From dc32f63453f56d07a1073a697dcd843dd3098c09 Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Mon, 30 Jul 2012 14:39:04 -0700 Subject: mm: fix wrong argument of migrate_huge_pages() in soft_offline_huge_page() Commit a6bc32b89922 ("mm: compaction: introduce sync-light migration for use by compaction") changed the declaration of migrate_pages() and migrate_huge_pages(). But it missed changing the argument of migrate_huge_pages() in soft_offline_huge_page(). In this case, we should call migrate_huge_pages() with MIGRATE_SYNC. Additionally, there is a mismatch between type the of argument and the function declaration for migrate_pages(). Signed-off-by: Joonsoo Kim Cc: Christoph Lameter Cc: Mel Gorman Acked-by: David Rientjes Cc: "Aneesh Kumar K.V" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memory-failure.c b/mm/memory-failure.c index de4ce70..6de0d61 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1433,8 +1433,8 @@ static int soft_offline_huge_page(struct page *page, int flags) /* Keep page count to indicate a given hugepage is isolated. */ list_add(&hpage->lru, &pagelist); - ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0, - true); + ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, false, + MIGRATE_SYNC); if (ret) { struct page *page1, *page2; list_for_each_entry_safe(page1, page2, &pagelist, lru) @@ -1563,7 +1563,7 @@ int soft_offline_page(struct page *page, int flags) page_is_file_cache(page)); list_add(&page->lru, &pagelist); ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, - 0, MIGRATE_SYNC); + false, MIGRATE_SYNC); if (ret) { putback_lru_pages(&pagelist); pr_info("soft offline: %#lx: migration failed %d, type %lx\n", -- cgit v0.10.2 From 6c4088ac3a4d82779903433bcd5f048c58fb1aca Mon Sep 17 00:00:00 2001 From: Greg Pearson Date: Mon, 30 Jul 2012 14:39:05 -0700 Subject: pcdp: use early_ioremap/early_iounmap to access pcdp table efi_setup_pcdp_console() is called during boot to parse the HCDP/PCDP EFI system table and setup an early console for printk output. The routine uses ioremap/iounmap to setup access to the HCDP/PCDP table information. The call to ioremap is happening early in the boot process which leads to a panic on x86_64 systems: panic+0x01ca do_exit+0x043c oops_end+0x00a7 no_context+0x0119 __bad_area_nosemaphore+0x0138 bad_area_nosemaphore+0x000e do_page_fault+0x0321 page_fault+0x0020 reserve_memtype+0x02a1 __ioremap_caller+0x0123 ioremap_nocache+0x0012 efi_setup_pcdp_console+0x002b setup_arch+0x03a9 start_kernel+0x00d4 x86_64_start_reservations+0x012c x86_64_start_kernel+0x00fe This replaces the calls to ioremap/iounmap in efi_setup_pcdp_console() with calls to early_ioremap/early_iounmap which can be called during early boot. This patch was tested on an x86_64 prototype system which uses the HCDP/PCDP table for early console setup. Signed-off-by: Greg Pearson Acked-by: Khalid Aziz Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/firmware/pcdp.c b/drivers/firmware/pcdp.c index 51e0e2d..a330492 100644 --- a/drivers/firmware/pcdp.c +++ b/drivers/firmware/pcdp.c @@ -95,7 +95,7 @@ efi_setup_pcdp_console(char *cmdline) if (efi.hcdp == EFI_INVALID_TABLE_ADDR) return -ENODEV; - pcdp = ioremap(efi.hcdp, 4096); + pcdp = early_ioremap(efi.hcdp, 4096); printk(KERN_INFO "PCDP: v%d at 0x%lx\n", pcdp->rev, efi.hcdp); if (strstr(cmdline, "console=hcdp")) { @@ -131,6 +131,6 @@ efi_setup_pcdp_console(char *cmdline) } out: - iounmap(pcdp); + early_iounmap(pcdp, 4096); return rc; } -- cgit v0.10.2 From f1fd75bfa07822b1de314062baff3280419a8bf4 Mon Sep 17 00:00:00 2001 From: Sasikantha babu Date: Mon, 30 Jul 2012 14:39:08 -0700 Subject: prctl: remove redunant assignment of "error" to zero Just setting the "error" to error number is enough on failure and It doesn't require to set "error" variable to zero in each switch case, since it was already initialized with zero. And also removed return 0 in switch case with break statement Signed-off-by: Sasikantha babu Acked-by: Kees Cook Acked-by: Serge E. Hallyn Cc: Cyrill Gorcunov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sys.c b/kernel/sys.c index 2d39a84..b04ae03 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2015,7 +2015,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; } me->pdeath_signal = arg2; - error = 0; break; case PR_GET_PDEATHSIG: error = put_user(me->pdeath_signal, (int __user *)arg2); @@ -2029,7 +2028,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; } set_dumpable(me->mm, arg2); - error = 0; break; case PR_SET_UNALIGN: @@ -2056,10 +2054,7 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SET_TIMING: if (arg2 != PR_TIMING_STATISTICAL) error = -EINVAL; - else - error = 0; break; - case PR_SET_NAME: comm[sizeof(me->comm)-1] = 0; if (strncpy_from_user(comm, (char __user *)arg2, @@ -2067,20 +2062,19 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, return -EFAULT; set_task_comm(me, comm); proc_comm_connector(me); - return 0; + break; case PR_GET_NAME: get_task_comm(comm, me); if (copy_to_user((char __user *)arg2, comm, sizeof(comm))) return -EFAULT; - return 0; + break; case PR_GET_ENDIAN: error = GET_ENDIAN(me, arg2); break; case PR_SET_ENDIAN: error = SET_ENDIAN(me, arg2); break; - case PR_GET_SECCOMP: error = prctl_get_seccomp(); break; @@ -2108,7 +2102,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, current->default_timer_slack_ns; else current->timer_slack_ns = arg2; - error = 0; break; case PR_MCE_KILL: if (arg4 | arg5) @@ -2134,7 +2127,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, default: return -EINVAL; } - error = 0; break; case PR_MCE_KILL_GET: if (arg2 | arg3 | arg4 | arg5) @@ -2153,7 +2145,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; case PR_SET_CHILD_SUBREAPER: me->signal->is_child_subreaper = !!arg2; - error = 0; break; case PR_GET_CHILD_SUBREAPER: error = put_user(me->signal->is_child_subreaper, -- cgit v0.10.2 From 32b4560b04af6e4fee241ea6de6db780eaf354f2 Mon Sep 17 00:00:00 2001 From: Fernando Luis Vazquez Cao Date: Mon, 30 Jul 2012 14:39:10 -0700 Subject: ntfs: remove references to long gone super operations and unimplemented methods ->delete_inode(), ->write_super_lockfs(), ->unlockfs() are gone so remove refereces to them in the NTFS code. Remove unnecessary comments about unimplemented methods while at it (suggested by Christoph Hellwig). Noticed while cleaning up the fsfreeze mess. Signed-off-by: Fernando Luis Vazquez Cao Cc: Anton Altaparmakov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index b341492..2bc149d 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -2660,31 +2660,14 @@ static const struct super_operations ntfs_sops = { .alloc_inode = ntfs_alloc_big_inode, /* VFS: Allocate new inode. */ .destroy_inode = ntfs_destroy_big_inode, /* VFS: Deallocate inode. */ #ifdef NTFS_RW - //.dirty_inode = NULL, /* VFS: Called from - // __mark_inode_dirty(). */ .write_inode = ntfs_write_inode, /* VFS: Write dirty inode to disk. */ - //.drop_inode = NULL, /* VFS: Called just after the - // inode reference count has - // been decreased to zero. - // NOTE: The inode lock is - // held. See fs/inode.c:: - // generic_drop_inode(). */ - //.delete_inode = NULL, /* VFS: Delete inode from disk. - // Called when i_count becomes - // 0 and i_nlink is also 0. */ - //.write_super = NULL, /* Flush dirty super block to - // disk. */ - //.sync_fs = NULL, /* ? */ - //.write_super_lockfs = NULL, /* ? */ - //.unlockfs = NULL, /* ? */ #endif /* NTFS_RW */ .put_super = ntfs_put_super, /* Syscall: umount. */ .statfs = ntfs_statfs, /* Syscall: statfs */ .remount_fs = ntfs_remount, /* Syscall: mount -o remount. */ .evict_inode = ntfs_evict_big_inode, /* VFS: Called when an inode is removed from memory. */ - //.umount_begin = NULL, /* Forced umount. */ .show_options = ntfs_show_options, /* Show mount options in proc. */ }; -- cgit v0.10.2 From 779302e67835fe9a6b74327e54969ba59cb3478a Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Mon, 30 Jul 2012 14:39:13 -0700 Subject: fs/xattr.c:getxattr(): improve handling of allocation failures This allocation can be as large as 64k. - Add __GFP_NOWARN so the falied kmalloc() is silent - Fall back to vmalloc() if the kmalloc() failed Signed-off-by: Sasha Levin Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/xattr.c b/fs/xattr.c index 1d7ac37..4d45b71 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -427,6 +427,7 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, { ssize_t error; void *kvalue = NULL; + void *vvalue = NULL; char kname[XATTR_NAME_MAX + 1]; error = strncpy_from_user(kname, name, sizeof(kname)); @@ -438,9 +439,13 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, if (size) { if (size > XATTR_SIZE_MAX) size = XATTR_SIZE_MAX; - kvalue = kzalloc(size, GFP_KERNEL); - if (!kvalue) - return -ENOMEM; + kvalue = kzalloc(size, GFP_KERNEL | __GFP_NOWARN); + if (!kvalue) { + vvalue = vmalloc(size); + if (!vvalue) + return -ENOMEM; + kvalue = vvalue; + } } error = vfs_getxattr(d, kname, kvalue, size); @@ -452,7 +457,10 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, than XATTR_SIZE_MAX bytes. Not possible. */ error = -E2BIG; } - kfree(kvalue); + if (vvalue) + vfree(vvalue); + else + kfree(kvalue); return error; } -- cgit v0.10.2 From 9520628e8ceb69fa9a4aee6b57f22675d9e1b709 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 30 Jul 2012 14:39:15 -0700 Subject: fs: make dumpable=2 require fully qualified path When the suid_dumpable sysctl is set to "2", and there is no core dump pipe defined in the core_pattern sysctl, a local user can cause core files to be written to root-writable directories, potentially with user-controlled content. This means an admin can unknowningly reintroduce a variation of CVE-2006-2451, allowing local users to gain root privileges. $ cat /proc/sys/fs/suid_dumpable 2 $ cat /proc/sys/kernel/core_pattern core $ ulimit -c unlimited $ cd / $ ls -l core ls: cannot access core: No such file or directory $ touch core touch: cannot touch `core': Permission denied $ OHAI="evil-string-here" ping localhost >/dev/null 2>&1 & $ pid=$! $ sleep 1 $ kill -SEGV $pid $ ls -l core -rw------- 1 root kees 458752 Jun 21 11:35 core $ sudo strings core | grep evil OHAI=evil-string-here While cron has been fixed to abort reading a file when there is any parse error, there are still other sensitive directories that will read any file present and skip unparsable lines. Instead of introducing a suid_dumpable=3 mode and breaking all users of mode 2, this only disables the unsafe portion of mode 2 (writing to disk via relative path). Most users of mode 2 (e.g. Chrome OS) already use a core dump pipe handler, so this change will not break them. For the situations where a pipe handler is not defined but mode 2 is still active, crash dumps will only be written to fully qualified paths. If a relative path is defined (e.g. the default "core" pattern), dump attempts will trigger a printk yelling about the lack of a fully qualified path. Signed-off-by: Kees Cook Cc: Alexander Viro Cc: Alan Cox Cc: "Eric W. Biederman" Cc: Doug Ledford Cc: Serge Hallyn Cc: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/sysctl/fs.txt b/Documentation/sysctl/fs.txt index 13d6166..8c235b6 100644 --- a/Documentation/sysctl/fs.txt +++ b/Documentation/sysctl/fs.txt @@ -163,16 +163,22 @@ This value can be used to query and set the core dump mode for setuid or otherwise protected/tainted binaries. The modes are 0 - (default) - traditional behaviour. Any process which has changed - privilege levels or is execute only will not be dumped + privilege levels or is execute only will not be dumped. 1 - (debug) - all processes dump core when possible. The core dump is owned by the current user and no security is applied. This is intended for system debugging situations only. Ptrace is unchecked. + This is insecure as it allows regular users to examine the memory + contents of privileged processes. 2 - (suidsafe) - any binary which normally would not be dumped is dumped - readable by root only. This allows the end user to remove - such a dump but not access it directly. For security reasons - core dumps in this mode will not overwrite one another or - other files. This mode is appropriate when administrators are - attempting to debug problems in a normal environment. + anyway, but only if the "core_pattern" kernel sysctl is set to + either a pipe handler or a fully qualified path. (For more details + on this limitation, see CVE-2006-2451.) This mode is appropriate + when administrators are attempting to debug problems in a normal + environment, and either have a core dump pipe handler that knows + to treat privileged core dumps with care, or specific directory + defined for catching core dumps. If a core dump happens without + a pipe handler or fully qualifid path, a message will be emitted + to syslog warning about the lack of a correct setting. ============================================================== diff --git a/fs/exec.c b/fs/exec.c index e95aeed..95aae3f 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2111,6 +2111,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) int retval = 0; int flag = 0; int ispipe; + bool need_nonrelative = false; static atomic_t core_dump_count = ATOMIC_INIT(0); struct coredump_params cprm = { .signr = signr, @@ -2136,14 +2137,16 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) if (!cred) goto fail; /* - * We cannot trust fsuid as being the "true" uid of the - * process nor do we know its entire history. We only know it - * was tainted so we dump it as root in mode 2. + * We cannot trust fsuid as being the "true" uid of the process + * nor do we know its entire history. We only know it was tainted + * so we dump it as root in mode 2, and only into a controlled + * environment (pipe handler or fully qualified path). */ if (__get_dumpable(cprm.mm_flags) == 2) { /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ cred->fsuid = GLOBAL_ROOT_UID; /* Dump root private */ + need_nonrelative = true; } retval = coredump_wait(exit_code, &core_state); @@ -2223,6 +2226,14 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) if (cprm.limit < binfmt->min_coredump) goto fail_unlock; + if (need_nonrelative && cn.corename[0] != '/') { + printk(KERN_WARNING "Pid %d(%s) can only dump core "\ + "to fully qualified path!\n", + task_tgid_vnr(current), current->comm); + printk(KERN_WARNING "Skipping core dump\n"); + goto fail_unlock; + } + cprm.file = filp_open(cn.corename, O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, 0600); -- cgit v0.10.2 From 54b501992dd2a839e94e76aa392c392b55080ce8 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 30 Jul 2012 14:39:18 -0700 Subject: coredump: warn about unsafe suid_dumpable / core_pattern combo When suid_dumpable=2, detect unsafe core_pattern settings and warn when they are seen. Signed-off-by: Kees Cook Suggested-by: Andrew Morton Cc: Alexander Viro Cc: Alan Cox Cc: "Eric W. Biederman" Cc: Doug Ledford Cc: Serge Hallyn Cc: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/exec.c b/fs/exec.c index 95aae3f..5af8390 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2002,17 +2002,17 @@ static void coredump_finish(struct mm_struct *mm) void set_dumpable(struct mm_struct *mm, int value) { switch (value) { - case 0: + case SUID_DUMPABLE_DISABLED: clear_bit(MMF_DUMPABLE, &mm->flags); smp_wmb(); clear_bit(MMF_DUMP_SECURELY, &mm->flags); break; - case 1: + case SUID_DUMPABLE_ENABLED: set_bit(MMF_DUMPABLE, &mm->flags); smp_wmb(); clear_bit(MMF_DUMP_SECURELY, &mm->flags); break; - case 2: + case SUID_DUMPABLE_SAFE: set_bit(MMF_DUMP_SECURELY, &mm->flags); smp_wmb(); set_bit(MMF_DUMPABLE, &mm->flags); @@ -2025,7 +2025,7 @@ static int __get_dumpable(unsigned long mm_flags) int ret; ret = mm_flags & MMF_DUMPABLE_MASK; - return (ret >= 2) ? 2 : ret; + return (ret > SUID_DUMPABLE_ENABLED) ? SUID_DUMPABLE_SAFE : ret; } int get_dumpable(struct mm_struct *mm) @@ -2142,7 +2142,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) * so we dump it as root in mode 2, and only into a controlled * environment (pipe handler or fully qualified path). */ - if (__get_dumpable(cprm.mm_flags) == 2) { + if (__get_dumpable(cprm.mm_flags) == SUID_DUMPABLE_SAFE) { /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ cred->fsuid = GLOBAL_ROOT_UID; /* Dump root private */ diff --git a/include/linux/sched.h b/include/linux/sched.h index a721cef..1e26a5e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -406,6 +406,11 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm) {} extern void set_dumpable(struct mm_struct *mm, int value); extern int get_dumpable(struct mm_struct *mm); +/* get/set_dumpable() values */ +#define SUID_DUMPABLE_DISABLED 0 +#define SUID_DUMPABLE_ENABLED 1 +#define SUID_DUMPABLE_SAFE 2 + /* mm flags */ /* dumpable bits */ #define MMF_DUMPABLE 0 /* core dump is permitted */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4ab1187..b46f496 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -174,6 +174,11 @@ static int proc_dointvec_minmax_sysadmin(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); #endif +static int proc_dointvec_minmax_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); +static int proc_dostring_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); + #ifdef CONFIG_MAGIC_SYSRQ /* Note: sysrq code uses it's own private copy */ static int __sysrq_enabled = SYSRQ_DEFAULT_ENABLE; @@ -410,7 +415,7 @@ static struct ctl_table kern_table[] = { .data = core_pattern, .maxlen = CORENAME_MAX_SIZE, .mode = 0644, - .proc_handler = proc_dostring, + .proc_handler = proc_dostring_coredump, }, { .procname = "core_pipe_limit", @@ -1498,7 +1503,7 @@ static struct ctl_table fs_table[] = { .data = &suid_dumpable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dointvec_minmax_coredump, .extra1 = &zero, .extra2 = &two, }, @@ -2009,6 +2014,34 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, do_proc_dointvec_minmax_conv, ¶m); } +static void validate_coredump_safety(void) +{ + if (suid_dumpable == SUID_DUMPABLE_SAFE && + core_pattern[0] != '/' && core_pattern[0] != '|') { + printk(KERN_WARNING "Unsafe core_pattern used with "\ + "suid_dumpable=2. Pipe handler or fully qualified "\ + "core dump path required.\n"); + } +} + +static int proc_dointvec_minmax_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + if (!error) + validate_coredump_safety(); + return error; +} + +static int proc_dostring_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dostring(table, write, buffer, lenp, ppos); + if (!error) + validate_coredump_safety(); + return error; +} + static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos, -- cgit v0.10.2 From f107701f011e681a21b95259f950bf9c836884e0 Mon Sep 17 00:00:00 2001 From: Kautuk Consul Date: Mon, 30 Jul 2012 14:39:21 -0700 Subject: xtensa/mm/fault.c: port OOM changes to do_page_fault Commits d065bd810b6d ("mm: retry page fault when blocking on disk transfer") and 37b23e0525d3 ("x86,mm: make pagefault killable") introduced changes into the x86 pagefault handler for making the page fault handler retryable as well as killable. These changes reduce the mmap_sem hold time, which is crucial during OOM killer invocation. Port these changes to xtensa. Signed-off-by: Kautuk Consul Acked-by: Chris Zankel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c index b17885a..5a74c53 100644 --- a/arch/xtensa/mm/fault.c +++ b/arch/xtensa/mm/fault.c @@ -44,6 +44,7 @@ void do_page_fault(struct pt_regs *regs) int is_write, is_exec; int fault; + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; info.si_code = SEGV_MAPERR; @@ -71,6 +72,7 @@ void do_page_fault(struct pt_regs *regs) address, exccause, regs->pc, is_write? "w":"", is_exec? "x":""); #endif +retry: down_read(&mm->mmap_sem); vma = find_vma(mm, address); @@ -93,6 +95,7 @@ good_area: if (is_write) { if (!(vma->vm_flags & VM_WRITE)) goto bad_area; + flags |= FAULT_FLAG_WRITE; } else if (is_exec) { if (!(vma->vm_flags & VM_EXEC)) goto bad_area; @@ -104,7 +107,11 @@ good_area: * make sure we exit gracefully rather than endlessly redo * the fault. */ - fault = handle_mm_fault(mm, vma, address, is_write ? FAULT_FLAG_WRITE : 0); + fault = handle_mm_fault(mm, vma, address, flags); + + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) + return; + if (unlikely(fault & VM_FAULT_ERROR)) { if (fault & VM_FAULT_OOM) goto out_of_memory; @@ -112,10 +119,22 @@ good_area: goto do_sigbus; BUG(); } - if (fault & VM_FAULT_MAJOR) - current->maj_flt++; - else - current->min_flt++; + if (flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; + if (fault & VM_FAULT_RETRY) { + flags &= ~FAULT_FLAG_ALLOW_RETRY; + + /* No need to up_read(&mm->mmap_sem) as we would + * have already released it in __lock_page_or_retry + * in mm/filemap.c. + */ + + goto retry; + } + } up_read(&mm->mmap_sem); return; -- cgit v0.10.2 From 7eafef6d1d712aa9973c3cbd0013a5b66d5a6604 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Mon, 30 Jul 2012 14:39:24 -0700 Subject: alpha: remove mysterious if zero-ed out includes There's a small group of odd looking includes in smc37c669.c. These includes appear to be if zero-ed out ever since they were added to the tree (in v2.1.89). Their purpose is unclear to me. Perhaps they were used in someones build system. Whatever their purpose was, nothing else uses something comparable. This entire if zero-ed out block might as well be removed. Signed-off-by: Paul Bolle Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Matt Turner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/alpha/kernel/smc37c669.c b/arch/alpha/kernel/smc37c669.c index 0435921d..c803fc7 100644 --- a/arch/alpha/kernel/smc37c669.c +++ b/arch/alpha/kernel/smc37c669.c @@ -933,18 +933,6 @@ void SMC37c669_display_device_info( * *-- */ -#if 0 -/* $INCLUDE_OPTIONS$ */ -#include "cp$inc:platform_io.h" -/* $INCLUDE_OPTIONS_END$ */ -#include "cp$src:common.h" -#include "cp$inc:prototypes.h" -#include "cp$src:kernel_def.h" -#include "cp$src:msg_def.h" -#include "cp$src:smcc669_def.h" -/* Platform-specific includes */ -#include "cp$src:platform.h" -#endif #ifndef TRUE #define TRUE 1 -- cgit v0.10.2 From bf7c27e9887af48952743753916f9cfbe900d0e9 Mon Sep 17 00:00:00 2001 From: Kautuk Consul Date: Mon, 30 Jul 2012 14:39:25 -0700 Subject: avr32/mm/fault.c: port OOM changes to do_page_fault Commits d065bd810b6d ("mm: retry page fault when blocking on disk transfer") and 37b23e0525d3 ("x86,mm: make pagefault killable") introduced changes into the x86 pagefault handler for making the page fault handler retryable as well as killable. These changes reduce the mmap_sem hold time, which is crucial during OOM killer invocation. Port these changes to AVR32. [akpm@linux-foundation.org: fix comment layout] Signed-off-by: Mohd. Faris Signed-off-by: Kautuk Consul Acked-by: Havard Skinnemoen Cc: Hans-Christian Egtvedt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c index f7040a1..b92e609 100644 --- a/arch/avr32/mm/fault.c +++ b/arch/avr32/mm/fault.c @@ -61,10 +61,10 @@ asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs) const struct exception_table_entry *fixup; unsigned long address; unsigned long page; - int writeaccess; long signr; int code; int fault; + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; if (notify_page_fault(regs, ecr)) return; @@ -86,6 +86,7 @@ asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs) local_irq_enable(); +retry: down_read(&mm->mmap_sem); vma = find_vma(mm, address); @@ -104,7 +105,6 @@ asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs) */ good_area: code = SEGV_ACCERR; - writeaccess = 0; switch (ecr) { case ECR_PROTECTION_X: @@ -121,7 +121,7 @@ good_area: case ECR_TLB_MISS_W: if (!(vma->vm_flags & VM_WRITE)) goto bad_area; - writeaccess = 1; + flags |= FAULT_FLAG_WRITE; break; default: panic("Unhandled case %lu in do_page_fault!", ecr); @@ -132,7 +132,11 @@ good_area: * sure we exit gracefully rather than endlessly redo the * fault. */ - fault = handle_mm_fault(mm, vma, address, writeaccess ? FAULT_FLAG_WRITE : 0); + fault = handle_mm_fault(mm, vma, address, flags); + + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) + return; + if (unlikely(fault & VM_FAULT_ERROR)) { if (fault & VM_FAULT_OOM) goto out_of_memory; @@ -140,10 +144,23 @@ good_area: goto do_sigbus; BUG(); } - if (fault & VM_FAULT_MAJOR) - tsk->maj_flt++; - else - tsk->min_flt++; + + if (flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault & VM_FAULT_MAJOR) + tsk->maj_flt++; + else + tsk->min_flt++; + if (fault & VM_FAULT_RETRY) { + flags &= ~FAULT_FLAG_ALLOW_RETRY; + + /* + * No need to up_read(&mm->mmap_sem) as we would have + * already released it in __lock_page_or_retry() in + * mm/filemap.c. + */ + goto retry; + } + } up_read(&mm->mmap_sem); return; -- cgit v0.10.2 From 93abe8e4b13ae9a0428ce940a8a03ac72a7626f1 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:27 -0700 Subject: clk: add non CONFIG_HAVE_CLK routines Many drivers are shared between architectures that may or may not have HAVE_CLK selected for them. To remove compilation errors for them we enclose clk_*() calls in these drivers within #ifdef CONFIG_HAVE_CLK, #endif. This patch removes the need of these CONFIG_HAVE_CLK statements, by introducing dummy routines when HAVE_CLK is not selected by platforms. So, definition of these routines will always be available. These calls will return error for platforms that don't select HAVE_CLK. Signed-off-by: Viresh Kumar Cc: Wolfram Sang Cc: Greg Kroah-Hartman Cc: Jeff Garzik Cc: Andrew Lunn Cc: Bhupesh Sharma Cc: Giuseppe Cavallaro Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/clk.h b/include/linux/clk.h index 2fd6a42..b3ac22d 100644 --- a/include/linux/clk.h +++ b/include/linux/clk.h @@ -85,6 +85,43 @@ int clk_notifier_unregister(struct clk *clk, struct notifier_block *nb); #endif /** + * clk_prepare - prepare a clock source + * @clk: clock source + * + * This prepares the clock source for use. + * + * Must not be called from within atomic context. + */ +#ifdef CONFIG_HAVE_CLK_PREPARE +int clk_prepare(struct clk *clk); +#else +static inline int clk_prepare(struct clk *clk) +{ + might_sleep(); + return 0; +} +#endif + +/** + * clk_unprepare - undo preparation of a clock source + * @clk: clock source + * + * This undoes a previously prepared clock. The caller must balance + * the number of prepare and unprepare calls. + * + * Must not be called from within atomic context. + */ +#ifdef CONFIG_HAVE_CLK_PREPARE +void clk_unprepare(struct clk *clk); +#else +static inline void clk_unprepare(struct clk *clk) +{ + might_sleep(); +} +#endif + +#ifdef CONFIG_HAVE_CLK +/** * clk_get - lookup and obtain a reference to a clock producer. * @dev: device for clock "consumer" * @id: clock consumer ID @@ -122,24 +159,6 @@ struct clk *clk_get(struct device *dev, const char *id); struct clk *devm_clk_get(struct device *dev, const char *id); /** - * clk_prepare - prepare a clock source - * @clk: clock source - * - * This prepares the clock source for use. - * - * Must not be called from within atomic context. - */ -#ifdef CONFIG_HAVE_CLK_PREPARE -int clk_prepare(struct clk *clk); -#else -static inline int clk_prepare(struct clk *clk) -{ - might_sleep(); - return 0; -} -#endif - -/** * clk_enable - inform the system when the clock source should be running. * @clk: clock source * @@ -167,47 +186,6 @@ int clk_enable(struct clk *clk); */ void clk_disable(struct clk *clk); - -/** - * clk_unprepare - undo preparation of a clock source - * @clk: clock source - * - * This undoes a previously prepared clock. The caller must balance - * the number of prepare and unprepare calls. - * - * Must not be called from within atomic context. - */ -#ifdef CONFIG_HAVE_CLK_PREPARE -void clk_unprepare(struct clk *clk); -#else -static inline void clk_unprepare(struct clk *clk) -{ - might_sleep(); -} -#endif - -/* clk_prepare_enable helps cases using clk_enable in non-atomic context. */ -static inline int clk_prepare_enable(struct clk *clk) -{ - int ret; - - ret = clk_prepare(clk); - if (ret) - return ret; - ret = clk_enable(clk); - if (ret) - clk_unprepare(clk); - - return ret; -} - -/* clk_disable_unprepare helps cases using clk_disable in non-atomic context. */ -static inline void clk_disable_unprepare(struct clk *clk) -{ - clk_disable(clk); - clk_unprepare(clk); -} - /** * clk_get_rate - obtain the current clock rate (in Hz) for a clock source. * This is only valid once the clock source has been enabled. @@ -298,6 +276,78 @@ struct clk *clk_get_parent(struct clk *clk); */ struct clk *clk_get_sys(const char *dev_id, const char *con_id); +#else /* !CONFIG_HAVE_CLK */ + +static inline struct clk *clk_get(struct device *dev, const char *id) +{ + return NULL; +} + +static inline struct clk *devm_clk_get(struct device *dev, const char *id) +{ + return NULL; +} + +static inline void clk_put(struct clk *clk) {} + +static inline void devm_clk_put(struct device *dev, struct clk *clk) {} + +static inline int clk_enable(struct clk *clk) +{ + return 0; +} + +static inline void clk_disable(struct clk *clk) {} + +static inline unsigned long clk_get_rate(struct clk *clk) +{ + return 0; +} + +static inline int clk_set_rate(struct clk *clk, unsigned long rate) +{ + return 0; +} + +static inline long clk_round_rate(struct clk *clk, unsigned long rate) +{ + return 0; +} + +static inline int clk_set_parent(struct clk *clk, struct clk *parent) +{ + return 0; +} + +static inline struct clk *clk_get_parent(struct clk *clk) +{ + return NULL; +} + +#endif + +/* clk_prepare_enable helps cases using clk_enable in non-atomic context. */ +static inline int clk_prepare_enable(struct clk *clk) +{ + int ret; + + ret = clk_prepare(clk); + if (ret) + return ret; + ret = clk_enable(clk); + if (ret) + clk_unprepare(clk); + + return ret; +} + +/* clk_disable_unprepare helps cases using clk_disable in non-atomic context. */ +static inline void clk_disable_unprepare(struct clk *clk) +{ + clk_disable(clk); + clk_unprepare(clk); +} + /** * clk_add_alias - add a new clock alias * @alias: name for clock alias -- cgit v0.10.2 From 01da40bf8791434455d6e56c711eb224ddbf3161 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:29 -0700 Subject: clk: remove redundant depends on from drivers/Kconfig menu "Common Clock Framework" has "depends on COMMON_CLK" and so configs defined within menu don't require these "depends on COMMON_CLK again". Signed-off-by: Viresh Kumar Cc: Wolfram Sang Cc: Greg Kroah-Hartman Cc: Jeff Garzik Cc: Andrew Lunn Cc: Bhupesh Sharma Cc: Giuseppe Cavallaro Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/clk/Kconfig b/drivers/clk/Kconfig index 3f99b90..7f0b5ca 100644 --- a/drivers/clk/Kconfig +++ b/drivers/clk/Kconfig @@ -25,7 +25,6 @@ menu "Common Clock Framework" config COMMON_CLK_DEBUG bool "DebugFS representation of clock tree" - depends on COMMON_CLK select DEBUG_FS ---help--- Creates a directory hierchy in debugfs for visualizing the clk -- cgit v0.10.2 From dd995c9d82b6eedb2e704daf44e903da0b7f6ac1 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:30 -0700 Subject: i2c/i2c-pxa: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. pxa i2c also has these dummy macros defined locally. Remove them as they aren't required anymore. Signed-off-by: Viresh Kumar Acked-by: Wolfram Sang Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index a997c7d..1034d93 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -41,13 +41,6 @@ #include -#ifndef CONFIG_HAVE_CLK -#define clk_get(dev, id) NULL -#define clk_put(clk) do { } while (0) -#define clk_disable(clk) do { } while (0) -#define clk_enable(clk) do { } while (0) -#endif - struct pxa_reg_layout { u32 ibmr; u32 idbr; -- cgit v0.10.2 From 714904e1c3f48b17ceff06b827485f8b1b5d8a91 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:32 -0700 Subject: usb/marvell: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Marvell usb also has these dummy macros defined locally. Remove them as they aren't required anymore. Signed-off-by: Viresh Kumar Cc: Greg Kroah-Hartman Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/platform_data/mv_usb.h b/include/linux/platform_data/mv_usb.h index d94804a..944b01d 100644 --- a/include/linux/platform_data/mv_usb.h +++ b/include/linux/platform_data/mv_usb.h @@ -52,13 +52,4 @@ struct mv_usb_platform_data { int (*set_vbus)(unsigned int vbus); int (*private_init)(void __iomem *opregs, void __iomem *phyregs); }; - -#ifndef CONFIG_HAVE_CLK -/* Dummy stub for clk framework */ -#define clk_get(dev, id) NULL -#define clk_put(clock) do {} while (0) -#define clk_enable(clock) do {} while (0) -#define clk_disable(clock) do {} while (0) -#endif - #endif -- cgit v0.10.2 From 08aff5358207ee80e4fe162845c17ea125295eec Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:34 -0700 Subject: usb/musb: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. musb also has these dummy macros defined locally. Remove them as they aren't required anymore. Signed-off-by: Viresh Kumar Acked-by: Jeff Garzik Cc: Greg Kroah-Hartman Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/usb/musb/musb_core.h b/drivers/usb/musb/musb_core.h index dbcdeea..586105b 100644 --- a/drivers/usb/musb/musb_core.h +++ b/drivers/usb/musb/musb_core.h @@ -81,14 +81,6 @@ struct musb_ep; #define is_peripheral_active(m) (!