NUMBUF], *end; int oom_adjust; - if (!capable(CAP_SYS_RESOURCE)) - return -EPERM; memset(buffer, 0, sizeof(buffer)); if (count > sizeof(buffer) - 1) count = sizeof(buffer) - 1; @@ -699,6 +697,10 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, task = get_proc_task(file->f_dentry->d_inode); if (!task) return -ESRCH; + if (oom_adjust < task->oomkilladj && !capable(CAP_SYS_RESOURCE)) { + put_task_struct(task); + return -EACCES; + } task->oomkilladj = oom_adjust; put_task_struct(task); if (end - buffer == 0) @@ -1883,8 +1885,9 @@ out: return; } -struct dentry *proc_pid_instantiate(struct inode *dir, - struct dentry * dentry, struct task_struct *task, void *ptr) +static struct dentry *proc_pid_instantiate(struct inode *dir, + struct dentry * dentry, + struct task_struct *task, void *ptr) { struct dentry *error = ERR_PTR(-ENOENT); struct inode *inode; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 49dfb2a..e26945b 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -81,14 +81,14 @@ static void proc_read_inode(struct inode * inode) inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; } -static kmem_cache_t * proc_inode_cachep; +static struct kmem_cache * proc_inode_cachep; static struct inode *proc_alloc_inode(struct super_block *sb) { struct proc_inode *ei; struct inode *inode; - ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, SLAB_KERNEL); + ei = (struct proc_inode *)kmem_cache_alloc(proc_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->pid = NULL; @@ -105,7 +105,7 @@ static void proc_destroy_inode(struct inode *inode) kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c index 1294eda..1be7308 100644 --- a/fs/proc/kcore.c +++ b/fs/proc/kcore.c @@ -22,6 +22,7 @@ #include <asm/uaccess.h> #include <asm/io.h> +#define CORE_STR "CORE" static int open_kcore(struct inode * inode, struct file * filp) { @@ -82,10 +83,11 @@ static size_t get_kcore_size(int *nphdr, size_t *elf_buflen) } *elf_buflen = sizeof(struct elfhdr) + (*nphdr + 2)*sizeof(struct elf_phdr) + - 3 * (sizeof(struct elf_note) + 4) + - sizeof(struct elf_prstatus) + - sizeof(struct elf_prpsinfo) + - sizeof(struct task_struct); + 3 * ((sizeof(struct elf_note)) + + roundup(sizeof(CORE_STR), 4)) + + roundup(sizeof(struct elf_prstatus), 4) + + roundup(sizeof(struct elf_prpsinfo), 4) + + roundup(sizeof(struct task_struct), 4); *elf_buflen = PAGE_ALIGN(*elf_buflen); return size + *elf_buflen; } @@ -210,7 +212,7 @@ static void elf_kcore_store_hdr(char *bufp, int nphdr, int dataoff) nhdr->p_offset = offset; /* set up the process status */ - notes[0].name = "CORE"; + notes[0].name = CORE_STR; notes[0].type = NT_PRSTATUS; notes[0].datasz = sizeof(struct elf_prstatus); notes[0].data = &prstatus; @@ -221,7 +223,7 @@ static void elf_kcore_store_hdr(char *bufp, int nphdr, int dataoff) bufp = storenote(¬es[0], bufp); /* set up the process info */ - notes[1].name = "CORE"; + notes[1].name = CORE_STR; notes[1].type = NT_PRPSINFO; notes[1].datasz = sizeof(struct elf_prpsinfo); notes[1].data = &prpsinfo; @@ -238,7 +240,7 @@ static void elf_kcore_store_hdr(char *bufp, int nphdr, int dataoff) bufp = storenote(¬es[1], bufp); /* set up the task structure */ - notes[2].name = "CORE"; + notes[2].name = CORE_STR; notes[2].type = NT_TASKSTRUCT; notes[2].datasz = sizeof(struct task_struct); notes[2].data = current; diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index 93c43b6..51815ce 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -696,9 +696,11 @@ void __init proc_misc_init(void) proc_symlink("mounts", NULL, "self/mounts"); /* And now for trickier ones */ +#ifdef CONFIG_PRINTK entry = create_proc_entry("kmsg", S_IRUSR, &proc_root); if (entry) entry->proc_fops = &proc_kmsg_operations; +#endif create_seq_entry("devices", 0, &proc_devinfo_operations); create_seq_entry("cpuinfo", 0, &proc_cpuinfo_operations); #ifdef CONFIG_BLOCK diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5a41db2..c047dc6 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -515,12 +515,12 @@ static void qnx4_read_inode(struct inode *inode) brelse(bh); } -static kmem_cache_t *qnx4_inode_cachep; +static