From f629299b544b6cc12b4e3e85fec96f4ce5809482 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 24 Jul 2011 23:15:42 +0200 Subject: trace events: Update version number reference to new 3.x scheme for EVENT_POWER_TRACING_DEPRECATED What was scheduled to be 2.6.41 is now going to be 3.1 . Signed-off-by: Jesper Juhl Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker Cc: Steven Rostedt Link: http://lkml.kernel.org/r/alpine.LNX.2.00.1107250929370.8080@swampdragon.chaosbits.net Signed-off-by: Ingo Molnar --- kernel/trace/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 2ad39e556cb4..cd3134510f3d 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -82,7 +82,7 @@ config EVENT_POWER_TRACING_DEPRECATED power:power_frequency This is for userspace compatibility and will vanish after 5 kernel iterations, - namely 2.6.41. + namely 3.1. config CONTEXT_SWITCH_TRACER bool -- cgit v1.3-8-gc7d7 From 1dd75f91ae713049eb6baaa640078f3a6549e522 Mon Sep 17 00:00:00 2001 From: "jhbird.choi@samsung.com" Date: Thu, 21 Jul 2011 15:29:14 +0900 Subject: genirq: Fix wrong bit operation (!msk & 0x01) should be !(msk & 0x01) Signed-off-by: Jonghwan Choi Link: http://lkml.kernel.org/r/1311229754-6003-1-git-send-email-jhbird.choi@samsung.com Signed-off-by: Thomas Gleixner Cc: stable@kernel.org --- kernel/irq/generic-chip.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index 3a2cab407b93..e38544dddb18 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -246,7 +246,7 @@ void irq_setup_generic_chip(struct irq_chip_generic *gc, u32 msk, gc->mask_cache = irq_reg_readl(gc->reg_base + ct->regs.mask); for (i = gc->irq_base; msk; msk >>= 1, i++) { - if (!msk & 0x01) + if (!(msk & 0x01)) continue; if (flags & IRQ_GC_INIT_NESTED_LOCK) @@ -301,7 +301,7 @@ void irq_remove_generic_chip(struct irq_chip_generic *gc, u32 msk, raw_spin_unlock(&gc_lock); for (; msk; msk >>= 1, i++) { - if (!msk & 0x01) + if (!(msk & 0x01)) continue; /* Remove handler first. That will mask the irq line */ -- cgit v1.3-8-gc7d7 From f3637a5f2e2eb391ff5757bc83fb5de8f9726464 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Thu, 7 Jul 2011 22:32:17 +0200 Subject: irq: Always set IRQF_ONESHOT if no primary handler is specified If no primary handler is specified then a default one is assigned which always returns IRQ_WAKE_THREAD. This handler requires the IRQF_ONESHOT flag on LEVEL / EIO typed irqs because the source of interrupt is not disabled. Since it is required for those users and there is no difference for others it makes sense to add this flag unconditionally. Signed-off-by: Sebastian Andrzej Siewior Link: http://lkml.kernel.org/r/1310070737-18514-1-git-send-email-bigeasy@linutronix.de Signed-off-by: Thomas Gleixner --- kernel/irq/manage.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 0a7840aeb0fb..3f9cd4799da7 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1322,6 +1322,7 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, if (!thread_fn) return -EINVAL; handler = irq_default_primary_handler; + irqflags |= IRQF_ONESHOT; } action = kzalloc(sizeof(struct irqaction), GFP_KERNEL); -- cgit v1.3-8-gc7d7 From b6873807a7143b7d6d8b06809295e559d07d7deb Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Mon, 11 Jul 2011 12:17:31 +0200 Subject: irq: Track the owner of irq descriptor Interrupt descriptors can be allocated from modules. The interrupts are used by other modules, but we have no refcount on the module which provides the interrupts and there is no way to establish one on the device level as the interrupt using module is agnostic to the fact that the interrupt is provided by a module rather than by some builtin interrupt controller. To prevent removal of the interrupt providing module, we can track the owner of the interrupt descriptor, which also provides the relevant irq chip functions in the irq descriptor. request/setup_irq() can now acquire a refcount on the owner module to prevent unloading. free_irq() drops the refcount. Signed-off-by: Sebastian Andrzej Siewior Link: http://lkml.kernel.org/r/20110711101731.GA13804@Chamillionaire.breakpoint.cc Signed-off-by: Thomas Gleixner --- include/linux/irq.h | 11 ++++++++++- include/linux/irqdesc.h | 1 + kernel/irq/irqdesc.c | 36 ++++++++++++++++++++++++------------ kernel/irq/manage.c | 17 +++++++++++++---- 4 files changed, 48 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/include/linux/irq.h b/include/linux/irq.h index baa397eb9c33..16d6f54ef1dd 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -546,7 +547,15 @@ static inline struct msi_desc *irq_data_get_msi(struct irq_data *d) return d->msi_desc; } -int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node); +int __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, + struct module *owner); + +static inline int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, + int node) +{ + return __irq_alloc_descs(irq, from, cnt, node, THIS_MODULE); +} + void irq_free_descs(unsigned int irq, unsigned int cnt); int irq_reserve_irqs(unsigned int from, unsigned int cnt); diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 2d921b35212c..150134ac709a 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -66,6 +66,7 @@ struct irq_desc { #ifdef CONFIG_PROC_FS struct proc_dir_entry *dir; #endif + struct module *owner; const char *name; } ____cacheline_internodealigned_in_smp; diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 4c60a50e66b2..cb65d0360e31 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -70,7 +70,8 @@ static inline void desc_smp_init(struct irq_desc *desc, int node) { } static inline int desc_node(struct irq_desc *desc) { return 0; } #endif -static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node) +static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node, + struct module *owner) { int cpu; @@ -86,6 +87,7 @@ static void desc_set_defaults(unsigned int irq, struct irq_desc *desc, int node) desc->irq_count = 0; desc->irqs_unhandled = 0; desc->name = NULL; + desc->owner = owner; for_each_possible_cpu(cpu) *per_cpu_ptr(desc->kstat_irqs, cpu) = 0; desc_smp_init(desc, node); @@ -128,7 +130,7 @@ static void free_masks(struct irq_desc *desc) static inline void free_masks(struct irq_desc *desc) { } #endif -static struct irq_desc *alloc_desc(int irq, int node) +static struct irq_desc *alloc_desc(int irq, int node, struct module *owner) { struct irq_desc *desc; gfp_t gfp = GFP_KERNEL; @@ -147,7 +149,7 @@ static struct irq_desc *alloc_desc(int irq, int node) raw_spin_lock_init(&desc->lock); lockdep_set_class(&desc->lock, &irq_desc_lock_class); - desc_set_defaults(irq, desc, node); + desc_set_defaults(irq, desc, node, owner); return desc; @@ -173,13 +175,14 @@ static void free_desc(unsigned int irq) kfree(desc); } -static int alloc_descs(unsigned int start, unsigned int cnt, int node) +static int alloc_descs(unsigned int start, unsigned int cnt, int node, + struct module *owner) { struct irq_desc *desc; int i; for (i = 0; i < cnt; i++) { - desc = alloc_desc(start + i, node); + desc = alloc_desc(start + i, node, owner); if (!desc) goto err; mutex_lock(&sparse_irq_lock); @@ -227,7 +230,7 @@ int __init early_irq_init(void) nr_irqs = initcnt; for (i = 0; i < initcnt; i++) { - desc = alloc_desc(i, node); + desc = alloc_desc(i, node, NULL); set_bit(i, allocated_irqs); irq_insert_desc(i, desc); } @@ -261,7 +264,7 @@ int __init early_irq_init(void) alloc_masks(&desc[i], GFP_KERNEL, node); raw_spin_lock_init(&desc[i].lock); lockdep_set_class(&desc[i].lock, &irq_desc_lock_class); - desc_set_defaults(i, &desc[i], node); + desc_set_defaults(i, &desc[i], node, NULL); } return arch_early_irq_init(); } @@ -276,8 +279,16 @@ static void free_desc(unsigned int irq) dynamic_irq_cleanup(irq); } -static inline int alloc_descs(unsigned int start, unsigned int cnt, int node) +static inline int alloc_descs(unsigned int start, unsigned int cnt, int node, + struct module *owner) { + u32 i; + + for (i = 0; i < cnt; i++) { + struct irq_desc *desc = irq_to_desc(start + i); + + desc->owner = owner; + } return start; } @@ -337,7 +348,8 @@ EXPORT_SYMBOL_GPL(irq_free_descs); * Returns the first irq number or error code */ int __ref -irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node) +__irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, + struct module *owner) { int start, ret; @@ -366,13 +378,13 @@ irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node) bitmap_set(allocated_irqs, start, cnt); mutex_unlock(&sparse_irq_lock); - return alloc_descs(start, cnt, node); + return alloc_descs(start, cnt, node, owner); err: mutex_unlock(&sparse_irq_lock); return ret; } -EXPORT_SYMBOL_GPL(irq_alloc_descs); +EXPORT_SYMBOL_GPL(__irq_alloc_descs); /** * irq_reserve_irqs - mark irqs allocated @@ -440,7 +452,7 @@ void dynamic_irq_cleanup(unsigned int irq) unsigned long flags; raw_spin_lock_irqsave(&desc->lock, flags); - desc_set_defaults(irq, desc, desc_node(desc)); + desc_set_defaults(irq, desc, desc_node(desc), NULL); raw_spin_unlock_irqrestore(&desc->lock, flags); } diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 3f9cd4799da7..2e9425889fa8 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -883,6 +883,8 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (desc->irq_data.chip == &no_irq_chip) return -ENOSYS; + if (!try_module_get(desc->owner)) + return -ENODEV; /* * Some drivers like serial.c use request_irq() heavily, * so we have to be careful not to interfere with a @@ -906,8 +908,10 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) */ nested = irq_settings_is_nested_thread(desc); if (nested) { - if (!new->thread_fn) - return -EINVAL; + if (!new->thread_fn) { + ret = -EINVAL; + goto