(m)->is_host) #define is_host_active(m) ((m)->is_host) -#ifndef CONFIG_HAVE_CLK -/* Dummy stub for clk framework */ -#define clk_get(dev, id) NULL -#define clk_put(clock) do {} while (0) -#define clk_enable(clock) do {} while (0) -#define clk_disable(clock) do {} while (0) -#endif - #ifdef CONFIG_PROC_FS #include #define MUSB_CONFIG_PROC_FS -- cgit v0.10.2 From d14fb1e93086ec1acec34d5e17c989fa528c65e5 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:35 -0700 Subject: ata/pata_arasan: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Signed-off-by: Viresh Kumar Cc: Jeff Garzik Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/ata/pata_arasan_cf.c b/drivers/ata/pata_arasan_cf.c index ac6a5be..bfaa5cb 100644 --- a/drivers/ata/pata_arasan_cf.c +++ b/drivers/ata/pata_arasan_cf.c @@ -184,10 +184,8 @@ struct arasan_cf_dev { /* pointer to ata_host structure */ struct ata_host *host; - /* clk structure, only if HAVE_CLK is defined */ -#ifdef CONFIG_HAVE_CLK + /* clk structure */ struct clk *clk; -#endif /* physical base address of controller */ dma_addr_t pbase; @@ -312,13 +310,11 @@ static int cf_init(struct arasan_cf_dev *acdev) unsigned long flags; int ret = 0; -#ifdef CONFIG_HAVE_CLK ret = clk_enable(acdev->clk); if (ret) { dev_dbg(acdev->host->dev, "clock enable failed"); return ret; } -#endif spin_lock_irqsave(&acdev->host->lock, flags); /* configure CF interface clock */ @@ -344,9 +340,7 @@ static void cf_exit(struct arasan_cf_dev *acdev) writel(readl(acdev->vbase + OP_MODE) & ~CFHOST_ENB, acdev->vbase + OP_MODE); spin_unlock_irqrestore(&acdev->host->lock, flags); -#ifdef CONFIG_HAVE_CLK clk_disable(acdev->clk); -#endif } static void dma_callback(void *dev) @@ -828,13 +822,11 @@ static int __devinit arasan_cf_probe(struct platform_device *pdev) return -ENOMEM; } -#ifdef CONFIG_HAVE_CLK acdev->clk = clk_get(&pdev->dev, NULL); if (IS_ERR(acdev->clk)) { dev_warn(&pdev->dev, "Clock not found\n"); return PTR_ERR(acdev->clk); } -#endif /* allocate host */ host = ata_host_alloc(&pdev->dev, 1); @@ -899,9 +891,7 @@ static int __devinit arasan_cf_probe(struct platform_device *pdev) &arasan_cf_sht); free_clk: -#ifdef CONFIG_HAVE_CLK clk_put(acdev->clk); -#endif return ret; } @@ -912,9 +902,7 @@ static int __devexit arasan_cf_remove(struct platform_device *pdev) ata_host_detach(host); cf_exit(acdev); -#ifdef CONFIG_HAVE_CLK clk_put(acdev->clk); -#endif return 0; } -- cgit v0.10.2 From 2b1463ddde35a720de2fb4013057396e7f8011f0 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:38 -0700 Subject: net/c_can: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Signed-off-by: Viresh Kumar Acked-by: David S. Miller Cc: Bhupesh Sharma Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/can/c_can/c_can_platform.c b/drivers/net/can/c_can/c_can_platform.c index f0921d1..6ff7ad0 100644 --- a/drivers/net/can/c_can/c_can_platform.c +++ b/drivers/net/can/c_can/c_can_platform.c @@ -74,7 +74,6 @@ static int __devinit c_can_plat_probe(struct platform_device *pdev) const struct platform_device_id *id; struct resource *mem; int irq; -#ifdef CONFIG_HAVE_CLK struct clk *clk; /* get the appropriate clk */ @@ -84,7 +83,6 @@ static int __devinit c_can_plat_probe(struct platform_device *pdev) ret = -ENODEV; goto exit; } -#endif /* get the platform data */ mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); @@ -145,10 +143,8 @@ static int __devinit c_can_plat_probe(struct platform_device *pdev) dev->irq = irq; priv->base = addr; -#ifdef CONFIG_HAVE_CLK priv->can.clock.freq = clk_get_rate(clk); priv->priv = clk; -#endif platform_set_drvdata(pdev, dev); SET_NETDEV_DEV(dev, &pdev->dev); @@ -172,10 +168,8 @@ exit_iounmap: exit_release_mem: release_mem_region(mem->start, resource_size(mem)); exit_free_clk: -#ifdef CONFIG_HAVE_CLK clk_put(clk); exit: -#endif dev_err(&pdev->dev, "probe failed\n"); return ret; @@ -196,9 +190,7 @@ static int __devexit c_can_plat_remove(struct platform_device *pdev) mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); release_mem_region(mem->start, resource_size(mem)); -#ifdef CONFIG_HAVE_CLK clk_put(priv->priv); -#endif return 0; } -- cgit v0.10.2 From 6a81c26f868b2fc5f88b9f78a6f82aeab1191e61 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:41 -0700 Subject: net/stmmac: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. This also fixes error paths of probe(), as a goto is required in this patch. Signed-off-by: Viresh Kumar Cc: Giuseppe Cavallaro Acked-by: David S. Miller Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index ab4c376..f2d3665 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -82,9 +82,7 @@ struct stmmac_priv { struct stmmac_counters mmc; struct dma_features dma_cap; int hw_cap_support; -#ifdef CONFIG_HAVE_CLK struct clk *stmmac_clk; -#endif int clk_csr; int synopsys_id; struct timer_list eee_ctrl_timer; @@ -113,46 +111,6 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, void stmmac_disable_eee_mode(struct stmmac_priv *priv); bool stmmac_eee_init(struct stmmac_priv *priv); -#ifdef CONFIG_HAVE_CLK -static inline int stmmac_clk_enable(struct stmmac_priv *priv) -{ - if (!IS_ERR(priv->stmmac_clk)) - return clk_prepare_enable(priv->stmmac_clk); - - return 0; -} - -static inline void stmmac_clk_disable(struct stmmac_priv *priv) -{ - if (IS_ERR(priv->stmmac_clk)) - return; - - clk_disable_unprepare(priv->stmmac_clk); -} -static inline int stmmac_clk_get(struct stmmac_priv *priv) -{ - priv->stmmac_clk = clk_get(priv->device, NULL); - - if (IS_ERR(priv->stmmac_clk)) - return PTR_ERR(priv->stmmac_clk); - - return 0; -} -#else -static inline int stmmac_clk_enable(struct stmmac_priv *priv) -{ - return 0; -} -static inline void stmmac_clk_disable(struct stmmac_priv *priv) -{ -} -static inline int stmmac_clk_get(struct stmmac_priv *priv) -{ - return 0; -} -#endif /* CONFIG_HAVE_CLK */ - - #ifdef CONFIG_STMMAC_PLATFORM extern struct platform_driver stmmac_pltfr_driver; static inline int stmmac_register_platform(void) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index f6b04c1..fd8882f 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -28,6 +28,7 @@ https://bugzilla.stlinux.com/ *******************************************************************************/ +#include #include #include #include @@ -173,12 +174,8 @@ static void stmmac_verify_args(void) static void stmmac_clk_csr_set(struct stmmac_priv *priv) { -#ifdef CONFIG_HAVE_CLK u32 clk_rate; - if (IS_ERR(priv->stmmac_clk)) - return; - clk_rate = clk_get_rate(priv->stmmac_clk); /* Platform provided default clk_csr would be assumed valid @@ -200,7 +197,6 @@ static void stmmac_clk_csr_set(struct stmmac_priv *priv) * we can not estimate the proper divider as it is not known * the frequency of clk_csr_i. So we do not change the default * divider. */ -#endif } #if defined(STMMAC_XMIT_DEBUG) || defined(STMMAC_RX_DEBUG) @@ -1070,7 +1066,7 @@ static int stmmac_open(struct net_device *dev) } else priv->tm->enable = 1; #endif - stmmac_clk_enable(priv); + clk_enable(priv->stmmac_clk); stmmac_check_ether_addr(priv); @@ -1192,7 +1188,7 @@ open_error: if (priv->phydev) phy_disconnect(priv->phydev); - stmmac_clk_disable(priv); + clk_disable(priv->stmmac_clk); return ret; } @@ -1250,7 +1246,7 @@ static int stmmac_release(struct net_device *dev) #ifdef CONFIG_STMMAC_DEBUG_FS stmmac_exit_fs(); #endif - stmmac_clk_disable(priv); + clk_disable(priv->stmmac_clk); return 0; } @@ -2078,11 +2074,14 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, ret = register_netdev(ndev); if (ret) { pr_err("%s: ERROR %i registering the device\n", __func__, ret); - goto error; + goto error_netdev_register; } - if (stmmac_clk_get(priv)) + priv->stmmac_clk = clk_get(priv->device, NULL); + if (IS_ERR(priv->stmmac_clk)) { pr_warning("%s: warning: cannot get CSR clock\n", __func__); + goto error_clk_get; + } /* If a specific clk_csr value is passed from the platform * this means that the CSR Clock Range selection cannot be @@ -2100,15 +2099,17 @@ struct stmmac_priv *stmmac_dvr_probe(struct device *device, if (ret < 0) { pr_debug("%s: MDIO bus (id: %d) registration failed", __func__, priv->plat->bus_id); - goto error; + goto error_mdio_register; } return priv; -error: - netif_napi_del(&priv->napi); - +error_mdio_register: + clk_put(priv->stmmac_clk); +error_clk_get: unregister_netdev(ndev); +error_netdev_register: + netif_napi_del(&priv->napi); free_netdev(ndev); return NULL; @@ -2177,7 +2178,7 @@ int stmmac_suspend(struct net_device *ndev) else { stmmac_set_mac(priv->ioaddr, false); /* Disable clock in case of PWM is off */ - stmmac_clk_disable(priv); + clk_disable(priv->stmmac_clk); } spin_unlock_irqrestore(&priv->lock, flags); return 0; @@ -2202,7 +2203,7 @@ int stmmac_resume(struct net_device *ndev) priv->hw->mac->pmt(priv->ioaddr, 0); else /* enable the clk prevously disabled */ - stmmac_clk_enable(priv); + clk_enable(priv->stmmac_clk); netif_device_attach(ndev); -- cgit v0.10.2 From f12a86a0987a443fb1bc2d4298e5db9c17f3e7cd Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:43 -0700 Subject: gadget/m66592: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Signed-off-by: Viresh Kumar Cc: Greg Kroah-Hartman Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/usb/gadget/m66592-udc.c b/drivers/usb/gadget/m66592-udc.c index 8981fbb..cf6bd62 100644 --- a/drivers/usb/gadget/m66592-udc.c +++ b/drivers/usb/gadget/m66592-udc.c @@ -1583,12 +1583,10 @@ static int __exit m66592_remove(struct platform_device *pdev) iounmap(m66592->reg); free_irq(platform_get_irq(pdev, 0), m66592); m66592_free_request(&m66592->ep[0].ep, m66592->ep0_req); -#ifdef CONFIG_HAVE_CLK if (m66592->pdata->on_chip) { clk_disable(m66592->clk); clk_put(m66592->clk); } -#endif kfree(m66592); return 0; } @@ -1602,9 +1600,7 @@ static int __init m66592_probe(struct platform_device *pdev) struct resource *res, *ires; void __iomem *reg = NULL; struct m66592 *m66592 = NULL; -#ifdef CONFIG_HAVE_CLK char clk_name[8]; -#endif int ret = 0; int i; @@ -1671,7 +1667,6 @@ static int __init m66592_probe(struct platform_device *pdev) goto clean_up; } -#ifdef CONFIG_HAVE_CLK if (m66592->pdata->on_chip) { snprintf(clk_name, sizeof(clk_name), "usbf%d", pdev->id); m66592->clk = clk_get(&pdev->dev, clk_name); @@ -1683,7 +1678,7 @@ static int __init m66592_probe(struct platform_device *pdev) } clk_enable(m66592->clk); } -#endif + INIT_LIST_HEAD(&m66592->gadget.ep_list); m66592->gadget.ep0 = &m66592->ep[0].ep; INIT_LIST_HEAD(&m66592->gadget.ep0->ep_list); @@ -1731,13 +1726,11 @@ err_add_udc: m66592_free_request(&m66592->ep[0].ep, m66592->ep0_req); clean_up3: -#ifdef CONFIG_HAVE_CLK if (m66592->pdata->on_chip) { clk_disable(m66592->clk); clk_put(m66592->clk); } clean_up2: -#endif free_irq(ires->start, m66592); clean_up: if (m66592) { diff --git a/drivers/usb/gadget/m66592-udc.h b/drivers/usb/gadget/m66592-udc.h index 88c85b4..16c7e14 100644 --- a/drivers/usb/gadget/m66592-udc.h +++ b/drivers/usb/gadget/m66592-udc.h @@ -13,10 +13,7 @@ #ifndef __M66592_UDC_H__ #define __M66592_UDC_H__ -#ifdef CONFIG_HAVE_CLK #include -#endif - #include #define M66592_SYSCFG 0x00 @@ -468,9 +465,7 @@ struct m66592_ep { struct m66592 { spinlock_t lock; void __iomem *reg; -#ifdef CONFIG_HAVE_CLK struct clk *clk; -#endif struct m66592_platdata *pdata; unsigned long irq_trigger; -- cgit v0.10.2 From 569a50da7fe5144ac3831ffa6f400aa276b9781a Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:46 -0700 Subject: gadget/r8a66597: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Signed-off-by: Viresh Kumar Cc: Greg Kroah-Hartman Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/usb/gadget/r8a66597-udc.c b/drivers/usb/gadget/r8a66597-udc.c index f3ac2a2..5a80751 100644 --- a/drivers/usb/gadget/r8a66597-udc.c +++ b/drivers/usb/gadget/r8a66597-udc.c @@ -1831,12 +1831,12 @@ static int __exit r8a66597_remove(struct platform_device *pdev) iounmap(r8a66597->sudmac_reg); free_irq(platform_get_irq(pdev, 0), r8a66597); r8a66597_free_request(&r8a66597->ep[0].ep, r8a66597->ep0_req); -#ifdef CONFIG_HAVE_CLK + if (r8a66597->pdata->on_chip) { clk_disable(r8a66597->clk); clk_put(r8a66597->clk); } -#endif + device_unregister(&r8a66597->gadget.dev); kfree(r8a66597); return 0; @@ -1868,9 +1868,7 @@ static int __init r8a66597_sudmac_ioremap(struct r8a66597 *r8a66597, static int __init r8a66597_probe(struct platform_device *pdev) { -#ifdef CONFIG_HAVE_CLK char clk_name[8]; -#endif struct resource *res, *ires; int irq; void __iomem *reg = NULL; @@ -1934,7 +1932,6 @@ static int __init r8a66597_probe(struct platform_device *pdev) r8a66597->timer.data = (unsigned long)r8a66597; r8a66597->reg = reg; -#ifdef CONFIG_HAVE_CLK if (r8a66597->pdata->on_chip) { snprintf(clk_name, sizeof(clk_name), "usb%d", pdev->id); r8a66597->clk = clk_get(&pdev->dev, clk_name); @@ -1946,7 +1943,7 @@ static int __init r8a66597_probe(struct platform_device *pdev) } clk_enable(r8a66597->clk); } -#endif + if (r8a66597->pdata->sudmac) { ret = r8a66597_sudmac_ioremap(r8a66597, pdev); if (ret < 0) @@ -2006,13 +2003,11 @@ err_add_udc: clean_up3: free_irq(irq, r8a66597); clean_up2: -#ifdef CONFIG_HAVE_CLK if (r8a66597->pdata->on_chip) { clk_disable(r8a66597->clk); clk_put(r8a66597->clk); } clean_up_dev: -#endif device_unregister(&r8a66597->gadget.dev); clean_up: if (r8a66597) { diff --git a/drivers/usb/gadget/r8a66597-udc.h b/drivers/usb/gadget/r8a66597-udc.h index 99908c7..45c4b2d 100644 --- a/drivers/usb/gadget/r8a66597-udc.h +++ b/drivers/usb/gadget/r8a66597-udc.h @@ -13,10 +13,7 @@ #ifndef __R8A66597_H__ #define __R8A66597_H__ -#ifdef CONFIG_HAVE_CLK #include -#endif - #include #define R8A66597_MAX_SAMPLING 10 @@ -92,9 +89,7 @@ struct r8a66597 { void __iomem *reg; void __iomem *sudmac_reg; -#ifdef CONFIG_HAVE_CLK struct clk *clk; -#endif struct r8a66597_platdata *pdata; struct usb_gadget gadget; -- cgit v0.10.2 From c311b78b2a2fc644da2e748b7999b9ff081ec00e Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:49 -0700 Subject: usb/host/r8a66597: remove conditional compilation of clk code With addition of dummy clk_*() calls for non CONFIG_HAVE_CLK cases in clk.h, there is no need to have clk code enclosed in #ifdef CONFIG_HAVE_CLK, #endif macros. Signed-off-by: Viresh Kumar Cc: Greg Kroah-Hartman Cc: Russell King Cc: Mike Turquette Cc: Sergei Shtylyov Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/usb/host/r8a66597-hcd.c b/drivers/usb/host/r8a66597-hcd.c index c868be6..4c634eb 100644 --- a/drivers/usb/host/r8a66597-hcd.c +++ b/drivers/usb/host/r8a66597-hcd.c @@ -95,9 +95,7 @@ static int r8a66597_clock_enable(struct r8a66597 *r8a66597) int i = 0; if (r8a66597->pdata->on_chip) { -#ifdef CONFIG_HAVE_CLK clk_enable(r8a66597->clk); -#endif do { r8a66597_write(r8a66597, SCKE, SYSCFG0); tmp = r8a66597_read(r8a66597, SYSCFG0); @@ -141,9 +139,7 @@ static void r8a66597_clock_disable(struct r8a66597 *r8a66597) udelay(1); if (r8a66597->pdata->on_chip) { -#ifdef CONFIG_HAVE_CLK clk_disable(r8a66597->clk); -#endif } else { r8a66597_bclr(r8a66597, PLLC, SYSCFG0); r8a66597_bclr(r8a66597, XCKE, SYSCFG0); @@ -2406,19 +2402,15 @@ static int __devexit r8a66597_remove(struct platform_device *pdev) del_timer_sync(&r8a66597->rh_timer); usb_remove_hcd(hcd); iounmap(r8a66597->reg); -#ifdef CONFIG_HAVE_CLK if (r8a66597->pdata->on_chip) clk_put(r8a66597->clk); -#endif usb_put_hcd(hcd); return 0; } static int __devinit r8a66597_probe(struct platform_device *pdev) { -#ifdef CONFIG_HAVE_CLK char clk_name[8]; -#endif struct resource *res = NULL, *ires; int irq = -1; void __iomem *reg = NULL; @@ -2482,7 +2474,6 @@ static int __devinit r8a66597_probe(struct platform_device *pdev) r8a66597->irq_sense_low = irq_trigger == IRQF_TRIGGER_LOW; if (r8a66597->pdata->on_chip) { -#ifdef CONFIG_HAVE_CLK snprintf(clk_name, sizeof(clk_name), "usb%d", pdev->id); r8a66597->clk = clk_get(&pdev->dev, clk_name); if (IS_ERR(r8a66597->clk)) { @@ -2491,7 +2482,6 @@ static int __devinit r8a66597_probe(struct platform_device *pdev) ret = PTR_ERR(r8a66597->clk); goto clean_up2; } -#endif r8a66597->max_root_hub = 1; } else r8a66597->max_root_hub = 2; @@ -2531,11 +2521,9 @@ static int __devinit r8a66597_probe(struct platform_device *pdev) return 0; clean_up3: -#ifdef CONFIG_HAVE_CLK if (r8a66597->pdata->on_chip) clk_put(r8a66597->clk); clean_up2: -#endif usb_put_hcd(hcd); clean_up: diff --git a/drivers/usb/host/r8a66597.h b/drivers/usb/host/r8a66597.h index f28782d..672cea3 100644 --- a/drivers/usb/host/r8a66597.h +++ b/drivers/usb/host/r8a66597.h @@ -26,10 +26,7 @@ #ifndef __R8A66597_H__ #define __R8A66597_H__ -#ifdef CONFIG_HAVE_CLK #include -#endif - #include #define R8A66597_MAX_NUM_PIPE 10 @@ -113,9 +110,7 @@ struct r8a66597_root_hub { struct r8a66597 { spinlock_t lock; void __iomem *reg; -#ifdef CONFIG_HAVE_CLK struct clk *clk; -#endif struct r8a66597_platdata *pdata; struct r8a66597_device device0; struct r8a66597_root_hub root_hub[R8A66597_MAX_ROOT_HUB]; -- cgit v0.10.2 From 7041717e7c6ca33310d0c7333d9f7cebc65b2534 Mon Sep 17 00:00:00 2001 From: Viresh Kumar Date: Mon, 30 Jul 2012 14:39:52 -0700 Subject: arch/arm/mach-netx/fb.c: reuse dummy clk routines for CONFIG_HAVE_CLK=n mach-netx had its own implementation of clk routines like, clk_get{put}, clk_enable{disable}, etc. And with introduction of following patchset: https://lkml.org/lkml/2012/4/24/154 we get compilation error for multiple definition of these routines. Sascha had following suggestion to deal with it: http://www.spinics.net/lists/arm-kernel/msg179369.html So, remove this code completely. Signed-off-by: Viresh Kumar Reported-by: Paul Gortmaker Acked-by: Sascha Hauer Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/arm/mach-netx/fb.c b/arch/arm/mach-netx/fb.c index 2cdf6ef..d122ee6 100644 --- a/arch/arm/mach-netx/fb.c +++ b/arch/arm/mach-netx/fb.c @@ -69,29 +69,6 @@ void netx_clcd_remove(struct clcd_fb *fb) fb->fb.screen_base, fb->fb.fix.smem_start); } -void clk_disable(struct clk *clk) -{ -} - -int clk_set_rate(struct clk *clk, unsigned long rate) -{ - return 0; -} - -int clk_enable(struct clk *clk) -{ - return 0; -} - -struct clk *clk_get(struct device *dev, const char *id) -{ - return dev && strcmp(dev_name(dev), "fb") == 0 ? NULL : ERR_PTR(-ENOENT); -} - -void clk_put(struct clk *clk) -{ -} - static AMBA_AHB_DEVICE(fb, "fb", 0, 0x00104000, { NETX_IRQ_LCD }, NULL); int netx_fb_init(struct clcd_board *board, struct clcd_panel *panel) -- cgit v0.10.2 From e47c6a3408ea6bcfc5204f23d5a5b209de6e68b9 Mon Sep 17 00:00:00 2001 From: Fengguang Wu Date: Mon, 30 Jul 2012 14:39:54 -0700 Subject: clk: validate pointer in __clk_disable() clk_get() returns -ENOENT on error and some careless caller might dereference it without error checking: In mxc_rnga_remove(): struct clk *clk = clk_get(&pdev->dev, "rng"); // ... clk_disable(clk); Since it's insane to audit the lots of existing and future clk users, let's add a check in the callee to avoid kernel panic and warn about any buggy user. Cc: Russell King Cc: Paul Gortmaker Cc: Viresh Kumar Cc: viresh kumar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index c87fdd7..efdfd00 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -465,6 +465,9 @@ static void __clk_disable(struct clk *clk) if (!clk) return; + if (WARN_ON(IS_ERR(clk))) + return; + if (WARN_ON(clk->enable_count == 0)) return; -- cgit v0.10.2 From 6c55845e71ba7a862395f504a240a5b62632e0d6 Mon Sep 17 00:00:00 2001 From: Fengguang Wu Date: Mon, 30 Jul 2012 14:39:57 -0700 Subject: avr32: fix build error in ATSTK1002 code Fix the error arch/avr32/boards/atstk1000/atstk1002.c:100: error: 'num_partitions' undeclared here (not in a function) which was introduced by commit 1754aab9bb86 ("mtd: ATMEL, AVR32: inline nand partition table access "). Signed-off-by: Fengguang Wu Cc: Dmitry Eremin-Solenikov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c index dc52633..6c80aba 100644 --- a/arch/avr32/boards/atstk1000/atstk1002.c +++ b/arch/avr32/boards/atstk1000/atstk1002.c @@ -97,7 +97,7 @@ static struct atmel_nand_data atstk1006_nand_data __initdata = { .enable_pin = GPIO_PIN_PB(29), .ecc_mode = NAND_ECC_SOFT, .parts = nand_partitions, - .num_parts = ARRAY_SIZE(num_partitions), + .num_parts = ARRAY_SIZE(nand_partitions), }; #endif -- cgit v0.10.2 From 190320c3b6640d4104650f55ff69611e050ea06b Mon Sep 17 00:00:00 2001 From: Vikram Mulukutla Date: Mon, 30 Jul 2012 14:39:58 -0700 Subject: panic: fix a possible deadlock in panic() panic_lock is meant to ensure that panic processing takes place only on one cpu; if any of the other cpus encounter a panic, they will spin waiting to be shut down. However, this causes a regression in this scenario: 1. Cpu 0 encounters a panic and acquires the panic_lock and proceeds with the panic processing. 2. There is an interrupt on cpu 0 that also encounters an error condition and invokes panic. 3. This second invocation fails to acquire the panic_lock and enters the infinite while loop in panic_smp_self_stop. Thus all panic processing is stopped, and the cpu is stuck for eternity in the while(1) inside panic_smp_self_stop. To address this, disable local interrupts with local_irq_disable before acquiring the panic_lock. This will prevent interrupt handlers from executing during the panic processing, thus avoiding this particular problem. Signed-off-by: Vikram Mulukutla Reviewed-by: Stephen Boyd Cc: Michael Holzheu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/panic.c b/kernel/panic.c index d2a5f4e..e1b2822 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -75,6 +75,14 @@ void panic(const char *fmt, ...) int state = 0; /* + * Disable local interrupts. This will prevent panic_smp_self_stop + * from deadlocking the first cpu that invokes the panic, since + * there is nothing to prevent an interrupt handler (that runs + * after the panic_lock is acquired) from invoking panic again. + */ + local_irq_disable(); + + /* * It's possible to come here directly from a panic-assertion and * not have preempt disabled. Some functions called from here want * preempt to be disabled. No point enabling it later though... -- cgit v0.10.2 From 45226e944ce071d0231949f2fea90969437cd2dc Mon Sep 17 00:00:00 2001 From: Sameer Nanda Date: Mon, 30 Jul 2012 14:40:00 -0700 Subject: NMI watchdog: fix for lockup detector breakage on resume On the suspend/resume path the boot CPU does not go though an offline->online transition. This breaks the NMI detector post-resume since it depends on PMU state that is lost when the system gets suspended. Fix this by forcing a CPU offline->online transition for the lockup detector on the boot CPU during resume. To provide more context, we enable NMI watchdog on Chrome OS. We have seen several reports of systems freezing up completely which indicated that the NMI watchdog was not firing for some reason. Debugging further, we found a simple way of repro'ing system freezes -- issuing the command 'tasket 1 sh -c "echo nmilockup > /proc/breakme"' after the system has been suspended/resumed one or more times. With this patch in place, the system freeze result in panics, as expected. These panics provide a nice stack trace for us to debug the actual issue causing the freeze. [akpm@linux-foundation.org: fiddle with code comment] [akpm@linux-foundation.org: make lockup_detector_bootcpu_resume() conditional on CONFIG_SUSPEND] [akpm@linux-foundation.org: fix section errors] Signed-off-by: Sameer Nanda Cc: Ingo Molnar Cc: Peter Zijlstra Cc: "Rafael J. Wysocki" Cc: Don Zickus Cc: Mandeep Singh Baines Cc: Srivatsa S. Bhat Cc: Anshuman Khandual Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/sched.h b/include/linux/sched.h index 1e26a5e..68dcffa 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -334,6 +334,14 @@ static inline void lockup_detector_init(void) } #endif +#if defined(CONFIG_LOCKUP_DETECTOR) && defined(CONFIG_SUSPEND) +void lockup_detector_bootcpu_resume(void); +#else +static inline void lockup_detector_bootcpu_resume(void) +{ +} +#endif + #ifdef CONFIG_DETECT_HUNG_TASK extern unsigned int sysctl_hung_task_panic; extern unsigned long sysctl_hung_task_check_count; diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index c8b7446..1da39ea 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -178,6 +178,9 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) arch_suspend_enable_irqs(); BUG_ON(irqs_disabled()); + /* Kick the lockup detector */ + lockup_detector_bootcpu_resume(); + Enable_cpus: enable_nonboot_cpus(); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 4b1dfba..69add8a 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -575,7 +575,7 @@ out: /* * Create/destroy watchdog threads as CPUs come and go: */ -static int __cpuinit +static int cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; @@ -610,10 +610,27 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block __cpuinitdata cpu_nfb = { +static struct notifier_block cpu_nfb = { .notifier_call = cpu_callback }; +#ifdef CONFIG_SUSPEND +/* + * On exit from suspend we force an offline->online transition on the boot CPU + * so that the PMU state that was lost while in suspended state gets set up + * properly for the boot CPU. This information is required for restarting the + * NMI watchdog. + */ +void lockup_detector_bootcpu_resume(void) +{ + void *cpu = (void *)(long)smp_processor_id(); + + cpu_callback(&cpu_nfb, CPU_DEAD_FROZEN, cpu); + cpu_callback(&cpu_nfb, CPU_UP_PREPARE_FROZEN, cpu); + cpu_callback(&cpu_nfb, CPU_ONLINE_FROZEN, cpu); +} +#endif + void __init lockup_detector_init(void) { void *cpu = (void *)(long)smp_processor_id(); -- cgit v0.10.2 From b57b44ae698944ffc6161352b8ff5c9cf9c592e2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:40:03 -0700 Subject: kernel/sys.c: avoid argv_free(NULL) If argv_split() failed, the code will end up calling argv_free(NULL). Fix it up and clean things up a bit. Addresses Coverity report 703573. Cc: Cyrill Gorcunov Cc: Kees Cook Cc: Serge Hallyn Cc: "Eric W. Biederman" Cc: WANG Cong Cc: Alan Cox Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sys.c b/kernel/sys.c index b04ae03..241507f 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2186,46 +2186,52 @@ static void argv_cleanup(struct subprocess_info *info) argv_free(info->argv); } -/** - * orderly_poweroff - Trigger an orderly system poweroff - * @force: force poweroff if command execution fails - * - * This may be called from any context to trigger a system shutdown. - * If the orderly shutdown fails, it will force an immediate shutdown. - */ -int orderly_poweroff(bool force) +static int __orderly_poweroff(void) { int argc; - char **argv = argv_split(GFP_ATOMIC, poweroff_cmd, &argc); + char **argv; static char *envp[] = { "HOME=/", "PATH=/sbin:/bin:/usr/sbin:/usr/bin", NULL }; - int ret = -ENOMEM; + int ret; + argv = argv_split(GFP_ATOMIC, poweroff_cmd, &argc); if (argv == NULL) { printk(KERN_WARNING "%s failed to allocate memory for \"%s\"\n", __func__, poweroff_cmd); - goto out; + return -ENOMEM; } ret = call_usermodehelper_fns(argv[0], argv, envp, UMH_NO_WAIT, NULL, argv_cleanup, NULL); -out: - if (likely(!ret)) - return 0; - if (ret == -ENOMEM) argv_free(argv); - if (force) { + return ret; +} + +/** + * orderly_poweroff - Trigger an orderly system poweroff + * @force: force poweroff if command execution fails + * + * This may be called from any context to trigger a system shutdown. + * If the orderly shutdown fails, it will force an immediate shutdown. + */ +int orderly_poweroff(bool force) +{ + int ret = __orderly_poweroff(); + + if (ret && force) { printk(KERN_WARNING "Failed to start orderly shutdown: " "forcing the issue\n"); - /* I guess this should try to kick off some daemon to - sync and poweroff asap. Or not even bother syncing - if we're doing an emergency shutdown? */ + /* + * I guess this should try to kick off some daemon to sync and + * poweroff asap. Or not even bother syncing if we're doing an + * emergency shutdown? + */ emergency_sync(); kernel_power_off(); } -- cgit v0.10.2 From 5353cf089bb32b69a7515be909c14bf05fe2e81e Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Jul 2012 14:40:06 -0700 Subject: drivers/misc/ti-st/st_core.c: missing NULL check Addresses https://bugzilla.kernel.org/show_bug.cgi?44431 Reported-by: Signed-off-by: Alan Cox Cc: Pavan Savoy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index 2b62232..acfaeeb 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -349,6 +349,11 @@ void st_int_recv(void *disc_data, st_gdata->rx_skb = alloc_skb( st_gdata->list[type]->max_frame_size, GFP_ATOMIC); + if (st_gdata->rx_skb == NULL) { + pr_err("out of memory: dropping\n"); + goto done; + } + skb_reserve(st_gdata->rx_skb, st_gdata->list[type]->reserve); /* next 2 required for BT only */ -- cgit v0.10.2 From cdf53441368cc02ee4aa8a8343a5dc25132836f0 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 30 Jul 2012 14:40:08 -0700 Subject: kmsg: /dev/kmsg - properly return possible copy_from_user() failure Reported-by: Andrew Morton Signed-off-by: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/printk.c b/kernel/printk.c index 50c96b5..852269a 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -389,8 +389,10 @@ static ssize_t devkmsg_writev(struct kiocb *iocb, const struct iovec *iv, line = buf; for (i = 0; i < count; i++) { - if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len)) + if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len)) { + ret = -EFAULT; goto out; + } line += iv[i].iov_len; } -- cgit v0.10.2 From acc8fa41ad31c576cdbc569cc3e0e443b1b98b44 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:09 -0700 Subject: printk: add generic functions to find KERN_ headers The current form of a KERN_ is "<.>". Add printk_get_level and printk_skip_level functions to handle these formats. These functions centralize tests of KERN_ so a future modification can change the KERN_ style and shorten the number of bytes consumed by these headers. [akpm@linux-foundation.org: fix build error and warning] Signed-off-by: Joe Perches Cc: Kay Sievers Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/printk.h b/include/linux/printk.h index 1bec2f7..6e12e1f 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -24,6 +24,32 @@ extern const char linux_proc_banner[]; */ #define KERN_CONT "" +static inline int printk_get_level(const char *buffer) +{ + if (buffer[0] == '<' && buffer[1] && buffer[2] == '>') { + switch (buffer[1]) { + case '0' ... '7': + case 'd': /* KERN_DEFAULT */ + case 'c': /* KERN_CONT */ + return buffer[1]; + } + } + return 0; +} + +static inline const char *printk_skip_level(const char *buffer) +{ + if (printk_get_level(buffer)) { + switch (buffer[1]) { + case '0' ... '7': + case 'd': /* KERN_DEFAULT */ + case 'c': /* KERN_CONT */ + return buffer + 3; + } + } + return buffer; +} + extern int console_printk[]; #define console_loglevel (console_printk[0]) diff --git a/kernel/printk.c b/kernel/printk.c index 852269a..0d882a2 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1487,6 +1487,7 @@ asmlinkage int vprintk_emit(int facility, int level, size_t text_len; enum log_flags lflags = 0; unsigned long flags; + int kern_level; int this_cpu; int printed_len = 0; @@ -1543,17 +1544,20 @@ asmlinkage int vprintk_emit(int facility, int level, } /* strip syslog prefix and extract log level or control flags */ - if (text[0] == '<' && text[1] && text[2] == '>') { - switch (text[1]) { + kern_level = printk_get_level(text); + if (kern_level) { + const char *end_of_header = printk_skip_level(text); + switch (kern_level) { case '0' ... '7': if (level == -1) - level = text[1] - '0'; + level = kern_level - '0'; case 'd': /* KERN_DEFAULT */ lflags |= LOG_PREFIX; case 'c': /* KERN_CONT */ - text += 3; - text_len -= 3; + break; } + text_len -= end_of_header - text; + text = (char *)end_of_header; } if (level == -1) -- cgit v0.10.2 From 314ba3520e513a78be80e8c2ddbd65c91e78a114 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:11 -0700 Subject: printk: add kern_levels.h to make KERN_ available for asm use Separate the printk.h file into 2 pieces so the definitions can be used in asm files. Signed-off-by: Joe Perches Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/kern_levels.h b/include/linux/kern_levels.h new file mode 100644 index 0000000..337e56f --- /dev/null +++ b/include/linux/kern_levels.h @@ -0,0 +1,22 @@ +#ifndef __KERN_LEVELS_H__ +#define __KERN_LEVELS_H__ + +#define KERN_EMERG "<0>" /* system is unusable */ +#define KERN_ALERT "<1>" /* action must be taken immediately */ +#define KERN_CRIT "<2>" /* critical conditions */ +#define KERN_ERR "<3>" /* error conditions */ +#define KERN_WARNING "<4>" /* warning conditions */ +#define KERN_NOTICE "<5>" /* normal but significant condition */ +#define KERN_INFO "<6>" /* informational */ +#define KERN_DEBUG "<7>" /* debug-level messages */ + +/* Use the default kernel loglevel */ +#define KERN_DEFAULT "" +/* + * Annotation for a "continued" line of log printout (only done after a + * line that had no enclosing \n). Only to be used by core/arch code + * during early bootup (a continued line is not SMP-safe otherwise). + */ +#define KERN_CONT "" + +#endif diff --git a/include/linux/printk.h b/include/linux/printk.h index 6e12e1f..fea2de3 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -2,28 +2,11 @@ #define __KERNEL_PRINTK__ #include +#include extern const char linux_banner[]; extern const char linux_proc_banner[]; -#define KERN_EMERG "<0>" /* system is unusable */ -#define KERN_ALERT "<1>" /* action must be taken immediately */ -#define KERN_CRIT "<2>" /* critical conditions */ -#define KERN_ERR "<3>" /* error conditions */ -#define KERN_WARNING "<4>" /* warning conditions */ -#define KERN_NOTICE "<5>" /* normal but significant condition */ -#define KERN_INFO "<6>" /* informational */ -#define KERN_DEBUG "<7>" /* debug-level messages */ - -/* Use the default kernel loglevel */ -#define KERN_DEFAULT "" -/* - * Annotation for a "continued" line of log printout (only done after a - * line that had no enclosing \n). Only to be used by core/arch code - * during early bootup (a continued line is not SMP-safe otherwise). - */ -#define KERN_CONT "" - static inline int printk_get_level(const char *buffer) { if (buffer[0] == '<' && buffer[1] && buffer[2] == '>') { -- cgit v0.10.2 From 0cc41e4a21d43695154fe6a151abf3b6f27b0bb0 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:12 -0700 Subject: arch: remove direct definitions of KERN_ uses Add #include so that the #define KERN_ macros don't have to be duplicated. Signed-off-by: Joe Perches Cc: Kay Sievers Cc: Russell King Cc: Kay Sievers Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/arm/lib/io-acorn.S b/arch/arm/lib/io-acorn.S index 1b197ea..69719ba 100644 --- a/arch/arm/lib/io-acorn.S +++ b/arch/arm/lib/io-acorn.S @@ -11,13 +11,14 @@ * */ #include +#include #include .text .align .Liosl_warning: - .ascii "<4>insl/outsl not implemented, called from %08lX\0" + .ascii KERN_WARNING "insl/outsl not implemented, called from %08lX\0" .align /* diff --git a/arch/arm/vfp/vfphw.S b/arch/arm/vfp/vfphw.S index 2d30c7f..d50f0e4 100644 --- a/arch/arm/vfp/vfphw.S +++ b/arch/arm/vfp/vfphw.S @@ -16,6 +16,7 @@ */ #include #include +#include #include "../kernel/entry-header.S" .macro DBGSTR, str @@ -24,7 +25,7 @@ add r0, pc, #4 bl printk b 1f - .asciz "<7>VFP: \str\n" + .asciz KERN_DEBUG "VFP: \str\n" .balign 4 1: ldmfd sp!, {r0-r3, ip, lr} #endif @@ -37,7 +38,7 @@ add r0, pc, #4 bl printk b 1f - .asciz "<7>VFP: \str\n" + .asciz KERN_DEBUG "VFP: \str\n" .balign 4 1: ldmfd sp!, {r0-r3, ip, lr} #endif @@ -52,7 +53,7 @@ add r0, pc, #4 bl printk b 1f - .asciz "<7>VFP: \str\n" + .asciz KERN_DEBUG "VFP: \str\n" .balign 4 1: ldmfd sp!, {r0-r3, ip, lr} #endif diff --git a/arch/frv/kernel/kernel_thread.S b/arch/frv/kernel/kernel_thread.S index 4531c83..f0e5294 100644 --- a/arch/frv/kernel/kernel_thread.S +++ b/arch/frv/kernel/kernel_thread.S @@ -10,10 +10,10 @@ */ #include +#include #include #define CLONE_VM 0x00000100 /* set if VM shared between processes */ -#define KERN_ERR "<3>" .section .rodata kernel_thread_emsg: -- cgit v0.10.2 From 533574c6bc30cf526cc1c41bde050c854a945efb Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:13 -0700 Subject: btrfs: use printk_get_level and printk_skip_level, add __printf, fix fallout Use the generic printk_get_level() to search a message for a kern_level. Add __printf to verify format and arguments. Fix a few messages that had mismatches in format and arguments. Add #ifdef CONFIG_PRINTK blocks to shrink the object size a bit when not using printk. [akpm@linux-foundation.org: whitespace tweak] Signed-off-by: Joe Perches Cc: Kay Sievers Cc: Chris Mason Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index adb1cd7..4bab807 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3342,10 +3342,22 @@ ssize_t btrfs_listxattr(struct dentry *dentry, char *buffer, size_t size); /* super.c */ int btrfs_parse_options(struct btrfs_root *root, char *options); int btrfs_sync_fs(struct super_block *sb, int wait); + +#ifdef CONFIG_PRINTK +__printf(2, 3) void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...); +#else +static inline __printf(2, 3) +void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...) +{ +} +#endif + +__printf(5, 6) void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function, unsigned int line, int errno, const char *fmt, ...); + void __btrfs_abort_transaction(struct btrfs_trans_handle *trans, struct btrfs_root *root, const char *function, unsigned int line, int errno); @@ -3386,6 +3398,7 @@ do { \ (errno), fmt, ##args); \ } while (0) +__printf(5, 6) void __btrfs_panic(struct btrfs_fs_info *fs_info, const char *function, unsigned int line, int errno, const char *fmt, ...); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 502b20c..fadeba6 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1114,7 +1114,7 @@ void clean_tree_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, spin_unlock(&root->fs_info->delalloc_lock); btrfs_panic(root->fs_info, -EOVERFLOW, "Can't clear %lu bytes from " - " dirty_mdatadata_bytes (%lu)", + " dirty_mdatadata_bytes (%llu)", buf->len, root->fs_info->dirty_metadata_bytes); } diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index c5dbd91..4da0865 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -1241,7 +1241,7 @@ static int __must_check __add_reloc_root(struct btrfs_root *root) if (rb_node) { btrfs_panic(root->fs_info, -EEXIST, "Duplicate root found " "for start=%llu while inserting into relocation " - "tree\n"); + "tree\n", node->bytenr); kfree(node); return -EEXIST; } diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index fa61ef5..8c6e61d 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -125,6 +125,7 @@ static void btrfs_handle_error(struct btrfs_fs_info *fs_info) } } +#ifdef CONFIG_PRINTK /* * __btrfs_std_error decodes expected errors from the caller and * invokes the approciate error response. @@ -167,7 +168,7 @@ void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function, va_end(args); } -const char *logtypes[] = { +static const char * const logtypes[] = { "emergency", "alert", "critical", @@ -185,22 +186,50 @@ void btrfs_printk(struct btrfs_fs_info *fs_info, const char *fmt, ...) struct va_format vaf; va_list args; const char *type = logtypes[4]; + int kern_level; va_start(args, fmt); - if (fmt[0] == '<' && isdigit(fmt[1]) && fmt[2] == '>') { - memcpy(lvl, fmt, 3); - lvl[3] = '\0'; - fmt += 3; - type = logtypes[fmt[1] - '0']; + kern_level = printk_get_level(fmt); + if (kern_level) { + size_t size = printk_skip_level(fmt) - fmt; + memcpy(lvl, fmt, size); + lvl[size] = '\0'; + fmt += size; + type = logtypes[kern_level - '0']; } else *lvl = '\0'; vaf.fmt = fmt; vaf.va = &args; + printk("%sBTRFS %s (device %s): %pV", lvl, type, sb->s_id, &vaf); + + va_end(args); } +#else + +void __btrfs_std_error(struct btrfs_fs_info *fs_info, const char *function, + unsigned int line, int errno, const char *fmt, ...) +{ + struct super_block *sb = fs_info->sb; + + /* + * Special case: if the error is EROFS, and we're already + * under MS_RDONLY, then it is safe here. + */ + if (errno == -EROFS && (sb->s_flags & MS_RDONLY)) + return; + + /* Don't go through full error handling during mount */ + if (sb->s_flags & MS_BORN) { + save_error_info(fs_info); + btrfs_handle_error(fs_info); + } +} +#endif + /* * We only mark the transaction aborted and then set the file system read-only. * This will prevent new transactions from starting or trying to join this -- cgit v0.10.2 From b778b3f25baf46dd88f16d51f1682828be0aa2ad Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:15 -0700 Subject: sound: use printk_get_level and printk_skip_level Make the output logging routine independent of the KERN_ style. Signed-off-by: Joe Perches Cc: Kay Sievers Cc: Takashi Iwai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/sound/core/misc.c b/sound/core/misc.c index 7681679..30e027e 100644 --- a/sound/core/misc.c +++ b/sound/core/misc.c @@ -68,6 +68,7 @@ void __snd_printk(unsigned int level, const char *path, int line, { va_list args; #ifdef CONFIG_SND_VERBOSE_PRINTK + int kern_level; struct va_format vaf; char verbose_fmt[] = KERN_DEFAULT "ALSA %s:%d %pV"; #endif @@ -81,12 +82,16 @@ void __snd_printk(unsigned int level, const char *path, int line, #ifdef CONFIG_SND_VERBOSE_PRINTK vaf.fmt = format; vaf.va = &args; - if (format[0] == '<' && format[2] == '>') { - memcpy(verbose_fmt, format, 3); - vaf.fmt = format + 3; + + kern_level = printk_get_level(format); + if (kern_level) { + const char *end_of_header = printk_skip_level(format); + memcpy(verbose_fmt, format, end_of_header - format); + vaf.fmt = end_of_header; } else if (level) - memcpy(verbose_fmt, KERN_DEBUG, 3); + memcpy(verbose_fmt, KERN_DEBUG, sizeof(KERN_DEBUG) - 1); printk(verbose_fmt, sanity_file_name(path), line, &vaf); + #else vprintk(format, args); #endif -- cgit v0.10.2 From 04d2c8c83d0e3ac5f78aeede51babb3236200112 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:17 -0700 Subject: printk: convert the format for KERN_ to a 2 byte pattern Instead of "<.>", use an ASCII SOH for the KERN_ prefix initiator. This saves 1 byte per printk, thousands of bytes in a normal kernel. No output changes are produced as vprintk_emit converts these uses to "<.