struct kmem_cache *qnx4_inode_cachep; static struct inode *qnx4_alloc_inode(struct super_block *sb) { struct qnx4_inode_info *ei; - ei = kmem_cache_alloc(qnx4_inode_cachep, SLAB_KERNEL); + ei = kmem_cache_alloc(qnx4_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -531,7 +531,7 @@ static void qnx4_destroy_inode(struct inode *inode) kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index ac14318..373d862 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -317,12 +317,11 @@ static int reiserfs_allocate_blocks_for_region(struct reiserfs_transaction_handl /* area filled with zeroes, to supply as list of zero blocknumbers We allocate it outside of loop just in case loop would spin for several iterations. */ - char *zeros = kmalloc(to_paste * UNFM_P_SIZE, GFP_ATOMIC); // We cannot insert more than MAX_ITEM_LEN bytes anyway. + char *zeros = kzalloc(to_paste * UNFM_P_SIZE, GFP_ATOMIC); // We cannot insert more than MAX_ITEM_LEN bytes anyway. if (!zeros) { res = -ENOMEM; goto error_exit_free_blocks; } - memset(zeros, 0, to_paste * UNFM_P_SIZE); do { to_paste = min_t(__u64, hole_size, @@ -407,6 +406,8 @@ static int reiserfs_allocate_blocks_for_region(struct reiserfs_transaction_handl we restart it. This will also free the path. */ if (journal_transaction_should_end (th, th->t_blocks_allocated)) { + inode->i_size = cpu_key_k_offset(&key) + + (to_paste << inode->i_blkbits); res = restart_transaction(th, inode, &path); @@ -1045,6 +1046,7 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode char *kaddr = kmap_atomic(prepared_pages[0], KM_USER0); memset(kaddr, 0, from); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(prepared_pages[0]); } if (to != PAGE_CACHE_SIZE) { /* Last page needs to be partially zeroed */ char *kaddr = @@ -1052,6 +1054,7 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode KM_USER0); memset(kaddr + to, 0, PAGE_CACHE_SIZE - to); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(prepared_pages[num_pages - 1]); } /* Since all blocks are new - use already calculated value */ @@ -1185,6 +1188,7 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode memset(kaddr + block_start, 0, from - block_start); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(prepared_pages[0]); set_buffer_uptodate(bh); } } @@ -1222,6 +1226,7 @@ static int reiserfs_prepare_file_region_for_write(struct inode *inode KM_USER0); memset(kaddr + to, 0, block_end - to); kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(prepared_pages[num_pages - 1]); set_buffer_uptodate(bh); } } @@ -1307,56 +1312,8 @@ static ssize_t reiserfs_file_write(struct file *file, /* the file we are going t count = MAX_NON_LFS - (unsigned long)*ppos; } - if (file->f_flags & O_DIRECT) { // Direct IO needs treatment - ssize_t result, after_file_end = 0; - if ((*ppos + count >= inode->i_size) - || (file->f_flags & O_APPEND)) { - /* If we are appending a file, we need to put this savelink in here. - If we will crash while doing direct io, finish_unfinished will - cut the garbage from the file end. */ - reiserfs_write_lock(inode->i_sb); - err = - journal_begin(&th, inode->i_sb, - JOURNAL_PER_BALANCE_CNT); - if (err) { - reiserfs_write_unlock(inode->i_sb); - return err; - } - reiserfs_update_inode_transaction(inode); - add_save_link(&th, inode, 1 /* Truncate */ ); - after_file_end = 1; - err = - journal_end(&th, inode->i_sb, - JOURNAL_PER_BALANCE_CNT); - reiserfs_write_unlock(inode->i_sb); - if (err) - return err; - } - result = do_sync_write(file, buf, count, ppos); - - if (after_file_end) { /* Now update i_size and remove the savelink */ - struct reiserfs_transaction_handle th; - reiserfs_write_lock(inode->i_sb); - err = journal_begin(&th, inode->i_sb, 1); - if (err) { - reiserfs_write_unlock(inode->i_sb); - return err; - } - reiserfs_update_inode_transaction(inode); - mark_inode_dirty(inode); - err = journal_end(&th, inode->i_sb, 1); - if (err) { - reiserfs_write_unlock(inode->i_sb); - return err; - } - err = remove_save_link(inode, 1 /* truncate */ ); - reiserfs_write_unlock(inode->i_sb); - if (err) - return err; - } - - return result; - } + if (file->f_flags & O_DIRECT) + return do_sync_write(file, buf, count, ppos); if (unlikely((ssize_t) count < 0)) return -EINVAL; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 9c69bca..254239e 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -216,11 +216,12 @@ static int file_capable(struct inode *inode, long block) BUG_ON(!th->t_trans_id); BUG_ON(!th->t_refcount); + pathrelse(path); + /* we cannot restart while nested */ if (th->t_refcount > 1) { return 0; } - pathrelse(path); reiserfs_update_sd(th, inode); err = journal_end(th, s, len); if (!err) { @@ -928,15 +929,12 @@ int reiserfs_get_block(struct inode *inode, sector_t block, if (blocks_needed == 1) { un = &unf_single; } else { - un = kmalloc(min(blocks_needed, max_to_insert) * UNFM_P_SIZE, GFP_ATOMIC); // We need to avoid scheduling. + un = kzalloc(min(blocks_needed, max_to_insert) * UNFM_P_SIZE, GFP_ATOMIC); // We need to avoid scheduling. if (!un) { un = &unf_single; blocks_needed = 1; max_to_insert = 0; - } else - memset(un, 0, - UNFM_P_SIZE * min(blocks_needed, - max_to_insert)); + } } if (blocks_needed <= max_to_insert) { /* we are going to add target block to the file. Use allocated diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 1724999..7fb5fb0 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -490,13 +490,13 @@ static void reiserfs_put_super(struct super_block *s) return; } -static kmem_cache_t *reiserfs_inode_cachep; +static struct kmem_cache *reiserfs_inode_cachep; static struct inode *reiserfs_alloc_inode(struct super_block *sb) { struct reiserfs_inode_info *ei; ei = (struct reiserfs_inode_info *) - kmem_cache_alloc(reiserfs_inode_cachep, SLAB_KERNEL); + kmem_cache_alloc(reiserfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -507,7 +507,7 @@ static void reiserfs_destroy_inode(struct inode *inode) kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } -static void init_once(void *foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void *foo, struct kmem_cache * cachep, unsigned long flags) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; @@ -1549,13 +1549,12 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) struct reiserfs_sb_info *sbi; int errval = -EINVAL; - sbi = kmalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL); + sbi = kzalloc(sizeof(struct reiserfs_sb_info), GFP_KERNEL); if (!sbi) { errval = -ENOMEM; goto error; } s->s_fs_info = sbi; - memset(sbi, 0, sizeof(struct reiserfs_sb_info)); /* Set default values for options: non-aggressive tails, RO on errors */ REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_SMALLTAIL); REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_ERROR_RO); diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c index ddcd9e1..c5af088 100644 --- a/fs/romfs/inode.c +++ b/fs/romfs/inode.c @@ -550,12 +550,12 @@ romfs_read_inode(struct inode *i) } } -static kmem_cache_t * romfs_inode_cachep; +static struct kmem_cache * romfs_inode_cachep; static struct inode *romfs_alloc_inode(struct super_block *sb) { struct romfs_inode_info *ei; - ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, SLAB_KERNEL); + ei = (struct romfs_inode_info *)kmem_cache_alloc(romfs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -566,7 +566,7 @@ static void romfs_destroy_inode(struct inode *inode) kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct romfs_inode_info *ei = (struct romfs_inode_info *) foo; diff --git a/fs/seq_file.c b/fs/seq_file.c index 555b9ac..10690aa 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -26,7 +26,7 @@ * ERR_PTR(error). In the end of sequence they return %NULL. ->show() * returns 0 in case of success and negative number in case of error. */ -int seq_open(struct file *file, struct seq_operations *op) +int seq_open(struct file *file, const struct seq_operations *op) { struct seq_file *p = file->private_data; @@ -408,7 +408,7 @@ EXPORT_SYMBOL(single_open); int single_release(struct inode *inode, struct file *file) { - struct seq_operations *op = ((struct seq_file *)file->private_data)->op; + const struct seq_operations *op = ((struct seq_file *)file->private_data)->op; int res = seq_release(inode, file); kfree(op); return res; diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 2c122ee..4af4cd7 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -50,12 +50,12 @@ static void smb_put_super(struct super_block *); static int smb_statfs(struct dentry *, struct kstatfs *); static int smb_show_options(struct seq_file *, struct vfsmount *); -static kmem_cache_t *smb_inode_cachep; +static struct kmem_cache *smb_inode_cachep; static struct inode *smb_alloc_inode(struct super_block *sb) { struct smb_inode_info *ei; - ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, SLAB_KERNEL); + ei = (struct smb_inode_info *)kmem_cache_alloc(smb_inode_cachep, GFP_KERNEL); if (!ei) return NULL; return &ei->vfs_inode; @@ -66,7 +66,7 @@ static void smb_destroy_inode(struct inode *inode) kmem_cache_free(smb_inode_cachep, SMB_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct smb_inode_info *ei = (struct smb_inode_info *) foo; unsigned long flagmask = SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR; diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 0fb7469..a4bcae8 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -25,7 +25,7 @@ #define ROUND_UP(x) (((x)+3) & ~3) /* cache for request structures */ -static kmem_cache_t *req_cachep; +static struct kmem_cache *req_cachep; static int smb_request_send_req(struct smb_request *req); @@ -61,7 +61,7 @@ static struct smb_request *smb_do_alloc_request(struct smb_sb_info *server, struct smb_request *req; unsigned char *buf = NULL; - req = kmem_cache_alloc(req_cachep, SLAB_KERNEL); + req = kmem_cache_alloc(req_cachep, GFP_KERNEL); VERBOSE("allocating request: %p\n", req); if (!req) goto out; @@ -51,13 +51,6 @@ int vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) return inode->i_op->getattr(mnt, dentry, stat); generic_fillattr(inode, stat); - if (!stat->blksize) { - struct super_block *s = inode->i_sb; - unsigned blocks; - blocks = (stat->size+s->s_blocksize-1) >> s->s_blocksize_bits; - stat->blocks = (s->s_blocksize / 512) * blocks; - stat->blksize = s->s_blocksize; - } return 0; } diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index 20551a1..e503f85 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -16,7 +16,7 @@ struct vfsmount *sysfs_mount; struct super_block * sysfs_sb = NULL; -kmem_cache_t *sysfs_dir_cachep; +struct kmem_cache *sysfs_dir_cachep; static struct super_operations sysfs_ops = { .statfs = simple_statfs, diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 6f3d6bd..bd7cec2 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -1,6 +1,6 @@ extern struct vfsmount * sysfs_mount; -extern kmem_cache_t *sysfs_dir_cachep; +extern struct kmem_cache *sysfs_dir_cachep; extern struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent *); extern int sysfs_create(struct dentry *, int mode, int (*init)(struct inode *)); diff --git a/fs/sysv/CHANGES b/fs/sysv/CHANGES deleted file mode 100644 index 66ea6e9..0000000 --- a/fs/sysv/CHANGES +++ /dev/null @@ -1,60 +0,0 @@ -Mon, 15 Dec 1997 Krzysztof G. Baranowski <kgb@manjak.knm.org.pl> - * namei.c: struct sysv_dir_inode_operations updated to use dentries. - -Fri, 23 Jan 1998 Krzysztof G. Baranowski <kgb@manjak.knm.org.pl> - * inode.c: corrected 1 track offset setting (in sb->sv_block_base). - Originally it was overridden (by setting to zero) - in detected_[xenix,sysv4,sysv2,coherent]. Thanks - to Andrzej Krzysztofowicz <ankry@mif.pg.gda.pl> - for identifying the problem. - -Tue, 27 Jan 1998 Krzysztof G. Baranowski <kgb@manjak.knm.org.pl> - * inode.c: added 2048-byte block support to SystemV FS. - Merged detected_bs[512,1024,2048]() into one function: - void detected_bs (u_char type, struct super_block *sb). - Thanks to Andrzej Krzysztofowicz <ankry@mif.pg.gda.pl> - for the patch. - -Wed, 4 Feb 1998 Krzysztof G. Baranowski <kgb@manjak.knm.org.pl> - * namei.c: removed static subdir(); is_subdir() from dcache.c - is used instead. Cosmetic changes. - -Thu, 3 Dec 1998 Al Viro (viro@parcelfarce.linux.theplanet.co.uk) - * namei.c (sysv_rmdir): - Bugectomy: old check for victim being busy - (inode->i_count) wasn't replaced (with checking - dentry->d_count) and escaped Linus in the last round - of changes. Shot and buried. - -Wed, 9 Dec 1998 AV - * namei.c (do_sysv_rename): - Fixed incorrect check for other owners + race. - Removed checks that went to VFS. - * namei.c (sysv_unlink): - Removed checks that went to VFS. - -Thu, 10 Dec 1998 AV - * namei.c (do_mknod): - Removed dead code - mknod is never asked to - create a symlink or directory. Incidentially, - it wouldn't do it right if it would be called. - -Sat, 26 Dec 1998 KGB - * inode.c (detect_sysv4): - Added detection of expanded s_type field (0x10, - 0x20 and 0x30). Forced read-only access in this case. - -Sun, 21 Mar 1999 AV - * namei.c (sysv_link): - Fixed i_count usage that resulted in dcache corruption. - * inode.c: - Filled ->delete_inode() method with sysv_delete_inode(). - sysv_put_inode() is gone, as it tried to do ->delete_ - _inode()'s job. - * ialloc.c: (sysv_free_inode): - Fixed race. - -Sun, 30 Apr 1999 AV - * namei.c (sysv_mknod): - Removed dead code (S_IFREG case is now passed to - ->create() by VFS). diff --git a/fs/sysv/ChangeLog b/fs/sysv/ChangeLog deleted file mode 100644 index f403f8b..0000000 --- a/fs/sysv/ChangeLog +++ /dev/null @@ -1,106 +0,0 @@ -Thu Feb 14 2002 Andrew Morton <akpm@zip.com.au> - - * dir_commit_chunk(): call writeout_one_page() as well as - waitfor_one_page() for IS_SYNC directories, so that we - actually do sync the directory. (forward-port from 2.4). - -Thu Feb 7 2002 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - - * super.c: switched to ->get_sb() - * ChangeLog: fixed dates ;-) - -2002-01-24 David S. Miller <davem@redhat.com> - - * inode.c: Include linux/init.h - -Mon Jan 21 2002 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - * ialloc.c (sysv_new_inode): zero SYSV_I(inode)->i_data out. - * i_vnode renamed to vfs_inode. Sorry, but let's keep that - consistent. - -Sat Jan 19 2002 Christoph Hellwig <hch@infradead.org> - - * include/linux/sysv_fs.h (SYSV_I): Get fs-private inode data using - list_entry() instead of inode->u. - * include/linux/sysv_fs_i.h: Add 'struct inode i_vnode' field to - sysv_inode_info structure. - * inode.c: Include <linux/slab.h>, implement alloc_inode/destroy_inode - sop methods, add infrastructure for per-fs inode slab cache. - * super.c (init_sysv_fs): Initialize inode cache, recover properly - in the case of failed register_filesystem for V7. - (exit_sysv_fs): Destroy inode cache. - -Sat Jan 19 2002 Christoph Hellwig <hch@infradead.org> - - * include/linux/sysv_fs.h: Include <linux/sysv_fs_i.h>, declare SYSV_I(). - * dir.c (sysv_find_entry): Use SYSV_I() instead of ->u.sysv_i to - access fs-private inode data. - * ialloc.c (sysv_new_inode): Likewise. - * inode.c (sysv_read_inode): Likewise. - (sysv_update_inode): Likewise. - * itree.c (get_branch): Likewise. - (sysv_truncate): Likewise. - * symlink.c (sysv_readlink): Likewise. - (sysv_follow_link): Likewise. - -Fri Jan 4 2002 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - - * ialloc.c (sysv_free_inode): Use sb->s_id instead of bdevname(). - * inode.c (sysv_read_inode): Likewise. - (sysv_update_inode): Likewise. - (sysv_sync_inode): Likewise. - * super.c (detect_sysv): Likewise. - (complete_read_super): Likewise. - (sysv_read_super): Likewise. - (v7_read_super): Likewise. - -Sun Dec 30 2001 Manfred Spraul <manfred@colorfullife.com> - - * dir.c (dir_commit_chunk): Do not set dir->i_version. - (sysv_readdir): Likewise. - -Thu Dec 27 2001 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - - * itree.c (get_block): Use map_bh() to fill out bh_result. - -Tue Dec 25 2001 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - - * super.c (sysv_read_super): Use sb_set_blocksize() to set blocksize. - (v7_read_super): Likewise. - -Tue Nov 27 2001 Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk> - - * itree.c (get_block): Change type for iblock argument to sector_t. - * super.c (sysv_read_super): Set s_blocksize early. - (v7_read_super): Likewise. - * balloc.c (sysv_new_block): Use sb_bread(). instead of bread(). - (sysv_count_free_blocks): Likewise. - * ialloc.c (sysv_raw_inode): Likewise. - * itree.c (get_branch): Likewise. - (free_branches): Likewise. - * super.c (sysv_read_super): Likewise. - (v7_read_super): Likewise. - -Sat Dec 15 2001 Christoph Hellwig <hch@infradead.org> - - * inode.c (sysv_read_inode): Mark inode as bad in case of failure. - * super.c (complete_read_super): Check for bad root inode. - -Wed Nov 21 2001 Andrew Morton <andrewm@uow.edu.au> - - * file.c (sysv_sync_file): Call fsync_inode_data_buffers. - -Fri Oct 26 2001 Christoph Hellwig <hch@infradead.org> - - * dir.c, ialloc.c, namei.c, include/linux/sysv_fs_i.h: - Implement per-Inode lookup offset cache. - Modelled after Ted's ext2 patch. - -Fri Oct 26 2001 Christoph Hellwig <hch@infradead.org> - - * inode.c, super.c, include/linux/sysv_fs.h, - include/linux/sysv_fs_sb.h: - Remove symlink faking. Noone really wants to use these as - linux filesystems and native OSes don't support it anyway. - - diff --git a/fs/sysv/INTRO b/fs/sysv/INTRO deleted file mode 100644 index de4e4d1..0000000 --- a/fs/sysv/INTRO +++ /dev/null @@ -1,182 +0,0 @@ -This is the implementation of the SystemV/Coherent filesystem for Linux. -It grew out of separate filesystem implementations - - Xenix FS Doug Evans <dje@cygnus.com> June 1992 - SystemV FS Paul B. Monday <pmonday@eecs.wsu.edu> March-June 1993 - Coherent FS B. Haible <haible@ma2s2.mathematik.uni-karlsruhe.de> June 1993 - -and was merged together in July 1993. - -These filesystems are rather similar. Here is a comparison with Minix FS: - -* Linux fdisk reports on partitions - - Minix FS 0x81 Linux/Minix - - Xenix FS ?? - - SystemV FS ?? - - Coherent FS 0x08 AIX bootable - -* Size of a block or zone (data allocation unit on disk) - - Minix FS 1024 - - Xenix FS 1024 (also 512 ??) - - SystemV FS 1024 (also 512 and 2048) - - Coherent FS 512 - -* General layout: all have one boot block, one super block and - separate areas for inodes and for directories/data. - On SystemV Release 2 FS (e.g. Microport) the first track is reserved and - all the block numbers (including the super block) are offset by one track. - -* Byte ordering of "short" (16 bit entities) on disk: - - Minix FS little endian 0 1 - - Xenix FS little endian 0 1 - - SystemV FS little endian 0 1 - - Coherent FS little endian 0 1 - Of course, this affects only the file system, not the data of files on it! - -* Byte ordering of "long" (32 bit entities) on disk: - - Minix FS little endian 0 1 2 3 - - Xenix FS little endian 0 1 2 3 - - SystemV FS little endian 0 1 2 3 - - Coherent FS PDP-11 2 3 0 1 - Of course, this affects only the file system, not the data of files on it! - -* Inode on disk: "short", 0 means non-existent, the root dir ino is: - - Minix FS 1 - - Xenix FS, SystemV FS, Coherent FS 2 - -* Maximum number of hard links to a file: - - Minix FS 250 - - Xenix FS ?? - - SystemV FS ?? - - Coherent FS >=10000 - -* Free inode management: - - Minix FS a bitmap - - Xenix FS, SystemV FS, Coherent FS - There is a cache of a certain number of free inodes in the super-block. - When it is exhausted, new free inodes are found using a linear search. - -* Free block management: - - Minix FS a bitmap - - Xenix FS, SystemV FS, Coherent FS - Free blocks are organized in a "free list". Maybe a misleading term, - since it is not true that every free block contains a pointer to - the next free block. Rather, the free blocks are organized in chunks - of limited size, and every now and then a free block contains pointers - to the free blocks pertaining to the next chunk; the first of these - contains pointers and so on. The list terminates with a "block number" - 0 on Xenix FS and SystemV FS, with a block zeroed out on Coherent FS. - -* Super-block location: - - Minix FS block 1 = bytes 1024..2047 - - Xenix FS block 1 = bytes 1024..2047 - - SystemV FS bytes 512..1023 - - Coherent FS block 1 = bytes 512..1023 - -* Super-block layout: - - Minix FS - unsigned short s_ninodes; - unsigned short s_nzones; - unsigned short s_imap_blocks; - unsigned short s_zmap_blocks; - unsigned short s_firstdatazone; - unsigned short s_log_zone_size; - unsigned long s_max_size; - unsigned short s_magic; - - Xenix