out_mput; + } /* * Replace the primary handler which was provided from * the driver for non nested interrupt handling by the @@ -929,8 +933,10 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) t = kthread_create(irq_thread, new, "irq/%d-%s", irq, new->name); - if (IS_ERR(t)) - return PTR_ERR(t); + if (IS_ERR(t)) { + ret = PTR_ERR(t); + goto out_mput; + } /* * We keep the reference to the task struct even if * the thread dies to avoid that the interrupt code @@ -1095,6 +1101,8 @@ out_thread: kthread_stop(t); put_task_struct(t); } +out_mput: + module_put(desc->owner); return ret; } @@ -1203,6 +1211,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) put_task_struct(action->thread); } + module_put(desc->owner); return action; } -- cgit v1.3-8-gc7d7 From b77f0f3c1f587791aa5d9bd1b0012c9a89eb9258 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Fri, 5 Aug 2011 16:40:40 -0400 Subject: jump label: Reduce the cycle count by changing the link order In the course of testing jump labels for use with the CFS bandwidth controller, Paul Turner, discovered that using jump labels reduced the branch count and the instruction count, but did not reduce the cycle count or wall time. I noticed that having the jump_label.o included in the kernel but not used in any way still caused this increase in cycle count and wall time. Thus, I moved jump_label.o in the kernel/Makefile, thus changing the link order, and presumably moving it out of hot icache areas. This brought down the cycle count/time as expected. In addition to Paul's testing, I've tested the patch using a single 'static_branch()' in the getppid() path, and basically running tight loops of calls to getppid(). Here are my results for the branch disabled case: With jump labels turned on (CONFIG_JUMP_LABEL), branch disabled: Performance counter stats for 'bash -c /tmp/getppid;true' (50 runs): 3,969,510,217 instructions # 0.864 IPC ( +-0.000% ) 4,592,334,954 cycles ( +- 0.046% ) 751,634,470 branches ( +- 0.000% ) 1.722635797 seconds time elapsed ( +- 0.046% ) Jump labels turned off (CONFIG_JUMP_LABEL not set), branch disabled: Performance counter stats for 'bash -c /tmp/getppid;true' (50 runs): 4,009,611,846 instructions # 0.867 IPC ( +-0.000% ) 4,622,210,580 cycles ( +- 0.012% ) 771,662,904 branches ( +- 0.000% ) 1.734341454 seconds time elapsed ( +- 0.022% ) Signed-off-by: Jason Baron Cc: rth@redhat.com Cc: a.p.zijlstra@chello.nl Cc: rostedt@goodmis.org Link: http://lkml.kernel.org/r/20110805204040.GG2522@redhat.com Signed-off-by: Ingo Molnar Tested-by: Paul Turner --- kernel/Makefile | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/Makefile b/kernel/Makefile index d06467fc8f7c..eca595e2fd52 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -10,7 +10,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ - async.o range.o jump_label.o + async.o range.o obj-y += groups.o ifdef CONFIG_FUNCTION_TRACER @@ -107,6 +107,7 @@ obj-$(CONFIG_PERF_EVENTS) += events/ obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o obj-$(CONFIG_PADATA) += padata.o obj-$(CONFIG_CRASH_DUMP) += crash_dump.o +obj-$(CONFIG_JUMP_LABEL) += jump_label.o ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) # According to Alan Modra , the -fno-omit-frame-pointer is -- cgit v1.3-8-gc7d7 From 80e0401e35410a69bfae05b454db8a7187edd6b8 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 5 Aug 2011 14:26:17 +0200 Subject: lockdep: Fix wrong assumption in match_held_lock match_held_lock() was assuming it was being called on a lock class that had already seen usage. This condition was true for bug-free code using lockdep_assert_held(), since you're in fact holding the lock when calling it. However the assumption fails the moment you assume the assertion can fail, which is the whole point of having the assertion in the first place. Anyway, now that there's more lockdep_is_held() users, notably __rcu_dereference_check(), its much easier to trigger this since we test for a number of locks and we only need to hold any one of them to be good. Reported-by: Sergey Senozhatsky Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1312547787.28695.2.camel@twins Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 8c24294e477f..91d67ce3a8d5 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -3111,7 +3111,13 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) if (!class) class = look_up_lock_class(lock, 0); - if (DEBUG_LOCKS_WARN_ON(!class)) + /* + * If look_up_lock_class() failed to find a class, we're trying + * to test if we hold a lock that has never yet been acquired. + * Clearly if the lock hasn't been acquired _ever_, we're not + * holding it either, so report failure. + */ + if (!class) return 0; if (DEBUG_LOCKS_WARN_ON(!hlock->nest_lock)) -- cgit v1.3-8-gc7d7 From f2c0d0266cc5eb36a4aa44944b4096ec121490aa Mon Sep 17 00:00:00 2001 From: Jonathan Nieder Date: Mon, 8 Aug 2011 06:22:43 +0200 Subject: cap_syslog: don't use WARN_ONCE for CAP_SYS_ADMIN deprecation warning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit syslog-ng versions before 3.3.0beta1 (2011-05-12) assume that CAP_SYS_ADMIN is sufficient to access syslog, so ever since CAP_SYSLOG was introduced (2010-11-25) they have triggered a warning. Commit ee24aebffb75 ("cap_syslog: accept CAP_SYS_ADMIN for now") improved matters a little by making syslog-ng work again, just keeping the WARN_ONCE(). But still, this is a warning that writes a stack trace we don't care about to syslog, sets a taint flag, and alarms sysadmins when nothing worse has happened than use of an old userspace with a recent kernel. Convert the WARN_ONCE to a printk_once to avoid that while continuing to give userspace developers a hint that this is an unwanted backward-compatibility feature and won't be around forever. Reported-by: Ralf Hildebrandt Reported-by: Niels Reported-by: Paweł Sikora Signed-off-by: Jonathan Nieder Liked-by: Gergely Nagy Acked-by: Serge Hallyn Acked-by: James Morris Signed-off-by: Linus Torvalds --- kernel/printk.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 37dff3429adb..836a2ae0ac31 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -318,8 +318,10 @@ static int check_syslog_permissions(int type, bool from_file) return 0; /* For historical reasons, accept CAP_SYS_ADMIN too, with a warning */ if (capable(CAP_SYS_ADMIN)) { - WARN_ONCE(1, "Attempt to access syslog with CAP_SYS_ADMIN " - "but no CAP_SYSLOG (deprecated).\n"); + printk_once(KERN_WARNING "%s (%d): " + "Attempt to access syslog with CAP_SYS_ADMIN " + "but no CAP_SYSLOG (deprecated).\n", + current->comm, task_pid_nr(current)); return 0; } return -EPERM; -- cgit v1.3-8-gc7d7 From 971c90bfa2f0b4fe52d6d9002178d547706f1343 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 4 Aug 2011 07:25:35 -0700 Subject: alarmtimers: Avoid possible null pointer traversal We don't check if old_setting is non null before assigning it, so correct this. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 59f369f98a04..1dee3f62a6a7 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -479,11 +479,8 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, if (!rtcdev) return -ENOTSUPP; - /* Save old values */ - old_setting->it_interval = - ktime_to_timespec(timr->it.alarmtimer.period); - old_setting->it_value = - ktime_to_timespec(timr->it.alarmtimer.node.expires); + if (old_setting) + alarm_timer_get(timr, old_setting); /* If the timer was already set, cancel it */ alarm_cancel(&timr->it.alarmtimer); -- cgit v1.3-8-gc7d7 From ea7802f630d356acaf66b3c0b28c00a945fc35dc Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 4 Aug 2011 07:51:56 -0700 Subject: alarmtimers: Memset itimerspec passed into alarm_timer_get Following common_timer_get, zero out the itimerspec passed in. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 1dee3f62a6a7..0e9263f6fd09 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -441,6 +441,8 @@ static int alarm_timer_create(struct k_itimer *new_timer) static void alarm_timer_get(struct k_itimer *timr, struct itimerspec *cur_setting) { + memset(cur_setting, 0, sizeof(struct itimerspec)); + cur_setting->it_interval = ktime_to_timespec(timr->it.alarmtimer.period); cur_setting->it_value = -- cgit v1.3-8-gc7d7 From 6af7e471e5a7746b8024d70b4363d3dfe41d36b8 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 10:26:09 -0700 Subject: alarmtimers: Avoid possible denial of service with high freq periodic timers Its possible to jam up the alarm timers by setting very small interval timers, which will cause the alarmtimer subsystem to spend all of its time firing and restarting timers. This can effectivly lock up a box. A deeper fix is needed, closely mimicking the hrtimer code, but for now just cap the interval to 100us to avoid userland hanging the system. CC: Thomas Gleixner CC: stable@kernel.org Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 0e9263f6fd09..ea5e1a928d5b 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -481,6 +481,15 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, if (!rtcdev) return -ENOTSUPP; + /* + * XXX HACK! Currently we can DOS a system if the interval + * period on alarmtimers is too small. Cap the interval here + * to 100us and solve this properly in a future patch! -jstultz + */ + if ((new_setting->it_interval.tv_sec == 0) && + (new_setting->it_interval.tv_nsec < 100000)) + new_setting->it_interval.tv_nsec = 100000; + if (old_setting) alarm_timer_get(timr, old_setting); -- cgit v1.3-8-gc7d7 From 4b41308d2d0398409620613c7eaaaf52c738b042 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 10:37:59 -0700 Subject: alarmtimers: Change alarmtimer functions to return alarmtimer_restart values In