>". Signed-off-by: Joe Perches Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/kern_levels.h b/include/linux/kern_levels.h index 337e56f..8c719a9 100644 --- a/include/linux/kern_levels.h +++ b/include/linux/kern_levels.h @@ -1,22 +1,25 @@ #ifndef __KERN_LEVELS_H__ #define __KERN_LEVELS_H__ -#define KERN_EMERG "<0>" /* system is unusable */ -#define KERN_ALERT "<1>" /* action must be taken immediately */ -#define KERN_CRIT "<2>" /* critical conditions */ -#define KERN_ERR "<3>" /* error conditions */ -#define KERN_WARNING "<4>" /* warning conditions */ -#define KERN_NOTICE "<5>" /* normal but significant condition */ -#define KERN_INFO "<6>" /* informational */ -#define KERN_DEBUG "<7>" /* debug-level messages */ +#define KERN_SOH "\001" /* ASCII Start Of Header */ +#define KERN_SOH_ASCII '\001' + +#define KERN_EMERG KERN_SOH "0" /* system is unusable */ +#define KERN_ALERT KERN_SOH "1" /* action must be taken immediately */ +#define KERN_CRIT KERN_SOH "2" /* critical conditions */ +#define KERN_ERR KERN_SOH "3" /* error conditions */ +#define KERN_WARNING KERN_SOH "4" /* warning conditions */ +#define KERN_NOTICE KERN_SOH "5" /* normal but significant condition */ +#define KERN_INFO KERN_SOH "6" /* informational */ +#define KERN_DEBUG KERN_SOH "7" /* debug-level messages */ + +#define KERN_DEFAULT KERN_SOH "d" /* the default kernel loglevel */ -/* Use the default kernel loglevel */ -#define KERN_DEFAULT "" /* * Annotation for a "continued" line of log printout (only done after a * line that had no enclosing \n). Only to be used by core/arch code * during early bootup (a continued line is not SMP-safe otherwise). */ -#define KERN_CONT "" +#define KERN_CONT KERN_SOH "c" #endif diff --git a/include/linux/printk.h b/include/linux/printk.h index fea2de3..93a231f 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -9,7 +9,7 @@ extern const char linux_proc_banner[]; static inline int printk_get_level(const char *buffer) { - if (buffer[0] == '<' && buffer[1] && buffer[2] == '>') { + if (buffer[0] == KERN_SOH_ASCII && buffer[1]) { switch (buffer[1]) { case '0' ... '7': case 'd': /* KERN_DEFAULT */ @@ -27,7 +27,7 @@ static inline const char *printk_skip_level(const char *buffer) case '0' ... '7': case 'd': /* KERN_DEFAULT */ case 'c': /* KERN_CONT */ - return buffer + 3; + return buffer + 2; } } return buffer; -- cgit v0.10.2 From 088a52aac810655c1db1e40331e4936946701e9c Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:19 -0700 Subject: printk: only look for prefix levels in kernel messages vprintk_emit() prefix parsing should only be done for internal kernel messages. This allows existing behavior to be kept in all cases. Signed-off-by: Joe Perches Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/printk.c b/kernel/printk.c index 0d882a2..6a76ab9 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1487,7 +1487,6 @@ asmlinkage int vprintk_emit(int facility, int level, size_t text_len; enum log_flags lflags = 0; unsigned long flags; - int kern_level; int this_cpu; int printed_len = 0; @@ -1543,21 +1542,24 @@ asmlinkage int vprintk_emit(int facility, int level, lflags |= LOG_NEWLINE; } - /* strip syslog prefix and extract log level or control flags */ - kern_level = printk_get_level(text); - if (kern_level) { - const char *end_of_header = printk_skip_level(text); - switch (kern_level) { - case '0' ... '7': - if (level == -1) - level = kern_level - '0'; - case 'd': /* KERN_DEFAULT */ - lflags |= LOG_PREFIX; - case 'c': /* KERN_CONT */ - break; + /* strip kernel syslog prefix and extract log level or control flags */ + if (facility == 0) { + int kern_level = printk_get_level(text); + + if (kern_level) { + const char *end_of_header = printk_skip_level(text); + switch (kern_level) { + case '0' ... '7': + if (level == -1) + level = kern_level - '0'; + case 'd': /* KERN_DEFAULT */ + lflags |= LOG_PREFIX; + case 'c': /* KERN_CONT */ + break; + } + text_len -= end_of_header - text; + text = (char *)end_of_header; } - text_len -= end_of_header - text; - text = (char *)end_of_header; } if (level == -1) -- cgit v0.10.2 From 61e99ab8e35a88b8c4d0f80d3df9ee16df471be5 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:21 -0700 Subject: printk: remove the now unnecessary "C" annotation for KERN_CONT Now that all KERN_ uses are prefixed with ASCII SOH, there is no need for a KERN_CONT. Keep it backward compatible by adding #define KERN_CONT "" Reduces kernel image size a thousand bytes. Signed-off-by: Joe Perches Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/kern_levels.h b/include/linux/kern_levels.h index 8c719a9..866caaa 100644 --- a/include/linux/kern_levels.h +++ b/include/linux/kern_levels.h @@ -20,6 +20,6 @@ * line that had no enclosing \n). Only to be used by core/arch code * during early bootup (a continued line is not SMP-safe otherwise). */ -#define KERN_CONT KERN_SOH "c" +#define KERN_CONT "" #endif diff --git a/include/linux/printk.h b/include/linux/printk.h index 93a231f..9afc01e 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -13,7 +13,6 @@ static inline int printk_get_level(const char *buffer) switch (buffer[1]) { case '0' ... '7': case 'd': /* KERN_DEFAULT */ - case 'c': /* KERN_CONT */ return buffer[1]; } } @@ -26,7 +25,6 @@ static inline const char *printk_skip_level(const char *buffer) switch (buffer[1]) { case '0' ... '7': case 'd': /* KERN_DEFAULT */ - case 'c': /* KERN_CONT */ return buffer + 2; } } -- cgit v0.10.2 From 76597ff989a1fbaa9b9a1e54007cd759bf257ab7 Mon Sep 17 00:00:00 2001 From: Andrei Emeltchenko Date: Mon, 30 Jul 2012 14:40:23 -0700 Subject: vsprintf: add %pMR for Bluetooth MAC address Bluetooth uses mostly LE byte order which is reversed for visual interpretation. Currently in Bluetooth in use unsafe batostr function. This is a slightly modified version of Joe's patch (sent Sat, Dec 4, 2010). Signed-off-by: Andrei Emeltchenko Cc: Joe Perches Cc: Marcel Holtmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/printk-formats.txt b/Documentation/printk-formats.txt index 5df176e..d8d168f 100644 --- a/Documentation/printk-formats.txt +++ b/Documentation/printk-formats.txt @@ -56,6 +56,7 @@ Struct Resources: MAC/FDDI addresses: %pM 00:01:02:03:04:05 + %pMR 05:04:03:02:01:00 %pMF 00-01-02-03-04-05 %pm 000102030405 @@ -67,6 +68,10 @@ MAC/FDDI addresses: the 'M' specifier to use dash ('-') separators instead of the default separator. + For Bluetooth addresses the 'R' specifier shall be used after the 'M' + specifier to use reversed byte order suitable for visual interpretation + of Bluetooth addresses which are in the little endian order. + IPv4 addresses: %pI4 1.2.3.4 diff --git a/lib/vsprintf.c b/lib/vsprintf.c index c3f36d41..7369745 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -662,15 +662,28 @@ char *mac_address_string(char *buf, char *end, u8 *addr, char *p = mac_addr; int i; char separator; + bool reversed = false; - if (fmt[1] == 'F') { /* FDDI canonical format */ + switch (fmt[1]) { + case 'F': separator = '-'; - } else { + break; + + case 'R': + reversed = true; + /* fall through */ + + default: separator = ':'; + break; } for (i = 0; i < 6; i++) { - p = hex_byte_pack(p, addr[i]); + if (reversed) + p = hex_byte_pack(p, addr[5 - i]); + else + p = hex_byte_pack(p, addr[i]); + if (fmt[0] == 'M' && i != 5) *p++ = separator; } @@ -933,6 +946,7 @@ int kptr_restrict __read_mostly; * - 'm' For a 6-byte MAC address, it prints the hex address without colons * - 'MF' For a 6-byte MAC FDDI address, it prints the address * with a dash-separated hex notation + * - '[mM]R For a 6-byte MAC address, Reverse order (Bluetooth) * - 'I' [46] for IPv4/IPv6 addresses printed in the usual way * IPv4 uses dot-separated decimal without leading 0's (1.2.3.4) * IPv6 uses colon separated network-order 16 bit hex with leading 0's @@ -995,7 +1009,8 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, return resource_string(buf, end, ptr, spec, fmt); case 'M': /* Colon separated: 00:01:02:03:04:05 */ case 'm': /* Contiguous: 000102030405 */ - /* [mM]F (FDDI, bit reversed) */ + /* [mM]F (FDDI) */ + /* [mM]R (Reverse order; Bluetooth) */ return mac_address_string(buf, end, ptr, spec, fmt); case 'I': /* Formatted IP supported * 4: 1.2.3.4 -- cgit v0.10.2 From 80f548e04d0b1d67d4fa8f59dbecc247f7b71c92 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:40:25 -0700 Subject: lib/vsprintf.c: remind people to update Documentation/printk-formats.txt when adding printk formats Cc: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 7369745..a0b5f15 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -1297,6 +1297,8 @@ qualifier: * case. * %n is ignored * + * ** Please update Documentation/printk-formats.txt when making changes ** + * * The return value is the number of characters which would * be generated for the given input, excluding the trailing * '\0', as per ISO C99. If you want to have the exact -- cgit v0.10.2 From 3715c5309f6d175c3053672b73fd4f73be16fd07 Mon Sep 17 00:00:00 2001 From: Dan Rosenberg Date: Mon, 30 Jul 2012 14:40:26 -0700 Subject: lib/vsprintf.c: kptr_restrict: fix pK-error in SysRq show-all-timers(Q) When using ALT+SysRq+Q all the pointers are replaced with "pK-error" like this: [23153.208033] .base: pK-error with echo h > /proc/sysrq-trigger it works: [23107.776363] .base: ffff88023e60d540 The intent behind this behavior was to return "pK-error" in cases where the %pK format specifier was used in interrupt context, because the CAP_SYSLOG check wouldn't be meaningful. Clearly this should only apply when kptr_restrict is actually enabled though. Reported-by: Stevie Trujillo Signed-off-by: Dan Rosenberg Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/vsprintf.c b/lib/vsprintf.c index a0b5f15..225aa68 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -1045,7 +1045,8 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, * %pK cannot be used in IRQ context because its test * for CAP_SYSLOG would be meaningless. */ - if (in_irq() || in_serving_softirq() || in_nmi()) { + if (kptr_restrict && (in_irq() || in_serving_softirq() || + in_nmi())) { if (spec.field_width == -1) spec.field_width = default_width; return string(buf, end, "pK-error", spec); -- cgit v0.10.2 From 31550a16a5d2af859e8a11839e8c6c6c9c92dfa7 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 30 Jul 2012 14:40:27 -0700 Subject: vsprintf: add support of '%*ph[CDN]' There are many places in the kernel where the drivers print small buffers as a hex string. This patch adds a support of the variable width buffer to print it as a hex string with a delimiter. The idea came from Pavel Roskin here: http://www.digipedia.pl/usenet/thread/18835/17449/ Sample output of pr_info("buf[%d:%d] %*phC\n", from, len, len, &buf[from]); could be look like this: [ 0.726130] buf[51:8] e8:16:b6:ef:e3:74:45:6e [ 0.750736] buf[59:15] 31:81:b8:3f:35:49:06:ae:df:32:06:05:4a:af:55 [ 0.757602] buf[17:5] ac:16:d5:2c:ef Signed-off-by: Andy Shevchenko Cc: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/printk-formats.txt b/Documentation/printk-formats.txt index d8d168f..7561d7e 100644 --- a/Documentation/printk-formats.txt +++ b/Documentation/printk-formats.txt @@ -53,6 +53,16 @@ Struct Resources: For printing struct resources. The 'R' and 'r' specifiers result in a printed resource with ('R') or without ('r') a decoded flags member. +Raw buffer as a hex string: + %*ph 00 01 02 ... 3f + %*phC 00:01:02: ... :3f + %*phD 00-01-02- ... -3f + %*phN 000102 ... 3f + + For printing a small buffers (up to 64 bytes long) as a hex string with + certain separator. For the larger buffers consider to use + print_hex_dump(). + MAC/FDDI addresses: %pM 00:01:02:03:04:05 diff --git a/lib/vsprintf.c b/lib/vsprintf.c index 225aa68..0e33754 100644 --- a/lib/vsprintf.c +++ b/lib/vsprintf.c @@ -655,6 +655,50 @@ char *resource_string(char *buf, char *end, struct resource *res, } static noinline_for_stack +char *hex_string(char *buf, char *end, u8 *addr, struct printf_spec spec, + const char *fmt) +{ + int i, len = 1; /* if we pass '%ph[CDN]', field witdh remains + negative value, fallback to the default */ + char separator; + + if (spec.field_width == 0) + /* nothing to print */ + return buf; + + if (ZERO_OR_NULL_PTR(addr)) + /* NULL pointer */ + return string(buf, end, NULL, spec); + + switch (fmt[1]) { + case 'C': + separator = ':'; + break; + case 'D': + separator = '-'; + break; + case 'N': + separator = 0; + break; + default: + separator = ' '; + break; + } + + if (spec.field_width > 0) + len = min_t(int, spec.field_width, 64); + + for (i = 0; i < len && buf < end - 1; i++) { + buf = hex_byte_pack(buf, addr[i]); + + if (buf < end && separator && i != len - 1) + *buf++ = separator; + } + + return buf; +} + +static noinline_for_stack char *mac_address_string(char *buf, char *end, u8 *addr, struct printf_spec spec, const char *fmt) { @@ -974,6 +1018,13 @@ int kptr_restrict __read_mostly; * correctness of the format string and va_list arguments. * - 'K' For a kernel pointer that should be hidden from unprivileged users * - 'NF' For a netdev_features_t + * - 'h[CDN]' For a variable-length buffer, it prints it as a hex string with + * a certain separator (' ' by default): + * C colon + * D dash + * N no separator + * The maximum supported length is 64 bytes of the input. Consider + * to use print_hex_dump() for the larger input. * * Note: The difference between 'S' and 'F' is that on ia64 and ppc64 * function pointers are really function descriptors, which contain a @@ -1007,6 +1058,8 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr, case 'R': case 'r': return resource_string(buf, end, ptr, spec, fmt); + case 'h': + return hex_string(buf, end, ptr, spec, fmt); case 'M': /* Colon separated: 00:01:02:03:04:05 */ case 'm': /* Contiguous: 000102030405 */ /* [mM]F (FDDI) */ @@ -1296,6 +1349,8 @@ qualifier: * %pI6c print an IPv6 address as specified by RFC 5952 * %pU[bBlL] print a UUID/GUID in big or little endian using lower or upper * case. + * %*ph[CDN] a variable-length hex string with a separator (supports up to 64 + * bytes of the input) * %n is ignored * * ** Please update Documentation/printk-formats.txt when making changes ** -- cgit v0.10.2 From a824c73cc33e149f79914bcebd90ace1433c0721 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:29 -0700 Subject: MAINTAINERS: update EXYNOS DP DRIVER F: patterns Add patterns for Exynos DP header to MAINTAINERS file. Signed-off-by: Jingoo Han Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index bd45164..c8804c0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2750,6 +2750,7 @@ M: Jingoo Han L: linux-fbdev@vger.kernel.org S: Maintained F: drivers/video/exynos/exynos_dp* +F: include/video/exynos_dp* EXYNOS MIPI DISPLAY DRIVERS M: Inki Dae -- cgit v0.10.2 From a4be29ac6df4f7c7a0225f7f72c78f4fb10114c0 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:31 -0700 Subject: drivers/video/backlight/atmel-pwm-bl.c: use devm_ functions The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_kzalloc of these functions. Signed-off-by: Jingoo Han Acked-by: Hans-Christian Egtvedt Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/atmel-pwm-bl.c b/drivers/video/backlight/atmel-pwm-bl.c index 0443a4f..cd64073 100644 --- a/drivers/video/backlight/atmel-pwm-bl.c +++ b/drivers/video/backlight/atmel-pwm-bl.c @@ -127,7 +127,8 @@ static int atmel_pwm_bl_probe(struct platform_device *pdev) struct atmel_pwm_bl *pwmbl; int retval; - pwmbl = kzalloc(sizeof(struct atmel_pwm_bl), GFP_KERNEL); + pwmbl = devm_kzalloc(&pdev->dev, sizeof(struct atmel_pwm_bl), + GFP_KERNEL); if (!pwmbl) return -ENOMEM; @@ -202,7 +203,6 @@ err_free_gpio: err_free_pwm: pwm_channel_free(&pwmbl->pwmc); err_free_mem: - kfree(pwmbl); return retval; } @@ -218,7 +218,6 @@ static int __exit atmel_pwm_bl_remove(struct platform_device *pdev) pwm_channel_free(&pwmbl->pwmc); backlight_device_unregister(pwmbl->bldev); platform_set_drvdata(pdev, NULL); - kfree(pwmbl); return 0; } -- cgit v0.10.2 From f1b60d46b6f1274f1eacfc89ddbf7f6b63348196 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:33 -0700 Subject: drivers/video/backlight/ot200_bl.c: use devm_ functions The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_kzalloc of these functions Signed-off-by: Jingoo Han Cc: Christian Gmeiner Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/ot200_bl.c b/drivers/video/backlight/ot200_bl.c index f519d55..ef637ff 100644 --- a/drivers/video/backlight/ot200_bl.c +++ b/drivers/video/backlight/ot200_bl.c @@ -97,10 +97,10 @@ static int ot200_backlight_probe(struct platform_device *pdev) goto error_mfgpt_alloc; } - data = kzalloc(sizeof(*data), GFP_KERNEL); + data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL); if (!data) { retval = -ENOMEM; - goto error_kzalloc; + goto error_devm_kzalloc; } /* setup gpio */ @@ -122,16 +122,14 @@ static int ot200_backlight_probe(struct platform_device *pdev) if (IS_ERR(bl)) { dev_err(&pdev->dev, "failed to register backlight\n"); retval = PTR_ERR(bl); - goto error_backlight_device_register; + goto error_devm_kzalloc; } platform_set_drvdata(pdev, bl); return 0; -error_backlight_device_register: - kfree(data); -error_kzalloc: +error_devm_kzalloc: cs5535_mfgpt_free_timer(pwm_timer); error_mfgpt_alloc: gpio_free(GPIO_DIMM); @@ -141,7 +139,6 @@ error_mfgpt_alloc: static int ot200_backlight_remove(struct platform_device *pdev) { struct backlight_device *bl = platform_get_drvdata(pdev); - struct ot200_backlight_data *data = bl_get_data(bl); backlight_device_unregister(bl); @@ -154,7 +151,6 @@ static int ot200_backlight_remove(struct platform_device *pdev) cs5535_mfgpt_free_timer(pwm_timer); gpio_free(GPIO_DIMM); - kfree(data); return 0; } -- cgit v0.10.2 From b4a74615a4729ca4e6903bd3027339c4e6e7ae03 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:34 -0700 Subject: drivers/video/backlight/lm3533_bl.c: use devm_ functions The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_kzalloc of these functions. Signed-off-by: Jingoo Han Acked-by: Johan Hovold Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/lm3533_bl.c b/drivers/video/backlight/lm3533_bl.c index bebeb63..18dca0c 100644 --- a/drivers/video/backlight/lm3533_bl.c +++ b/drivers/video/backlight/lm3533_bl.c @@ -295,7 +295,7 @@ static int __devinit lm3533_bl_probe(struct platform_device *pdev) return -EINVAL; } - bl = kzalloc(sizeof(*bl), GFP_KERNEL); + bl = devm_kzalloc(&pdev->dev, sizeof(*bl), GFP_KERNEL); if (!bl) { dev_err(&pdev->dev, "failed to allocate memory for backlight\n"); @@ -317,8 +317,7 @@ static int __devinit lm3533_bl_probe(struct platform_device *pdev) &lm3533_bl_ops, &props); if (IS_ERR(bd)) { dev_err(&pdev->dev, "failed to register backlight device\n"); - ret = PTR_ERR(bd); - goto err_free; + return PTR_ERR(bd); } bl->bd = bd; @@ -348,8 +347,6 @@ err_sysfs_remove: sysfs_remove_group(&bd->dev.kobj, &lm3533_bl_attribute_group); err_unregister: backlight_device_unregister(bd); -err_free: - kfree(bl); return ret; } @@ -367,7 +364,6 @@ static int __devexit lm3533_bl_remove(struct platform_device *pdev) lm3533_ctrlbank_disable(&bl->cb); sysfs_remove_group(&bd->dev.kobj, &lm3533_bl_attribute_group); backlight_device_unregister(bd); - kfree(bl); return 0; } -- cgit v0.10.2 From 5f9dd5c9fd4efacc9f8c36c0b3c17a4308bbe4d9 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:36 -0700 Subject: backlight: atmel-pwm-bl: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Cc: Hans-Christian Egtvedt Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/atmel-pwm-bl.c b/drivers/video/backlight/atmel-pwm-bl.c index cd64073..df1cbb7 100644 --- a/drivers/video/backlight/atmel-pwm-bl.c +++ b/drivers/video/backlight/atmel-pwm-bl.c @@ -155,7 +155,8 @@ static int atmel_pwm_bl_probe(struct platform_device *pdev) goto err_free_mem; if (pwmbl->gpio_on != -1) { - retval = gpio_request(pwmbl->gpio_on, "gpio_atmel_pwm_bl"); + retval = devm_gpio_request(&pdev->dev, pwmbl->gpio_on, + "gpio_atmel_pwm_bl"); if (retval) { pwmbl->gpio_on = -1; goto err_free_pwm; @@ -165,7 +166,7 @@ static int atmel_pwm_bl_probe(struct platform_device *pdev) retval = gpio_direction_output(pwmbl->gpio_on, 0 ^ pdata->on_active_low); if (retval) - goto err_free_gpio; + goto err_free_pwm; } memset(&props, 0, sizeof(struct backlight_properties)); @@ -175,7 +176,7 @@ static int atmel_pwm_bl_probe(struct platform_device *pdev) &atmel_pwm_bl_ops, &props); if (IS_ERR(bldev)) { retval = PTR_ERR(bldev); - goto err_free_gpio; + goto err_free_pwm; } pwmbl->bldev = bldev; @@ -197,9 +198,6 @@ static int atmel_pwm_bl_probe(struct platform_device *pdev) err_free_bl_dev: platform_set_drvdata(pdev, NULL); backlight_device_unregister(bldev); -err_free_gpio: - if (pwmbl->gpio_on != -1) - gpio_free(pwmbl->gpio_on); err_free_pwm: pwm_channel_free(&pwmbl->pwmc); err_free_mem: @@ -210,10 +208,8 @@ static int __exit atmel_pwm_bl_remove(struct platform_device *pdev) { struct atmel_pwm_bl *pwmbl = platform_get_drvdata(pdev); - if (pwmbl->gpio_on != -1) { + if (pwmbl->gpio_on != -1) gpio_set_value(pwmbl->gpio_on, 0); - gpio_free(pwmbl->gpio_on); - } pwm_channel_disable(&pwmbl->pwmc); pwm_channel_free(&pwmbl->pwmc); backlight_device_unregister(pwmbl->bldev); -- cgit v0.10.2 From f673934c615987abe16919ede8fcd3ff6805d0d6 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:37 -0700 Subject: backlight: ot200_bl: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Acked-by: Christian Gmeiner Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/ot200_bl.c b/drivers/video/backlight/ot200_bl.c index ef637ff..469cf0f 100644 --- a/drivers/video/backlight/ot200_bl.c +++ b/drivers/video/backlight/ot200_bl.c @@ -84,7 +84,8 @@ static int ot200_backlight_probe(struct platform_device *pdev) int retval = 0; /* request gpio */ - if (gpio_request(GPIO_DIMM, "ot200 backlight dimmer") < 0) { + if (devm_gpio_request(&pdev->dev, GPIO_DIMM, + "ot200 backlight dimmer") < 0) { dev_err(&pdev->dev, "failed to request GPIO %d\n", GPIO_DIMM); return -ENODEV; } @@ -93,8 +94,7 @@ static int ot200_backlight_probe(struct platform_device *pdev) pwm_timer = cs5535_mfgpt_alloc_timer(7, MFGPT_DOMAIN_ANY); if (!pwm_timer) { dev_err(&pdev->dev, "MFGPT 7 not available\n"); - retval = -ENODEV; - goto error_mfgpt_alloc; + return -ENODEV; } data = devm_kzalloc(&pdev->dev, sizeof(*data), GFP_KERNEL); @@ -131,8 +131,6 @@ static int ot200_backlight_probe(struct platform_device *pdev) error_devm_kzalloc: cs5535_mfgpt_free_timer(pwm_timer); -error_mfgpt_alloc: - gpio_free(GPIO_DIMM); return retval; } @@ -149,7 +147,6 @@ static int ot200_backlight_remove(struct platform_device *pdev) MAX_COMP2 - dim_table[100]); cs5535_mfgpt_free_timer(pwm_timer); - gpio_free(GPIO_DIMM); return 0; } -- cgit v0.10.2 From 402ff9a26427c2c8fa5424c24ba04a7f6ecc25f1 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:39 -0700 Subject: backlight: tosa_lcd: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Cc: Dmitry Baryshkov Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/tosa_lcd.c b/drivers/video/backlight/tosa_lcd.c index 47823b8..33047a66 100644 --- a/drivers/video/backlight/tosa_lcd.c +++ b/drivers/video/backlight/tosa_lcd.c @@ -193,7 +193,7 @@ static int __devinit tosa_lcd_probe(struct spi_device *spi) data->spi = spi; dev_set_drvdata(&spi->dev, data); - ret = gpio_request(TOSA_GPIO_TG_ON, "tg #pwr"); + ret = devm_gpio_request(&spi->dev, TOSA_GPIO_TG_ON, "tg #pwr"); if (ret < 0) goto err_gpio_tg; @@ -201,7 +201,7 @@ static int __devinit tosa_lcd_probe(struct spi_device *spi) ret = gpio_direction_output(TOSA_GPIO_TG_ON, 0); if (ret < 0) - goto err_gpio_dir; + goto err_gpio_tg; mdelay(60); tosa_lcd_tg_init(data); @@ -221,8 +221,6 @@ static int __devinit tosa_lcd_probe(struct spi_device *spi) err_register: tosa_lcd_tg_off(data); -err_gpio_dir: - gpio_free(TOSA_GPIO_TG_ON); err_gpio_tg: dev_set_drvdata(&spi->dev, NULL); return ret; @@ -239,7 +237,6 @@ static int __devexit tosa_lcd_remove(struct spi_device *spi) tosa_lcd_tg_off(data); - gpio_free(TOSA_GPIO_TG_ON); dev_set_drvdata(&spi->dev, NULL); return 0; -- cgit v0.10.2 From f5b71941a5359cd6702be26715fd57d47187c0d9 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:40 -0700 Subject: backlight: tosa_bl: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Cc: Dmitry Baryshkov Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/tosa_bl.c b/drivers/video/backlight/tosa_bl.c index 0d54e60..49342e1 100644 --- a/drivers/video/backlight/tosa_bl.c +++ b/drivers/video/backlight/tosa_bl.c @@ -92,14 +92,14 @@ static int __devinit tosa_bl_probe(struct i2c_client *client, data->comadj = sharpsl_param.comadj == -1 ? COMADJ_DEFAULT : sharpsl_param.comadj; - ret = gpio_request(TOSA_GPIO_BL_C20MA, "backlight"); + ret = devm_gpio_request(&client->dev, TOSA_GPIO_BL_C20MA, "backlight"); if (ret) { dev_dbg(&data->bl->dev, "Unable to request gpio!\n"); return ret; } ret = gpio_direction_output(TOSA_GPIO_BL_C20MA, 0); if (ret) - goto err_gpio_dir; + return ret; i2c_set_clientdata(client, data); data->i2c = client; @@ -123,8 +123,6 @@ static int __devinit tosa_bl_probe(struct i2c_client *client, err_reg: data->bl = NULL; -err_gpio_dir: - gpio_free(TOSA_GPIO_BL_C20MA); return ret; } @@ -135,8 +133,6 @@ static int __devexit tosa_bl_remove(struct i2c_client *client) backlight_device_unregister(data->bl); data->bl = NULL; - gpio_free(TOSA_GPIO_BL_C20MA); - return 0; } -- cgit v0.10.2 From 04e961fb1da31258f7bf4480a59518c3de3caa60 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:43 -0700 Subject: backlight: lms283gf05: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Acked-by: Marek Vasut Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/lms283gf05.c b/drivers/video/backlight/lms283gf05.c index a9f2c36..ea43f22 100644 --- a/drivers/video/backlight/lms283gf05.c +++ b/drivers/video/backlight/lms283gf05.c @@ -158,29 +158,27 @@ static int __devinit lms283gf05_probe(struct spi_device *spi) int ret = 0; if (pdata != NULL) { - ret = gpio_request(pdata->reset_gpio, "LMS285GF05 RESET"); + ret = devm_gpio_request(&spi->dev, pdata->reset_gpio, + "LMS285GF05 RESET"); if (ret) return ret; ret = gpio_direction_output(pdata->reset_gpio, !pdata->reset_inverted); if (ret) - goto err; + return ret; } st = devm_kzalloc(&spi->dev, sizeof(struct lms283gf05_state), GFP_KERNEL); if (st == NULL) { dev_err(&spi->dev, "No memory for device state\n"); - ret = -ENOMEM; - goto err; + return -ENOMEM; } ld = lcd_device_register("lms283gf05", &spi->dev, st, &lms_ops); - if (IS_ERR(ld)) { - ret = PTR_ERR(ld); - goto err; - } + if (IS_ERR(ld)) + return PTR_ERR(ld); st->spi = spi; st->ld = ld; @@ -193,24 +191,14 @@ static int __devinit lms283gf05_probe(struct spi_device *spi) lms283gf05_toggle(spi, disp_initseq, ARRAY_SIZE(disp_initseq)); return 0; - -err: - if (pdata != NULL) - gpio_free(pdata->reset_gpio); - - return ret; } static int __devexit lms283gf05_remove(struct spi_device *spi) { struct lms283gf05_state *st = dev_get_drvdata(&spi->dev); - struct lms283gf05_pdata *pdata = st->spi->dev.platform_data; lcd_device_unregister(st->ld); - if (pdata != NULL) - gpio_free(pdata->reset_gpio); - return 0; } -- cgit v0.10.2 From 4405d5d9dd76740bfb715fc6a32343e087a13e1e Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:45 -0700 Subject: backlight: corgi_lcd: use devm_gpio_request() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request() for these functions. Signed-off-by: Jingoo Han Cc: Eric Miao Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/corgi_lcd.c b/drivers/video/backlight/corgi_lcd.c index 23d7326..c781768 100644 --- a/drivers/video/backlight/corgi_lcd.c +++ b/drivers/video/backlight/corgi_lcd.c @@ -492,7 +492,8 @@ static int setup_gpio_backlight(struct corgi_lcd *lcd, lcd->gpio_backlight_cont = -1; if (gpio_is_valid(pdata->gpio_backlight_on)) { - err = gpio_request(pdata->gpio_backlight_on, "BL_ON"); + err = devm_gpio_request(&spi->dev, pdata->gpio_backlight_on, + "BL_ON"); if (err) { dev_err(&spi->dev, "failed to request GPIO%d for " "backlight_on\n", pdata->gpio_backlight_on); @@ -504,11 +505,12 @@ static int setup_gpio_backlight(struct corgi_lcd *lcd, } if (gpio_is_valid(pdata->gpio_backlight_cont)) { - err = gpio_request(pdata->gpio_backlight_cont, "BL_CONT"); + err = devm_gpio_request(&spi->dev, pdata->gpio_backlight_cont, + "BL_CONT"); if (err) { dev_err(&spi->dev, "failed to request GPIO%d for " "backlight_cont\n", pdata->gpio_backlight_cont); - goto err_free_backlight_on; + return err; } lcd->gpio_backlight_cont = pdata->gpio_backlight_cont; @@ -525,11 +527,6 @@ static int setup_gpio_backlight(struct corgi_lcd *lcd, } } return 0; - -err_free_backlight_on: - if (gpio_is_valid(lcd->gpio_backlight_on)) - gpio_free(lcd->gpio_backlight_on); - return err; } static int __devinit corgi_lcd_probe(struct spi_device *spi) @@ -602,12 +599,6 @@ static int __devexit corgi_lcd_remove(struct spi_device *spi) backlight_update_status(lcd->bl_dev); backlight_device_unregister(lcd->bl_dev); - if (gpio_is_valid(lcd->gpio_backlight_on)) - gpio_free(lcd->gpio_backlight_on); - - if (gpio_is_valid(lcd->gpio_backlight_cont)) - gpio_free(lcd->gpio_backlight_cont); - corgi_lcd_set_power(lcd->lcd_dev, FB_BLANK_POWERDOWN); lcd_device_unregister(lcd->lcd_dev); -- cgit v0.10.2 From 62a5c075bafb91895506bc68f76db1ff1b5d1a02 Mon Sep 17 00:00:00 2001 From: Jingoo Han Date: Mon, 30 Jul 2012 14:40:47 -0700 Subject: backlight: l4f00242t03: export and use devm_gpio_request_one() The devm_ functions allocate memory that is released when a driver detaches. This patch uses devm_gpio_request_one() for these functions. Signed-off-by: Jingoo Han Cc: Alberto Panizzo Cc: Richard Purdie Cc: Grant Likely Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/l4f00242t03.c b/drivers/video/backlight/l4f00242t03.c index 40f606a..2d90c06 100644 --- a/drivers/video/backlight/l4f00242t03.c +++ b/drivers/video/backlight/l4f00242t03.c @@ -175,28 +175,27 @@ static int __devinit l4f00242t03_probe(struct spi_device *spi) priv->spi = spi; - ret = gpio_request_one(pdata->reset_gpio, GPIOF_OUT_INIT_HIGH, - "lcd l4f00242t03 reset"); + ret = devm_gpio_request_one(&spi->dev, pdata->reset_gpio, + GPIOF_OUT_INIT_HIGH, "lcd l4f00242t03 reset"); if (ret) { dev_err(&spi->dev, "Unable to get the lcd l4f00242t03 reset gpio.\n"); return ret; } - ret = gpio_request_one(pdata->data_enable_gpio, GPIOF_OUT_INIT_LOW, - "lcd l4f00242t03 data enable"); + ret = devm_gpio_request_one(&spi->dev, pdata->data_enable_gpio, + GPIOF_OUT_INIT_LOW, "lcd l4f00242t03 data enable"); if (ret) { dev_err(&spi->dev, "Unable to get the lcd l4f00242t03 data en gpio.\n"); - goto err; + return ret; } priv->io_reg = regulator_get(&spi->dev, "vdd"); if (IS_ERR(priv->io_reg)) { - ret = PTR_ERR(priv->io_reg); dev_err(&spi->dev, "%s: Unable to get the IO regulator\n", __func__); - goto err2; + return PTR_ERR(priv->io_reg); } priv->core_reg = regulator_get(&spi->dev, "vcore"); @@ -204,14 +203,14 @@ static int __devinit l4f00242t03_probe(struct spi_device *spi) ret = PTR_ERR(priv->core_reg); dev_err(&spi->dev, "%s: Unable to get the core regulator\n", __func__); - goto err3; + goto err1; } priv->ld = lcd_device_register("l4f00242t03", &spi->dev, priv, &l4f_ops); if (IS_ERR(priv->ld)) { ret = PTR_ERR(priv->ld); - goto err4; + goto err2; } /* Init the LCD */ @@ -223,14 +222,10 @@ static int __devinit l4f00242t03_probe(struct spi_device *spi) return 0; -err4: +err2: regulator_put(priv->core_reg); -err3: +err1: regulator_put(priv->io_reg); -err2: - gpio_free(pdata->data_enable_gpio); -err: - gpio_free(pdata->reset_gpio); return ret; } @@ -238,16 +233,12 @@ err: static int __devexit l4f00242t03_remove(struct spi_device *spi) { struct l4f00242t03_priv *priv = dev_get_drvdata(&spi->dev); - struct l4f00242t03_pdata *pdata = priv->spi->dev.platform_data; l4f00242t03_lcd_power_set(priv->ld, FB_BLANK_POWERDOWN); lcd_device_unregister(priv->ld); dev_set_drvdata(&spi->dev, NULL); - gpio_free(pdata->data_enable_gpio); - gpio_free(pdata->reset_gpio); - regulator_put(priv->io_reg); regulator_put(priv->core_reg); -- cgit v0.10.2 From a1fcb2e31822c0617c6e274de4af2a2bb5dc7d3f Mon Sep 17 00:00:00 2001 From: "Kim, Milo" Date: Mon, 30 Jul 2012 14:40:50 -0700 Subject: backlight: move register definitions from header to source ROM boundary definitions do not need to be exported because these are used only internally in the lp855x driver. And few code cosmetic changes Signed-off-by: Milo(Woogyom) Kim Cc: Richard Purdie Cc: Bryan Wu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/lp855x_bl.c b/drivers/video/backlight/lp855x_bl.c index 72a0e0c..3d24314 100644 --- a/drivers/video/backlight/lp855x_bl.c +++ b/drivers/video/backlight/lp855x_bl.c @@ -17,8 +17,12 @@ #include /* Registers */ -#define BRIGHTNESS_CTRL (0x00) -#define DEVICE_CTRL (0x01) +#define BRIGHTNESS_CTRL 0x00 +#define DEVICE_CTRL 0x01 +#define EEPROM_START 0xA0 +#define EEPROM_END 0xA7 +#define EPROM_START 0xA0 +#define EPROM_END 0xAF #define BUF_SIZE 20 #define DEFAULT_BL_NAME "lcd-backlight" diff --git a/include/linux/lp855x.h b/include/linux/lp855x.h index 781a490..cc76f1f 100644 --- a/include/linux/lp855x.h +++ b/include/linux/lp855x.h @@ -47,12 +47,6 @@ (LP8556_I2C_ONLY << BRT_MODE_SHFT)) #define LP8556_COMB2_CONFIG (LP8556_COMBINED2 << BRT_MODE_SHFT) -/* ROM area boundary */ -#define EEPROM_START (0xA0) -#define EEPROM_END (0xA7) -#define EPROM_START (0xA0) -#define EPROM_END (0xAF) - enum lp855x_chip_id { LP8550, LP8551, -- cgit v0.10.2 From f7f95056779eb69c5fc3ac30e5cb6fd28bdbba43 Mon Sep 17 00:00:00 2001 From: "Kim, Milo" Date: Mon, 30 Jul 2012 14:40:53 -0700 Subject: backlight: move lp855x header into platform_data directory The lp855x header is used only in the platform side, so it can be moved into platform_data directory Signed-off-by: Milo(Woogyom) Kim Cc: Richard Purdie Cc: Bryan Wu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/video/backlight/lp855x_bl.c b/drivers/video/backlight/lp855x_bl.c index 3d24314..aa6d4f7 100644 --- a/drivers/video/backlight/lp855x_bl.c +++ b/drivers/video/backlight/lp855x_bl.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include /* Registers */ #define BRIGHTNESS_CTRL 0x00 diff --git a/include/linux/lp855x.h b/include/linux/lp855x.h deleted file mode 100644 index cc76f1f..0000000 --- a/include/linux/lp855x.h +++ /dev/null @@ -1,125 +0,0 @@ -/* - * LP855x Backlight Driver - * - * Copyright (C) 2011 Texas Instruments - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - * - */ - -#ifndef _LP855X_H -#define _LP855X_H - -#define BL_CTL_SHFT (0) -#define BRT_MODE_SHFT (1) -#define BRT_MODE_MASK (0x06) - -/* Enable backlight. Only valid when BRT_MODE=10(I2C only) */ -#define ENABLE_BL (1) -#define DISABLE_BL (0) - -#define I2C_CONFIG(id) id ## _I2C_CONFIG -#define PWM_CONFIG(id) id ## _PWM_CONFIG - -/* DEVICE CONTROL register - LP8550 */ -#define LP8550_PWM_CONFIG (LP8550_PWM_ONLY << BRT_MODE_SHFT) -#define LP8550_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ - (LP8550_I2C_ONLY << BRT_MODE_SHFT)) - -/* DEVICE CONTROL register - LP8551 */ -#define LP8551_PWM_CONFIG LP8550_PWM_CONFIG -#define LP8551_I2C_CONFIG LP8550_I2C_CONFIG - -/* DEVICE CONTROL register - LP8552 */ -#define LP8552_PWM_CONFIG LP8550_PWM_CONFIG -#define LP8552_I2C_CONFIG LP8550_I2C_CONFIG - -/* DEVICE CONTROL register - LP8553 */ -#define LP8553_PWM_CONFIG LP8550_PWM_CONFIG -#define LP8553_I2C_CONFIG LP8550_I2C_CONFIG - -/* DEVICE CONTROL register - LP8556 */ -#define LP8556_PWM_CONFIG (LP8556_PWM_ONLY << BRT_MODE_SHFT) -#define LP8556_COMB1_CONFIG (LP8556_COMBINED1 << BRT_MODE_SHFT) -#define LP8556_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ - (LP8556_I2C_ONLY << BRT_MODE_SHFT)) -#define LP8556_COMB2_CONFIG (LP8556_COMBINED2 << BRT_MODE_SHFT) - -enum lp855x_chip_id { - LP8550, - LP8551, - LP8552, - LP8553, - LP8556, -}; - -enum lp855x_brightness_ctrl_mode { - PWM_BASED = 1, - REGISTER_BASED, -}; - -enum lp8550_brighntess_source { - LP8550_PWM_ONLY, - LP8550_I2C_ONLY = 2, -}; - -enum lp8551_brighntess_source { - LP8551_PWM_ONLY = LP8550_PWM_ONLY, - LP8551_I2C_ONLY = LP8550_I2C_ONLY, -}; - -enum lp8552_brighntess_source { - LP8552_PWM_ONLY = LP8550_PWM_ONLY, - LP8552_I2C_ONLY = LP8550_I2C_ONLY, -}; - -enum lp8553_brighntess_source { - LP8553_PWM_ONLY = LP8550_PWM_ONLY, - LP8553_I2C_ONLY = LP8550_I2C_ONLY, -}; - -enum lp8556_brightness_source { - LP8556_PWM_ONLY, - LP8556_COMBINED1, /* pwm + i2c before the shaper block */ - LP8556_I2C_ONLY, - LP8556_COMBINED2, /* pwm + i2c after the shaper block */ -}; - -struct lp855x_pwm_data { - void (*pwm_set_intensity) (int brightness, int max_brightness); - int (*pwm_get_intensity) (int max_brightness); -}; - -struct lp855x_rom_data { - u8 addr; - u8 val; -}; - -/** - * struct lp855x_platform_data - * @name : Backlight driver name. If it is not defined, default name is set. - * @mode : brightness control by pwm or lp855x register - * @device_control : value of DEVICE CONTROL register - * @initial_brightness : initial value of backlight brightness - * @pwm_data : platform specific pwm generation functions. - Only valid when mode is PWM_BASED. - * @load_new_rom_data : - 0 : use default configuration data - 1 : update values of eeprom or eprom registers on loading driver - * @size_program : total size of lp855x_rom_data - * @rom_data : list of new eeprom/eprom registers - */ -struct lp855x_platform_data { - char *name; - enum lp855x_brightness_ctrl_mode mode; - u8 device_control; - int initial_brightness; - struct lp855x_pwm_data pwm_data; - u8 load_new_rom_data; - int size_program; - struct lp855x_rom_data *rom_data; -}; - -#endif diff --git a/include/linux/platform_data/lp855x.h b/include/linux/platform_data/lp855x.h new file mode 100644 index 0000000..cc76f1f --- /dev/null +++ b/include/linux/platform_data/lp855x.h @@ -0,0 +1,125 @@ +/* + * LP855x Backlight Driver + * + * Copyright (C) 2011 Texas Instruments + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#ifndef _LP855X_H +#define _LP855X_H + +#define BL_CTL_SHFT (0) +#define BRT_MODE_SHFT (1) +#define BRT_MODE_MASK (0x06) + +/* Enable backlight. Only valid when BRT_MODE=10(I2C only) */ +#define ENABLE_BL (1) +#define DISABLE_BL (0) + +#define I2C_CONFIG(id) id ## _I2C_CONFIG +#define PWM_CONFIG(id) id ## _PWM_CONFIG + +/* DEVICE CONTROL register - LP8550 */ +#define LP8550_PWM_CONFIG (LP8550_PWM_ONLY << BRT_MODE_SHFT) +#define LP8550_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ + (LP8550_I2C_ONLY << BRT_MODE_SHFT)) + +/* DEVICE CONTROL register - LP8551 */ +#define LP8551_PWM_CONFIG LP8550_PWM_CONFIG +#define LP8551_I2C_CONFIG LP8550_I2C_CONFIG + +/* DEVICE CONTROL register - LP8552 */ +#define LP8552_PWM_CONFIG LP8550_PWM_CONFIG +#define LP8552_I2C_CONFIG LP8550_I2C_CONFIG + +/* DEVICE CONTROL register - LP8553 */ +#define LP8553_PWM_CONFIG LP8550_PWM_CONFIG +#define LP8553_I2C_CONFIG LP8550_I2C_CONFIG + +/* DEVICE CONTROL register - LP8556 */ +#define LP8556_PWM_CONFIG (LP8556_PWM_ONLY << BRT_MODE_SHFT) +#define LP8556_COMB1_CONFIG (LP8556_COMBINED1 << BRT_MODE_SHFT) +#define LP8556_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ + (LP8556_I2C_ONLY << BRT_MODE_SHFT)) +#define LP8556_COMB2_CONFIG (LP8556_COMBINED2 << BRT_MODE_SHFT) + +enum lp855x_chip_id { + LP8550, + LP8551, + LP8552, + LP8553, + LP8556, +}; + +enum lp855x_brightness_ctrl_mode { + PWM_BASED = 1, + REGISTER_BASED, +}; + +enum lp8550_brighntess_source { + LP8550_PWM_ONLY, + LP8550_I2C_ONLY = 2, +}; + +enum lp8551_brighntess_source { + LP8551_PWM_ONLY = LP8550_PWM_ONLY, + LP8551_I2C_ONLY = LP8550_I2C_ONLY, +}; + +enum lp8552_brighntess_source { + LP8552_PWM_ONLY = LP8550_PWM_ONLY, + LP8552_I2C_ONLY = LP8550_I2C_ONLY, +}; + +enum lp8553_brighntess_source { + LP8553_PWM_ONLY = LP8550_PWM_ONLY, + LP8553_I2C_ONLY = LP8550_I2C_ONLY, +}; + +enum lp8556_brightness_source { + LP8556_PWM_ONLY, + LP8556_COMBINED1, /* pwm + i2c before the shaper block */ + LP8556_I2C_ONLY, + LP8556_COMBINED2, /* pwm + i2c after the shaper block */ +}; + +struct lp855x_pwm_data { + void (*pwm_set_intensity) (int brightness, int max_brightness); + int (*pwm_get_intensity) (int max_brightness); +}; + +struct lp855x_rom_data { + u8 addr; + u8 val; +}; + +/** + * struct lp855x_platform_data + * @name : Backlight driver name. If it is not defined, default name is set. + * @mode : brightness control by pwm or lp855x register + * @device_control : value of DEVICE CONTROL register + * @initial_brightness : initial value of backlight brightness + * @pwm_data : platform specific pwm generation functions. + Only valid when mode is PWM_BASED. + * @load_new_rom_data : + 0 : use default configuration data + 1 : update values