diff options
198 files changed, 2754 insertions, 649 deletions
diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl index 90dc2de8e0af..158ffe9bfade 100644 --- a/Documentation/DocBook/kernel-locking.tmpl +++ b/Documentation/DocBook/kernel-locking.tmpl @@ -222,7 +222,7 @@ <title>Two Main Types of Kernel Locks: Spinlocks and Semaphores</title> <para> - There are two main types of kernel locks. The fundamental type + There are three main types of kernel locks. The fundamental type is the spinlock (<filename class="headerfile">include/asm/spinlock.h</filename>), which is a very simple single-holder lock: if you can't get the @@ -230,16 +230,22 @@ very small and fast, and can be used anywhere. </para> <para> - The second type is a semaphore + The second type is a mutex + (<filename class="headerfile">include/linux/mutex.h</filename>): it + is like a spinlock, but you may block holding a mutex. + If you can't lock a mutex, your task will suspend itself, and be woken + up when the mutex is released. This means the CPU can do something + else while you are waiting. There are many cases when you simply + can't sleep (see <xref linkend="sleeping-things"/>), and so have to + use a spinlock instead. + </para> + <para> + The third type is a semaphore (<filename class="headerfile">include/asm/semaphore.h</filename>): it can have more than one holder at any time (the number decided at initialization time), although it is most commonly used as a - single-holder lock (a mutex). If you can't get a semaphore, - your task will put itself on the queue, and be woken up when the - semaphore is released. This means the CPU will do something - else while you are waiting, but there are many cases when you - simply can't sleep (see <xref linkend="sleeping-things"/>), and so - have to use a spinlock instead. + single-holder lock (a mutex). If you can't get a semaphore, your + task will be suspended and later on woken up - just like for mutexes. </para> <para> Neither type of lock is recursive: see diff --git a/Documentation/mutex-design.txt b/Documentation/mutex-design.txt new file mode 100644 index 000000000000..cbf79881a41c --- /dev/null +++ b/Documentation/mutex-design.txt @@ -0,0 +1,135 @@ +Generic Mutex Subsystem + +started by Ingo Molnar <mingo@redhat.com> + + "Why on earth do we need a new mutex subsystem, and what's wrong + with semaphores?" + +firstly, there's nothing wrong with semaphores. But if the simpler +mutex semantics are sufficient for your code, then there are a couple +of advantages of mutexes: + + - 'struct mutex' is smaller on most architectures: .e.g on x86, + 'struct semaphore' is 20 bytes, 'struct mutex' is 16 bytes. + A smaller structure size means less RAM footprint, and better + CPU-cache utilization. + + - tighter code. On x86 i get the following .text sizes when + switching all mutex-alike semaphores in the kernel to the mutex + subsystem: + + text data bss dec hex filename + 3280380 868188 396860 4545428 455b94 vmlinux-semaphore + 3255329 865296 396732 4517357 44eded vmlinux-mutex + + that's 25051 bytes of code saved, or a 0.76% win - off the hottest + codepaths of the kernel. (The .data savings are 2892 bytes, or 0.33%) + Smaller code means better icache footprint, which is one of the + major optimization goals in the Linux kernel currently. + + - the mutex subsystem is slightly faster and has better scalability for + contended workloads. On an 8-way x86 system, running a mutex-based + kernel and testing creat+unlink+close (of separate, per-task files) + in /tmp with 16 parallel tasks, the average number of ops/sec is: + + Semaphores: Mutexes: + + $ ./test-mutex V 16 10 $ ./test-mutex V 16 10 + 8 CPUs, running 16 tasks. 8 CPUs, running 16 tasks. + checking VFS performance. checking VFS performance. + avg loops/sec: 34713 avg loops/sec: 84153 + CPU utilization: 63% CPU utilization: 22% + + i.e. in this workload, the mutex based kernel was 2.4 times faster + than the semaphore based kernel, _and_ it also had 2.8 times less CPU + utilization. (In terms of 'ops per CPU cycle', the semaphore kernel + performed 551 ops/sec per 1% of CPU time used, while the mutex kernel + performed 3825 ops/sec per 1% of CPU time used - it was 6.9 times + more efficient.) + + the scalability difference is visible even on a 2-way P4 HT box: + + Semaphores: Mutexes: + + $ ./test-mutex V 16 10 $ ./test-mutex V 16 10 + 4 CPUs, running 16 tasks. 8 CPUs, running 16 tasks. + checking VFS performance. checking VFS performance. + avg loops/sec: 127659 avg loops/sec: 181082 + CPU utilization: 100% CPU utilization: 34% + + (the straight performance advantage of mutexes is 41%, the per-cycle + efficiency of mutexes is 4.1 times better.) + + - there are no fastpath tradeoffs, the mutex fastpath is just as tight + as the semaphore fastpath. On x86, the locking fastpath is 2 + instructions: + + c0377ccb <mutex_lock>: + c0377ccb: f0 ff 08 lock decl (%eax) + c0377cce: 78 0e js c0377cde <.text.lock.mutex> + c0377cd0: c3 ret + + the unlocking fastpath is equally tight: + + c0377cd1 <mutex_unlock>: + c0377cd1: f0 ff 00 lock incl (%eax) + c0377cd4: 7e 0f jle c0377ce5 <.text.lock.mutex+0x7> + c0377cd6: c3 ret + + - 'struct mutex' semantics are well-defined and are enforced if + CONFIG_DEBUG_MUTEXES is turned on. Semaphores on the other hand have + virtually no debugging code or instrumentation. The mutex subsystem + checks and enforces the following rules: + + * - only one task can hold the mutex at a time + * - only the owner can unlock the mutex + * - multiple unlocks are not permitted + * - recursive locking is not permitted + * - a mutex object must be initialized via the API + * - a mutex object must not be initialized via memset or copying + * - task may not exit with mutex held + * - memory areas where held locks reside must not be freed + * - held mutexes must not be reinitialized + * - mutexes may not be used in irq contexts + + furthermore, there are also convenience features in the debugging + code: + + * - uses symbolic names of mutexes, whenever they are printed in debug output + * - point-of-acquire tracking, symbolic lookup of function names + * - list of all locks held in the system, printout of them + * - owner tracking + * - detects self-recursing locks and prints out all relevant info + * - detects multi-task circular deadlocks and prints out all affected + * locks and tasks (and only those tasks) + +Disadvantages +------------- + +The stricter mutex API means you cannot use mutexes the same way you +can use semaphores: e.g. they cannot be used from an interrupt context, +nor can they be unlocked from a different context that which acquired +it. [ I'm not aware of any other (e.g. performance) disadvantages from +using mutexes at the moment, please let me know if you find any. ] + +Implementation of mutexes +------------------------- + +'struct mutex' is the new mutex type, defined in include/linux/mutex.h +and implemented in kernel/mutex.c. It is a counter-based mutex with a +spinlock and a wait-list. The counter has 3 states: 1 for "unlocked", +0 for "locked" and negative numbers (usually -1) for "locked, potential +waiters queued". + +the APIs of 'struct mutex' have been streamlined: + + DEFINE_MUTEX(name); + + mutex_init(mutex); + + void mutex_lock(struct mutex *lock); + int mutex_lock_interruptible(struct mutex *lock); + int mutex_trylock(struct mutex *lock); + void mutex_unlock(struct mutex *lock); + int mutex_is_locked(struct mutex *lock); + diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c index c30a16df6440..e8a53552b13d 100644 --- a/arch/i386/mm/pageattr.c +++ b/arch/i386/mm/pageattr.c @@ -222,6 +222,10 @@ void kernel_map_pages(struct page *page, int numpages, int enable) { if (PageHighMem(page)) return; + if (!enable) + mutex_debug_check_no_locks_freed(page_address(page), + page_address(page+numpages)); + /* the return value is ignored - the calls cannot fail, * large pages are disabled at boot time. */ diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 1f3507c75e90..d2ba358c6e38 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -137,7 +137,7 @@ spufs_delete_inode(struct inode *inode) static void spufs_prune_dir(struct dentry *dir) { struct dentry *dentry, *tmp; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); list_for_each_entry_safe(dentry, tmp, &dir->d_subdirs, d_child) { spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); @@ -154,7 +154,7 @@ static void spufs_prune_dir(struct dentry *dir) } } shrink_dcache_parent(dir); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); } static int spufs_rmdir(struct inode *root, struct dentry *dir_dentry) @@ -162,15 +162,15 @@ static int spufs_rmdir(struct inode *root, struct dentry *dir_dentry) struct spu_context *ctx; /* remove all entries */ - down(&root->i_sem); + mutex_lock(&root->i_mutex); spufs_prune_dir(dir_dentry); - up(&root->i_sem); + mutex_unlock(&root->i_mutex); /* We have to give up the mm_struct */ ctx = SPUFS_I(dir_dentry->d_inode)->i_ctx; spu_forget(ctx); - /* XXX Do we need to hold i_sem here ? */ + /* XXX Do we need to hold i_mutex here ? */ return simple_rmdir(root, dir_dentry); } @@ -330,7 +330,7 @@ long spufs_create_thread(struct nameidata *nd, out_dput: dput(dentry); out_dir: - up(&nd->dentry->d_inode->i_sem); + mutex_unlock(&nd->dentry->d_inode->i_mutex); out: return ret; } diff --git a/drivers/block/loop.c b/drivers/block/loop.c index a452b13620a2..864729046e22 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -215,7 +215,7 @@ static int do_lo_send_aops(struct loop_device *lo, struct bio_vec *bvec, unsigned offset, bv_offs; int len, ret; - down(&mapping->host->i_sem); + mutex_lock(&mapping->host->i_mutex); index = pos >> PAGE_CACHE_SHIFT; offset = pos & ((pgoff_t)PAGE_CACHE_SIZE - 1); bv_offs = bvec->bv_offset; @@ -278,7 +278,7 @@ static int do_lo_send_aops(struct loop_device *lo, struct bio_vec *bvec, } ret = 0; out: - up(&mapping->host->i_sem); + mutex_unlock(&mapping->host->i_mutex); return ret; unlock: unlock_page(page); @@ -527,12 +527,12 @@ static int loop_make_request(request_queue_t *q, struct bio *old_bio) lo->lo_pending++; loop_add_bio(lo, old_bio); spin_unlock_irq(&lo->lo_lock); - up(&lo->lo_bh_mutex); + complete(&lo->lo_bh_done); return 0; out: if (lo->lo_pending == 0) - up(&lo->lo_bh_mutex); + complete(&lo->lo_bh_done); spin_unlock_irq(&lo->lo_lock); bio_io_error(old_bio, old_bio->bi_size); return 0; @@ -593,23 +593,20 @@ static int loop_thread(void *data) lo->lo_pending = 1; /* - * up sem, we are running + * complete it, we are running */ - up(&lo->lo_sem); + complete(&lo->lo_done); for (;;) { int pending; - /* - * interruptible just to not contribute to load avg - */ - if (down_interruptible(&lo->lo_bh_mutex)) + if (wait_for_completion_interruptible(&lo->lo_bh_done)) continue; spin_lock_irq(&lo->lo_lock); /* - * could be upped because of tear-down, not pending work + * could be completed because of tear-down, not pending work */ if (unlikely(!lo->lo_pending)) { spin_unlock_irq(&lo->lo_lock); @@ -632,7 +629,7 @@ static int loop_thread(void *data) break; } - up(&lo->lo_sem); + complete(&lo->lo_done); return 0; } @@ -843,7 +840,7 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file, set_blocksize(bdev, lo_blocksize); kernel_thread(loop_thread, lo, CLONE_KERNEL); - down(&lo->lo_sem); + wait_for_completion(&lo->lo_done); return 0; out_putf: @@ -909,10 +906,10 @@ static int loop_clr_fd(struct loop_device *lo, struct block_device *bdev) lo->lo_state = Lo_rundown; lo->lo_pending--; if (!lo->lo_pending) - up(&lo->lo_bh_mutex); + complete(&lo->lo_bh_done); spin_unlock_irq(&lo->lo_lock); - down(&lo->lo_sem); + wait_for_completion(&lo->lo_done); lo->lo_backing_file = NULL; @@ -1289,8 +1286,8 @@ static int __init loop_init(void) if (!lo->lo_queue) goto out_mem4; init_MUTEX(&lo->lo_ctl_mutex); - init_MUTEX_LOCKED(&lo->lo_sem); - init_MUTEX_LOCKED(&lo->lo_bh_mutex); + init_completion(&lo->lo_done); + init_completion(&lo->lo_bh_done); lo->lo_number = i; spin_lock_init(&lo->lo_lock); disk->major = LOOP_MAJOR; diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c index c0cdc182a8b0..4bdf95716e2b 100644 --- a/drivers/block/sx8.c +++ b/drivers/block/sx8.c @@ -27,8 +27,8 @@ #include <linux/time.h> #include <linux/hdreg.h> #include <linux/dma-mapping.h> +#include <linux/completion.h> #include <asm/io.h> -#include <asm/semaphore.h> #include <asm/uaccess.h> #if 0 @@ -303,7 +303,7 @@ struct carm_host { struct work_struct fsm_task; - struct semaphore probe_sem; + struct completion probe_comp; }; struct carm_response { @@ -1346,7 +1346,7 @@ static void carm_fsm_task (void *_data) } case HST_PROBE_FINISHED: - up(&host->probe_sem); + complete(&host->probe_comp); break; case HST_ERROR: @@ -1622,7 +1622,7 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) host->flags = pci_dac ? FL_DAC : 0; spin_lock_init(&host->lock); INIT_WORK(&host->fsm_task, carm_fsm_task, host); - init_MUTEX_LOCKED(&host->probe_sem); + init_completion(&host->probe_comp); for (i = 0; i < ARRAY_SIZE(host->req); i++) host->req[i].tag = i; @@ -1691,8 +1691,8 @@ static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) goto err_out_free_irq; - DPRINTK("waiting for probe_sem\n"); - down(&host->probe_sem); + DPRINTK("waiting for probe_comp\n"); + wait_for_completion(&host->probe_comp); printk(KERN_INFO "%s: pci %s, ports %d, io %lx, irq %u, major %d\n", host->name, pci_name(pdev), (int) CARM_MAX_PORTS, diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 5b2d18035073..704c3c07f0ab 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -741,7 +741,7 @@ static loff_t memory_lseek(struct file * file, loff_t offset, int orig) { loff_t ret; - down(&file->f_dentry->d_inode->i_sem); + mutex_lock(&file->f_dentry->d_inode->i_mutex); switch (orig) { case 0: file->f_pos = offset; @@ -756,7 +756,7 @@ static loff_t memory_lseek(struct file * file, loff_t offset, int orig) default: ret = -EINVAL; } - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return ret; } diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index 145275ebdd7e..5765f672e853 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -153,6 +153,21 @@ static struct sysrq_key_op sysrq_mountro_op = { /* END SYNC SYSRQ HANDLERS BLOCK */ +#ifdef CONFIG_DEBUG_MUTEXES + +static void +sysrq_handle_showlocks(int key, struct pt_regs *pt_regs, struct tty_struct *tty) +{ + mutex_debug_show_all_locks(); +} + +static struct sysrq_key_op sysrq_showlocks_op = { + .handler = sysrq_handle_showlocks, + .help_msg = "show-all-locks(D)", + .action_msg = "Show Locks Held", +}; + +#endif /* SHOW SYSRQ HANDLERS BLOCK */ @@ -294,7 +309,11 @@ static struct sysrq_key_op *sysrq_key_table[SYSRQ_KEY_TABLE_LENGTH] = { #else /* c */ NULL, #endif +#ifdef CONFIG_DEBUG_MUTEXES +/* d */ &sysrq_showlocks_op, +#else /* d */ NULL, +#endif /* e */ &sysrq_term_op, /* f */ &sysrq_moom_op, /* g */ NULL, diff --git a/drivers/char/watchdog/cpu5wdt.c b/drivers/char/watchdog/cpu5wdt.c index e75045fe2641..3e8410b5a65e 100644 --- a/drivers/char/watchdog/cpu5wdt.c +++ b/drivers/char/watchdog/cpu5wdt.c @@ -28,6 +28,7 @@ #include <linux/init.h> #include <linux/ioport.h> #include <linux/timer.h> +#include <linux/completion.h> #include <linux/jiffies.h> #include <asm/io.h> #include <asm/uaccess.h> @@ -57,7 +58,7 @@ static int ticks = 10000; /* some device data */ static struct { - struct semaphore stop; + struct completion stop; volatile int running; struct timer_list timer; volatile int queue; @@ -85,7 +86,7 @@ static void cpu5wdt_trigger(unsigned long unused) } else { /* ticks doesn't matter anyway */ - up(&cpu5wdt_device.stop); + complete(&cpu5wdt_device.stop); } } @@ -239,7 +240,7 @@ static int __devinit cpu5wdt_init(void) if ( !val ) printk(KERN_INFO PFX "sorry, was my fault\n"); - init_MUTEX_LOCKED(&cpu5wdt_device.stop); + init_completion(&cpu5wdt_device.stop); cpu5wdt_device.queue = 0; clear_bit(0, &cpu5wdt_device.inuse); @@ -269,7 +270,7 @@ static void __devexit cpu5wdt_exit(void) { if ( cpu5wdt_device.queue ) { cpu5wdt_device.queue = 0; - down(&cpu5wdt_device.stop); + wait_for_completion(&cpu5wdt_device.stop); } misc_deregister(&cpu5wdt_misc); diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c index 1ddaa71a8f45..7cb2d86601db 100644 --- a/drivers/ide/ide-probe.c +++ b/drivers/ide/ide-probe.c @@ -655,7 +655,7 @@ static void hwif_release_dev (struct device *dev) { ide_hwif_t *hwif = container_of(dev, ide_hwif_t, gendev); - up(&hwif->gendev_rel_sem); + complete(&hwif->gendev_rel_comp); } static void hwif_register (ide_hwif_t *hwif) @@ -1327,7 +1327,7 @@ static void drive_release_dev (struct device *dev) drive->queue = NULL; spin_unlock_irq(&ide_lock); - up(&drive->gendev_rel_sem); + complete(&drive->gendev_rel_comp); } /* diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c index b069b13b75a7..ec5a4cb173b0 100644 --- a/drivers/ide/ide.c +++ b/drivers/ide/ide.c @@ -222,7 +222,7 @@ static void init_hwif_data(ide_hwif_t *hwif, unsigned int index) hwif->mwdma_mask = 0x80; /* disable all mwdma */ hwif->swdma_mask = 0x80; /* disable all swdma */ - sema_init(&hwif->gendev_rel_sem, 0); + init_completion(&hwif->gendev_rel_comp); default_hwif_iops(hwif); default_hwif_transport(hwif); @@ -245,7 +245,7 @@ static void init_hwif_data(ide_hwif_t *hwif, unsigned int index) drive->is_flash = 0; drive->vdma = 0; INIT_LIST_HEAD(&drive->list); - sema_init(&drive->gendev_rel_sem, 0); + init_completion(&drive->gendev_rel_comp); } } @@ -602,7 +602,7 @@ void ide_unregister(unsigned int index) } spin_unlock_irq(&ide_lock); device_unregister(&drive->gendev); - down(&drive->gendev_rel_sem); + wait_for_completion(&drive->gendev_rel_comp); spin_lock_irq(&ide_lock); } hwif->present = 0; @@ -662,7 +662,7 @@ void ide_unregister(unsigned int index) /* More messed up locking ... */ spin_unlock_irq(&ide_lock); device_unregister(&hwif->gendev); - down(&hwif->gendev_rel_sem); + wait_for_completion(&hwif->gendev_rel_comp); /* * Remove us from the kernel's knowledge diff --git a/drivers/isdn/capi/capifs.c b/drivers/isdn/capi/capifs.c index 207cae366256..0a37aded4b54 100644 --- a/drivers/isdn/capi/capifs.c +++ b/drivers/isdn/capi/capifs.c @@ -138,7 +138,7 @@ static struct dentry *get_node(int num) { char s[10]; struct dentry *root = capifs_root; - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); return lookup_one_len(s, root, sprintf(s, "%d", num)); } @@ -159,7 +159,7 @@ void capifs_new_ncci(unsigned int number, dev_t device) dentry = get_node(number); if (!IS_ERR(dentry) && !dentry->d_inode) d_instantiate(dentry, inode); - up(&capifs_root->d_inode->i_sem); + mutex_unlock(&capifs_root->d_inode->i_mutex); } void capifs_free_ncci(unsigned int number) @@ -175,7 +175,7 @@ void capifs_free_ncci(unsigned int number) } dput(dentry); } - up(&capifs_root->d_inode->i_sem); + mutex_unlock(&capifs_root->d_inode->i_mutex); } static int __init capifs_init(void) diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 0e481512f918..5c210b0a4cb0 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -837,9 +837,9 @@ static void __set_size(struct mapped_device *md, sector_t size) { set_capacity(md->disk, size); - down(&md->suspended_bdev->bd_inode->i_sem); + mutex_lock(&md->suspended_bdev->bd_inode->i_mutex); i_size_write(md->suspended_bdev->bd_inode, (loff_t)size << SECTOR_SHIFT); - up(&md->suspended_bdev->bd_inode->i_sem); + mutex_unlock(&md->suspended_bdev->bd_inode->i_mutex); } static int __bind(struct mapped_device *md, struct dm_table *t) diff --git a/drivers/md/md.c b/drivers/md/md.c index e423a16ba3c9..0302723fa21f 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -3460,9 +3460,9 @@ static int update_size(mddev_t *mddev, unsigned long size) bdev = bdget_disk(mddev->gendisk, 0); if (bdev) { - down(&bdev->bd_inode->i_sem); + mutex_lock(&bdev->bd_inode->i_mutex); i_size_write(bdev->bd_inode, mddev->array_size << 10); - up(&bdev->bd_inode->i_sem); + mutex_unlock(&bdev->bd_inode->i_mutex); bdput(bdev); } } @@ -3486,9 +3486,9 @@ static int update_raid_disks(mddev_t *mddev, int raid_disks) bdev = bdget_disk(mddev->gendisk, 0); if (bdev) { - down(&bdev->bd_inode->i_sem); + mutex_lock(&bdev->bd_inode->i_mutex); i_size_write(bdev->bd_inode, mddev->array_size << 10); - up(&bdev->bd_inode->i_sem); + mutex_unlock(&bdev->bd_inode->i_mutex); bdput(bdev); } } diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 9eb465727fce..9cb6dd0834be 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -25,7 +25,7 @@ proc_bus_pci_lseek(struct file *file, loff_t off, int whence) loff_t new = -1; struct inode *inode = file->f_dentry->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); switch (whence) { case 0: new = off; @@ -41,7 +41,7 @@ proc_bus_pci_lseek(struct file *file, loff_t off, int whence) new = -EINVAL; else file->f_pos = new; - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return new; } diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c index 4ddc453023a2..3cf945cc5b9a 100644 --- a/drivers/usb/core/inode.c +++ b/drivers/usb/core/inode.c @@ -184,13 +184,13 @@ static void update_bus(struct dentry *bus) bus->d_inode->i_gid = busgid; bus->d_inode->i_mode = S_IFDIR | busmode; - down(&bus->d_inode->i_sem); + mutex_lock(&bus->d_inode->i_mutex); list_for_each_entry(dev, &bus->d_subdirs, d_u.d_child) if (dev->d_inode) update_dev(dev); - up(&bus->d_inode->i_sem); + mutex_unlock(&bus->d_inode->i_mutex); } static void update_sb(struct super_block *sb) @@ -201,7 +201,7 @@ static void update_sb(struct super_block *sb) if (!root) return; - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); list_for_each_entry(bus, &root->d_subdirs, d_u.d_child) { if (bus->d_inode) { @@ -219,7 +219,7 @@ static void update_sb(struct super_block *sb) } } - up(&root->d_inode->i_sem); + mutex_unlock(&root->d_inode->i_mutex); } static int remount(struct super_block *sb, int *flags, char *data) @@ -333,10 +333,10 @@ static int usbfs_empty (struct dentry *dentry) static int usbfs_unlink (struct inode *dir, struct dentry *dentry) { struct inode *inode = dentry->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); dentry->d_inode->i_nlink--; dput(dentry); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); d_delete(dentry); return 0; } @@ -346,7 +346,7 @@ static int usbfs_rmdir(struct inode *dir, struct dentry *dentry) int error = -ENOTEMPTY; struct inode * inode = dentry->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); dentry_unhash(dentry); if (usbfs_empty(dentry)) { dentry->d_inode->i_nlink -= 2; @@ -355,7 +355,7 @@ static int usbfs_rmdir(struct inode *dir, struct dentry *dentry) dir->i_nlink--; error = 0; } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (!error) d_delete(dentry); dput(dentry); @@ -380,7 +380,7 @@ static loff_t default_file_lseek (struct file *file, loff_t offset, int orig) { loff_t retval = -EINVAL; - down(&file->f_dentry->d_inode->i_sem); + mutex_lock(&file->f_dentry->d_inode->i_mutex); switch(orig) { case 0: if (offset > 0) { @@ -397,7 +397,7 @@ static loff_t default_file_lseek (struct file *file, loff_t offset, int orig) default: break; } - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return retval; } @@ -480,7 +480,7 @@ static int fs_create_by_name (const char *name, mode_t mode, } *dentry = NULL; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); *dentry = lookup_one_len(name, parent, strlen(name)); if (!IS_ERR(dentry)) { if ((mode & S_IFMT) == S_IFDIR) @@ -489,7 +489,7 @@ static int fs_create_by_name (const char *name, mode_t mode, error = usbfs_create (parent->d_inode, *dentry, mode); } else error = PTR_ERR(dentry); - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); return error; } @@ -528,7 +528,7 @@ static void fs_remove_file (struct dentry *dentry) if (!parent || !parent->d_inode) return; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); if (usbfs_positive(dentry)) { if (dentry->d_inode) { if (S_ISDIR(dentry->d_inode->i_mode)) @@ -538,7 +538,7 @@ static void fs_remove_file (struct dentry *dentry) dput(dentry); } } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); } /* --------------------------------------------------------------------- */ diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 0cea9782d7d4..de59c58896d6 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -1891,7 +1891,7 @@ static int fsync_sub(struct lun *curlun) return -EINVAL; inode = filp->f_dentry->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); current->flags |= PF_SYNCWRITE; rc = filemap_fdatawrite(inode->i_mapping); err = filp->f_op->fsync(filp, filp->f_dentry, 1); @@ -1901,7 +1901,7 @@ static int fsync_sub(struct lun *curlun) if (!rc) rc = err; current->flags &= ~PF_SYNCWRITE; - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); VLDBG(curlun, "fdatasync -> %d\n", rc); return rc; } diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index 5c40980a5bd9..c6c279de832e 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -1562,10 +1562,10 @@ restart: spin_unlock_irq (&dev->lock); /* break link to dcache */ - down (&parent->i_sem); + mutex_lock (&parent->i_mutex); d_delete (dentry); dput (dentry); - up (&parent->i_sem); + mutex_unlock (&parent->i_mutex); /* fds may still be open */ goto restart; diff --git a/fs/affs/inode.c b/fs/affs/inode.c index 9ebe881c6786..44d439cb69f4 100644 --- a/fs/affs/inode.c +++ b/fs/affs/inode.c @@ -244,10 +244,10 @@ affs_put_inode(struct inode *inode) pr_debug("AFFS: put_inode(ino=%lu, nlink=%u)\n", inode->i_ino, inode->i_nlink); affs_free_prealloc(inode); if (atomic_read(&inode->i_count) == 1) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (inode->i_size != AFFS_I(inode)->mmu_private) affs_truncate(inode); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } } diff --git a/fs/autofs/root.c b/fs/autofs/root.c index a1ab1c0ed215..808134a5a2fa 100644 --- a/fs/autofs/root.c +++ b/fs/autofs/root.c @@ -229,9 +229,9 @@ static struct dentry *autofs_root_lookup(struct inode *dir, struct dentry *dentr dentry->d_flags |= DCACHE_AUTOFS_PENDING; d_add(dentry, NULL); - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); autofs_revalidate(dentry, nd); - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); /* * If we are still pending, check if we had to handle diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 2241405ffc41..541b19e6fec9 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -489,9 +489,9 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s d_add(dentry, NULL); if (dentry->d_op && dentry->d_op->d_revalidate) { - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); (dentry->d_op->d_revalidate)(dentry, nd); - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); } /* diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 2568eb41cb3a..9ccc7d8275b8 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -588,11 +588,11 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, case 2: set_bit(Enabled, &e->flags); break; case 3: root = dget(file->f_vfsmnt->mnt_sb->s_root); - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); kill_node(e); - up(&root->d_inode->i_sem); + mutex_unlock(&root->d_inode->i_mutex); dput(root); break; default: return res; @@ -622,7 +622,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, return PTR_ERR(e); root = dget(sb->s_root); - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); dentry = lookup_one_len(e->name, root, strlen(e->name)); err = PTR_ERR(dentry); if (IS_ERR(dentry)) @@ -658,7 +658,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, out2: dput(dentry); out: - up(&root->d_inode->i_sem); + mutex_unlock(&root->d_inode->i_mutex); dput(root); if (err) { @@ -703,12 +703,12 @@ static ssize_t bm_status_write(struct file * file, const char __user * buffer, case 1: enabled = 0; break; case 2: enabled = 1; break; case 3: root = dget(file->f_vfsmnt->mnt_sb->s_root); - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); while (!list_empty(&entries)) kill_node(list_entry(entries.next, Node, list)); - up(&root->d_inode->i_sem); + mutex_unlock(&root->d_inode->i_mutex); dput(root); default: return res; } diff --git a/fs/block_dev.c b/fs/block_dev.c index e0df94c37b7e..6e50346fb1ee 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -202,7 +202,7 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin) loff_t size; loff_t retval; - down(&bd_inode->i_sem); + mutex_lock(&bd_inode->i_mutex); size = i_size_read(bd_inode); switch (origin) { @@ -219,7 +219,7 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin) } retval = offset; } - up(&bd_inode->i_sem); + mutex_unlock(&bd_inode->i_mutex); return retval; } diff --git a/fs/buffer.c b/fs/buffer.c index 55f0975a9b15..6466bc8a3dc7 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -352,11 +352,11 @@ static long do_fsync(unsigned int fd, int datasync) * We need to protect against concurrent writers, * which could cause livelocks in fsync_buffers_list */ - down(&mapping->host->i_sem); + mutex_lock(&mapping->host->i_mutex); err = file->f_op->fsync(file, file->f_dentry, datasync); if (!ret) ret = err; - up(&mapping->host->i_sem); + mutex_unlock(&mapping->host->i_mutex); err = filemap_fdatawait(mapping); if (!ret) ret = err; @@ -2338,7 +2338,7 @@ int generic_commit_write(struct file *file, struct page *page, __block_commit_write(inode,page,from,to); /* * No need to use i_size_read() here, the i_size - * cannot change under us because we hold i_sem. + * cannot change under us because we hold i_mutex. */ if (pos > inode->i_size) { i_size_write(inode, pos); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 2a13a2bac8f1..e10213b7541e 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -860,9 +860,9 @@ static int cifs_oplock_thread(void * dummyarg) DeleteOplockQEntry(oplock_item); /* can not grab inode sem here since it would deadlock when oplock received on delete - since vfs_unlink holds the i_sem across + since vfs_unlink holds the i_mutex across the call */ - /* down(&inode->i_sem);*/ + /* mutex_lock(&inode->i_mutex);*/ if (S_ISREG(inode->i_mode)) { rc = filemap_fdatawrite(inode->i_mapping); if(CIFS_I(inode)->clientCanCacheRead == 0) { @@ -871,7 +871,7 @@ static int cifs_oplock_thread(void * dummyarg) } } else rc = 0; - /* up(&inode->i_sem);*/ + /* mutex_unlock(&inode->i_mutex);*/ if (rc) CIFS_I(inode)->write_behind_rc = rc; cFYI(1,("Oplock flush inode %p rc %d",inode,rc)); diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 9558f51bca55..3ebce9430f4a 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1040,9 +1040,9 @@ int cifs_revalidate(struct dentry *direntry) } /* can not grab this sem since kernel filesys locking documentation - indicates i_sem may be taken by the kernel on lookup and rename - which could deadlock if we grab the i_sem here as well */ -/* down(&direntry->d_inode->i_sem);*/ + indicates i_mutex may be taken by the kernel on lookup and rename + which could deadlock if we grab the i_mutex here as well */ +/* mutex_lock(&direntry->d_inode->i_mutex);*/ /* need to write out dirty pages here */ if (direntry->d_inode->i_mapping) { /* do we need to lock inode until after invalidate completes @@ -1066,7 +1066,7 @@ int cifs_revalidate(struct dentry *direntry) } } } -/* up(&direntry->d_inode->i_sem); */ +/* mutex_unlock(&direntry->d_inode->i_mutex); */ kfree(full_path); FreeXid(xid); diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 2391766e9c7c..8f1a517f8b4e 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -453,7 +453,7 @@ int coda_readdir(struct file *coda_file, void *dirent, filldir_t filldir) coda_vfs_stat.readdir++; host_inode = host_file->f_dentry->d_inode; - down(&host_inode->i_sem); + mutex_lock(&host_inode->i_mutex); host_file->f_pos = coda_file->f_pos; if (!host_file->f_op->readdir) { @@ -475,7 +475,7 @@ int coda_readdir(struct file *coda_file, void *dirent, filldir_t filldir) } out: coda_file->f_pos = host_file->f_pos; - up(&host_inode->i_sem); + mutex_unlock(&host_inode->i_mutex); return ret; } diff --git a/fs/coda/file.c b/fs/coda/file.c index e6bc022568f3..30b4630bd735 100644 --- a/fs/coda/file.c +++ b/fs/coda/file.c @@ -77,14 +77,14 @@ coda_file_write(struct file *coda_file, const char __user *buf, size_t count, lo return -EINVAL; host_inode = host_file->f_dentry->d_inode; - down(&coda_inode->i_sem); + mutex_lock(&coda_inode->i_mutex); ret = host_file->f_op->write(host_file, buf, count, ppos); coda_inode->i_size = host_inode->i_size; coda_inode->i_blocks = (coda_inode->i_size + 511) >> 9; coda_inode->i_mtime = coda_inode->i_ctime = CURRENT_TIME_SEC; - up(&coda_inode->i_sem); + mutex_unlock(&coda_inode->i_mutex); return ret; } @@ -272,9 +272,9 @@ int coda_fsync(struct file *coda_file, struct dentry *coda_dentry, int datasync) if (host_file->f_op && host_file->f_op->fsync) { host_dentry = host_file->f_dentry; host_inode = host_dentry->d_inode; - down(&host_inode->i_sem); + mutex_lock(&host_inode->i_mutex); err = host_file->f_op->fsync(host_file, host_dentry, datasync); - up(&host_inode->i_sem); + mutex_unlock(&host_inode->i_mutex); } if ( !err && !datasync ) { diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index e48b539243a1..b668ec61527e 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -288,10 +288,10 @@ static struct dentry * configfs_lookup(struct inode *dir, /* * Only subdirectories count here. Files (CONFIGFS_NOT_PINNED) are - * attributes and are removed by rmdir(). We recurse, taking i_sem + * attributes and are removed by rmdir(). We recurse, taking i_mutex * on all children that are candidates for default detach. If the * result is clean, then configfs_detach_group() will handle dropping - * i_sem. If there is an error, the caller will clean up the i_sem + * i_mutex. If there is an error, the caller will clean up the i_mutex * holders via configfs_detach_rollback(). */ static int configfs_detach_prep(struct dentry *dentry) @@ -309,8 +309,8 @@ static int configfs_detach_prep(struct dentry *dentry) if (sd->s_type & CONFIGFS_NOT_PINNED) continue; if (sd->s_type & CONFIGFS_USET_DEFAULT) { - down(&sd->s_dentry->d_inode->i_sem); - /* Mark that we've taken i_sem */ + mutex_lock(&sd->s_dentry->d_inode->i_mutex); + /* Mark that we've taken i_mutex */ sd->s_type |= CONFIGFS_USET_DROPPING; ret = configfs_detach_prep(sd->s_dentry); @@ -327,7 +327,7 @@ out: } /* - * Walk the tree, dropping i_sem wherever CONFIGFS_USET_DROPPING is + * Walk the tree, dropping i_mutex wherever CONFIGFS_USET_DROPPING is * set. */ static void configfs_detach_rollback(struct dentry *dentry) @@ -341,7 +341,7 @@ static void configfs_detach_rollback(struct dentry *dentry) if (sd->s_type & CONFIGFS_USET_DROPPING) { sd->s_type &= ~CONFIGFS_USET_DROPPING; - up(&sd->s_dentry->d_inode->i_sem); + mutex_unlock(&sd->s_dentry->d_inode->i_mutex); } } } @@ -424,11 +424,11 @@ static void detach_groups(struct config_group *group) /* * From rmdir/unregister, a configfs_detach_prep() pass - * has taken our i_sem for us. Drop it. + * has taken our i_mutex for us. Drop it. * From mkdir/register cleanup, there is no sem held. */ if (sd->s_type & CONFIGFS_USET_DROPPING) - up(&child->d_inode->i_sem); + mutex_unlock(&child->d_inode->i_mutex); d_delete(child); dput(child); @@ -493,11 +493,11 @@ static int populate_groups(struct config_group *group) /* FYI, we're faking mkdir here * I'm not sure we need this semaphore, as we're called * from our parent's mkdir. That holds our parent's - * i_sem, so afaik lookup cannot continue through our + * i_mutex, so afaik lookup cannot continue through our * parent to find us, let alone mess with our tree. - * That said, taking our i_sem is closer to mkdir + * That said, taking our i_mutex is closer to mkdir * emulation, and shouldn't hurt. */ - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); for (i = 0; group->default_groups[i]; i++) { new_group = group->default_groups[i]; @@ -507,7 +507,7 @@ static int populate_groups(struct config_group *group) break; } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); } if (ret) @@ -856,7 +856,7 @@ int configfs_rename_dir(struct config_item * item, const char *new_name) down_write(&configfs_rename_sem); parent = item->parent->dentry; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); new_dentry = lookup_one_len(new_name, parent, strlen(new_name)); if (!IS_ERR(new_dentry)) { @@ -872,7 +872,7 @@ int configfs_rename_dir(struct config_item * item, const char *new_name) error = -EEXIST; dput(new_dentry); } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); up_write(&configfs_rename_sem); return error; @@ -884,9 +884,9 @@ static int configfs_dir_open(struct inode *inode, struct file *file) struct dentry * dentry = file->f_dentry; struct configfs_dirent * parent_sd = dentry->d_fsdata; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); file->private_data = configfs_new_dirent(parent_sd, NULL); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return file->private_data ? 0 : -ENOMEM; @@ -897,9 +897,9 @@ static int configfs_dir_close(struct inode *inode, struct file *file) struct dentry * dentry = file->f_dentry; struct configfs_dirent * cursor = file->private_data; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); list_del_init(&cursor->s_sibling); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); release_configfs_dirent(cursor); @@ -975,7 +975,7 @@ static loff_t configfs_dir_lseek(struct file * file, loff_t offset, int origin) { struct dentry * dentry = file->f_dentry; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); switch (origin) { case 1: offset += file->f_pos; @@ -983,7 +983,7 @@ static loff_t configfs_dir_lseek(struct file * file, loff_t offset, int origin) if (offset >= 0) break; default: - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -1007,7 +1007,7 @@ static loff_t configfs_dir_lseek(struct file * file, loff_t offset, int origin) list_add_tail(&cursor->s_sibling, p); } } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return offset; } @@ -1037,7 +1037,7 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) sd = configfs_sb->s_root->d_fsdata; link_group(to_config_group(sd->s_element), group); - down(&configfs_sb->s_root->d_inode->i_sem); + mutex_lock(&configfs_sb->s_root->d_inode->i_mutex); name.name = group->cg_item.ci_name; name.len = strlen(name.name); @@ -1057,7 +1057,7 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) else d_delete(dentry); - up(&configfs_sb->s_root->d_inode->i_sem); + mutex_unlock(&configfs_sb->s_root->d_inode->i_mutex); if (dentry) { dput(dentry); @@ -1079,18 +1079,18 @@ void configfs_unregister_subsystem(struct configfs_subsystem *subsys) return; } - down(&configfs_sb->s_root->d_inode->i_sem); - down(&dentry->d_inode->i_sem); + mutex_lock(&configfs_sb->s_root->d_inode->i_mutex); + mutex_lock(&dentry->d_inode->i_mutex); if (configfs_detach_prep(dentry)) { printk(KERN_ERR "configfs: Tried to unregister non-empty subsystem!\n"); } configfs_detach_group(&group->cg_item); dentry->d_inode->i_flags |= S_DEAD; - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); d_delete(dentry); - up(&configfs_sb->s_root->d_inode->i_sem); + mutex_unlock(&configfs_sb->s_root->d_inode->i_mutex); dput(dentry); diff --git a/fs/configfs/file.c b/fs/configfs/file.c index af1ffc9a15c0..c26cd61f13af 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -336,9 +336,9 @@ int configfs_add_file(struct dentry * dir, const struct configfs_attribute * att umode_t mode = (attr->ca_mode & S_IALLUGO) | S_IFREG; int error = 0; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); error = configfs_make_dirent(parent_sd, NULL, (void *) attr, mode, type); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return error; } diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 6b274c6d428f..6577c588de9d 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -122,7 +122,7 @@ const unsigned char * configfs_get_name(struct configfs_dirent *sd) /* * Unhashes the dentry corresponding to given configfs_dirent - * Called with parent inode's i_sem held. + * Called with parent inode's i_mutex held. */ void configfs_drop_dentry(struct configfs_dirent * sd, struct dentry * parent) { @@ -145,7 +145,7 @@ void configfs_hash_and_remove(struct dentry * dir, const char * name) struct configfs_dirent * sd; struct configfs_dirent * parent_sd = dir->d_fsdata; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); list_for_each_entry(sd, &parent_sd->s_children, s_sibling) { if (!sd->s_element) continue; @@ -156,7 +156,7 @@ void configfs_hash_and_remove(struct dentry * dir, const char * name) break; } } - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index a86ac4aeaedb..d4f1a2cddd47 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -146,7 +146,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode, } *dentry = NULL; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); *dentry = lookup_one_len(name, parent, strlen(name)); if (!IS_ERR(dentry)) { if ((mode & S_IFMT) == S_IFDIR) @@ -155,7 +155,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode, error = debugfs_create(parent->d_inode, *dentry, mode); } else error = PTR_ERR(dentry); - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); return error; } @@ -273,7 +273,7 @@ void debugfs_remove(struct dentry *dentry) if (!parent || !parent->d_inode) return; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); if (debugfs_positive(dentry)) { if (dentry->d_inode) { if (S_ISDIR(dentry->d_inode->i_mode)) @@ -283,7 +283,7 @@ void debugfs_remove(struct dentry *dentry) dput(dentry); } } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); simple_release_fs(&debugfs_mount, &debugfs_mount_count); } EXPORT_SYMBOL_GPL(debugfs_remove); diff --git a/fs/devfs/base.c b/fs/devfs/base.c index 1274422a5384..b621521e09d4 100644 --- a/fs/devfs/base.c +++ b/fs/devfs/base.c @@ -2162,27 +2162,27 @@ static int devfs_d_revalidate_wait(struct dentry *dentry, struct nameidata *nd) * * make sure that * d_instantiate always runs under lock - * we release i_sem lock before going to sleep + * we release i_mutex lock before going to sleep * * unfortunately sometimes d_revalidate is called with - * and sometimes without i_sem lock held. The following checks + * and sometimes without i_mutex lock held. The following checks * attempt to deduce when we need to add (and drop resp.) lock * here. This relies on current (2.6.2) calling coventions: * - * lookup_hash is always run under i_sem and is passing NULL + * lookup_hash is always run under i_mutex and is passing NULL * as nd * - * open(...,O_CREATE,...) calls _lookup_hash under i_sem + * open(...,O_CREATE,...) calls _lookup_hash under i_mutex * and sets flags to LOOKUP_OPEN|LOOKUP_CREATE * * all other invocations of ->d_revalidate seem to happen - * outside of i_sem + * outside of i_mutex */ need_lock = nd && (!(nd->flags & LOOKUP_CREATE) || (nd->flags & LOOKUP_PARENT)); if (need_lock) - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); if (is_devfsd_or_child(fs_info)) { devfs_handle_t de = lookup_info->de; @@ -2221,9 +2221,9 @@ static int devfs_d_revalidate_wait(struct dentry *dentry, struct nameidata *nd) add_wait_queue(&lookup_info->wait_queue, &wait); read_unlock(&parent->u.dir.lock); /* at this point it is always (hopefully) locked */ - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); schedule(); - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); /* * This does not need nor should remove wait from wait_queue. * Wait queue head is never reused - nothing is ever added to it @@ -2238,7 +2238,7 @@ static int devfs_d_revalidate_wait(struct dentry *dentry, struct nameidata *nd) out: if (need_lock) - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); return 1; } /* End Function devfs_d_revalidate_wait */ @@ -2284,9 +2284,9 @@ static struct dentry *devfs_lookup(struct inode *dir, struct dentry *dentry, /* Unlock directory semaphore, which will release any waiters. They will get the hashed dentry, and may be forced to wait for revalidation */ - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); wait_for_devfsd_finished(fs_info); /* If I'm not devfsd, must wait */ - down(&dir->i_sem); /* Grab it again because them's the rules */ + mutex_lock(&dir->i_mutex); /* Grab it again because them's the rules */ de = lookup_info.de; /* If someone else has been so kind as to make the inode, we go home early */ diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index f2be44d4491f..bfb8a230bac9 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -130,7 +130,7 @@ static struct dentry *get_node(int num) { char s[12]; struct dentry *root = devpts_root; - down(&root->d_inode->i_sem); + mutex_lock(&root->d_inode->i_mutex); return lookup_one_len(s, root, sprintf(s, "%d", num)); } @@ -161,7 +161,7 @@ int devpts_pty_new(struct tty_struct *tty) if (!IS_ERR(dentry) && !dentry->d_inode) d_instantiate(dentry, inode); - up(&devpts_root->d_inode->i_sem); + mutex_unlock(&devpts_root->d_inode->i_mutex); return 0; } @@ -178,7 +178,7 @@ struct tty_struct *devpts_get_tty(int number) dput(dentry); } - up(&devpts_root->d_inode->i_sem); + mutex_unlock(&devpts_root->d_inode->i_mutex); return tty; } @@ -196,7 +196,7 @@ void devpts_pty_kill(int number) } dput(dentry); } - up(&devpts_root->d_inode->i_sem); + mutex_unlock(&devpts_root->d_inode->i_mutex); } static int __init init_devpts_fs(void) diff --git a/fs/direct-io.c b/fs/direct-io.c index 3931e7f1e6bf..30dbbd1df511 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -56,7 +56,7 @@ * lock_type is DIO_LOCKING for regular files on direct-IO-naive filesystems. * This determines whether we need to do the fancy locking which prevents * direct-IO from being able to read uninitialised disk blocks. If its zero - * (blockdev) this locking is not done, and if it is DIO_OWN_LOCKING i_sem is + * (blockdev) this locking is not done, and if it is DIO_OWN_LOCKING i_mutex is * not held for the entire direct write (taken briefly, initially, during a * direct read though, but its never held for the duration of a direct-IO). */ @@ -930,7 +930,7 @@ out: } /* - * Releases both i_sem and i_alloc_sem + * Releases both i_mutex and i_alloc_sem */ static ssize_t direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode, @@ -1062,11 +1062,11 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode, /* * All block lookups have been performed. For READ requests - * we can let i_sem go now that its achieved its purpose + * we can let i_mutex go now that its achieved its purpose * of protecting us from looking up uninitialized blocks. */ if ((rw == READ) && (dio->lock_type == DIO_LOCKING)) - up(&dio->inode->i_sem); + mutex_unlock(&dio->inode->i_mutex); /* * OK, all BIOs are submitted, so we can decrement bio_count to truly @@ -1145,18 +1145,18 @@ direct_io_worker(int rw, struct kiocb *iocb, struct inode *inode, * The locking rules are governed by the dio_lock_type parameter. * * DIO_NO_LOCKING (no locking, for raw block device access) - * For writes, i_sem is not held on entry; it is never taken. + * For writes, i_mutex is not held on entry; it is never taken. * * DIO_LOCKING (simple locking for regular files) - * For writes we are called under i_sem and return with i_sem held, even though + * For writes we are called under i_mutex and return with i_mutex held, even though * it is internally dropped. - * For reads, i_sem is not held on entry, but it is taken and dropped before + * For reads, i_mutex is not held on entry, but it is taken and dropped before * returning. * * DIO_OWN_LOCKING (filesystem provides synchronisation and handling of * uninitialised data, allowing parallel direct readers and writers) - * For writes we are called without i_sem, return without it, never touch it. - * For reads, i_sem is held on entry and will be released before returning. + * For writes we are called without i_mutex, return without it, never touch it. + * For reads, i_mutex is held on entry and will be released before returning. * * Additional i_alloc_sem locking requirements described inline below. */ @@ -1214,11 +1214,11 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, * For block device access DIO_NO_LOCKING is used, * neither readers nor writers do any locking at all * For regular files using DIO_LOCKING, - * readers need to grab i_sem and i_alloc_sem - * writers need to grab i_alloc_sem only (i_sem is already held) + * readers need to grab i_mutex and i_alloc_sem + * writers need to grab i_alloc_sem only (i_mutex is already held) * For regular files using DIO_OWN_LOCKING, * neither readers nor writers take any locks here - * (i_sem is already held and release for writers here) + * (i_mutex is already held and release for writers here) */ dio->lock_type = dio_lock_type; if (dio_lock_type != DIO_NO_LOCKING) { @@ -1228,7 +1228,7 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, mapping = iocb->ki_filp->f_mapping; if (dio_lock_type != DIO_OWN_LOCKING) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); reader_with_isem = 1; } @@ -1240,7 +1240,7 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, } if (dio_lock_type == DIO_OWN_LOCKING) { - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); reader_with_isem = 0; } } @@ -1266,7 +1266,7 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, out: if (reader_with_isem) - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (rw & WRITE) current->flags &= ~PF_SYNCWRITE; return retval; diff --git a/fs/dquot.c b/fs/dquot.c index 2a62b3dc20ec..cb6d5bfbdfd5 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -100,7 +100,7 @@ * operation is just reading pointers from inode (or not using them at all) the * read lock is enough. If pointers are altered function must hold write lock * (these locking rules also apply for S_NOQUOTA flag in the inode - note that - * for altering the flag i_sem is also needed). If operation is holding + * for altering the flag i_mutex is also needed). If operation is holding * reference to dquot in other way (e.g. quotactl ops) it must be guarded by * dqonoff_sem. * This locking assures that: @@ -117,9 +117,9 @@ * spinlock to internal buffers before writing. * * Lock ordering (including related VFS locks) is the following: - * i_sem > dqonoff_sem > iprune_sem > journal_lock > dqptr_sem > + * i_mutex > dqonoff_sem > iprune_sem > journal_lock > dqptr_sem > * > dquot->dq_lock > dqio_sem - * i_sem on quota files is special (it's below dqio_sem) + * i_mutex on quota files is special (it's below dqio_sem) */ static DEFINE_SPINLOCK(dq_list_lock); @@ -1369,11 +1369,11 @@ int vfs_quota_off(struct super_block *sb, int type) /* If quota was reenabled in the meantime, we have * nothing to do */ if (!sb_has_quota_enabled(sb, cnt)) { - down(&toputinode[cnt]->i_sem); + mutex_lock(&toputinode[cnt]->i_mutex); toputinode[cnt]->i_flags &= ~(S_IMMUTABLE | S_NOATIME | S_NOQUOTA); truncate_inode_pages(&toputinode[cnt]->i_data, 0); - up(&toputinode[cnt]->i_sem); + mutex_unlock(&toputinode[cnt]->i_mutex); mark_inode_dirty(toputinode[cnt]); iput(toputinode[cnt]); } @@ -1417,7 +1417,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id) write_inode_now(inode, 1); /* And now flush the block cache so that kernel sees the changes */ invalidate_bdev(sb->s_bdev, 0); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); down(&dqopt->dqonoff_sem); if (sb_has_quota_enabled(sb, type)) { error = -EBUSY; @@ -1449,7 +1449,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id) goto out_file_init; } up(&dqopt->dqio_sem); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); set_enable_flags(dqopt, type); add_dquot_ref(sb, type); @@ -1470,7 +1470,7 @@ out_lock: inode->i_flags |= oldflags; up_write(&dqopt->dqptr_sem); } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out_fmt: put_quota_format(fmt); diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index c49d6254379a..5bfe40085fbc 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -177,9 +177,9 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, struct dentry *ppd; struct dentry *npd; - down(&pd->d_inode->i_sem); + mutex_lock(&pd->d_inode->i_mutex); ppd = CALL(nops,get_parent)(pd); - up(&pd->d_inode->i_sem); + mutex_unlock(&pd->d_inode->i_mutex); if (IS_ERR(ppd)) { err = PTR_ERR(ppd); @@ -201,9 +201,9 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, break; } dprintk("find_exported_dentry: found name: %s\n", nbuf); - down(&ppd->d_inode->i_sem); + mutex_lock(&ppd->d_inode->i_mutex); npd = lookup_one_len(nbuf, ppd, strlen(nbuf)); - up(&ppd->d_inode->i_sem); + mutex_unlock(&ppd->d_inode->i_mutex); if (IS_ERR(npd)) { err = PTR_ERR(npd); dprintk("find_exported_dentry: lookup failed: %d\n", err); @@ -242,9 +242,9 @@ find_exported_dentry(struct super_block *sb, void *obj, void *parent, struct dentry *nresult; err = CALL(nops,get_name)(target_dir, nbuf, result); if (!err) { - down(&target_dir->d_inode->i_sem); + mutex_lock(&target_dir->d_inode->i_mutex); nresult = lookup_one_len(nbuf, target_dir, strlen(nbuf)); - up(&target_dir->d_inode->i_sem); + mutex_unlock(&target_dir->d_inode->i_mutex); if (!IS_ERR(nresult)) { if (nresult->d_inode) { dput(result); diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index 6af2f4130290..239133d01d91 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -149,7 +149,7 @@ ext2_iset_acl(struct inode *inode, struct posix_acl **i_acl, } /* - * inode->i_sem: don't care + * inode->i_mutex: don't care */ static struct posix_acl * ext2_get_acl(struct inode *inode, int type) @@ -211,7 +211,7 @@ ext2_get_acl(struct inode *inode, int type) } /* - * inode->i_sem: down + * inode->i_mutex: down */ static int ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl) @@ -301,8 +301,8 @@ ext2_permission(struct inode *inode, int mask, struct nameidata *nd) /* * Initialize the ACLs of a new inode. Called from ext2_new_inode. * - * dir->i_sem: down - * inode->i_sem: up (access to inode is still exclusive) + * dir->i_mutex: down + * inode->i_mutex: up (access to inode is still exclusive) */ int ext2_init_acl(struct inode *inode, struct inode *dir) @@ -361,7 +361,7 @@ cleanup: * for directories) are added. There are no more bits available in the * file mode. * - * inode->i_sem: down + * inode->i_mutex: down */ int ext2_acl_chmod(struct inode *inode) diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index e977f8566d14..00de0a7312a2 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -53,7 +53,7 @@ struct ext2_inode_info { #ifdef CONFIG_EXT2_FS_XATTR /* * Extended attributes can be read independently of the main file - * data. Taking i_sem even when reading would cause contention + * data. Taking i_mutex even when reading would cause contention * between readers of EAs and writers of regular file data, so * instead we synchronize on xattr_sem when reading or changing * EAs. diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 522fa70dd8ea..8d6819846fc9 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1152,7 +1152,7 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type, struct buffer_head tmp_bh; struct buffer_head *bh; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; @@ -1189,7 +1189,7 @@ out: inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME; mark_inode_dirty(inode); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return len - towrite; } diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c index 0099462d4271..f7a3b5fee274 100644 --- a/fs/ext2/xattr.c +++ b/fs/ext2/xattr.c @@ -325,7 +325,7 @@ cleanup: /* * Inode operation listxattr() * - * dentry->d_inode->i_sem: don't care + * dentry->d_inode->i_mutex: don't care */ ssize_t ext2_listxattr(struct dentry *dentry, char *buffer, size_t size) diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index 3ac38266fc9e..9ed132c96034 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -152,7 +152,7 @@ ext3_iset_acl(struct inode *inode, struct posix_acl **i_acl, /* * Inode operation get_posix_acl(). * - * inode->i_sem: don't care + * inode->i_mutex: don't care */ static struct posix_acl * ext3_get_acl(struct inode *inode, int type) @@ -216,7 +216,7 @@ ext3_get_acl(struct inode *inode, int type) /* * Set the access or default ACL of an inode. * - * inode->i_sem: down unless called from ext3_new_inode + * inode->i_mutex: down unless called from ext3_new_inode */ static int ext3_set_acl(handle_t *handle, struct inode *inode, int type, @@ -306,8 +306,8 @@ ext3_permission(struct inode *inode, int mask, struct nameidata *nd) /* * Initialize the ACLs of a new inode. Called from ext3_new_inode. * - * dir->i_sem: down - * inode->i_sem: up (access to inode is still exclusive) + * dir->i_mutex: down + * inode->i_mutex: up (access to inode is still exclusive) */ int ext3_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) @@ -368,7 +368,7 @@ cleanup: * for directories) are added. There are no more bits available in the * file mode. * - * inode->i_sem: down + * inode->i_mutex: down */ int ext3_acl_chmod(struct inode *inode) diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 7c45acf94589..56bf76586019 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -2150,7 +2150,7 @@ int ext3_force_commit(struct super_block *sb) static void ext3_write_super (struct super_block * sb) { - if (down_trylock(&sb->s_lock) == 0) + if (mutex_trylock(&sb->s_lock) != 0) BUG(); sb->s_dirt = 0; } @@ -2601,7 +2601,7 @@ static ssize_t ext3_quota_write(struct super_block *sb, int type, struct buffer_head *bh; handle_t *handle = journal_current_handle(); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; @@ -2644,7 +2644,7 @@ out: inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME; ext3_mark_inode_dirty(handle, inode); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return len - towrite; } diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 430de9f63be3..238199d82ce5 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -140,7 +140,7 @@ ext3_xattr_handler(int name_index) /* * Inode operation listxattr() * - * dentry->d_inode->i_sem: don't care + * dentry->d_inode->i_mutex: don't care */ ssize_t ext3_listxattr(struct dentry *dentry, char *buffer, size_t size) diff --git a/fs/fat/dir.c b/fs/fat/dir.c index eef1b81aa294..db0de5c621c7 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -729,13 +729,13 @@ static int fat_dir_ioctl(struct inode * inode, struct file * filp, buf.dirent = d1; buf.result = 0; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = -ENOENT; if (!IS_DEADDIR(inode)) { ret = __fat_readdir(inode, filp, &buf, fat_ioctl_filldir, short_only, both); } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret >= 0) ret = buf.result; return ret; diff --git a/fs/fat/file.c b/fs/fat/file.c index 9b07c328a6fc..d30876cf35f5 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -41,7 +41,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, if (err) return err; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (IS_RDONLY(inode)) { err = -EROFS; @@ -103,7 +103,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp, MSDOS_I(inode)->i_attrs = attr & ATTR_UNUSED; mark_inode_dirty(inode); up: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return err; } default: diff --git a/fs/fifo.c b/fs/fifo.c index 5455916241f0..923371b753ab 100644 --- a/fs/fifo.c +++ b/fs/fifo.c @@ -35,7 +35,7 @@ static int fifo_open(struct inode *inode, struct file *filp) int ret; ret = -ERESTARTSYS; - if (down_interruptible(PIPE_SEM(*inode))) + if (mutex_lock_interruptible(PIPE_MUTEX(*inode))) goto err_nolock_nocleanup; if (!inode->i_pipe) { @@ -119,7 +119,7 @@ static int fifo_open(struct inode *inode, struct file *filp) } /* Ok! */ - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return 0; err_rd: @@ -139,7 +139,7 @@ err: free_pipe_info(inode); err_nocleanup: - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); err_nolock_nocleanup: return ret; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 05dedddf4289..63d2980df5c9 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -560,9 +560,9 @@ static ssize_t fuse_direct_write(struct file *file, const char __user *buf, struct inode *inode = file->f_dentry->d_inode; ssize_t res; /* Don't allow parallel writes to the same file */ - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); res = fuse_direct_io(file, buf, count, ppos, 1); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return res; } diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index d499393a8ae7..050a49276499 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -547,13 +547,13 @@ static int hfs_file_release(struct inode *inode, struct file *file) if (atomic_read(&file->f_count) != 0) return 0; if (atomic_dec_and_test(&HFS_I(inode)->opencnt)) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); hfs_file_truncate(inode); //if (inode->i_flags & S_DEAD) { // hfs_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL); // hfs_delete_inode(inode); //} - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } return 0; } diff --git a/fs/hfsplus/bitmap.c b/fs/hfsplus/bitmap.c index c7d316455fa0..9fb51632303c 100644 --- a/fs/hfsplus/bitmap.c +++ b/fs/hfsplus/bitmap.c @@ -29,7 +29,7 @@ int hfsplus_block_allocate(struct super_block *sb, u32 size, u32 offset, u32 *ma return size; dprint(DBG_BITMAP, "block_allocate: %u,%u,%u\n", size, offset, len); - down(&HFSPLUS_SB(sb).alloc_file->i_sem); + mutex_lock(&HFSPLUS_SB(sb).alloc_file->i_mutex); mapping = HFSPLUS_SB(sb).alloc_file->i_mapping; page = read_cache_page(mapping, offset / PAGE_CACHE_BITS, (filler_t *)mapping->a_ops->readpage, NULL); @@ -143,7 +143,7 @@ done: sb->s_dirt = 1; dprint(DBG_BITMAP, "-> %u,%u\n", start, *max); out: - up(&HFSPLUS_SB(sb).alloc_file->i_sem); + mutex_unlock(&HFSPLUS_SB(sb).alloc_file->i_mutex); return start; } @@ -164,7 +164,7 @@ int hfsplus_block_free(struct super_block *sb, u32 offset, u32 count) if ((offset + count) > HFSPLUS_SB(sb).total_blocks) return -2; - down(&HFSPLUS_SB(sb).alloc_file->i_sem); + mutex_lock(&HFSPLUS_SB(sb).alloc_file->i_mutex); mapping = HFSPLUS_SB(sb).alloc_file->i_mapping; pnr = offset / PAGE_CACHE_BITS; page = read_cache_page(mapping, pnr, (filler_t *)mapping->a_ops->readpage, NULL); @@ -215,7 +215,7 @@ out: kunmap(page); HFSPLUS_SB(sb).free_blocks += len; sb->s_dirt = 1; - up(&HFSPLUS_SB(sb).alloc_file->i_sem); + mutex_unlock(&HFSPLUS_SB(sb).alloc_file->i_mutex); return 0; } diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index fc98583cf045..983bcd02ac1c 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -276,13 +276,13 @@ static int hfsplus_file_release(struct inode *inode, struct file *file) if (atomic_read(&file->f_count) != 0) return 0; if (atomic_dec_and_test(&HFSPLUS_I(inode).opencnt)) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); hfsplus_file_truncate(inode); if (inode->i_flags & S_DEAD) { hfsplus_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL); hfsplus_delete_inode(inode); } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } return 0; } diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c index 0217c3a04441..5591f9623aa2 100644 --- a/fs/hpfs/dir.c +++ b/fs/hpfs/dir.c @@ -32,19 +32,19 @@ static loff_t hpfs_dir_lseek(struct file *filp, loff_t off, int whence) /*printk("dir lseek\n");*/ if (new_off == 0 || new_off == 1 || new_off == 11 || new_off == 12 || new_off == 13) goto ok; - down(&i->i_sem); + mutex_lock(&i->i_mutex); pos = ((loff_t) hpfs_de_as_down_as_possible(s, hpfs_inode->i_dno) << 4) + 1; while (pos != new_off) { if (map_pos_dirent(i, &pos, &qbh)) hpfs_brelse4(&qbh); else goto fail; if (pos == 12) goto fail; } - up(&i->i_sem); + mutex_unlock(&i->i_mutex); ok: unlock_kernel(); return filp->f_pos = new_off; fail: - up(&i->i_sem); + mutex_unlock(&i->i_mutex); /*printk("illegal lseek: %016llx\n", new_off);*/ unlock_kernel(); return -ESPIPE; diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c index 52930915bad8..a44dc5897399 100644 --- a/fs/hppfs/hppfs_kern.c +++ b/fs/hppfs/hppfs_kern.c @@ -171,12 +171,12 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, err = -ENOMEM; parent = HPPFS_I(ino)->proc_dentry; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); proc_dentry = d_lookup(parent, &dentry->d_name); if(proc_dentry == NULL){ proc_dentry = d_alloc(parent, &dentry->d_name); if(proc_dentry == NULL){ - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); goto out; } new = (*parent->d_inode->i_op->lookup)(parent->d_inode, @@ -186,7 +186,7 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, proc_dentry = new; } } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); if(IS_ERR(proc_dentry)) return(proc_dentry); diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 8c41315a6e42..ff1b7d108bd0 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -118,7 +118,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) vma_len = (loff_t)(vma->vm_end - vma->vm_start); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); file_accessed(file); vma->vm_flags |= VM_HUGETLB | VM_RESERVED; vma->vm_ops = &hugetlb_vm_ops; @@ -133,7 +133,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) if (inode->i_size < len) inode->i_size = len; out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return ret; } diff --git a/fs/inode.c b/fs/inode.c index fd568caf7f74..e08767fd57b0 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -192,7 +192,7 @@ void inode_init_once(struct inode *inode) INIT_HLIST_NODE(&inode->i_hash); INIT_LIST_HEAD(&inode->i_dentry); INIT_LIST_HEAD(&inode->i_devices); - sema_init(&inode->i_sem, 1); + mutex_init(&inode->i_mutex); init_rwsem(&inode->i_alloc_sem); INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC); rwlock_init(&inode->i_data.tree_lock); diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 2559ee10beda..fc3855a1aef3 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -1415,7 +1415,7 @@ jffs_file_write(struct file *filp, const char *buf, size_t count, * This will never trigger with sane page sizes. leave it in * anyway, since I'm thinking about how to merge larger writes * (the current idea is to poke a thread that does the actual - * I/O and starts by doing a down(&inode->i_sem). then we + * I/O and starts by doing a mutex_lock(&inode->i_mutex). then we * would need to get the page cache pages and have a list of * I/O requests and do write-merging here. * -- prumpf diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h index c0fd7b3eadc6..dc21a5bd54d4 100644 --- a/fs/jfs/jfs_incore.h +++ b/fs/jfs/jfs_incore.h @@ -58,7 +58,7 @@ struct jfs_inode_info { /* * rdwrlock serializes xtree between reads & writes and synchronizes * changes to special inodes. It's use would be redundant on - * directories since the i_sem taken in the VFS is sufficient. + * directories since the i_mutex taken in the VFS is sufficient. */ struct rw_semaphore rdwrlock; /* @@ -68,7 +68,7 @@ struct jfs_inode_info { * inode is blocked in txBegin or TxBeginAnon */ struct semaphore commit_sem; - /* xattr_sem allows us to access the xattrs without taking i_sem */ + /* xattr_sem allows us to access the xattrs without taking i_mutex */ struct rw_semaphore xattr_sem; lid_t xtlid; /* lid of xtree lock on directory */ #ifdef CONFIG_JFS_POSIX_ACL diff --git a/fs/libfs.c b/fs/libfs.c index 9c50523382e7..63c020e6589e 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -74,7 +74,7 @@ int dcache_dir_close(struct inode *inode, struct file *file) loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) { - down(&file->f_dentry->d_inode->i_sem); + mutex_lock(&file->f_dentry->d_inode->i_mutex); switch (origin) { case 1: offset += file->f_pos; @@ -82,7 +82,7 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) if (offset >= 0) break; default: - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -106,7 +106,7 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) spin_unlock(&dcache_lock); } } - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return offset; } @@ -356,7 +356,7 @@ int simple_commit_write(struct file *file, struct page *page, /* * No need to use i_size_read() here, the i_size - * cannot change under us because we hold the i_sem. + * cannot change under us because we hold the i_mutex. */ if (pos > inode->i_size) i_size_write(inode, pos); diff --git a/fs/namei.c b/fs/namei.c index 300eae088d5f..0a8f073435af 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -438,7 +438,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s struct dentry * result; struct inode *dir = parent->d_inode; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); /* * First re-do the cached lookup just in case it was created * while we waited for the directory semaphore.. @@ -464,7 +464,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s else result = dentry; } - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); return result; } @@ -472,7 +472,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s * Uhhuh! Nasty case: the cache was re-populated while * we waited on the semaphore. Need to revalidate. */ - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); if (result->d_op && result->d_op->d_revalidate) { if (!result->d_op->d_revalidate(result, nd) && !d_invalidate(result)) { dput(result); @@ -1366,7 +1366,7 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2) struct dentry *p; if (p1 == p2) { - down(&p1->d_inode->i_sem); + mutex_lock(&p1->d_inode->i_mutex); return NULL; } @@ -1374,30 +1374,30 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2) for (p = p1; p->d_parent != p; p = p->d_parent) { if (p->d_parent == p2) { - down(&p2->d_inode->i_sem); - down(&p1->d_inode->i_sem); + mutex_lock(&p2->d_inode->i_mutex); + mutex_lock(&p1->d_inode->i_mutex); return p; } } for (p = p2; p->d_parent != p; p = p->d_parent) { if (p->d_parent == p1) { - down(&p1->d_inode->i_sem); - down(&p2->d_inode->i_sem); + mutex_lock(&p1->d_inode->i_mutex); + mutex_lock(&p2->d_inode->i_mutex); return p; } } - down(&p1->d_inode->i_sem); - down(&p2->d_inode->i_sem); + mutex_lock(&p1->d_inode->i_mutex); + mutex_lock(&p2->d_inode->i_mutex); return NULL; } void unlock_rename(struct dentry *p1, struct dentry *p2) { - up(&p1->d_inode->i_sem); + mutex_unlock(&p1->d_inode->i_mutex); if (p1 != p2) { - up(&p2->d_inode->i_sem); + mutex_unlock(&p2->d_inode->i_mutex); up(&p1->d_inode->i_sb->s_vfs_rename_sem); } } @@ -1563,14 +1563,14 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) dir = nd->dentry; nd->flags &= ~LOOKUP_PARENT; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); path.dentry = lookup_hash(nd); path.mnt = nd->mnt; do_last: error = PTR_ERR(path.dentry); if (IS_ERR(path.dentry)) { - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); goto exit; } @@ -1579,7 +1579,7 @@ do_last: if (!IS_POSIXACL(dir->d_inode)) mode &= ~current->fs->umask; error = vfs_create(dir->d_inode, path.dentry, mode, nd); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); dput(nd->dentry); nd->dentry = path.dentry; if (error) @@ -1593,7 +1593,7 @@ do_last: /* * It already exists. */ - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); error = -EEXIST; if (flag & O_EXCL) @@ -1665,7 +1665,7 @@ do_link: goto exit; } dir = nd->dentry; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); path.dentry = lookup_hash(nd); path.mnt = nd->mnt; __putname(nd->last.name); @@ -1680,13 +1680,13 @@ do_link: * Simple function to lookup and return a dentry and create it * if it doesn't exist. Is SMP-safe. * - * Returns with nd->dentry->d_inode->i_sem locked. + * Returns with nd->dentry->d_inode->i_mutex locked. */ struct dentry *lookup_create(struct nameidata *nd, int is_dir) { struct dentry *dentry = ERR_PTR(-EEXIST); - down(&nd->dentry->d_inode->i_sem); + mutex_lock(&nd->dentry->d_inode->i_mutex); /* * Yucky last component or no last component at all? * (foo/., foo/.., /////) @@ -1784,7 +1784,7 @@ asmlinkage long sys_mknod(const char __user * filename, int mode, unsigned dev) } dput(dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); path_release(&nd); out: putname(tmp); @@ -1836,7 +1836,7 @@ asmlinkage long sys_mkdir(const char __user * pathname, int mode) error = vfs_mkdir(nd.dentry->d_inode, dentry, mode); dput(dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); path_release(&nd); out: putname(tmp); @@ -1885,7 +1885,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) DQUOT_INIT(dir); - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); dentry_unhash(dentry); if (d_mountpoint(dentry)) error = -EBUSY; @@ -1897,7 +1897,7 @@ int vfs_rmdir(struct inode *dir, struct dentry *dentry) dentry->d_inode->i_flags |= S_DEAD; } } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); if (!error) { d_delete(dentry); } @@ -1932,14 +1932,14 @@ asmlinkage long sys_rmdir(const char __user * pathname) error = -EBUSY; goto exit1; } - down(&nd.dentry->d_inode->i_sem); + mutex_lock(&nd.dentry->d_inode->i_mutex); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { error = vfs_rmdir(nd.dentry->d_inode, dentry); dput(dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); exit1: path_release(&nd); exit: @@ -1959,7 +1959,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) DQUOT_INIT(dir); - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); if (d_mountpoint(dentry)) error = -EBUSY; else { @@ -1967,7 +1967,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) if (!error) error = dir->i_op->unlink(dir, dentry); } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); /* We don't d_delete() NFS sillyrenamed files--they still exist. */ if (!error && !(dentry->d_flags & DCACHE_NFSFS_RENAMED)) { @@ -1979,7 +1979,7 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry) /* * Make sure that the actual truncation of the file will occur outside its - * directory's i_sem. Truncate can take a long time if there is a lot of + * directory's i_mutex. Truncate can take a long time if there is a lot of * writeout happening, and we don't want to prevent access to the directory * while waiting on the I/O. */ @@ -2001,7 +2001,7 @@ asmlinkage long sys_unlink(const char __user * pathname) error = -EISDIR; if (nd.last_type != LAST_NORM) goto exit1; - down(&nd.dentry->d_inode->i_sem); + mutex_lock(&nd.dentry->d_inode->i_mutex); dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { @@ -2015,7 +2015,7 @@ asmlinkage long sys_unlink(const char __user * pathname) exit2: dput(dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); if (inode) iput(inode); /* truncate the inode here */ exit1: @@ -2075,7 +2075,7 @@ asmlinkage long sys_symlink(const char __user * oldname, const char __user * new error = vfs_symlink(nd.dentry->d_inode, dentry, from, S_IALLUGO); dput(dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); path_release(&nd); out: putname(to); @@ -2113,10 +2113,10 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de if (error) return error; - down(&old_dentry->d_inode->i_sem); + mutex_lock(&old_dentry->d_inode->i_mutex); DQUOT_INIT(dir); error = dir->i_op->link(old_dentry, dir, new_dentry); - up(&old_dentry->d_inode->i_sem); + mutex_unlock(&old_dentry->d_inode->i_mutex); if (!error) fsnotify_create(dir, new_dentry->d_name.name); return error; @@ -2157,7 +2157,7 @@ asmlinkage long sys_link(const char __user * oldname, const char __user * newnam error = vfs_link(old_nd.dentry, nd.dentry->d_inode, new_dentry); dput(new_dentry); } - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); out_release: path_release(&nd); out: @@ -2178,7 +2178,7 @@ exit: * sb->s_vfs_rename_sem. We might be more accurate, but that's another * story. * c) we have to lock _three_ objects - parents and victim (if it exists). - * And that - after we got ->i_sem on parents (until then we don't know + * And that - after we got ->i_mutex on parents (until then we don't know * whether the target exists). Solution: try to be smart with locking * order for inodes. We rely on the fact that tree topology may change * only under ->s_vfs_rename_sem _and_ that parent of the object we @@ -2195,9 +2195,9 @@ exit: * stuff into VFS), but the former is not going away. Solution: the same * trick as in rmdir(). * e) conversion from fhandle to dentry may come in the wrong moment - when - * we are removing the target. Solution: we will have to grab ->i_sem + * we are removing the target. Solution: we will have to grab ->i_mutex * in the fhandle_to_dentry code. [FIXME - current nfsfh.c relies on - * ->i_sem on parents, which works but leads to some truely excessive + * ->i_mutex on parents, which works but leads to some truely excessive * locking]. */ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, @@ -2222,7 +2222,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, target = new_dentry->d_inode; if (target) { - down(&target->i_sem); + mutex_lock(&target->i_mutex); dentry_unhash(new_dentry); } if (d_mountpoint(old_dentry)||d_mountpoint(new_dentry)) @@ -2232,7 +2232,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry, if (target) { if (!error) target->i_flags |= S_DEAD; - up(&target->i_sem); + mutex_unlock(&target->i_mutex); if (d_unhashed(new_dentry)) d_rehash(new_dentry); dput(new_dentry); @@ -2255,7 +2255,7 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, dget(new_dentry); target = new_dentry->d_inode; if (target) - down(&target->i_sem); + mutex_lock(&target->i_mutex); if (d_mountpoint(old_dentry)||d_mountpoint(new_dentry)) error = -EBUSY; else @@ -2266,7 +2266,7 @@ static int vfs_rename_other(struct inode *old_dir, struct dentry *old_dentry, d_move(old_dentry, new_dentry); } if (target) - up(&target->i_sem); + mutex_unlock(&target->i_mutex); dput(new_dentry); return error; } diff --git a/fs/namespace.c b/fs/namespace.c index 3e8fb61ad597..f0e353f5bc30 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -814,7 +814,7 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) return -ENOTDIR; err = -ENOENT; - down(&nd->dentry->d_inode->i_sem); + mutex_lock(&nd->dentry->d_inode->i_mutex); if (IS_DEADDIR(nd->dentry->d_inode)) goto out_unlock; @@ -826,7 +826,7 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) err = attach_recursive_mnt(mnt, nd, NULL); out_unlock: - up(&nd->dentry->d_inode->i_sem); + mutex_unlock(&nd->dentry->d_inode->i_mutex); if (!err) security_sb_post_addmount(mnt, nd); return err; @@ -962,7 +962,7 @@ static int do_move_mount(struct nameidata *nd, char *old_name) goto out; err = -ENOENT; - down(&nd->dentry->d_inode->i_sem); + mutex_lock(&nd->dentry->d_inode->i_mutex); if (IS_DEADDIR(nd->dentry->d_inode)) goto out1; @@ -1004,7 +1004,7 @@ static int do_move_mount(struct nameidata *nd, char *old_name) list_del_init(&old_nd.mnt->mnt_expire); spin_unlock(&vfsmount_lock); out1: - up(&nd->dentry->d_inode->i_sem); + mutex_unlock(&nd->dentry->d_inode->i_mutex); out: up_write(&namespace_sem); if (!err) @@ -1573,7 +1573,7 @@ asmlinkage long sys_pivot_root(const char __user * new_root, user_nd.dentry = dget(current->fs->root); read_unlock(¤t->fs->lock); down_write(&namespace_sem); - down(&old_nd.dentry->d_inode->i_sem); + mutex_lock(&old_nd.dentry->d_inode->i_mutex); error = -EINVAL; if (IS_MNT_SHARED(old_nd.mnt) || IS_MNT_SHARED(new_nd.mnt->mnt_parent) || @@ -1626,7 +1626,7 @@ asmlinkage long sys_pivot_root(const char __user * new_root, path_release(&root_parent); path_release(&parent_nd); out2: - up(&old_nd.dentry->d_inode->i_sem); + mutex_unlock(&old_nd.dentry->d_inode->i_mutex); up_write(&namespace_sem); path_release(&user_nd); path_release(&old_nd); diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index e9255198f767..a1554bead692 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -194,7 +194,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) spin_unlock(&inode->i_lock); /* Ensure consistent page alignment of the data. * Note: assumes we have exclusive access to this mapping either - * through inode->i_sem or some other mechanism. + * through inode->i_mutex or some other mechanism. */ if (page->index == 0) invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1); @@ -573,7 +573,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin) { - down(&filp->f_dentry->d_inode->i_sem); + mutex_lock(&filp->f_dentry->d_inode->i_mutex); switch (origin) { case 1: offset += filp->f_pos; @@ -589,7 +589,7 @@ loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int origin) ((struct nfs_open_context *)filp->private_data)->dir_cookie = 0; } out: - up(&filp->f_dentry->d_inode->i_sem); + mutex_unlock(&filp->f_dentry->d_inode->i_mutex); return offset; } @@ -1001,7 +1001,7 @@ static int nfs_open_revalidate(struct dentry *dentry, struct nameidata *nd) openflags &= ~(O_CREAT|O_TRUNC); /* - * Note: we're not holding inode->i_sem and so may be racing with + * Note: we're not holding inode->i_mutex and so may be racing with * operations that change the directory. We therefore save the * change attribute *before* we do the RPC call. */ @@ -1051,7 +1051,7 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc) return dentry; if (!desc->plus || !(entry->fattr->valid & NFS_ATTR_FATTR)) return NULL; - /* Note: caller is already holding the dir->i_sem! */ + /* Note: caller is already holding the dir->i_mutex! */ dentry = d_alloc(parent, &name); if (dentry == NULL) return NULL; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 954cf893d50c..be963a133aaa 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -121,9 +121,9 @@ out: static void nfsd4_sync_rec_dir(void) { - down(&rec_dir.dentry->d_inode->i_sem); + mutex_lock(&rec_dir.dentry->d_inode->i_mutex); nfsd_sync_dir(rec_dir.dentry); - up(&rec_dir.dentry->d_inode->i_sem); + mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); } int @@ -143,7 +143,7 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) nfs4_save_user(&uid, &gid); /* lock the parent */ - down(&rec_dir.dentry->d_inode->i_sem); + mutex_lock(&rec_dir.dentry->d_inode->i_mutex); dentry = lookup_one_len(dname, rec_dir.dentry, HEXDIR_LEN-1); if (IS_ERR(dentry)) { @@ -159,7 +159,7 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) out_put: dput(dentry); out_unlock: - up(&rec_dir.dentry->d_inode->i_sem); + mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); if (status == 0) { clp->cl_firststate = 1; nfsd4_sync_rec_dir(); @@ -259,9 +259,9 @@ nfsd4_remove_clid_file(struct dentry *dir, struct dentry *dentry) printk("nfsd4: non-file found in client recovery directory\n"); return -EINVAL; } - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); status = vfs_unlink(dir->d_inode, dentry); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return status; } @@ -274,9 +274,9 @@ nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) * any regular files anyway, just in case the directory was created by * a kernel from the future.... */ nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); status = vfs_rmdir(dir->d_inode, dentry); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return status; } @@ -288,9 +288,9 @@ nfsd4_unlink_clid_dir(char *name, int namlen) dprintk("NFSD: nfsd4_unlink_clid_dir. name %.*s\n", namlen, name); - down(&rec_dir.dentry->d_inode->i_sem); + mutex_lock(&rec_dir.dentry->d_inode->i_mutex); dentry = lookup_one_len(name, rec_dir.dentry, namlen); - up(&rec_dir.dentry->d_inode->i_sem); + mutex_unlock(&rec_dir.dentry->d_inode->i_mutex); if (IS_ERR(dentry)) { status = PTR_ERR(dentry); return status; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index df4019f04560..bb36b4304491 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -390,12 +390,12 @@ set_nfsv4_acl_one(struct dentry *dentry, struct posix_acl *pacl, char *key) error = -EOPNOTSUPP; if (inode->i_op && inode->i_op->setxattr) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); security_inode_setxattr(dentry, key, buf, len, 0); error = inode->i_op->setxattr(dentry, key, buf, len, 0); if (!error) security_inode_post_setxattr(dentry, key, buf, len, 0); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } out: kfree(buf); @@ -739,9 +739,9 @@ nfsd_sync(struct file *filp) int err; struct inode *inode = filp->f_dentry->d_inode; dprintk("nfsd: sync file %s\n", filp->f_dentry->d_name.name); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); err=nfsd_dosync(filp, filp->f_dentry, filp->f_op); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return err; } @@ -885,9 +885,9 @@ static void kill_suid(struct dentry *dentry) struct iattr ia; ia.ia_valid = ATTR_KILL_SUID | ATTR_KILL_SGID; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); notify_change(dentry, &ia); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); } static inline int diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index eda056bac256..9480a0526cd3 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -1532,7 +1532,7 @@ int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a, * NOTE to self: No changes in the attribute list are required to move from * a resident to a non-resident attribute. * - * Locking: - The caller must hold i_sem on the inode. + * Locking: - The caller must hold i_mutex on the inode. */ int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size) { @@ -1728,7 +1728,7 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size) /* * This needs to be last since the address space operations ->readpage * and ->writepage can run concurrently with us as they are not - * serialized on i_sem. Note, we are not allowed to fail once we flip + * serialized on i_mutex. Note, we are not allowed to fail once we flip * this switch, which is another reason to do this last. */ NInoSetNonResident(ni); diff --git a/fs/ntfs/dir.c b/fs/ntfs/dir.c index 795c3d1930f5..b0690d4c8906 100644 --- a/fs/ntfs/dir.c +++ b/fs/ntfs/dir.c @@ -69,7 +69,7 @@ ntfschar I30[5] = { const_cpu_to_le16('$'), const_cpu_to_le16('I'), * work but we don't care for how quickly one can access them. This also fixes * the dcache aliasing issues. * - * Locking: - Caller must hold i_sem on the directory. + * Locking: - Caller must hold i_mutex on the directory. * - Each page cache page in the index allocation mapping must be * locked whilst being accessed otherwise we may find a corrupt * page due to it being under ->writepage at the moment which @@ -1085,11 +1085,11 @@ static inline int ntfs_filldir(ntfs_volume *vol, loff_t fpos, * While this will return the names in random order this doesn't matter for * ->readdir but OTOH results in a faster ->readdir. * - * VFS calls ->readdir without BKL but with i_sem held. This protects the VFS + * VFS calls ->readdir without BKL but with i_mutex held. This protects the VFS * parts (e.g. ->f_pos and ->i_size, and it also protects against directory * modifications). * - * Locking: - Caller must hold i_sem on the directory. + * Locking: - Caller must hold i_mutex on the directory. * - Each page cache page in the index allocation mapping must be * locked whilst being accessed otherwise we may find a corrupt * page due to it being under ->writepage at the moment which @@ -1520,7 +1520,7 @@ static int ntfs_dir_open(struct inode *vi, struct file *filp) * Note: In the past @filp could be NULL so we ignore it as we don't need it * anyway. * - * Locking: Caller must hold i_sem on the inode. + * Locking: Caller must hold i_mutex on the inode. * * TODO: We should probably also write all attribute/index inodes associated * with this inode but since we have no simple way of getting to them we ignore diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 727533891813..30f71acdc1cb 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -106,7 +106,7 @@ static int ntfs_file_open(struct inode *vi, struct file *filp) * this is the case, the necessary zeroing will also have happened and that all * metadata is self-consistent. * - * Locking: i_sem on the vfs inode corrseponsind to the ntfs inode @ni must be + * Locking: i_mutex on the vfs inode corrseponsind to the ntfs inode @ni must be * held by the caller. */ static int ntfs_attr_extend_initialized(ntfs_inode *ni, const s64 new_init_size, @@ -473,7 +473,7 @@ static inline int ntfs_submit_bh_for_read(struct buffer_head *bh) * @bytes: number of bytes to be written * * This is called for non-resident attributes from ntfs_file_buffered_write() - * with i_sem held on the inode (@pages[0]->mapping->host). There are + * with i_mutex held on the inode (@pages[0]->mapping->host). There are * @nr_pages pages in @pages which are locked but not kmap()ped. The source * data has not yet been copied into the @pages. * @@ -1637,7 +1637,7 @@ err_out: * @pos: byte position in file at which the write begins * @bytes: number of bytes to be written * - * This is called from ntfs_file_buffered_write() with i_sem held on the inode + * This is called from ntfs_file_buffered_write() with i_mutex held on the inode * (@pages[0]->mapping->host). There are @nr_pages pages in @pages which are * locked but not kmap()ped. The source data has already been copied into the * @page. ntfs_prepare_pages_for_non_resident_write() has been called before @@ -1814,7 +1814,7 @@ err_out: /** * ntfs_file_buffered_write - * - * Locking: The vfs is holding ->i_sem on the inode. + * Locking: The vfs is holding ->i_mutex on the inode. */ static ssize_t ntfs_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, unsigned long nr_segs, @@ -2196,9 +2196,9 @@ static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const char __user *buf, BUG_ON(iocb->ki_pos != pos); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = ntfs_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { int err = sync_page_range(inode, mapping, pos, ret); if (err < 0) @@ -2221,12 +2221,12 @@ static ssize_t ntfs_file_writev(struct file *file, const struct iovec *iov, struct kiocb kiocb; ssize_t ret; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); init_sync_kiocb(&kiocb, file); ret = ntfs_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos); if (ret == -EIOCBQUEUED) ret = wait_on_sync_kiocb(&kiocb); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { int err = sync_page_range(inode, mapping, *ppos - ret, ret); if (err < 0) @@ -2269,7 +2269,7 @@ static ssize_t ntfs_file_write(struct file *file, const char __user *buf, * Note: In the past @filp could be NULL so we ignore it as we don't need it * anyway. * - * Locking: Caller must hold i_sem on the inode. + * Locking: Caller must hold i_mutex on the inode. * * TODO: We should probably also write all attribute/index