order to properly fix the denial of service issue with high freq periodic alarm timers, we need to push the re-arming logic into the alarm timer handler, much as the hrtimer code does. This patch introduces alarmtimer_restart enum and changes the alarmtimer handler declarations to use it as a return value. Further, to ease following changes, it extends the alarmtimer handler functions to also take the time at expiration. No logic is yet modified. CC: Thomas Gleixner Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 9 +++++++-- kernel/time/alarmtimer.c | 13 +++++++++---- 2 files changed, 16 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index c5d6095b46f8..0289eb29e794 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -13,6 +13,11 @@ enum alarmtimer_type { ALARM_NUMTYPE, }; +enum alarmtimer_restart { + ALARMTIMER_NORESTART, + ALARMTIMER_RESTART, +}; + /** * struct alarm - Alarm timer structure * @node: timerqueue node for adding to the event list this value @@ -26,14 +31,14 @@ enum alarmtimer_type { struct alarm { struct timerqueue_node node; ktime_t period; - void (*function)(struct alarm *); + enum alarmtimer_restart (*function)(struct alarm *, ktime_t now); enum alarmtimer_type type; bool enabled; void *data; }; void alarm_init(struct alarm *alarm, enum alarmtimer_type type, - void (*function)(struct alarm *)); + enum alarmtimer_restart (*function)(struct alarm *, ktime_t)); void alarm_start(struct alarm *alarm, ktime_t start, ktime_t period); void alarm_cancel(struct alarm *alarm); diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index ea5e1a928d5b..9e786053ffa2 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -196,7 +196,7 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) } spin_unlock_irqrestore(&base->lock, flags); if (alarm->function) - alarm->function(alarm); + alarm->function(alarm, now); spin_lock_irqsave(&base->lock, flags); } @@ -299,7 +299,7 @@ static void alarmtimer_freezerset(ktime_t absexp, enum alarmtimer_type type) * @function: callback that is run when the alarm fires */ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, - void (*function)(struct alarm *)) + enum alarmtimer_restart (*function)(struct alarm *, ktime_t)) { timerqueue_init(&alarm->node); alarm->period = ktime_set(0, 0); @@ -365,12 +365,15 @@ static enum alarmtimer_type clock2alarm(clockid_t clockid) * * Posix timer callback for expired alarm timers. */ -static void alarm_handle_timer(struct alarm *alarm) +static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, + ktime_t now) { struct k_itimer *ptr = container_of(alarm, struct k_itimer, it.alarmtimer); if (posix_timer_event(ptr, 0) != 0) ptr->it_overrun++; + + return ALARMTIMER_NORESTART; } /** @@ -509,13 +512,15 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, * * Wakes up the task that set the alarmtimer */ -static void alarmtimer_nsleep_wakeup(struct alarm *alarm) +static enum alarmtimer_restart alarmtimer_nsleep_wakeup(struct alarm *alarm, + ktime_t now) { struct task_struct *task = (struct task_struct *)alarm->data; alarm->data = NULL; if (task) wake_up_process(task); + return ALARMTIMER_NORESTART; } /** -- cgit v1.3-8-gc7d7 From 54da23b720d5d612f8f1669f9ed3744008fb7382 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 11:08:07 -0700 Subject: alarmtimers: Push rearming peroidic timers down into alamrtimer handler This patch pushes the periodic alarmtimer re-arming down into the alarmtimer handler, mimicking how hrtimers handle this. CC: Thomas Gleixner Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 9e786053ffa2..55e634ea6054 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -174,6 +174,7 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) unsigned long flags; ktime_t now; int ret = HRTIMER_NORESTART; + int restart = ALARMTIMER_NORESTART; spin_lock_irqsave(&base->lock, flags); now = base->gettime(); @@ -188,16 +189,16 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) timerqueue_del(&base->timerqueue, &alarm->node); alarm->enabled = 0; - /* Re-add periodic timers */ - if (alarm->period.tv64) { - alarm->node.expires = ktime_add(expired, alarm->period); - timerqueue_add(&base->timerqueue, &alarm->node); - alarm->enabled = 1; - } + spin_unlock_irqrestore(&base->lock, flags); if (alarm->function) - alarm->function(alarm, now); + restart = alarm->function(alarm, now); spin_lock_irqsave(&base->lock, flags); + + if (restart != ALARMTIMER_NORESTART) { + timerqueue_add(&base->timerqueue, &alarm->node); + alarm->enabled = 1; + } } if (next) { @@ -373,6 +374,11 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, if (posix_timer_event(ptr, 0) != 0) ptr->it_overrun++; + /* Re-add periodic timers */ + if (alarm->period.tv64) { + alarm->node.expires = ktime_add(now, alarm->period); + return ALARMTIMER_RESTART; + } return ALARMTIMER_NORESTART; } -- cgit v1.3-8-gc7d7 From dce75a8c71819ed4c7efdcd53c9b6f6356dc8cb5 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 11:31:03 -0700 Subject: alarmtimers: Add alarm_forward functionality In order to avoid wasting time expiring and re-adding very high freq periodic alarmtimers, introduce alarm_forward() which is similar to hrtimer_forward and moves the timer to the next future expiration time and returns the number of overruns. CC: Thomas Gleixner Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 2 ++ kernel/time/alarmtimer.c | 37 ++++++++++++++++++++++++++++++++++++- 2 files changed, 38 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index 0289eb29e794..17535963b274 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -42,4 +42,6 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, void alarm_start(struct alarm *alarm, ktime_t start, ktime_t period); void alarm_cancel(struct alarm *alarm); +u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval); + #endif diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 55e634ea6054..f03b04291b6f 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -347,6 +347,41 @@ void alarm_cancel(struct alarm *alarm) } + +u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval) +{ + u64 overrun = 1; + ktime_t delta; + + delta = ktime_sub(now, alarm->node.expires); + + if (delta.tv64 < 0) + return 0; + + if (unlikely(delta.tv64 >= interval.tv64)) { + s64 incr = ktime_to_ns(interval); + + overrun = ktime_divns(delta, incr); + + alarm->node.expires = ktime_add_ns(alarm->node.expires, + incr*overrun); + + if (alarm->node.expires.tv64 > now.tv64) + return overrun; + /* + * This (and the ktime_add() below) is the + * correction for exact: + */ + overrun++; + } + + alarm->node.expires = ktime_add(alarm->node.expires, interval); + return overrun; +} + + + + /** * clock2alarm - helper that converts from clockid to alarmtypes * @clockid: clockid. @@ -376,7 +411,7 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, /* Re-add periodic timers */ if (alarm->period.tv64) { - alarm->node.expires = ktime_add(now, alarm->period); + ptr->it_overrun += alarm_forward(alarm, now, alarm->period); return ALARMTIMER_RESTART; } return ALARMTIMER_NORESTART; -- cgit v1.3-8-gc7d7 From d77e23accec56bf2ba12187fe77a2f500a511282 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 11:40:23 -0700 Subject: alarmtimers: Remove interval cap limit hack Now that the alarmtimers code has been refactored, the interval cap limit can be removed. CC: Thomas Gleixner Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 9 --------- 1 file changed, 9 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index f03b04291b6f..a522c007e6fd 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -525,15 +525,6 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, if (!rtcdev) return -ENOTSUPP; - /* - * XXX HACK! Currently we can DOS a system if the interval - * period on alarmtimers is too small. Cap the interval here - * to 100us and solve this properly in a future patch! -jstultz - */ - if ((new_setting->it_interval.tv_sec == 0) && - (new_setting->it_interval.tv_nsec < 100000)) - new_setting->it_interval.tv_nsec = 100000; - if (old_setting) alarm_timer_get(timr, old_setting); -- cgit v1.3-8-gc7d7 From 9e26476243e438f4534a562660c1296a15a9e202 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 12:09:24 -0700 Subject: alarmtimers: Remove period from alarm structure Now that periodic alarmtimers are managed by the handler function, remove the period value from the alarm structure and let the handlers manage the interval on their own. CC: Thomas Gleixner Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 3 +-- include/linux/posix-timers.h | 5 ++++- kernel/time/alarmtimer.c | 30 ++++++++++++++---------------- 3 files changed, 19 insertions(+), 19 deletions(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index 17535963b274..c854a8efa863 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -30,7 +30,6 @@ enum alarmtimer_restart { */ struct alarm { struct timerqueue_node node; - ktime_t period; enum alarmtimer_restart (*function)(struct alarm *, ktime_t now); enum alarmtimer_type type; bool enabled; @@ -39,7 +38,7 @@ struct alarm { void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)); -void alarm_start(struct alarm *alarm, ktime_t start, ktime_t period); +void alarm_start(struct alarm *alarm, ktime_t start); void alarm_cancel(struct alarm *alarm); u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval); diff --git a/include/linux/posix-timers.h b/include/linux/posix-timers.h index 959c14132f46..042058fdb0af 100644 --- a/include/linux/posix-timers.h +++ b/include/linux/posix-timers.h @@ -81,7 +81,10 @@ struct k_itimer { unsigned long incr; unsigned long expires; } mmtimer; - struct alarm alarmtimer; + struct { + struct alarm alarmtimer; + ktime_t