FS, SystemV FS, Coherent FS - unsigned short s_firstdatazone; - unsigned long s_nzones; - unsigned short s_fzone_count; - unsigned long s_fzones[NICFREE]; - unsigned short s_finode_count; - unsigned short s_finodes[NICINOD]; - char s_flock; - char s_ilock; - char s_modified; - char s_rdonly; - unsigned long s_time; - short s_dinfo[4]; -- SystemV FS only - unsigned long s_free_zones; - unsigned short s_free_inodes; - short s_dinfo[4]; -- Xenix FS only - unsigned short s_interleave_m,s_interleave_n; -- Coherent FS only - char s_fname[6]; - char s_fpack[6]; - then they differ considerably: - Xenix FS - char s_clean; - char s_fill[371]; - long s_magic; - long s_type; - SystemV FS - long s_fill[12 or 14]; - long s_state; - long s_magic; - long s_type; - Coherent FS - unsigned long s_unique; - Note that Coherent FS has no magic. - -* Inode layout: - - Minix FS - unsigned short i_mode; - unsigned short i_uid; - unsigned long i_size; - unsigned long i_time; - unsigned char i_gid; - unsigned char i_nlinks; - unsigned short i_zone[7+1+1]; - - Xenix FS, SystemV FS, Coherent FS - unsigned short i_mode; - unsigned short i_nlink; - unsigned short i_uid; - unsigned short i_gid; - unsigned long i_size; - unsigned char i_zone[3*(10+1+1+1)]; - unsigned long i_atime; - unsigned long i_mtime; - unsigned long i_ctime; - -* Regular file data blocks are organized as - - Minix FS - 7 direct blocks - 1 indirect block (pointers to blocks) - 1 double-indirect block (pointer to pointers to blocks) - - Xenix FS, SystemV FS, Coherent FS - 10 direct blocks - 1 indirect block (pointers to blocks) - 1 double-indirect block (pointer to pointers to blocks) - 1 triple-indirect block (pointer to pointers to pointers to blocks) - -* Inode size, inodes per block - - Minix FS 32 32 - - Xenix FS 64 16 - - SystemV FS 64 16 - - Coherent FS 64 8 - -* Directory entry on disk - - Minix FS - unsigned short inode; - char name[14/30]; - - Xenix FS, SystemV FS, Coherent FS - unsigned short inode; - char name[14]; - -* Dir entry size, dir entries per block - - Minix FS 16/32 64/32 - - Xenix FS 16 64 - - SystemV FS 16 64 - - Coherent FS 16 32 - -* How to implement symbolic links such that the host fsck doesn't scream: - - Minix FS normal - - Xenix FS kludge: as regular files with chmod 1000 - - SystemV FS ?? - - Coherent FS kludge: as regular files with chmod 1000 - - -Notation: We often speak of a "block" but mean a zone (the allocation unit) -and not the disk driver's notion of "block". - - -Bruno Haible <haible@ma2s2.mathematik.uni-karlsruhe.de> diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index d63c5e4..ead9864 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -301,13 +301,13 @@ static void sysv_delete_inode(struct inode *inode) unlock_kernel(); } -static kmem_cache_t *sysv_inode_cachep; +static struct kmem_cache *sysv_inode_cachep; static struct inode *sysv_alloc_inode(struct super_block *sb) { struct sysv_inode_info *si; - si = kmem_cache_alloc(sysv_inode_cachep, SLAB_KERNEL); + si = kmem_cache_alloc(sysv_inode_cachep, GFP_KERNEL); if (!si) return NULL; return &si->vfs_inode; @@ -318,7 +318,7 @@ static void sysv_destroy_inode(struct inode *inode) kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } -static void init_once(void *p, kmem_cache_t *cachep, unsigned long flags) +static void init_once(void *p, struct kmem_cache *cachep, unsigned long flags) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/udf/super.c b/fs/udf/super.c index 1aea6a4..1dbc295 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -107,12 +107,12 @@ static struct file_system_type udf_fstype = { .fs_flags = FS_REQUIRES_DEV, }; -static kmem_cache_t * udf_inode_cachep; +static struct kmem_cache * udf_inode_cachep; static struct inode *udf_alloc_inode(struct super_block *sb) { struct udf_inode_info *ei; - ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, SLAB_KERNEL); + ei = (struct udf_inode_info *)kmem_cache_alloc(udf_inode_cachep, GFP_KERNEL); if (!ei) return NULL; @@ -130,7 +130,7 @@ static void udf_destroy_inode(struct inode *inode) kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct udf_inode_info *ei = (struct udf_inode_info *) foo; @@ -1709,7 +1709,7 @@ void udf_error(struct super_block *sb, const char *function, sb->s_dirt = 1; } va_start(args, fmt); - vsprintf(error_buf, fmt, args); + vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); printk (KERN_CRIT "UDF-fs error (device %s): %s: %s\n", sb->s_id, function, error_buf); @@ -1721,7 +1721,7 @@ void udf_warning(struct super_block *sb, const char *function, va_list args; va_start (args, fmt); - vsprintf(error_buf, fmt, args); + vsnprintf(error_buf, sizeof(error_buf), fmt, args); va_end(args); printk(KERN_WARNING "UDF-fs warning (device %s): %s: %s\n", sb->s_id, function, error_buf); diff --git a/fs/ufs/super.c b/fs/ufs/super.c index ec79e30..8a8e938 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -224,7 +224,7 @@ void ufs_error (struct super_block * sb, const char * function, sb->s_flags |= MS_RDONLY; } va_start (args, fmt); - vsprintf (error_buf, fmt, args); + vsnprintf (error_buf, sizeof(error_buf), fmt, args); va_end (args); switch (UFS_SB(sb)->s_mount_opt & UFS_MOUNT_ONERROR) { case UFS_MOUNT_ONERROR_PANIC: @@ -255,7 +255,7 @@ void ufs_panic (struct super_block * sb, const char * function, sb->s_dirt = 1; } va_start (args, fmt); - vsprintf (error_buf, fmt, args); + vsnprintf (error_buf, sizeof(error_buf), fmt, args); va_end (args); sb->s_flags |= MS_RDONLY; printk (KERN_CRIT "UFS-fs panic (device %s): %s: %s\n", @@ -268,7 +268,7 @@ void ufs_warning (struct super_block * sb, const char * function, va_list args; va_start (args, fmt); - vsprintf (error_buf, fmt, args); + vsnprintf (error_buf, sizeof(error_buf), fmt, args); va_end (args); printk (KERN_WARNING "UFS-fs warning (device %s): %s: %s\n", sb->s_id, function, error_buf); @@ -1204,12 +1204,12 @@ static int ufs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static kmem_cache_t * ufs_inode_cachep; +static struct kmem_cache * ufs_inode_cachep; static struct inode *ufs_alloc_inode(struct super_block *sb) { struct ufs_inode_info *ei; - ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, SLAB_KERNEL); + ei = (struct ufs_inode_info *)kmem_cache_alloc(ufs_inode_cachep, GFP_KERNEL); if (!ei) return NULL; ei->vfs_inode.i_version = 1; @@ -1221,7 +1221,7 @@ static void ufs_destroy_inode(struct inode *inode) kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } -static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +static void init_once(void * foo, struct kmem_cache * cachep, unsigned long flags) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/fs/ufs/util.h b/fs/ufs/util.h index 28fce6c..7dd12bb 100644 --- a/fs/ufs/util.h +++ b/fs/ufs/util.h @@ -299,7 +299,7 @@ static inline void *get_usb_offset(struct ufs_sb_private_info *uspi, #define ubh_get_addr16(ubh,begin) \ (((__fs16*)((ubh)->bh[(begin) >> (uspi->s_fshift-1)]->b_data)) + \ - ((begin) & (uspi->fsize>>1) - 1))) + ((begin) & ((uspi->fsize>>1) - 1))) #define ubh_get_addr32(ubh,begin) \ (((__fs32*)((ubh)->bh[(begin) >> (uspi->s_fshift-2)]->b_data)) + \ diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index eef4a0b..4fb01ff 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -32,6 +32,7 @@ #include <linux/kthread.h> #include <linux/migrate.h> #include <linux/backing-dev.h> +#include <linux/freezer.h> STATIC kmem_zone_t *xfs_buf_zone; STATIC kmem_shaker_t xfs_buf_shake; @@ -1826,11 +1827,11 @@ xfs_buf_init(void) if (!xfs_buf_zone) goto out_free_trace_buf; - xfslogd_workqueue = create_workqueue("xfslogd"); + xfslogd_workqueue = create_freezeable_workqueue("xfslogd"); if (!xfslogd_workqueue) goto out_free_buf_zone; - xfsdatad_workqueue = create_workqueue("xfsdatad"); + xfsdatad_workqueue = create_freezeable_workqueue("xfsdatad"); if (!xfsdatad_workqueue) goto out_destroy_xfslogd_workqueue; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index de05abb..b93265b 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -56,6 +56,7 @@ #include <linux/mempool.h> #include <linux/writeback.h> #include <linux/kthread.h> +#include <linux/freezer.h> STATIC struct quotactl_ops xfs_quotactl_operations; STATIC struct super_operations xfs_super_operations; |