inodes associated * with this inode but since we have no simple way of getting to them we ignore diff --git a/fs/ntfs/index.c b/fs/ntfs/index.c index 8f2d5727546f..9f5427c2d105 100644 --- a/fs/ntfs/index.c +++ b/fs/ntfs/index.c @@ -32,7 +32,7 @@ * Allocate a new index context, initialize it with @idx_ni and return it. * Return NULL if allocation failed. * - * Locking: Caller must hold i_sem on the index inode. + * Locking: Caller must hold i_mutex on the index inode. */ ntfs_index_context *ntfs_index_ctx_get(ntfs_inode *idx_ni) { @@ -50,7 +50,7 @@ ntfs_index_context *ntfs_index_ctx_get(ntfs_inode *idx_ni) * * Release the index context @ictx, releasing all associated resources. * - * Locking: Caller must hold i_sem on the index inode. + * Locking: Caller must hold i_mutex on the index inode. */ void ntfs_index_ctx_put(ntfs_index_context *ictx) { @@ -106,7 +106,7 @@ void ntfs_index_ctx_put(ntfs_index_context *ictx) * or ntfs_index_entry_write() before the call to ntfs_index_ctx_put() to * ensure that the changes are written to disk. * - * Locking: - Caller must hold i_sem on the index inode. + * Locking: - Caller must hold i_mutex on the index inode. * - Each page cache page in the index allocation mapping must be * locked whilst being accessed otherwise we may find a corrupt * page due to it being under ->writepage at the moment which diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index b24f4c4b2c5c..bda7a08911a5 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -2125,13 +2125,13 @@ void ntfs_put_inode(struct inode *vi) ntfs_inode *ni = NTFS_I(vi); if (NInoIndexAllocPresent(ni)) { struct inode *bvi = NULL; - down(&vi->i_sem); + mutex_lock(&vi->i_mutex); if (atomic_read(&vi->i_count) == 2) { bvi = ni->itype.index.bmp_ino; if (bvi) ni->itype.index.bmp_ino = NULL; } - up(&vi->i_sem); + mutex_unlock(&vi->i_mutex); if (bvi) iput(bvi); } @@ -2311,7 +2311,7 @@ static const char *es = " Leaving inconsistent metadata. Unmount and run " * * Returns 0 on success or -errno on error. * - * Called with ->i_sem held. In all but one case ->i_alloc_sem is held for + * Called with ->i_mutex held. In all but one case ->i_alloc_sem is held for * writing. The only case in the kernel where ->i_alloc_sem is not held is * mm/filemap.c::generic_file_buffered_write() where vmtruncate() is called * with the current i_size as the offset. The analogous place in NTFS is in @@ -2831,7 +2831,7 @@ void ntfs_truncate_vfs(struct inode *vi) { * We also abort all changes of user, group, and mode as we do not implement * the NTFS ACLs yet. * - * Called with ->i_sem held. For the ATTR_SIZE (i.e. ->truncate) case, also + * Called with ->i_mutex held. For the ATTR_SIZE (i.e. ->truncate) case, also * called with ->i_alloc_sem held for writing. * * Basically this is a copy of generic notify_change() and inode_setattr() diff --git a/fs/ntfs/namei.c b/fs/ntfs/namei.c index 351dbc3b6e40..5ea9eb93af62 100644 --- a/fs/ntfs/namei.c +++ b/fs/ntfs/namei.c @@ -96,7 +96,7 @@ * name. We then convert the name to the current NLS code page, and proceed * searching for a dentry with this name, etc, as in case 2), above. * - * Locking: Caller must hold i_sem on the directory. + * Locking: Caller must hold i_mutex on the directory. */ static struct dentry *ntfs_lookup(struct inode *dir_ino, struct dentry *dent, struct nameidata *nd) @@ -254,7 +254,7 @@ handle_name: nls_name.hash = full_name_hash(nls_name.name, nls_name.len); /* - * Note: No need for dent->d_lock lock as i_sem is held on the + * Note: No need for dent->d_lock lock as i_mutex is held on the * parent inode. */ @@ -374,7 +374,7 @@ struct inode_operations ntfs_dir_inode_ops = { * The code is based on the ext3 ->get_parent() implementation found in * fs/ext3/namei.c::ext3_get_parent(). * - * Note: ntfs_get_parent() is called with @child_dent->d_inode->i_sem down. + * Note: ntfs_get_parent() is called with @child_dent->d_inode->i_mutex down. * * Return the dentry of the parent directory on success or the error code on * error (IS_ERR() is true). diff --git a/fs/ntfs/quota.c b/fs/ntfs/quota.c index 833df2a4e9fb..d0ef4182147b 100644 --- a/fs/ntfs/quota.c +++ b/fs/ntfs/quota.c @@ -48,7 +48,7 @@ BOOL ntfs_mark_quotas_out_of_date(ntfs_volume *vol) ntfs_error(vol->sb, "Quota inodes are not open."); return FALSE; } - down(&vol->quota_q_ino->i_sem); + mutex_lock(&vol->quota_q_ino->i_mutex); ictx = ntfs_index_ctx_get(NTFS_I(vol->quota_q_ino)); if (!ictx) { ntfs_error(vol->sb, "Failed to get index context."); @@ -98,7 +98,7 @@ BOOL ntfs_mark_quotas_out_of_date(ntfs_volume *vol) ntfs_index_entry_mark_dirty(ictx); set_done: ntfs_index_ctx_put(ictx); - up(&vol->quota_q_ino->i_sem); + mutex_unlock(&vol->quota_q_ino->i_mutex); /* * We set the flag so we do not try to mark the quotas out of date * again on remount. @@ -110,7 +110,7 @@ done: err_out: if (ictx) ntfs_index_ctx_put(ictx); - up(&vol->quota_q_ino->i_sem); + mutex_unlock(&vol->quota_q_ino->i_mutex); return FALSE; } diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 6c16db9e1a8a..280e383fc84e 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -1213,10 +1213,10 @@ static int check_windows_hibernation_status(ntfs_volume *vol) * Find the inode number for the hibernation file by looking up the * filename hiberfil.sys in the root directory. */ - down(&vol->root_ino->i_sem); + mutex_lock(&vol->root_ino->i_mutex); mref = ntfs_lookup_inode_by_name(NTFS_I(vol->root_ino), hiberfil, 12, &name); - up(&vol->root_ino->i_sem); + mutex_unlock(&vol->root_ino->i_mutex); if (IS_ERR_MREF(mref)) { ret = MREF_ERR(mref); /* If the file does not exist, Windows is not hibernated. */ @@ -1307,10 +1307,10 @@ static BOOL load_and_init_quota(ntfs_volume *vol) * Find the inode number for the quota file by looking up the filename * $Quota in the extended system files directory $Extend. */ - down(&vol->extend_ino->i_sem); + mutex_lock(&vol->extend_ino->i_mutex); mref = ntfs_lookup_inode_by_name(NTFS_I(vol->extend_ino), Quota, 6, &name); - up(&vol->extend_ino->i_sem); + mutex_unlock(&vol->extend_ino->i_mutex); if (IS_ERR_MREF(mref)) { /* * If the file does not exist, quotas are disabled and have @@ -1390,10 +1390,10 @@ static BOOL load_and_init_usnjrnl(ntfs_volume *vol) * Find the inode number for the transaction log file by looking up the * filename $UsnJrnl in the extended system files directory $Extend. */ - down(&vol->extend_ino->i_sem); + mutex_lock(&vol->extend_ino->i_mutex); mref = ntfs_lookup_inode_by_name(NTFS_I(vol->extend_ino), UsnJrnl, 8, &name); - up(&vol->extend_ino->i_sem); + mutex_unlock(&vol->extend_ino->i_mutex); if (IS_ERR_MREF(mref)) { /* * If the file does not exist, transaction logging is disabled, @@ -2312,9 +2312,9 @@ static void ntfs_put_super(struct super_block *sb) if (!list_empty(&sb->s_dirty)) { const char *s1, *s2; - down(&vol->mft_ino->i_sem); + mutex_lock(&vol->mft_ino->i_mutex); truncate_inode_pages(vol->mft_ino->i_mapping, 0); - up(&vol->mft_ino->i_sem); + mutex_unlock(&vol->mft_ino->i_mutex); write_inode_now(vol->mft_ino, 1); if (!list_empty(&sb->s_dirty)) { static const char *_s1 = "inodes"; diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 465f797451ee..6b9812db3779 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -966,7 +966,7 @@ static int ocfs2_truncate_log_append(struct ocfs2_super *osb, mlog_entry("start_blk = %"MLFu64", num_clusters = %u\n", start_blk, num_clusters); - BUG_ON(!down_trylock(&tl_inode->i_sem)); + BUG_ON(mutex_trylock(&tl_inode->i_mutex)); start_cluster = ocfs2_blocks_to_clusters(osb->sb, start_blk); @@ -1108,7 +1108,7 @@ bail: return status; } -/* Expects you to already be holding tl_inode->i_sem */ +/* Expects you to already be holding tl_inode->i_mutex */ static int __ocfs2_flush_truncate_log(struct ocfs2_super *osb) { int status; @@ -1123,7 +1123,7 @@ static int __ocfs2_flush_truncate_log(struct ocfs2_super *osb) mlog_entry_void(); - BUG_ON(!down_trylock(&tl_inode->i_sem)); + BUG_ON(mutex_trylock(&tl_inode->i_mutex)); di = (struct ocfs2_dinode *) tl_bh->b_data; tl = &di->id2.i_dealloc; @@ -1198,9 +1198,9 @@ int ocfs2_flush_truncate_log(struct ocfs2_super *osb) int status; struct inode *tl_inode = osb->osb_tl_inode; - down(&tl_inode->i_sem); + mutex_lock(&tl_inode->i_mutex); status = __ocfs2_flush_truncate_log(osb); - up(&tl_inode->i_sem); + mutex_unlock(&tl_inode->i_mutex); return status; } @@ -1363,7 +1363,7 @@ int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb, mlog(0, "cleanup %u records from %"MLFu64"\n", num_recs, tl_copy->i_blkno); - down(&tl_inode->i_sem); + mutex_lock(&tl_inode->i_mutex); for(i = 0; i < num_recs; i++) { if (ocfs2_truncate_log_needs_flush(osb)) { status = __ocfs2_flush_truncate_log(osb); @@ -1395,7 +1395,7 @@ int ocfs2_complete_truncate_log_recovery(struct ocfs2_super *osb, } bail_up: - up(&tl_inode->i_sem); + mutex_unlock(&tl_inode->i_mutex); mlog_exit(status); return status; @@ -1840,7 +1840,7 @@ start: mlog(0, "clusters_to_del = %u in this pass\n", clusters_to_del); - down(&tl_inode->i_sem); + mutex_lock(&tl_inode->i_mutex); tl_sem = 1; /* ocfs2_truncate_log_needs_flush guarantees us at least one * record is free for use. If there isn't any, we flush to get @@ -1875,7 +1875,7 @@ start: goto bail; } - up(&tl_inode->i_sem); + mutex_unlock(&tl_inode->i_mutex); tl_sem = 0; ocfs2_commit_trans(handle); @@ -1890,7 +1890,7 @@ bail: ocfs2_schedule_truncate_log_flush(osb, 1); if (tl_sem) - up(&tl_inode->i_sem); + mutex_unlock(&tl_inode->i_mutex); if (handle) ocfs2_commit_trans(handle); @@ -1994,7 +1994,7 @@ int ocfs2_prepare_truncate(struct ocfs2_super *osb, goto bail; } - down(&ext_alloc_inode->i_sem); + mutex_lock(&ext_alloc_inode->i_mutex); (*tc)->tc_ext_alloc_inode = ext_alloc_inode; status = ocfs2_meta_lock(ext_alloc_inode, @@ -2026,7 +2026,7 @@ static void ocfs2_free_truncate_context(struct ocfs2_truncate_context *tc) if (tc->tc_ext_alloc_locked) ocfs2_meta_unlock(tc->tc_ext_alloc_inode, 1); - up(&tc->tc_ext_alloc_inode->i_sem); + mutex_unlock(&tc->tc_ext_alloc_inode->i_mutex); iput(tc->tc_ext_alloc_inode); } diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c index 5fd60c105913..cf7828f23361 100644 --- a/fs/ocfs2/cluster/nodemanager.c +++ b/fs/ocfs2/cluster/nodemanager.c @@ -653,7 +653,7 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g struct config_group *o2hb_group = NULL, *ret = NULL; void *defs = NULL; - /* this runs under the parent dir's i_sem; there can be only + /* this runs under the parent dir's i_mutex; there can be only * one caller in here at a time */ if (o2nm_single_cluster) goto out; /* ENOSPC */ diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index 856e20ae8263..57158fa75d91 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -202,7 +202,7 @@ bail: } /* - * NOTE: this should always be called with parent dir i_sem taken. + * NOTE: this should always be called with parent dir i_mutex taken. */ int ocfs2_find_files_on_disk(const char *name, int namelen, @@ -245,7 +245,7 @@ leave: * Return 0 if the name does not exist * Return -EEXIST if the directory contains the name * - * Callers should have i_sem + a cluster lock on dir + * Callers should have i_mutex + a cluster lock on dir */ int ocfs2_check_dir_for_entry(struct inode *dir, const char *name, diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 72ae9e3306f4..ca5f9f90d794 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -492,7 +492,7 @@ restart_all: } /* blocks peope in read/write from reading our allocation - * until we're done changing it. We depend on i_sem to block + * until we're done changing it. We depend on i_mutex to block * other extend/truncate calls while we're here. Ordering wrt * start_trans is important here -- always do it before! */ down_write(&OCFS2_I(inode)->ip_alloc_sem); @@ -958,8 +958,8 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, filp->f_flags &= ~O_DIRECT; #endif - down(&inode->i_sem); - /* to match setattr's i_sem -> i_alloc_sem -> rw_lock ordering */ + mutex_lock(&inode->i_mutex); + /* to match setattr's i_mutex -> i_alloc_sem -> rw_lock ordering */ if (filp->f_flags & O_DIRECT) { have_alloc_sem = 1; down_read(&inode->i_alloc_sem); @@ -1123,7 +1123,7 @@ out: up_read(&inode->i_alloc_sem); if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); mlog_exit(ret); return ret; diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index a91ba4dec936..d4ecc0627716 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -485,10 +485,10 @@ static int ocfs2_remove_inode(struct inode *inode, goto bail; } - down(&inode_alloc_inode->i_sem); + mutex_lock(&inode_alloc_inode->i_mutex); status = ocfs2_meta_lock(inode_alloc_inode, NULL, &inode_alloc_bh, 1); if (status < 0) { - up(&inode_alloc_inode->i_sem); + mutex_unlock(&inode_alloc_inode->i_mutex); mlog_errno(status); goto bail; @@ -536,7 +536,7 @@ bail_commit: ocfs2_commit_trans(handle); bail_unlock: ocfs2_meta_unlock(inode_alloc_inode, 1); - up(&inode_alloc_inode->i_sem); + mutex_unlock(&inode_alloc_inode->i_mutex); brelse(inode_alloc_bh); bail: iput(inode_alloc_inode); @@ -567,10 +567,10 @@ static int ocfs2_wipe_inode(struct inode *inode, /* Lock the orphan dir. The lock will be held for the entire * delete_inode operation. We do this now to avoid races with * recovery completion on other nodes. */ - down(&orphan_dir_inode->i_sem); + mutex_lock(&orphan_dir_inode->i_mutex); status = ocfs2_meta_lock(orphan_dir_inode, NULL, &orphan_dir_bh, 1); if (status < 0) { - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); mlog_errno(status); goto bail; @@ -593,7 +593,7 @@ static int ocfs2_wipe_inode(struct inode *inode, bail_unlock_dir: ocfs2_meta_unlock(orphan_dir_inode, 1); - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); brelse(orphan_dir_bh); bail: iput(orphan_dir_inode); diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 04428042e5e5..303c8d96457f 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -216,7 +216,7 @@ void ocfs2_handle_add_inode(struct ocfs2_journal_handle *handle, atomic_inc(&inode->i_count); /* we're obviously changing it... */ - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); /* sanity check */ BUG_ON(OCFS2_I(inode)->ip_handle); @@ -241,7 +241,7 @@ static void ocfs2_handle_unlock_inodes(struct ocfs2_journal_handle *handle) OCFS2_I(inode)->ip_handle = NULL; list_del_init(&OCFS2_I(inode)->ip_handle_list); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); iput(inode); } } @@ -1433,10 +1433,10 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb, goto out; } - down(&orphan_dir_inode->i_sem); + mutex_lock(&orphan_dir_inode->i_mutex); status = ocfs2_meta_lock(orphan_dir_inode, NULL, NULL, 0); if (status < 0) { - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); mlog_errno(status); goto out; } @@ -1451,7 +1451,7 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb, if (!bh) status = -EINVAL; if (status < 0) { - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); if (bh) brelse(bh); mlog_errno(status); @@ -1465,7 +1465,7 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb, if (!ocfs2_check_dir_entry(orphan_dir_inode, de, bh, local)) { - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); status = -EINVAL; mlog_errno(status); brelse(bh); @@ -1509,7 +1509,7 @@ static int ocfs2_recover_orphans(struct ocfs2_super *osb, } brelse(bh); } - up(&orphan_dir_inode->i_sem); + mutex_unlock(&orphan_dir_inode->i_mutex); ocfs2_meta_unlock(orphan_dir_inode, 0); have_disk_lock = 0; diff --git a/fs/ocfs2/localalloc.c b/fs/ocfs2/localalloc.c index fe373a2101d9..149b35181666 100644 --- a/fs/ocfs2/localalloc.c +++ b/fs/ocfs2/localalloc.c @@ -334,7 +334,7 @@ int ocfs2_begin_local_alloc_recovery(struct ocfs2_super *osb, goto bail; } - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); status = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, &alloc_bh, 0, inode); @@ -367,7 +367,7 @@ bail: brelse(alloc_bh); if (inode) { - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); iput(inode); } @@ -446,7 +446,7 @@ bail: /* * make sure we've got at least bitswanted contiguous bits in the - * local alloc. You lose them when you drop i_sem. + * local alloc. You lose them when you drop i_mutex. * * We will add ourselves to the transaction passed in, but may start * our own in order to shift windows. diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 48bf7f0ce544..364d64bd5f10 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -169,7 +169,7 @@ static match_table_t tokens = { */ static void ocfs2_write_super(struct super_block *sb) { - if (down_trylock(&sb->s_lock) == 0) + if (mutex_trylock(&sb->s_lock) != 0) BUG(); sb->s_dirt = 0; } diff --git a/fs/open.c b/fs/open.c index 75f3329e8a67..a3b3a9b5c2ff 100644 --- a/fs/open.c +++ b/fs/open.c @@ -211,9 +211,9 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs, newattrs.ia_valid |= ATTR_FILE; } - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); err = notify_change(dentry, &newattrs); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return err; } @@ -398,9 +398,9 @@ asmlinkage long sys_utime(char __user * filename, struct utimbuf __user * times) (error = vfs_permission(&nd, MAY_WRITE)) != 0) goto dput_and_out; } - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); error = notify_change(nd.dentry, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); dput_and_out: path_release(&nd); out: @@ -451,9 +451,9 @@ long do_utimes(char __user * filename, struct timeval * times) (error = vfs_permission(&nd, MAY_WRITE)) != 0) goto dput_and_out; } - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); error = notify_change(nd.dentry, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); dput_and_out: path_release(&nd); out: @@ -620,13 +620,13 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode) err = -EPERM; if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) goto out_putf; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; err = notify_change(dentry, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out_putf: fput(file); @@ -654,13 +654,13 @@ asmlinkage long sys_chmod(const char __user * filename, mode_t mode) if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) goto dput_and_out; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (mode == (mode_t) -1) mode = inode->i_mode; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; error = notify_change(nd.dentry, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); dput_and_out: path_release(&nd); @@ -696,9 +696,9 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group) } if (!S_ISDIR(inode->i_mode)) newattrs.ia_valid |= ATTR_KILL_SUID|ATTR_KILL_SGID; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); error = notify_change(dentry, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out: return error; } diff --git a/fs/pipe.c b/fs/pipe.c index 66aa0b938d6a..acb030b61fb0 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -44,10 +44,10 @@ void pipe_wait(struct inode * inode) * is considered a noninteractive wait: */ prepare_to_wait(PIPE_WAIT(*inode), &wait, TASK_INTERRUPTIBLE|TASK_NONINTERACTIVE); - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); schedule(); finish_wait(PIPE_WAIT(*inode), &wait); - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); } static inline int @@ -136,7 +136,7 @@ pipe_readv(struct file *filp, const struct iovec *_iov, do_wakeup = 0; ret = 0; - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); info = inode->i_pipe; for (;;) { int bufs = info->nrbufs; @@ -200,7 +200,7 @@ pipe_readv(struct file *filp, const struct iovec *_iov, } pipe_wait(inode); } - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); /* Signal writers asynchronously that there is more room. */ if (do_wakeup) { wake_up_interruptible(PIPE_WAIT(*inode)); @@ -237,7 +237,7 @@ pipe_writev(struct file *filp, const struct iovec *_iov, do_wakeup = 0; ret = 0; - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); info = inode->i_pipe; if (!PIPE_READERS(*inode)) { @@ -341,7 +341,7 @@ pipe_writev(struct file *filp, const struct iovec *_iov, PIPE_WAITING_WRITERS(*inode)--; } out: - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); if (do_wakeup) { wake_up_interruptible(PIPE_WAIT(*inode)); kill_fasync(PIPE_FASYNC_READERS(*inode), SIGIO, POLL_IN); @@ -381,7 +381,7 @@ pipe_ioctl(struct inode *pino, struct file *filp, switch (cmd) { case FIONREAD: - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); info = inode->i_pipe; count = 0; buf = info->curbuf; @@ -390,7 +390,7 @@ pipe_ioctl(struct inode *pino, struct file *filp, count += info->bufs[buf].len; buf = (buf+1) & (PIPE_BUFFERS-1); } - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return put_user(count, (int __user *)arg); default: return -EINVAL; @@ -433,7 +433,7 @@ pipe_poll(struct file *filp, poll_table *wait) static int pipe_release(struct inode *inode, int decr, int decw) { - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); PIPE_READERS(*inode) -= decr; PIPE_WRITERS(*inode) -= decw; if (!PIPE_READERS(*inode) && !PIPE_WRITERS(*inode)) { @@ -443,7 +443,7 @@ pipe_release(struct inode *inode, int decr, int decw) kill_fasync(PIPE_FASYNC_READERS(*inode), SIGIO, POLL_IN); kill_fasync(PIPE_FASYNC_WRITERS(*inode), SIGIO, POLL_OUT); } - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return 0; } @@ -454,9 +454,9 @@ pipe_read_fasync(int fd, struct file *filp, int on) struct inode *inode = filp->f_dentry->d_inode; int retval; - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); retval = fasync_helper(fd, filp, on, PIPE_FASYNC_READERS(*inode)); - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); if (retval < 0) return retval; @@ -471,9 +471,9 @@ pipe_write_fasync(int fd, struct file *filp, int on) struct inode *inode = filp->f_dentry->d_inode; int retval; - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); retval = fasync_helper(fd, filp, on, PIPE_FASYNC_WRITERS(*inode)); - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); if (retval < 0) return retval; @@ -488,14 +488,14 @@ pipe_rdwr_fasync(int fd, struct file *filp, int on) struct inode *inode = filp->f_dentry->d_inode; int retval; - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); retval = fasync_helper(fd, filp, on, PIPE_FASYNC_READERS(*inode)); if (retval >= 0) retval = fasync_helper(fd, filp, on, PIPE_FASYNC_WRITERS(*inode)); - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); if (retval < 0) return retval; @@ -534,9 +534,9 @@ pipe_read_open(struct inode *inode, struct file *filp) { /* We could have perhaps used atomic_t, but this and friends below are the only places. So it doesn't seem worthwhile. */ - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); PIPE_READERS(*inode)++; - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return 0; } @@ -544,9 +544,9 @@ pipe_read_open(struct inode *inode, struct file *filp) static int pipe_write_open(struct inode *inode, struct file *filp) { - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); PIPE_WRITERS(*inode)++; - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return 0; } @@ -554,12 +554,12 @@ pipe_write_open(struct inode *inode, struct file *filp) static int pipe_rdwr_open(struct inode *inode, struct file *filp) { - down(PIPE_SEM(*inode)); + mutex_lock(PIPE_MUTEX(*inode)); if (filp->f_mode & FMODE_READ) PIPE_READERS(*inode)++; if (filp->f_mode & FMODE_WRITE) PIPE_WRITERS(*inode)++; - up(PIPE_SEM(*inode)); + mutex_unlock(PIPE_MUTEX(*inode)); return 0; } diff --git a/fs/quota.c b/fs/quota.c index 612e04db4b93..d14d872646d4 100644 --- a/fs/quota.c +++ b/fs/quota.c @@ -168,7 +168,7 @@ static void quota_sync_sb(struct super_block *sb, int type) sync_blockdev(sb->s_bdev); /* Now when everything is written we can discard the pagecache so - * that userspace sees the changes. We need i_sem and so we could + * that userspace sees the changes. We need i_mutex and so we could * not do it inside dqonoff_sem. Moreover we need to be carefull * about races with quotaoff() (that is the reason why we have own * reference to inode). */ @@ -184,9 +184,9 @@ static void quota_sync_sb(struct super_block *sb, int type) up(&sb_dqopt(sb)->dqonoff_sem); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (discard[cnt]) { - down(&discard[cnt]->i_sem); + mutex_lock(&discard[cnt]->i_mutex); truncate_inode_pages(&discard[cnt]->i_data, 0); - up(&discard[cnt]->i_sem); + mutex_unlock(&discard[cnt]->i_mutex); iput(discard[cnt]); } } diff --git a/fs/read_write.c b/fs/read_write.c index df3468a22fea..3f7a1a62165f 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -33,7 +33,7 @@ loff_t generic_file_llseek(struct file *file, loff_t offset, int origin) long long retval; struct inode *inode = file->f_mapping->host; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); switch (origin) { case 2: offset += inode->i_size; @@ -49,7 +49,7 @@ loff_t generic_file_llseek(struct file *file, loff_t offset, int origin) } retval = offset; } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return retval; } diff --git a/fs/readdir.c b/fs/readdir.c index b03579bc0210..b6109329b607 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -30,13 +30,13 @@ int vfs_readdir(struct file *file, filldir_t filler, void *buf) if (res) goto out; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); res = -ENOENT; if (!IS_DEADDIR(inode)) { res = file->f_op->readdir(file, buf, filler); file_accessed(file); } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out: return res; } diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 7892a865b58a..127e7d2cabdd 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -49,7 +49,7 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp) } reiserfs_write_lock(inode->i_sb); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); /* freeing preallocation only involves relogging blocks that * are already in the current transaction. preallocation gets * freed at the end of each transaction, so it is impossible for @@ -100,7 +100,7 @@ static int reiserfs_file_release(struct inode *inode, struct file *filp) err = reiserfs_truncate_file(inode, 0); } out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); reiserfs_write_unlock(inode->i_sb); return err; } @@ -1342,7 +1342,7 @@ static ssize_t reiserfs_file_write(struct file *file, /* the file we are going t if (unlikely(!access_ok(VERIFY_READ, buf, count))) return -EFAULT; - down(&inode->i_sem); // locks the entire file for just us + mutex_lock(&inode->i_mutex); // locks the entire file for just us pos = *ppos; @@ -1532,12 +1532,12 @@ static ssize_t reiserfs_file_write(struct file *file, /* the file we are going t generic_osync_inode(inode, file->f_mapping, OSYNC_METADATA | OSYNC_DATA); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); reiserfs_async_progress_wait(inode->i_sb); return (already_written != 0) ? already_written : res; out: - up(&inode->i_sem); // unlock the file on exit. + mutex_unlock(&inode->i_mutex); // unlock the file on exit. return res; } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index a5e3a0ddbe53..ffa34b861bdb 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -40,12 +40,12 @@ void reiserfs_delete_inode(struct inode *inode) /* The = 0 happens when we abort creating a new inode for some reason like lack of space.. */ if (!(inode->i_state & I_NEW) && INODE_PKEY(inode)->k_objectid != 0) { /* also handles bad_inode case */ - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); reiserfs_delete_xattrs(inode); if (journal_begin(&th, inode->i_sb, jbegin_count)) { - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); goto out; } reiserfs_update_inode_transaction(inode); @@ -59,11 +59,11 @@ void reiserfs_delete_inode(struct inode *inode) DQUOT_FREE_INODE(inode); if (journal_end(&th, inode->i_sb, jbegin_count)) { - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); goto out; } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); /* check return value from reiserfs_delete_object after * ending the transaction @@ -551,7 +551,7 @@ static int convert_tail_for_hole(struct inode *inode, /* we don't have to make sure the conversion did not happen while ** we were locking the page because anyone that could convert - ** must first take i_sem. + ** must first take i_mutex. ** ** We must fix the tail page for writing because it might have buffers ** that are mapped, but have a block number of 0. This indicates tail @@ -586,7 +586,7 @@ static inline int _allocate_block(struct reiserfs_transaction_handle *th, BUG_ON(!th->t_trans_id); #ifdef REISERFS_PREALLOCATE - if (!(flags & GET_BLOCK_NO_ISEM)) { + if (!