interval; + } alarm; struct rcu_head rcu; } it; }; diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index a522c007e6fd..90935591dd44 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -303,7 +303,6 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)) { timerqueue_init(&alarm->node); - alarm->period = ktime_set(0, 0); alarm->function = function; alarm->type = type; alarm->enabled = 0; @@ -313,9 +312,8 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, * alarm_start - Sets an alarm to fire * @alarm: ptr to alarm to set * @start: time to run the alarm - * @period: period at which the alarm will recur */ -void alarm_start(struct alarm *alarm, ktime_t start, ktime_t period) +void alarm_start(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; @@ -324,7 +322,6 @@ void alarm_start(struct alarm *alarm, ktime_t start, ktime_t period) if (alarm->enabled) alarmtimer_remove(base, alarm); alarm->node.expires = start; - alarm->period = period; alarmtimer_enqueue(base, alarm); alarm->enabled = 1; spin_unlock_irqrestore(&base->lock, flags); @@ -405,13 +402,14 @@ static enum alarmtimer_restart alarm_handle_timer(struct alarm *alarm, ktime_t now) { struct k_itimer *ptr = container_of(alarm, struct k_itimer, - it.alarmtimer); + it.alarm.alarmtimer); if (posix_timer_event(ptr, 0) != 0) ptr->it_overrun++; /* Re-add periodic timers */ - if (alarm->period.tv64) { - ptr->it_overrun += alarm_forward(alarm, now, alarm->period); + if (ptr->it.alarm.interval.tv64) { + ptr->it_overrun += alarm_forward(alarm, now, + ptr->it.alarm.interval); return ALARMTIMER_RESTART; } return ALARMTIMER_NORESTART; @@ -471,7 +469,7 @@ static int alarm_timer_create(struct k_itimer *new_timer) type = clock2alarm(new_timer->it_clock); base = &alarm_bases[type]; - alarm_init(&new_timer->it.alarmtimer, type, alarm_handle_timer); + alarm_init(&new_timer->it.alarm.alarmtimer, type, alarm_handle_timer); return 0; } @@ -488,9 +486,9 @@ static void alarm_timer_get(struct k_itimer *timr, memset(cur_setting, 0, sizeof(struct itimerspec)); cur_setting->it_interval = - ktime_to_timespec(timr->it.alarmtimer.period); + ktime_to_timespec(timr->it.alarm.interval); cur_setting->it_value = - ktime_to_timespec(timr->it.alarmtimer.node.expires); + ktime_to_timespec(timr->it.alarm.alarmtimer.node.expires); return; } @@ -505,7 +503,7 @@ static int alarm_timer_del(struct k_itimer *timr) if (!rtcdev) return -ENOTSUPP; - alarm_cancel(&timr->it.alarmtimer); + alarm_cancel(&timr->it.alarm.alarmtimer); return 0; } @@ -529,12 +527,12 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, alarm_timer_get(timr, old_setting); /* If the timer was already set, cancel it */ - alarm_cancel(&timr->it.alarmtimer); + alarm_cancel(&timr->it.alarm.alarmtimer); /* start the timer */ - alarm_start(&timr->it.alarmtimer, - timespec_to_ktime(new_setting->it_value), - timespec_to_ktime(new_setting->it_interval)); + timr->it.alarm.interval = timespec_to_ktime(new_setting->it_interval); + alarm_start(&timr->it.alarm.alarmtimer, + timespec_to_ktime(new_setting->it_value)); return 0; } @@ -567,7 +565,7 @@ static int alarmtimer_do_nsleep(struct alarm *alarm, ktime_t absexp) alarm->data = (void *)current; do { set_current_state(TASK_INTERRUPTIBLE); - alarm_start(alarm, absexp, ktime_set(0, 0)); + alarm_start(alarm, absexp); if (likely(alarm->data)) schedule(); -- cgit v1.3-8-gc7d7 From a28cde81ab13cc251748a4c4ef06883dd09a10ea Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 12:30:21 -0700 Subject: alarmtimers: Add more refined alarm state tracking In order to allow for functionality like try_to_cancel, add more refined state tracking (similar to hrtimers). CC: Thomas Gleixner Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 34 +++++++++++++++++++++++++++++++++- kernel/time/alarmtimer.c | 21 ++++++++++++++------- 2 files changed, 47 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index c854a8efa863..304124a6c982 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -18,6 +18,11 @@ enum alarmtimer_restart { ALARMTIMER_RESTART, }; + +#define ALARMTIMER_STATE_INACTIVE 0x00 +#define ALARMTIMER_STATE_ENQUEUED 0x01 +#define ALARMTIMER_STATE_CALLBACK 0x02 + /** * struct alarm - Alarm timer structure * @node: timerqueue node for adding to the event list this value @@ -32,7 +37,7 @@ struct alarm { struct timerqueue_node node; enum alarmtimer_restart (*function)(struct alarm *, ktime_t now); enum alarmtimer_type type; - bool enabled; + int state; void *data; }; @@ -43,4 +48,31 @@ void alarm_cancel(struct alarm *alarm); u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval); +/* + * A alarmtimer is active, when it is enqueued into timerqueue or the + * callback function is running. + */ +static inline int alarmtimer_active(const struct alarm *timer) +{ + return timer->state != ALARMTIMER_STATE_INACTIVE; +} + +/* + * Helper function to check, whether the timer is on one of the queues + */ +static inline int alarmtimer_is_queued(struct alarm *timer) +{ + return timer->state & ALARMTIMER_STATE_ENQUEUED; +} + +/* + * Helper function to check, whether the timer is running the callback + * function + */ +static inline int alarmtimer_callback_running(struct alarm *timer) +{ + return timer->state & ALARMTIMER_STATE_CALLBACK; +} + + #endif diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 90935591dd44..5b14cc29b6a6 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -126,6 +126,8 @@ static struct rtc_device *alarmtimer_get_rtcdev(void) static void alarmtimer_enqueue(struct alarm_base *base, struct alarm *alarm) { timerqueue_add(&base->timerqueue, &alarm->node); + alarm->state |= ALARMTIMER_STATE_ENQUEUED; + if (&alarm->node == timerqueue_getnext(&base->timerqueue)) { hrtimer_try_to_cancel(&base->timer); hrtimer_start(&base->timer, alarm->node.expires, @@ -147,7 +149,12 @@ static void alarmtimer_remove(struct alarm_base *base, struct alarm *alarm) { struct timerqueue_node *next = timerqueue_getnext(&base->timerqueue); + if (!(alarm->state & ALARMTIMER_STATE_ENQUEUED)) + return; + timerqueue_del(&base->timerqueue, &alarm->node); + alarm->state &= ~ALARMTIMER_STATE_ENQUEUED; + if (next == &alarm->node) { hrtimer_try_to_cancel(&base->timer); next = timerqueue_getnext(&base->timerqueue); @@ -188,16 +195,18 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) alarm = container_of(next, struct alarm, node); timerqueue_del(&base->timerqueue, &alarm->node); - alarm->enabled = 0; + alarm->state &= ~ALARMTIMER_STATE_ENQUEUED; + alarm->state |= ALARMTIMER_STATE_CALLBACK; spin_unlock_irqrestore(&base->lock, flags); if (alarm->function) restart = alarm->function(alarm, now); spin_lock_irqsave(&base->lock, flags); + alarm->state &= ~ALARMTIMER_STATE_CALLBACK; if (restart != ALARMTIMER_NORESTART) { timerqueue_add(&base->timerqueue, &alarm->node); - alarm->enabled = 1; + alarm->state |= ALARMTIMER_STATE_ENQUEUED; } } @@ -305,7 +314,7 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, timerqueue_init(&alarm->node); alarm->function = function; alarm->type = type; - alarm->enabled = 0; + alarm->state = ALARMTIMER_STATE_INACTIVE; } /** @@ -319,11 +328,10 @@ void alarm_start(struct alarm *alarm, ktime_t start) unsigned long flags; spin_lock_irqsave(&base->lock, flags); - if (alarm->enabled) + if (alarmtimer_active(alarm)) alarmtimer_remove(base, alarm); alarm->node.expires = start; alarmtimer_enqueue(base, alarm); - alarm->enabled = 1; spin_unlock_irqrestore(&base->lock, flags); } @@ -337,9 +345,8 @@ void alarm_cancel(struct alarm *alarm) unsigned long flags; spin_lock_irqsave(&base->lock, flags); - if (alarm->enabled) + if (alarmtimer_is_queued(alarm)) alarmtimer_remove(base, alarm); - alarm->enabled = 0; spin_unlock_irqrestore(&base->lock, flags); } -- cgit v1.3-8-gc7d7 From 9082c465a5403f4a98734193e078552991a2e283 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 10 Aug 2011 12:41:36 -0700 Subject: alarmtimers: Add try_to_cancel functionality There's a number of edge cases when cancelling a alarm, so to be sure we accurately do so, introduce try_to_cancel, which returns proper failure errors if it cannot. Also modify cancel to spin until the alarm is properly disabled. CC: Thomas Gleixner Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 3 ++- kernel/time/alarmtimer.c | 43 +++++++++++++++++++++++++++++++++++++------ 2 files changed, 39 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index 304124a6c982..975009e1cbe6 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -44,7 +44,8 @@ struct alarm { void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)); void alarm_start(struct alarm *alarm, ktime_t start); -void alarm_cancel(struct alarm *alarm); +int alarm_try_to_cancel(struct alarm *alarm); +int alarm_cancel(struct alarm *alarm); u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval); diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 5b14cc29b6a6..bdb7342b6896 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -336,21 +336,49 @@ void alarm_start(struct alarm *alarm, ktime_t start) } /** - * alarm_cancel - Tries to cancel an alarm timer + * alarm_try_to_cancel - Tries to cancel an alarm timer * @alarm: ptr to alarm to be canceled + * + * Returns 1 if the timer was canceled, 0 if it was not running, + * and -1 if the callback was running */ -void alarm_cancel(struct alarm *alarm) +int alarm_try_to_cancel(struct alarm *alarm) { struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; - + int ret = -1; spin_lock_irqsave(&base->lock, flags); - if (alarmtimer_is_queued(alarm)) + + if (alarmtimer_callback_running(alarm)) + goto out; + + if (alarmtimer_is_queued(alarm)) { alarmtimer_remove(base, alarm); + ret = 1; + } else + ret = 