of eeprom or eprom registers on loading driver + * @size_program : total size of lp855x_rom_data + * @rom_data : list of new eeprom/eprom registers + */ +struct lp855x_platform_data { + char *name; + enum lp855x_brightness_ctrl_mode mode; + u8 device_control; + int initial_brightness; + struct lp855x_pwm_data pwm_data; + u8 load_new_rom_data; + int size_program; + struct lp855x_rom_data *rom_data; +}; + +#endif -- cgit v0.10.2 From 639b9e34f15e4b2c30068a4e4485586af0cdf709 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:40:55 -0700 Subject: string: introduce memweight() memweight() is the function that counts the total number of bits set in memory area. Unlike bitmap_weight(), memweight() takes pointer and size in bytes to specify a memory area which does not need to be aligned to long-word boundary. [akpm@linux-foundation.org: rename `w' to `ret'] Signed-off-by: Akinobu Mita Cc: Anders Larsen Cc: Alasdair Kergon Cc: Laurent Pinchart Cc: Mark Fasheh Cc: Joel Becker Cc: Jan Kara Cc: Andreas Dilger Cc: "Theodore Ts'o" Cc: Matthew Wilcox Cc: Mauro Carvalho Chehab Cc: Tony Luck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/string.h b/include/linux/string.h index e033564..ffe0442 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -145,4 +145,7 @@ static inline bool strstarts(const char *str, const char *prefix) return strncmp(str, prefix, strlen(prefix)) == 0; } #endif + +extern size_t memweight(const void *ptr, size_t bytes); + #endif /* _LINUX_STRING_H_ */ diff --git a/lib/Makefile b/lib/Makefile index 8c31a0c..df663cc 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -22,7 +22,7 @@ lib-y += kobject.o klist.o obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \ bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o \ string_helpers.o gcd.o lcm.o list_sort.o uuid.o flex_array.o \ - bsearch.o find_last_bit.o find_next_bit.o llist.o + bsearch.o find_last_bit.o find_next_bit.o llist.o memweight.o obj-y += kstrtox.o obj-$(CONFIG_TEST_KSTRTOX) += test-kstrtox.o diff --git a/lib/memweight.c b/lib/memweight.c new file mode 100644 index 0000000..e35fc87 --- /dev/null +++ b/lib/memweight.c @@ -0,0 +1,38 @@ +#include +#include +#include + +/** + * memweight - count the total number of bits set in memory area + * @ptr: pointer to the start of the area + * @bytes: the size of the area + */ +size_t memweight(const void *ptr, size_t bytes) +{ + size_t ret = 0; + size_t longs; + const unsigned char *bitmap = ptr; + + for (; bytes > 0 && ((unsigned long)bitmap) % sizeof(long); + bytes--, bitmap++) + ret += hweight8(*bitmap); + + longs = bytes / sizeof(long); + if (longs) { + BUG_ON(longs >= INT_MAX / BITS_PER_LONG); + ret += bitmap_weight((unsigned long *)bitmap, + longs * BITS_PER_LONG); + bytes -= longs * sizeof(long); + bitmap += longs * sizeof(long); + } + /* + * The reason that this last loop is distinct from the preceding + * bitmap_weight() call is to compute 1-bits in the last region smaller + * than sizeof(long) properly on big-endian systems. + */ + for (; bytes > 0; bytes--, bitmap++) + ret += hweight8(*bitmap); + + return ret; +} +EXPORT_SYMBOL(memweight); -- cgit v0.10.2 From 9b58f6d4aaef070bf6e0744713b6d2b6fc1b3578 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:40:57 -0700 Subject: qnx4fs: use memweight() Use memweight() to count the total number of bits clear in memory area. Note that this memweight() call can't be replaced with a single bitmap_weight() call, although the pointer to the memory area is aligned to long-word boundary. Because the size of the memory area may not be a multiple of BITS_PER_LONG, then it returns wrong value on big-endian architecture. Signed-off-by: Akinobu Mita Acked-by: Anders Larsen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/qnx4/bitmap.c b/fs/qnx4/bitmap.c index 22e0d60..76a7a69 100644 --- a/fs/qnx4/bitmap.c +++ b/fs/qnx4/bitmap.c @@ -17,23 +17,6 @@ #include #include "qnx4.h" -static void count_bits(register const char *bmPart, register int size, - int *const tf) -{ - char b; - int tot = *tf; - - if (size > QNX4_BLOCK_SIZE) { - size = QNX4_BLOCK_SIZE; - } - do { - b = *bmPart++; - tot += 8 - hweight8(b); - size--; - } while (size != 0); - *tf = tot; -} - unsigned long qnx4_count_free_blocks(struct super_block *sb) { int start = le32_to_cpu(qnx4_sb(sb)->BitMap->di_first_xtnt.xtnt_blk) - 1; @@ -44,13 +27,16 @@ unsigned long qnx4_count_free_blocks(struct super_block *sb) struct buffer_head *bh; while (total < size) { + int bytes = min(size - total, QNX4_BLOCK_SIZE); + if ((bh = sb_bread(sb, start + offset)) == NULL) { printk(KERN_ERR "qnx4: I/O error in counting free blocks\n"); break; } - count_bits(bh->b_data, size - total, &total_free); + total_free += bytes * BITS_PER_BYTE - + memweight(bh->b_data, bytes); brelse(bh); - total += QNX4_BLOCK_SIZE; + total += bytes; offset++; } -- cgit v0.10.2 From 8fb980e35b41bbb79aa47f48839d95f0354af3c4 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:40:59 -0700 Subject: dm: use memweight() Use memweight() to count the total number of bits set in memory area. Signed-off-by: Akinobu Mita Cc: Alasdair Kergon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/md/dm-log.c b/drivers/md/dm-log.c index 65ebaeb..627d191 100644 --- a/drivers/md/dm-log.c +++ b/drivers/md/dm-log.c @@ -571,16 +571,6 @@ static void disk_dtr(struct dm_dirty_log *log) destroy_log_context(lc); } -static int count_bits32(uint32_t *addr, unsigned size) -{ - int count = 0, i; - - for (i = 0; i < size; i++) { - count += hweight32(*(addr+i)); - } - return count; -} - static void fail_log_device(struct log_c *lc) { if (lc->log_dev_failed) @@ -629,7 +619,8 @@ static int disk_resume(struct dm_dirty_log *log) /* copy clean across to sync */ memcpy(lc->sync_bits, lc->clean_bits, size); - lc->sync_count = count_bits32(lc->clean_bits, lc->bitset_uint32_count); + lc->sync_count = memweight(lc->clean_bits, + lc->bitset_uint32_count * sizeof(uint32_t)); lc->sync_search = 0; /* set the correct number of regions in the header */ -- cgit v0.10.2 From 0121ad62c20ed779e38ad689071da2805f03249f Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:00 -0700 Subject: affs: use memweight() Use memweight() to count the total number of bits set in memory area. Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/affs/bitmap.c b/fs/affs/bitmap.c index 6e0be43..a32246b 100644 --- a/fs/affs/bitmap.c +++ b/fs/affs/bitmap.c @@ -10,30 +10,6 @@ #include #include "affs.h" -/* This is, of course, shamelessly stolen from fs/minix */ - -static const int nibblemap[] = { 0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4 }; - -static u32 -affs_count_free_bits(u32 blocksize, const void *data) -{ - const u32 *map; - u32 free; - u32 tmp; - - map = data; - free = 0; - for (blocksize /= 4; blocksize > 0; blocksize--) { - tmp = *map++; - while (tmp) { - free += nibblemap[tmp & 0xf]; - tmp >>= 4; - } - } - - return free; -} - u32 affs_count_free_blocks(struct super_block *sb) { @@ -317,7 +293,7 @@ int affs_init_bitmap(struct super_block *sb, int *flags) goto out; } pr_debug("AFFS: read bitmap block %d: %d\n", blk, bm->bm_key); - bm->bm_free = affs_count_free_bits(sb->s_blocksize - 4, bh->b_data + 4); + bm->bm_free = memweight(bh->b_data + 4, sb->s_blocksize - 4); /* Don't try read the extension if this is the last block, * but we also need the right bm pointer below @@ -367,7 +343,7 @@ int affs_init_bitmap(struct super_block *sb, int *flags) /* recalculate bitmap count for last block */ bm--; - bm->bm_free = affs_count_free_bits(sb->s_blocksize - 4, bh->b_data + 4); + bm->bm_free = memweight(bh->b_data + 4, sb->s_blocksize - 4); out: affs_brelse(bh); -- cgit v0.10.2 From 9a601396e8681d9884a9f429cba7ceb2b2a4d21c Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:02 -0700 Subject: video/uvc: use memweight() Use memweight() to count the total number of bits set in memory area. Signed-off-by: Akinobu Mita Acked-by: Laurent Pinchart Acked-by: Mauro Carvalho Chehab Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/media/video/uvc/uvc_ctrl.c b/drivers/media/video/uvc/uvc_ctrl.c index af26bbe..f7061a5 100644 --- a/drivers/media/video/uvc/uvc_ctrl.c +++ b/drivers/media/video/uvc/uvc_ctrl.c @@ -2083,7 +2083,7 @@ int uvc_ctrl_init_device(struct uvc_device *dev) /* Walk the entities list and instantiate controls */ list_for_each_entry(entity, &dev->entities, list) { struct uvc_control *ctrl; - unsigned int bControlSize = 0, ncontrols = 0; + unsigned int bControlSize = 0, ncontrols; __u8 *bmControls = NULL; if (UVC_ENTITY_TYPE(entity) == UVC_VC_EXTENSION_UNIT) { @@ -2101,8 +2101,7 @@ int uvc_ctrl_init_device(struct uvc_device *dev) uvc_ctrl_prune_entity(dev, entity); /* Count supported controls and allocate the controls array */ - for (i = 0; i < bControlSize; ++i) - ncontrols += hweight8(bmControls[i]); + ncontrols = memweight(bmControls, bControlSize); if (ncontrols == 0) continue; -- cgit v0.10.2 From a75613ec73ec87726a81fe421385a13c25fdcfc4 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:03 -0700 Subject: ocfs2: use memweight() Use memweight to count the total number of bits set in memory area. Signed-off-by: Akinobu Mita Cc: Mark Fasheh Cc: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index 210c352..a9f78c7 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -784,14 +784,10 @@ bail: static u32 ocfs2_local_alloc_count_bits(struct ocfs2_dinode *alloc) { - int i; - u8 *buffer; - u32 count = 0; + u32 count; struct ocfs2_local_alloc *la = OCFS2_LOCAL_ALLOC(alloc); - buffer = la->la_bitmap; - for (i = 0; i < le16_to_cpu(la->la_size); i++) - count += hweight8(buffer[i]); + count = memweight(la->la_bitmap, le16_to_cpu(la->la_size)); trace_ocfs2_local_alloc_count_bits(count); return count; -- cgit v0.10.2 From ecd0afa3ced0ebf36901b53fd9ee431f8a34a161 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:05 -0700 Subject: ext2: use memweight() Convert ext2_count_free() to use memweight() instead of table lookup based counting clear bits implementation. This change only affects the code segments enabled by EXT2FS_DEBUG. Note that this memweight() call can't be replaced with a single bitmap_weight() call, although the pointer to the memory area is aligned to long-word boundary. Because the size of the memory area may not be a multiple of BITS_PER_LONG, then it returns wrong value on big-endian architecture. This also includes the following changes. - Remove unnecessary map == NULL check in ext2_count_free() which always takes non-null pointer as the memory area. - Fix printk format warning that only reveals with EXT2FS_DEBUG. Signed-off-by: Akinobu Mita Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index 1c36139..376aa77 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -1444,19 +1444,9 @@ ext2_fsblk_t ext2_new_block(struct inode *inode, unsigned long goal, int *errp) #ifdef EXT2FS_DEBUG -static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; - -unsigned long ext2_count_free (struct buffer_head * map, unsigned int numchars) +unsigned long ext2_count_free(struct buffer_head *map, unsigned int numchars) { - unsigned int i; - unsigned long sum = 0; - - if (!map) - return (0); - for (i = 0; i < numchars; i++) - sum += nibblemap[map->b_data[i] & 0xf] + - nibblemap[(map->b_data[i] >> 4) & 0xf]; - return (sum); + return numchars * BITS_PER_BYTE - memweight(map->b_data, numchars); } #endif /* EXT2FS_DEBUG */ diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index c13eb7b..8f370e01 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -644,6 +644,7 @@ unsigned long ext2_count_free_inodes (struct super_block * sb) } brelse(bitmap_bh); printk("ext2_count_free_inodes: stored = %lu, computed = %lu, %lu\n", + (unsigned long) percpu_counter_read(&EXT2_SB(sb)->s_freeinodes_counter), desc_count, bitmap_count); return desc_count; -- cgit v0.10.2 From 10d470849a7c6dd360e8ad4770160ad7af9adb4b Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:06 -0700 Subject: ext3: use memweight() Convert ext3_count_free() to use memweight() instead of table lookup based counting clear bits implementation. This change only affects the code segments enabled by EXT3FS_DEBUG. Note that this memweight() call can't be replaced with a single bitmap_weight() call, although the pointer to the memory area is aligned to long-word boundary. Because the size of the memory area may not be a multiple of BITS_PER_LONG, then it returns wrong value on big-endian architecture. This also includes the following changes. - Remove unnecessary map == NULL check in ext3_count_free() which always takes non-null pointer as the memory area. - Fix printk format warning that only reveals with EXT3FS_DEBUG. Signed-off-by: Akinobu Mita Acked-by: Jan Kara Cc: Andreas Dilger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index 25cd608..90d901f 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -1813,7 +1813,7 @@ ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb) brelse(bitmap_bh); printk("ext3_count_free_blocks: stored = "E3FSBLK ", computed = "E3FSBLK", "E3FSBLK"\n", - le32_to_cpu(es->s_free_blocks_count), + (ext3_fsblk_t)le32_to_cpu(es->s_free_blocks_count), desc_count, bitmap_count); return bitmap_count; #else diff --git a/fs/ext3/bitmap.c b/fs/ext3/bitmap.c index 909d13e..ef9c643 100644 --- a/fs/ext3/bitmap.c +++ b/fs/ext3/bitmap.c @@ -11,19 +11,9 @@ #ifdef EXT3FS_DEBUG -static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; - unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars) { - unsigned int i; - unsigned long sum = 0; - - if (!map) - return (0); - for (i = 0; i < numchars; i++) - sum += nibblemap[map->b_data[i] & 0xf] + - nibblemap[(map->b_data[i] >> 4) & 0xf]; - return (sum); + return numchars * BITS_PER_BYTE - memweight(map->b_data, numchars); } #endif /* EXT3FS_DEBUG */ -- cgit v0.10.2 From 6017b485caeae5915956190b4f3d8307021e785d Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:41:08 -0700 Subject: ext4: use memweight() Convert ext4_count_free() to use memweight() instead of table lookup based counting clear bits implementation. This change only affects the code segments enabled by EXT4FS_DEBUG. Note that this memweight() call can't be replaced with a single bitmap_weight() call, although the pointer to the memory area is aligned to long-word boundary. Because the size of the memory area may not be a multiple of BITS_PER_LONG, then it returns wrong value on big-endian architecture. This also includes the following change. - Remove unnecessary map == NULL check in ext4_count_free() which always takes non-null pointer as the memory area. Signed-off-by: Akinobu Mita Cc: "Theodore Ts'o" Cc: Andreas Dilger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c index a94b9c6..f8716ea 100644 --- a/fs/ext4/bitmap.c +++ b/fs/ext4/bitmap.c @@ -11,16 +11,9 @@ #include #include "ext4.h" -static const int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; - unsigned int ext4_count_free(char *bitmap, unsigned int numchars) { - unsigned int i, sum = 0; - - for (i = 0; i < numchars; i++) - sum += nibblemap[bitmap[i] & 0xf] + - nibblemap[(bitmap[i] >> 4) & 0xf]; - return sum; + return numchars * BITS_PER_BYTE - memweight(bitmap, numchars); } int ext4_inode_bitmap_csum_verify(struct super_block *sb, ext4_group_t group, -- cgit v0.10.2 From 7463449b8287162454d7e00bf7fd2c64f72c1dc8 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Mon, 30 Jul 2012 14:41:09 -0700 Subject: atomic64_test: simplify the #ifdef for atomic64_dec_if_positive() test Introduce CONFIG_ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE and use this instead of the multitude of #if defined() checks in atomic64_test.c Signed-off-by: Catalin Marinas Cc: Russell King Cc: Ralf Baechle Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index fbdd853..4172c3c 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -11,6 +11,7 @@ config ARM select RTC_LIB select SYS_SUPPORTS_APM_EMULATION select GENERIC_ATOMIC64 if (CPU_V6 || !CPU_32v6K || !AEABI) + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select HAVE_OPROFILE if (HAVE_PERF_EVENTS) select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL select HAVE_ARCH_KGDB diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index b3e10fd..5e238d0 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -20,6 +20,7 @@ config MIPS select ARCH_BINFMT_ELF_RANDOMIZE_PIE select RTC_LIB if !MACH_LOONGSON select GENERIC_ATOMIC64 if !64BIT + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select HAVE_DMA_ATTRS select HAVE_DMA_API_DEBUG select HAVE_GENERIC_HARDIRQS diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 9a5d3cd..4f681b7 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -115,6 +115,7 @@ config PPC select HAVE_OPROFILE select HAVE_SYSCALL_WRAPPERS if PPC64 select GENERIC_ATOMIC64 if PPC32 + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select HAVE_IRQ_WORK select HAVE_PERF_EVENTS select HAVE_REGS_AND_STACK_ACCESS_API diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index a39b469..d0a5e92 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -85,6 +85,7 @@ config S390 select HAVE_ARCH_MUTEX_CPU_RELAX select HAVE_ARCH_JUMP_LABEL if !MARCH_G5 select ARCH_SAVE_PAGE_KEYS if HIBERNATION + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP select HAVE_CMPXCHG_LOCAL diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index c70684f..ca4fdef 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -70,6 +70,7 @@ config X86 select HAVE_ARCH_JUMP_LABEL select HAVE_TEXT_POKE_SMP select HAVE_GENERIC_HARDIRQS + select ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE select SPARSE_IRQ select GENERIC_FIND_FIRST_BIT select GENERIC_IRQ_PROBE diff --git a/lib/Kconfig b/lib/Kconfig index 8269d56..72c1d46 100644 --- a/lib/Kconfig +++ b/lib/Kconfig @@ -340,6 +340,9 @@ config NLATTR config GENERIC_ATOMIC64 bool +config ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE + def_bool y if GENERIC_ATOMIC64 + config LRU_CACHE tristate diff --git a/lib/atomic64_test.c b/lib/atomic64_test.c index cb99b91..00bca22 100644 --- a/lib/atomic64_test.c +++ b/lib/atomic64_test.c @@ -114,8 +114,7 @@ static __init int test_atomic64(void) r += one; BUG_ON(v.counter != r); -#if defined(CONFIG_X86) || defined(CONFIG_MIPS) || defined(CONFIG_PPC) || \ - defined(CONFIG_S390) || defined(_ASM_GENERIC_ATOMIC64_H) || defined(CONFIG_ARM) +#ifdef CONFIG_ARCH_HAS_ATOMIC64_DEC_IF_POSITIVE INIT(onestwos); BUG_ON(atomic64_dec_if_positive(&v) != (onestwos - 1)); r -= one; @@ -129,7 +128,7 @@ static __init int test_atomic64(void) BUG_ON(atomic64_dec_if_positive(&v) != (-one - one)); BUG_ON(v.counter != r); #else -#warning Please implement atomic64_dec_if_positive for your architecture, and add it to the IF above +#warning Please implement atomic64_dec_if_positive for your architecture and select the above Kconfig symbol #endif INIT(onestwos); -- cgit v0.10.2 From 4b0681487bf72fdb86b42c93b7d8a607e5a426b1 Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Mon, 30 Jul 2012 14:41:11 -0700 Subject: spinlock_debug: print offset in addition to symbol name If there are two spinlocks embedded in a structure that kallsyms knows about and one of the spinlocks locks up we will print the name of the containing structure instead of the address of the lock. This is quite bad, so let's use %pS instead of %ps so we get an offset in addition to the symbol so we can determine which particular lock is having problems. Signed-off-by: Stephen Boyd Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index e91fbc2..eb10578 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -58,7 +58,7 @@ static void spin_dump(raw_spinlock_t *lock, const char *msg) printk(KERN_EMERG "BUG: spinlock %s on CPU#%d, %s/%d\n", msg, raw_smp_processor_id(), current->comm, task_pid_nr(current)); - printk(KERN_EMERG " lock: %ps, .magic: %08x, .owner: %s/%d, " + printk(KERN_EMERG " lock: %pS, .magic: %08x, .owner: %s/%d, " ".owner_cpu: %d\n", lock, lock->magic, owner ? owner->comm : "", -- cgit v0.10.2 From 4ed940d4c34c21a1a356969a923f2815d608e0bf Mon Sep 17 00:00:00 2001 From: Yasuaki Ishimatsu Date: Mon, 30 Jul 2012 14:41:13 -0700 Subject: firmware_map: make firmware_map_add_early() argument consistent with firmware_map_add_hotplug() There are two ways to create /sys/firmware/memmap/X sysfs: - firmware_map_add_early When the system starts, it is calledd from e820_reserve_resources() - firmware_map_add_hotplug When the memory is hot plugged, it is called from add_memory() But these functions are called without unifying value of end argument as below: - end argument of firmware_map_add_early() : start + size - 1 - end argument of firmware_map_add_hogplug() : start + size The patch unifies them to "start + size". Even if applying the patch, /sys/firmware/memmap/X/end file content does not change. [akpm@linux-foundation.org: clarify comments] Signed-off-by: Yasuaki Ishimatsu Reviewed-by: Dave Hansen Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 4185797..ed858e9 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -944,7 +944,7 @@ void __init e820_reserve_resources(void) for (i = 0; i < e820_saved.nr_map; i++) { struct e820entry *entry = &e820_saved.map[i]; firmware_map_add_early(entry->addr, - entry->addr + entry->size - 1, + entry->addr + entry->size, e820_type_to_string(entry->type)); } } diff --git a/drivers/firmware/memmap.c b/drivers/firmware/memmap.c index adc0710..c1cdc92 100644 --- a/drivers/firmware/memmap.c +++ b/drivers/firmware/memmap.c @@ -98,7 +98,7 @@ static LIST_HEAD(map_entries); /** * firmware_map_add_entry() - Does the real work to add a firmware memmap entry. * @start: Start of the memory range. - * @end: End of the memory range (inclusive). + * @end: End of the memory range (exclusive). * @type: Type of the memory range. * @entry: Pre-allocated (either kmalloc() or bootmem allocator), uninitialised * entry. @@ -113,7 +113,7 @@ static int firmware_map_add_entry(u64 start, u64 end, BUG_ON(start > end); entry->start = start; - entry->end = end; + entry->end = end - 1; entry->type = type; INIT_LIST_HEAD(&entry->list); kobject_init(&entry->kobj, &memmap_ktype); @@ -148,7 +148,7 @@ static int add_sysfs_fw_map_entry(struct firmware_map_entry *entry) * firmware_map_add_hotplug() - Adds a firmware mapping entry when we do * memory hotplug. * @start: Start of the memory range. - * @end: End of the memory range (inclusive). + * @end: End of the memory range (exclusive) * @type: Type of the memory range. * * Adds a firmware mapping entry. This function is for memory hotplug, it is @@ -175,7 +175,7 @@ int __meminit firmware_map_add_hotplug(u64 start, u64 end, const char *type) /** * firmware_map_add_early() - Adds a firmware mapping entry. * @start: Start of the memory range. - * @end: End of the memory range (inclusive). + * @end: End of the memory range. * @type: Type of the memory range. * * Adds a firmware mapping entry. This function uses the bootmem allocator -- cgit v0.10.2 From b34a26f3158cca932b6f1427b97fe12c7838ee68 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:41:16 -0700 Subject: checkpatch: Update alignment check Parenthesis alignment doesn't correctly check an existing line after an inserted or modified line with an open parenthesis. Fix it. Signed-off-by: Joe Perches Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index e5bd60f..630319a 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1848,8 +1848,8 @@ sub process { my $pos = pos_last_openparen($rest); if ($pos >= 0) { - $line =~ /^\+([ \t]*)/; - my $newindent = $1; + $line =~ /^(\+| )([ \t]*)/; + my $newindent = $2; my $goodtabindent = $oldindent . "\t" x ($pos / 8) . -- cgit v0.10.2 From ce0338df3c9a43e709b8a478265b32b9edcc7ccc Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:41:18 -0700 Subject: checkpatch: test for non-standard signatures Warn on non-standard signature styles. Signed-off-by: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 630319a..3e04f80 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -1600,13 +1600,17 @@ sub process { # Check signature styles if (!$in_header_lines && - $line =~ /^(\s*)($signature_tags)(\s*)(.*)/) { + $line =~ /^(\s*)([a-z0-9_-]+by:|$signature_tags)(\s*)(.*)/i) { my $space_before = $1; my $sign_off = $2; my $space_after = $3; my $email = $4; my $ucfirst_sign_off = ucfirst(lc($sign_off)); + if ($sign_off !~ /$signature_tags/) { + WARN("BAD_SIGN_OFF", + "Non-standard signature: $sign_off\n" . $herecurr); + } if (defined $space_before && $space_before ne "") { WARN("BAD_SIGN_OFF", "Do not use whitespace before $ucfirst_sign_off\n" . $herecurr); -- cgit v0.10.2 From 4a273195a551a27a9a3ebed072c8df16c853da7f Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:41:20 -0700 Subject: checkpatch: check usleep_range() arguments usleep_range() shouldn't use the same args for min and max. Report it when it happens and when both args are decimal and min > max. Signed-off-by: Joe Perches Cc: Yuval Mintz Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 3e04f80..4bad570 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -3313,6 +3313,22 @@ sub process { } } +# check usleep_range arguments + if ($^V && $^V ge 5.10.0 && + defined $stat && + $stat =~ /^\+(?:.*?)\busleep_range\s*\(\s*($FuncArg)\s*,\s*($FuncArg)\s*\)/) { + my $min = $1; + my $max = $7; + if ($min eq $max) { + WARN("USLEEP_RANGE", + "usleep_range should not use min == max args; see Documentation/timers/timers-howto.txt\n" . "$here\n$stat\n"); + } elsif ($min =~ /^\d+$/ && $max =~ /^\d+$/ && + $min > $max) { + WARN("USLEEP_RANGE", + "usleep_range args reversed, use min then max; see Documentation/timers/timers-howto.txt\n" . "$here\n$stat\n"); + } + } + # check for new externs in .c files. if ($realfile =~ /\.c$/ && defined $stat && $stat =~ /^.\s*(?:extern\s+)?$Type\s+($Ident)(\s*)\(/s) -- cgit v0.10.2 From 66c80b6077256898df948ac6acf547b47ddb1fcf Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:41:22 -0700 Subject: checkpatch: Add acheck for use of sizeof without parenthesis Kernel style uses parenthesis around sizeof. Signed-off-by: Joe Perches Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 4bad570..40b0627 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -3265,6 +3265,12 @@ sub process { "sizeof(& should be avoided\n" . $herecurr); } +# check for sizeof without parenthesis + if ($line =~ /\bsizeof\s+((?:\*\s*|)$Lval|$Type(?:\s+$Lval|))/) { + WARN("SIZEOF_PARENTHESIS", + "sizeof $1 should be sizeof($1)\n" . $herecurr); + } + # check for line continuations in quoted strings with odd counts of " if ($rawline =~ /\\$/ && $rawline =~ tr/"/"/ % 2) { WARN("LINE_CONTINUATIONS", -- cgit v0.10.2 From b13edf7ff2dd0fef95e981170fa71fa6b60421b0 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:41:24 -0700 Subject: checkpatch: add checks for do {} while (0) macro misuses These types of macros should not be used for either a single statement nor should the macro end with a semi-colon. Add tests for these conditions. Signed-off-by: Joe Perches Cc: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 40b0627..913d6bd 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -2988,6 +2988,45 @@ sub process { } } +# do {} while (0) macro tests: +# single-statement macros do not need to be enclosed in do while (0) loop, +# macro should not end with a semicolon + if ($^V && $^V ge 5.10.0 && + $realfile !~ m@/vmlinux.lds.h$@ && + $line =~ /^.\s*\#\s*define\s+$Ident(\()?/) { + my $ln = $linenr; + my $cnt = $realcnt; + my ($off, $dstat, $dcond, $rest); + my $ctx = ''; + ($dstat, $dcond, $ln, $cnt, $off) = + ctx_statement_block($linenr, $realcnt, 0); + $ctx = $dstat; + + $dstat =~ s/\\\n.//g; + + if ($dstat =~ /^\+\s*#\s*define\s+$Ident\s*${balanced_parens}\s*do\s*{(.*)\s*}\s*while\s*\(\s*0\s*\)\s*([;\s]*)\s*$/) { + my $stmts = $2; + my $semis = $3; + + $ctx =~ s/\n*$//; + my $cnt = statement_rawlines($ctx); + my $herectx = $here . "\n"; + + for (my $n = 0; $n < $cnt; $n++) { + $herectx .= raw_line($linenr, $n) . "\n"; + } + + if (($stmts =~ tr/;/;/) == 1) { + WARN("SINGLE_STATEMENT_DO_WHILE_MACRO", + "Single statement macros should not use a do {} while (0) loop\n" . "$herectx"); + } + if (defined $semis && $semis ne "") { + WARN("DO_WHILE_MACRO_WITH_TRAILING_SEMICOLON", + "do {} while (0) macros should not be semicolon terminated\n" . "$herectx"); + } + } + } + # make sure symbols are always wrapped with VMLINUX_SYMBOL() ... # all assignments may have only one of the following with an assignment: # . -- cgit v0.10.2 From 49ac572b93832210dc1895839692b3a51e5d0e27 Mon Sep 17 00:00:00 2001 From: Thiago Rafael Becker Date: Mon, 30 Jul 2012 14:41:26 -0700 Subject: lib/crc32.c: fix unused variables warnings Variables t4, t5, t6 and t7 are only used when CRC_LE_BITS != 32. Fix the following compilation warnings: lib/crc32.c: In function 'crc32_body': lib/crc32.c:77:55: warning: unused variable 't7' lib/crc32.c:77:41: warning: unused variable 't6' lib/crc32.c:77:27: warning: unused variable 't5' lib/crc32.c:77:13: warning: unused variable 't4' Signed-off-by: Thiago Rafael Becker Cc: "Darrick J. Wong" Cc: Bob Pearson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/crc32.c b/lib/crc32.c index b0d278f..61774b8 100644 --- a/lib/crc32.c +++ b/lib/crc32.c @@ -74,7 +74,9 @@ crc32_body(u32 crc, unsigned char const *buf, size_t len, const u32 (*tab)[256]) size_t i; # endif const u32 *t0=tab[0], *t1=tab[1], *t2=tab[2], *t3=tab[3]; +# if CRC_LE_BITS != 32 const u32 *t4 = tab[4], *t5 = tab[5], *t6 = tab[6], *t7 = tab[7]; +# endif u32 q; /* Align it */ -- cgit v0.10.2 From 40251b8eb46e48c011939a3ddf056fe13a223319 Mon Sep 17 00:00:00 2001 From: Kamil Dudka Date: Mon, 30 Jul 2012 14:41:29 -0700 Subject: drivers/message/i2o/i2o_proc.c: the pointer returned from chtostr() points to an array which is no longer valid ... when being used in the calling function. Although it may work, the behavior is undefined. Detected by cppcheck. Signed-off-by: Kamil Dudka Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/message/i2o/i2o_proc.c b/drivers/message/i2o/i2o_proc.c index 506c36f..8001aa6 100644 --- a/drivers/message/i2o/i2o_proc.c +++ b/drivers/message/i2o/i2o_proc.c @@ -255,9 +255,8 @@ static char *scsi_devices[] = { "Array Controller Device" }; -static char *chtostr(u8 * chars, int n) +static char *chtostr(char *tmp, u8 *chars, int n) { - char tmp[256]; tmp[0] = 0; return strncat(tmp, (char *)chars, n); } @@ -791,6 +790,7 @@ static int i2o_seq_show_ddm_table(struct seq_file *seq, void *v) } *result; i2o_exec_execute_ddm_table ddm_table; + char tmp[28 + 1]; result = kmalloc(sizeof(*result), GFP_KERNEL); if (!result) @@ -826,7 +826,7 @@ static int i2o_seq_show_ddm_table(struct seq_file *seq, void *v) seq_printf(seq, "%-#7x", ddm_table.i2o_vendor_id); seq_printf(seq, "%-#8x", ddm_table.module_id); seq_printf(seq, "%-29s", - chtostr(ddm_table.module_name_version, 28)); + chtostr(tmp, ddm_table.module_name_version, 28)); seq_printf(seq, "%9d ", ddm_table.data_size); seq_printf(seq, "%8d", ddm_table.code_size); @@ -893,6 +893,7 @@ static int i2o_seq_show_drivers_stored(struct seq_file *seq, void *v) i2o_driver_result_table *result; i2o_driver_store_table *dst; + char tmp[28 + 1]; result = kmalloc(sizeof(i2o_driver_result_table), GFP_KERNEL); if (result == NULL) @@ -927,8 +928,9 @@ static int i2o_seq_show_drivers_stored(struct seq_file *seq, void *v) seq_printf(seq, "%-#7x", dst->i2o_vendor_id); seq_printf(seq, "%-#8x", dst->module_id); - seq_printf(seq, "%-29s", chtostr(dst->module_name_version, 28)); - seq_printf(seq, "%-9s", chtostr(dst->date, 8)); + seq_printf(seq, "%-29s", + chtostr(tmp, dst->module_name_version, 28)); + seq_printf(seq, "%-9s", chtostr(tmp, dst->date, 8)); seq_printf(seq, "%8d ", dst->module_size); seq_printf(seq, "%8d ", dst->mpb_size); seq_printf(seq, "0x%04x", dst->module_flags); @@ -1248,6 +1250,7 @@ static int i2o_seq_show_dev_identity(struct seq_file *seq, void *v) // == (allow) 512d bytes (max) static u16 *work16 = (u16 *) work32; int token; + char tmp[16 + 1]; token = i2o_parm_field_get(d, 0xF100, -1, &work32, sizeof(work32)); @@ -1260,13 +1263,13 @@ static int i2o_seq_show_dev_identity(struct seq_file *seq, void *v) seq_printf(seq, "Owner TID : %0#5x\n", work16[2]); seq_printf(seq, "Parent TID : %0#5x\n", work16[3]); seq_printf(seq, "Vendor info : %s\n", - chtostr((u8 *) (work32 + 2), 16)); + chtostr(tmp, (u8 *) (work32 + 2), 16)); seq_printf(seq, "Product info : %s\n", - chtostr((u8 *) (work32 + 6), 16)); + chtostr(tmp, (u8 *) (work32 + 6), 16)); seq_printf(seq, "Description : %s\n", - chtostr((u8 *) (work32 + 10), 16)); + chtostr(tmp, (u8 *) (work32 + 10), 16)); seq_printf(seq, "Product rev. : %s\n", - chtostr((u8 *) (work32 + 14), 8)); + chtostr(tmp, (u8 *) (work32 + 14), 8)); seq_printf(seq, "Serial number : "); print_serial_number(seq, (u8 *) (work32 + 16), @@ -1303,6 +1306,8 @@ static int i2o_seq_show_ddm_identity(struct seq_file *seq, void *v) u8 pad[256]; // allow up to 256 byte (max) serial number } result; + char tmp[24 + 1]; + token = i2o_parm_field_get(d, 0xF101, -1, &result, sizeof(result)); if (token < 0) { @@ -1312,9 +1317,9 @@ static int i2o_seq_show_ddm_identity(struct seq_file *seq, void *v) seq_printf(seq, "Registering DDM TID : 0x%03x\n", result.ddm_tid); seq_printf(seq, "Module name : %s\n", - chtostr(result.module_name, 24)); + chtostr(tmp, result.module_name, 24)); seq_printf(seq, "Module revision : %s\n", - chtostr(result.module_rev, 8)); + chtostr(tmp, result.module_rev, 8)); seq_printf(seq, "Serial number : "); print_serial_number(seq, result.serial_number, sizeof(result) - 36); @@ -1338,6 +1343,8 @@ static int i2o_seq_show_uinfo(struct seq_file *seq, void *v) u8 instance_number[4]; } result; + char tmp[64 + 1]; + token = i2o_parm_field_get(d, 0xF102, -1, &result, sizeof(result)); if (token < 0) { @@ -1346,13 +1353,13 @@ static int i2o_seq_show_uinfo(struct seq_file *seq, void *v) } seq_printf(seq, "Device name : %s\n", - chtostr(result.device_name, 64)); + chtostr(tmp, result.device_name, 64)); seq_printf(seq, "Service name : %s\n", - chtostr(result.service_name, 64)); + chtostr(tmp, result.service_name, 64)); seq_printf(seq, "Physical name : %s\n", - chtostr(result.physical_location, 64)); + chtostr(tmp, result.physical_location, 64)); seq_printf(seq, "Instance number : %s\n", - chtostr(result.instance_number, 4)); + chtostr(tmp, result.instance_number, 4)); return 0; } -- cgit v0.10.2 From 261eba73353edd48b0c0cb7aad59553dfc712ebc Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Jul 2012 14:41:30 -0700 Subject: drivers/message/i2o/i2o_config.c: bound allocation Fix a case where users can try to allocate arbitarily large amounts of memory. 64K is overkill for a config request so apply an upper bound. Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/message/i2o/i2o_config.c b/drivers/message/i2o/i2o_config.c index 098de2b..9a49c24 100644 --- a/drivers/message/i2o/i2o_config.c +++ b/drivers/message/i2o/i2o_config.c @@ -188,6 +188,13 @@ static int i2o_cfg_parms(unsigned long arg, unsigned int type) if (!dev) return -ENXIO; + /* + * Stop users being able to try and allocate arbitary amounts + * of DMA space. 64K is way more than sufficient for this. + */ + if (kcmd.oplen > 65536) + return -EMSGSIZE; + ops = memdup_user(kcmd.opbuf, kcmd.oplen); if (IS_ERR(ops)) return PTR_ERR(ops); -- cgit v0.10.2 From 8384dfeb759acac48ea24b81f134cac3cbb3e6fc Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 30 Jul 2012 14:41:31 -0700 Subject: drivers/rtc/rtc-coh901331.c: use clk_prepare/unprepare Make sure we prepare/unprepare the clock for the COH901331 RTC driver as is required by the clk API especially if you use common clock. Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-coh901331.c b/drivers/rtc/rtc-coh901331.c index a5b8a0c..5866502 100644 --- a/drivers/rtc/rtc-coh901331.c +++ b/drivers/rtc/rtc-coh901331.c @@ -157,6 +157,7 @@ static int __exit coh901331_remove(struct platform_device *pdev) if (rtap) { free_irq(rtap->irq, rtap); rtc_device_unregister(rtap->rtc); + clk_unprepare(rtap->clk); clk_put(rtap->clk); iounmap(rtap->virtbase); release_mem_region(rtap->phybase, rtap->physize); @@ -213,10 +214,10 @@ static int __init coh901331_probe(struct platform_device *pdev) } /* We enable/disable the clock only to assure it works */ - ret = clk_enable(rtap->clk); + ret = clk_prepare_enable(rtap->clk); if (ret) { dev_err(&pdev->dev, "could not enable clock\n"); - goto out_no_clk_enable; + goto out_no_clk_prepenable; } clk_disable(rtap->clk); @@ -232,7 +233,8 @@ static int __init coh901331_probe(struct platform_device *pdev) out_no_rtc: platform_set_drvdata(pdev, NULL); - out_no_clk_enable: + clk_unprepare(rtap->clk); + out_no_clk_prepenable: clk_put(rtap->clk); out_no_clk: free_irq(rtap->irq, rtap); @@ -265,6 +267,7 @@ static int coh901331_suspend(struct platform_device *pdev, pm_message_t state) writel(0, rtap->virtbase + COH901331_IRQ_MASK); clk_disable(rtap->clk); } + clk_unprepare(rtap->clk); return 0; } @@ -272,6 +275,7 @@ static int coh901331_resume(struct platform_device *pdev) { struct coh901331_port *rtap = dev_get_drvdata(&pdev->dev); + clk_prepare(rtap->clk); if (device_may_wakeup(&pdev->dev)) { disable_irq_wake(rtap->irq); } else { @@ -293,6 +297,7 @@ static void coh901331_shutdown(struct platform_device *pdev) clk_enable(rtap->clk); writel(0, rtap->virtbase + COH901331_IRQ_MASK); clk_disable(rtap->clk); + clk_unprepare(rtap->clk); } static struct platform_driver coh901331_driver = { -- cgit v0.10.2 From 36ac1d24f1488c32b85a1718a4edfda615b2ef77 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 30 Jul 2012 14:41:32 -0700 Subject: drivers/rtc/rtc-coh901331.c: use devm allocation Allocate memory, region, remap and irq for device state using devm_* helpers to simplify memory accounting. Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-coh901331.c b/drivers/rtc/rtc-coh901331.c index 5866502..76b2156 100644 --- a/drivers/rtc/rtc-coh901331.c +++ b/drivers/rtc/rtc-coh901331.c @@ -155,14 +155,10 @@ static int __exit coh901331_remove(struct platform_device *pdev) struct coh901331_port *rtap = dev_get_drvdata(&pdev->dev); if (rtap) { - free_irq(rtap->irq, rtap); rtc_device_unregister(rtap->rtc); clk_unprepare(rtap->clk); clk_put(rtap->clk); - iounmap(rtap->virtbase); - release_mem_region(rtap->phybase, rtap->physize); platform_set_drvdata(pdev, NULL); - kfree(rtap); } return 0; @@ -175,42 +171,36 @@ static int __init coh901331_probe(struct platform_device *pdev) struct coh901331_port *rtap; struct resource *res; - rtap = kzalloc(sizeof(struct coh901331_port), GFP_KERNEL); + rtap = devm_kzalloc(&pdev->dev, + sizeof(struct coh901331_port), GFP_KERNEL); if (!rtap) return -ENOMEM; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) { - ret = -ENOENT; - goto out_no_resource; - } + if (!res) + return -ENOENT; + rtap->phybase = res->start; rtap->physize = resource_size(res); - if (request_mem_region(rtap->phybase, rtap->physize, - "rtc-coh901331") == NULL) { - ret = -EBUSY; - goto out_no_memregion; - } + if (devm_request_mem_region(&pdev->dev, rtap->phybase, rtap->physize, + "rtc-coh901331") == NULL) + return -EBUSY; - rtap->virtbase = ioremap(rtap->phybase, rtap->physize); - if (!rtap->virtbase) { - ret = -ENOMEM; - goto out_no_remap; - } + rtap->virtbase = devm_ioremap(&pdev->dev, rtap->phybase, rtap->physize); + if (!rtap->virtbase) + return -ENOMEM; rtap->irq = platform_get_irq(pdev, 0); - if (request_irq(rtap->irq, coh901331_interrupt, 0, - "RTC COH 901 331 Alarm", rtap)) { - ret = -EIO; - goto out_no_irq; - } + if (devm_request_irq(&pdev->dev, rtap->irq, coh901331_interrupt, 0, + "RTC COH 901 331 Alarm", rtap)) + return -EIO; rtap->clk = clk_get(&pdev->dev, NULL); if (IS_ERR(rtap->clk)) { ret = PTR_ERR(rtap->clk); dev_err(&pdev->dev, "could not get clock\n"); - goto out_no_clk; + return ret; } /* We enable/disable the clock only to assure it works */ @@ -236,16 +226,6 @@ static int __init coh901331_probe(struct platform_device *pdev) clk_unprepare(rtap->clk); out_no_clk_prepenable: clk_put(rtap->clk); - out_no_clk: - free_irq(rtap->irq, rtap); - out_no_irq: - iounmap(rtap->virtbase); - out_no_remap: - platform_set_drvdata(pdev, NULL); - out_no_memregion: - release_mem_region(rtap->phybase, SZ_4K); - out_no_resource: - kfree(rtap); return ret; } -- cgit v0.10.2 From aff05ed5d103524bd69bd9d7b621c5c8a6c63198 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 30 Jul 2012 14:41:34 -0700 Subject: rtc: pl031: encapsulate per-vendor ops Move the per-vendor operations for this RTC into a encapsulating struct so we can have more per-vendor variables than just the ops. Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c index cc05339..575fbbf 100644 --- a/drivers/rtc/rtc-pl031.c +++ b/drivers/rtc/rtc-pl031.c @@ -68,7 +68,16 @@ #define RTC_TIMER_FREQ 32768 +/** + * struct pl031_vendor_data - per-vendor variations + * @ops: the vendor-specific operations used on this silicon version + */ +struct pl031_vendor_data { + struct rtc_class_ops ops; +}; + struct pl031_local { + struct pl031_vendor_data *vendor; struct rtc_device *rtc; void __iomem *base; u8 hw_designer; @@ -303,7 +312,8 @@ static int pl031_probe(struct amba_device *adev, const struct amba_id *id) { int ret; struct pl031_local *ldata; - struct rtc_class_ops *ops = id->data; + struct pl031_vendor_data *vendor = id->data; + struct rtc_class_ops *ops = &vendor->ops; unsigned long time; ret = amba_request_regions(adev, NULL); @@ -315,6 +325,7 @@ static int pl031_probe(struct amba_device *adev, const struct amba_id *id) ret = -ENOMEM; goto out; } + ldata->vendor = vendor; ldata->base = ioremap(adev->res.start, resource_size(&adev->res)); @@ -383,48 +394,54 @@ err_req: } /* Operations for the original ARM version */ -static struct rtc_class_ops arm_pl031_ops = { - .read_time = pl031_read_time, - .set_time = pl031_set_time, - .read_alarm = pl031_read_alarm, - .set_alarm = pl031_set_alarm, - .alarm_irq_enable = pl031_alarm_irq_enable, +static struct pl031_vendor_data arm_pl031 = { + .ops = { + .read_time = pl031_read_time, + .set_time = pl031_set_time, + .read_alarm = pl031_read_alarm, + .set_alarm = pl031_set_alarm, + .alarm_irq_enable = pl031_alarm_irq_enable, + }, }; /* The First ST derivative */ -static struct rtc_class_ops stv1_pl031_ops = { - .read_time = pl031_read_time, - .set_time = pl031_set_time, - .read_alarm = pl031_read_alarm, - .set_alarm = pl031_set_alarm, - .alarm_irq_enable = pl031_alarm_irq_enable, +static struct pl031_vendor_data stv1_pl031 = { + .ops = { + .read_time = pl031_read_time, + .set_time = pl031_set_time, + .read_alarm = pl031_read_alarm, + .set_alarm = pl031_set_alarm, + .alarm_irq_enable = pl031_alarm_irq_enable, + }, }; /* And the second ST derivative */ -static struct rtc_class_ops stv2_pl031_ops = { - .read_time = pl031_stv2_read_time, - .set_time = pl031_stv2_set_time, - .read_alarm = pl031_stv2_read_alarm, - .set_alarm = pl031_stv2_set_alarm, - .alarm_irq_enable = pl031_alarm_irq_enable, +static struct pl031_vendor_data stv2_pl031 = { + .ops = { + .read_time = pl031_stv2_read_time, + .set_time = pl031_stv2_set_time, + .read_alarm = pl031_stv2_read_alarm, + .set_alarm = pl031_stv2_set_alarm, + .alarm_irq_enable = pl031_alarm_irq_enable, + }, }; static struct amba_id pl031_ids[] = { { .id = 0x00041031, .mask = 0x000fffff, - .data = &arm_pl031_ops, + .data = &arm_pl031, }, /* ST Micro variants */ { .id = 0x00180031, .mask = 0x00ffffff, - .data = &stv1_pl031_ops, + .data = &stv1_pl031, }, { .id = 0x00280031, .mask = 0x00ffffff, - .data = &stv2_pl031_ops, + .data = &stv2_pl031, }, {0, 0}, }; -- cgit v0.10.2 From 1bb457fc9268bb75156af8db78b72a4463baaec4 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Mon, 30 Jul 2012 14:41:36 -0700 Subject: rtc: pl031: use per-vendor variables for special init Instead of hard-checking for certain vendor codes, follow the pattern of other AMBA (PrimeCell) drivers and use variables in the vendor data. Get rid of the locally cached vendor and hardware revision since we already have the nice vendor data variable in the state. Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c index 575fbbf..e66afb8 100644 --- a/drivers/rtc/rtc-pl031.c +++ b/drivers/rtc/rtc-pl031.c @@ -71,17 +71,21 @@ /** * struct pl031_vendor_data - per-vendor variations * @ops: the vendor-specific operations used on this silicon version + * @clockwatch: if this is an ST Microelectronics silicon version with a + * clockwatch function + * @st_weekday: if this is an ST Microelectronics silicon version that need + * the weekday fix */ struct pl031_vendor_data { struct rtc_class_ops ops; + bool clockwatch; + bool st_weekday; }; struct pl031_local { struct pl031_vendor_data *vendor; struct rtc_device *rtc; void __iomem *base; - u8 hw_designer; - u8 hw_revision:4; }; static int pl031_alarm_irq_enable(struct device *dev, @@ -336,14 +340,11 @@ static int pl031_probe(struct amba_device *adev, const struct amba_id *id) amba_set_drvdata(adev, ldata); - ldata->hw_designer = amba_manf(adev); - ldata->hw_revision = amba_rev(adev); - - dev_dbg(&adev->dev, "designer ID = 0x%02x\n", ldata->hw_designer); - dev_dbg(&adev->dev, "revision = 0x%01x\n", ldata->hw_revision); + dev_dbg(&adev->dev, "designer ID = 0x%02x\n", amba_manf(adev)); + dev_dbg(&adev->dev, "revision = 0x%01x\n", amba_rev(adev)); /* Enable the clockwatch on ST Variants */ - if (ldata->hw_designer == AMBA_VENDOR_ST) + if (vendor->clockwatch) writel(readl(ldata->base + RTC_CR) | RTC_CR_CWEN, ldata->base + RTC_CR); @@ -351,7 +352,7 @@ static int pl031_probe(struct amba_device *adev, const struct amba_id *id) * On ST PL031 variants, the RTC reset value does not provide correct * weekday for 2000-01-01. Correct the erroneous sunday to saturday. */ - if (ldata->hw_designer == AMBA_VENDOR_ST) { + if (vendor->st_weekday) { if (readl(ldata->base + RTC_YDR) == 0x2000) { time = readl(ldata->base + RTC_DR); if ((time & @@ -413,6 +414,8 @@ static struct pl031_vendor_data stv1_pl031 = { .set_alarm = pl031_set_alarm, .alarm_irq_enable = pl031_alarm_irq_enable, }, + .clockwatch = true, + .st_weekday = true, }; /* And the second ST derivative */ @@ -424,6 +427,8 @@ static struct pl031_vendor_data stv2_pl031 = { .set_alarm = pl031_stv2_set_alarm, .alarm_irq_enable = pl031_alarm_irq_enable, }, + .clockwatch = true, + .st_weekday = true, }; static struct amba_id pl031_ids[] = { -- cgit v0.10.2 From 559a6fc0508392a665a26d93db9ff875bfdc6540 Mon Sep 17 00:00:00 2001 From: Mattias Wallin Date: Mon, 30 Jul 2012 14:41:39 -0700 Subject: rtc: pl031: fix up IRQ flags The pl031 interrupt is shared between the timer part and the clockwatch part of the same HW block on the ux500, so mark it IRQF_SHARED on this variant. This patch also adds the IRQF_NO_SUSPEND flag to the rtc irq on all variants as we don't want this pretty important IRQ to be disabled in suspend. Signed-off-by: Mattias Wallin Signed-off-by: Linus Walleij Reviewed-by: Rickard Andersson Reviewed-by: Jonas Aberg Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c index e66afb8..08378e3 100644 --- a/drivers/rtc/rtc-pl031.c +++ b/drivers/rtc/rtc-pl031.c @@ -75,11 +75,13 @@ * clockwatch function * @st_weekday: if this is an ST Microelectronics silicon version that need * the weekday fix + * @irqflags: special IRQ flags per variant */ struct pl031_vendor_data { struct rtc_class_ops ops; bool clockwatch; bool st_weekday; + unsigned long irqflags; }; struct pl031_local { @@ -373,7 +375,7 @@ static int pl031_probe(struct amba_device *adev, const struct amba_id *id) } if (request_irq(adev->irq[0], pl031_interrupt, - 0, "rtc-pl031", ldata)) { + vendor->irqflags, "rtc-pl031", ldata)) { ret = -EIO; goto out_no_irq; } @@ -403,6 +405,7 @@ static struct pl031_vendor_data arm_pl031 = { .set_alarm = pl031_set_alarm, .alarm_irq_enable = pl031_alarm_irq_enable, }, + .irqflags = IRQF_NO_SUSPEND, }; /* The First ST derivative */ @@ -416,6 +419,7 @@ static struct pl031_vendor_data stv1_pl031 = { }, .clockwatch = true, .st_weekday = true, + .irqflags = IRQF_NO_SUSPEND, }; /* And the second ST derivative */ @@ -429,6 +433,11 @@ static struct pl031_vendor_data stv2_pl031 = { }, .clockwatch = true, .st_weekday = true, + /* + * This variant shares the IRQ with another block and must not + * suspend that IRQ line. + */ + .irqflags = IRQF_SHARED | IRQF_NO_SUSPEND, }; static struct amba_id pl031_ids[] = { -- cgit v0.10.2 From dc43d4a26b808240410ce3ea1c72a8047f90d1d8 Mon Sep 17 00:00:00 2001 From: Ramesh Chandrasekaran Date: Mon, 30 Jul 2012 14:41:41 -0700 Subject: drivers/rtc/rtc-ab8500.c: use UIE emulation RTC: Fix to correct improper implementation of clock update irq (RTC_UIE) and enable UIE Emulation. [akpm@linux-foundation.org: checkpatch fixes] Signed-off-by: Ramesh Chandrasekaran Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig index 08cbdb9..668da59 100644 --- a/drivers/rtc/Kconfig +++ b/drivers/rtc/Kconfig @@ -694,6 +694,7 @@ config RTC_DRV_AB3100 config RTC_DRV_AB8500 tristate "ST-Ericsson AB8500 RTC" depends on AB8500_CORE + select RTC_INTF_DEV_UIE_EMUL help Select this to enable the ST-Ericsson AB8500 power management IC RTC support. This chip contains a battery- and capacitor-backed RTC. diff --git a/drivers/rtc/rtc-ab8500.c b/drivers/rtc/rtc-ab8500.c index 370889d..1a57e03 100644 --- a/drivers/rtc/rtc-ab8500.c +++ b/drivers/rtc/rtc-ab8500.c @@ -225,7 +225,8 @@ static int ab8500_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) { int retval, i; unsigned char buf[ARRAY_SIZE(ab8500_rtc_alarm_regs)]; - unsigned long mins, secs = 0; + unsigned long mins, secs = 0, cursec = 0; + struct rtc_time curtm; if (alarm->time.tm_year < (AB8500_RTC_EPOCH - 1900)) { dev_dbg(dev, "year should be equal to or greater than %d\n", @@ -237,6 +238,18 @@ static int ab8500_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alarm) rtc_tm_to_time(&alarm->time, &secs); /* + * Check whether alarm is set less than 1min. + * Since our RTC doesn't support alarm resolution less than 1min, + * return -EINVAL, so UIE EMUL can take it up, incase of UIE_ON + */ + ab8500_rtc_read_time(dev, &curtm); /* Read current time */ + rtc_tm_to_time(&curtm, &cursec); + if ((secs - cursec) < 59) { + dev_dbg(dev, "Alarm less than 1 minute not supported\r\n"); + return -EINVAL; + } + + /* * Convert it to the number of seconds since 01-01-2000 00:00:00, since * we only have a small counter in the RTC. */ -- cgit v0.10.2 From 064407f1c0d54ad215927b4c14b3f4e68fa486f4 Mon Sep 17 00:00:00 2001 From: Bengt Jonsson Date: Mon, 30 Jul 2012 14:41:43 -0700 Subject: drivers/rtc/rtc-ab8500.c: remove fix for AB8500 ED version AB8500 ED (Early Drop) is no longer supported by the kernel. Signed-off-by: Bengt Jonsson Signed-off-by: Linus Walleij Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-ab8500.c b/drivers/rtc/rtc-ab8500.c index 1a57e03..bf3c2f6 100644 --- a/drivers/rtc/rtc-ab8500.c +++ b/drivers/rtc/rtc-ab8500.c @@ -89,22 +89,17 @@ static int ab8500_rtc_read_time(struct device *dev, struct rtc_time *tm) if (retval < 0) return retval; - /* Early AB8500 chips will not clear the rtc read request bit */ - if (abx500_get_chip_id(dev) == 0) { - usleep_range(1000, 1000); - } else { - /* Wait for some cycles after enabling the rtc read in ab8500 */ - while (time_before(jiffies, timeout)) { - retval = abx500_get_register_interruptible(dev, - AB8500_RTC, AB8500_RTC_READ_REQ_REG, &value); - if (retval < 0) - return retval; - - if (!(value & RTC_READ_REQUEST)) - break; - - usleep_range(1000, 5000); - } + /* Wait for some cycles after enabling the rtc read in ab8500 */ + while (time_before(jiffies, timeout)) { + retval = abx500_get_register_interruptible(dev, + AB8500_RTC, AB8500_RTC_READ_REQ_REG, &value); + if (retval < 0) + return retval; + + if (!(value & RTC_READ_REQUEST)) + break; + + usleep_range(1000, 5000); } /* Read the Watchtime registers */ -- cgit v0.10.2 From aa11ec593e2c68d57728d2ebb2da265468e410c9 Mon Sep 17 00:00:00 2001 From: Devendra Naga Date: Mon, 30 Jul 2012 14:41:45 -0700 Subject: drivers/rtc/rtc-r9701.c: avoid second call to rtc_valid_tm() r9701_get_datetime() calls rtc_valid_tm() and returns the value returned by rtc_valid_tm(), which can be used in the `if', so calling rtc_valid_tm() a second time is not required. Signed-off-by: Devendra Naga Cc: Alessandro Zummo Cc: Anatolij Gustschin Cc: Andreas Dumberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-r9701.c b/drivers/rtc/rtc-r9701.c index 33b6ba0..e6c34c0 100644 --- a/drivers/rtc/rtc-r9701.c +++ b/drivers/rtc/rtc-r9701.c @@ -138,8 +138,7 @@ static int __devinit r9701_probe(struct spi_device *spi) * contain invalid values. If so, try to write a default date: * 2000/1/1 00:00:00 */ - r9701_get_datetime(&spi->dev, &dt); - if (rtc_valid_tm(&dt)) { + if (r9701_get_datetime(&spi->dev, &dt)) { dev_info(&spi->dev, "trying to repair invalid date/time\n"); dt.tm_sec = 0; dt.tm_min = 0; -- cgit v0.10.2 From 8e74b6edc9b7befd9f7a61b9271384eeb1eb8251 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:41:47 -0700 Subject: drivers/rtc/rtc-r9701.c: check that r9701_set_datetime() succeeded When the driver detects that the clock time is invalid, it attempts to write a sane time into the hardware. We curently assume that everything is OK if those writes succeeded. But it is better to re-read the time from the hardware to ensure that the new settings got there OK. Cc: Devendra Naga Cc: Alessandro Zummo Cc: Anatolij Gustschin Cc: Andreas Dumberger Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-r9701.c b/drivers/rtc/rtc-r9701.c index e6c34c0..2c183eb 100644 --- a/drivers/rtc/rtc-r9701.c +++ b/drivers/rtc/rtc-r9701.c @@ -147,7 +147,8 @@ static int __devinit r9701_probe(struct spi_device *spi) dt.tm_mon = 0; dt.tm_year = 100; - if (r9701_set_datetime(&spi->dev, &dt)) { + if (r9701_set_datetime(&spi->dev, &dt) || + r9701_get_datetime(&spi->dev, &dt)) { dev_err(&spi->dev, "cannot repair RTC register\n"); return -ENODEV; } -- cgit v0.10.2 From dbd9acbef2f178e74c2f2def23eb4e8c2bd14270 Mon Sep 17 00:00:00 2001 From: Sachin Kamat Date: Mon, 30 Jul 2012 14:41:48 -0700 Subject: drivers/rtc/rtc-s3c.c: replace #include header files from asm/* to linux/* Fixes the following checkpatch warnings: WARNING: Use #include instead of WARNING: Use #include instead of Signed-off-by: Sachin Kamat Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c index 7e6af0b..bfbd92c 100644 --- a/drivers/rtc/rtc-s3c.c +++ b/drivers/rtc/rtc-s3c.c @@ -26,10 +26,10 @@ #include #include #include +#include +#include #include -#include -#include #include #include -- cgit v0.10.2 From 0f636fc16f80fe64261ce39440e49a259ff2b0da Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 30 Jul 2012 14:41:50 -0700 Subject: rtc/mc13xxx: use MODULE_DEVICE_TABLE instead of MODULE_ALIAS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This allows automatic driver loading for all supported device types. Signed-off-by: Uwe Kleine-König Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-mc13xxx.c b/drivers/rtc/rtc-mc13xxx.c index 546f685..0ce0602 100644 --- a/drivers/rtc/rtc-mc13xxx.c +++ b/drivers/rtc/rtc-mc13xxx.c @@ -405,8 +405,9 @@ static const struct platform_device_id mc13xxx_rtc_idtable[] = { }, { .name = "mc13892-rtc", }, - { } + { /* sentinel */ } }; +MODULE_DEVICE_TABLE(platform, mc13xxx_rtc_idtable); static struct platform_driver mc13xxx_rtc_driver = { .id_table = mc13xxx_rtc_idtable, @@ -432,4 +433,3 @@ module_exit(mc13xxx_rtc_exit); MODULE_AUTHOR("Sascha Hauer "); MODULE_DESCRIPTION("RTC driver for Freescale MC13XXX PMIC"); MODULE_LICENSE("GPL v2"); -MODULE_ALIAS("platform:" DRIVER_NAME); -- cgit v0.10.2 From 19a1ac505d5f3545ebc5b0a195d65cef4da94bb5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Mon, 30 Jul 2012 14:41:52 -0700 Subject: rtc/mc13xxx: add support for the rtc in the mc34708 pmic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Uwe Kleine-König Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-mc13xxx.c b/drivers/rtc/rtc-mc13xxx.c index 0ce0602..2643d88 100644 --- a/drivers/rtc/rtc-mc13xxx.c +++ b/drivers/rtc/rtc-mc13xxx.c @@ -404,6 +404,8 @@ static const struct platform_device_id mc13xxx_rtc_idtable[] = { .name = "mc13783-rtc", }, { .name = "mc13892-rtc", + }, { + .name = "mc34708-rtc", }, { /* sentinel */ } }; -- cgit v0.10.2 From 007def046711479f3d19bec4016b65fd73deed6c Mon Sep 17 00:00:00 2001 From: Devendra Naga Date: Mon, 30 Jul 2012 14:41:54 -0700 Subject: rtc/rtc-da9052: remove unneed devm_kfree call Freeing will trigger when driver unloads, so using devm_kfree() is not needed. Signed-off-by: Devendra Naga Cc: Alessandro Zummo Cc: Ashish Jangam Cc: David Dajun Chen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-da9052.c b/drivers/rtc/rtc-da9052.c index da6ab52..7807025 100644 --- a/drivers/rtc/rtc-da9052.c +++ b/drivers/rtc/rtc-da9052.c @@ -245,7 +245,7 @@ static int __devinit da9052_rtc_probe(struct platform_device *pdev) "ALM", rtc); if (ret != 0) { rtc_err(rtc->da9052, "irq registration failed: %d\n", ret); - goto err_mem; + return ret; } rtc->rtc = rtc_device_register(pdev->name, &pdev->dev, @@ -259,8 +259,6 @@ static int __devinit da9052_rtc_probe(struct platform_device *pdev) err_free_irq: free_irq(rtc->irq, rtc); -err_mem: - devm_kfree(&pdev->dev, rtc); return ret; } @@ -271,7 +269,6 @@ static int __devexit da9052_rtc_remove(struct platform_device *pdev) rtc_device_unregister(rtc->rtc); free_irq(rtc->irq, rtc); platform_set_drvdata(pdev, NULL); - devm_kfree(&pdev->dev, rtc); return 0; } -- cgit v0.10.2 From 0a25bf40d8ea9aeb8412666a7bc8ab95b4cddcb1 Mon Sep 17 00:00:00 2001 From: Nick Bowler Date: Mon, 30 Jul 2012 14:41:57 -0700 Subject: drivers/rtc/rtc-pcf8563.c: set owner field in driver struct The owner member is supposed to be set to the module implementing the device driver, i.e., THIS_MODULE. This enables the appropriate module link in sysfs. Signed-off-by: Nick Bowler Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c index 97a3284..24a9d6a 100644 --- a/drivers/rtc/rtc-pcf8563.c +++ b/drivers/rtc/rtc-pcf8563.c @@ -288,6 +288,7 @@ MODULE_DEVICE_TABLE(i2c, pcf8563_id); static struct i2c_driver pcf8563_driver = { .driver = { .name = "rtc-pcf8563", + .owner = THIS_MODULE, }, .probe = pcf8563_probe, .remove = pcf8563_remove, -- cgit v0.10.2 From 8dccaf0607b3e047680952d0baf4da0486fbbcc1 Mon Sep 17 00:00:00 2001 From: Nick Bowler Date: Mon, 30 Jul 2012 14:41:59 -0700 Subject: drivers/rtc/rtc-pcf8563.c: add device tree support Set the of_match_table for this driver so that devices can be described in the device tree. Signed-off-by: Nick Bowler Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c index 24a9d6a..c2fe426 100644 --- a/drivers/rtc/rtc-pcf8563.c +++ b/drivers/rtc/rtc-pcf8563.c @@ -19,6 +19,7 @@ #include #include #include +#include #define DRV_VERSION "0.4.3" @@ -285,10 +286,19 @@ static const struct i2c_device_id pcf8563_id[] = { }; MODULE_DEVICE_TABLE(i2c, pcf8563_id); +#ifdef CONFIG_OF +static const struct of_device_id pcf8563_of_match[] __devinitconst = { + { .compatible = "nxp,pcf8563" }, + {} +}; +MODULE_DEVICE_TABLE(of, pcf8563_of_match); +#endif + static struct i2c_driver pcf8563_driver = { .driver = { .name = "rtc-pcf8563", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(pcf8563_of_match), }, .probe = pcf8563_probe, .remove = pcf8563_remove, -- cgit v0.10.2 From 6ed6a722f9abac25b0549e7507a2b745ede4475c Mon Sep 17 00:00:00 2001 From: Vladimir Serbinenko Date: Mon, 30 Jul 2012 14:42:00 -0700 Subject: minixfs: fix block limit check On minix2 and minix3 usually max_size is 7fffffff and the check in question prohibits creation of last block spanning right before 7fffffff, due to downward rounding during the division. Fix it by using multiplication instead. [akpm@linux-foundation.org: fix up code layout, use local `sb'] Signed-off-by: Vladimir Serbinenko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/minix/itree_v2.c b/fs/minix/itree_v2.c index 13487ad..78e2d93 100644 --- a/fs/minix/itree_v2.c +++ b/fs/minix/itree_v2.c @@ -32,7 +32,8 @@ static int block_to_path(struct inode * inode, long block, int offsets[DEPTH]) if (block < 0) { printk("MINIX-fs: block_to_path: block %ld < 0 on dev %s\n", block, bdevname(sb->s_bdev, b)); - } else if (block >= (minix_sb(inode->i_sb)->s_max_size/sb->s_blocksize)) { + } else if ((u64)block * (u64)sb->s_blocksize >= + minix_sb(sb)->s_max_size) { if (printk_ratelimit()) printk("MINIX-fs: block_to_path: " "block %ld too big on dev %s\n", -- cgit v0.10.2 From 6b0f3393e38584ec22bab62fe01df58ae5a73ee7 Mon Sep 17 00:00:00 2001 From: Vyacheslav Dubeyko Date: Mon, 30 Jul 2012 14:42:02 -0700 Subject: nilfs2: add omitted comment for ns_mount_state field of the_nilfs structure Add omitted comment for ns_mount_state field of the_nilfs structure. Signed-off-by: Vyacheslav Dubeyko Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index 9992b11..ef40a51 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -52,6 +52,7 @@ enum { * @ns_sbwtime: previous write time of super block * @ns_sbwcount: write count of super block * @ns_sbsize: size of valid data in super block + * @ns_mount_state: file system state * @ns_seg_seq: segment sequence counter * @ns_segnum: index number of the latest full segment. * @ns_nextnum: index number of the full segment index to be used next -- cgit v0.10.2 From 278038ac53c6c4f53d1d34f978beb9aba1410b2c Mon Sep 17 00:00:00 2001 From: Fernando Luis Vazquez Cao Date: Mon, 30 Jul 2012 14:42:03 -0700 Subject: nilfs2: remove references to long gone super operations ->delete_inode(), ->write_super_lockfs(), ->unlockfs() are gone so remove references to them in the NTFS code. Noticed while cleaning up the fsfreeze mess. Signed-off-by: Fernando Luis Vazquez Cao Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index d57c42f9..a76d6ea 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -677,7 +677,6 @@ static const struct super_operations nilfs_sops = { .destroy_inode = nilfs_destroy_inode, .dirty_inode = nilfs_dirty_inode, /* .write_inode = nilfs_write_inode, */ - /* .put_inode = nilfs_put_inode, */ /* .drop_inode = nilfs_drop_inode, */ .evict_inode = nilfs_evict_inode, .put_super = nilfs_put_super, @@ -685,8 +684,6 @@ static const struct super_operations nilfs_sops = { .sync_fs = nilfs_sync_fs, .freeze_fs = nilfs_freeze, .unfreeze_fs = nilfs_unfreeze, - /* .write_super_lockfs */ - /* .unlockfs */ .statfs = nilfs_statfs, .remount_fs = nilfs_remount, /* .umount_begin */ -- cgit v0.10.2 From fe0627e7b3d32a41c16fac6e0af091991545865e Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Mon, 30 Jul 2012 14:42:05 -0700 Subject: nilfs2: fix timing issue between rmcp and chcp ioctls The checkpoint deletion ioctl (rmcp ioctl) has potential for breaking snapshot because it is not fully exclusive with checkpoint mode change ioctl (chcp ioctl). The rmcp ioctl first tests if the specified checkpoint is a snapshot or not within nilfs_cpfile_delete_checkpoint function, and then calls nilfs_cpfile_delete_checkpoints function to actually invalidate the checkpoint only if it's not a snapshot. However, the checkpoint can be changed into a snapshot by the chcp ioctl between these two operations. In that case, calling nilfs_cpfile_delete_checkpoints() wrongly invalidates the snapshot, which leads to snapshot list corruption and snapshot count mismatch. This fixes the issue by changing nilfs_cpfile_delete_checkpoints() so that it reconfirms the target checkpoints are snapshot or not. This second check is exclusive with the chcp operation since it is protected by an existing semaphore. Signed-off-by: Ryusuke Konishi Cc: Fernando Luis Vazquez Cao Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c index dab5c4c..deaa3d3 100644 --- a/fs/nilfs2/cpfile.c +++ b/fs/nilfs2/cpfile.c @@ -286,7 +286,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, __u64 cno; void *kaddr; unsigned long tnicps; - int ret, ncps, nicps, count, i; + int ret, ncps, nicps, nss, count, i; if (unlikely(start == 0 || start > end)) { printk(KERN_ERR "%s: invalid range of checkpoint numbers: " @@ -301,6 +301,7 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, if (ret < 0) goto out_sem; tnicps = 0; + nss = 0; for (cno = start; cno < end; cno += ncps) { ncps = nilfs_cpfile_checkpoints_in_block(cpfile, cno, end); @@ -318,8 +319,9 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, cpfile, cno, cp_bh, kaddr); nicps = 0; for (i = 0; i < ncps; i++, cp = (void *)cp + cpsz) { - WARN_ON(nilfs_checkpoint_snapshot(cp)); - if (!nilfs_checkpoint_invalid(cp)) { + if (nilfs_checkpoint_snapshot(cp)) { + nss++; + } else if (!nilfs_checkpoint_invalid(cp)) { nilfs_checkpoint_set_invalid(cp); nicps++; } @@ -364,6 +366,8 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, } brelse(header_bh); + if (nss > 0) + ret = -EBUSY; out_sem: up_write(&NILFS_MDT(cpfile)->mi_sem); -- cgit v0.10.2 From 572d8b3945a31bee7c40d21556803e4807fd9141 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Mon, 30 Jul 2012 14:42:07 -0700 Subject: nilfs2: fix deadlock issue between chcp and thaw ioctls An fs-thaw ioctl causes deadlock with a chcp or mkcp -s command: chcp D ffff88013870f3d0 0 1325 1324 0x00000004 ... Call Trace: nilfs_transaction_begin+0x11c/0x1a0 [nilfs2] wake_up_bit+0x20/0x20 copy_from_user+0x18/0x30 [nilfs2] nilfs_ioctl_change_cpmode+0x7d/0xcf [nilfs2] nilfs_ioctl+0x252/0x61a [nilfs2] do_page_fault+0x311/0x34c get_unmapped_area+0x132/0x14e do_vfs_ioctl+0x44b/0x490 __set_task_blocked+0x5a/0x61 vm_mmap_pgoff+0x76/0x87 __set_current_blocked+0x30/0x4a sys_ioctl+0x4b/0x6f system_call_fastpath+0x16/0x1b thaw D ffff88013870d890 0 1352 1351 0x00000004 ... Call Trace: rwsem_down_failed_common+0xdb/0x10f call_rwsem_down_write_failed+0x13/0x20 down_write+0x25/0x27 thaw_super+0x13/0x9e do_vfs_ioctl+0x1f5/0x490 vm_mmap_pgoff+0x76/0x87 sys_ioctl+0x4b/0x6f filp_close+0x64/0x6c system_call_fastpath+0x16/0x1b where the thaw ioctl deadlocked at thaw_super() when called while chcp was waiting at nilfs_transaction_begin() called from nilfs_ioctl_change_cpmode(). This deadlock is 100% reproducible. This is because nilfs_ioctl_change_cpmode() first locks sb->s_umount in read mode and then waits for unfreezing in nilfs_transaction_begin(), whereas thaw_super() locks sb->s_umount in write mode. The locking of sb->s_umount here was intended to make snapshot mounts and the downgrade of snapshots to checkpoints exclusive. This fixes the deadlock issue by replacing the sb->s_umount usage in nilfs_ioctl_change_cpmode() with a dedicated mutex which protects snapshot mounts. Signed-off-by: Ryusuke Konishi Cc: Fernando Luis Vazquez Cao Tested-by: Ryusuke Konishi Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 06658ca..0b6387c 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -182,7 +182,7 @@ static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp, if (copy_from_user(&cpmode, argp, sizeof(cpmode))) goto out; - down_read(&inode->i_sb->s_umount); + mutex_lock(&nilfs->ns_snapshot_mount_mutex); nilfs_transaction_begin(inode->i_sb, &ti, 0); ret = nilfs_cpfile_change_cpmode( @@ -192,7 +192,7 @@ static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp, else nilfs_transaction_commit(inode->i_sb); /* never fails */ - up_read(&inode->i_sb->s_umount); + mutex_unlock(&nilfs->ns_snapshot_mount_mutex); out: mnt_drop_write_file(filp); return ret; diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index a76d6ea..6522cac 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -945,6 +945,8 @@ static int nilfs_attach_snapshot(struct super_block *s, __u64 cno, struct nilfs_root *root; int ret; + mutex_lock(&nilfs->ns_snapshot_mount_mutex); + down_read(&nilfs->ns_segctor_sem); ret = nilfs_cpfile_is_snapshot(nilfs->ns_cpfile, cno); up_read(&nilfs->ns_segctor_sem); @@ -969,6 +971,7 @@ static int nilfs_attach_snapshot(struct super_block *s, __u64 cno, ret = nilfs_get_root_dentry(s, root, root_dentry); nilfs_put_root(root); out: + mutex_unlock(&nilfs->ns_snapshot_mount_mutex); return ret; } diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index 501b7f8..41e6a04 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -76,6 +76,7 @@ struct the_nilfs *alloc_nilfs(struct block_device *bdev) nilfs->ns_bdev = bdev; atomic_set(&nilfs->ns_ndirtyblks, 0); init_rwsem(&nilfs->ns_sem); + mutex_init(&nilfs->ns_snapshot_mount_mutex); INIT_LIST_HEAD(&nilfs->ns_dirty_files); INIT_LIST_HEAD(&nilfs->ns_gc_inodes); spin_lock_init(&nilfs->ns_inode_lock); diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index ef40a51..2558f32 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -47,6 +47,7 @@ enum { * @ns_flags: flags * @ns_bdev: block device * @ns_sem: semaphore for shared states + * @ns_snapshot_mount_mutex: mutex to protect snapshot mounts * @ns_sbh: buffer heads of on-disk super blocks * @ns_sbp: pointers to super block data * @ns_sbwtime: previous write time of super block @@ -100,6 +101,7 @@ struct the_nilfs { struct block_device *ns_bdev; struct rw_semaphore ns_sem; + struct mutex ns_snapshot_mount_mutex; /* * used for -- cgit v0.10.2 From 8c74ac0557b4cabc9017c2740c5f62b330192416 Mon Sep 17 00:00:00 2001 From: Vyacheslav Dubeyko Date: Mon, 30 Jul 2012 14:42:09 -0700 Subject: nilfs2: add omitted comments for structures in nilfs2_fs.h Add omitted comments for structures in nilfs2_fs.h. Signed-off-by: Vyacheslav Dubeyko Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/nilfs2_fs.h b/include/linux/nilfs2_fs.h index 89bd4a4..9875576 100644 --- a/include/linux/nilfs2_fs.h +++ b/include/linux/nilfs2_fs.h @@ -293,7 +293,7 @@ struct nilfs_dir_entry { __le64 inode; /* Inode number */ __le16 rec_len; /* Directory entry length */ __u8 name_len; /* Name length */ - __u8 file_type; + __u8 file_type; /* Dir entry type (file, dir, etc) */ char name[NILFS_NAME_LEN]; /* File name */ char pad; }; @@ -395,7 +395,7 @@ union nilfs_binfo { }; /** - * struct nilfs_segment_summary - segment summary + * struct nilfs_segment_summary - segment summary header * @ss_datasum: checksum of data * @ss_sumsum: checksum of segment summary * @ss_magic: magic number @@ -683,9 +683,9 @@ struct nilfs_sufile_header { /** * nilfs_suinfo - segment usage information - * @sui_lastmod: - * @sui_nblocks: - * @sui_flags: + * @sui_lastmod: timestamp of last modification + * @sui_nblocks: number of written blocks in segment + * @sui_flags: segment usage flags */ struct nilfs_suinfo { __u64 sui_lastmod; @@ -716,9 +716,10 @@ enum { }; /** - * struct nilfs_cpmode - - * @cc_cno: - * @cc_mode: + * struct nilfs_cpmode - change checkpoint mode structure + * @cm_cno: checkpoint number + * @cm_mode: mode of checkpoint + * @cm_pad: padding */ struct nilfs_cpmode { __u64 cm_cno; @@ -728,11 +729,11 @@ struct nilfs_cpmode { /** * struct nilfs_argv - argument vector - * @v_base: - * @v_nmembs: - * @v_size: - * @v_flags: - * @v_index: + * @v_base: pointer on data array from userspace + * @v_nmembs: number of members in data array + * @v_size: size of data array in bytes + * @v_flags: flags + * @v_index: start number of target data items */ struct nilfs_argv { __u64 v_base; @@ -743,9 +744,9 @@ struct nilfs_argv { }; /** - * struct nilfs_period - - * @p_start: - * @p_end: + * struct nilfs_period - period of checkpoint numbers + * @p_start: start checkpoint number (inclusive) + * @p_end: end checkpoint number (exclusive) */ struct nilfs_period { __u64 p_start; @@ -753,7 +754,7 @@ struct nilfs_period { }; /** - * struct nilfs_cpstat - + * struct nilfs_cpstat - checkpoint statistics * @cs_cno: checkpoint number * @cs_ncps: number of checkpoints * @cs_nsss: number of snapshots @@ -765,7 +766,7 @@ struct nilfs_cpstat { }; /** - * struct nilfs_sustat - + * struct nilfs_sustat - segment usage statistics * @ss_nsegs: number of segments * @ss_ncleansegs: number of clean segments * @ss_ndirtysegs: number of dirty segments @@ -784,10 +785,10 @@ struct nilfs_sustat { /** * struct nilfs_vinfo - virtual block number information - * @vi_vblocknr: - * @vi_start: - * @vi_end: - * @vi_blocknr: + * @vi_vblocknr: virtual block number + * @vi_start: start checkpoint number (inclusive) + * @vi_end: end checkpoint number (exclusive) + * @vi_blocknr: disk block number */ struct nilfs_vinfo { __u64 vi_vblocknr; @@ -797,7 +798,15 @@ struct nilfs_vinfo { }; /** - * struct nilfs_vdesc - + * struct nilfs_vdesc - descriptor of virtual block number + * @vd_ino: inode number + * @vd_cno: checkpoint number + * @vd_vblocknr: virtual block number + * @vd_period: period of checkpoint numbers + * @vd_blocknr: disk block number + * @vd_offset: logical block offset inside a file + * @vd_flags: flags (data or node block) + * @vd_pad: padding */ struct nilfs_vdesc { __u64 vd_ino; @@ -811,7 +820,13 @@ struct nilfs_vdesc { }; /** - * struct nilfs_bdesc - + * struct nilfs_bdesc - descriptor of disk block number + * @bd_ino: inode number + * @bd_oblocknr: disk block address (for skipping dead blocks) + * @bd_blocknr: disk block address + * @bd_offset: logical block offset inside a file + * @bd_level: level in the b-tree organization + * @bd_pad: padding */ struct nilfs_bdesc { __u64 bd_ino; -- cgit v0.10.2 From f5974c8f8cf431baf44e7127b669e3b1960f184f Mon Sep 17 00:00:00 2001 From: Vyacheslav Dubeyko Date: Mon, 30 Jul 2012 14:42:10 -0700 Subject: nilfs2: add omitted comments for different structures in driver implementation Add omitted comments for different structures in driver implementation. Signed-off-by: Vyacheslav Dubeyko Signed-off-by: Ryusuke Konishi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/nilfs2/alloc.h b/fs/nilfs2/alloc.h index f5fde36..fb72381 100644 --- a/fs/nilfs2/alloc.h +++ b/fs/nilfs2/alloc.h @@ -76,15 +76,23 @@ int nilfs_palloc_freev(struct inode *, __u64 *, size_t); #define nilfs_clear_bit_atomic ext2_clear_bit_atomic #define nilfs_find_next_zero_bit find_next_zero_bit_le -/* - * persistent object allocator cache +/** + * struct nilfs_bh_assoc - block offset and buffer head association + * @blkoff: block offset + * @bh: buffer head */ - struct nilfs_bh_assoc { unsigned long blkoff; struct buffer_head *bh; }; +/** + * struct nilfs_palloc_cache - persistent object allocator cache + * @lock: cache protecting lock + * @prev_desc: blockgroup descriptors cache + * @prev_bitmap: blockgroup bitmap cache + * @prev_entry: translation entries cache + */ struct nilfs_palloc_cache { spinlock_t lock; struct nilfs_bh_assoc prev_desc; diff --git a/fs/nilfs2/bmap.h b/fs/nilfs2/bmap.h index 40d9f45..b89e680 100644 --- a/fs/nilfs2/bmap.h +++ b/fs/nilfs2/bmap.h @@ -135,6 +135,13 @@ struct nilfs_bmap { /* state */ #define NILFS_BMAP_DIRTY 0x00000001 +/** + * struct nilfs_bmap_store - shadow copy of bmap state + * @data: cached raw block mapping of on-disk inode + * @last_allocated_key: cached value of last allocated key for data block + * @last_allocated_ptr: cached value of last allocated ptr for data block + * @state: cached value of state field of bmap structure + */ struct nilfs_bmap_store { __le64 data[NILFS_BMAP_SIZE / sizeof(__le64)]; __u64 last_allocated_key; diff --git a/fs/nilfs2/btnode.h b/fs/nilfs2/btnode.h index 3a4dd2d..d876b56 100644 --- a/fs/nilfs2/btnode.h +++ b/fs/nilfs2/btnode.h @@ -29,7 +29,13 @@ #include #include - +/** + * struct nilfs_btnode_chkey_ctxt - change key context + * @oldkey: old key of block's moving content + * @newkey: new key for block's content + * @bh: buffer head of old buffer + * @newbh: buffer head of new buffer + */ struct nilfs_btnode_chkey_ctxt { __u64 oldkey; __u64 newkey; diff --git a/fs/nilfs2/dat.c b/fs/nilfs2/dat.c index b5c13f3..fa0f803 100644 --- a/fs/nilfs2/dat.c +++ b/fs/nilfs2/dat.c @@ -33,6 +33,12 @@ #define NILFS_CNO_MIN ((__u64)1) #define NILFS_CNO_MAX (~(__u64)0) +/** + * struct nilfs_dat_info - on-memory private data of DAT file + * @mi: on-memory private data of metadata file + * @palloc_cache: persistent object allocator cache of DAT file + * @shadow: shadow map of DAT file + */ struct nilfs_dat_info { struct nilfs_mdt_info mi; struct nilfs_palloc_cache palloc_cache; diff --git a/fs/nilfs2/export.h b/fs/nilfs2/export.h index a71cc41..19ccbf9 100644 --- a/fs/nilfs2/export.h +++ b/fs/nilfs2/export.h @@ -5,6 +5,14 @@ extern const struct export_operations nilfs_export_ops; +/** + * struct nilfs_fid - NILFS file id type + * @cno: checkpoint number + * @ino: inode number + * @gen: file generation (version) for NFS + * @parent_gen: parent generation (version) for NFS + * @parent_ino: parent inode number + */ struct nilfs_fid { u64 cno; u64 ino; diff --git a/fs/nilfs2/ifile.c b/fs/nilfs2/ifile.c index 5a48df7..d8e65bd 100644 --- a/fs/nilfs2/ifile.c +++ b/fs/nilfs2/ifile.c @@ -29,7 +29,11 @@ #include "alloc.h" #include "ifile.h" - +/** + * struct nilfs_ifile_info - on-memory private data of ifile + * @mi: on-memory private data of metadata file + * @palloc_cache: persistent object allocator cache of ifile + */ struct nilfs_ifile_info { struct nilfs_mdt_info mi; struct nilfs_palloc_cache palloc_cache; diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index 7cc6446..6e2c3db 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -34,6 +34,13 @@ #include "cpfile.h" #include "ifile.h" +/** + * struct nilfs_iget_args - arguments used during comparison between inodes + * @ino: inode number + * @cno: checkpoint number + * @root: pointer on NILFS root object (mounted checkpoint) + * @for_gc: inode for GC flag + */ struct nilfs_iget_args { u64 ino; __u64 cno; diff --git a/fs/nilfs2/mdt.h b/fs/nilfs2/mdt.h index ab20a4b..ab172e8 100644 --- a/fs/nilfs2/mdt.h +++ b/fs/nilfs2/mdt.h @@ -28,6 +28,13 @@ #include "nilfs.h" #include "page.h" +/** + * struct nilfs_shadow_map - shadow mapping of meta data file + * @bmap_store: shadow copy of bmap state + * @frozen_data: shadowed dirty data pages + * @frozen_btnodes: shadowed dirty b-tree nodes' pages + * @frozen_buffers: list of frozen buffers + */ struct nilfs_shadow_map { struct nilfs_bmap_store bmap_store; struct address_space frozen_data; diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index 250add8..74cece8 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -32,8 +32,21 @@ #include "the_nilfs.h" #include "bmap.h" -/* - * nilfs inode data in memory +/** + * struct nilfs_inode_info - nilfs inode data in memory + * @i_flags: inode flags + * @i_state: dynamic state flags + * @i_bmap: pointer on i_bmap_data + * @i_bmap_data: raw block mapping + * @i_xattr: + * @i_dir_start_lookup: page index of last successful search + * @i_cno: checkpoint number for GC inode + * @i_btnode_cache: cached pages of b-tree nodes + * @i_dirty: list for connecting dirty files + * @xattr_sem: semaphore for extended attributes processing + * @i_bh: buffer contains disk inode + * @i_root: root object of the current filesystem tree + * @vfs_inode: VFS inode object */ struct nilfs_inode_info { __u32 i_flags; diff --git a/fs/nilfs2/sufile.c b/fs/nilfs2/sufile.c index c5b7653..3127e9f 100644 --- a/fs/nilfs2/sufile.c +++ b/fs/nilfs2/sufile.c @@ -30,7 +30,13 @@ #include "mdt.h" #include "sufile.h" - +/** + * struct nilfs_sufile_info - on-memory private data of sufile + * @mi: on-memory private data of metadata file + * @ncleansegs: number of clean segments + * @allocmin: lower limit of allocatable segment range + * @allocmax: upper limit of allocatable segment range + */ struct nilfs_sufile_info { struct nilfs_mdt_info mi; unsigned long ncleansegs;/* number of clean segments */ diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index 2558f32..6eee417 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -232,9 +232,8 @@ THE_NILFS_FNS(SB_DIRTY, sb_dirty) * @count: refcount of this structure * @nilfs: nilfs object * @ifile: inode file - * @root: root inode * @inodes_count: number of inodes - * @blocks_count: number of blocks (Reserved) + * @blocks_count: number of blocks */ struct nilfs_root { __u64 cno; -- cgit v0.10.2 From 497d48bd27ec1c44b4600e8e98a776188f2e11f2 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Mon, 30 Jul 2012 14:42:11 -0700 Subject: hfsplus: use -ENOMEM when kzalloc() fails Use -ENOMEM return value instead of -EINVAL when kzalloc() fails. Signed-off-by: Namjae Jeon Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 4733320..fdafb2d 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -365,7 +365,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) u64 last_fs_block, last_fs_page; int err; - err = -EINVAL; + err = -ENOMEM; sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) goto out; -- cgit v0.10.2 From a943ed71c9171fb5e3b256e8022bbedff95cc826 Mon Sep 17 00:00:00 2001 From: "Steven J. Magnani" Date: Mon, 30 Jul 2012 14:42:13 -0700 Subject: fat: accessors for msdos_dir_entry 'start' fields Simplify code by providing accessor functions for the directory entry start cluster fields. Signed-off-by: Steven J. Magnani Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fat/fat.h b/fs/fat/fat.h index fc35c5c..2deeeb8 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -217,6 +217,21 @@ static inline void fat16_towchar(wchar_t *dst, const __u8 *src, size_t len) #endif } +static inline int fat_get_start(const struct msdos_sb_info *sbi, + const struct msdos_dir_entry *de) +{ + int cluster = le16_to_cpu(de->start); + if (sbi->fat_bits == 32) + cluster |= (le16_to_cpu(de->starthi) << 16); + return cluster; +} + +static inline void fat_set_start(struct msdos_dir_entry *de, int cluster) +{ + de->start = cpu_to_le16(cluster); + de->starthi = cpu_to_le16(cluster >> 16); +} + static inline void fatwchar_to16(__u8 *dst, const wchar_t *src, size_t len) { #ifdef __BIG_ENDIAN diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 0038b32..05e897f 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -369,10 +369,7 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de) inode->i_op = sbi->dir_ops; inode->i_fop = &fat_dir_operations; - MSDOS_I(inode)->i_start = le16_to_cpu(de->start); - if (sbi->fat_bits == 32) - MSDOS_I(inode)->i_start |= (le16_to_cpu(de->starthi) << 16); - + MSDOS_I(inode)->i_start = fat_get_start(sbi, de); MSDOS_I(inode)->i_logstart = MSDOS_I(inode)->i_start; error = fat_calc_dir_size(inode); if (error < 0) @@ -385,9 +382,7 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de) inode->i_mode = fat_make_mode(sbi, de->attr, ((sbi->options.showexec && !is_exec(de->name + 8)) ? S_IRUGO|S_IWUGO : S_IRWXUGO)); - MSDOS_I(inode)->i_start = le16_to_cpu(de->start); - if (sbi->fat_bits == 32) - MSDOS_I(inode)->i_start |= (le16_to_cpu(de->starthi) << 16); + MSDOS_I(inode)->i_start = fat_get_start(sbi, de); MSDOS_I(inode)->i_logstart = MSDOS_I(inode)->i_start; inode->i_size = le32_to_cpu(de->size); @@ -613,8 +608,7 @@ retry: else raw_entry->size = cpu_to_le32(inode->i_size); raw_entry->attr = fat_make_attrs(inode); - raw_entry->start = cpu_to_le16(MSDOS_I(inode)->i_logstart); - raw_entry->starthi = cpu_to_le16(MSDOS_I(inode)->i_logstart >> 16); + fat_set_start(raw_entry, MSDOS_I(inode)->i_logstart); fat_time_unix2fat(sbi, &inode->i_mtime, &raw_entry->time, &raw_entry->date, NULL); if (sbi->options.isvfat) { diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 70d993a..b0e12bf 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -246,8 +246,7 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name, de.ctime_cs = 0; de.time = time; de.date = date; - de.start = cpu_to_le16(cluster); - de.starthi = cpu_to_le16(cluster >> 16); + fat_set_start(&de, cluster); de.size = 0; err = fat_add_entries(dir, &de, 1, sinfo); @@ -530,9 +529,7 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name, mark_inode_dirty(old_inode); if (update_dotdot) { - int start = MSDOS_I(new_dir)->i_logstart; - dotdot_de->start = cpu_to_le16(start); - dotdot_de->starthi = cpu_to_le16(start >> 16); + fat_set_start(dotdot_de, MSDOS_I(new_dir)->i_logstart); mark_buffer_dirty_inode(dotdot_bh, old_inode); if (IS_DIRSYNC(new_dir)) { err = sync_dirty_buffer(dotdot_bh); @@ -572,9 +569,7 @@ error_dotdot: corrupt = 1; if (update_dotdot) { - int start = MSDOS_I(old_dir)->i_logstart; - dotdot_de->start = cpu_to_le16(start); - dotdot_de->starthi = cpu_to_le16(start >> 16); + fat_set_start(dotdot_de, MSDOS_I(old_dir)->i_logstart); mark_buffer_dirty_inode(dotdot_bh, old_inode); corrupt |= sync_dirty_buffer(dotdot_bh); } diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 6cc4806..6a6d8c0 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -651,8 +651,7 @@ shortname: de->time = de->ctime = time; de->date = de->cdate = de->adate = date; de->ctime_cs = time_cs; - de->start = cpu_to_le16(cluster); - de->starthi = cpu_to_le16(cluster >> 16); + fat_set_start(de, cluster); de->size = 0; out_free: __putname(uname); @@ -965,9 +964,7 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, mark_inode_dirty(old_inode); if (update_dotdot) { - int start = MSDOS_I(new_dir)->i_logstart; - dotdot_de->start = cpu_to_le16(start); - dotdot_de->starthi = cpu_to_le16(start >> 16); + fat_set_start(dotdot_de, MSDOS_I(new_dir)->i_logstart); mark_buffer_dirty_inode(dotdot_bh, old_inode); if (IS_DIRSYNC(new_dir)) { err = sync_dirty_buffer(dotdot_bh); @@ -1009,9 +1006,7 @@ error_dotdot: corrupt = 1; if (update_dotdot) { - int start = MSDOS_I(old_dir)->i_logstart; - dotdot_de->start = cpu_to_le16(start); - dotdot_de->starthi = cpu_to_le16(start >> 16); + fat_set_start(dotdot_de, MSDOS_I(old_dir)->i_logstart); mark_buffer_dirty_inode(dotdot_bh, old_inode); corrupt |= sync_dirty_buffer(dotdot_bh); } -- cgit v0.10.2 From deb8274a0cf44827ec260330cc1d94d0f3dcfb94 Mon Sep 17 00:00:00 2001 From: "Steven J. Magnani" Date: Mon, 30 Jul 2012 14:42:16 -0700 Subject: fat: refactor shortname parsing Nearly identical shortname parsing is performed in fat_search_long() and __fat_readdir(). Extract this code into a function that may be called by both. Signed-off-by: Steven J. Magnani Acked-by: OGAWA Hirofumi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 6eaa28c..dc49ed2 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -35,6 +35,11 @@ #define FAT_MAX_UNI_CHARS ((MSDOS_SLOTS - 1) * 13 + 1) #define FAT_MAX_UNI_SIZE (FAT_MAX_UNI_CHARS * sizeof(wchar_t)) +static inline unsigned char fat_tolower(unsigned char c) +{ + return ((c >= 'A') && (c <= 'Z')) ? c+32 : c; +} + static inline loff_t fat_make_i_pos(struct super_block *sb, struct buffer_head *bh, struct msdos_dir_entry *de) @@ -333,6 +338,124 @@ parse_long: return 0; } +/** + * fat_parse_short - Parse MS-DOS (short) directory entry. + * @sb: superblock + * @de: directory entry to parse + * @name: FAT_MAX_SHORT_SIZE array in which to place extracted name + * @dot_hidden: Nonzero == prepend '.' to names with ATTR_HIDDEN + * + * Returns the number of characters extracted into 'name'. + */ +static int fat_parse_short(struct super_block *sb, + const struct msdos_dir_entry *de, + unsigned char *name, int dot_hidden) +{ + const struct msdos_sb_info *sbi = MSDOS_SB(sb); + int isvfat = sbi->options.isvfat; + int nocase = sbi->options.nocase; + unsigned short opt_shortname = sbi->options.shortname; + struct nls_table *nls_disk = sbi->nls_disk; + wchar_t uni_name[14]; + unsigned char c, work[MSDOS_NAME]; + unsigned char *ptname = name; + int chi, chl, i, j, k; + int dotoffset = 0; + int name_len = 0, uni_len = 0; + + if (!isvfat && dot_hidden && (de->attr & ATTR_HIDDEN)) { + *ptname++ = '.'; + dotoffset = 1; + } + + memcpy(work, de->name, sizeof(work)); + /* see namei.c, msdos_format_name */ + if (work[0] == 0x05) + work[0] = 0xE5; + + /* Filename */ + for (i = 0, j = 0; i < 8;) { + c = work[i]; + if (!c) + break; + chl = fat_shortname2uni(nls_disk, &work[i], 8 - i, + &uni_name[j++], opt_shortname, + de->lcase & CASE_LOWER_BASE); + if (chl <= 1) { + if (!isvfat) + ptname[i] = nocase ? c : fat_tolower(c); + i++; + if (c != ' ') { + name_len = i; + uni_len = j; + } + } else { + uni_len = j; + if (isvfat) + i += min(chl, 8-i); + else { + for (chi = 0; chi < chl && i < 8; chi++, i++) + ptname[i] = work[i]; + } + if (chl) + name_len = i; + } + } + + i = name_len; + j = uni_len; + fat_short2uni(nls_disk, ".", 1, &uni_name[j++]); + if (!isvfat) + ptname[i] = '.'; + i++; + + /* Extension */ + for (k = 8; k < MSDOS_NAME;) { + c = work[k]; + if (!c) + break; + chl = fat_shortname2uni(nls_disk, &work[k], MSDOS_NAME - k, + &uni_name[j++], opt_shortname, + de->lcase & CASE_LOWER_EXT); + if (chl <= 1) { + k++; + if (!isvfat) + ptname[i] = nocase ? c : fat_tolower(c); + i++; + if (c != ' ') { + name_len = i; + uni_len = j; + } + } else { + uni_len = j; + if (isvfat) { + int offset = min(chl, MSDOS_NAME-k); + k += offset; + i += offset; + } else { + for (chi = 0; chi < chl && k < MSDOS_NAME; + chi++, i++, k++) { + ptname[i] = work[k]; + } + } + if (chl) + name_len = i; + } + } + + if (name_len > 0) { + name_len += dotoffset; + + if (sbi->options.isvfat) { + uni_name[uni_len] = 0x0000; + name_len = fat_uni_to_x8(sb, uni_name, name, + FAT_MAX_SHORT_SIZE); + } + } + + return name_len; +} + /* * Return values: negative -> error, 0 -> not found, positive -> found, * value is the total amount of slots, including the shortname entry. @@ -344,15 +467,11 @@ int fat_search_long(struct inode *inode, const unsigned char *name, struct msdos_sb_info *sbi = MSDOS_SB(sb); struct buffer_head *bh = NULL; struct msdos_dir_entry *de; - struct nls_table *nls_disk = sbi->nls_disk; unsigned char nr_slots; - wchar_t bufuname[14]; wchar_t *unicode = NULL; - unsigned char work[MSDOS_NAME]; unsigned char bufname[FAT_MAX_SHORT_SIZE]; - unsigned short opt_shortname = sbi->options.shortname; loff_t cpos = 0; - int chl, i, j, last_u, err, len; + int err, len; err = -ENOENT; while (1) { @@ -380,47 +499,16 @@ parse_record: goto end_of_dir; } - memcpy(work, de->name, sizeof(de->name)); - /* see namei.c, msdos_format_name */ - if (work[0] == 0x05) - work[0] = 0xE5; - for (i = 0, j = 0, last_u = 0; i < 8;) { - if (!work[i]) - break; - chl = fat_shortname2uni(nls_disk, &work[i], 8 - i, - &bufuname[j++], opt_shortname, - de->lcase & CASE_LOWER_BASE); - if (chl <= 1) { - if (work[i] != ' ') - last_u = j; - } else { - last_u = j; - } - i += chl; - } - j = last_u; - fat_short2uni(nls_disk, ".", 1, &bufuname[j++]); - for (i = 8; i < MSDOS_NAME;) { - if (!work[i]) - break; - chl = fat_shortname2uni(nls_disk, &work[i], - MSDOS_NAME - i, - &bufuname[j++], opt_shortname, - de->lcase & CASE_LOWER_EXT); - if (chl <= 1) { - if (work[i] != ' ') - last_u = j; - } else { - last_u = j; - } - i += chl; - } - if (!last_u) + /* Never prepend '.' to hidden files here. + * That is done only for msdos mounts (and only when + * 'dotsOK=yes'); if we are executing here, it is in the + * context of a vfat mount. + */ + len = fat_parse_short(sb, de, bufname, 0); + if (len == 0) continue; /* Compare shortname */ - bufuname[last_u] = 0x0000; - len = fat_uni_to_x8(sb, bufuname, bufname, sizeof(bufname)); if (fat_name_match(sbi, name, name_len, bufname, len)) goto found; @@ -469,20 +557,15 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent, struct msdos_sb_info *sbi = MSDOS_SB(sb); struct buffer_head *bh; struct msdos_dir_entry *de; - struct nls_table *nls_disk = sbi->nls_disk; unsigned char nr_slots; - wchar_t bufuname[14]; wchar_t *unicode = NULL; - unsigned char c, work[MSDOS_NAME]; - unsigned char bufname[FAT_MAX_SHORT_SIZE], *ptname = bufname; - unsigned short opt_shortname = sbi->options.shortname; + unsigned char bufname[FAT_MAX_SHORT_SIZE]; int isvfat = sbi->options.isvfat; - int nocase = sbi->options.nocase; const char *fill_name = NULL; unsigned long inum; unsigned long lpos, dummy, *furrfu = &lpos; loff_t cpos; - int chi, chl, i, i2, j, last, last_u, dotoffset = 0, fill_len = 0; + int short_len = 0, fill_len = 0; int ret = 0; lock_super(sb); @@ -556,74 +639,10 @@ parse_record: } } - if (sbi->options.dotsOK) { - ptname = bufname; - dotoffset = 0; - if (de->attr & ATTR_HIDDEN) { - *ptname++ = '.'; - dotoffset = 1; - } - } - - memcpy(work, de->name, sizeof(de->name)); - /* see namei.c, msdos_format_name */ - if (work[0] == 0x05) - work[0] = 0xE5; - for (i = 0, j = 0, last = 0, last_u = 0; i < 8;) { - if (!(c = work[i])) - break; - chl = fat_shortname2uni(nls_disk, &work[i], 8 - i, - &bufuname[j++], opt_shortname, - de->lcase & CASE_LOWER_BASE); - if (chl <= 1) { - ptname[i++] = (!nocase && c>='A' && c<='Z') ? c+32 : c; - if (c != ' ') { - last = i; - last_u = j; - } - } else { - last_u = j; - for (chi = 0; chi < chl && i < 8; chi++) { - ptname[i] = work[i]; - i++; last = i; - } - } - } - i = last; - j = last_u; - fat_short2uni(nls_disk, ".", 1, &bufuname[j++]); - ptname[i++] = '.'; - for (i2 = 8; i2 < MSDOS_NAME;) { - if (!(c = work[i2])) - break; - chl = fat_shortname2uni(nls_disk, &work[i2], MSDOS_NAME - i2, - &bufuname[j++], opt_shortname, - de->lcase & CASE_LOWER_EXT); - if (chl <= 1) { - i2++; - ptname[i++] = (!nocase && c>='A' && c<='Z') ? c+32 : c; - if (c != ' ') { - last = i; - last_u = j; - } - } else { - last_u = j; - for (chi = 0; chi < chl && i2 < MSDOS_NAME; chi++) { - ptname[i++] = work[i2++]; - last = i; - } - } - } - if (!last) + short_len = fat_parse_short(sb, de, bufname, sbi->options.dotsOK); + if (short_len == 0) goto record_end; - i = last + dotoffset; - j = last_u; - - if (isvfat) { - bufuname[j] = 0x0000; - i = fat_uni_to_x8(sb, bufuname, bufname, sizeof(bufname)); - } if (nr_slots) { /* hack for fat_ioctl_filldir() */ struct fat_ioctl_filldir_callback *p = dirent; @@ -631,12 +650,12 @@ parse_record: p->longname = fill_name; p->long_len = fill_len; p->shortname = bufname; - p->short_len = i; + p->short_len = short_len; fill_name = NULL; fill_len = 0; } else { fill_name = bufname; - fill_len = i; + fill_len = short_len; } start_filldir: -- cgit v0.10.2 From 79c743dd1e8de61c31f484c0a1b48930543044b3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:42:17 -0700 Subject: kernel/kmod.c: document call_usermodehelper_fns() a bit This function's interface is, uh, subtle. Attempt to apologise for it. Cc: WANG Cong Cc: Cyrill Gorcunov Cc: Kees Cook Cc: Serge Hallyn Cc: "Eric W. Biederman" Cc: Alan Cox Cc: Oleg Nesterov Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/kmod.c b/kernel/kmod.c index ff2c7cb..2a83515 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -577,6 +577,12 @@ unlock: return retval; } +/* + * call_usermodehelper_fns() will not run the caller-provided cleanup function + * if a memory allocation failure is experienced. So the caller might need to + * check the call_usermodehelper_fns() return value: if it is -ENOMEM, perform + * the necessaary cleanup within the caller. + */ int call_usermodehelper_fns( char *path, char **argv, char **envp, int wait, int (*init)(struct subprocess_info *info, struct cred *new), -- cgit v0.10.2 From 0f20784d4ba3f88ca33b703b23372d8ccf6dbd42 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Mon, 30 Jul 2012 14:42:20 -0700 Subject: kmod: avoid deadlock from recursive kmod call The system deadlocks (at least since 2.6.10) when call_usermodehelper(UMH_WAIT_EXEC) request triggers call_usermodehelper(UMH_WAIT_PROC) request. This is because "khelper thread is waiting for the worker thread at wait_for_completion() in do_fork() since the worker thread was created with CLONE_VFORK flag" and "the worker thread cannot call complete() because do_execve() is blocked at UMH_WAIT_PROC request" and "the khelper thread cannot start processing UMH_WAIT_PROC request because the khelper thread is waiting for the worker thread at wait_for_completion() in do_fork()". The easiest example to observe this deadlock is to use a corrupted /sbin/hotplug binary (like shown below). # : > /tmp/dummy # chmod 755 /tmp/dummy # echo /tmp/dummy > /proc/sys/kernel/hotplug # modprobe whatever call_usermodehelper("/tmp/dummy", UMH_WAIT_EXEC) is called from kobject_uevent_env() in lib/kobject_uevent.c upon loading/unloading a module. do_execve("/tmp/dummy") triggers a call to request_module("binfmt-0000") from search_binary_handler() which in turn calls call_usermodehelper(UMH_WAIT_PROC). In order to avoid deadlock, as a for-now and easy-to-backport solution, do not try to call wait_for_completion() in call_usermodehelper_exec() if the worker thread was created by khelper thread with CLONE_VFORK flag. Future and fundamental solution might be replacing singleton khelper thread with some workqueue so that recursive calls up to max_active dependency loop can be handled without deadlock. [akpm@linux-foundation.org: add comment to kmod_thread_locker] Signed-off-by: Tetsuo Handa Cc: Arjan van de Ven Acked-by: Rusty Russell Cc: Tejun Heo Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/kmod.c b/kernel/kmod.c index 2a83515..6f99aea 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -45,6 +45,13 @@ extern int max_threads; static struct workqueue_struct *khelper_wq; +/* + * kmod_thread_locker is used for deadlock avoidance. There is no explicit + * locking to protect this global - it is private to the singleton khelper + * thread and should only ever be modified by that thread. + */ +static const struct task_struct *kmod_thread_locker; + #define CAP_BSET (void *)1 #define CAP_PI (void *)2 @@ -221,6 +228,13 @@ fail: return 0; } +static int call_helper(void *data) +{ + /* Worker thread started blocking khelper thread. */ + kmod_thread_locker = current; + return ____call_usermodehelper(data); +} + static void call_usermodehelper_freeinfo(struct subprocess_info *info) { if (info->cleanup) @@ -295,9 +309,12 @@ static void __call_usermodehelper(struct work_struct *work) if (wait == UMH_WAIT_PROC) pid = kernel_thread(wait_for_helper, sub_info, CLONE_FS | CLONE_FILES | SIGCHLD); - else - pid = kernel_thread(____call_usermodehelper, sub_info, + else { + pid = kernel_thread(call_helper, sub_info, CLONE_VFORK | SIGCHLD); + /* Worker thread stopped blocking khelper thread. */ + kmod_thread_locker = NULL; + } switch (wait) { case UMH_NO_WAIT: @@ -548,6 +565,16 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) retval = -EBUSY; goto out; } + /* + * Worker thread must not wait for khelper thread at below + * wait_for_completion() if the thread was created with CLONE_VFORK + * flag, for khelper thread is already waiting for the thread at + * wait_for_completion() in do_fork(). + */ + if (wait != UMH_NO_WAIT && current == kmod_thread_locker) { + retval = -EBUSY; + goto out; + } sub_info->complete = &done; sub_info->wait = wait; -- cgit v0.10.2 From 108ceeb020bb3558fe175a3fc8b60fd6c1a2a279 Mon Sep 17 00:00:00 2001 From: Jovi Zhang Date: Mon, 30 Jul 2012 14:42:23 -0700 Subject: coredump: fix wrong comments on core limits of pipe coredump case In commit 898b374af6f7 ("exec: replace call_usermodehelper_pipe with use of umh init function and resolve limit"), the core limits recursive check value was changed from 0 to 1, but the corresponding comments were not updated. Signed-off-by: Jovi Zhang Cc: Oleg Nesterov Cc: Neil Horman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/exec.c b/fs/exec.c index 5af8390..3684353 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2174,15 +2174,16 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) } if (cprm.limit == 1) { - /* + /* See umh_pipe_setup() which sets RLIMIT_CORE = 1. + * * Normally core limits are irrelevant to pipes, since * we're not writing to the file system, but we use - * cprm.limit of 1 here as a speacial value. Any - * non-1 limit gets set to RLIM_INFINITY below, but - * a limit of 0 skips the dump. This is a consistent - * way to catch recursive crashes. We can still crash - * if the core_pattern binary sets RLIM_CORE = !1 - * but it runs as root, and can do lots of stupid things + * cprm.limit of 1 here as a speacial value, this is a + * consistent way to catch recursive crashes. + * We can still crash if the core_pattern binary sets + * RLIM_CORE = !1, but it runs as root, and can do + * lots of stupid things. + * * Note that we use task_tgid_vnr here to grab the pid * of the process group leader. That way we get the * right pid if a thread in a multi-threaded -- cgit v0.10.2 From e8905ec27e2f4ea1b9f7e03df68a060b3ae6fca8 Mon Sep 17 00:00:00 2001 From: Djalal Harouni Date: Mon, 30 Jul 2012 14:42:26 -0700 Subject: proc: environ_read() make sure offset points to environment address range Currently the following offset and environment address range check in environ_read() of /proc//environ is buggy: int this_len = mm->env_end - (mm->env_start + src); if (this_len <= 0) break; Large or negative offsets on /proc//environ converted to 'unsigned long' may pass this check since '(mm->env_start + src)' can overflow and 'this_len' will be positive. This can turn /proc//environ to act like /proc//mem since (mm->env_start + src) will point and read from another VMA. There are two fixes here plus some code cleaning: 1) Fix the overflow by checking if the offset that was converted to unsigned long will always point to the [mm->env_start, mm->env_end] address range. 2) Remove the truncation that was made to the result of the check, storing the result in 'int this_len' will alter its value and we can not depend on it. For kernels that have commit b409e578d ("proc: clean up /proc//environ handling") which adds the appropriate ptrace check and saves the 'mm' at ->open() time, this is not a security issue. This patch is taken from the grsecurity patch since it was just made available. Signed-off-by: Djalal Harouni Cc: Oleg Nesterov Cc: Brad Spengler Acked-by: Kees Cook Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 2772208..39ee093 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -827,15 +827,16 @@ static ssize_t environ_read(struct file *file, char __user *buf, if (!atomic_inc_not_zero(&mm->mm_users)) goto free; while (count > 0) { - int this_len, retval, max_len; + size_t this_len, max_len; + int retval; - this_len = mm->env_end - (mm->env_start + src); - - if (this_len <= 0) + if (src >= (mm->env_end - mm->env_start)) break; - max_len = (count > PAGE_SIZE) ? PAGE_SIZE : count; - this_len = (this_len > max_len) ? max_len : this_len; + this_len = mm->env_end - (mm->env_start + src); + + max_len = min_t(size_t, PAGE_SIZE, count); + this_len = min(max_len, this_len); retval = access_remote_vm(mm, (mm->env_start + src), page, this_len, 0); -- cgit v0.10.2 From bc452b4b65bd589083a7a7ba4f14f85dfc8454fa Mon Sep 17 00:00:00 2001 From: Djalal Harouni Date: Mon, 30 Jul 2012 14:42:28 -0700 Subject: proc: do not allow negative offsets on /proc//environ __mem_open() which is called by both /proc//environ and /proc//mem ->open() handlers will allow the use of negative offsets. /proc//mem has negative offsets but not /proc//environ. Clean this by moving the 'force FMODE_UNSIGNED_OFFSET flag' to mem_open() to allow negative offsets only on /proc//mem. Signed-off-by: Djalal Harouni Cc: Oleg Nesterov Cc: Brad Spengler Acked-by: Kees Cook Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/base.c b/fs/proc/base.c index 39ee093..1b6c84c 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -695,8 +695,6 @@ static int __mem_open(struct inode *inode, struct file *file, unsigned int mode) mmput(mm); } - /* OK to pass negative loff_t, we can catch out-of-range */ - file->f_mode |= FMODE_UNSIGNED_OFFSET; file->private_data = mm; return 0; @@ -704,7 +702,12 @@ static int __mem_open(struct inode *inode, struct file *file, unsigned int mode) static int mem_open(struct inode *inode, struct file *file) { - return __mem_open(inode, file, PTRACE_MODE_ATTACH); + int ret = __mem_open(inode, file, PTRACE_MODE_ATTACH); + + /* OK to pass negative loff_t, we can catch out-of-range */ + file->f_mode |= FMODE_UNSIGNED_OFFSET; + + return ret; } static ssize_t mem_rw(struct file *file, char __user *buf, -- cgit v0.10.2 From b2412b7fa7a3816fa8633dc2ff19f1a90aabe423 Mon Sep 17 00:00:00 2001 From: Huang Shijie Date: Mon, 30 Jul 2012 14:42:30 -0700 Subject: fork: use vma_pages() to simplify the code The current code can be replaced by vma_pages(). So use it to simplify the code. [akpm@linux-foundation.org: initialise `len' at its definition site] Signed-off-by: Huang Shijie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index ff1cad3..2c18029 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -391,8 +391,8 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) } charge = 0; if (mpnt->vm_flags & VM_ACCOUNT) { - unsigned long len; - len = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT; + unsigned long len = vma_pages(mpnt); + if (security_vm_enough_memory_mm(oldmm, len)) /* sic */ goto fail_nomem; charge = len; -- cgit v0.10.2 From 87bec58a52652e2eb2a575692a40f9466c7bd31b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:42:31 -0700 Subject: revert "sched: Fix fork() error path to not crash" To make way for "fork: fix error handling in dup_task()", which fixes the errors more completely. Cc: Salman Qazi Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index 2c18029..088025b 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -304,17 +304,12 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) } err = arch_dup_task_struct(tsk, orig); - - /* - * We defer looking at err, because we will need this setup - * for the clean up path to work correctly. - */ - tsk->stack = ti; - setup_thread_stack(tsk, orig); - if (err) goto out; + tsk->stack = ti; + + setup_thread_stack(tsk, orig); clear_user_return_notifier(tsk); clear_tsk_need_resched(tsk); stackend = end_of_stack(tsk); -- cgit v0.10.2 From f19b9f74b7ea3b21ddcee55d852a6488239608a4 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:42:33 -0700 Subject: fork: fix error handling in dup_task() The function dup_task() may fail at the following function calls in the following order. 0) alloc_task_struct_node() 1) alloc_thread_info_node() 2) arch_dup_task_struct() Error by 0) is not a matter, it can just return. But error by 1) requires releasing task_struct allocated by 0) before it returns. Likewise, error by 2) requires releasing task_struct and thread_info allocated by 0) and 1). The existing error handling calls free_task_struct() and free_thread_info() which do not only release task_struct and thread_info, but also call architecture specific arch_release_task_struct() and arch_release_thread_info(). The problem is that task_struct and thread_info are not fully initialized yet at this point, but arch_release_task_struct() and arch_release_thread_info() are called with them. For example, x86 defines its own arch_release_task_struct() that releases a task_xstate. If alloc_thread_info_node() fails in dup_task(), arch_release_task_struct() is called with task_struct which is just allocated and filled with garbage in this error handling. This actually happened with tools/testing/fault-injection/failcmd.sh # env FAILCMD_TYPE=fail_page_alloc \ ./tools/testing/fault-injection/failcmd.sh --times=100 \ --min-order=0 --ignore-gfp-wait=0 \ -- make -C tools/testing/selftests/ run_tests In order to fix this issue, make free_{task_struct,thread_info}() not to call arch_release_{task_struct,thread_info}() and call arch_release_{task_struct,thread_info}() implicitly where needed. Default arch_release_task_struct() and arch_release_thread_info() are defined as empty by default. So this change only affects the architectures which implement their own arch_release_task_struct() or arch_release_thread_info() as listed below. arch_release_task_struct(): x86, sh arch_release_thread_info(): mn10300, tile Signed-off-by: Akinobu Mita Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: David Howells Cc: Koichi Yasutake Cc: Paul Mundt Cc: Chris Metcalf Cc: Salman Qazi Cc: Peter Zijlstra Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/fork.c b/kernel/fork.c index 088025b..8efac1f 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -114,6 +114,10 @@ int nr_processes(void) return total; } +void __weak arch_release_task_struct(struct task_struct *tsk) +{ +} + #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR static struct kmem_cache *task_struct_cachep; @@ -122,17 +126,17 @@ static inline struct task_struct *alloc_task_struct_node(int node) return kmem_cache_alloc_node(task_struct_cachep, GFP_KERNEL, node); } -void __weak arch_release_task_struct(struct task_struct *tsk) { } - static inline void free_task_struct(struct task_struct *tsk) { - arch_release_task_struct(tsk); kmem_cache_free(task_struct_cachep, tsk); } #endif +void __weak arch_release_thread_info(struct thread_info *ti) +{ +} + #ifndef CONFIG_ARCH_THREAD_INFO_ALLOCATOR -void __weak arch_release_thread_info(struct thread_info *ti) { } /* * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a @@ -150,7 +154,6 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, static inline void free_thread_info(struct thread_info *ti) { - arch_release_thread_info(ti); free_pages((unsigned long)ti, THREAD_SIZE_ORDER); } # else @@ -164,7 +167,6 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, static void free_thread_info(struct thread_info *ti) { - arch_release_thread_info(ti); kmem_cache_free(thread_info_cache, ti); } @@ -205,10 +207,12 @@ static void account_kernel_stack(struct thread_info *ti, int account) void free_task(struct task_struct *tsk) { account_kernel_stack(tsk->stack, -1); + arch_release_thread_info(tsk->stack); free_thread_info(tsk->stack); rt_mutex_debug_task_free(tsk); ftrace_graph_exit_task(tsk); put_seccomp_filter(tsk); + arch_release_task_struct(tsk); free_task_struct(tsk); } EXPORT_SYMBOL(free_task); @@ -298,14 +302,12 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) return NULL; ti = alloc_thread_info_node(tsk, node); - if (!ti) { - free_task_struct(tsk); - return NULL; - } + if (!ti) + goto free_tsk; err = arch_dup_task_struct(tsk, orig); if (err) - goto out; + goto free_ti; tsk->stack = ti; @@ -333,8 +335,9 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) return tsk; -out: +free_ti: free_thread_info(ti); +free_tsk: free_task_struct(tsk); return NULL; } -- cgit v0.10.2 From 63dca8d5b5ef7effb58b8d6892a024125c0fab0b Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 30 Jul 2012 14:42:36 -0700 Subject: kdump: append newline to the last lien of vmcoreinfo note The last line of vmcoreinfo note does not end with \n. Parsing all the lines in note becomes easier if all lines end with \n instead of trying to special case the last line. I know at least one tool, vmcore-dmesg in kexec-tools tree which made the assumption that all lines end with \n. I think it is a good idea to fix it. Signed-off-by: Vivek Goyal Cc: "Eric W. Biederman" Cc: Atsushi Kumagai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/kexec.c b/kernel/kexec.c index 4e2e472..0668d58 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1424,7 +1424,7 @@ static void update_vmcoreinfo_note(void) void crash_save_vmcoreinfo(void) { - vmcoreinfo_append_str("CRASHTIME=%ld", get_seconds()); + vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); update_vmcoreinfo_note(); } -- cgit v0.10.2 From 079a96ae3871f0ed9083aac2218136ccec5b9877 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 30 Jul 2012 14:42:38 -0700 Subject: ipc: add COMPAT_SHMLBA support If the SHMLBA definition for a native task differs from the definition for a compat task, the do_shmat() function would need to handle both. This patch introduces COMPAT_SHMLBA, which is used by the compat shmat syscall when calling the ipc code and allows architectures such as AArch64 (where the native SHMLBA is 64k but the compat (AArch32) definition is 16k) to provide the correct semantics for compat IPC system calls. Cc: David S. Miller Cc: Chris Zankel Cc: Arnd Bergmann Acked-by: Catalin Marinas Signed-off-by: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c index c38e5aa..0dc1f57 100644 --- a/arch/sparc/kernel/sys_sparc_64.c +++ b/arch/sparc/kernel/sys_sparc_64.c @@ -470,7 +470,7 @@ SYSCALL_DEFINE6(sparc_ipc, unsigned int, call, int, first, unsigned long, second switch (call) { case SHMAT: { ulong raddr; - err = do_shmat(first, ptr, (int)second, &raddr); + err = do_shmat(first, ptr, (int)second, &raddr, SHMLBA); if (!err) { if (put_user(raddr, (ulong __user *) third)) diff --git a/arch/xtensa/kernel/syscall.c b/arch/xtensa/kernel/syscall.c index 816e6d0..05b3f09 100644 --- a/arch/xtensa/kernel/syscall.c +++ b/arch/xtensa/kernel/syscall.c @@ -44,7 +44,7 @@ asmlinkage long xtensa_shmat(int shmid, char __user *shmaddr, int shmflg) unsigned long ret; long err; - err = do_shmat(shmid, shmaddr, shmflg, &ret); + err = do_shmat(shmid, shmaddr, shmflg, &ret, SHMLBA); if (err) return err; return (long)ret; diff --git a/include/linux/shm.h b/include/linux/shm.h index 92808b8..edd0868 100644 --- a/include/linux/shm.h +++ b/include/linux/shm.h @@ -107,12 +107,14 @@ struct shmid_kernel /* private to the kernel */ #define SHM_NORESERVE 010000 /* don't check for reservations */ #ifdef CONFIG_SYSVIPC -long do_shmat(int shmid, char __user *shmaddr, int shmflg, unsigned long *addr); +long do_shmat(int shmid, char __user *shmaddr, int shmflg, unsigned long *addr, + unsigned long shmlba); extern int is_file_shm_hugepages(struct file *file); extern void exit_shm(struct task_struct *task); #else static inline long do_shmat(int shmid, char __user *shmaddr, - int shmflg, unsigned long *addr) + int shmflg, unsigned long *addr, + unsigned long shmlba) { return -ENOSYS; } diff --git a/ipc/compat.c b/ipc/compat.c index a6df704..53cebdf 100644 --- a/ipc/compat.c +++ b/ipc/compat.c @@ -514,6 +514,10 @@ long compat_sys_msgctl(int first, int second, void __user *uptr) return err; } +#ifndef COMPAT_SHMLBA +#define COMPAT_SHMLBA SHMLBA +#endif + #ifdef CONFIG_ARCH_WANT_OLD_COMPAT_IPC long compat_sys_shmat(int first, int second, compat_uptr_t third, int version, void __user *uptr) @@ -524,7 +528,7 @@ long compat_sys_shmat(int first, int second, compat_uptr_t third, int version, if (version == 1) return -EINVAL; - err = do_shmat(first, uptr, second, &raddr); + err = do_shmat(first, uptr, second, &raddr, COMPAT_SHMLBA); if (err < 0) return err; uaddr = compat_ptr(third); @@ -536,7 +540,7 @@ long compat_sys_shmat(int shmid, compat_uptr_t shmaddr, int shmflg) unsigned long ret; long err; - err = do_shmat(shmid, compat_ptr(shmaddr), shmflg, &ret); + err = do_shmat(shmid, compat_ptr(shmaddr), shmflg, &ret, COMPAT_SHMLBA); if (err) return err; force_successful_syscall_return(); diff --git a/ipc/shm.c b/ipc/shm.c index 41c1285..00faa05 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -953,7 +953,8 @@ out: * "raddr" thing points to kernel space, and there has to be a wrapper around * this. */ -long do_shmat(int shmid, char __user *shmaddr, int shmflg, ulong *raddr) +long do_shmat(int shmid, char __user *shmaddr, int shmflg, ulong *raddr, + unsigned long shmlba) { struct shmid_kernel *shp; unsigned long addr; @@ -973,9 +974,9 @@ long do_shmat(int shmid, char __user *shmaddr, int shmflg, ulong *raddr) if (shmid < 0) goto out; else if ((addr = (ulong)shmaddr)) { - if (addr & (SHMLBA-1)) { + if (addr & (shmlba - 1)) { if (shmflg & SHM_RND) - addr &= ~(SHMLBA-1); /* round down */ + addr &= ~(shmlba - 1); /* round down */ else #ifndef __ARCH_FORCE_SHMLBA if (addr & ~PAGE_MASK) @@ -1107,7 +1108,7 @@ SYSCALL_DEFINE3(shmat, int, shmid, char __user *, shmaddr, int, shmflg) unsigned long ret; long err; - err = do_shmat(shmid, shmaddr, shmflg, &ret); + err = do_shmat(shmid, shmaddr, shmflg, &ret, SHMLBA); if (err) return err; force_successful_syscall_return(); diff --git a/ipc/syscall.c b/ipc/syscall.c index 1d6f53f..0d1e32ce 100644 --- a/ipc/syscall.c +++ b/ipc/syscall.c @@ -73,7 +73,7 @@ SYSCALL_DEFINE6(ipc, unsigned int, call, int, first, unsigned long, second, default: { unsigned long raddr; ret = do_shmat(first, (char __user *)ptr, - second, &raddr); + second, &raddr, SHMLBA); if (ret) return ret; return put_user(raddr, (unsigned long __user *) third); -- cgit v0.10.2 From b610c04c667f3c056243fd64041c7f152a512ee4 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 30 Jul 2012 14:42:40 -0700 Subject: ipc: allow compat IPC version field parsing if !ARCH_WANT_OLD_COMPAT_IPC Commit 48b25c43e6ee ("ipc: provide generic compat versions of IPC syscalls") added a new ARCH_WANT_OLD_COMPAT_IPC config option for architectures to select if their compat target requires the old IPC syscall interface. For architectures (such as AArch64) that do not require the internal calling conventions provided by this option, but have a compat target where the C library passes