(flags & GET_BLOCK_NO_IMUX)) { return reiserfs_new_unf_blocknrs2(th, inode, allocated_block_nr, path, block); } @@ -2318,7 +2318,7 @@ static int map_block_for_writepage(struct inode *inode, /* this is where we fill in holes in the file. */ if (use_get_block) { retval = reiserfs_get_block(inode, block, bh_result, - GET_BLOCK_CREATE | GET_BLOCK_NO_ISEM + GET_BLOCK_CREATE | GET_BLOCK_NO_IMUX | GET_BLOCK_NO_DANGLE); if (!retval) { if (!buffer_mapped(bh_result) diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index 81fc00285f60..ba8bf8df6dc7 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -120,7 +120,7 @@ static int reiserfs_unpack(struct inode *inode, struct file *filp) /* we need to make sure nobody is changing the file size beneath ** us */ - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); write_from = inode->i_size & (blocksize - 1); /* if we are on a block boundary, we are already unpacked. */ @@ -156,7 +156,7 @@ static int reiserfs_unpack(struct inode *inode, struct file *filp) page_cache_release(page); out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); reiserfs_write_unlock(inode->i_sb); return retval; } diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 42afb5bef111..397d9590c8f2 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -2211,7 +2211,7 @@ static ssize_t reiserfs_quota_write(struct super_block *sb, int type, size_t towrite = len; struct buffer_head tmp_bh, *bh; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; @@ -2250,7 +2250,7 @@ static ssize_t reiserfs_quota_write(struct super_block *sb, int type, inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME; mark_inode_dirty(inode); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return len - towrite; } diff --git a/fs/reiserfs/tail_conversion.c b/fs/reiserfs/tail_conversion.c index c92e124f628e..196e971c03c9 100644 --- a/fs/reiserfs/tail_conversion.c +++ b/fs/reiserfs/tail_conversion.c @@ -205,7 +205,7 @@ int indirect2direct(struct reiserfs_transaction_handle *th, struct inode *p_s_in 1) * p_s_sb->s_blocksize; pos1 = pos; - // we are protected by i_sem. The tail can not disapper, not + // we are protected by i_mutex. The tail can not disapper, not // append can be done either // we are in truncate or packing tail in file_release diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 02091eaac0b4..f1895f0a278e 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -67,11 +67,11 @@ static struct dentry *create_xa_root(struct super_block *sb) goto out; } else if (!xaroot->d_inode) { int err; - down(&privroot->d_inode->i_sem); + mutex_lock(&privroot->d_inode->i_mutex); err = privroot->d_inode->i_op->mkdir(privroot->d_inode, xaroot, 0700); - up(&privroot->d_inode->i_sem); + mutex_unlock(&privroot->d_inode->i_mutex); if (err) { dput(xaroot); @@ -219,7 +219,7 @@ static struct dentry *get_xa_file_dentry(const struct inode *inode, } else if (flags & XATTR_REPLACE || flags & FL_READONLY) { goto out; } else { - /* inode->i_sem is down, so nothing else can try to create + /* inode->i_mutex is down, so nothing else can try to create * the same xattr */ err = xadir->d_inode->i_op->create(xadir->d_inode, xafile, 0700 | S_IFREG, NULL); @@ -268,7 +268,7 @@ static struct file *open_xa_file(const struct inode *inode, const char *name, * and don't mess with f->f_pos, but the idea is the same. Do some * action on each and every entry in the directory. * - * we're called with i_sem held, so there are no worries about the directory + * we're called with i_mutex held, so there are no worries about the directory * changing underneath us. */ static int __xattr_readdir(struct file *filp, void *dirent, filldir_t filldir) @@ -426,7 +426,7 @@ int xattr_readdir(struct file *file, filldir_t filler, void *buf) int res = -ENOTDIR; if (!file->f_op || !file->f_op->readdir) goto out; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); // down(&inode->i_zombie); res = -ENOENT; if (!IS_DEADDIR(inode)) { @@ -435,7 +435,7 @@ int xattr_readdir(struct file *file, filldir_t filler, void *buf) unlock_kernel(); } // up(&inode->i_zombie); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out: return res; } @@ -480,7 +480,7 @@ static inline __u32 xattr_hash(const char *msg, int len) /* Generic extended attribute operations that can be used by xa plugins */ /* - * inode->i_sem: down + * inode->i_mutex: down */ int reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, @@ -535,7 +535,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, /* Resize it so we're ok to write there */ newattrs.ia_size = buffer_size; newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; - down(&xinode->i_sem); + mutex_lock(&xinode->i_mutex); err = notify_change(fp->f_dentry, &newattrs); if (err) goto out_filp; @@ -598,7 +598,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, } out_filp: - up(&xinode->i_sem); + mutex_unlock(&xinode->i_mutex); fput(fp); out: @@ -606,7 +606,7 @@ reiserfs_xattr_set(struct inode *inode, const char *name, const void *buffer, } /* - * inode->i_sem: down + * inode->i_mutex: down */ int reiserfs_xattr_get(const struct inode *inode, const char *name, void *buffer, @@ -793,7 +793,7 @@ reiserfs_delete_xattrs_filler(void *buf, const char *name, int namelen, } -/* This is called w/ inode->i_sem downed */ +/* This is called w/ inode->i_mutex downed */ int reiserfs_delete_xattrs(struct inode *inode) { struct file *fp; @@ -946,7 +946,7 @@ int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs) /* * Inode operation getxattr() - * Preliminary locking: we down dentry->d_inode->i_sem + * Preliminary locking: we down dentry->d_inode->i_mutex */ ssize_t reiserfs_getxattr(struct dentry * dentry, const char *name, void *buffer, @@ -970,7 +970,7 @@ reiserfs_getxattr(struct dentry * dentry, const char *name, void *buffer, /* * Inode operation setxattr() * - * dentry->d_inode->i_sem down + * dentry->d_inode->i_mutex down */ int reiserfs_setxattr(struct dentry *dentry, const char *name, const void *value, @@ -1008,7 +1008,7 @@ reiserfs_setxattr(struct dentry *dentry, const char *name, const void *value, /* * Inode operation removexattr() * - * dentry->d_inode->i_sem down + * dentry->d_inode->i_mutex down */ int reiserfs_removexattr(struct dentry *dentry, const char *name) { @@ -1091,7 +1091,7 @@ reiserfs_listxattr_filler(void *buf, const char *name, int namelen, /* * Inode operation listxattr() * - * Preliminary locking: we down dentry->d_inode->i_sem + * Preliminary locking: we down dentry->d_inode->i_mutex */ ssize_t reiserfs_listxattr(struct dentry * dentry, char *buffer, size_t size) { @@ -1289,9 +1289,9 @@ int reiserfs_xattr_init(struct super_block *s, int mount_flags) if (!IS_ERR(dentry)) { if (!(mount_flags & MS_RDONLY) && !dentry->d_inode) { struct inode *inode = dentry->d_parent->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); err = inode->i_op->mkdir(inode, dentry, 0700); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (err) { dput(dentry); dentry = NULL; diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index a47ac9aac8b2..2dc953504cc0 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -174,7 +174,7 @@ static void *posix_acl_to_disk(const struct posix_acl *acl, size_t * size) /* * Inode operation get_posix_acl(). * - * inode->i_sem: down + * inode->i_mutex: down * BKL held [before 2.5.x] */ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type) @@ -237,7 +237,7 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type) /* * Inode operation set_posix_acl(). * - * inode->i_sem: down + * inode->i_mutex: down * BKL held [before 2.5.x] */ static int @@ -312,7 +312,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) return error; } -/* dir->i_sem: down, +/* dir->i_mutex: locked, * inode is new and not released into the wild yet */ int reiserfs_inherit_default_acl(struct inode *dir, struct dentry *dentry, diff --git a/fs/relayfs/inode.c b/fs/relayfs/inode.c index 7b7f2cb5f0e1..383523011aad 100644 --- a/fs/relayfs/inode.c +++ b/fs/relayfs/inode.c @@ -109,7 +109,7 @@ static struct dentry *relayfs_create_entry(const char *name, } parent = dget(parent); - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); d = lookup_one_len(name, parent, strlen(name)); if (IS_ERR(d)) { d = NULL; @@ -139,7 +139,7 @@ release_mount: simple_release_fs(&relayfs_mount, &relayfs_mount_count); exit: - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); dput(parent); return d; } @@ -204,7 +204,7 @@ int relayfs_remove(struct dentry *dentry) return -EINVAL; parent = dget(parent); - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); if (dentry->d_inode) { if (S_ISDIR(dentry->d_inode->i_mode)) error = simple_rmdir(parent->d_inode, dentry); @@ -215,7 +215,7 @@ int relayfs_remove(struct dentry *dentry) } if (!error) dput(dentry); - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); dput(parent); if (!error) @@ -476,7 +476,7 @@ static ssize_t relay_file_read(struct file *filp, ssize_t ret = 0; void *from; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if(!relay_file_read_avail(buf, *ppos)) goto out; @@ -494,7 +494,7 @@ static ssize_t relay_file_read(struct file *filp, relay_file_read_consume(buf, read_start, count); *ppos = relay_file_read_end_pos(buf, read_start, count); out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return ret; } diff --git a/fs/super.c b/fs/super.c index 0a30e51692cf..c177b92419c5 100644 --- a/fs/super.c +++ b/fs/super.c @@ -72,7 +72,7 @@ static struct super_block *alloc_super(void) INIT_HLIST_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); init_rwsem(&s->s_umount); - sema_init(&s->s_lock, 1); + mutex_init(&s->s_lock); down_write(&s->s_umount); s->s_count = S_BIAS; atomic_set(&s->s_active, 1); diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index d36780382176..49bd219275db 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -99,7 +99,7 @@ static int create_dir(struct kobject * k, struct dentry * p, int error; umode_t mode = S_IFDIR| S_IRWXU | S_IRUGO | S_IXUGO; - down(&p->d_inode->i_sem); + mutex_lock(&p->d_inode->i_mutex); *d = lookup_one_len(n, p, strlen(n)); if (!IS_ERR(*d)) { error = sysfs_make_dirent(p->d_fsdata, *d, k, mode, SYSFS_DIR); @@ -122,7 +122,7 @@ static int create_dir(struct kobject * k, struct dentry * p, dput(*d); } else error = PTR_ERR(*d); - up(&p->d_inode->i_sem); + mutex_unlock(&p->d_inode->i_mutex); return error; } @@ -246,7 +246,7 @@ static void remove_dir(struct dentry * d) struct dentry * parent = dget(d->d_parent); struct sysfs_dirent * sd; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); d_delete(d); sd = d->d_fsdata; list_del_init(&sd->s_sibling); @@ -257,7 +257,7 @@ static void remove_dir(struct dentry * d) pr_debug(" o %s removing done (%d)\n",d->d_name.name, atomic_read(&d->d_count)); - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); dput(parent); } @@ -286,7 +286,7 @@ void sysfs_remove_dir(struct kobject * kobj) return; pr_debug("sysfs %s: removing dir\n",dentry->d_name.name); - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); parent_sd = dentry->d_fsdata; list_for_each_entry_safe(sd, tmp, &parent_sd->s_children, s_sibling) { if (!sd->s_element || !(sd->s_type & SYSFS_NOT_PINNED)) @@ -295,7 +295,7 @@ void sysfs_remove_dir(struct kobject * kobj) sysfs_drop_dentry(sd, dentry); sysfs_put(sd); } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); remove_dir(dentry); /** @@ -318,7 +318,7 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name) down_write(&sysfs_rename_sem); parent = kobj->parent->dentry; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); new_dentry = lookup_one_len(new_name, parent, strlen(new_name)); if (!IS_ERR(new_dentry)) { @@ -334,7 +334,7 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name) error = -EEXIST; dput(new_dentry); } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); up_write(&sysfs_rename_sem); return error; @@ -345,9 +345,9 @@ static int sysfs_dir_open(struct inode *inode, struct file *file) struct dentry * dentry = file->f_dentry; struct sysfs_dirent * parent_sd = dentry->d_fsdata; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); file->private_data = sysfs_new_dirent(parent_sd, NULL); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return file->private_data ? 0 : -ENOMEM; @@ -358,9 +358,9 @@ static int sysfs_dir_close(struct inode *inode, struct file *file) struct dentry * dentry = file->f_dentry; struct sysfs_dirent * cursor = file->private_data; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); list_del_init(&cursor->s_sibling); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); release_sysfs_dirent(cursor); @@ -436,7 +436,7 @@ static loff_t sysfs_dir_lseek(struct file * file, loff_t offset, int origin) { struct dentry * dentry = file->f_dentry; - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); switch (origin) { case 1: offset += file->f_pos; @@ -444,7 +444,7 @@ static loff_t sysfs_dir_lseek(struct file * file, loff_t offset, int origin) if (offset >= 0) break; default: - up(&file->f_dentry->d_inode->i_sem); + mutex_unlock(&file->f_dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -468,7 +468,7 @@ static loff_t sysfs_dir_lseek(struct file * file, loff_t offset, int origin) list_add_tail(&cursor->s_sibling, p); } } - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return offset; } @@ -483,4 +483,3 @@ struct file_operations sysfs_dir_operations = { EXPORT_SYMBOL_GPL(sysfs_create_dir); EXPORT_SYMBOL_GPL(sysfs_remove_dir); EXPORT_SYMBOL_GPL(sysfs_rename_dir); - diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 4013d7905e84..d0e3d8495165 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -364,9 +364,9 @@ int sysfs_add_file(struct dentry * dir, const struct attribute * attr, int type) umode_t mode = (attr->mode & S_IALLUGO) | S_IFREG; int error = 0; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); error = sysfs_make_dirent(parent_sd, NULL, (void *) attr, mode, type); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return error; } @@ -398,7 +398,7 @@ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) struct dentry * victim; int res = -ENOENT; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); victim = lookup_one_len(attr->name, dir, strlen(attr->name)); if (!IS_ERR(victim)) { /* make sure dentry is really there */ @@ -420,7 +420,7 @@ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) */ dput(victim); } - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return res; } @@ -441,22 +441,22 @@ int sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode) struct iattr newattrs; int res = -ENOENT; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); victim = lookup_one_len(attr->name, dir, strlen(attr->name)); if (!IS_ERR(victim)) { if (victim->d_inode && (victim->d_parent->d_inode == dir->d_inode)) { inode = victim->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; res = notify_change(victim, &newattrs); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } dput(victim); } - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return res; } @@ -480,4 +480,3 @@ void sysfs_remove_file(struct kobject * kobj, const struct attribute * attr) EXPORT_SYMBOL_GPL(sysfs_create_file); EXPORT_SYMBOL_GPL(sysfs_remove_file); EXPORT_SYMBOL_GPL(sysfs_update_file); - diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index 970a33f03299..c3133219941c 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -201,7 +201,7 @@ const unsigned char * sysfs_get_name(struct sysfs_dirent *sd) /* * Unhashes the dentry corresponding to given sysfs_dirent - * Called with parent inode's i_sem held. + * Called with parent inode's i_mutex held. */ void sysfs_drop_dentry(struct sysfs_dirent * sd, struct dentry * parent) { @@ -232,7 +232,7 @@ void sysfs_hash_and_remove(struct dentry * dir, const char * name) /* no inode means this hasn't been made visible yet */ return; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); list_for_each_entry(sd, &parent_sd->s_children, s_sibling) { if (!sd->s_element) continue; @@ -243,7 +243,5 @@ void sysfs_hash_and_remove(struct dentry * dir, const char * name) break; } } - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); } - - diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c index de402fa915f2..e38d6338a20d 100644 --- a/fs/sysfs/symlink.c +++ b/fs/sysfs/symlink.c @@ -86,9 +86,9 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char BUG_ON(!kobj || !kobj->dentry || !name); - down(&dentry->d_inode->i_sem); + mutex_lock(&dentry->d_inode->i_mutex); error = sysfs_add_link(dentry, name, target); - up(&dentry->d_inode->i_sem); + mutex_unlock(&dentry->d_inode->i_mutex); return error; } @@ -177,4 +177,3 @@ struct inode_operations sysfs_symlink_inode_operations = { EXPORT_SYMBOL_GPL(sysfs_create_link); EXPORT_SYMBOL_GPL(sysfs_remove_link); - diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 2ba11a9aa995..e9a42c711a9e 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1275,7 +1275,7 @@ static ssize_t ufs_quota_write(struct super_block *sb, int type, size_t towrite = len; struct buffer_head *bh; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); while (towrite > 0) { tocopy = sb->s_blocksize - offset < towrite ? sb->s_blocksize - offset : towrite; @@ -1297,7 +1297,7 @@ static ssize_t ufs_quota_write(struct super_block *sb, int type, } out: if (len == towrite) { - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return err; } if (inode->i_size < off+len-towrite) @@ -1305,7 +1305,7 @@ out: inode->i_version++; inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return len - towrite; } diff --git a/fs/xattr.c b/fs/xattr.c index bcc2156d4d28..386a532ee5a9 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -51,7 +51,7 @@ setxattr(struct dentry *d, char __user *name, void __user *value, } } - down(&d->d_inode->i_sem); + mutex_lock(&d->d_inode->i_mutex); error = security_inode_setxattr(d, kname, kvalue, size, flags); if (error) goto out; @@ -73,7 +73,7 @@ setxattr(struct dentry *d, char __user *name, void __user *value, fsnotify_xattr(d); } out: - up(&d->d_inode->i_sem); + mutex_unlock(&d->d_inode->i_mutex); kfree(kvalue); return error; } @@ -323,9 +323,9 @@ removexattr(struct dentry *d, char __user *name) error = security_inode_removexattr(d, kname); if (error) goto out; - down(&d->d_inode->i_sem); + mutex_lock(&d->d_inode->i_mutex); error = d->d_inode->i_op->removexattr(d, kname); - up(&d->d_inode->i_sem); + mutex_unlock(&d->d_inode->i_mutex); if (!error) fsnotify_xattr(d); } diff --git a/fs/xfs/linux-2.6/mutex.h b/fs/xfs/linux-2.6/mutex.h index ce773d89a923..d3369b6ca168 100644 --- a/fs/xfs/linux-2.6/mutex.h +++ b/fs/xfs/linux-2.6/mutex.h @@ -19,7 +19,7 @@ #define __XFS_SUPPORT_MUTEX_H__ #include <linux/spinlock.h> -#include <asm/semaphore.h> +#include <linux/mutex.h> /* * Map the mutex'es from IRIX to Linux semaphores. @@ -28,12 +28,8 @@ * callers. */ #define MUTEX_DEFAULT 0x0 -typedef struct semaphore mutex_t; -#define mutex_init(lock, type, name) sema_init(lock, 1) -#define mutex_destroy(lock) sema_init(lock, -99) -#define mutex_lock(lock, num) down(lock) -#define mutex_trylock(lock) (down_trylock(lock) ? 0 : 1) -#define mutex_unlock(lock) up(lock) +typedef struct mutex mutex_t; +//#define mutex_destroy(lock) do{}while(0) #endif /* __XFS_SUPPORT_MUTEX_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 14215a7db59f..41c478bb1ffc 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -203,7 +203,7 @@ validate_fields( ip->i_nlink = va.va_nlink; ip->i_blocks = va.va_nblocks; - /* we're under i_sem so i_size can't change under us */ + /* we're under i_mutex so i_size can't change under us */ if (i_size_read(ip) != va.va_size) i_size_write(ip, va.va_size); } diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 279e9bc92aba..5675117ef227 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -254,7 +254,7 @@ xfs_read( } if (unlikely(ioflags & IO_ISDIRECT)) - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); xfs_ilock(ip, XFS_IOLOCK_SHARED); if (DM_EVENT_ENABLED(vp->v_vfsp, ip, DM_EVENT_READ) && @@ -286,7 +286,7 @@ xfs_read( unlock_isem: if (unlikely(ioflags & IO_ISDIRECT)) - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return ret; } @@ -655,7 +655,7 @@ relock: iolock = XFS_IOLOCK_EXCL; locktype = VRWLOCK_WRITE; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); } else { iolock = XFS_IOLOCK_SHARED; locktype = VRWLOCK_WRITE_DIRECT; @@ -686,7 +686,7 @@ start: int dmflags = FILP_DELAY_FLAG(file); if (need_isem) - dmflags |= DM_FLAGS_ISEM; + dmflags |= DM_FLAGS_IMUX; xfs_iunlock(xip, XFS_ILOCK_EXCL); error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, vp, @@ -772,7 +772,7 @@ retry: if (need_isem) { /* demote the lock now the cached pages are gone */ XFS_ILOCK_DEMOTE(mp, io, XFS_IOLOCK_EXCL); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); iolock = XFS_IOLOCK_SHARED; locktype = VRWLOCK_WRITE_DIRECT; @@ -817,14 +817,14 @@ retry: xfs_rwunlock(bdp, locktype); if (need_isem) - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, vp, DM_RIGHT_NULL, vp, DM_RIGHT_NULL, NULL, NULL, 0, 0, 0); /* Delay flag intentionally unused */ if (error) goto out_nounlocks; if (need_isem) - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); xfs_rwlock(bdp, locktype); pos = xip->i_d.di_size; ret = 0; @@ -926,7 +926,7 @@ retry: xfs_rwunlock(bdp, locktype); if (need_isem) - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); error = sync_page_range(inode, mapping, pos, ret); if (!error) @@ -938,7 +938,7 @@ retry: xfs_rwunlock(bdp, locktype); out_unlock_isem: if (need_isem) - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); out_nounlocks: return -error; } diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index 00b5043dfa5a..772ac48329ea 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -104,7 +104,7 @@ xfs_qm_dqinit( */ if (brandnewdquot) { dqp->dq_flnext = dqp->dq_flprev = dqp; - mutex_init(&dqp->q_qlock, MUTEX_DEFAULT, "xdq"); + mutex_init(&dqp->q_qlock); initnsema(&dqp->q_flock, 1, "fdq"); sv_init(&dqp->q_pinwait, SV_DEFAULT, "pdq"); @@ -1382,7 +1382,7 @@ void xfs_dqlock( xfs_dquot_t *dqp) { - mutex_lock(&(dqp->q_qlock), PINOD); + mutex_lock(&(dqp->q_qlock)); } void diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 5328a2937127..bb6991a7a617 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -167,7 +167,7 @@ xfs_Gqm_init(void) xqm->qm_dqfree_ratio = XFS_QM_DQFREE_RATIO; xqm->qm_nrefs = 0; #ifdef DEBUG - mutex_init(&qcheck_lock, MUTEX_DEFAULT, "qchk"); + xfs_mutex_init(&qcheck_lock, MUTEX_DEFAULT, "qchk"); #endif return xqm; } @@ -1166,7 +1166,7 @@ xfs_qm_init_quotainfo( qinf->qi_dqreclaims = 0; /* mutex used to serialize quotaoffs */ - mutex_init(&qinf->qi_quotaofflock, MUTEX_DEFAULT, "qoff"); + mutex_init(&qinf->qi_quotaofflock); /* Precalc some constants */ qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); @@ -1285,7 +1285,7 @@ xfs_qm_list_init( char *str, int n) { - mutex_init(&list->qh_lock, MUTEX_DEFAULT, str); + mutex_init(&list->qh_lock); list->qh_next = NULL; list->qh_version = 0; list->qh_nelems = 0; @@ -2762,7 +2762,7 @@ STATIC void xfs_qm_freelist_init(xfs_frlist_t *ql) { ql->qh_next = ql->qh_prev = (xfs_dquot_t *) ql; - mutex_init(&ql->qh_lock, MUTEX_DEFAULT, "dqf"); + mutex_init(&ql->qh_lock); ql->qh_version = 0; ql->qh_nelems = 0; } @@ -2772,7 +2772,7 @@ xfs_qm_freelist_destroy(xfs_frlist_t *ql) { xfs_dquot_t *dqp, *nextdqp; - mutex_lock(&ql->qh_lock, PINOD); + mutex_lock(&ql->qh_lock); for (dqp = ql->qh_next; dqp != (xfs_dquot_t *)ql; ) { xfs_dqlock(dqp); diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h index 12da259f2fcb..4568deb6da86 100644 --- a/fs/xfs/quota/xfs_qm.h +++ b/fs/xfs/quota/xfs_qm.h @@ -165,7 +165,7 @@ typedef struct xfs_dquot_acct { #define XFS_QM_IWARNLIMIT 5 #define XFS_QM_RTBWARNLIMIT 5 -#define XFS_QM_LOCK(xqm) (mutex_lock(&xqm##_lock, PINOD)) +#define XFS_QM_LOCK(xqm) (mutex_lock(&xqm##_lock)) #define XFS_QM_UNLOCK(xqm) (mutex_unlock(&xqm##_lock)) #define XFS_QM_HOLD(xqm) ((xqm)->qm_nrefs++) #define XFS_QM_RELE(xqm) ((xqm)->qm_nrefs--) diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index d9d2993de435..90402a1c3983 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -363,7 +363,7 @@ xfs_qm_init(void) KERN_INFO "SGI XFS Quota Management subsystem\n"; printk(message); - mutex_init(&xfs_Gqm_lock, MUTEX_DEFAULT, "xfs_qmlock"); + mutex_init(&xfs_Gqm_lock); vfs_bhv_set_custom(&xfs_qmops, &xfs_qmcore_xfs); xfs_qm_init_procfs(); } diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 24690e1af659..86a1d09f48d5 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -233,7 +233,7 @@ xfs_qm_scall_quotaoff( */ ASSERT(mp->m_quotainfo); if (mp->m_quotainfo) - mutex_lock(&(XFS_QI_QOFFLOCK(mp)), PINOD); + mutex_lock(&(XFS_QI_QOFFLOCK(mp))); ASSERT(mp->m_quotainfo); @@ -508,7 +508,7 @@ xfs_qm_scall_quotaon( /* * Switch on quota enforcement in core. */ - mutex_lock(&(XFS_QI_QOFFLOCK(mp)), PINOD); + mutex_lock(&(XFS_QI_QOFFLOCK(mp))); mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD); mutex_unlock(&(XFS_QI_QOFFLOCK(mp))); @@ -617,7 +617,7 @@ xfs_qm_scall_setqlim( * a quotaoff from happening). (XXXThis doesn't currently happen * because we take the vfslock before calling xfs_qm_sysent). */ - mutex_lock(&(XFS_QI_QOFFLOCK(mp)), PINOD); + mutex_lock(&(XFS_QI_QOFFLOCK(mp))); /* * Get the dquot (locked), and join it to the transaction. @@ -1426,7 +1426,7 @@ xfs_qm_internalqcheck( xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC); XFS_bflush(mp->m_ddev_targp); - mutex_lock(&qcheck_lock, PINOD); + mutex_lock(&qcheck_lock); /* There should be absolutely no quota activity while this is going on. */ qmtest_udqtab = kmem_zalloc(qmtest_hashmask * diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h index 7a9f3beb818c..b7ddd04aae32 100644 --- a/fs/xfs/quota/xfs_quota_priv.h +++ b/fs/xfs/quota/xfs_quota_priv.h @@ -51,7 +51,7 @@ #define XFS_QI_MPLNEXT(mp) ((mp)->m_quotainfo->qi_dqlist.qh_next) #define XFS_QI_MPLNDQUOTS(mp) ((mp)->m_quotainfo->qi_dqlist.qh_nelems) -#define XQMLCK(h) (mutex_lock(&((h)->qh_lock), PINOD)) +#define XQMLCK(h) (mutex_lock(&((h)->qh_lock))) #define XQMUNLCK(h) (mutex_unlock(&((h)->qh_lock))) #ifdef DEBUG struct xfs_dqhash; diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c index 70ce40914c8a..69ec4f540c3a 100644 --- a/fs/xfs/support/uuid.c +++ b/fs/xfs/support/uuid.c @@ -24,7 +24,7 @@ static uuid_t *uuid_table; void uuid_init(void) { - mutex_init(&uuid_monitor, MUTEX_DEFAULT, "uuid_monitor"); + mutex_init(&uuid_monitor); } /* @@ -94,7 +94,7 @@ uuid_table_insert(uuid_t *uuid) { int i, hole; - mutex_lock(&uuid_monitor, PVFS); + mutex_lock(&uuid_monitor); for (i = 0, hole = -1; i < uuid_table_size; i++) { if (uuid_is_nil(&uuid_table[i])) { hole = i; @@ -122,7 +122,7 @@ uuid_table_remove(uuid_t *uuid) { int i; - mutex_lock(&uuid_monitor, PVFS); + mutex_lock(&uuid_monitor); for (i = 0; i < uuid_table_size; i++) { if (uuid_is_nil(&uuid_table[i])) continue; diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h index 864bf6955689..b4c7f2bc55a0 100644 --- a/fs/xfs/xfs_dmapi.h +++ b/fs/xfs/xfs_dmapi.h @@ -152,7 +152,7 @@ typedef enum { #define DM_FLAGS_NDELAY 0x001 /* return EAGAIN after dm_pending() */ #define DM_FLAGS_UNWANTED 0x002 /* event not in fsys dm_eventset_t */ -#define DM_FLAGS_ISEM 0x004 /* thread holds i_sem */ +#define DM_FLAGS_IMUX 0x004 /* thread holds i_mutex */ #define DM_FLAGS_IALLOCSEM_RD 0x010 /* thread holds i_alloc_sem rd */ #define DM_FLAGS_IALLOCSEM_WR 0x020 /* thread holds i_alloc_sem wr */ @@ -161,21 +161,21 @@ typedef enum { */ #if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,0) #define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - DM_FLAGS_ISEM : 0) -#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_ISEM) + DM_FLAGS_IMUX : 0) +#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX) #endif #if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,0)) && \ (LINUX_VERSION_CODE >= KERNEL_VERSION(2,4,22)) #define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - DM_FLAGS_IALLOCSEM_RD : DM_FLAGS_ISEM) -#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_ISEM) + DM_FLAGS_IALLOCSEM_RD : DM_FLAGS_IMUX) +#define DM_SEM_FLAG_WR (DM_FLAGS_IALLOCSEM_WR | DM_FLAGS_IMUX) #endif #if LINUX_VERSION_CODE <= KERNEL_VERSION(2,4,21) #define DM_SEM_FLAG_RD(ioflags) (((ioflags) & IO_ISDIRECT) ? \ - 0 : DM_FLAGS_ISEM) -#define DM_SEM_FLAG_WR (DM_FLAGS_ISEM) + 0 : DM_FLAGS_IMUX) +#define DM_SEM_FLAG_WR (DM_FLAGS_IMUX) #endif diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 541d5dd474be..303af86739bf 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -117,7 +117,7 @@ xfs_mount_init(void) AIL_LOCKINIT(&mp->m_ail_lock, "xfs_ail"); spinlock_init(&mp->m_sb_lock, "xfs_sb"); - mutex_init(&mp->m_ilock, MUTEX_DEFAULT, "xfs_ilock"); + mutex_init(&mp->m_ilock); initnsema(&mp->m_growlock, 1, "xfs_grow"); /* * Initialize the AIL. diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 08b2e0a5d807..3432fd5a3986 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -533,7 +533,7 @@ typedef struct xfs_mod_sb { int msb_delta; /* Change to make to specified field */ } xfs_mod_sb_t; -#define XFS_MOUNT_ILOCK(mp) mutex_lock(&((mp)->m_ilock), PINOD) +#define XFS_MOUNT_ILOCK(mp) mutex_lock(&((mp)->m_ilock)) #define XFS_MOUNT_IUNLOCK(mp) mutex_unlock(&((mp)->m_ilock)) #define XFS_SB_LOCK(mp) mutex_spinlock(&(mp)->m_sb_lock) #define XFS_SB_UNLOCK(mp,s) mutex_spinunlock(&(mp)->m_sb_lock,(s)) diff --git a/include/asm-alpha/atomic.h b/include/asm-alpha/atomic.h index cb03bbe92cdf..fc77f7413083 100644 --- a/include/asm-alpha/atomic.h +++ b/include/asm-alpha/atomic.h @@ -176,6 +176,7 @@ static __inline__ long atomic64_sub_return(long i, atomic64_t * v) } #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-alpha/mutex.h b/include/asm-alpha/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-alpha/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-arm/atomic.h b/include/asm-arm/atomic.h index f72b63309bc5..3d7283d84405 100644 --- a/include/asm-arm/atomic.h +++ b/include/asm-arm/atomic.h @@ -175,6 +175,8 @@ static inline void atomic_clear_mask(unsigned long mask, unsigned long *addr) #endif /* __LINUX_ARM_ARCH__ */ +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int c, old; diff --git a/include/asm-arm/mutex.h b/include/asm-arm/mutex.h new file mode 100644 index 000000000000..6caa59f1f595 --- /dev/null +++ b/include/asm-arm/mutex.h @@ -0,0 +1,128 @@ +/* + * include/asm-arm/mutex.h + * + * ARM optimized mutex locking primitives + * + * Please look into asm-generic/mutex-xchg.h for a formal definition. + */ +#ifndef _ASM_MUTEX_H +#define _ASM_MUTEX_H + +#if __LINUX_ARM_ARCH__ < 6 +/* On pre-ARMv6 hardware the swp based implementation is the most efficient. */ +# include <asm-generic/mutex-xchg.h> +#else + +/* + * Attempting to lock a mutex on ARMv6+ can be done with a bastardized + * atomic decrement (it is not a reliable atomic decrement but it satisfies + * the defined semantics for our purpose, while being smaller and faster + * than a real atomic decrement or atomic swap. The idea is to attempt + * decrementing the lock value only once. If once decremented it isn't zero, + * or if its store-back fails due to a dispute on the exclusive store, we + * simply bail out immediately through the slow path where the lock will be + * reattempted until it succeeds. + */ +#define __mutex_fastpath_lock(count, fail_fn) \ +do { \ + int __ex_flag, __res; \ + \ + typecheck(atomic_t *, count); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ ( \ + "ldrex %0, [%2] \n" \ + "sub %0, %0, #1 \n" \ + "strex %1, %0, [%2] \n" \ + \ + : "=&r" (__res), "=&r" (__ex_flag) \ + : "r" (&(count)->counter) \ + : "cc","memory" ); \ + \ + if (unlikely(__res || __ex_flag)) \ + fail_fn(count); \ +} while (0) + +#define __mutex_fastpath_lock_retval(count, fail_fn) \ +({ \ + int __ex_flag, __res; \ + \ + typecheck(atomic_t *, count); \ + typecheck_fn(fastcall int (*)(atomic_t *), fail_fn); \ + \ + __asm__ ( \ + "ldrex %0, [%2] \n" \ + "sub %0, %0, #1 \n" \ + "strex %1, %0, [%2] \n" \ + \ + : "=&r" (__res), "=&r" (__ex_flag) \ + : "r" (&(count)->counter) \ + : "cc","memory" ); \ + \ + __res |= __ex_flag; \ + if (unlikely(__res != 0)) \ + __res = fail_fn(count); \ + __res; \ +}) + +/* + * Same trick is used for the unlock fast path. However the original value, + * rather than the result, is used to test for success in order to have + * better generated assembly. + */ +#define __mutex_fastpath_unlock(count, fail_fn) \ +do { \ + int __ex_flag, __res, __orig; \ + \ + typecheck(atomic_t *, count); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ ( \ + "ldrex %0, [%3] \n" \ + "add %1, %0, #1 \n" \ + "strex %2, %1, [%3] \n" \ + \ + : "=&r" (__orig), "=&r" (__res), "=&r" (__ex_flag) \ + : "r" (&(count)->counter) \ + : "cc","memory" ); \ + \ + if (unlikely(__orig || __ex_flag)) \ + fail_fn(count); \ +} while (0) + +/* + * If the unlock was done on a contended lock, or if the unlock simply fails + * then the mutex remains locked. + */ +#define __mutex_slowpath_needs_to_unlock() 1 + +/* + * For __mutex_fastpath_trylock we use another construct which could be + * described as a "single value cmpxchg". + * + * This provides the needed trylock semantics like cmpxchg would, but it is + * lighter and less generic than a true cmpxchg implementation. + */ +static inline int +__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + int __ex_flag, __res, __orig; + + __asm__ ( + + "1: ldrex %0, [%3] \n" + "subs %1, %0, #1 \n" + "strexeq %2, %1, [%3] \n" + "movlt %0, #0 \n" + "cmpeq %2, #0 \n" + "bgt 1b \n" + + : "=&r" (__orig), "=&r" (__res), "=&r" (__ex_flag) + : "r" (&count->counter) + : "cc", "memory" ); + + return __orig; +} + +#endif +#endif diff --git a/include/asm-arm26/atomic.h b/include/asm-arm26/atomic.h index 3074b0e76343..1552c8653990 100644 --- a/include/asm-arm26/atomic.h +++ b/include/asm-arm26/atomic.h @@ -76,6 +76,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-cris/atomic.h b/include/asm-cris/atomic.h index 2df2c7aa19b7..0b51a87e5532 100644 --- a/include/asm-cris/atomic.h +++ b/include/asm-cris/atomic.h @@ -136,6 +136,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-cris/mutex.h b/include/asm-cris/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-cris/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-frv/atomic.h b/include/asm-frv/atomic.h index 9c9e9499cfd8..a59f684b4f33 100644 --- a/include/asm-frv/atomic.h +++ b/include/asm-frv/atomic.h @@ -328,6 +328,7 @@ extern uint32_t __cmpxchg_32(uint32_t *v, uint32_t test, uint32_t new); #endif #define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), old, new)) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-frv/mutex.h b/include/asm-frv/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-frv/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-generic/mutex-dec.h b/include/asm-generic/mutex-dec.h new file mode 100644 index 000000000000..74b18cda169f --- /dev/null +++ b/include/asm-generic/mutex-dec.h @@ -0,0 +1,110 @@ +/* + * asm-generic/mutex-dec.h + * + * Generic implementation of the mutex fastpath, based on atomic + * decrement/increment. + */ +#ifndef _ASM_GENERIC_MUTEX_DEC_H +#define _ASM_GENERIC_MUTEX_DEC_H + +/** + * __mutex_fastpath_lock - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if + * it wasn't 1 originally. This function MUST leave the value lower than + * 1 even when the "1" assertion wasn't true. + */ +#define __mutex_fastpath_lock(count, fail_fn) \ +do { \ + if (unlikely(atomic_dec_return(count) < 0)) \ + fail_fn(count); \ + else \ + smp_mb(); \ +} while (0) + +/** + * __mutex_fastpath_lock_retval - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if + * it wasn't 1 originally. This function returns 0 if the fastpath succeeds, + * or anything the slow path function returns. + */ +static inline int +__mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + if (unlikely(atomic_dec_return(count) < 0)) + return fail_fn(count); + else { + smp_mb(); + return 0; + } +} + +/** + * __mutex_fastpath_unlock - try to promote the count from 0 to 1 + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 0 + * + * Try to promote the count from 0 to 1. If it wasn't 0, call <fail_fn>. + * In the failure case, this function is allowed to either set the value to + * 1, or to set it to a value lower than 1. + * + * If the implementation sets it to a value of lower than 1, then the + * __mutex_slowpath_needs_to_unlock() macro needs to return 1, it needs + * to return 0 otherwise. + */ +#define __mutex_fastpath_unlock(count, fail_fn) \ +do { \ + smp_mb(); \ + if (unlikely(atomic_inc_return(count) <= 0)) \ + fail_fn(count); \ +} while (0) + +#define __mutex_slowpath_needs_to_unlock() 1 + +/** + * __mutex_fastpath_trylock - try to acquire the mutex, without waiting + * + * @count: pointer of type atomic_t + * @fail_fn: fallback function + * + * Change the count from 1 to a value lower than 1, and return 0 (failure) + * if it wasn't 1 originally, or return 1 (success) otherwise. This function + * MUST leave the value lower than 1 even when the "1" assertion wasn't true. + * Additionally, if the value was < 0 originally, this function must not leave + * it to 0 on failure. + * + * If the architecture has no effective trylock variant, it should call the + * <fail_fn> spinlock-based trylock variant unconditionally. + */ +static inline int +__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + /* + * We have two variants here. The cmpxchg based one is the best one + * because it never induce a false contention state. It is included + * here because architectures using the inc/dec algorithms over the + * xchg ones are much more likely to support cmpxchg natively. + * + * If not we fall back to the spinlock based variant - that is + * just as efficient (and simpler) as a 'destructive' probing of + * the mutex state would be. + */ +#ifdef __HAVE_ARCH_CMPXCHG + if (likely(atomic_cmpxchg(count, 1, 0)) == 1) { + smp_mb(); + return 1; + } + return 0; +#else + return fail_fn(count); +#endif +} + +#endif diff --git a/include/asm-generic/mutex-null.h b/include/asm-generic/mutex-null.h new file mode 100644 index 000000000000..5cf8b7ce0c45 --- /dev/null +++ b/include/asm-generic/mutex-null.h @@ -0,0 +1,24 @@ +/* + * asm-generic/mutex-null.h + * + * Generic implementation of the mutex fastpath, based on NOP :-) + * + * This is used by the mutex-debugging infrastructure, but it can also + * be used by architectures that (for whatever reason) want to use the + * spinlock based slowpath. + */ +#ifndef _ASM_GENERIC_MUTEX_NULL_H +#define _ASM_GENERIC_MUTEX_NULL_H + +/* extra parameter only needed for mutex debugging: */ +#ifndef __IP__ +# define __IP__ +#endif + +#define __mutex_fastpath_lock(count, fail_fn) fail_fn(count __RET_IP__) +#define __mutex_fastpath_lock_retval(count, fail_fn) fail_fn(count __RET_IP__) +#define __mutex_fastpath_unlock(count, fail_fn) fail_fn(count __RET_IP__) +#define __mutex_fastpath_trylock(count, fail_fn) fail_fn(count) +#define __mutex_slowpath_needs_to_unlock() 1 + +#endif diff --git a/include/asm-generic/mutex-xchg.h b/include/asm-generic/mutex-xchg.h new file mode 100644 index 000000000000..1d24f47e6c48 --- /dev/null +++ b/include/asm-generic/mutex-xchg.h @@ -0,0 +1,117 @@ +/* + * asm-generic/mutex-xchg.h + * + * Generic implementation of the mutex fastpath, based on xchg(). + * + * NOTE: An xchg based implementation is less optimal than an atomic + * decrement/increment based implementation. If your architecture + * has a reasonable atomic dec/inc then you should probably use + * asm-generic/mutex-dec.h instead, or you could open-code an + * optimized version in asm/mutex.h. + */ +#ifndef _ASM_GENERIC_MUTEX_XCHG_H +#define _ASM_GENERIC_MUTEX_XCHG_H + +/** + * __mutex_fastpath_lock - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if it + * wasn't 1 originally. This function MUST leave the value lower than 1 + * even when the "1" assertion wasn't true. + */ +#define __mutex_fastpath_lock(count, fail_fn) \ +do { \ + if (unlikely(atomic_xchg(count, 0) != 1)) \ + fail_fn(count); \ + else \ + smp_mb(); \ +} while (0) + + +/** + * __mutex_fastpath_lock_retval - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if it + * wasn't 1 originally. This function returns 0 if the fastpath succeeds, + * or anything the slow path function returns + */ +static inline int +__mutex_fastpath_lock_retval(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + if (unlikely(atomic_xchg(count, 0) != 1)) + return fail_fn(count); + else { + smp_mb(); + return 0; + } +} + +/** + * __mutex_fastpath_unlock - try to promote the mutex from 0 to 1 + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 0 + * + * try to promote the mutex from 0 to 1. if it wasn't 0, call <function> + * In the failure case, this function is allowed to either set the value to + * 1, or to set it to a value lower than one. + * If the implementation sets it to a value of lower than one, the + * __mutex_slowpath_needs_to_unlock() macro needs to return 1, it needs + * to return 0 otherwise. + */ +#define __mutex_fastpath_unlock(count, fail_fn) \ +do { \ + smp_mb(); \ + if (unlikely(atomic_xchg(count, 1) != 0)) \ + fail_fn(count); \ +} while (0) + +#define __mutex_slowpath_needs_to_unlock() 0 + +/** + * __mutex_fastpath_trylock - try to acquire the mutex, without waiting + * + * @count: pointer of type atomic_t + * @fail_fn: spinlock based trylock implementation + * + * Change the count from 1 to a value lower than 1, and return 0 (failure) + * if it wasn't 1 originally, or return 1 (success) otherwise. This function + * MUST leave the value lower than 1 even when the "1" assertion wasn't true. + * Additionally, if the value was < 0 originally, this function must not leave + * it to 0 on failure. + * + * If the architecture has no effective trylock variant, it should call the + * <fail_fn> spinlock-based trylock variant unconditionally. + */ +static inline int +__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + int prev = atomic_xchg(count, 0); + + if (unlikely(prev < 0)) { + /* + * The lock was marked contended so we must restore that + * state. If while doing so we get back a prev value of 1 + * then we just own it. + * + * [ In the rare case of the mutex going to 1, to 0, to -1 + * and then back to 0 in this few-instructions window, + * this has the potential to trigger the slowpath for the + * owner's unlock path needlessly, but that's not a problem + * in practice. ] + */ + prev = atomic_xchg(count, prev); + if (prev < 0) + prev = 0; + } + smp_mb(); + + return prev; +} + +#endif diff --git a/include/asm-h8300/atomic.h b/include/asm-h8300/atomic.h index d891541e89c3..21f54428c86b 100644 --- a/include/asm-h8300/atomic.h +++ b/include/asm-h8300/atomic.h @@ -95,6 +95,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-h8300/mutex.h b/include/asm-h8300/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-h8300/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-i386/atomic.h b/include/asm-i386/atomic.h index 7a5472d77091..de649d3aa2d4 100644 --- a/include/asm-i386/atomic.h +++ b/include/asm-i386/atomic.h @@ -216,6 +216,7 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v) } #define atomic_cmpxchg(v, old, new) ((int)cmpxchg(&((v)->counter), old, new)) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-i386/mutex.h b/include/asm-i386/mutex.h new file mode 100644 index 000000000000..4e5e3de1b9a6 --- /dev/null +++ b/include/asm-i386/mutex.h @@ -0,0 +1,124 @@ +/* + * Assembly implementation of the mutex fastpath, based on atomic + * decrement/increment. + * + * started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + */ +#ifndef _ASM_MUTEX_H +#define _ASM_MUTEX_H + +/** + * __mutex_fastpath_lock - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fn> if it + * wasn't 1 originally. This function MUST leave the value lower than 1 + * even when the "1" assertion wasn't true. + */ +#define __mutex_fastpath_lock(count, fail_fn) \ +do { \ + unsigned int dummy; \ + \ + typecheck(atomic_t *, count); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ __volatile__( \ + LOCK " decl (%%eax) \n" \ + " js "#fail_fn" \n" \ + \ + :"=a" (dummy) \ + : "a" (count) \ + : "memory", "ecx", "edx"); \ +} while (0) + + +/** + * __mutex_fastpath_lock_retval - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if it + * wasn't 1 originally. This function returns 0 if the fastpath succeeds, + * or anything the slow path function returns + */ +static inline int +__mutex_fastpath_lock_retval(atomic_t *count, + int fastcall (*fail_fn)(atomic_t *)) +{ + if (unlikely(atomic_dec_return(count) < 0)) + return fail_fn(count); + else + return 0; +} + +/** + * __mutex_fastpath_unlock - try to promote the mutex from 0 to 1 + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 0 + * + * try to promote the mutex from 0 to 1. if it wasn't 0, call <fail_fn>. + * In the failure case, this function is allowed to either set the value + * to 1, or to set it to a value lower than 1. + * + * If the implementation sets it to a value of lower than 1, the + * __mutex_slowpath_needs_to_unlock() macro needs to return 1, it needs + * to return 0 otherwise. + */ +#define __mutex_fastpath_unlock(count, fail_fn) \ +do { \ + unsigned int dummy; \ + \ + typecheck(atomic_t *, count); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ __volatile__( \ + LOCK " incl (%%eax) \n" \ + " jle "#fail_fn" \n" \ + \ + :"=a" (dummy) \ + : "a" (count) \ + : "memory", "ecx", "edx"); \ +} while (0) + +#define __mutex_slowpath_needs_to_unlock() 1 + +/** + * __mutex_fastpath_trylock - try to acquire the mutex, without waiting + * + * @count: pointer of type atomic_t + * @fail_fn: fallback function + * + * Change the count from 1 to a value lower than 1, and return 0 (failure) + * if it wasn't 1 originally, or return 1 (success) otherwise. This function + * MUST leave the value lower than 1 even when the "1" assertion wasn't true. + * Additionally, if the value was < 0 originally, this function must not leave + * it to 0 on failure. + */ +static inline int +__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + /* + * We have two variants here. The cmpxchg based one is the best one + * because it never induce a false contention state. It is included + * here because architectures using the inc/dec algorithms over the + * xchg ones are much more likely to support cmpxchg natively. + * + * If not we fall back to the spinlock based variant - that is + * just as efficient (and simpler) as a 'destructive' probing of + * the mutex state would be. + */ +#ifdef __HAVE_ARCH_CMPXCHG + if (likely(atomic_cmpxchg(count, 1, 0)) == 1) + return 1; + return 0; +#else + return fail_fn(count); +#endif +} + +#endif diff --git a/include/asm-ia64/atomic.h b/include/asm-ia64/atomic.h index 15cf7984c48e..d3e0dfa99e1f 100644 --- a/include/asm-ia64/atomic.h +++ b/include/asm-ia64/atomic.h @@ -89,6 +89,7 @@ ia64_atomic64_sub (__s64 i, atomic64_t *v) } #define atomic_cmpxchg(v, old, new) ((int)cmpxchg(&((v)->counter), old, new)) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-ia64/mutex.h b/include/asm-ia64/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-ia64/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-m32r/atomic.h b/include/asm-m32r/atomic.h index 70761278b6cb..3122fe106f05 100644 --- a/include/asm-m32r/atomic.h +++ b/include/asm-m32r/atomic.h @@ -243,6 +243,7 @@ static __inline__ int atomic_dec_return(atomic_t *v) #define atomic_add_negative(i,v) (atomic_add_return((i), (v)) < 0) #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-m32r/mutex.h b/include/asm-m32r/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-m32r/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-m68k/atomic.h b/include/asm-m68k/atomic.h index b8a4e75d679d..a4a84d5c65d5 100644 --- a/include/asm-m68k/atomic.h +++ b/include/asm-m68k/atomic.h @@ -140,6 +140,7 @@ static inline void atomic_set_mask(unsigned long mask, unsigned long *v) } #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-m68k/mutex.h b/include/asm-m68k/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-m68k/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-m68knommu/atomic.h b/include/asm-m68knommu/atomic.h index 1702dbe9318c..6c4e4b63e454 100644 --- a/include/asm-m68knommu/atomic.h +++ b/include/asm-m68knommu/atomic.h @@ -129,6 +129,7 @@ static inline int atomic_sub_return(int i, atomic_t * v) } #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-m68knommu/mutex.h b/include/asm-m68knommu/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-m68knommu/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-mips/atomic.h b/include/asm-mips/atomic.h index 92256e43a938..94a95872d727 100644 --- a/include/asm-mips/atomic.h +++ b/include/asm-mips/atomic.h @@ -289,6 +289,7 @@ static __inline__ int atomic_sub_if_positive(int i, atomic_t * v) } #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-mips/mutex.h b/include/asm-mips/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-mips/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-parisc/atomic.h b/include/asm-parisc/atomic.h index 64ebd086c40d..2ca56d34aaad 100644 --- a/include/asm-parisc/atomic.h +++ b/include/asm-parisc/atomic.h @@ -165,6 +165,7 @@ static __inline__ int atomic_read(const atomic_t *v) /* exported interface */ #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-parisc/mutex.h b/include/asm-parisc/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-parisc/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-powerpc/atomic.h b/include/asm-powerpc/atomic.h index ae395a0632a6..248f9aec959c 100644 --- a/include/asm-powerpc/atomic.h +++ b/include/asm-powerpc/atomic.h @@ -165,6 +165,7 @@ static __inline__ int atomic_dec_return(atomic_t *v) } #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-powerpc/mutex.h b/include/asm-powerpc/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-powerpc/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-s390/atomic.h b/include/asm-s390/atomic.h index d82aedf616fe..be6fefe223d6 100644 --- a/include/asm-s390/atomic.h +++ b/include/asm-s390/atomic.h @@ -75,6 +75,8 @@ static __inline__ void atomic_set_mask(unsigned long mask, atomic_t * v) __CS_LOOP(v, mask, "or"); } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static __inline__ int atomic_cmpxchg(atomic_t *v, int old, int new) { __asm__ __volatile__(" cs %0,%3,0(%2)\n" diff --git a/include/asm-s390/mutex.h b/include/asm-s390/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-s390/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h index 618d8e0de348..fb627de217f2 100644 --- a/include/asm-sh/atomic.h +++ b/include/asm-sh/atomic.h @@ -101,6 +101,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-sh/mutex.h b/include/asm-sh/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-sh/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-sh64/atomic.h b/include/asm-sh64/atomic.h index f3ce5c0df13a..28f2ea9b567b 100644 --- a/include/asm-sh64/atomic.h +++ b/include/asm-sh64/atomic.h @@ -113,6 +113,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-sh64/mutex.h b/include/asm-sh64/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-sh64/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-sparc/atomic.h b/include/asm-sparc/atomic.h index accb4967e9d2..e1033170bd3a 100644 --- a/include/asm-sparc/atomic.h +++ b/include/asm-sparc/atomic.h @@ -20,6 +20,7 @@ typedef struct { volatile int counter; } atomic_t; extern int __atomic_add_return(int, atomic_t *); extern int atomic_cmpxchg(atomic_t *, int, int); +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) extern int atomic_add_unless(atomic_t *, int, int); extern void atomic_set(atomic_t *, int); diff --git a/include/asm-sparc/mutex.h b/include/asm-sparc/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-sparc/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-sparc64/atomic.h b/include/asm-sparc64/atomic.h index 11f5aa5d108c..25256bdc8aae 100644 --- a/include/asm-sparc64/atomic.h +++ b/include/asm-sparc64/atomic.h @@ -72,6 +72,7 @@ extern int atomic64_sub_ret(int, atomic64_t *); #define atomic64_add_negative(i, v) (atomic64_add_ret(i, v) < 0) #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) #define atomic_add_unless(v, a, u) \ ({ \ diff --git a/include/asm-sparc64/mutex.h b/include/asm-sparc64/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-sparc64/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-um/mutex.h b/include/asm-um/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-um/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-v850/atomic.h b/include/asm-v850/atomic.h index f5b9ab6f4e70..166df00457ea 100644 --- a/include/asm-v850/atomic.h +++ b/include/asm-v850/atomic.h @@ -104,6 +104,8 @@ static inline int atomic_cmpxchg(atomic_t *v, int old, int new) return ret; } +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) + static inline int atomic_add_unless(atomic_t *v, int a, int u) { int ret; diff --git a/include/asm-v850/mutex.h b/include/asm-v850/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-v850/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/asm-x86_64/atomic.h b/include/asm-x86_64/atomic.h index 72eb071488c7..6b540237a2f8 100644 --- a/include/asm-x86_64/atomic.h +++ b/include/asm-x86_64/atomic.h @@ -389,6 +389,7 @@ static __inline__ long atomic64_sub_return(long i, atomic64_t *v) #define atomic64_dec_return(v) (atomic64_sub_return(1,v)) #define atomic_cmpxchg(v, old, new) ((int)cmpxchg(&((v)->counter), old, new)) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-x86_64/mutex.h b/include/asm-x86_64/mutex.h new file mode 100644 index 000000000000..818abfd262d1 --- /dev/null +++ b/include/asm-x86_64/mutex.h @@ -0,0 +1,113 @@ +/* + * Assembly implementation of the mutex fastpath, based on atomic + * decrement/increment. + * + * started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + */ +#ifndef _ASM_MUTEX_H +#define _ASM_MUTEX_H + +/** + * __mutex_fastpath_lock - decrement and call function if negative + * @v: pointer of type atomic_t + * @fail_fn: function to call if the result is negative + * + * Atomically decrements @v and calls <fail_fn> if the result is negative. + */ +#define __mutex_fastpath_lock(v, fail_fn) \ +do { \ + unsigned long dummy; \ + \ + typecheck(atomic_t *, v); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ __volatile__( \ + LOCK " decl (%%rdi) \n" \ + " js 2f \n" \ + "1: \n" \ + \ + LOCK_SECTION_START("") \ + "2: call "#fail_fn" \n" \ + " jmp 1b \n" \ + LOCK_SECTION_END \ + \ + :"=D" (dummy) \ + : "D" (v) \ + : "rax", "rsi", "rdx", "rcx", \ + "r8", "r9", "r10", "r11", "memory"); \ +} while (0) + +/** + * __mutex_fastpath_lock_retval - try to take the lock by moving the count + * from 1 to a 0 value + * @count: pointer of type atomic_t + * @fail_fn: function to call if the original value was not 1 + * + * Change the count from 1 to a value lower than 1, and call <fail_fn> if + * it wasn't 1 originally. This function returns 0 if the fastpath succeeds, + * or anything the slow path function returns + */ +static inline int +__mutex_fastpath_lock_retval(atomic_t *count, + int fastcall (*fail_fn)(atomic_t *)) +{ + if (unlikely(atomic_dec_return(count) < 0)) + return fail_fn(count); + else + return 0; +} + +/** + * __mutex_fastpath_unlock - increment and call function if nonpositive + * @v: pointer of type atomic_t + * @fail_fn: function to call if the result is nonpositive + * + * Atomically increments @v and calls <fail_fn> if the result is nonpositive. + */ +#define __mutex_fastpath_unlock(v, fail_fn) \ +do { \ + unsigned long dummy; \ + \ + typecheck(atomic_t *, v); \ + typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \ + \ + __asm__ __volatile__( \ + LOCK " incl (%%rdi) \n" \ + " jle 2f \n" \ + "1: \n" \ + \ + LOCK_SECTION_START("") \ + "2: call "#fail_fn" \n" \ + " jmp 1b \n" \ + LOCK_SECTION_END \ + \ + :"=D" (dummy) \ + : "D" (v) \ + : "rax", "rsi", "rdx", "rcx", \ + "r8", "r9", "r10", "r11", "memory"); \ +} while (0) + +#define __mutex_slowpath_needs_to_unlock() 1 + +/** + * __mutex_fastpath_trylock - try to acquire the mutex, without waiting + * + * @count: pointer of type atomic_t + * @fail_fn: fallback function + * + * Change the count from 1 to 0 and return 1 (success), or return 0 (failure) + * if it wasn't 1 originally. [the fallback function is never used on + * x86_64, because all x86_64 CPUs have a CMPXCHG instruction.] + */ +static inline int +__mutex_fastpath_trylock(atomic_t *count, int (*fail_fn)(atomic_t *)) +{ + if (likely(atomic_cmpxchg(count, 1, 0)) == 1) + return 1; + else + return 0; +} + +#endif diff --git a/include/asm-xtensa/atomic.h b/include/asm-xtensa/atomic.h index e2ce06b101ad..fe105a123924 100644 --- a/include/asm-xtensa/atomic.h +++ b/include/asm-xtensa/atomic.h @@ -224,6 +224,7 @@ static inline int atomic_sub_return(int i, atomic_t * v) #define atomic_add_negative(i,v) (atomic_add_return((i),(v)) < 0) #define atomic_cmpxchg(v, o, n) ((int)cmpxchg(&((v)->counter), (o), (n))) +#define atomic_xchg(v, new) (xchg(&((v)->counter), new)) /** * atomic_add_unless - add unless the number is a given value diff --git a/include/asm-xtensa/mutex.h b/include/asm-xtensa/mutex.h new file mode 100644 index 000000000000..458c1f7fbc18 --- /dev/null +++ b/include/asm-xtensa/mutex.h @@ -0,0 +1,9 @@ +/* + * Pull in the generic implementation for the mutex fastpath. + * + * TODO: implement optimized primitives instead, or leave the generic + * implementation in place, or pick the atomic_xchg() based generic + * implementation. (see asm-generic/mutex-xchg.h for details) + */ + +#include <asm-generic/mutex-dec.h> diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h index 2914f7b07156..e71dd98dbcae 100644 --- a/include/linux/ext3_fs_i.h +++ b/include/linux/ext3_fs_i.h @@ -87,7 +87,7 @@ struct ext3_inode_info { #ifdef CONFIG_EXT3_FS_XATTR /* * Extended attributes can be read independently of the main file - * data. Taking i_sem even when reading would cause contention + * data. Taking i_mutex even when reading would cause contention * between readers of EAs and writers of regular file data, so * instead we synchronize on xattr_sem when reading or changing * EAs. diff --git a/include/linux/fs.h b/include/linux/fs.h index 4c82219b0fae..92ae3e2067b0 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -219,6 +219,7 @@ extern int dir_notify_enable; #include <linux/prio_tree.h> #include <linux/init.h> #include <linux/sched.h> +#include <linux/mutex.h> #include <asm/atomic.h> #include <asm/semaphore.h> @@ -484,7 +485,7 @@ struct inode { unsigned long i_blocks; unsigned short i_bytes; spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ - struct semaphore i_sem; + struct mutex i_mutex; struct rw_semaphore i_alloc_sem; struct inode_operations *i_op; struct file_operations *i_fop; /* former ->i_op->default_file_ops */ @@ -820,7 +821,7 @@ struct super_block { unsigned long s_magic; struct dentry *s_root; struct rw_semaphore s_umount; - struct semaphore s_lock; + struct mutex s_lock; int s_count; int s_syncing; int s_need_sync_fs; @@ -892,13 +893,13 @@ static inline int has_fs_excl(void) static inline void lock_super(struct super_block * sb) { get_fs_excl(); - down(&sb->s_lock); + mutex_lock(&sb->s_lock); } static inline void unlock_super(struct super_block * sb) { put_fs_excl(); - up(&sb->s_lock); + mutex_unlock(&sb->s_lock); } /* @@ -1191,7 +1192,7 @@ int sync_inode(struct inode *inode, struct writeback_control *wbc); * directory. The name should be stored in the @name (with the * understanding that it is already pointing to a a %NAME_MAX+1 sized * buffer. get_name() should return %0 on success, a negative error code - * or error. @get_name will be called without @parent->i_sem held. + * or error. @get_name will be called without @parent->i_mutex held. * * get_parent: * @get_parent should find the parent directory for the given @child which @@ -1213,7 +1214,7 @@ int sync_inode(struct inode *inode, struct writeback_control *wbc); * nfsd_find_fh_dentry() in either the @obj or @parent parameters. * * Locking rules: - * get_parent is called with child->d_inode->i_sem down + * get_parent is called with child->d_inode->i_mutex down * get_name is not (which is possibly inconsistent) */ diff --git a/include/linux/ide.h b/include/linux/ide.h index ef8d0cbb832f..9a8c05dbe4f3 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -18,6 +18,7 @@ #include <linux/bio.h> #include <linux/device.h> #include <linux/pci.h> +#include <linux/completion.h> #include <asm/byteorder.h> #include <asm/system.h> #include <asm/io.h> @@ -638,7 +639,7 @@ typedef struct ide_drive_s { int crc_count; /* crc counter to reduce drive speed */ struct list_head list; struct device gendev; - struct semaphore gendev_rel_sem; /* to deal with device release() */ + struct completion gendev_rel_comp; /* to deal with device release() */ } ide_drive_t; #define to_ide_device(dev)container_of(dev, ide_drive_t, gendev) @@ -794,7 +795,7 @@ typedef struct hwif_s { unsigned sg_mapped : 1; /* sg_table and sg_nents are ready */ struct device gendev; - struct semaphore gendev_rel_sem; /* To deal with device release() */ + struct completion gendev_rel_comp; /* To deal with device release() */ void *hwif_data; /* extra hwif data */ diff --git a/include/linux/jffs2_fs_i.h b/include/linux/jffs2_fs_i.h index ef85ab56302b..ad565bf9dcc1 100644 --- a/include/linux/jffs2_fs_i.h +++ b/include/linux/jffs2_fs_i.h @@ -8,11 +8,11 @@ #include <asm/semaphore.h> struct jffs2_inode_info { - /* We need an internal semaphore similar to inode->i_sem. + /* We need an internal mutex similar to inode->i_mutex. Unfortunately, we can't used the existing one, because either the GC would deadlock, or we'd have to release it before letting GC proceed. Or we'd have to put ugliness - into the GC code so it didn't attempt to obtain the i_sem + into the GC code so it didn't attempt to obtain the i_mutex for the inode(s) which are already locked */ struct semaphore sem; diff --git a/include/linux/kernel.h b/include/linux/kernel.h index ca7ff8fdd090..d0e6ca3b00ef 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -286,6 +286,15 @@ extern void dump_stack(void); 1; \ }) +/* + * Check at compile time that 'function' is a certain type, or is a pointer + * to that type (needs to use typedef for the function type.) + */ +#define typecheck_fn(type,function) \ +({ typeof(type) __tmp = function; \ + (void)__tmp; \ +}) + #endif /* __KERNEL__ */ #define SI_LOAD_SHIFT 16 diff --git a/include/linux/loop.h b/include/linux/loop.h index 40f63c9879d2..f96506782ebe 100644 --- a/include/linux/loop.h +++ b/include/linux/loop.h @@ -58,9 +58,9 @@ struct loop_device { struct bio *lo_bio; struct bio *lo_biotail; int lo_state; - struct semaphore lo_sem; + struct completion lo_done; + struct completion lo_bh_done; struct semaphore lo_ctl_mutex; - struct semaphore lo_bh_mutex; int lo_pending; request_queue_t *lo_queue; diff --git a/include/linux/mm.h b/include/linux/mm.h index df80e63903b5..3f1fafc0245e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -13,6 +13,7 @@ #include <linux/rbtree.h> #include <linux/prio_tree.h> #include <linux/fs.h> +#include <linux/mutex.h> struct mempolicy; struct anon_vma; @@ -1024,6 +1025,9 @@ static inline void vm_stat_account(struct mm_struct *mm, static inline void kernel_map_pages(struct page *page, int numpages, int enable) { + if (!PageHighMem(page) && !enable) + mutex_debug_check_no_locks_freed(page_address(page), + page_address(page + numpages)); } #endif diff --git a/include/linux/mutex-debug.h b/include/linux/mutex-debug.h new file mode 100644 index 000000000000..0ccd8f983b50 --- /dev/null +++ b/include/linux/mutex-debug.h @@ -0,0 +1,21 @@ +#ifndef __LINUX_MUTEX_DEBUG_H +#define __LINUX_MUTEX_DEBUG_H + +/* + * Mutexes - debugging helpers: + */ + +#define __DEBUG_MUTEX_INITIALIZER(lockname) \ + , .held_list = LIST_HEAD_INIT(lockname.held_list), \ + .name = #lockname , .magic = &lockname + +#define mutex_init(sem) __mutex_init(sem, __FUNCTION__) + +extern void FASTCALL(mutex_destroy(struct mutex *lock)); + +extern void mutex_debug_show_all_locks(void); +extern void mutex_debug_show_held_locks(struct task_struct *filter); +extern void mutex_debug_check_no_locks_held(struct task_struct *task); +extern void mutex_debug_check_no_locks_freed(const void *from, const void *to); + +#endif diff --git a/include/linux/mutex.h b/include/linux/mutex.h new file mode 100644 index 000000000000..9bce0fee68d4 --- /dev/null +++ b/include/linux/mutex.h @@ -0,0 +1,119 @@ +/* + * Mutexes: blocking mutual exclusion locks + * + * started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * This file contains the main data structure and API definitions. + */ +#ifndef __LINUX_MUTEX_H +#define __LINUX_MUTEX_H + +#include <linux/list.h> +#include <linux/spinlock_types.h> + +#include <asm/atomic.h> + +/* + * Simple, straightforward mutexes with strict semantics: + * + * - only one task can hold the mutex at a time + * - only the owner can unlock the mutex + * - multiple unlocks are not permitted + * - recursive locking is not permitted + * - a mutex object must be initialized via the API + * - a mutex object must not be initialized via memset or copying + * - task may not exit with mutex held + * - memory areas where held locks reside must not be freed + * - held mutexes must not be reinitialized + * - mutexes may not be used in irq contexts + * + * These semantics are fully enforced when DEBUG_MUTEXES is + * enabled. Furthermore, besides enforcing the above rules, the mutex + * debugging code also implements a number of additional features + * that make lock debugging easier and faster: + * + * - uses symbolic names of mutexes, whenever they are printed in debug output + * - point-of-acquire tracking, symbolic lookup of function names + * - list of all locks held in the system, printout of them + * - owner tracking + * - detects self-recursing locks and prints out all relevant info + * - detects multi-task circular deadlocks and prints out all affected + * locks and tasks (and only those tasks) + */ +struct mutex { + /* 1: unlocked, 0: locked, negative: locked, possible waiters */ + atomic_t count; + spinlock_t wait_lock; + struct list_head wait_list; +#ifdef CONFIG_DEBUG_MUTEXES + struct thread_info *owner; + struct list_head held_list; + unsigned long acquire_ip; + const char *name; + void *magic; +#endif +}; + +/* + * This is the control structure for tasks blocked on mutex, + * which resides on the blocked task's kernel stack: + */ +struct mutex_waiter { + struct list_head list; + struct task_struct *task; +#ifdef CONFIG_DEBUG_MUTEXES + struct mutex *lock; + void *magic; +#endif +}; + +#ifdef CONFIG_DEBUG_MUTEXES +# include <linux/mutex-debug.h> +#else +# define __DEBUG_MUTEX_INITIALIZER(lockname) +# define mutex_init(mutex) __mutex_init(mutex, NULL) +# define mutex_destroy(mutex) do { } while (0) +# define mutex_debug_show_all_locks() do { } while (0) +# define mutex_debug_show_held_locks(p) do { } while (0) +# define mutex_debug_check_no_locks_held(task) do { } while (0) +# define mutex_debug_check_no_locks_freed(from, to) do { } while (0) +#endif + +#define __MUTEX_INITIALIZER(lockname) \ + { .count = ATOMIC_INIT(1) \ + , .wait_lock = SPIN_LOCK_UNLOCKED \ + , .wait_list = LIST_HEAD_INIT(lockname.wait_list) \ + __DEBUG_MUTEX_INITIALIZER(lockname) } + +#define DEFINE_MUTEX(mutexname) \ + struct mutex mutexname = __MUTEX_INITIALIZER(mutexname) + +extern void fastcall __mutex_init(struct mutex *lock, const char *name); + +/*** + * mutex_is_locked - is the mutex locked + * @lock: the mutex to be queried + * + * Returns 1 if the mutex is locked, 0 if unlocked. + */ +static inline int fastcall mutex_is_locked(struct mutex *lock) +{ + return atomic_read(&lock->count) != 1; +} + +/* + * See kernel/mutex.c for detailed documentation of these APIs. + * Also see Documentation/mutex-design.txt. + */ +extern void fastcall mutex_lock(struct mutex *lock); +extern int fastcall mutex_lock_interruptible(struct mutex *lock); +/* + * NOTE: mutex_trylock() follows the spin_trylock() convention, + * not the down_trylock() convention! + */ +extern int fastcall mutex_trylock(struct mutex *lock); +extern void fastcall mutex_unlock(struct mutex *lock); + +#endif diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index bb842ea41033..0798b7781a6e 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -294,7 +294,7 @@ fill_post_wcc(struct svc_fh *fhp) /* * Lock a file handle/inode * NOTE: both fh_lock and fh_unlock are done "by hand" in - * vfs.c:nfsd_rename as it needs to grab 2 i_sem's at once + * vfs.c:nfsd_rename as it needs to grab 2 i_mutex's at once * so, any changes here should be reflected there. */ static inline void @@ -317,7 +317,7 @@ fh_lock(struct svc_fh *fhp) } inode = dentry->d_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); fill_pre_wcc(fhp); fhp->fh_locked = 1; } @@ -333,7 +333,7 @@ fh_unlock(struct svc_fh *fhp) if (fhp->fh_locked) { fill_post_wcc(fhp); - up(&fhp->fh_dentry->d_inode->i_sem); + mutex_unlock(&fhp->fh_dentry->d_inode->i_mutex); fhp->fh_locked = 0; } } diff --git a/include/linux/pipe_fs_i.h b/include/linux/pipe_fs_i.h index 1767073df26f..b12e59c75752 100644 --- a/include/linux/pipe_fs_i.h +++ b/include/linux/pipe_fs_i.h @@ -37,7 +37,7 @@ struct pipe_inode_info { memory allocation, whereas PIPE_BUF makes atomicity guarantees. */ #define PIPE_SIZE PAGE_SIZE -#define PIPE_SEM(inode) (&(inode).i_sem) +#define PIPE_MUTEX(inode) (&(inode).i_mutex) #define PIPE_WAIT(inode) (&(inode).i_pipe->wait) #define PIPE_READERS(inode) ((inode).i_pipe->readers) #define PIPE_WRITERS(inode) ((inode).i_pipe->writers) diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 001ab82df051..e276c5ba2bb7 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1857,7 +1857,7 @@ void padd_item(char *item, int total_length, int length); #define GET_BLOCK_CREATE 1 /* add anything you need to find block */ #define GET_BLOCK_NO_HOLE 2 /* return -ENOENT for file holes */ #define GET_BLOCK_READ_DIRECT 4 /* read the tail if indirect item not found */ -#define GET_BLOCK_NO_ISEM 8 /* i_sem is not held, don't preallocate */ +#define GET_BLOCK_NO_IMUX 8 /* i_mutex is not held, don't preallocate */ #define GET_BLOCK_NO_DANGLE 16 /* don't leave any transactions running */ int restart_transaction(struct reiserfs_transaction_handle *th, diff --git a/include/linux/sched.h b/include/linux/sched.h index 78eb92ae4d94..85b53f87c703 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -817,6 +817,11 @@ struct task_struct { /* Protection of proc_dentry: nesting proc_lock, dcache_lock, write_lock_irq(&tasklist_lock); */ spinlock_t proc_lock; +#ifdef CONFIG_DEBUG_MUTEXES + /* mutex deadlock detection */ + struct mutex_waiter *blocked_on; +#endif + /* journalling filesystem info */ void *journal_info; diff --git a/ipc/mqueue.c b/ipc/mqueue.c index c8943b53d8e6..a8aa6152eea6 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -660,7 +660,7 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode, if (fd < 0) goto out_putname; - down(&mqueue_mnt->mnt_root->d_inode->i_sem); + mutex_lock(&mqueue_mnt->mnt_root->d_inode->i_mutex); dentry = lookup_one_len(name, mqueue_mnt->mnt_root, strlen(name)); if (IS_ERR(dentry)) { error = PTR_ERR(dentry); @@ -697,7 +697,7 @@ out_putfd: out_err: fd = error; out_upsem: - up(&mqueue_mnt->mnt_root->d_inode->i_sem); + mutex_unlock(&mqueue_mnt->mnt_root->d_inode->i_mutex); out_putname: putname(name); return fd; @@ -714,7 +714,7 @@ asmlinkage long sys_mq_unlink(const char __user *u_name) if (IS_ERR(name)) return PTR_ERR(name); - down(&mqueue_mnt->mnt_root->d_inode->i_sem); + mutex_lock(&mqueue_mnt->mnt_root->d_inode->i_mutex); dentry = lookup_one_len(name, mqueue_mnt->mnt_root, strlen(name)); if (IS_ERR(dentry)) { err = PTR_ERR(dentry); @@ -735,7 +735,7 @@ out_err: dput(dentry); out_unlock: - up(&mqueue_mnt->mnt_root->d_inode->i_sem); + mutex_unlock(&mqueue_mnt->mnt_root->d_inode->i_mutex); putname(name); if (inode) iput(inode); diff --git a/kernel/Makefile b/kernel/Makefile index 4f5a1453093a..a940bac02837 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -7,8 +7,9 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o profile.o \ sysctl.o capability.o ptrace.o timer.o user.o \ signal.o sys.o kmod.o workqueue.o pid.o \ rcupdate.o intermodule.o extable.o params.o posix-timers.o \ - kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o + kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o +obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o obj-$(CONFIG_FUTEX) += futex.o obj-$(CONFIG_GENERIC_ISA_DMA) += dma.o obj-$(CONFIG_SMP) += cpu.o spinlock.o diff --git a/kernel/cpuset.c b/kernel/cpuset.c index eab64e23bcae..2a75e44e1a41 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -1513,7 +1513,7 @@ static int cpuset_add_file(struct dentry *dir, const struct cftype *cft) struct dentry *dentry; int error; - down(&dir->d_inode->i_sem); + mutex_lock(&dir->d_inode->i_mutex); dentry = cpuset_get_dentry(dir, cft->name); if (!IS_ERR(dentry)) { error = cpuset_create_file(dentry, 0644 | S_IFREG); @@ -1522,7 +1522,7 @@ static int cpuset_add_file(struct dentry *dir, const struct cftype *cft) dput(dentry); } else error = PTR_ERR(dentry); - up(&dir->d_inode->i_sem); + mutex_unlock(&dir->d_inode->i_mutex); return error; } @@ -1793,7 +1793,7 @@ static long cpuset_create(struct cpuset *parent, const char *name, int mode) /* * Release manage_sem before cpuset_populate_dir() because it - * will down() this new directory's i_sem and if we race with + * will down() this new directory's i_mutex and if we race with * another mkdir, we might deadlock. */ up(&manage_sem); @@ -1812,7 +1812,7 @@ static int cpuset_mkdir(struct inode *dir, struct dentry *dentry, int mode) { struct cpuset *c_parent = dentry->d_parent->d_fsdata; - /* the vfs holds inode->i_sem already */ + /* the vfs holds inode->i_mutex already */ return cpuset_create(c_parent, dentry->d_name.name, mode | S_IFDIR); } @@ -1823,7 +1823,7 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry) struct cpuset *parent; char *pathbuf = NULL; - /* the vfs holds both inode->i_sem already */ + /* the vfs holds both inode->i_mutex already */ down(&manage_sem); cpuset_update_task_memory_state(); diff --git a/kernel/exit.c b/kernel/exit.c index caceabf3f230..309a46fa16f8 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -29,6 +29,7 @@ #include <linux/syscalls.h> #include <linux/signal.h> #include <linux/cn_proc.h> +#include <linux/mutex.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -869,6 +870,10 @@ fastcall NORET_TYPE void do_exit(long code) mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; #endif + /* + * If DEBUG_MUTEXES is on, make sure we are holding no locks: + */ + mutex_debug_check_no_locks_held(tsk); /* PF_DEAD causes final put_task_struct after we schedule. */ preempt_disable(); diff --git a/kernel/fork.c b/kernel/fork.c index 72e3252c6763..b18d64554feb 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -979,6 +979,10 @@ static task_t *copy_process(unsigned long clone_flags, } #endif +#ifdef CONFIG_DEBUG_MUTEXES + p->blocked_on = NULL; /* not blocked yet */ +#endif + p->tgid = p->pid; if (clone_flags & CLONE_THREAD) p->tgid = current->tgid; diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c new file mode 100644 index 000000000000..4fcb051a8b9e --- /dev/null +++ b/kernel/mutex-debug.c @@ -0,0 +1,464 @@ +/* + * kernel/mutex-debug.c + * + * Debugging code for mutexes + * + * Started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * lock debugging, locking tree, deadlock detection started by: + * + * Copyright (C) 2004, LynuxWorks, Inc., Igor Manyilov, Bill Huey + * Released under the General Public License (GPL). + */ +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/delay.h> +#include <linux/module.h> +#include <linux/spinlock.h> +#include <linux/kallsyms.h> +#include <linux/interrupt.h> + +#include <asm/mutex.h> + +#include "mutex-debug.h" + +/* + * We need a global lock when we walk through the multi-process + * lock tree. Only used in the deadlock-debugging case. + */ +DEFINE_SPINLOCK(debug_mutex_lock); + +/* + * All locks held by all tasks, in a single global list: + */ +LIST_HEAD(debug_mutex_held_locks); + +/* + * In the debug case we carry the caller's instruction pointer into + * other functions, but we dont want the function argument overhead + * in the nondebug case - hence these macros: + */ +#define __IP_DECL__ , unsigned long ip +#define __IP__ , ip +#define __RET_IP__ , (unsigned long)__builtin_return_address(0) + +/* + * "mutex debugging enabled" flag. We turn it off when we detect + * the first problem because we dont want to recurse back + * into the tracing code when doing error printk or + * executing a BUG(): + */ +int debug_mutex_on = 1; + +static void printk_task(struct task_struct *p) +{ + if (p) + printk("%16s:%5d [%p, %3d]", p->comm, p->pid, p, p->prio); + else + printk("<none>"); +} + +static void printk_ti(struct thread_info *ti) +{ + if (ti) + printk_task(ti->task); + else + printk("<none>"); +} + +static void printk_task_short(struct task_struct *p) +{ + if (p) + printk("%s/%d [%p, %3d]", p->comm, p->pid, p, p->prio); + else + printk("<none>"); +} + +static void printk_lock(struct mutex *lock, int print_owner) +{ + printk(" [%p] {%s}\n", lock, lock->name); + + if (print_owner && lock->owner) { + printk(".. held by: "); + printk_ti(lock->owner); + printk("\n"); + } + if (lock->owner) { + printk("... acquired at: "); + print_symbol("%s\n", lock->acquire_ip); + } +} + +/* + * printk locks held by a task: + */ +static void show_task_locks(struct task_struct *p) +{ + switch (p->state) { + case TASK_RUNNING: printk("R"); break; + case TASK_INTERRUPTIBLE: printk("S"); break; + case TASK_UNINTERRUPTIBLE: printk("D"); break; + case TASK_STOPPED: printk("T"); break; + case EXIT_ZOMBIE: printk("Z"); break; + case EXIT_DEAD: printk("X"); break; + default: printk("?"); break; + } + printk_task(p); + if (p->blocked_on) { + struct mutex *lock = p->blocked_on->lock; + + printk(" blocked on mutex:"); + printk_lock(lock, 1); + } else + printk(" (not blocked on mutex)\n"); +} + +/* + * printk all locks held in the system (if filter == NULL), + * or all locks belonging to a single task (if filter != NULL): + */ +void show_held_locks(struct task_struct *filter) +{ + struct list_head *curr, *cursor = NULL; + struct mutex *lock; + struct thread_info *t; + unsigned long flags; + int count = 0; + + if (filter) { + printk("------------------------------\n"); + printk("| showing all locks held by: | ("); + printk_task_short(filter); + printk("):\n"); + printk("------------------------------\n"); + } else { + printk("---------------------------\n"); + printk("| showing all locks held: |\n"); + printk("---------------------------\n"); + } + + /* + * Play safe and acquire the global trace lock. We + * cannot printk with that lock held so we iterate + * very carefully: + */ +next: + debug_spin_lock_save(&debug_mutex_lock, flags); + list_for_each(curr, &debug_mutex_held_locks) { + if (cursor && curr != cursor) + continue; + lock = list_entry(curr, struct mutex, held_list); + t = lock->owner; + if (filter && (t != filter->thread_info)) + continue; + count++; + cursor = curr->next; + debug_spin_lock_restore(&debug_mutex_lock, flags); + + printk("\n#%03d: ", count); + printk_lock(lock, filter ? 0 : 1); + goto next; + } + debug_spin_lock_restore(&debug_mutex_lock, flags); + printk("\n"); +} + +void mutex_debug_show_all_locks(void) +{ + struct task_struct *g, *p; + int count = 10; + int unlock = 1; + + printk("\nShowing all blocking locks in the system:\n"); + + /* + * Here we try to get the tasklist_lock as hard as possible, + * if not successful after 2 seconds we ignore it (but keep + * trying). This is to enable a debug printout even if a + * tasklist_lock-holding task deadlocks or crashes. + */ +retry: + if (!read_trylock(&tasklist_lock)) { + if (count == 10) + printk("hm, tasklist_lock locked, retrying... "); + if (count) { + count--; + printk(" #%d", 10-count); + mdelay(200); + goto retry; + } + printk(" ignoring it.\n"); + unlock = 0; + } + if (count != 10) + printk(" locked it.\n"); + + do_each_thread(g, p) { + show_task_locks(p); + if (!unlock) + if (read_trylock(&tasklist_lock)) + unlock = 1; + } while_each_thread(g, p); + + printk("\n"); + show_held_locks(NULL); + printk("=============================================\n\n"); + + if (unlock) + read_unlock(&tasklist_lock); +} + +static void report_deadlock(struct task_struct *task, struct mutex *lock, + struct mutex *lockblk, unsigned long ip) +{ + printk("\n%s/%d is trying to acquire this lock:\n", + current->comm, current->pid); + printk_lock(lock, 1); + printk("... trying at: "); + print_symbol("%s\n", ip); + show_held_locks(current); + + if (lockblk) { + printk("but %s/%d is deadlocking current task %s/%d!\n\n", + task->comm, task->pid, current->comm, current->pid); + printk("\n%s/%d is blocked on this lock:\n", + task->comm, task->pid); + printk_lock(lockblk, 1); + + show_held_locks(task); + + printk("\n%s/%d's [blocked] stackdump:\n\n", + task->comm, task->pid); + show_stack(task, NULL); + } + + printk("\n%s/%d's [current] stackdump:\n\n", + current->comm, current->pid); + dump_stack(); + mutex_debug_show_all_locks(); + printk("[ turning off deadlock detection. Please report this. ]\n\n"); + local_irq_disable(); +} + +/* + * Recursively check for mutex deadlocks: + */ +static int check_deadlock(struct mutex *lock, int depth, + struct thread_info *ti, unsigned long ip) +{ + struct mutex *lockblk; + struct task_struct *task; + + if (!debug_mutex_on) + return 0; + + ti = lock->owner; + if (!ti) + return 0; + + task = ti->task; + lockblk = NULL; + if (task->blocked_on) + lockblk = task->blocked_on->lock; + + /* Self-deadlock: */ + if (current == task) { + DEBUG_OFF(); + if (depth) + return 1; + printk("\n==========================================\n"); + printk( "[ BUG: lock recursion deadlock detected! |\n"); + printk( "------------------------------------------\n"); + report_deadlock(task, lock, NULL, ip); + return 0; + } + + /* Ugh, something corrupted the lock data structure? */ + if (depth > 20) { + DEBUG_OFF(); + printk("\n===========================================\n"); + printk( "[ BUG: infinite lock dependency detected!? |\n"); + printk( "-------------------------------------------\n"); + report_deadlock(task, lock, lockblk, ip); + return 0; + } + + /* Recursively check for dependencies: */ + if (lockblk && check_deadlock(lockblk, depth+1, ti, ip)) { + printk("\n============================================\n"); + printk( "[ BUG: circular locking deadlock detected! ]\n"); + printk( "--------------------------------------------\n"); + report_deadlock(task, lock, lockblk, ip); + return 0; + } + return 0; +} + +/* + * Called when a task exits, this function checks whether the + * task is holding any locks, and reports the first one if so: + */ +void mutex_debug_check_no_locks_held(struct task_struct *task) +{ + struct list_head *curr, *next; + struct thread_info *t; + unsigned long flags; + struct mutex *lock; + + if (!debug_mutex_on) + return; + + debug_spin_lock_save(&debug_mutex_lock, flags); + list_for_each_safe(curr, next, &debug_mutex_held_locks) { + lock = list_entry(curr, struct mutex, held_list); + t = lock->owner; + if (t != task->thread_info) + continue; + list_del_init(curr); + DEBUG_OFF(); + debug_spin_lock_restore(&debug_mutex_lock, flags); + + printk("BUG: %s/%d, lock held at task exit time!\n", + task->comm, task->pid); + printk_lock(lock, 1); + if (lock->owner != task->thread_info) + printk("exiting task is not even the owner??\n"); + return; + } + debug_spin_lock_restore(&debug_mutex_lock, flags); +} + +/* + * Called when kernel memory is freed (or unmapped), or if a mutex + * is destroyed or reinitialized - this code checks whether there is + * any held lock in the memory range of <from> to <to>: + */ +void mutex_debug_check_no_locks_freed(const void *from, const void *to) +{ + struct list_head *curr, *next; + unsigned long flags; + struct mutex *lock; + void *lock_addr; + + if (!debug_mutex_on) + return; + + debug_spin_lock_save(&debug_mutex_lock, flags); + list_for_each_safe(curr, next, &debug_mutex_held_locks) { + lock = list_entry(curr, struct mutex, held_list); + lock_addr = lock; + if (lock_addr < from || lock_addr >= to) + continue; + list_del_init(curr); + DEBUG_OFF(); + debug_spin_lock_restore(&debug_mutex_lock, flags); + + printk("BUG: %s/%d, active lock [%p(%p-%p)] freed!\n", + current->comm, current->pid, lock, from, to); + dump_stack(); + printk_lock(lock, 1); + if (lock->owner != current_thread_info()) + printk("freeing task is not even the owner??\n"); + return; + } + debug_spin_lock_restore(&debug_mutex_lock, flags); +} + +/* + * Must be called with lock->wait_lock held. + */ +void debug_mutex_set_owner(struct mutex *lock, + struct thread_info *new_owner __IP_DECL__) +{ + lock->owner = new_owner; + DEBUG_WARN_ON(!list_empty(&lock->held_list)); + if (debug_mutex_on) { + list_add_tail(&lock->held_list, &debug_mutex_held_locks); + lock->acquire_ip = ip; + } +} + +void debug_mutex_init_waiter(struct mutex_waiter *waiter) +{ + memset(waiter, 0x11, sizeof(*waiter)); + waiter->magic = waiter; + INIT_LIST_HEAD(&waiter->list); +} + +void debug_mutex_wake_waiter(struct mutex *lock, struct mutex_waiter *waiter) +{ + SMP_DEBUG_WARN_ON(!spin_is_locked(&lock->wait_lock)); + DEBUG_WARN_ON(list_empty(&lock->wait_list)); + DEBUG_WARN_ON(waiter->magic != waiter); + DEBUG_WARN_ON(list_empty(&waiter->list)); +} + +void debug_mutex_free_waiter(struct mutex_waiter *waiter) +{ + DEBUG_WARN_ON(!list_empty(&waiter->list)); + memset(waiter, 0x22, sizeof(*waiter)); +} + +void debug_mutex_add_waiter(struct mutex *lock, struct mutex_waiter *waiter, + struct thread_info *ti __IP_DECL__) +{ + SMP_DEBUG_WARN_ON(!spin_is_locked(&lock->wait_lock)); + check_deadlock(lock, 0, ti, ip); + /* Mark the current thread as blocked on the lock: */ + ti->task->blocked_on = waiter; + waiter->lock = lock; +} + +void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, + struct thread_info *ti) +{ + DEBUG_WARN_ON(list_empty(&waiter->list)); + DEBUG_WARN_ON(waiter->task != ti->task); + DEBUG_WARN_ON(ti->task->blocked_on != waiter); + ti->task->blocked_on = NULL; + + list_del_init(&waiter->list); + waiter->task = NULL; +} + +void debug_mutex_unlock(struct mutex *lock) +{ + DEBUG_WARN_ON(lock->magic != lock); + DEBUG_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next); + DEBUG_WARN_ON(lock->owner != current_thread_info()); + if (debug_mutex_on) { + DEBUG_WARN_ON(list_empty(&lock->held_list)); + list_del_init(&lock->held_list); + } +} + +void debug_mutex_init(struct mutex *lock, const char *name) +{ + /* + * Make sure we are not reinitializing a held lock: + */ + mutex_debug_check_no_locks_freed((void *)lock, (void *)(lock + 1)); + lock->owner = NULL; + INIT_LIST_HEAD(&lock->held_list); + lock->name = name; + lock->magic = lock; +} + +/*** + * mutex_destroy - mark a mutex unusable + * @lock: the mutex to be destroyed + * + * This function marks the mutex uninitialized, and any subsequent + * use of the mutex is forbidden. The mutex must not be locked when + * this function is called. + */ +void fastcall mutex_destroy(struct mutex *lock) +{ + DEBUG_WARN_ON(mutex_is_locked(lock)); + lock->magic = NULL; +} + +EXPORT_SYMBOL_GPL(mutex_destroy); + diff --git a/kernel/mutex-debug.h b/kernel/mutex-debug.h new file mode 100644 index 000000000000..fd384050acb1 --- /dev/null +++ b/kernel/mutex-debug.h @@ -0,0 +1,134 @@ +/* + * Mutexes: blocking mutual exclusion locks + * + * started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * This file contains mutex debugging related internal declarations, + * prototypes and inline functions, for the CONFIG_DEBUG_MUTEXES case. + * More details are in kernel/mutex-debug.c. + */ + +extern spinlock_t debug_mutex_lock; +extern struct list_head debug_mutex_held_locks; +extern int debug_mutex_on; + +/* + * In the debug case we carry the caller's instruction pointer into + * other functions, but we dont want the function argument overhead + * in the nondebug case - hence these macros: + */ +#define __IP_DECL__ , unsigned long ip +#define __IP__ , ip +#define __RET_IP__ , (unsigned long)__builtin_return_address(0) + +/* + * This must be called with lock->wait_lock held. + */ +extern void debug_mutex_set_owner(struct mutex *lock, + struct thread_info *new_owner __IP_DECL__); + +static inline void debug_mutex_clear_owner(struct mutex *lock) +{ + lock->owner = NULL; +} + +extern void debug_mutex_init_waiter(struct mutex_waiter *waiter); +extern void debug_mutex_wake_waiter(struct mutex *lock, + struct mutex_waiter *waiter); +extern void debug_mutex_free_waiter(struct mutex_waiter *waiter); +extern void debug_mutex_add_waiter(struct mutex *lock, + struct mutex_waiter *waiter, + struct thread_info *ti __IP_DECL__); +extern void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, + struct thread_info *ti); +extern void debug_mutex_unlock(struct mutex *lock); +extern void debug_mutex_init(struct mutex *lock, const char *name); + +#define debug_spin_lock(lock) \ + do { \ + local_irq_disable(); \ + if (debug_mutex_on) \ + spin_lock(lock); \ + } while (0) + +#define debug_spin_unlock(lock) \ + do { \ + if (debug_mutex_on) \ + spin_unlock(lock); \ + local_irq_enable(); \ + preempt_check_resched(); \ + } while (0) + +#define debug_spin_lock_save(lock, flags) \ + do { \ + local_irq_save(flags); \ + if (debug_mutex_on) \ + spin_lock(lock); \ + } while (0) + +#define debug_spin_lock_restore(lock, flags) \ + do { \ + if (debug_mutex_on) \ + spin_unlock(lock); \ + local_irq_restore(flags); \ + preempt_check_resched(); \ + } while (0) + +#define spin_lock_mutex(lock) \ + do { \ + struct mutex *l = container_of(lock, struct mutex, wait_lock); \ + \ + DEBUG_WARN_ON(in_interrupt()); \ + debug_spin_lock(&debug_mutex_lock); \ + spin_lock(lock); \ + DEBUG_WARN_ON(l->magic != l); \ + } while (0) + +#define spin_unlock_mutex(lock) \ + do { \ + spin_unlock(lock); \ + debug_spin_unlock(&debug_mutex_lock); \ + } while (0) + +#define DEBUG_OFF() \ +do { \ + if (debug_mutex_on) { \ + debug_mutex_on = 0; \ + console_verbose(); \ + if (spin_is_locked(&debug_mutex_lock)) \ + spin_unlock(&debug_mutex_lock); \ + } \ +} while (0) + +#define DEBUG_BUG() \ +do { \ + if (debug_mutex_on) { \ + DEBUG_OFF(); \ + BUG(); \ + } \ +} while (0) + +#define DEBUG_WARN_ON(c) \ +do { \ + if (unlikely(c && debug_mutex_on)) { \ + DEBUG_OFF(); \ + WARN_ON(1); \ + } \ +} while (0) + +# define DEBUG_BUG_ON(c) \ +do { \ + if (unlikely(c)) \ + DEBUG_BUG(); \ +} while (0) + +#ifdef CONFIG_SMP +# define SMP_DEBUG_WARN_ON(c) DEBUG_WARN_ON(c) +# define SMP_DEBUG_BUG_ON(c) DEBUG_BUG_ON(c) +#else +# define SMP_DEBUG_WARN_ON(c) do { } while (0) +# define SMP_DEBUG_BUG_ON(c) do { } while (0) +#endif + diff --git a/kernel/mutex.c b/kernel/mutex.c new file mode 100644 index 000000000000..7eb960661441 --- /dev/null +++ b/kernel/mutex.c @@ -0,0 +1,325 @@ +/* + * kernel/mutex.c + * + * Mutexes: blocking mutual exclusion locks + * + * Started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * Many thanks to Arjan van de Ven, Thomas Gleixner, Steven Rostedt and + * David Howells for suggestions and improvements. + * + * Also see