0; +out: spin_unlock_irqrestore(&base->lock, flags); + return ret; } +/** + * alarm_cancel - Spins trying to cancel an alarm timer until it is done + * @alarm: ptr to alarm to be canceled + * + * Returns 1 if the timer was canceled, 0 if it was not active. + */ +int alarm_cancel(struct alarm *alarm) +{ + for (;;) { + int ret = alarm_try_to_cancel(alarm); + if (ret >= 0) + return ret; + cpu_relax(); + } +} + u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval) { @@ -510,7 +538,9 @@ static int alarm_timer_del(struct k_itimer *timr) if (!rtcdev) return -ENOTSUPP; - alarm_cancel(&timr->it.alarm.alarmtimer); + if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) + return TIMER_RETRY; + return 0; } @@ -534,7 +564,8 @@ static int alarm_timer_set(struct k_itimer *timr, int flags, alarm_timer_get(timr, old_setting); /* If the timer was already set, cancel it */ - alarm_cancel(&timr->it.alarm.alarmtimer); + if (alarm_try_to_cancel(&timr->it.alarm.alarmtimer) < 0) + return TIMER_RETRY; /* start the timer */ timr->it.alarm.interval = timespec_to_ktime(new_setting->it_interval); -- cgit v1.3-8-gc7d7 From 8bc0dafb5cf38a19484dfb16e2c6d29e85820046 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 14 Jul 2011 18:35:13 -0700 Subject: alarmtimers: Rework RTC device selection using class interface This allows cleaner detection of the RTC device being registered, rather then probing any time someone calls alarmtimer_get_rtcdev. CC: Thomas Gleixner Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 78 +++++++++++++++++++++++++----------------------- 1 file changed, 40 insertions(+), 38 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index bdb7342b6896..154d5563ab1b 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -52,27 +52,6 @@ static struct rtc_timer rtctimer; static struct rtc_device *rtcdev; static DEFINE_SPINLOCK(rtcdev_lock); -/** - * has_wakealarm - check rtc device has wakealarm ability - * @dev: current device - * @name_ptr: name to be returned - * - * This helper function checks to see if the rtc device can wake - * from suspend. - */ -static int has_wakealarm(struct device *dev, void *name_ptr) -{ - struct rtc_device *candidate = to_rtc_device(dev); - - if (!candidate->ops->set_alarm) - return 0; - if (!device_may_wakeup(candidate->dev.parent)) - return 0; - - *(const char **)name_ptr = dev_name(dev); - return 1; -} - /** * alarmtimer_get_rtcdev - Return selected rtcdevice * @@ -82,37 +61,58 @@ static int has_wakealarm(struct device *dev, void *name_ptr) */ static struct rtc_device *alarmtimer_get_rtcdev(void) { - struct device *dev; - char *str; unsigned long flags; struct rtc_device *ret; spin_lock_irqsave(&rtcdev_lock, flags); - if (!rtcdev) { - /* Find an rtc device and init the rtc_timer */ - dev = class_find_device(rtc_class, NULL, &str, has_wakealarm); - /* If we have a device then str is valid. See has_wakealarm() */ - if (dev) { - rtcdev = rtc_class_open(str); - /* - * Drop the reference we got in class_find_device, - * rtc_open takes its own. - */ - put_device(dev); - rtc_timer_init(&rtctimer, NULL, NULL); - } - } ret = rtcdev; spin_unlock_irqrestore(&rtcdev_lock, flags); return ret; } + + +static int alarmtimer_rtc_add_device(struct device *dev, + struct class_interface *class_intf) +{ + unsigned long flags; + struct rtc_device *rtc = to_rtc_device(dev); + + if (rtcdev) + return -EBUSY; + + if (!rtc->ops->set_alarm) + return -1; + if (!device_may_wakeup(rtc->dev.parent)) + return -1; + + spin_lock_irqsave(&rtcdev_lock, flags); + if (!rtcdev) { + rtcdev = rtc; + /* hold a reference so it doesn't go away */ + get_device(dev); + } + spin_unlock_irqrestore(&rtcdev_lock, flags); + return 0; +} + +static struct class_interface alarmtimer_rtc_interface = { + .add_dev = &alarmtimer_rtc_add_device, +}; + +static void alarmtimer_rtc_interface_setup(void) +{ + alarmtimer_rtc_interface.class = rtc_class; + class_interface_register(&alarmtimer_rtc_interface); +} #else #define alarmtimer_get_rtcdev() (0) #define rtcdev (0) +#define alarmtimer_rtc_interface_setup() #endif + /** * alarmtimer_enqueue - Adds an alarm timer to an alarm_base timerqueue * @base: pointer to the base where the timer is being run @@ -244,7 +244,7 @@ static int alarmtimer_suspend(struct device *dev) freezer_delta = ktime_set(0, 0); spin_unlock_irqrestore(&freezer_delta_lock, flags); - rtc = rtcdev; + rtc = alarmtimer_get_rtcdev(); /* If we have no rtcdev, just return */ if (!rtc) return 0; @@ -792,6 +792,8 @@ static int __init alarmtimer_init(void) HRTIMER_MODE_ABS); alarm_bases[i].timer.function = alarmtimer_fired; } + + alarmtimer_rtc_interface_setup(); error = platform_driver_register(&alarmtimer_driver); platform_device_register_simple("alarmtimer", -1, NULL, 0); -- cgit v1.3-8-gc7d7 From 3a301d7c1c68fd96bbcf82db82d9508918e0dc22 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 8 Aug 2011 21:39:39 -0400 Subject: tracing: Clean up tb_fmt to not give faulty compile warning gcc incorrectly states that the variable "fmt" is uninitialized when CC_OPITMIZE_FOR_SIZE is set. Instead of just blindly setting fmt to NULL, the code is cleaned up a little to be a bit easier for humans to follow, as well as gcc to know the variables are initialized. Signed-off-by: Steven Rostedt --- kernel/trace/trace_printk.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_printk.c b/kernel/trace/trace_printk.c index 1f06468a10d7..6fd4ffd042f9 100644 --- a/kernel/trace/trace_printk.c +++ b/kernel/trace/trace_printk.c @@ -59,18 +59,19 @@ void hold_module_trace_bprintk_format(const char **start, const char **end) continue; } + fmt = NULL; tb_fmt = kmalloc(sizeof(*tb_fmt), GFP_KERNEL); - if (tb_fmt) + if (tb_fmt) { fmt = kmalloc(strlen(*iter) + 1, GFP_KERNEL); - if (tb_fmt && fmt) { - list_add_tail(&tb_fmt->list, &trace_bprintk_fmt_list); - strcpy(fmt, *iter); - tb_fmt->fmt = fmt; - *iter = tb_fmt->fmt; - } else { - kfree(tb_fmt); - *iter = NULL; + if (fmt) { + list_add_tail(&tb_fmt->list, &trace_bprintk_fmt_list); + strcpy(fmt, *iter); + tb_fmt->fmt = fmt; + } else + kfree(tb_fmt); } + *iter = fmt; + } mutex_unlock(&btrace_mutex); } -- cgit v1.3-8-gc7d7 From b75ef8b44b1cb95f5a26484b0e2fe37a63b12b44 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Wed, 10 Aug 2011 15:18:39 -0400 Subject: Tracepoint: Dissociate from module mutex Copy the information needed from struct module into a local module list held within tracepoint.c from within the module coming/going notifier. This vastly simplifies locking of tracepoint registration / unregistration, because we don't have to take the module mutex to register and unregister tracepoints anymore. Steven Rostedt ran into dependency problems related to modules mutex vs kprobes mutex vs ftrace mutex vs tracepoint mutex that seems to be hard to fix without removing this dependency between tracepoint and module mutex. (note: it should be investigated whether kprobes could benefit of being dissociated from the modules mutex too.) This also fixes module handling of tracepoint list iterators, because it was expecting the list to be sorted by pointer address. Given we have control on our own list now, it's OK to sort this list which has tracepoints as its only purpose. The reason why this sorting is required is to handle the fact that seq files (and any read() operation from user-space) cannot hold the tracepoint mutex across multiple calls, so list entries may vanish between calls. With sorting, the tracepoint iterator becomes usable even if the list don't contain the exact item pointed to by the iterator anymore. Signed-off-by: Mathieu Desnoyers Acked-by: Jason Baron CC: Ingo Molnar CC: Lai Jiangshan CC: Peter Zijlstra CC: Thomas Gleixner CC: Masami Hiramatsu Link: http://lkml.kernel.org/r/20110810191839.GC8525@Krystal Signed-off-by: Steven Rostedt --- include/linux/module.h | 12 ---- include/linux/tracepoint.h | 25 +++---- kernel/module.c | 47 ------------- kernel/tracepoint.c | 169 +++++++++++++++++++++++++++++++++++++++------ 4 files changed, 158 insertions(+), 95 deletions(-) (limited to 'kernel') diff --git a/include/linux/module.h b/include/linux/module.h index 1c30087a2d81..863921637d9f 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -580,9 +580,6 @@ int unregister_module_notifier(struct notifier_block * nb); extern void print_modules(void); -extern void module_update_tracepoints(void); -extern int module_get_iter_tracepoints(struct tracepoint_iter *iter); - #else /* !CONFIG_MODULES... */ #define EXPORT_SYMBOL(sym) #define EXPORT_SYMBOL_GPL(sym) @@ -698,15 +695,6 @@ static inline int unregister_module_notifier(struct notifier_block * nb) static inline void print_modules(void) { } - -static inline void module_update_tracepoints(void) -{ -} - -static inline int module_get_iter_tracepoints(struct tracepoint_iter *iter) -{ - return 0; -} #endif /* CONFIG_MODULES */ #ifdef CONFIG_SYSFS diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index d530a4460a0b..df0a779c1bbd 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -54,8 +54,18 @@ extern int tracepoint_probe_unregister_noupdate(const char *name, void *probe, void *data); extern void tracepoint_probe_update_all(void); +#ifdef CONFIG_MODULES +struct tp_module { + struct list_head list; + unsigned int num_tracepoints; + struct tracepoint * const *tracepoints_ptrs; +}; +#endif /* CONFIG_MODULES */ + struct tracepoint_iter { - struct module *module; +#ifdef CONFIG_MODULES + struct tp_module *module; +#endif /* CONFIG_MODULES */ struct tracepoint * const *tracepoint; }; @@ -63,8 +73,6 @@ extern void tracepoint_iter_start(struct tracepoint_iter *iter); extern void tracepoint_iter_next(struct tracepoint_iter *iter); extern void tracepoint_iter_stop(struct tracepoint_iter *iter); extern void tracepoint_iter_reset(struct tracepoint_iter *iter); -extern int tracepoint_get_iter_range(struct tracepoint * const **tracepoint, - struct tracepoint * const *begin, struct tracepoint * const *end); /* * tracepoint_synchronize_unregister must be called between the last tracepoint @@ -78,17 +86,6 @@ static inline void tracepoint_synchronize_unregister(void) #define PARAMS(args...) args -#ifdef CONFIG_TRACEPOINTS -extern -void tracepoint_update_probe_range(struct tracepoint * const *begin, - struct tracepoint * const *end); -#else -static inline -void tracepoint_update_probe_range(struct tracepoint * const *begin, - struct tracepoint * const *end) -{ } -#endif /* CONFIG_TRACEPOINTS */ - #endif /* _LINUX_TRACEPOINT_H */ /* diff --git a/kernel/module.c b/kernel/module.c index 04379f92f843..93342d992f34 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3487,50 +3487,3 @@ void module_layout(struct module *mod, } EXPORT_SYMBOL(module_layout); #endif - -#ifdef CONFIG_TRACEPOINTS -void module_update_tracepoints(void) -{ - struct module *mod; - - mutex_lock(&module_mutex); - list_for_each_entry(mod, &modules, list) - if (!mod->taints) - tracepoint_update_probe_range(mod->tracepoints_ptrs, - mod->tracepoints_ptrs + mod->num_tracepoints); - mutex_unlock(&module_mutex); -} - -/* - * Returns 0 if current not found. - * Returns 1 if current found. - */ -int module_get_iter_tracepoints(struct tracepoint_iter *iter) -{ - struct module *iter_mod; - int found = 0; - - mutex_lock(&module_mutex); - list_for_each_entry(iter_mod, &modules, list) { - if (!iter_mod->taints) { - /* - * Sorted module list - */ - if (iter_mod < iter->module) - continue; - else if (iter_mod > iter->module) - iter->tracepoint = NULL; - found = tracepoint_get_iter_range(&iter->tracepoint, - iter_mod->tracepoints_ptrs, - iter_mod->tracepoints_ptrs - + iter_mod->num_tracepoints); - if (found) { - iter->module = iter_mod; - break; - } - } - } - mutex_unlock(&module_mutex); - return found; -} -#endif diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index b219f1449c54..db110b8ae030 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -34,11 +34,16 @@ extern struct tracepoint * const __stop___tracepoints_ptrs[]; static const int tracepoint_debug; /* - * tracepoints_mutex nests inside module_mutex. Tracepoints mutex protects the - * builtin and module tracepoints and the hash table. + * Tracepoints mutex protects the builtin and module tracepoints and the hash + * table, as well as the local module list. */ static DEFINE_MUTEX(tracepoints_mutex); +#ifdef CONFIG_MODULES +/* Local list of struct module */ +static LIST_HEAD(tracepoint_module_list); +#endif /* CONFIG_MODULES */ + /* * Tracepoint hash table, containing the active tracepoints. * Protected by tracepoints_mutex. @@ -292,9 +297,10 @@ static void disable_tracepoint(struct tracepoint *elem) * @end: end of the range * * Updates the probe callback corresponding to a range of tracepoints. + * Called with tracepoints_mutex held. */ -void tracepoint_update_probe_range(struct tracepoint * const *begin, - struct tracepoint * const *end) +static void tracepoint_update_probe_range(struct tracepoint * const *begin, + struct tracepoint * const *end) { struct tracepoint * const *iter; struct tracepoint_entry *mark_entry; @@ -302,7 +308,6 @@ void tracepoint_update_probe_range(struct tracepoint * const *begin, if (!begin) return; - mutex_lock(&tracepoints_mutex); for (iter = begin; iter < end; iter++) { mark_entry = get_tracepoint((*iter)->name); if (mark_entry) { @@ -312,11 +317,27 @@ void tracepoint_update_probe_range(struct tracepoint * const *begin, disable_tracepoint(*iter); } } - mutex_unlock(&tracepoints_mutex); } +#ifdef CONFIG_MODULES +void module_update_tracepoints(void) +{ + struct tp_module *tp_mod; + + list_for_each_entry(tp_mod, &tracepoint_module_list, list) + tracepoint_update_probe_range(tp_mod->tracepoints_ptrs, + tp_mod->tracepoints_ptrs + tp_mod->num_tracepoints); +} +#else /* CONFIG_MODULES */ +void module_update_tracepoints(void) +{ +} +#endif /* CONFIG_MODULES */ + + /* * Update probes, removing the faulty probes. + * Called with tracepoints_mutex held. */ static void tracepoint_update_probes(void) { @@ -359,11 +380,12 @@ int tracepoint_probe_register(const char *name, void *probe, void *data) mutex_lock(&tracepoints_mutex); old = tracepoint_add_probe(name, probe, data); - mutex_unlock(&tracepoints_mutex); - if (IS_ERR(old)) + if (IS_ERR(old)) { + mutex_unlock(&tracepoints_mutex); return PTR_ERR(old); - + } tracepoint_update_probes(); /* may update entry */ + mutex_unlock(&tracepoints_mutex); release_probes(old); return 0; } @@ -402,11 +424,12 @@ int tracepoint_probe_unregister(const char *name, void *probe, void *data) mutex_lock(&tracepoints_mutex); old = tracepoint_remove_probe(name, probe, data); - mutex_unlock(&tracepoints_mutex); - if (IS_ERR(old)) + if (IS_ERR(old)) { + mutex_unlock(&tracepoints_mutex); return PTR_ERR(old); - + } tracepoint_update_probes(); /* may update entry */ + mutex_unlock(&tracepoints_mutex); release_probes(old); return 0; } @@ -489,9 +512,8 @@ void tracepoint_probe_update_all(void) if (!list_empty(&old_probes)) list_replace_init(&old_probes, &release_probes); need_update = 0; - mutex_unlock(&tracepoints_mutex); - tracepoint_update_probes(); + mutex_unlock(&tracepoints_mutex); list_for_each_entry_safe(pos, next, &release_probes, u.list) { list_del(&pos->u.list); call_rcu_sched(&pos->u.rcu, rcu_free_old_probes); @@ -509,7 +531,7 @@ EXPORT_SYMBOL_GPL(tracepoint_probe_update_all); * Will return the first tracepoint in the range if the input tracepoint is * NULL. */ -int tracepoint_get_iter_range(struct tracepoint * const **tracepoint, +static int tracepoint_get_iter_range(struct tracepoint * const **tracepoint, struct tracepoint * const *begin, struct tracepoint * const *end) { if (!*tracepoint && begin != end) { @@ -520,11 +542,12 @@ int tracepoint_get_iter_range(struct tracepoint * const **tracepoint, return 1; return 0; } -EXPORT_SYMBOL_GPL(tracepoint_get_iter_range); +#ifdef CONFIG_MODULES static void tracepoint_get_iter(struct tracepoint_iter *iter) { int found = 0; + struct tp_module *iter_mod; /* Core kernel tracepoints */ if (!iter->module) { @@ -534,12 +557,43 @@ static void tracepoint_get_iter(struct tracepoint_iter *iter) if (found) goto end; } - /* tracepoints in modules. */ - found = module_get_iter_tracepoints(iter); + /* Tracepoints in modules */ + mutex_lock(&tracepoints_mutex); + list_for_each_entry(iter_mod, &tracepoint_module_list, list) { + /* + * Sorted module list + */ + if (iter_mod < iter->module) + continue; + else if (iter_mod > iter->module) + iter->tracepoint = NULL; + found = tracepoint_get_iter_range(&iter->tracepoint, + iter_mod->tracepoints_ptrs, + iter_mod->tracepoints_ptrs + + iter_mod->num_tracepoints); + if (found) { + iter->module = iter_mod; + break; + } + } + mutex_unlock(&tracepoints_mutex); end: if (!found) tracepoint_iter_reset(iter); } +#else /* CONFIG_MODULES */ +static void tracepoint_get_iter(struct tracepoint_iter *iter) +{ + int found = 0; + + /* Core kernel tracepoints */ + found = tracepoint_get_iter_range(&iter->tracepoint, + __start___tracepoints_ptrs, + __stop___tracepoints_ptrs); + if (!found) + tracepoint_iter_reset(iter); +} +#endif /* CONFIG_MODULES */ void tracepoint_iter_start(struct tracepoint_iter *iter) { @@ -566,26 +620,98 @@ EXPORT_SYMBOL_GPL(tracepoint_iter_stop); void tracepoint_iter_reset(struct tracepoint_iter *iter) { +#ifdef CONFIG_MODULES iter->module = NULL; +#endif /* CONFIG_MODULES */ iter->tracepoint = NULL; } EXPORT_SYMBOL_GPL(tracepoint_iter_reset); #ifdef CONFIG_MODULES +static int tracepoint_module_coming(struct module *mod) +{ + struct tp_module *tp_mod, *iter; + int ret = 0; + + /* + * We skip modules that tain the kernel, especially those with different + * module header (for forced load), to make sure we don't cause a crash. + */ + if (mod->taints) + return 0; + mutex_lock(&tracepoints_mutex); + tp_mod = kmalloc(sizeof(struct tp_module), GFP_KERNEL); + if (!tp_mod) { + ret = -ENOMEM; + goto end; + } + tp_mod->num_tracepoints = mod->num_tracepoints; + tp_mod->tracepoints_ptrs = mod->tracepoints_ptrs; + + /* + * tracepoint_module_list is kept sorted by struct module pointer + * address for iteration on tracepoints from a seq_file that can release + * the mutex between calls. + */ + list_for_each_entry_reverse(iter, &tracepoint_module_list, list) { + BUG_ON(iter == tp_mod); /* Should never be in the list twice */ + if (iter < tp_mod) { + /* We belong to the location right after iter. */ + list_add(&tp_mod->list, &iter->list); + goto module_added; + } + } + /* We belong to the beginning of the list */ + list_add(&tp_mod->list, &tracepoint_module_list); +module_added: + tracepoint_update_probe_range(mod->tracepoints_ptrs, + mod->tracepoints_ptrs + mod->num_tracepoints); +end: + mutex_unlock(&tracepoints_mutex); + return ret; +} + +static int tracepoint_module_going(struct module *mod) +{ + struct tp_module *pos; + + mutex_lock(&tracepoints_mutex); + tracepoint_update_probe_range(mod->tracepoints_ptrs, + mod->tracepoints_ptrs + mod->num_tracepoints); + list_for_each_entry(pos, &tracepoint_module_list, list) { + if (pos->tracepoints_ptrs == mod->tracepoints_ptrs) { + list_del(&pos->list); + kfree(pos); + break; + } + } + /* + * In the case of modules that were tainted at "coming", we'll simply + * walk through the list without finding it. We cannot use the "tainted" + * flag on "going", in case a module taints the kernel only after being + * loaded. + */ + mutex_unlock(&tracepoints_mutex); + return 0; +} int tracepoint_module_notify(struct notifier_block *self, unsigned long val, void *data) { struct module *mod = data; + int ret = 0; switch (val) { case MODULE_STATE_COMING: + ret = tracepoint_module_coming(mod); + break; + case MODULE_STATE_LIVE: + break; case MODULE_STATE_GOING: - tracepoint_update_probe_range(mod->tracepoints_ptrs, - mod->tracepoints_ptrs + mod->num_tracepoints); + ret = tracepoint_module_going(mod); break; } - return 0; + return ret; } struct notifier_block tracepoint_module_nb = { @@ -598,7 +724,6 @@ static int init_tracepoints(void) return register_module_notifier(&tracepoint_module_nb); } __initcall(init_tracepoints); - #endif /* CONFIG_MODULES */ #ifdef CONFIG_HAVE_SYSCALL_TRACEPOINTS -- cgit v1.3-8-gc7d7 From c09c47caedc9854d59378d6e34c989e51cfdd2b4 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Thu, 11 Aug 2011 10:36:05 +0200 Subject: blktrace: add FLUSH/FUA support Add FLUSH/FUA support to blktrace. As FLUSH precedes WRITE and/or FUA follows WRITE, use the same 'F' flag for both cases and distinguish them by their (relative) position. The end results look like (other flags might be shown also): - WRITE: W - WRITE_FLUSH: FW - WRITE_FUA: WF - WRITE_FLUSH_FUA: FWF Note that we reuse TC_BARRIER due to lack of bit space of act_mask so that the older versions of blktrace tools will report flush requests as barriers from now on. Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Signed-off-by: Namhyung Kim Reviewed-by: Jeff Moyer Signed-off-by: Jens Axboe --- include/linux/blktrace_api.h | 5 +++-- include/trace/events/block.h | 20 +++++++++++--------- kernel/trace/blktrace.c | 21 ++++++++++++++++----- 3 files changed, 30 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h index 8c7c2de7631a..8e9e4bc6d73b 100644 --- a/include/linux/blktrace_api.h +++ b/include/linux/blktrace_api.h @@ -14,7 +14,7 @@ enum blktrace_cat { BLK_TC_READ = 1 << 0, /* reads */ BLK_TC_WRITE = 1 << 1, /* writes */ - BLK_TC_BARRIER = 1 << 2, /* barrier */ + BLK_TC_FLUSH = 1 << 2, /* flush */ BLK_TC_SYNC = 1 << 3, /* sync IO */ BLK_TC_SYNCIO = BLK_TC_SYNC, BLK_TC_QUEUE = 1 << 4, /* queueing/merging */ @@ -28,8 +28,9 @@ enum blktrace_cat { BLK_TC_META = 1 << 12, /* metadata */ BLK_TC_DISCARD = 1 << 13, /* discard requests */ BLK_TC_DRV_DATA = 1 << 14, /* binary per-driver data */ + BLK_TC_FUA = 1 << 15, /* fua requests */ - BLK_TC_END = 1 << 15, /* only 16-bits, reminder */ + BLK_TC_END = 1 << 15, /* we've run out of bits! */ }; #define BLK_TC_SHIFT (16) diff --git a/include/trace/events/block.h b/include/trace/events/block.h index bf366547da25..05c5e61f0a7c 100644 --- a/include/trace/events/block.h +++ b/include/trace/events/block.h @@ -8,6 +8,8 @@ #include #include +#define RWBS_LEN 8 + DECLARE_EVENT_CLASS(block_rq_with_error, TP_PROTO(struct request_queue *q, struct request *rq), @@ -19,7 +21,7 @@ DECLARE_EVENT_CLASS(block_rq_with_error, __field( sector_t, sector ) __field( unsigned int, nr_sector ) __field( int, errors ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __dynamic_array( char, cmd, blk_cmd_buf_len(rq) ) ), @@ -104,7 +106,7 @@ DECLARE_EVENT_CLASS(block_rq, __field( sector_t, sector ) __field( unsigned int, nr_sector ) __field( unsigned int, bytes ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __array( char, comm, TASK_COMM_LEN ) __dynamic_array( char, cmd, blk_cmd_buf_len(rq) ) ), @@ -183,7 +185,7 @@ TRACE_EVENT(block_bio_bounce, __field( dev_t, dev ) __field( sector_t, sector ) __field( unsigned int, nr_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __array( char, comm, TASK_COMM_LEN ) ), @@ -222,7 +224,7 @@ TRACE_EVENT(block_bio_complete, __field( sector_t, sector ) __field( unsigned, nr_sector ) __field( int, error ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN) ), TP_fast_assign( @@ -249,7 +251,7 @@ DECLARE_EVENT_CLASS(block_bio, __field( dev_t, dev ) __field( sector_t, sector ) __field( unsigned int, nr_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __array( char, comm, TASK_COMM_LEN ) ), @@ -321,7 +323,7 @@ DECLARE_EVENT_CLASS(block_get_rq, __field( dev_t, dev ) __field( sector_t, sector ) __field( unsigned int, nr_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __array( char, comm, TASK_COMM_LEN ) ), @@ -456,7 +458,7 @@ TRACE_EVENT(block_split, __field( dev_t, dev ) __field( sector_t, sector ) __field( sector_t, new_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN ) __array( char, comm, TASK_COMM_LEN ) ), @@ -498,7 +500,7 @@ TRACE_EVENT(block_bio_remap, __field( unsigned int, nr_sector ) __field( dev_t, old_dev ) __field( sector_t, old_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN) ), TP_fast_assign( @@ -542,7 +544,7 @@ TRACE_EVENT(block_rq_remap, __field( unsigned int, nr_sector ) __field( dev_t, old_dev ) __field( sector_t, old_sector ) - __array( char, rwbs, 6 ) + __array( char, rwbs, RWBS_LEN) ), TP_fast_assign( diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 6957aa298dfa..7c910a5593a6 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -206,6 +206,8 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, what |= MASK_TC_BIT(rw, RAHEAD); what |= MASK_TC_BIT(rw, META); what |= MASK_TC_BIT(rw, DISCARD); + what |= MASK_TC_BIT(rw, FLUSH); + what |= MASK_TC_BIT(rw, FUA); pid = tsk->pid; if (act_log_check(bt, what, sector, pid)) @@ -1054,6 +1056,9 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) goto out; } + if (tc & BLK_TC_FLUSH) + rwbs[i++] = 'F'; + if (tc & BLK_TC_DISCARD) rwbs[i++] = 'D'; else if (tc & BLK_TC_WRITE) @@ -1063,10 +1068,10 @@ static void fill_rwbs(char *rwbs, const struct blk_io_trace *t) else rwbs[i++] = 'N'; + if (tc & BLK_TC_FUA) + rwbs[i++] = 'F'; if (tc & BLK_TC_AHEAD) rwbs[i++] = 'A'; - if (tc & BLK_TC_BARRIER) - rwbs[i++] = 'B'; if (tc & BLK_TC_SYNC) rwbs[i++] = 'S'; if (tc & BLK_TC_META) @@ -1132,7 +1137,7 @@ typedef int (blk_log_action_t) (struct trace_iterator *iter, const char *act); static int blk_log_action_classic(struct trace_iterator *iter, const char *act) { - char rwbs[6]; + char rwbs[RWBS_LEN]; unsigned long long ts = iter->ts; unsigned long nsec_rem = do_div(ts, NSEC_PER_SEC); unsigned secs = (unsigned long)ts; @@ -1148,7 +1153,7 @@ static int blk_log_action_classic(struct trace_iterator *iter, const char *act) static int blk_log_action(struct trace_iterator *iter, const char *act) { - char rwbs[6]; + char rwbs[RWBS_LEN]; const struct blk_io_trace *t = te_blk_io_trace(iter->ent); fill_rwbs(rwbs, t); @@ -1561,7 +1566,7 @@ static const struct { } mask_maps[] = { { BLK_TC_READ, "read" }, { BLK_TC_WRITE, "write" }, - { BLK_TC_BARRIER, "barrier" }, + { BLK_TC_FLUSH, "flush" }, { BLK_TC_SYNC, "sync" }, { BLK_TC_QUEUE, "queue" }, { BLK_TC_REQUEUE, "requeue" }, @@ -1573,6 +1578,7 @@ static const struct { { BLK_TC_META, "meta" }, { BLK_TC_DISCARD, "discard" }, { BLK_TC_DRV_DATA, "drv_data" }, + { BLK_TC_FUA, "fua" }, }; static int blk_trace_str2mask(const char *str) @@ -1788,6 +1794,9 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes) { int i = 0; + if (rw & REQ_FLUSH) + rwbs[i++] = 'F'; + if (rw & WRITE) rwbs[i++] = 'W'; else if (rw & REQ_DISCARD) @@ -1797,6 +1806,8 @@ void blk_fill_rwbs(char *rwbs, u32 rw, int bytes) else rwbs[i++] = 'N'; + if (rw & REQ_FUA) + rwbs[i++] = 'F'; if (rw & REQ_RAHEAD) rwbs[i++] = 'A'; if (rw & REQ_SYNC) -- cgit v1.3-8-gc7d7 From 72fa59970f8698023045ab0713d66f3f4f96945c Mon Sep 17 00:00:00 2001 From: Vasiliy Kulikov Date: Mon, 8 Aug 2011 19:02:04 +0400 Subject: move RLIMIT_NPROC check from set_user() to do_execve_common() The patch http://lkml.org/lkml/2003/7/13/226 introduced an RLIMIT_NPROC check in set_user() to check for NPROC exceeding via setuid() and similar functions. Before the check there was a possibility to greatly exceed the allowed number of processes by an unprivileged user if the program relied on rlimit only. But the check created new security threat: many poorly written programs simply don't check setuid() return code and believe it cannot fail if executed with root privileges. So, the check is removed in this patch because of too often privilege escalations related to buggy programs. The NPROC can still be enforced in the common code flow of daemons spawning user processes. Most of daemons do fork()+setuid()+execve(). The check introduced in execve() (1) enforces the same limit as in setuid() and (2) doesn't create similar security issues. Neil Brown suggested to track what specific process has exceeded the limit by setting PF_NPROC_EXCEEDED process flag. With the change only this process would fail on execve(), and other processes' execve() behaviour is not changed. Solar Designer suggested to re-check whether NPROC limit is still exceeded at the moment of execve(). If the process was sleeping for days between set*uid() and execve(), and the NPROC counter step down under the limit, the defered execve() failure because NPROC limit was exceeded days ago would be unexpected. If the limit is not exceeded anymore, we clear the flag on successful calls to execve() and fork(). The flag is also cleared on successful calls to set_user() as the limit was exceeded for the previous user, not the current one. Similar check was introduced in -ow patches (without the process flag). v3 - clear PF_NPROC_EXCEEDED on successful calls to set_user(). Reviewed-by: James Morris Signed-off-by: Vasiliy Kulikov Acked-by: NeilBrown Signed-off-by: Linus Torvalds --- fs/exec.c | 17 +++++++++++++++++ include/linux/sched.h | 1 + kernel/cred.c | 6 ++---- kernel/fork.c | 1 + kernel/sys.c | 15 +++++++++++---- 5 files changed, 32 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/fs/exec.c