the IPC_64 flag explicitly, compat_ipc_parse_version no longer strips out the flag before calling the native system call implementation, resulting in unknown SHM/IPC commands and -EINVAL being returned to userspace. This patch separates the selection of the internal calling conventions for the IPC syscalls from the version parsing, allowing architectures to select __ARCH_WANT_COMPAT_IPC_PARSE_VERSION if they want to use version parsing whilst retaining the newer syscall calling conventions. Acked-by: Chris Metcalf Cc: Arnd Bergmann Acked-by: Catalin Marinas Signed-off-by: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compat.h b/include/linux/compat.h index 4e89039..9f68e90 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -256,6 +256,7 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr, compat_size_t __user *len_ptr); #ifdef CONFIG_ARCH_WANT_OLD_COMPAT_IPC +#define __ARCH_WANT_COMPAT_IPC_PARSE_VERSION long compat_sys_semctl(int first, int second, int third, void __user *uptr); long compat_sys_msgsnd(int first, int second, int third, void __user *uptr); long compat_sys_msgrcv(int first, int second, int msgtyp, int third, diff --git a/ipc/compat.c b/ipc/compat.c index 53cebdf..a41600f 100644 --- a/ipc/compat.c +++ b/ipc/compat.c @@ -118,7 +118,7 @@ extern int sem_ctls[]; static inline int compat_ipc_parse_version(int *cmd) { -#ifdef CONFIG_ARCH_WANT_OLD_COMPAT_IPC +#ifdef __ARCH_WANT_COMPAT_IPC_PARSE_VERSION int version = *cmd & IPC_64; /* this is tricky: architectures that have support for the old -- cgit v0.10.2 From 05ba3f1aa1b04e921068249dd52a80bc84c2aeb4 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 30 Jul 2012 14:42:43 -0700 Subject: ipc: compat: use signed size_t types for msgsnd and msgrcv The msgsnd and msgrcv system calls use size_t to represent the size of the message being transferred. POSIX states that values of msgsz greater than SSIZE_MAX cause the result to be implementation-defined. On Linux, this equates to returning -EINVAL if (long) msgsz < 0. For compat tasks where !CONFIG_ARCH_WANT_OLD_COMPAT_IPC and compat_size_t is smaller than size_t, negative size values passed from userspace will be interpreted as positive values by do_msg{rcv,snd} and will fail to exit early with -EINVAL. This patch changes the compat prototypes for msg{rcv,snd} so that the message size is represented as a compat_ssize_t, which we cast to the native ssize_t type for the core IPC code. Cc: Arnd Bergmann Acked-by: Chris Metcalf Acked-by: Catalin Marinas Signed-off-by: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/compat.h b/include/linux/compat.h index 9f68e90..f2b8fe2 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -266,9 +266,9 @@ long compat_sys_shmat(int first, int second, compat_uptr_t third, int version, #else long compat_sys_semctl(int semid, int semnum, int cmd, int arg); long compat_sys_msgsnd(int msqid, struct compat_msgbuf __user *msgp, - size_t msgsz, int msgflg); + compat_ssize_t msgsz, int msgflg); long compat_sys_msgrcv(int msqid, struct compat_msgbuf __user *msgp, - size_t msgsz, long msgtyp, int msgflg); + compat_ssize_t msgsz, long msgtyp, int msgflg); long compat_sys_shmat(int shmid, compat_uptr_t shmaddr, int shmflg); #endif long compat_sys_msgctl(int first, int second, void __user *uptr); diff --git a/ipc/compat.c b/ipc/compat.c index a41600f..20f92b2 100644 --- a/ipc/compat.c +++ b/ipc/compat.c @@ -373,21 +373,21 @@ long compat_sys_semctl(int semid, int semnum, int cmd, int arg) } long compat_sys_msgsnd(int msqid, struct compat_msgbuf __user *msgp, - size_t msgsz, int msgflg) + compat_ssize_t msgsz, int msgflg) { compat_long_t mtype; if (get_user(mtype, &msgp->mtype)) return -EFAULT; - return do_msgsnd(msqid, mtype, msgp->mtext, msgsz, msgflg); + return do_msgsnd(msqid, mtype, msgp->mtext, (ssize_t)msgsz, msgflg); } long compat_sys_msgrcv(int msqid, struct compat_msgbuf __user *msgp, - size_t msgsz, long msgtyp, int msgflg) + compat_ssize_t msgsz, long msgtyp, int msgflg) { long err, mtype; - err = do_msgrcv(msqid, &mtype, msgp->mtext, msgsz, msgtyp, msgflg); + err = do_msgrcv(msqid, &mtype, msgp->mtext, (ssize_t)msgsz, msgtyp, msgflg); if (err < 0) goto out; -- cgit v0.10.2 From c1d7e01d7877a397655277a920aeaa3830ed9461 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 30 Jul 2012 14:42:46 -0700 Subject: ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION Rather than #define the options manually in the architecture code, add Kconfig options for them and select them there instead. This also allows us to select the compat IPC version parsing automatically for platforms using the old compat IPC interface. Reported-by: Andrew Morton Signed-off-by: Will Deacon Cc: Arnd Bergmann Cc: Chris Metcalf Cc: Catalin Marinas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/Kconfig b/arch/Kconfig index 8c3d957..72f2fa1 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -248,7 +248,14 @@ config HAVE_CMPXCHG_LOCAL config HAVE_CMPXCHG_DOUBLE bool +config ARCH_WANT_IPC_PARSE_VERSION + bool + +config ARCH_WANT_COMPAT_IPC_PARSE_VERSION + bool + config ARCH_WANT_OLD_COMPAT_IPC + select ARCH_WANT_COMPAT_IPC_PARSE_VERSION bool config HAVE_ARCH_SECCOMP_FILTER diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index 3de74c9..d5b9b5e 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -14,6 +14,7 @@ config ALPHA select AUTO_IRQ_AFFINITY if SMP select GENERIC_IRQ_SHOW select ARCH_WANT_OPTIONAL_GPIOLIB + select ARCH_WANT_IPC_PARSE_VERSION select ARCH_HAVE_NMI_SAFE_CMPXCHG select GENERIC_SMP_IDLE_THREAD select GENERIC_CMOS_UPDATE diff --git a/arch/alpha/include/asm/unistd.h b/arch/alpha/include/asm/unistd.h index d1f23b7..633b23b 100644 --- a/arch/alpha/include/asm/unistd.h +++ b/arch/alpha/include/asm/unistd.h @@ -470,7 +470,6 @@ #define NR_SYSCALLS 504 -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 4172c3c..5df1114 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -39,6 +39,7 @@ config ARM select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW select GENERIC_IRQ_PROBE + select ARCH_WANT_IPC_PARSE_VERSION select HARDIRQS_SW_RESEND select CPU_PM if (SUSPEND || CPU_IDLE) select GENERIC_PCI_IOMAP diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h index 512cd14..0cab47d 100644 --- a/arch/arm/include/asm/unistd.h +++ b/arch/arm/include/asm/unistd.h @@ -446,7 +446,6 @@ #ifdef __KERNEL__ -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_GETHOSTNAME #define __ARCH_WANT_SYS_PAUSE diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig index 71d38c7..5ade51c 100644 --- a/arch/avr32/Kconfig +++ b/arch/avr32/Kconfig @@ -12,6 +12,7 @@ config AVR32 select HARDIRQS_SW_RESEND select GENERIC_IRQ_SHOW select ARCH_HAVE_CUSTOM_GPIO_H + select ARCH_WANT_IPC_PARSE_VERSION select ARCH_HAVE_NMI_SAFE_CMPXCHG select GENERIC_CLOCKEVENTS help diff --git a/arch/avr32/include/asm/unistd.h b/arch/avr32/include/asm/unistd.h index f714544..1358e36 100644 --- a/arch/avr32/include/asm/unistd.h +++ b/arch/avr32/include/asm/unistd.h @@ -318,7 +318,6 @@ /* SMP stuff */ #define __IGNORE_getcpu -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig index 9b76510..fb9fe00 100644 --- a/arch/blackfin/Kconfig +++ b/arch/blackfin/Kconfig @@ -33,6 +33,7 @@ config BLACKFIN select HAVE_PERF_EVENTS select ARCH_HAVE_CUSTOM_GPIO_H select ARCH_WANT_OPTIONAL_GPIOLIB + select ARCH_WANT_IPC_PARSE_VERSION select HAVE_GENERIC_HARDIRQS select GENERIC_ATOMIC64 select GENERIC_IRQ_PROBE diff --git a/arch/blackfin/include/asm/unistd.h b/arch/blackfin/include/asm/unistd.h index 3287222..5b2a074 100644 --- a/arch/blackfin/include/asm/unistd.h +++ b/arch/blackfin/include/asm/unistd.h @@ -434,7 +434,6 @@ #define __IGNORE_getcpu #ifdef __KERNEL__ -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/cris/Kconfig b/arch/cris/Kconfig index bb34465..e922154 100644 --- a/arch/cris/Kconfig +++ b/arch/cris/Kconfig @@ -42,6 +42,7 @@ config CRIS select HAVE_IDE select GENERIC_ATOMIC64 select HAVE_GENERIC_HARDIRQS + select ARCH_WANT_IPC_PARSE_VERSION select GENERIC_IRQ_SHOW select GENERIC_IOMAP select GENERIC_SMP_IDLE_THREAD if ETRAX_ARCH_V32 diff --git a/arch/cris/include/asm/unistd.h b/arch/cris/include/asm/unistd.h index f921b8b..51873a4 100644 --- a/arch/cris/include/asm/unistd.h +++ b/arch/cris/include/asm/unistd.h @@ -347,7 +347,6 @@ #include -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT #define __ARCH_WANT_STAT64 diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index a685910..971c0a1 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -9,6 +9,7 @@ config FRV select GENERIC_IRQ_SHOW select ARCH_HAVE_NMI_SAFE_CMPXCHG select GENERIC_CPU_DEVICES + select ARCH_WANT_IPC_PARSE_VERSION config ZONE_DMA bool diff --git a/arch/frv/include/asm/unistd.h b/arch/frv/include/asm/unistd.h index a569dff..67f23a3 100644 --- a/arch/frv/include/asm/unistd.h +++ b/arch/frv/include/asm/unistd.h @@ -349,7 +349,6 @@ #define NR_syscalls 338 -#define __ARCH_WANT_IPC_PARSE_VERSION /* #define __ARCH_WANT_OLD_READDIR */ #define __ARCH_WANT_OLD_STAT #define __ARCH_WANT_STAT64 diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig index 56e890d..5e8a0d9 100644 --- a/arch/h8300/Kconfig +++ b/arch/h8300/Kconfig @@ -3,6 +3,7 @@ config H8300 default y select HAVE_IDE select HAVE_GENERIC_HARDIRQS + select ARCH_WANT_IPC_PARSE_VERSION select GENERIC_IRQ_SHOW select GENERIC_CPU_DEVICES diff --git a/arch/h8300/include/asm/unistd.h b/arch/h8300/include/asm/unistd.h index 7185113..5cd8828 100644 --- a/arch/h8300/include/asm/unistd.h +++ b/arch/h8300/include/asm/unistd.h @@ -331,7 +331,6 @@ #define NR_syscalls 321 -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT #define __ARCH_WANT_STAT64 diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index b638d5b..49498bb 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -7,6 +7,7 @@ config M32R select HAVE_KERNEL_GZIP select HAVE_KERNEL_BZIP2 select HAVE_KERNEL_LZMA + select ARCH_WANT_IPC_PARSE_VERSION select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW diff --git a/arch/m32r/include/asm/unistd.h b/arch/m32r/include/asm/unistd.h index 3e1db56..d5e66a4 100644 --- a/arch/m32r/include/asm/unistd.h +++ b/arch/m32r/include/asm/unistd.h @@ -336,7 +336,6 @@ #define NR_syscalls 326 -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig index 1471201..0b0f8b8 100644 --- a/arch/m68k/Kconfig +++ b/arch/m68k/Kconfig @@ -10,6 +10,7 @@ config M68K select GENERIC_STRNCPY_FROM_USER if MMU select GENERIC_STRNLEN_USER if MMU select FPU if MMU + select ARCH_WANT_IPC_PARSE_VERSION select ARCH_USES_GETTIMEOFFSET if MMU && !COLDFIRE config RWSEM_GENERIC_SPINLOCK diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h index ea0b502..045cfd6 100644 --- a/arch/m68k/include/asm/unistd.h +++ b/arch/m68k/include/asm/unistd.h @@ -357,7 +357,6 @@ #define NR_syscalls 347 -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT #define __ARCH_WANT_STAT64 diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig index 0bf4423..ab9afca 100644 --- a/arch/microblaze/Kconfig +++ b/arch/microblaze/Kconfig @@ -15,6 +15,7 @@ config MICROBLAZE select TRACING_SUPPORT select OF select OF_EARLY_FLATTREE + select ARCH_WANT_IPC_PARSE_VERSION select IRQ_DOMAIN select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_PROBE diff --git a/arch/microblaze/include/asm/unistd.h b/arch/microblaze/include/asm/unistd.h index d20ffbc..6985e6e 100644 --- a/arch/microblaze/include/asm/unistd.h +++ b/arch/microblaze/include/asm/unistd.h @@ -400,7 +400,6 @@ #ifdef __KERNEL__ #ifndef __ASSEMBLY__ -#define __ARCH_WANT_IPC_PARSE_VERSION /* #define __ARCH_WANT_OLD_READDIR */ /* #define __ARCH_WANT_OLD_STAT */ #define __ARCH_WANT_STAT64 diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 5e238d0..2d56cd5 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -27,6 +27,7 @@ config MIPS select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW select HAVE_ARCH_JUMP_LABEL + select ARCH_WANT_IPC_PARSE_VERSION select IRQ_FORCED_THREADING select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP diff --git a/arch/mips/include/asm/unistd.h b/arch/mips/include/asm/unistd.h index d8dad53..bebbde0 100644 --- a/arch/mips/include/asm/unistd.h +++ b/arch/mips/include/asm/unistd.h @@ -1034,7 +1034,6 @@ #ifndef __ASSEMBLY__ #define __ARCH_OMIT_COMPAT_SYS_GETDENTS64 -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/mn10300/Kconfig b/arch/mn10300/Kconfig index 687f9b4..5cfb086 100644 --- a/arch/mn10300/Kconfig +++ b/arch/mn10300/Kconfig @@ -3,6 +3,7 @@ config MN10300 select HAVE_OPROFILE select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW + select ARCH_WANT_IPC_PARSE_VERSION select HAVE_ARCH_TRACEHOOK select HAVE_ARCH_KGDB select HAVE_NMI_WATCHDOG if MN10300_WD_TIMER diff --git a/arch/mn10300/include/asm/unistd.h b/arch/mn10300/include/asm/unistd.h index 9051f92..866eb14 100644 --- a/arch/mn10300/include/asm/unistd.h +++ b/arch/mn10300/include/asm/unistd.h @@ -358,7 +358,6 @@ /* * specify the deprecated syscalls we want to support on this arch */ -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT #define __ARCH_WANT_STAT64 diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 4f681b7..352f416 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -121,6 +121,7 @@ config PPC select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_HW_BREAKPOINT if PERF_EVENTS && PPC_BOOK3S_64 select HAVE_GENERIC_HARDIRQS + select ARCH_WANT_IPC_PARSE_VERSION select SPARSE_IRQ select IRQ_PER_CPU select IRQ_DOMAIN diff --git a/arch/powerpc/include/asm/unistd.h b/arch/powerpc/include/asm/unistd.h index d3d1b5e..bd377a3 100644 --- a/arch/powerpc/include/asm/unistd.h +++ b/arch/powerpc/include/asm/unistd.h @@ -389,7 +389,6 @@ #include #include -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index d0a5e92..296cd32 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -118,6 +118,7 @@ config S390 select ARCH_INLINE_WRITE_UNLOCK_BH select ARCH_INLINE_WRITE_UNLOCK_IRQ select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE + select ARCH_WANT_IPC_PARSE_VERSION select GENERIC_SMP_IDLE_THREAD select GENERIC_TIME_VSYSCALL select GENERIC_CLOCKEVENTS diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h index 2e37157..6756e78 100644 --- a/arch/s390/include/asm/unistd.h +++ b/arch/s390/include/asm/unistd.h @@ -388,7 +388,6 @@ #define __IGNORE_recvmmsg #define __IGNORE_sendmmsg -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_SYS_ALARM #define __ARCH_WANT_SYS_GETHOSTNAME diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index a24595d..36f5141 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -21,6 +21,7 @@ config SUPERH select HAVE_KERNEL_LZMA select HAVE_KERNEL_XZ select HAVE_KERNEL_LZO + select ARCH_WANT_IPC_PARSE_VERSION select HAVE_SYSCALL_TRACEPOINTS select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_GENERIC_HARDIRQS @@ -50,6 +51,7 @@ config SUPERH32 select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE + select ARCH_WANT_IPC_PARSE_VERSION select HAVE_FUNCTION_GRAPH_TRACER select HAVE_ARCH_KGDB select HAVE_HW_BREAKPOINT diff --git a/arch/sh/include/asm/unistd.h b/arch/sh/include/asm/unistd.h index e800a38..7bc6707 100644 --- a/arch/sh/include/asm/unistd.h +++ b/arch/sh/include/asm/unistd.h @@ -6,7 +6,6 @@ # endif # define __ARCH_WANT_SYS_RT_SIGSUSPEND -# define __ARCH_WANT_IPC_PARSE_VERSION # define __ARCH_WANT_OLD_READDIR # define __ARCH_WANT_OLD_STAT # define __ARCH_WANT_STAT64 diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index e74ff13..67f1f6f 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -27,6 +27,7 @@ config SPARC select HAVE_ARCH_JUMP_LABEL select HAVE_GENERIC_HARDIRQS select GENERIC_IRQ_SHOW + select ARCH_WANT_IPC_PARSE_VERSION select USE_GENERIC_SMP_HELPERS if SMP select GENERIC_PCI_IOMAP select HAVE_NMI_WATCHDOG if SPARC64 diff --git a/arch/sparc/include/asm/unistd.h b/arch/sparc/include/asm/unistd.h index c7cb0af..fb26934 100644 --- a/arch/sparc/include/asm/unistd.h +++ b/arch/sparc/include/asm/unistd.h @@ -423,7 +423,6 @@ #endif #ifdef __KERNEL__ -#define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_STAT64 #define __ARCH_WANT_SYS_ALARM diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ca4fdef..ba2657c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -85,6 +85,7 @@ config X86 select GENERIC_IOMAP select DCACHE_WORD_ACCESS select GENERIC_SMP_IDLE_THREAD + select ARCH_WANT_IPC_PARSE_VERSION if X86_32 select HAVE_ARCH_SECCOMP_FILTER select BUILDTIME_EXTABLE_SORT select GENERIC_CMOS_UPDATE diff --git a/arch/x86/include/asm/unistd.h b/arch/x86/include/asm/unistd.h index 4437001..0d9776e 100644 --- a/arch/x86/include/asm/unistd.h +++ b/arch/x86/include/asm/unistd.h @@ -15,7 +15,6 @@ # ifdef CONFIG_X86_32 # include -# define __ARCH_WANT_IPC_PARSE_VERSION # define __ARCH_WANT_STAT64 # define __ARCH_WANT_SYS_IPC # define __ARCH_WANT_SYS_OLD_MMAP diff --git a/include/linux/compat.h b/include/linux/compat.h index f2b8fe2..09b28b7 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -256,7 +256,6 @@ compat_sys_get_robust_list(int pid, compat_uptr_t __user *head_ptr, compat_size_t __user *len_ptr); #ifdef CONFIG_ARCH_WANT_OLD_COMPAT_IPC -#define __ARCH_WANT_COMPAT_IPC_PARSE_VERSION long compat_sys_semctl(int first, int second, int third, void __user *uptr); long compat_sys_msgsnd(int first, int second, int third, void __user *uptr); long compat_sys_msgrcv(int first, int second, int msgtyp, int third, diff --git a/ipc/compat.c b/ipc/compat.c index 20f92b2..ad9518e 100644 --- a/ipc/compat.c +++ b/ipc/compat.c @@ -118,7 +118,7 @@ extern int sem_ctls[]; static inline int compat_ipc_parse_version(int *cmd) { -#ifdef __ARCH_WANT_COMPAT_IPC_PARSE_VERSION +#ifdef CONFIG_ARCH_WANT_COMPAT_IPC_PARSE_VERSION int version = *cmd & IPC_64; /* this is tricky: architectures that have support for the old diff --git a/ipc/util.c b/ipc/util.c index 75261a3..eb07fd3 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -804,7 +804,7 @@ out_up: return ERR_PTR(err); } -#ifdef __ARCH_WANT_IPC_PARSE_VERSION +#ifdef CONFIG_ARCH_WANT_IPC_PARSE_VERSION /** @@ -826,7 +826,7 @@ int ipc_parse_version (int *cmd) } } -#endif /* __ARCH_WANT_IPC_PARSE_VERSION */ +#endif /* CONFIG_ARCH_WANT_IPC_PARSE_VERSION */ #ifdef CONFIG_PROC_FS struct ipc_proc_iter { diff --git a/ipc/util.h b/ipc/util.h index 6f5c20b..850ef3e 100644 --- a/ipc/util.h +++ b/ipc/util.h @@ -130,7 +130,7 @@ struct kern_ipc_perm *ipcctl_pre_down(struct ipc_namespace *ns, struct ipc_ids *ids, int id, int cmd, struct ipc64_perm *perm, int extra_perm); -#ifndef __ARCH_WANT_IPC_PARSE_VERSION +#ifndef CONFIG_ARCH_WANT_IPC_PARSE_VERSION /* On IA-64, we always use the "64-bit version" of the IPC structures. */ # define ipc_parse_version(cmd) IPC_64 #else -- cgit v0.10.2 From fd4b616b0fbb77e3f349e7d60914f2b7c7e39f9c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 30 Jul 2012 14:42:48 -0700 Subject: sysctl: suppress kmemleak messages register_sysctl_table() is a strange function, as it makes internal allocations (a header) to register a sysctl_table. This header is a handle to the table that is created, and can be used to unregister the table. But if the table is permanent and never unregistered, the header acts the same as a static variable. Unfortunately, this allocation of memory that is never expected to be freed fools kmemleak in thinking that we have leaked memory. For those sysctl tables that are never unregistered, and have no pointer referencing them, kmemleak will think that these are memory leaks: unreferenced object 0xffff880079fb9d40 (size 192): comm "swapper/0", pid 0, jiffies 4294667316 (age 12614.152s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x73/0x98 [] kmemleak_alloc_recursive.constprop.42+0x16/0x18 [] __kmalloc+0x107/0x153 [] kzalloc.constprop.8+0xe/0x10 [] __register_sysctl_paths+0xe1/0x160 [] register_sysctl_paths+0x1b/0x1d [] register_sysctl_table+0x18/0x1a [] sysctl_init+0x10/0x14 [] proc_sys_init+0x2f/0x31 [] proc_root_init+0xa5/0xa7 [] start_kernel+0x3d0/0x40a [] x86_64_start_reservations+0xae/0xb2 [] x86_64_start_kernel+0x102/0x111 [] 0xffffffffffffffff The sysctl_base_table used by sysctl itself is one such instance that registers the table to never be unregistered. Use kmemleak_not_leak() to suppress the kmemleak false positive. Signed-off-by: Steven Rostedt Acked-by: Catalin Marinas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/sysctl.c b/kernel/sysctl.c index b46f496..97186b9 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -1556,7 +1557,10 @@ static struct ctl_table dev_table[] = { int __init sysctl_init(void) { - register_sysctl_table(sysctl_base_table); + struct ctl_table_header *hdr; + + hdr = register_sysctl_table(sysctl_base_table); + kmemleak_not_leak(hdr); return 0; } -- cgit v0.10.2 From 25353b3377d5a75d4b830477bb90a3691155de72 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Jul 2012 14:42:49 -0700 Subject: taskstats: check nla_reserve() return Addresses https://bugzilla.kernel.org/show_bug.cgi?id=44621 Reported-by: Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/taskstats.c b/kernel/taskstats.c index e660464..d0a3279 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -436,6 +436,11 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) na = nla_reserve(rep_skb, CGROUPSTATS_TYPE_CGROUP_STATS, sizeof(struct cgroupstats)); + if (na == NULL) { + rc = -EMSGSIZE; + goto err; + } + stats = nla_data(na); memset(stats, 0, sizeof(*stats)); -- cgit v0.10.2 From 668f06b9fb846ecedb73964ca5dd95ce441707be Mon Sep 17 00:00:00 2001 From: Emil Goode Date: Mon, 30 Jul 2012 14:42:51 -0700 Subject: pps: return PTR_ERR on error in device_create We should return PTR_ERR if the call to the device_create function fails. Without this patch we instead return the value from a successful call to cdev_add if the call to device_create fails. Signed-off-by: Emil Goode Acked-by: Devendra Naga Cc: Alexander Gordeev Cc: Rodolfo Giometti Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/pps/pps.c b/drivers/pps/pps.c index 98fbe62..e771487 100644 --- a/drivers/pps/pps.c +++ b/drivers/pps/pps.c @@ -327,8 +327,10 @@ int pps_register_cdev(struct pps_device *pps) } pps->dev = device_create(pps_class, pps->info.dev, devt, pps, "pps%d", pps->id); - if (IS_ERR(pps->dev)) + if (IS_ERR(pps->dev)) { + err = PTR_ERR(pps->dev); goto del_cdev; + } pps->dev->release = pps_device_destruct; -- cgit v0.10.2 From 98c350cda2c14a343d34ea01a3d9c24fea5ec66d Mon Sep 17 00:00:00 2001 From: Justin Lecher Date: Mon, 30 Jul 2012 14:42:53 -0700 Subject: fs: cachefiles: add support for large files in filesystem caching Support the caching of large files. Addresses https://bugzilla.kernel.org/show_bug.cgi?id=31182 Signed-off-by: Justin Lecher Signed-off-by: Suresh Jayaraman Tested-by: Suresh Jayaraman Acked-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/cachefiles/rdwr.c b/fs/cachefiles/rdwr.c index c0353df..c994691 100644 --- a/fs/cachefiles/rdwr.c +++ b/fs/cachefiles/rdwr.c @@ -919,7 +919,7 @@ int cachefiles_write_page(struct fscache_storage *op, struct page *page) * own time */ path.mnt = cache->mnt; path.dentry = object->backer; - file = dentry_open(&path, O_RDWR, cache->cache_cred); + file = dentry_open(&path, O_RDWR | O_LARGEFILE, cache->cache_cred); if (IS_ERR(file)) { ret = PTR_ERR(file); } else { -- cgit v0.10.2 From f7e1becb078c2b996420a61f2a411ef19335e2da Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:42:56 -0700 Subject: include/linux/aio.h: cpp->C conversions Convert init_sync_kiocb() from a nasty macro into a nice C function. The struct assignment trick takes care of zeroing all unmentioned fields. Shrinks fs/read_write.o's .text from 9857 bytes to 9714. Also demacroize is_sync_kiocb() and aio_ring_avail(). The latter fixes an arg-referenced-multiple-times hand grenade. Cc: Junxiao Bi Cc: Mark Fasheh Acked-by: Jeff Moyer Cc: Joel Becker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/aio.h b/include/linux/aio.h index b1a520e..31ff6db 100644 --- a/include/linux/aio.h +++ b/include/linux/aio.h @@ -126,22 +126,20 @@ struct kiocb { struct eventfd_ctx *ki_eventfd; }; -#define is_sync_kiocb(iocb) ((iocb)->ki_key == KIOCB_SYNC_KEY) -#define init_sync_kiocb(x, filp) \ - do { \ - struct task_struct *tsk = current; \ - (x)->ki_flags = 0; \ - (x)->ki_users = 1; \ - (x)->ki_key = KIOCB_SYNC_KEY; \ - (x)->ki_filp = (filp); \ - (x)->ki_ctx = NULL; \ - (x)->ki_cancel = NULL; \ - (x)->ki_retry = NULL; \ - (x)->ki_dtor = NULL; \ - (x)->ki_obj.tsk = tsk; \ - (x)->ki_user_data = 0; \ - (x)->private = NULL; \ - } while (0) +static inline bool is_sync_kiocb(struct kiocb *kiocb) +{ + return kiocb->ki_key == KIOCB_SYNC_KEY; +} + +static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp) +{ + *kiocb = (struct kiocb) { + .ki_users = 1, + .ki_key = KIOCB_SYNC_KEY, + .ki_filp = filp, + .ki_obj.tsk = current, + }; +} #define AIO_RING_MAGIC 0xa10a10a1 #define AIO_RING_COMPAT_FEATURES 1 @@ -161,8 +159,6 @@ struct aio_ring { struct io_event io_events[0]; }; /* 128 bytes + ring size */ -#define aio_ring_avail(info, ring) (((ring)->head + (info)->nr - 1 - (ring)->tail) % (info)->nr) - #define AIO_RING_PAGES 8 struct aio_ring_info { unsigned long mmap_base; @@ -177,6 +173,12 @@ struct aio_ring_info { struct page *internal_pages[AIO_RING_PAGES]; }; +static inline unsigned aio_ring_avail(struct aio_ring_info *info, + struct aio_ring *ring) +{ + return (ring->head + info->nr - 1 - ring->tail) % info->nr; +} + struct kioctx { atomic_t users; int dead; -- cgit v0.10.2 From 65fed8f6f23070b56d0ed3841173ddd410130a89 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Mon, 30 Jul 2012 14:42:58 -0700 Subject: resource: make sure requested range is included in the root range When the requested range is outside of the root range the logic in __reserve_region_with_split will cause an infinite recursion which will overflow the stack as seen in the warning bellow. This particular stack overflow was caused by requesting the (100000000-107ffffff) range while the root range was (0-ffffffff). In this case __request_resource would return the whole root range as conflict range (i.e. 0-ffffffff). Then, the logic in __reserve_region_with_split would continue the recursion requesting the new range as (conflict->end+1, end) which incidentally in this case equals the originally requested range. This patch aborts looking for an usable range when the request does not intersect with the root range. When the request partially overlaps with the root range, it ajust the request to fall in the root range and then continues with the new request. When the request is modified or aborted errors and a stack trace are logged to allow catching the errors in the upper layers. [ 5.968374] WARNING: at kernel/sched.c:4129 sub_preempt_count+0x63/0x89() [ 5.975150] Modules linked in: [ 5.978184] Pid: 1, comm: swapper Not tainted 3.0.22-mid27-00004-gb72c817 #46 [ 5.985324] Call Trace: [ 5.987759] [] ? console_unlock+0x17b/0x18d [ 5.992891] [] warn_slowpath_common+0x48/0x5d [ 5.998194] [] ? sub_preempt_count+0x63/0x89 [ 6.003412] [] warn_slowpath_null+0xf/0x13 [ 6.008453] [] sub_preempt_count+0x63/0x89 [ 6.013499] [] _raw_spin_unlock+0x27/0x3f [ 6.018453] [] add_partial+0x36/0x3b [ 6.022973] [] deactivate_slab+0x96/0xb4 [ 6.027842] [] __slab_alloc.isra.54.constprop.63+0x204/0x241 [ 6.034456] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.039842] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.045232] [] kmem_cache_alloc_trace+0x51/0xb0 [ 6.050710] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.056100] [] kzalloc.constprop.5+0x29/0x38 [ 6.061320] [] __reserve_region_with_split+0x1c/0xd1 [ 6.067230] [] __reserve_region_with_split+0xc6/0xd1 ... [ 7.179057] [] __reserve_region_with_split+0xc6/0xd1 [ 7.184970] [] reserve_region_with_split+0x30/0x42 [ 7.190709] [] e820_reserve_resources_late+0xd1/0xe9 [ 7.196623] [] pcibios_resource_survey+0x23/0x2a [ 7.202184] [] pcibios_init+0x23/0x35 [ 7.206789] [] pci_subsys_init+0x3f/0x44 [ 7.211659] [] do_one_initcall+0x72/0x122 [ 7.216615] [] ? pci_legacy_init+0x3d/0x3d [ 7.221659] [] kernel_init+0xa6/0x118 [ 7.226265] [] ? start_kernel+0x334/0x334 [ 7.231223] [] kernel_thread_helper+0x6/0x10 Signed-off-by: Octavian Purdila Signed-off-by: Ram Pai Cc: Jesse Barnes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/resource.c b/kernel/resource.c index dc8b477..34d4588 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -7,6 +7,8 @@ * Arbitrary resource management. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -791,8 +793,28 @@ void __init reserve_region_with_split(struct resource *root, resource_size_t start, resource_size_t end, const char *name) { + int abort = 0; + write_lock(&resource_lock); - __reserve_region_with_split(root, start, end, name); + if (root->start > start || root->end < end) { + pr_err("requested range [0x%llx-0x%llx] not in root %pr\n", + (unsigned long long)start, (unsigned long long)end, + root); + if (start > root->end || end < root->start) + abort = 1; + else { + if (end > root->end) + end = root->end; + if (start < root->start) + start = root->start; + pr_err("fixing request to [0x%llx-0x%llx]\n", + (unsigned long long)start, + (unsigned long long)end); + } + dump_stack(); + } + if (!abort) + __reserve_region_with_split(root, start, end, name); write_unlock(&resource_lock); } -- cgit v0.10.2 From 1d151c337d79fa3de88654d2514f58fbd916a8e0 Mon Sep 17 00:00:00 2001 From: Cyrill Gorcunov Date: Mon, 30 Jul 2012 14:43:00 -0700 Subject: c/r: fcntl: add F_GETOWNER_UIDS option When we restore file descriptors we would like them to look exactly as they were at dumping time. With help of fcntl it's almost possible, the missing snippet is file owners UIDs. To be able to read their values the F_GETOWNER_UIDS is introduced. This option is valid iif CONFIG_CHECKPOINT_RESTORE is turned on, otherwise returning -EINVAL. Signed-off-by: Cyrill Gorcunov Acked-by: "Eric W. Biederman" Cc: "Serge E. Hallyn" Cc: Oleg Nesterov Cc: Pavel Emelyanov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fcntl.c b/fs/fcntl.c index 81b70e6..887b5ba 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -340,6 +341,31 @@ static int f_getown_ex(struct file *filp, unsigned long arg) return ret; } +#ifdef CONFIG_CHECKPOINT_RESTORE +static int f_getowner_uids(struct file *filp, unsigned long arg) +{ + struct user_namespace *user_ns = current_user_ns(); + uid_t * __user dst = (void * __user)arg; + uid_t src[2]; + int err; + + read_lock(&filp->f_owner.lock); + src[0] = from_kuid(user_ns, filp->f_owner.uid); + src[1] = from_kuid(user_ns, filp->f_owner.euid); + read_unlock(&filp->f_owner.lock); + + err = put_user(src[0], &dst[0]); + err |= put_user(src[1], &dst[1]); + + return err; +} +#else +static int f_getowner_uids(struct file *filp, unsigned long arg) +{ + return -EINVAL; +} +#endif + static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, struct file *filp) { @@ -396,6 +422,9 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, case F_SETOWN_EX: err = f_setown_ex(filp, arg); break; + case F_GETOWNER_UIDS: + err = f_getowner_uids(filp, arg); + break; case F_GETSIG: err = filp->f_owner.signum; break; diff --git a/include/asm-generic/fcntl.h b/include/asm-generic/fcntl.h index 9e5b035..a48937d 100644 --- a/include/asm-generic/fcntl.h +++ b/include/asm-generic/fcntl.h @@ -120,6 +120,10 @@ #define F_GETOWN_EX 16 #endif +#ifndef F_GETOWNER_UIDS +#define F_GETOWNER_UIDS 17 +#endif + #define F_OWNER_TID 0 #define F_OWNER_PID 1 #define F_OWNER_PGRP 2 diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 94c45a1..ec43760 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -3180,6 +3180,7 @@ static int selinux_file_fcntl(struct file *file, unsigned int cmd, case F_GETFL: case F_GETOWN: case F_GETSIG: + case F_GETOWNER_UIDS: /* Just check FD__USE permission */ err = file_has_perm(cred, file, 0); break; -- cgit v0.10.2 From 8d438288145f19f253a82ca71290b44fce79e23f Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:02 -0700 Subject: fault-injection: notifier error injection This patchset provides kernel modules that can be used to test the error handling of notifier call chain failures by injecting artifical errors to the following notifier chain callbacks. * CPU notifier * PM notifier * memory hotplug notifier * powerpc pSeries reconfig notifier Example: Inject CPU offline error (-1 == -EPERM) # cd /sys/kernel/debug/notifier-error-inject/cpu # echo -1 > actions/CPU_DOWN_PREPARE/error # echo 0 > /sys/devices/system/cpu/cpu1/online bash: echo: write error: Operation not permitted The patchset also adds cpu and memory hotplug tests to tools/testing/selftests These tests first do simple online and offline test and then do fault injection tests if notifier error injection module is available. This patch: The notifier error injection provides the ability to inject artifical errors to specified notifier chain callbacks. It is useful to test the error handling of notifier call chain failures. This adds common basic functions to define which type of events can be fail and to initialize the debugfs interface to control what error code should be returned and which event should be failed. Signed-off-by: Akinobu Mita Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/fault-injection/notifier-error-inject.txt b/Documentation/fault-injection/notifier-error-inject.txt new file mode 100644 index 0000000..c83526c --- /dev/null +++ b/Documentation/fault-injection/notifier-error-inject.txt @@ -0,0 +1,99 @@ +Notifier error injection +======================== + +Notifier error injection provides the ability to inject artifical errors to +specified notifier chain callbacks. It is useful to test the error handling of +notifier call chain failures which is rarely executed. There are kernel +modules that can be used to test the following notifiers. + + * CPU notifier + * PM notifier + * Memory hotplug notifier + * powerpc pSeries reconfig notifier + +CPU notifier error injection module +----------------------------------- +This feature can be used to test the error handling of the CPU notifiers by +injecting artifical errors to CPU notifier chain callbacks. + +If the notifier call chain should be failed with some events notified, write +the error code to debugfs interface +/sys/kernel/debug/notifier-error-inject/cpu/actions//error + +Possible CPU notifier events to be failed are: + + * CPU_UP_PREPARE + * CPU_UP_PREPARE_FROZEN + * CPU_DOWN_PREPARE + * CPU_DOWN_PREPARE_FROZEN + +Example1: Inject CPU offline error (-1 == -EPERM) + + # cd /sys/kernel/debug/notifier-error-inject/cpu + # echo -1 > actions/CPU_DOWN_PREPARE/error + # echo 0 > /sys/devices/system/cpu/cpu1/online + bash: echo: write error: Operation not permitted + +Example2: inject CPU online error (-2 == -ENOENT) + + # echo -2 > actions/CPU_UP_PREPARE/error + # echo 1 > /sys/devices/system/cpu/cpu1/online + bash: echo: write error: No such file or directory + +PM notifier error injection module +---------------------------------- +This feature is controlled through debugfs interface +/sys/kernel/debug/notifier-error-inject/pm/actions//error + +Possible PM notifier events to be failed are: + + * PM_HIBERNATION_PREPARE + * PM_SUSPEND_PREPARE + * PM_RESTORE_PREPARE + +Example: Inject PM suspend error (-12 = -ENOMEM) + + # cd /sys/kernel/debug/notifier-error-inject/pm/ + # echo -12 > actions/PM_SUSPEND_PREPARE/error + # echo mem > /sys/power/state + bash: echo: write error: Cannot allocate memory + +Memory hotplug notifier error injection module +---------------------------------------------- +This feature is controlled through debugfs interface +/sys/kernel/debug/notifier-error-inject/memory/actions//error + +Possible memory notifier events to be failed are: + + * MEM_GOING_ONLINE + * MEM_GOING_OFFLINE + +Example: Inject memory hotplug offline error (-12 == -ENOMEM) + + # cd /sys/kernel/debug/notifier-error-inject/memory + # echo -12 > actions/MEM_GOING_OFFLINE/error + # echo offline > /sys/devices/system/memory/memoryXXX/state + bash: echo: write error: Cannot allocate memory + +powerpc pSeries reconfig notifier error injection module +-------------------------------------------------------- +This feature is controlled through debugfs interface +/sys/kernel/debug/notifier-error-inject/pSeries-reconfig/actions//error + +Possible pSeries reconfig notifier events to be failed are: + + * PSERIES_RECONFIG_ADD + * PSERIES_RECONFIG_REMOVE + * PSERIES_DRCONF_MEM_ADD + * PSERIES_DRCONF_MEM_REMOVE + +For more usage examples +----------------------- +There are tools/testing/selftests using the notifier error injection features +for CPU and memory notifiers. + + * tools/testing/selftests/cpu-hotplug/on-off-test.sh + * tools/testing/selftests/memory-hotplug/on-off-test.sh + +These scripts first do simple online and offline tests and then do fault +injection tests if notifier error injection module is available. diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 4a18650..bdee91d 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1084,6 +1084,17 @@ config LKDTM Documentation on how to use the module can be found in Documentation/fault-injection/provoke-crashes.txt +config NOTIFIER_ERROR_INJECTION + tristate "Notifier error injection" + depends on DEBUG_KERNEL + select DEBUG_FS + help + This option provides the ability to inject artifical errors to + specified notifier chain callbacks. It is useful to test the error + handling of notifier call chain failures. + + Say N if unsure. + config CPU_NOTIFIER_ERROR_INJECT tristate "CPU notifier error injection module" depends on HOTPLUG_CPU && DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index df663cc..71acfa44 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -90,6 +90,7 @@ obj-$(CONFIG_AUDIT_GENERIC) += audit.o obj-$(CONFIG_SWIOTLB) += swiotlb.o obj-$(CONFIG_IOMMU_HELPER) += iommu-helper.o obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o +obj-$(CONFIG_NOTIFIER_ERROR_INJECTION) += notifier-error-inject.o obj-$(CONFIG_CPU_NOTIFIER_ERROR_INJECT) += cpu-notifier-error-inject.o lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/lib/notifier-error-inject.c b/lib/notifier-error-inject.c new file mode 100644 index 0000000..44b92cb --- /dev/null +++ b/lib/notifier-error-inject.c @@ -0,0 +1,112 @@ +#include + +#include "notifier-error-inject.h" + +static int debugfs_errno_set(void *data, u64 val) +{ + *(int *)data = clamp_t(int, val, -MAX_ERRNO, 0); + return 0; +} + +static int debugfs_errno_get(void *data, u64 *val) +{ + *val = *(int *)data; + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(fops_errno, debugfs_errno_get, debugfs_errno_set, + "%lld\n"); + +static struct dentry *debugfs_create_errno(const char *name, mode_t mode, + struct dentry *parent, int *value) +{ + return debugfs_create_file(name, mode, parent, value, &fops_errno); +} + +static int notifier_err_inject_callback(struct notifier_block *nb, + unsigned long val, void *p) +{ + int err = 0; + struct notifier_err_inject *err_inject = + container_of(nb, struct notifier_err_inject, nb); + struct notifier_err_inject_action *action; + + for (action = err_inject->actions; action->name; action++) { + if (action->val == val) { + err = action->error; + break; + } + } + if (err) + pr_info("Injecting error (%d) to %s\n", err, action->name); + + return notifier_from_errno(err); +} + +struct dentry *notifier_err_inject_dir; +EXPORT_SYMBOL_GPL(notifier_err_inject_dir); + +struct dentry *notifier_err_inject_init(const char *name, struct dentry *parent, + struct notifier_err_inject *err_inject, int priority) +{ + struct notifier_err_inject_action *action; + mode_t mode = S_IFREG | S_IRUSR | S_IWUSR; + struct dentry *dir; + struct dentry *actions_dir; + + err_inject->nb.notifier_call = notifier_err_inject_callback; + err_inject->nb.priority = priority; + + dir = debugfs_create_dir(name, parent); + if (!dir) + return ERR_PTR(-ENOMEM); + + actions_dir = debugfs_create_dir("actions", dir); + if (!actions_dir) + goto fail; + + for (action = err_inject->actions; action->name; action++) { + struct dentry *action_dir; + + action_dir = debugfs_create_dir(action->name, actions_dir); + if (!action_dir) + goto fail; + + /* + * Create debugfs r/w file containing action->error. If + * notifier call chain is called with action->val, it will + * fail with the error code + */ + if (!debugfs_create_errno("error", mode, action_dir, + &action->error)) + goto fail; + } + return dir; +fail: + debugfs_remove_recursive(dir); + return ERR_PTR(-ENOMEM); +} +EXPORT_SYMBOL_GPL(notifier_err_inject_init); + +static int __init err_inject_init(void) +{ + notifier_err_inject_dir = + debugfs_create_dir("notifier-error-inject", NULL); + + if (!notifier_err_inject_dir) + return -ENOMEM; + + return 0; +} + +static void __exit err_inject_exit(void) +{ + debugfs_remove_recursive(notifier_err_inject_dir); +} + +module_init(err_inject_init); +module_exit(err_inject_exit); + +MODULE_DESCRIPTION("Notifier error injection module"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Akinobu Mita "); diff --git a/lib/notifier-error-inject.h b/lib/notifier-error-inject.h new file mode 100644 index 0000000..99b3b6f --- /dev/null +++ b/lib/notifier-error-inject.h @@ -0,0 +1,24 @@ +#include +#include +#include + +struct notifier_err_inject_action { + unsigned long val; + int error; + const char *name; +}; + +#define NOTIFIER_ERR_INJECT_ACTION(action) \ + .name = #action, .val = (action), + +struct notifier_err_inject { + struct notifier_block nb; + struct notifier_err_inject_action actions[]; + /* The last slot must be terminated with zero sentinel */ +}; + +extern struct dentry *notifier_err_inject_dir; + +extern struct dentry *notifier_err_inject_init(const char *name, + struct dentry *parent, struct notifier_err_inject *err_inject, + int priority); -- cgit v0.10.2 From f5a9f52e2c5654c3d212dbf7e7a169c60876691a Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:03 -0700 Subject: cpu: rewrite cpu-notifier-error-inject module Rewrite existing cpu-notifier-error-inject module to use debugfs based new framework. This change removes cpu_up_prepare_error and cpu_down_prepare_error module parameters which were used to specify error code to be injected. We could keep these module parameters for backward compatibility by module_param_cb but it seems overkill for this module. This provides the ability to inject artifical errors to CPU notifier chain callbacks. It is controlled through debugfs interface under /sys/kernel/debug/notifier-error-inject/cpu If the notifier call chain should be failed with some events notified, write the error code to "actions//error". Example1: inject CPU offline error (-1 == -EPERM) # cd /sys/kernel/debug/notifier-error-inject/cpu # echo -1 > actions/CPU_DOWN_PREPARE/error # echo 0 > /sys/devices/system/cpu/cpu1/online bash: echo: write error: Operation not permitted Example2: inject CPU online error (-2 == -ENOENT) # cd /sys/kernel/debug/notifier-error-inject/cpu # echo -2 > actions/CPU_UP_PREPARE/error # echo 1 > /sys/devices/system/cpu/cpu1/online bash: echo: write error: No such file or directory Signed-off-by: Akinobu Mita Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index bdee91d..2034148 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1097,10 +1097,22 @@ config NOTIFIER_ERROR_INJECTION config CPU_NOTIFIER_ERROR_INJECT tristate "CPU notifier error injection module" - depends on HOTPLUG_CPU && DEBUG_KERNEL + depends on HOTPLUG_CPU && NOTIFIER_ERROR_INJECTION help This option provides a kernel module that can be used to test - the error handling of the cpu notifiers + the error handling of the cpu notifiers by injecting artifical + errors to CPU notifier chain callbacks. It is controlled through + debugfs interface under /sys/kernel/debug/notifier-error-inject/cpu + + If the notifier call chain should be failed with some events + notified, write the error code to "actions//error". + + Example: Inject CPU offline error (-1 == -EPERM) + + # cd /sys/kernel/debug/notifier-error-inject/cpu + # echo -1 > actions/CPU_DOWN_PREPARE/error + # echo 0 > /sys/devices/system/cpu/cpu1/online + bash: echo: write error: Operation not permitted To compile this code as a module, choose M here: the module will be called cpu-notifier-error-inject. diff --git a/lib/cpu-notifier-error-inject.c b/lib/cpu-notifier-error-inject.c index 4dc2032..707ca24 100644 --- a/lib/cpu-notifier-error-inject.c +++ b/lib/cpu-notifier-error-inject.c @@ -1,58 +1,45 @@ #include -#include #include -#include +#include -static int priority; -static int cpu_up_prepare_error; -static int cpu_down_prepare_error; +#include "notifier-error-inject.h" +static int priority; module_param(priority, int, 0); MODULE_PARM_DESC(priority, "specify cpu notifier priority"); -module_param(cpu_up_prepare_error, int, 0644); -MODULE_PARM_DESC(cpu_up_prepare_error, - "specify error code to inject CPU_UP_PREPARE action"); - -module_param(cpu_down_prepare_error, int, 0644); -MODULE_PARM_DESC(cpu_down_prepare_error, - "specify error code to inject CPU_DOWN_PREPARE action"); - -static int err_inject_cpu_callback(struct notifier_block *nfb, - unsigned long action, void *hcpu) -{ - int err = 0; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - err = cpu_up_prepare_error; - break; - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - err = cpu_down_prepare_error; - break; +static struct notifier_err_inject cpu_notifier_err_inject = { + .actions = { + { NOTIFIER_ERR_INJECT_ACTION(CPU_UP_PREPARE) }, + { NOTIFIER_ERR_INJECT_ACTION(CPU_UP_PREPARE_FROZEN) }, + { NOTIFIER_ERR_INJECT_ACTION(CPU_DOWN_PREPARE) }, + { NOTIFIER_ERR_INJECT_ACTION(CPU_DOWN_PREPARE_FROZEN) }, + {} } - if (err) - printk(KERN_INFO "Injecting error (%d) at cpu notifier\n", err); - - return notifier_from_errno(err); -} - -static struct notifier_block err_inject_cpu_notifier = { - .notifier_call = err_inject_cpu_callback, }; +static struct dentry *dir; + static int err_inject_init(void) { - err_inject_cpu_notifier.priority = priority; + int err; + + dir = notifier_err_inject_init("cpu", notifier_err_inject_dir, + &cpu_notifier_err_inject, priority); + if (IS_ERR(dir)) + return PTR_ERR(dir); + + err = register_hotcpu_notifier(&cpu_notifier_err_inject.nb); + if (err) + debugfs_remove_recursive(dir); - return register_hotcpu_notifier(&err_inject_cpu_notifier); + return err; } static void err_inject_exit(void) { - unregister_hotcpu_notifier(&err_inject_cpu_notifier); + unregister_hotcpu_notifier(&cpu_notifier_err_inject.nb); + debugfs_remove_recursive(dir); } module_init(err_inject_init); -- cgit v0.10.2 From 048b9c3549790af21eabd06a5ebdad305e75b1c5 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:07 -0700 Subject: PM: PM notifier error injection module This provides the ability to inject artifical errors to PM notifier chain callbacks. It is controlled through debugfs interface under /sys/kernel/debug/notifier-error-inject/pm Each of the files in "error" directory represents an event which can be failed and contains the error code. If the notifier call chain should be failed with some events notified, write the error code to the files. If the notifier call chain should be failed with some events notified, write the error code to "actions//error". Example: Inject PM suspend error (-12 = -ENOMEM) # cd /sys/kernel/debug/notifier-error-inject/pm # echo -12 > actions/PM_SUSPEND_PREPARE/error # echo mem > /sys/power/state bash: echo: write error: Cannot allocate memory Signed-off-by: Akinobu Mita Acked-by: "Rafael J. Wysocki" Cc: Pavel Machek Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 2034148..c2024d8 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1119,6 +1119,30 @@ config CPU_NOTIFIER_ERROR_INJECT If unsure, say N. +config PM_NOTIFIER_ERROR_INJECT + tristate "PM notifier error injection module" + depends on PM && NOTIFIER_ERROR_INJECTION + default m if PM_DEBUG + help + This option provides the ability to inject artifical errors to + PM notifier chain callbacks. It is controlled through debugfs + interface /sys/kernel/debug/notifier-error-inject/pm + + If the notifier call chain should be failed with some events + notified, write the error code to "actions//error". + + Example: Inject PM suspend error (-12 = -ENOMEM) + + # cd /sys/kernel/debug/notifier-error-inject/pm/ + # echo -12 > actions/PM_SUSPEND_PREPARE/error + # echo mem > /sys/power/state + bash: echo: write error: Cannot allocate memory + + To compile this code as a module, choose M here: the module will + be called pm-notifier-error-inject. + + If unsure, say N. + config FAULT_INJECTION bool "Fault-injection framework" depends on DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index 71acfa44..c32c664 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -92,6 +92,7 @@ obj-$(CONFIG_IOMMU_HELPER) += iommu-helper.o obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o obj-$(CONFIG_NOTIFIER_ERROR_INJECTION) += notifier-error-inject.o obj-$(CONFIG_CPU_NOTIFIER_ERROR_INJECT) += cpu-notifier-error-inject.o +obj-$(CONFIG_PM_NOTIFIER_ERROR_INJECT) += pm-notifier-error-inject.o lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/lib/pm-notifier-error-inject.c b/lib/pm-notifier-error-inject.c new file mode 100644 index 0000000..c094b2d --- /dev/null +++ b/lib/pm-notifier-error-inject.c @@ -0,0 +1,49 @@ +#include +#include +#include + +#include "notifier-error-inject.h" + +static int priority; +module_param(priority, int, 0); +MODULE_PARM_DESC(priority, "specify PM notifier priority"); + +static struct notifier_err_inject pm_notifier_err_inject = { + .actions = { + { NOTIFIER_ERR_INJECT_ACTION(PM_HIBERNATION_PREPARE) }, + { NOTIFIER_ERR_INJECT_ACTION(PM_SUSPEND_PREPARE) }, + { NOTIFIER_ERR_INJECT_ACTION(PM_RESTORE_PREPARE) }, + {} + } +}; + +static struct dentry *dir; + +static int err_inject_init(void) +{ + int err; + + dir = notifier_err_inject_init("pm", notifier_err_inject_dir, + &pm_notifier_err_inject, priority); + if (IS_ERR(dir)) + return PTR_ERR(dir); + + err = register_pm_notifier(&pm_notifier_err_inject.nb); + if (err) + debugfs_remove_recursive(dir); + + return err; +} + +static void err_inject_exit(void) +{ + unregister_pm_notifier(&pm_notifier_err_inject.nb); + debugfs_remove_recursive(dir); +} + +module_init(err_inject_init); +module_exit(err_inject_exit); + +MODULE_DESCRIPTION("PM notifier error injection module"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Akinobu Mita "); -- cgit v0.10.2 From 9579f5bd31a04e80a87a7b58bd52dff6dc68bc99 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:10 -0700 Subject: memory: memory notifier error injection module This provides the ability to inject artifical errors to memory hotplug notifier chain callbacks. It is controlled through debugfs interface under /sys/kernel/debug/notifier-error-inject/memory If the notifier call chain should be failed with some events notified, write the error code to "actions//error". Example: Inject memory hotplug offline error (-12 == -ENOMEM) # cd /sys/kernel/debug/notifier-error-inject/memory # echo -12 > actions/MEM_GOING_OFFLINE/error # echo offline > /sys/devices/system/memory/memoryXXX/state bash: echo: write error: Cannot allocate memory Signed-off-by: Akinobu Mita Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index c2024d8..35cfd9e 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1143,6 +1143,29 @@ config PM_NOTIFIER_ERROR_INJECT If unsure, say N. +config MEMORY_NOTIFIER_ERROR_INJECT + tristate "Memory hotplug notifier error injection module" + depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION + help + This option provides the ability to inject artifical errors to + memory hotplug notifier chain callbacks. It is controlled through + debugfs interface under /sys/kernel/debug/notifier-error-inject/memory + + If the notifier call chain should be failed with some events + notified, write the error code to "actions//error". + + Example: Inject memory hotplug offline error (-12 == -ENOMEM) + + # cd /sys/kernel/debug/notifier-error-inject/memory + # echo -12 > actions/MEM_GOING_OFFLINE/error + # echo offline > /sys/devices/system/memory/memoryXXX/state + bash: echo: write error: Cannot allocate memory + + To compile this code as a module, choose M here: the module will + be called pSeries-reconfig-notifier-error-inject. + + If unsure, say N. + config FAULT_INJECTION bool "Fault-injection framework" depends on DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index c32c664..c347193 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -93,6 +93,7 @@ obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o obj-$(CONFIG_NOTIFIER_ERROR_INJECTION) += notifier-error-inject.o obj-$(CONFIG_CPU_NOTIFIER_ERROR_INJECT) += cpu-notifier-error-inject.o obj-$(CONFIG_PM_NOTIFIER_ERROR_INJECT) += pm-notifier-error-inject.o +obj-$(CONFIG_MEMORY_NOTIFIER_ERROR_INJECT) += memory-notifier-error-inject.o lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/lib/memory-notifier-error-inject.c b/lib/memory-notifier-error-inject.c new file mode 100644 index 0000000..e6239bf --- /dev/null +++ b/lib/memory-notifier-error-inject.c @@ -0,0 +1,48 @@ +#include +#include +#include + +#include "notifier-error-inject.h" + +static int priority; +module_param(priority, int, 0); +MODULE_PARM_DESC(priority, "specify memory notifier priority"); + +static struct notifier_err_inject memory_notifier_err_inject = { + .actions = { + { NOTIFIER_ERR_INJECT_ACTION(MEM_GOING_ONLINE) }, + { NOTIFIER_ERR_INJECT_ACTION(MEM_GOING_OFFLINE) }, + {} + } +}; + +static struct dentry *dir; + +static int err_inject_init(void) +{ + int err; + + dir = notifier_err_inject_init("memory", notifier_err_inject_dir, + &memory_notifier_err_inject, priority); + if (IS_ERR(dir)) + return PTR_ERR(dir); + + err = register_memory_notifier(&memory_notifier_err_inject.nb); + if (err) + debugfs_remove_recursive(dir); + + return err; +} + +static void err_inject_exit(void) +{ + unregister_memory_notifier(&memory_notifier_err_inject.nb); + debugfs_remove_recursive(dir); +} + +module_init(err_inject_init); +module_exit(err_inject_exit); + +MODULE_DESCRIPTION("memory notifier error injection module"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Akinobu Mita "); -- cgit v0.10.2 From 08dfb4ddeeeebdee4f3d5a08a87dc9aa68d26f81 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:13 -0700 Subject: powerpc: pSeries reconfig notifier error injection module This provides the ability to inject artifical errors to pSeries reconfig notifier chain callbacks. It is controlled through debugfs interface under /sys/kernel/debug/notifier-error-inject/pSeries-reconfig If the notifier call chain should be failed with some events notified, write the error code to "actions//error". Signed-off-by: Akinobu Mita Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 35cfd9e..2403a63 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1166,6 +1166,23 @@ config MEMORY_NOTIFIER_ERROR_INJECT If unsure, say N. +config PSERIES_RECONFIG_NOTIFIER_ERROR_INJECT + tristate "pSeries reconfig notifier error injection module" + depends on PPC_PSERIES && NOTIFIER_ERROR_INJECTION + help + This option provides the ability to inject artifical errors to + pSeries reconfig notifier chain callbacks. It is controlled + through debugfs interface under + /sys/kernel/debug/notifier-error-inject/pSeries-reconfig/ + + If the notifier call chain should be failed with some events + notified, write the error code to "actions//error". + + To compile this code as a module, choose M here: the module will + be called memory-notifier-error-inject. + + If unsure, say N. + config FAULT_INJECTION bool "Fault-injection framework" depends on DEBUG_KERNEL diff --git a/lib/Makefile b/lib/Makefile index c347193..1054de8 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -94,6 +94,8 @@ obj-$(CONFIG_NOTIFIER_ERROR_INJECTION) += notifier-error-inject.o obj-$(CONFIG_CPU_NOTIFIER_ERROR_INJECT) += cpu-notifier-error-inject.o obj-$(CONFIG_PM_NOTIFIER_ERROR_INJECT) += pm-notifier-error-inject.o obj-$(CONFIG_MEMORY_NOTIFIER_ERROR_INJECT) += memory-notifier-error-inject.o +obj-$(CONFIG_PSERIES_RECONFIG_NOTIFIER_ERROR_INJECT) += \ + pSeries-reconfig-notifier-error-inject.o lib-$(CONFIG_GENERIC_BUG) += bug.o diff --git a/lib/pSeries-reconfig-notifier-error-inject.c b/lib/pSeries-reconfig-notifier-error-inject.c new file mode 100644 index 0000000..7f7c98d --- /dev/null +++ b/lib/pSeries-reconfig-notifier-error-inject.c @@ -0,0 +1,51 @@ +#include +#include + +#include + +#include "notifier-error-inject.h" + +static int priority; +module_param(priority, int, 0); +MODULE_PARM_DESC(priority, "specify pSeries reconfig notifier priority"); + +static struct notifier_err_inject reconfig_err_inject = { + .actions = { + { NOTIFIER_ERR_INJECT_ACTION(PSERIES_RECONFIG_ADD) }, + { NOTIFIER_ERR_INJECT_ACTION(PSERIES_RECONFIG_REMOVE) }, + { NOTIFIER_ERR_INJECT_ACTION(PSERIES_DRCONF_MEM_ADD) }, + { NOTIFIER_ERR_INJECT_ACTION(PSERIES_DRCONF_MEM_REMOVE) }, + {} + } +}; + +static struct dentry *dir; + +static int err_inject_init(void) +{ + int err; + + dir = notifier_err_inject_init("pSeries-reconfig", + notifier_err_inject_dir, &reconfig_err_inject, priority); + if (IS_ERR(dir)) + return PTR_ERR(dir); + + err = pSeries_reconfig_notifier_register(&reconfig_err_inject.nb); + if (err) + debugfs_remove_recursive(dir); + + return err; +} + +static void err_inject_exit(void) +{ + pSeries_reconfig_notifier_unregister(&reconfig_err_inject.nb); + debugfs_remove_recursive(dir); +} + +module_init(err_inject_init); +module_exit(err_inject_exit); + +MODULE_DESCRIPTION("pSeries reconfig notifier error injection module"); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Akinobu Mita "); -- cgit v0.10.2 From d89dffa976bcd13fd87eb76e02e3b71c3a7868e3 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:17 -0700 Subject: fault-injection: add selftests for cpu and memory hotplug This adds two selftests * tools/testing/selftests/cpu-hotplug/on-off-test.sh is testing script for CPU hotplug 1. Online all hot-pluggable CPUs 2. Offline all hot-pluggable CPUs 3. Online all hot-pluggable CPUs again 4. Exit if cpu-notifier-error-inject.ko is not available 5. Offline all hot-pluggable CPUs in preparation for testing 6. Test CPU hot-add error handling by injecting notifier errors 7. Online all hot-pluggable CPUs in preparation for testing 8. Test CPU hot-remove error handling by injecting notifier errors * tools/testing/selftests/memory-hotplug/on-off-test.sh is doing the similar thing for memory hotplug. 1. Online all hot-pluggable memory 2. Offline 10% of hot-pluggable memory 3. Online all hot-pluggable memory again 4. Exit if memory-notifier-error-inject.ko is not available 5. Offline 10% of hot-pluggable memory in preparation for testing 6. Test memory hot-add error handling by injecting notifier errors 7. Online all hot-pluggable memory in preparation for testing 8. Test memory hot-remove error handling by injecting notifier errors Signed-off-by: Akinobu Mita Suggested-by: Andrew Morton Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Greg KH Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile index a4162e1..85baf11 100644 --- a/tools/testing/selftests/Makefile +++ b/tools/testing/selftests/Makefile @@ -1,4 +1,4 @@ -TARGETS = breakpoints kcmp mqueue vm +TARGETS = breakpoints kcmp mqueue vm cpu-hotplug memory-hotplug all: for TARGET in $(TARGETS); do \ diff --git a/tools/testing/selftests/cpu-hotplug/Makefile b/tools/testing/selftests/cpu-hotplug/Makefile new file mode 100644 index 0000000..7c9c20f --- /dev/null +++ b/tools/testing/selftests/cpu-hotplug/Makefile @@ -0,0 +1,6 @@ +all: + +run_tests: + ./on-off-test.sh + +clean: diff --git a/tools/testing/selftests/cpu-hotplug/on-off-test.sh b/tools/testing/selftests/cpu-hotplug/on-off-test.sh new file mode 100644 index 0000000..bdde7cf --- /dev/null +++ b/tools/testing/selftests/cpu-hotplug/on-off-test.sh @@ -0,0 +1,221 @@ +#!/bin/bash + +SYSFS= + +prerequisite() +{ + msg="skip all tests:" + + if [ $UID != 0 ]; then + echo $msg must be run as root >&2 + exit 0 + fi + + SYSFS=`mount -t sysfs | head -1 | awk '{ print $3 }'` + + if [ ! -d "$SYSFS" ]; then + echo $msg sysfs is not mounted >&2 + exit 0 + fi + + if ! ls $SYSFS/devices/system/cpu/cpu* > /dev/null 2>&1; then + echo $msg cpu hotplug is not supported >&2 + exit 0 + fi +} + +# +# list all hot-pluggable CPUs +# +hotpluggable_cpus() +{ + local state=${1:-.\*} + + for cpu in $SYSFS/devices/system/cpu/cpu*; do + if [ -f $cpu/online ] && grep -q $state $cpu/online; then + echo ${cpu##/*/cpu} + fi + done +} + +hotplaggable_offline_cpus() +{ + hotpluggable_cpus 0 +} + +hotpluggable_online_cpus() +{ + hotpluggable_cpus 1 +} + +cpu_is_online() +{ + grep -q 1 $SYSFS/devices/system/cpu/cpu$1/online +} + +cpu_is_offline() +{ + grep -q 0 $SYSFS/devices/system/cpu/cpu$1/online +} + +online_cpu() +{ + echo 1 > $SYSFS/devices/system/cpu/cpu$1/online +} + +offline_cpu() +{ + echo 0 > $SYSFS/devices/system/cpu/cpu$1/online +} + +online_cpu_expect_success() +{ + local cpu=$1 + + if ! online_cpu $cpu; then + echo $FUNCNAME $cpu: unexpected fail >&2 + elif ! cpu_is_online $cpu; then + echo $FUNCNAME $cpu: unexpected offline >&2 + fi +} + +online_cpu_expect_fail() +{ + local cpu=$1 + + if online_cpu $cpu 2> /dev/null; then + echo $FUNCNAME $cpu: unexpected success >&2 + elif ! cpu_is_offline $cpu; then + echo $FUNCNAME $cpu: unexpected online >&2 + fi +} + +offline_cpu_expect_success() +{ + local cpu=$1 + + if ! offline_cpu $cpu; then + echo $FUNCNAME $cpu: unexpected fail >&2 + elif ! cpu_is_offline $cpu; then + echo $FUNCNAME $cpu: unexpected offline >&2 + fi +} + +offline_cpu_expect_fail() +{ + local cpu=$1 + + if offline_cpu $cpu 2> /dev/null; then + echo $FUNCNAME $cpu: unexpected success >&2 + elif ! cpu_is_online $cpu; then + echo $FUNCNAME $cpu: unexpected offline >&2 + fi +} + +error=-12 +priority=0 + +while getopts e:hp: opt; do + case $opt in + e) + error=$OPTARG + ;; + h) + echo "Usage $0 [ -e errno ] [ -p notifier-priority ]" + exit + ;; + p) + priority=$OPTARG + ;; + esac +done + +if ! [ "$error" -ge -4095 -a "$error" -lt 0 ]; then + echo "error code must be -4095 <= errno < 0" >&2 + exit 1 +fi + +prerequisite + +# +# Online all hot-pluggable CPUs +# +for cpu in `hotplaggable_offline_cpus`; do + online_cpu_expect_success $cpu +done + +# +# Offline all hot-pluggable CPUs +# +for cpu in `hotpluggable_online_cpus`; do + offline_cpu_expect_success $cpu +done + +# +# Online all hot-pluggable CPUs again +# +for cpu in `hotplaggable_offline_cpus`; do + online_cpu_expect_success $cpu +done + +# +# Test with cpu notifier error injection +# + +DEBUGFS=`mount -t debugfs | head -1 | awk '{ print $3 }'` +NOTIFIER_ERR_INJECT_DIR=$DEBUGFS/notifier-error-inject/cpu + +prerequisite_extra() +{ + msg="skip extra tests:" + + /sbin/modprobe -q -r cpu-notifier-error-inject + /sbin/modprobe -q cpu-notifier-error-inject priority=$priority + + if [ ! -d "$DEBUGFS" ]; then + echo $msg debugfs is not mounted >&2 + exit 0 + fi + + if [ ! -d $NOTIFIER_ERR_INJECT_DIR ]; then + echo $msg cpu-notifier-error-inject module is not available >&2 + exit 0 + fi +} + +prerequisite_extra + +# +# Offline all hot-pluggable CPUs +# +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/CPU_DOWN_PREPARE/error +for cpu in `hotpluggable_online_cpus`; do + offline_cpu_expect_success $cpu +done + +# +# Test CPU hot-add error handling (offline => online) +# +echo $error > $NOTIFIER_ERR_INJECT_DIR/actions/CPU_UP_PREPARE/error +for cpu in `hotplaggable_offline_cpus`; do + online_cpu_expect_fail $cpu +done + +# +# Online all hot-pluggable CPUs +# +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/CPU_UP_PREPARE/error +for cpu in `hotplaggable_offline_cpus`; do + online_cpu_expect_success $cpu +done + +# +# Test CPU hot-remove error handling (online => offline) +# +echo $error > $NOTIFIER_ERR_INJECT_DIR/actions/CPU_DOWN_PREPARE/error +for cpu in `hotpluggable_online_cpus`; do + offline_cpu_expect_fail $cpu +done + +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/CPU_DOWN_PREPARE/error +/sbin/modprobe -q -r cpu-notifier-error-inject diff --git a/tools/testing/selftests/memory-hotplug/Makefile b/tools/testing/selftests/memory-hotplug/Makefile new file mode 100644 index 0000000..7c9c20f --- /dev/null +++ b/tools/testing/selftests/memory-hotplug/Makefile @@ -0,0 +1,6 @@ +all: + +run_tests: + ./on-off-test.sh + +clean: diff --git a/tools/testing/selftests/memory-hotplug/on-off-test.sh b/tools/testing/selftests/memory-hotplug/on-off-test.sh new file mode 100644 index 0000000..a2816f6 --- /dev/null +++ b/tools/testing/selftests/memory-hotplug/on-off-test.sh @@ -0,0 +1,230 @@ +#!/bin/bash + +SYSFS= + +prerequisite() +{ + msg="skip all tests:" + + if [ $UID != 0 ]; then + echo $msg must be run as root >&2 + exit 0 + fi + + SYSFS=`mount -t sysfs | head -1 | awk '{ print $3 }'` + + if [ ! -d "$SYSFS" ]; then + echo $msg sysfs is not mounted >&2 + exit 0 + fi + + if ! ls $SYSFS/devices/system/memory/memory* > /dev/null 2>&1; then + echo $msg memory hotplug is not supported >&2 + exit 0 + fi +} + +# +# list all hot-pluggable memory +# +hotpluggable_memory() +{ + local state=${1:-.\*} + + for memory in $SYSFS/devices/system/memory/memory*; do + if grep -q 1 $memory/removable && + grep -q $state $memory/state; then + echo ${memory##/*/memory} + fi + done +} + +hotplaggable_offline_memory() +{ + hotpluggable_memory offline +} + +hotpluggable_online_memory() +{ + hotpluggable_memory online +} + +memory_is_online() +{ + grep -q online $SYSFS/devices/system/memory/memory$1/state +} + +memory_is_offline() +{ + grep -q offline $SYSFS/devices/system/memory/memory$1/state +} + +online_memory() +{ + echo online > $SYSFS/devices/system/memory/memory$1/state +} + +offline_memory() +{ + echo offline > $SYSFS/devices/system/memory/memory$1/state +} + +online_memory_expect_success() +{ + local memory=$1 + + if ! online_memory $memory; then + echo $FUNCNAME $memory: unexpected fail >&2 + elif ! memory_is_online $memory; then + echo $FUNCNAME $memory: unexpected offline >&2 + fi +} + +online_memory_expect_fail() +{ + local memory=$1 + + if online_memory $memory 2> /dev/null; then + echo $FUNCNAME $memory: unexpected success >&2 + elif ! memory_is_offline $memory; then + echo $FUNCNAME $memory: unexpected online >&2 + fi +} + +offline_memory_expect_success() +{ + local memory=$1 + + if ! offline_memory $memory; then + echo $FUNCNAME $memory: unexpected fail >&2 + elif ! memory_is_offline $memory; then + echo $FUNCNAME $memory: unexpected offline >&2 + fi +} + +offline_memory_expect_fail() +{ + local memory=$1 + + if offline_memory $memory 2> /dev/null; then + echo $FUNCNAME $memory: unexpected success >&2 + elif ! memory_is_online $memory; then + echo $FUNCNAME $memory: unexpected offline >&2 + fi +} + +error=-12 +priority=0 +ratio=10 + +while getopts e:hp:r: opt; do + case $opt in + e) + error=$OPTARG + ;; + h) + echo "Usage $0 [ -e errno ] [ -p notifier-priority ] [ -r percent-of-memory-to-offline ]" + exit + ;; + p) + priority=$OPTARG + ;; + r) + ratio=$OPTARG + ;; + esac +done + +if ! [ "$error" -ge -4095 -a "$error" -lt 0 ]; then + echo "error code must be -4095 <= errno < 0" >&2 + exit 1 +fi + +prerequisite + +# +# Online all hot-pluggable memory +# +for memory in `hotplaggable_offline_memory`; do + online_memory_expect_success $memory +done + +# +# Offline $ratio percent of hot-pluggable memory +# +for memory in `hotpluggable_online_memory`; do + if [ $((RANDOM % 100)) -lt $ratio ]; then + offline_memory_expect_success $memory + fi +done + +# +# Online all hot-pluggable memory again +# +for memory in `hotplaggable_offline_memory`; do + online_memory_expect_success $memory +done + +# +# Test with memory notifier error injection +# + +DEBUGFS=`mount -t debugfs | head -1 | awk '{ print $3 }'` +NOTIFIER_ERR_INJECT_DIR=$DEBUGFS/notifier-error-inject/memory + +prerequisite_extra() +{ + msg="skip extra tests:" + + /sbin/modprobe -q -r memory-notifier-error-inject + /sbin/modprobe -q memory-notifier-error-inject priority=$priority + + if [ ! -d "$DEBUGFS" ]; then + echo $msg debugfs is not mounted >&2 + exit 0 + fi + + if [ ! -d $NOTIFIER_ERR_INJECT_DIR ]; then + echo $msg memory-notifier-error-inject module is not available >&2 + exit 0 + fi +} + +prerequisite_extra + +# +# Offline $ratio percent of hot-pluggable memory +# +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/MEM_GOING_OFFLINE/error +for memory in `hotpluggable_online_memory`; do + if [ $((RANDOM % 100)) -lt $ratio ]; then + offline_memory_expect_success $memory + fi +done + +# +# Test memory hot-add error handling (offline => online) +# +echo $error > $NOTIFIER_ERR_INJECT_DIR/actions/MEM_GOING_ONLINE/error +for memory in `hotplaggable_offline_memory`; do + online_memory_expect_fail $memory +done + +# +# Online all hot-pluggable memory +# +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/MEM_GOING_ONLINE/error +for memory in `hotplaggable_offline_memory`; do + online_memory_expect_success $memory +done + +# +# Test memory hot-remove error handling (online => offline) +# +echo $error > $NOTIFIER_ERR_INJECT_DIR/actions/MEM_GOING_OFFLINE/error +for memory in `hotpluggable_online_memory`; do + offline_memory_expect_fail $memory +done + +echo 0 > $NOTIFIER_ERR_INJECT_DIR/actions/MEM_GOING_OFFLINE/error +/sbin/modprobe -q -r memory-notifier-error-inject -- cgit v0.10.2 From c24aa64d169b7224f1a5bc6a4b1365da37ce861b Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:43:20 -0700 Subject: fault-injection: add tool to run command with failslab or fail_page_alloc This adds tools/testing/fault-injection/failcmd.sh to run a command while injecting slab/page allocation failures via fault injection. Example: Run a command "make -C tools/testing/selftests/ run_tests" with injecting slab allocation failure. # ./tools/testing/fault-injection/failcmd.sh \ -- make -C tools/testing/selftests/ run_tests Same as above except to specify 100 times failures at most instead of one time at most by default. # ./tools/testing/fault-injection/failcmd.sh --times=100 \ -- make -C tools/testing/selftests/ run_tests Same as above except to inject page allocation failure instead of slab allocation failure. # env FAILCMD_TYPE=fail_page_alloc \ ./tools/testing/fault-injection/failcmd.sh --times=100 \ -- make -C tools/testing/selftests/ run_tests Signed-off-by: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/fault-injection/fault-injection.txt b/Documentation/fault-injection/fault-injection.txt index ba4be8b..4cf1a2a 100644 --- a/Documentation/fault-injection/fault-injection.txt +++ b/Documentation/fault-injection/fault-injection.txt @@ -240,3 +240,30 @@ trap "echo 0 > /sys/kernel/debug/$FAILTYPE/probability" SIGINT SIGTERM EXIT echo "Injecting errors into the module $module... (interrupt to stop)" sleep 1000000 +Tool to run command with failslab or fail_page_alloc +---------------------------------------------------- +In order to make it easier to accomplish the tasks mentioned above, we can use +tools/testing/fault-injection/failcmd.sh. Please run a command +"./tools/testing/fault-injection/failcmd.sh --help" for more information and +see the following examples. + +Examples: + +Run a command "make -C tools/testing/selftests/ run_tests" with injecting slab +allocation failure. + + # ./tools/testing/fault-injection/failcmd.sh \ + -- make -C tools/testing/selftests/ run_tests + +Same as above except to specify 100 times failures at most instead of one time +at most by default. + + # ./tools/testing/fault-injection/failcmd.sh --times=100 \ + -- make -C tools/testing/selftests/ run_tests + +Same as above except to inject page allocation failure instead of slab +allocation failure. + + # env FAILCMD_TYPE=fail_page_alloc \ + ./tools/testing/fault-injection/failcmd.sh --times=100 \ + -- make -C tools/testing/selftests/ run_tests diff --git a/tools/testing/fault-injection/failcmd.sh b/tools/testing/fault-injection/failcmd.sh new file mode 100644 index 0000000..1776e92 --- /dev/null +++ b/tools/testing/fault-injection/failcmd.sh @@ -0,0 +1,219 @@ +#!/bin/bash +# +# NAME +# failcmd.sh - run a command with injecting slab/page allocation failures +# +# SYNOPSIS +# failcmd.sh --help +# failcmd.sh [] command [arguments] +# +# DESCRIPTION +# Run command with injecting slab/page allocation failures by fault +# injection. +# +# NOTE: you need to run this script as root. +# + +usage() +{ + cat >&2 <&2 + exit 1 +fi + +DEBUGFS=`mount -t debugfs | head -1 | awk '{ print $3}'` + +if [ ! -d "$DEBUGFS" ]; then + echo debugfs is not mounted >&2 + exit 1 +fi + +FAILCMD_TYPE=${FAILCMD_TYPE:-failslab} +FAULTATTR=$DEBUGFS/$FAILCMD_TYPE + +if [ ! -d $FAULTATTR ]; then + echo $FAILCMD_TYPE is not available >&2 + exit 1 +fi + +LONGOPTS=probability:,interval:,times:,space:,verbose:,task-filter: +LONGOPTS=$LONGOPTS,stacktrace-depth:,require-start:,require-end: +LONGOPTS=$LONGOPTS,reject-start:,reject-end:,oom-kill-allocating-task:,help + +if [ $FAILCMD_TYPE = failslab ]; then + LONGOPTS=$LONGOPTS,ignore-gfp-wait:,cache-filter: +elif [ $FAILCMD_TYPE = fail_page_alloc ]; then + LONGOPTS=$LONGOPTS,ignore-gfp-wait:,ignore-gfp-highmem:,min-order: +fi + +TEMP=`getopt -o p:i:t:s:v:h --long $LONGOPTS -n 'failcmd.sh' -- "$@"` + +if [ $? != 0 ]; then + usage + exit 1 +fi + +eval set -- "$TEMP" + +fault_attr_default() +{ + echo N > $FAULTATTR/task-filter + echo 0 > $FAULTATTR/probability + echo 1 > $FAULTATTR/times +} + +fault_attr_default + +oom_kill_allocating_task_saved=`cat /proc/sys/vm/oom_kill_allocating_task` + +restore_values() +{ + fault_attr_default + echo $oom_kill_allocating_task_saved \ + > /proc/sys/vm/oom_kill_allocating_task +} + +# +# Default options +# +declare -i oom_kill_allocating_task=1 +declare task_filter=Y +declare -i probability=1 +declare -i times=1 + +while true; do + case "$1" in + -p|--probability) + probability=$2 + shift 2 + ;; + -i|--interval) + echo $2 > $FAULTATTR/interval + shift 2 + ;; + -t|--times) + times=$2 + shift 2 + ;; + -s|--space) + echo $2 > $FAULTATTR/space + shift 2 + ;; + -v|--verbose) + echo $2 > $FAULTATTR/verbose + shift 2 + ;; + --task-filter) + task_filter=$2 + shift 2 + ;; + --stacktrace-depth) + echo $2 > $FAULTATTR/stacktrace-depth + shift 2 + ;; + --require-start) + echo $2 > $FAULTATTR/require-start + shift 2 + ;; + --require-end) + echo $2 > $FAULTATTR/require-end + shift 2 + ;; + --reject-start) + echo $2 > $FAULTATTR/reject-start + shift 2 + ;; + --reject-end) + echo $2 > $FAULTATTR/reject-end + shift 2 + ;; + --oom-kill-allocating-task) + oom_kill_allocating_task=$2 + shift 2 + ;; + --ignore-gfp-wait) + echo $2 > $FAULTATTR/ignore-gfp-wait + shift 2 + ;; + --cache-filter) + echo $2 > $FAULTATTR/cache_filter + shift 2 + ;; + --ignore-gfp-highmem) + echo $2 > $FAULTATTR/ignore-gfp-highmem + shift 2 + ;; + --min-order) + echo $2 > $FAULTATTR/min-order + shift 2 + ;; + -h|--help) + usage + exit 0 + shift + ;; + --) + shift + break + ;; + esac +done + +[ -z "$@" ] && exit 0 + +echo $oom_kill_allocating_task > /proc/sys/vm/oom_kill_allocating_task +echo $task_filter > $FAULTATTR/task-filter +echo $probability > $FAULTATTR/probability +echo $times > $FAULTATTR/times + +trap "restore_values" SIGINT SIGTERM EXIT + +cmd="echo 1 > /proc/self/make-it-fail && exec $@" +bash -c "$cmd" -- cgit v0.10.2 From e04f228335bde305d2833e40382ed37fc36f9efe Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Mon, 30 Jul 2012 14:43:22 -0700 Subject: lib/scatterlist: do not re-write gfp_flags in __sg_alloc_table() We are seeing a lot of sg_alloc_table allocation failures using the new drm prime infrastructure. We isolated the cause to code in __sg_alloc_table that was re-writing the gfp_flags. There is a comment in the code that suggest that there is an assumption about the allocation coming from a memory pool. This was likely true when sg lists were primarily used for disk I/O. Signed-off-by: Mandeep Singh Baines Cc: Jens Axboe Cc: Paul Gortmaker Cc: Cong Wang Cc: Daniel Vetter Cc: Rob Clark Cc: Sumit Semwal Cc: Inki Dae Cc: Dave Airlie Cc: Sonny Rao Cc: Olof Johansson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/lib/scatterlist.c b/lib/scatterlist.c index 6096e89..d09bdd8 100644 --- a/lib/scatterlist.c +++ b/lib/scatterlist.c @@ -279,14 +279,6 @@ int __sg_alloc_table(struct sg_table *table, unsigned int nents, if (!left) sg_mark_end(&sg[sg_size - 1]); - /* - * only really needed for mempool backed sg allocations (like - * SCSI), a possible improvement here would be to pass the - * table pointer into the allocator and let that clear these - * flags - */ - gfp_mask &= ~__GFP_WAIT; - gfp_mask |= __GFP_HIGH; prv = sg; } while (left); -- cgit v0.10.2 From 086ff4b3a7fb9cdf41e6a5d0ccd99b86d84633a1 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Jul 2012 14:43:24 -0700 Subject: drivers/misc/lkdtm.c: fix missing allocation failure check Addresses https://bugzilla.kernel.org/show_bug.cgi?id=44691 Reported-by: Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/misc/lkdtm.c b/drivers/misc/lkdtm.c index 28adefe..08aad69 100644 --- a/drivers/misc/lkdtm.c +++ b/drivers/misc/lkdtm.c @@ -477,6 +477,8 @@ static ssize_t lkdtm_debugfs_read(struct file *f, char __user *user_buf, int i, n, out; buf = (char *)__get_free_page(GFP_KERNEL); + if (buf == NULL) + return -ENOMEM; n = snprintf(buf, PAGE_SIZE, "Available crash types:\n"); for (i = 0; i < ARRAY_SIZE(cp_type); i++) -- cgit v0.10.2