Documentation/mutex-design.txt. + */ +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/module.h> +#include <linux/spinlock.h> +#include <linux/interrupt.h> + +/* + * In the DEBUG case we are using the "NULL fastpath" for mutexes, + * which forces all calls into the slowpath: + */ +#ifdef CONFIG_DEBUG_MUTEXES +# include "mutex-debug.h" +# include <asm-generic/mutex-null.h> +#else +# include "mutex.h" +# include <asm/mutex.h> +#endif + +/*** + * mutex_init - initialize the mutex + * @lock: the mutex to be initialized + * + * Initialize the mutex to unlocked state. + * + * It is not allowed to initialize an already locked mutex. + */ +void fastcall __mutex_init(struct mutex *lock, const char *name) +{ + atomic_set(&lock->count, 1); + spin_lock_init(&lock->wait_lock); + INIT_LIST_HEAD(&lock->wait_list); + + debug_mutex_init(lock, name); +} + +EXPORT_SYMBOL(__mutex_init); + +/* + * We split the mutex lock/unlock logic into separate fastpath and + * slowpath functions, to reduce the register pressure on the fastpath. + * We also put the fastpath first in the kernel image, to make sure the + * branch is predicted by the CPU as default-untaken. + */ +static void fastcall noinline __sched +__mutex_lock_slowpath(atomic_t *lock_count __IP_DECL__); + +/*** + * mutex_lock - acquire the mutex + * @lock: the mutex to be acquired + * + * Lock the mutex exclusively for this task. If the mutex is not + * available right now, it will sleep until it can get it. + * + * The mutex must later on be released by the same task that + * acquired it. Recursive locking is not allowed. The task + * may not exit without first unlocking the mutex. Also, kernel + * memory where the mutex resides mutex must not be freed with + * the mutex still locked. The mutex must first be initialized + * (or statically defined) before it can be locked. memset()-ing + * the mutex to 0 is not allowed. + * + * ( The CONFIG_DEBUG_MUTEXES .config option turns on debugging + * checks that will enforce the restrictions and will also do + * deadlock debugging. ) + * + * This function is similar to (but not equivalent to) down(). + */ +void fastcall __sched mutex_lock(struct mutex *lock) +{ + /* + * The locking fastpath is the 1->0 transition from + * 'unlocked' into 'locked' state. + * + * NOTE: if asm/mutex.h is included, then some architectures + * rely on mutex_lock() having _no other code_ here but this + * fastpath. That allows the assembly fastpath to do + * tail-merging optimizations. (If you want to put testcode + * here, do it under #ifndef CONFIG_MUTEX_DEBUG.) + */ + __mutex_fastpath_lock(&lock->count, __mutex_lock_slowpath); +} + +EXPORT_SYMBOL(mutex_lock); + +static void fastcall noinline __sched +__mutex_unlock_slowpath(atomic_t *lock_count __IP_DECL__); + +/*** + * mutex_unlock - release the mutex + * @lock: the mutex to be released + * + * Unlock a mutex that has been locked by this task previously. + * + * This function must not be used in interrupt context. Unlocking + * of a not locked mutex is not allowed. + * + * This function is similar to (but not equivalent to) up(). + */ +void fastcall __sched mutex_unlock(struct mutex *lock) +{ + /* + * The unlocking fastpath is the 0->1 transition from 'locked' + * into 'unlocked' state: + * + * NOTE: no other code must be here - see mutex_lock() . + */ + __mutex_fastpath_unlock(&lock->count, __mutex_unlock_slowpath); +} + +EXPORT_SYMBOL(mutex_unlock); + +/* + * Lock a mutex (possibly interruptible), slowpath: + */ +static inline int __sched +__mutex_lock_common(struct mutex *lock, long state __IP_DECL__) +{ + struct task_struct *task = current; + struct mutex_waiter waiter; + unsigned int old_val; + + debug_mutex_init_waiter(&waiter); + + spin_lock_mutex(&lock->wait_lock); + + debug_mutex_add_waiter(lock, &waiter, task->thread_info, ip); + + /* add waiting tasks to the end of the waitqueue (FIFO): */ + list_add_tail(&waiter.list, &lock->wait_list); + waiter.task = task; + + for (;;) { + /* + * Lets try to take the lock again - this is needed even if + * we get here for the first time (shortly after failing to + * acquire the lock), to make sure that we get a wakeup once + * it's unlocked. Later on, if we sleep, this is the + * operation that gives us the lock. We xchg it to -1, so + * that when we release the lock, we properly wake up the + * other waiters: + */ + old_val = atomic_xchg(&lock->count, -1); + if (old_val == 1) + break; + + /* + * got a signal? (This code gets eliminated in the + * TASK_UNINTERRUPTIBLE case.) + */ + if (unlikely(state == TASK_INTERRUPTIBLE && + signal_pending(task))) { + mutex_remove_waiter(lock, &waiter, task->thread_info); + spin_unlock_mutex(&lock->wait_lock); + + debug_mutex_free_waiter(&waiter); + return -EINTR; + } + __set_task_state(task, state); + + /* didnt get the lock, go to sleep: */ + spin_unlock_mutex(&lock->wait_lock); + schedule(); + spin_lock_mutex(&lock->wait_lock); + } + + /* got the lock - rejoice! */ + mutex_remove_waiter(lock, &waiter, task->thread_info); + debug_mutex_set_owner(lock, task->thread_info __IP__); + + /* set it to 0 if there are no waiters left: */ + if (likely(list_empty(&lock->wait_list))) + atomic_set(&lock->count, 0); + + spin_unlock_mutex(&lock->wait_lock); + + debug_mutex_free_waiter(&waiter); + + DEBUG_WARN_ON(list_empty(&lock->held_list)); + DEBUG_WARN_ON(lock->owner != task->thread_info); + + return 0; +} + +static void fastcall noinline __sched +__mutex_lock_slowpath(atomic_t *lock_count __IP_DECL__) +{ + struct mutex *lock = container_of(lock_count, struct mutex, count); + + __mutex_lock_common(lock, TASK_UNINTERRUPTIBLE __IP__); +} + +/* + * Release the lock, slowpath: + */ +static fastcall noinline void +__mutex_unlock_slowpath(atomic_t *lock_count __IP_DECL__) +{ + struct mutex *lock = container_of(lock_count, struct mutex, count); + + DEBUG_WARN_ON(lock->owner != current_thread_info()); + + spin_lock_mutex(&lock->wait_lock); + + /* + * some architectures leave the lock unlocked in the fastpath failure + * case, others need to leave it locked. In the later case we have to + * unlock it here + */ + if (__mutex_slowpath_needs_to_unlock()) + atomic_set(&lock->count, 1); + + debug_mutex_unlock(lock); + + if (!list_empty(&lock->wait_list)) { + /* get the first entry from the wait-list: */ + struct mutex_waiter *waiter = + list_entry(lock->wait_list.next, + struct mutex_waiter, list); + + debug_mutex_wake_waiter(lock, waiter); + + wake_up_process(waiter->task); + } + + debug_mutex_clear_owner(lock); + + spin_unlock_mutex(&lock->wait_lock); +} + +/* + * Here come the less common (and hence less performance-critical) APIs: + * mutex_lock_interruptible() and mutex_trylock(). + */ +static int fastcall noinline __sched +__mutex_lock_interruptible_slowpath(atomic_t *lock_count __IP_DECL__); + +/*** + * mutex_lock_interruptible - acquire the mutex, interruptable + * @lock: the mutex to be acquired + * + * Lock the mutex like mutex_lock(), and return 0 if the mutex has + * been acquired or sleep until the mutex becomes available. If a + * signal arrives while waiting for the lock then this function + * returns -EINTR. + * + * This function is similar to (but not equivalent to) down_interruptible(). + */ +int fastcall __sched mutex_lock_interruptible(struct mutex *lock) +{ + /* NOTE: no other code must be here - see mutex_lock() */ + return __mutex_fastpath_lock_retval + (&lock->count, __mutex_lock_interruptible_slowpath); +} + +EXPORT_SYMBOL(mutex_lock_interruptible); + +static int fastcall noinline __sched +__mutex_lock_interruptible_slowpath(atomic_t *lock_count __IP_DECL__) +{ + struct mutex *lock = container_of(lock_count, struct mutex, count); + + return __mutex_lock_common(lock, TASK_INTERRUPTIBLE __IP__); +} + +/* + * Spinlock based trylock, we take the spinlock and check whether we + * can get the lock: + */ +static inline int __mutex_trylock_slowpath(atomic_t *lock_count) +{ + struct mutex *lock = container_of(lock_count, struct mutex, count); + int prev; + + spin_lock_mutex(&lock->wait_lock); + + prev = atomic_xchg(&lock->count, -1); + if (likely(prev == 1)) + debug_mutex_set_owner(lock, current_thread_info() __RET_IP__); + /* Set it back to 0 if there are no waiters: */ + if (likely(list_empty(&lock->wait_list))) + atomic_set(&lock->count, 0); + + spin_unlock_mutex(&lock->wait_lock); + + return prev == 1; +} + +/*** + * mutex_trylock - try acquire the mutex, without waiting + * @lock: the mutex to be acquired + * + * Try to acquire the mutex atomically. Returns 1 if the mutex + * has been acquired successfully, and 0 on contention. + * + * NOTE: this function follows the spin_trylock() convention, so + * it is negated to the down_trylock() return values! Be careful + * about this when converting semaphore users to mutexes. + * + * This function must not be used in interrupt context. The + * mutex must be released by the same task that acquired it. + */ +int fastcall mutex_trylock(struct mutex *lock) +{ + return __mutex_fastpath_trylock(&lock->count, + __mutex_trylock_slowpath); +} + +EXPORT_SYMBOL(mutex_trylock); + + + diff --git a/kernel/mutex.h b/kernel/mutex.h new file mode 100644 index 000000000000..00fe84e7b672 --- /dev/null +++ b/kernel/mutex.h @@ -0,0 +1,35 @@ +/* + * Mutexes: blocking mutual exclusion locks + * + * started by Ingo Molnar: + * + * Copyright (C) 2004, 2005, 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com> + * + * This file contains mutex debugging related internal prototypes, for the + * !CONFIG_DEBUG_MUTEXES case. Most of them are NOPs: + */ + +#define spin_lock_mutex(lock) spin_lock(lock) +#define spin_unlock_mutex(lock) spin_unlock(lock) +#define mutex_remove_waiter(lock, waiter, ti) \ + __list_del((waiter)->list.prev, (waiter)->list.next) + +#define DEBUG_WARN_ON(c) do { } while (0) +#define debug_mutex_set_owner(lock, new_owner) do { } while (0) +#define debug_mutex_clear_owner(lock) do { } while (0) +#define debug_mutex_init_waiter(waiter) do { } while (0) +#define debug_mutex_wake_waiter(lock, waiter) do { } while (0) +#define debug_mutex_free_waiter(waiter) do { } while (0) +#define debug_mutex_add_waiter(lock, waiter, ti, ip) do { } while (0) +#define debug_mutex_unlock(lock) do { } while (0) +#define debug_mutex_init(lock, name) do { } while (0) + +/* + * Return-address parameters/declarations. They are very useful for + * debugging, but add overhead in the !DEBUG case - so we go the + * trouble of using this not too elegant but zero-cost solution: + */ +#define __IP_DECL__ +#define __IP__ +#define __RET_IP__ + diff --git a/kernel/sched.c b/kernel/sched.c index 92733091154c..34a945bcc022 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4386,6 +4386,7 @@ void show_state(void) } while_each_thread(g, p); read_unlock(&tasklist_lock); + mutex_debug_show_all_locks(); } /** diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index c48260fb8fd9..1fcd856edec1 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -95,6 +95,14 @@ config DEBUG_PREEMPT if kernel code uses it in a preemption-unsafe way. Also, the kernel will detect preemption count underflows. +config DEBUG_MUTEXES + bool "Mutex debugging, deadlock detection" + default y + depends on DEBUG_KERNEL + help + This allows mutex semantics violations and mutex related deadlocks + (lockups) to be detected and reported automatically. + config DEBUG_SPINLOCK bool "Spinlock debugging" depends on DEBUG_KERNEL diff --git a/mm/filemap.c b/mm/filemap.c index 478f4c74cc31..5fca2737c971 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -61,7 +61,7 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * ->swap_lock (exclusive_swap_page, others) * ->mapping->tree_lock * - * ->i_sem + * ->i_mutex * ->i_mmap_lock (truncate->unmap_mapping_range) * * ->mmap_sem @@ -73,9 +73,9 @@ generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, * ->lock_page (access_process_vm) * * ->mmap_sem - * ->i_sem (msync) + * ->i_mutex (msync) * - * ->i_sem + * ->i_mutex * ->i_alloc_sem (various) * * ->inode_lock @@ -276,7 +276,7 @@ static int wait_on_page_writeback_range(struct address_space *mapping, * integrity" operation. It waits upon in-flight writeout before starting and * waiting upon new writeout. If there was an IO error, return it. * - * We need to re-take i_sem during the generic_osync_inode list walk because + * We need to re-take i_mutex during the generic_osync_inode list walk because * it is otherwise livelockable. */ int sync_page_range(struct inode *inode, struct address_space *mapping, @@ -290,9 +290,9 @@ int sync_page_range(struct inode *inode, struct address_space *mapping, return 0; ret = filemap_fdatawrite_range(mapping, pos, pos + count - 1); if (ret == 0) { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = generic_osync_inode(inode, mapping, OSYNC_METADATA); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } if (ret == 0) ret = wait_on_page_writeback_range(mapping, start, end); @@ -301,7 +301,7 @@ int sync_page_range(struct inode *inode, struct address_space *mapping, EXPORT_SYMBOL(sync_page_range); /* - * Note: Holding i_sem across sync_page_range_nolock is not a good idea + * Note: Holding i_mutex across sync_page_range_nolock is not a good idea * as it forces O_SYNC writers to different parts of the same file * to be serialised right until io completion. */ @@ -1892,7 +1892,7 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov, /* * Sync the fs metadata but not the minor inode changes and * of course not the data as we did direct DMA for the IO. - * i_sem is held, which protects generic_osync_inode() from + * i_mutex is held, which protects generic_osync_inode() from * livelocking. */ if (written >= 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { @@ -2195,10 +2195,10 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const char __user *buf, BUG_ON(iocb->ki_pos != pos); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = __generic_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { ssize_t err; @@ -2220,9 +2220,9 @@ ssize_t generic_file_write(struct file *file, const char __user *buf, struct iovec local_iov = { .iov_base = (void __user *)buf, .iov_len = count }; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = __generic_file_write_nolock(file, &local_iov, 1, ppos); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { ssize_t err; @@ -2256,9 +2256,9 @@ ssize_t generic_file_writev(struct file *file, const struct iovec *iov, struct inode *inode = mapping->host; ssize_t ret; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); ret = __generic_file_write_nolock(file, iov, nr_segs, ppos); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { int err; @@ -2272,7 +2272,7 @@ ssize_t generic_file_writev(struct file *file, const struct iovec *iov, EXPORT_SYMBOL(generic_file_writev); /* - * Called under i_sem for writes to S_ISREG files. Returns -EIO if something + * Called under i_mutex for writes to S_ISREG files. Returns -EIO if something * went wrong during pagecache shootdown. */ static ssize_t diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c index 9cf687e4a29a..e2b34e95913e 100644 --- a/mm/filemap_xip.c +++ b/mm/filemap_xip.c @@ -338,7 +338,7 @@ __xip_file_write(struct file *filp, const char __user *buf, *ppos = pos; /* * No need to use i_size_read() here, the i_size - * cannot change under us because we hold i_sem. + * cannot change under us because we hold i_mutex. */ if (pos > inode->i_size) { i_size_write(inode, pos); @@ -358,7 +358,7 @@ xip_file_write(struct file *filp, const char __user *buf, size_t len, loff_t pos; ssize_t ret; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (!access_ok(VERIFY_READ, buf, len)) { ret=-EFAULT; @@ -390,7 +390,7 @@ xip_file_write(struct file *filp, const char __user *buf, size_t len, out_backing: current->backing_dev_info = NULL; out_up: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return ret; } EXPORT_SYMBOL_GPL(xip_file_write); diff --git a/mm/memory.c b/mm/memory.c index 3944fec38012..7a11ddd5060f 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1784,13 +1784,13 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end) if (!inode->i_op || !inode->i_op->truncate_range) return -ENOSYS; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); down_write(&inode->i_alloc_sem); unmap_mapping_range(mapping, offset, (end - offset), 1); truncate_inode_pages_range(mapping, offset, end); inode->i_op->truncate_range(inode, offset, end); up_write(&inode->i_alloc_sem); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return 0; } diff --git a/mm/msync.c b/mm/msync.c index 1b5b6f662dcf..3563a56e1a51 100644 --- a/mm/msync.c +++ b/mm/msync.c @@ -137,7 +137,7 @@ static int msync_interval(struct vm_area_struct *vma, ret = filemap_fdatawrite(mapping); if (file->f_op && file->f_op->fsync) { /* - * We don't take i_sem here because mmap_sem + * We don't take i_mutex here because mmap_sem * is already held. */ err = file->f_op->fsync(file,file->f_dentry,1); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e0e84924171b..a5e6891f7bb6 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -415,6 +415,9 @@ static void __free_pages_ok(struct page *page, unsigned int order) int reserved = 0; arch_free_page(page, order); + if (!PageHighMem(page)) + mutex_debug_check_no_locks_freed(page_address(page), + page_address(page+(1<<order))); #ifndef CONFIG_MMU for (i = 1 ; i < (1 << order) ; ++i) diff --git a/mm/rmap.c b/mm/rmap.c index 66ec43053a4d..dfbb89f99a15 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -20,13 +20,13 @@ /* * Lock ordering in mm: * - * inode->i_sem (while writing or truncating, not reading or faulting) + * inode->i_mutex (while writing or truncating, not reading or faulting) * inode->i_alloc_sem * * When a page fault occurs in writing from user to file, down_read - * of mmap_sem nests within i_sem; in sys_msync, i_sem nests within - * down_read of mmap_sem; i_sem and down_write of mmap_sem are never - * taken together; in truncation, i_sem is taken outermost. + * of mmap_sem nests within i_mutex; in sys_msync, i_mutex nests within + * down_read of mmap_sem; i_mutex and down_write of mmap_sem are never + * taken together; in truncation, i_mutex is taken outermost. * * mm->mmap_sem * page->flags PG_locked (lock_page) diff --git a/mm/shmem.c b/mm/shmem.c index a1f2f02af724..343b3c0937e5 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1370,7 +1370,7 @@ shmem_file_write(struct file *file, const char __user *buf, size_t count, loff_t if (!access_ok(VERIFY_READ, buf, count)) return -EFAULT; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); pos = *ppos; written = 0; @@ -1455,7 +1455,7 @@ shmem_file_write(struct file *file, const char __user *buf, size_t count, loff_t if (written) err = written; out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return err; } @@ -1491,7 +1491,7 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_ /* * We must evaluate after, since reads (unlike writes) - * are called without i_sem protection against truncate + * are called without i_mutex protection against truncate */ nr = PAGE_CACHE_SIZE; i_size = i_size_read(inode); diff --git a/mm/slab.c b/mm/slab.c index 1c46c6383552..33aab345cd4a 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -3071,6 +3071,7 @@ void kfree(const void *objp) local_irq_save(flags); kfree_debugcheck(objp); c = page_get_cache(virt_to_page(objp)); + mutex_debug_check_no_locks_freed(objp, objp+obj_reallen(c)); __cache_free(c, (void *)objp); local_irq_restore(flags); } diff --git a/mm/swapfile.c b/mm/swapfile.c index 80f948a2028b..6544565a7c0f 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -1187,9 +1187,9 @@ asmlinkage long sys_swapoff(const char __user * specialfile) set_blocksize(bdev, p->old_block_size); bd_release(bdev); } else { - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); inode->i_flags &= ~S_SWAPFILE; - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } filp_close(swap_file, NULL); err = 0; @@ -1406,7 +1406,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags) p->bdev = bdev; } else if (S_ISREG(inode->i_mode)) { p->bdev = inode->i_sb->s_bdev; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); did_down = 1; if (IS_SWAPFILE(inode)) { error = -EBUSY; @@ -1596,7 +1596,7 @@ out: if (did_down) { if (!error) inode->i_flags |= S_SWAPFILE; - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } return error; } diff --git a/mm/truncate.c b/mm/truncate.c index b1a463d0fe71..6cb3fff25f67 100644 --- a/mm/truncate.c +++ b/mm/truncate.c @@ -196,7 +196,7 @@ EXPORT_SYMBOL(truncate_inode_pages_range); * @mapping: mapping to truncate * @lstart: offset from which to truncate * - * Called under (and serialised by) inode->i_sem. + * Called under (and serialised by) inode->i_mutex. */ void truncate_inode_pages(struct address_space *mapping, loff_t lstart) { diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index e14c1cae7460..9764c80ab0b2 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -69,13 +69,13 @@ rpc_timeout_upcall_queue(void *data) struct rpc_inode *rpci = (struct rpc_inode *)data; struct inode *inode = &rpci->vfs_inode; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops == NULL) goto out; if (rpci->nreaders == 0 && !list_empty(&rpci->pipe)) __rpc_purge_upcall(inode, -ETIMEDOUT); out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } int @@ -84,7 +84,7 @@ rpc_queue_upcall(struct inode *inode, struct rpc_pipe_msg *msg) struct rpc_inode *rpci = RPC_I(inode); int res = -EPIPE; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops == NULL) goto out; if (rpci->nreaders) { @@ -100,7 +100,7 @@ rpc_queue_upcall(struct inode *inode, struct rpc_pipe_msg *msg) res = 0; } out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); wake_up(&rpci->waitq); return res; } @@ -116,7 +116,7 @@ rpc_close_pipes(struct inode *inode) { struct rpc_inode *rpci = RPC_I(inode); - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops != NULL) { rpci->nreaders = 0; __rpc_purge_list(rpci, &rpci->in_upcall, -EPIPE); @@ -127,7 +127,7 @@ rpc_close_pipes(struct inode *inode) rpci->ops = NULL; } rpc_inode_setowner(inode, NULL); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); cancel_delayed_work(&rpci->queue_timeout); flush_scheduled_work(); } @@ -154,7 +154,7 @@ rpc_pipe_open(struct inode *inode, struct file *filp) struct rpc_inode *rpci = RPC_I(inode); int res = -ENXIO; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops != NULL) { if (filp->f_mode & FMODE_READ) rpci->nreaders ++; @@ -162,7 +162,7 @@ rpc_pipe_open(struct inode *inode, struct file *filp) rpci->nwriters ++; res = 0; } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return res; } @@ -172,7 +172,7 @@ rpc_pipe_release(struct inode *inode, struct file *filp) struct rpc_inode *rpci = RPC_I(inode); struct rpc_pipe_msg *msg; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops == NULL) goto out; msg = (struct rpc_pipe_msg *)filp->private_data; @@ -190,7 +190,7 @@ rpc_pipe_release(struct inode *inode, struct file *filp) if (rpci->ops->release_pipe) rpci->ops->release_pipe(inode); out: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return 0; } @@ -202,7 +202,7 @@ rpc_pipe_read(struct file *filp, char __user *buf, size_t len, loff_t *offset) struct rpc_pipe_msg *msg; int res = 0; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); if (rpci->ops == NULL) { res = -EPIPE; goto out_unlock; @@ -229,7 +229,7 @@ rpc_pipe_read(struct file *filp, char __user *buf, size_t len, loff_t *offset) rpci->ops->destroy_msg(msg); } out_unlock: - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return res; } @@ -240,11 +240,11 @@ rpc_pipe_write(struct file *filp, const char __user *buf, size_t len, loff_t *of struct rpc_inode *rpci = RPC_I(inode); int res; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); res = -EPIPE; if (rpci->ops != NULL) res = rpci->ops->downcall(filp, buf, len); - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); return res; } @@ -322,7 +322,7 @@ rpc_info_open(struct inode *inode, struct file *file) if (!ret) { struct seq_file *m = file->private_data; - down(&inode->i_sem); + mutex_lock(&inode->i_mutex); clnt = RPC_I(inode)->private; if (clnt) { atomic_inc(&clnt->cl_users); @@ -331,7 +331,7 @@ rpc_info_open(struct inode *inode, struct file *file) single_release(inode, file); ret = -EINVAL; } - up(&inode->i_sem); + mutex_unlock(&inode->i_mutex); } return ret; } @@ -491,7 +491,7 @@ rpc_depopulate(struct dentry *parent) struct dentry *dentry, *dvec[10]; int n = 0; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); repeat: spin_lock(&dcache_lock); list_for_each_safe(pos, next, &parent->d_subdirs) { @@ -519,7 +519,7 @@ repeat: } while (n); goto repeat; } - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); } static int @@ -532,7 +532,7 @@ rpc_populate(struct dentry *parent, struct dentry *dentry; int mode, i; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); for (i = start; i < eof; i++) { dentry = d_alloc_name(parent, files[i].name); if (!dentry) @@ -552,10 +552,10 @@ rpc_populate(struct dentry *parent, dir->i_nlink++; d_add(dentry, inode); } - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); return 0; out_bad: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); printk(KERN_WARNING "%s: %s failed to populate directory %s\n", __FILE__, __FUNCTION__, parent->d_name.name); return -ENOMEM; @@ -609,7 +609,7 @@ rpc_lookup_negative(char *path, struct nameidata *nd) if ((error = rpc_lookup_parent(path, nd)) != 0) return ERR_PTR(error); dir = nd->dentry->d_inode; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); dentry = lookup_hash(nd); if (IS_ERR(dentry)) goto out_err; @@ -620,7 +620,7 @@ rpc_lookup_negative(char *path, struct nameidata *nd) } return dentry; out_err: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); rpc_release_path(nd); return dentry; } @@ -646,7 +646,7 @@ rpc_mkdir(char *path, struct rpc_clnt *rpc_client) if (error) goto err_depopulate; out: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); rpc_release_path(&nd); return dentry; err_depopulate: @@ -671,7 +671,7 @@ rpc_rmdir(char *path) if ((error = rpc_lookup_parent(path, &nd)) != 0) return error; dir = nd.dentry->d_inode; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); dentry = lookup_hash(&nd); if (IS_ERR(dentry)) { error = PTR_ERR(dentry); @@ -681,7 +681,7 @@ rpc_rmdir(char *path) error = __rpc_rmdir(dir, dentry); dput(dentry); out_release: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); rpc_release_path(&nd); return error; } @@ -710,7 +710,7 @@ rpc_mkpipe(char *path, void *private, struct rpc_pipe_ops *ops, int flags) rpci->ops = ops; inode_dir_notify(dir, DN_CREATE); out: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); rpc_release_path(&nd); return dentry; err_dput: @@ -732,7 +732,7 @@ rpc_unlink(char *path) if ((error = rpc_lookup_parent(path, &nd)) != 0) return error; dir = nd.dentry->d_inode; - down(&dir->i_sem); + mutex_lock(&dir->i_mutex); dentry = lookup_hash(&nd); if (IS_ERR(dentry)) { error = PTR_ERR(dentry); @@ -746,7 +746,7 @@ rpc_unlink(char *path) dput(dentry); inode_dir_notify(dir, DN_DELETE); out_release: - up(&dir->i_sem); + mutex_unlock(&dir->i_mutex); rpc_release_path(&nd); return error; } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 5f6ae79b8b16..1b5989b1b670 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -784,7 +784,7 @@ static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) err = vfs_mknod(nd.dentry->d_inode, dentry, mode, 0); if (err) goto out_mknod_dput; - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); dput(nd.dentry); nd.dentry = dentry; @@ -823,7 +823,7 @@ out: out_mknod_dput: dput(dentry); out_mknod_unlock: - up(&nd.dentry->d_inode->i_sem); + mutex_unlock(&nd.dentry->d_inode->i_mutex); path_release(&nd); out_mknod_parent: if (err==-EEXIST) diff --git a/security/inode.c b/security/inode.c index a5964502ae30..0f77b0223662 100644 --- a/security/inode.c +++ b/security/inode.c @@ -172,7 +172,7 @@ static int create_by_name(const char *name, mode_t mode, return -EFAULT; } - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); *dentry = lookup_one_len(name, parent, strlen(name)); if (!IS_ERR(dentry)) { if ((mode & S_IFMT) == S_IFDIR) @@ -181,7 +181,7 @@ static int create_by_name(const char *name, mode_t mode, error = create(parent->d_inode, *dentry, mode); } else error = PTR_ERR(dentry); - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); return error; } @@ -302,7 +302,7 @@ void securityfs_remove(struct dentry *dentry) if (!parent || !parent->d_inode) return; - down(&parent->d_inode->i_sem); + mutex_lock(&parent->d_inode->i_mutex); if (positive(dentry)) { if (dentry->d_inode) { if (S_ISDIR(dentry->d_inode->i_mode)) @@ -312,7 +312,7 @@ void securityfs_remove(struct dentry *dentry) dput(dentry); } } - up(&parent->d_inode->i_sem); + mutex_unlock(&parent->d_inode->i_mutex); simple_release_fs(&mount, &mount_count); } EXPORT_SYMBOL_GPL(securityfs_remove); diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index 16df1246a131..7fd072392c7e 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -2135,9 +2135,7 @@ static ssize_t snd_pcm_oss_write(struct file *file, const char __user *buf, size substream = pcm_oss_file->streams[SNDRV_PCM_STREAM_PLAYBACK]; if (substream == NULL) return -ENXIO; - up(&file->f_dentry->d_inode->i_sem); result = snd_pcm_oss_write1(substream, buf, count); - down(&file->f_dentry->d_inode->i_sem); #ifdef OSS_DEBUG printk("pcm_oss: write %li bytes (wrote %li bytes)\n", (long)count, (long)result); #endif diff --git a/sound/core/seq/seq_memory.c b/sound/core/seq/seq_memory.c index 9ee6c177db0c..40b4f679c80e 100644 --- a/sound/core/seq/seq_memory.c +++ b/sound/core/seq/seq_memory.c @@ -32,10 +32,6 @@ #include "seq_info.h" #include "seq_lock.h" -/* semaphore in struct file record */ -#define semaphore_of(fp) ((fp)->f_dentry->d_inode->i_sem) - - static inline int snd_seq_pool_available(struct snd_seq_pool *pool) { return pool->total_elements - atomic_read(&pool->counter); |