b/fs/exec.c index da80612a35f4..25dcbe5fc356 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1459,6 +1459,23 @@ static int do_execve_common(const char *filename, struct files_struct *displaced; bool clear_in_exec; int retval; + const struct cred *cred = current_cred(); + + /* + * We move the actual failure in case of RLIMIT_NPROC excess from + * set*uid() to execve() because too many poorly written programs + * don't check setuid() return code. Here we additionally recheck + * whether NPROC limit is still exceeded. + */ + if ((current->flags & PF_NPROC_EXCEEDED) && + atomic_read(&cred->user->processes) > rlimit(RLIMIT_NPROC)) { + retval = -EAGAIN; + goto out_ret; + } + + /* We're below the limit (still or again), so we don't want to make + * further execve() calls fail. */ + current->flags &= ~PF_NPROC_EXCEEDED; retval = unshare_files(&displaced); if (retval) diff --git a/include/linux/sched.h b/include/linux/sched.h index 20b03bf94748..4ac2c0578e0f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1767,6 +1767,7 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t * #define PF_DUMPCORE 0x00000200 /* dumped core */ #define PF_SIGNALED 0x00000400 /* killed by a signal */ #define PF_MEMALLOC 0x00000800 /* Allocating memory */ +#define PF_NPROC_EXCEEDED 0x00001000 /* set_user noticed that RLIMIT_NPROC was exceeded */ #define PF_USED_MATH 0x00002000 /* if unset the fpu must be initialized before use */ #define PF_FREEZING 0x00004000 /* freeze in progress. do not account to load */ #define PF_NOFREEZE 0x00008000 /* this thread should not be frozen */ diff --git a/kernel/cred.c b/kernel/cred.c index 174fa84eca30..8ef31f53c44c 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -508,10 +508,8 @@ int commit_creds(struct cred *new) key_fsgid_changed(task); /* do it - * - What if a process setreuid()'s and this brings the - * new uid over his NPROC rlimit? We can check this now - * cheaply with the new uid cache, so if it matters - * we should be checking for it. -DaveM + * RLIMIT_NPROC limits on user->processes have already been checked + * in set_user(). */ alter_cred_subscribers(new, 2); if (new->user != old->user) diff --git a/kernel/fork.c b/kernel/fork.c index e7ceaca89609..8e6b6f4fb272 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1111,6 +1111,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->real_cred->user != INIT_USER) goto bad_fork_free; } + current->flags &= ~PF_NPROC_EXCEEDED; retval = copy_creds(p, clone_flags); if (retval < 0) diff --git a/kernel/sys.c b/kernel/sys.c index a101ba36c444..dd948a1fca4c 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -621,11 +621,18 @@ static int set_user(struct cred *new) if (!new_user) return -EAGAIN; + /* + * We don't fail in case of NPROC limit excess here because too many + * poorly written programs don't check set*uid() return code, assuming + * it never fails if called by root. We may still enforce NPROC limit + * for programs doing set*uid()+execve() by harmlessly deferring the + * failure to the execve() stage. + */ if (atomic_read(&new_user->processes) >= rlimit(RLIMIT_NPROC) && - new_user != INIT_USER) { - free_uid(new_user); - return -EAGAIN; - } + new_user != INIT_USER) + current->flags |= PF_NPROC_EXCEEDED; + else + current->flags &= ~PF_NPROC_EXCEEDED; free_uid(new->user); new->user = new_user; -- cgit v1.3-8-gc7d7 From c59d87c460767bc35dafd490139d3cfe78fb8da4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Fri, 12 Aug 2011 16:21:35 -0500 Subject: xfs: remove subdirectories Use the move from Linux 2.6 to Linux 3.x as an excuse to kill the annoying subdirectories in the XFS source code. Besides the large amount of file rename the only changes are to the Makefile, a few files including headers with the subdirectory prefix, and the binary sysctl compat code that includes a header under fs/xfs/ from kernel/. Signed-off-by: Christoph Hellwig Signed-off-by: Alex Elder --- fs/xfs/Makefile | 117 +- fs/xfs/kmem.c | 132 +++ fs/xfs/kmem.h | 124 ++ fs/xfs/linux-2.6/kmem.c | 132 --- fs/xfs/linux-2.6/kmem.h | 124 -- fs/xfs/linux-2.6/mrlock.h | 90 -- fs/xfs/linux-2.6/time.h | 36 - fs/xfs/linux-2.6/xfs_acl.c | 420 ------- fs/xfs/linux-2.6/xfs_aops.c | 1499 ------------------------ fs/xfs/linux-2.6/xfs_aops.h | 68 -- fs/xfs/linux-2.6/xfs_buf.c | 1876 ------------------------------ fs/xfs/linux-2.6/xfs_buf.h | 326 ------ fs/xfs/linux-2.6/xfs_discard.c | 222 ---- fs/xfs/linux-2.6/xfs_discard.h | 10 - fs/xfs/linux-2.6/xfs_export.c | 250 ---- fs/xfs/linux-2.6/xfs_export.h | 72 -- fs/xfs/linux-2.6/xfs_file.c | 1096 ------------------ fs/xfs/linux-2.6/xfs_fs_subr.c | 96 -- fs/xfs/linux-2.6/xfs_globals.c | 43 - fs/xfs/linux-2.6/xfs_ioctl.c | 1556 ------------------------- fs/xfs/linux-2.6/xfs_ioctl.h | 85 -- fs/xfs/linux-2.6/xfs_ioctl32.c | 672 ----------- fs/xfs/linux-2.6/xfs_ioctl32.h | 237 ---- fs/xfs/linux-2.6/xfs_iops.c | 1210 -------------------- fs/xfs/linux-2.6/xfs_iops.h | 30 - fs/xfs/linux-2.6/xfs_linux.h | 309 ----- fs/xfs/linux-2.6/xfs_message.c | 108 -- fs/xfs/linux-2.6/xfs_message.h | 39 - fs/xfs/linux-2.6/xfs_quotaops.c | 139 --- fs/xfs/linux-2.6/xfs_stats.c | 122 -- fs/xfs/linux-2.6/xfs_stats.h | 223 ---- fs/xfs/linux-2.6/xfs_super.c | 1773 ---------------------------- fs/xfs/linux-2.6/xfs_super.h | 87 -- fs/xfs/linux-2.6/xfs_sync.c | 1065 ----------------- fs/xfs/linux-2.6/xfs_sync.h | 51 - fs/xfs/linux-2.6/xfs_sysctl.c | 252 ---- fs/xfs/linux-2.6/xfs_sysctl.h | 102 -- fs/xfs/linux-2.6/xfs_trace.c | 56 - fs/xfs/linux-2.6/xfs_trace.h | 1746 ---------------------------- fs/xfs/linux-2.6/xfs_vnode.h | 64 -- fs/xfs/linux-2.6/xfs_xattr.c | 241 ---- fs/xfs/mrlock.h | 90 ++ fs/xfs/quota/xfs_dquot.c | 1454 ----------------------- fs/xfs/quota/xfs_dquot.h | 137 --- fs/xfs/quota/xfs_dquot_item.c | 529 --------- fs/xfs/quota/xfs_dquot_item.h | 48 - fs/xfs/quota/xfs_qm.c | 2416 --------------------------------------- fs/xfs/quota/xfs_qm.h | 166 --- fs/xfs/quota/xfs_qm_bhv.c | 176 --- fs/xfs/quota/xfs_qm_stats.c | 105 -- fs/xfs/quota/xfs_qm_stats.h | 53 - fs/xfs/quota/xfs_qm_syscalls.c | 906 --------------- fs/xfs/quota/xfs_quota_priv.h | 53 - fs/xfs/quota/xfs_trans_dquot.c | 890 -------------- fs/xfs/support/uuid.c | 63 - fs/xfs/support/uuid.h | 29 - fs/xfs/time.h | 36 + fs/xfs/uuid.c | 63 + fs/xfs/uuid.h | 29 + fs/xfs/xfs_acl.c | 420 +++++++ fs/xfs/xfs_aops.c | 1499 ++++++++++++++++++++++++ fs/xfs/xfs_aops.h | 68 ++ fs/xfs/xfs_buf.c | 1876 ++++++++++++++++++++++++++++++ fs/xfs/xfs_buf.h | 326 ++++++ fs/xfs/xfs_discard.c | 222 ++++ fs/xfs/xfs_discard.h | 10 + fs/xfs/xfs_dquot.c | 1454 +++++++++++++++++++++++ fs/xfs/xfs_dquot.h | 137 +++ fs/xfs/xfs_dquot_item.c | 529 +++++++++ fs/xfs/xfs_dquot_item.h | 48 + fs/xfs/xfs_export.c | 250 ++++ fs/xfs/xfs_export.h | 72 ++ fs/xfs/xfs_file.c | 1096 ++++++++++++++++++ fs/xfs/xfs_fs_subr.c | 96 ++ fs/xfs/xfs_globals.c | 43 + fs/xfs/xfs_ioctl.c | 1556 +++++++++++++++++++++++++ fs/xfs/xfs_ioctl.h | 85 ++ fs/xfs/xfs_ioctl32.c | 672 +++++++++++ fs/xfs/xfs_ioctl32.h | 237 ++++ fs/xfs/xfs_iops.c | 1210 ++++++++++++++++++++ fs/xfs/xfs_iops.h | 30 + fs/xfs/xfs_linux.h | 309 +++++ fs/xfs/xfs_message.c | 108 ++ fs/xfs/xfs_message.h | 39 + fs/xfs/xfs_qm.c | 2416 +++++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_qm.h | 166 +++ fs/xfs/xfs_qm_bhv.c | 176 +++ fs/xfs/xfs_qm_stats.c | 105 ++ fs/xfs/xfs_qm_stats.h | 53 + fs/xfs/xfs_qm_syscalls.c | 906 +++++++++++++++ fs/xfs/xfs_quota_priv.h | 53 + fs/xfs/xfs_quotaops.c | 139 +++ fs/xfs/xfs_stats.c | 122 ++ fs/xfs/xfs_stats.h | 223 ++++ fs/xfs/xfs_super.c | 1773 ++++++++++++++++++++++++++++ fs/xfs/xfs_super.h | 87 ++ fs/xfs/xfs_sync.c | 1065 +++++++++++++++++ fs/xfs/xfs_sync.h | 51 + fs/xfs/xfs_sysctl.c | 252 ++++ fs/xfs/xfs_sysctl.h | 102 ++ fs/xfs/xfs_trace.c | 56 + fs/xfs/xfs_trace.h | 1746 ++++++++++++++++++++++++++++ fs/xfs/xfs_trans_dquot.c | 890 ++++++++++++++ fs/xfs/xfs_vnode.h | 64 ++ fs/xfs/xfs_xattr.c | 241 ++++ kernel/sysctl_binary.c | 2 +- kernel/sysctl_check.c | 2 +- 107 files changed, 23610 insertions(+), 23615 deletions(-) create mode 100644 fs/xfs/kmem.c create mode 100644 fs/xfs/kmem.h delete mode 100644 fs/xfs/linux-2.6/kmem.c delete mode 100644 fs/xfs/linux-2.6/kmem.h delete mode 100644 fs/xfs/linux-2.6/mrlock.h delete mode 100644 fs/xfs/linux-2.6/time.h delete mode 100644 fs/xfs/linux-2.6/xfs_acl.c delete mode 100644 fs/xfs/linux-2.6/xfs_aops.c delete mode 100644 fs/xfs/linux-2.6/xfs_aops.h delete mode 100644 fs/xfs/linux-2.6/xfs_buf.c delete mode 100644 fs/xfs/linux-2.6/xfs_buf.h delete mode 100644 fs/xfs/linux-2.6/xfs_discard.c delete mode 100644 fs/xfs/linux-2.6/xfs_discard.h delete mode 100644 fs/xfs/linux-2.6/xfs_export.c delete mode 100644 fs/xfs/linux-2.6/xfs_export.h delete mode 100644 fs/xfs/linux-2.6/xfs_file.c delete mode 100644 fs/xfs/linux-2.6/xfs_fs_subr.c delete mode 100644 fs/xfs/linux-2.6/xfs_globals.c delete mode 100644 fs/xfs/linux-2.6/xfs_ioctl.c delete mode 100644 fs/xfs/linux-2.6/xfs_ioctl.h delete mode 100644 fs/xfs/linux-2.6/xfs_ioctl32.c delete mode 100644 fs/xfs/linux-2.6/xfs_ioctl32.h delete mode 100644 fs/xfs/linux-2.6/xfs_iops.c delete mode 100644 fs/xfs/linux-2.6/xfs_iops.h delete mode 100644 fs/xfs/linux-2.6/xfs_linux.h delete mode 100644 fs/xfs/linux-2.6/xfs_message.c delete mode 100644 fs/xfs/linux-2.6/xfs_message.h delete mode 100644 fs/xfs/linux-2.6/xfs_quotaops.c delete mode 100644 fs/xfs/linux-2.6/xfs_stats.c delete mode 100644 fs/xfs/linux-2.6/xfs_stats.h delete mode 100644 fs/xfs/linux-2.6/xfs_super.c delete mode 100644 fs/xfs/linux-2.6/xfs_super.h delete mode 100644 fs/xfs/linux-2.6/xfs_sync.c delete mode 100644 fs/xfs/linux-2.6/xfs_sync.h delete mode 100644 fs/xfs/linux-2.6/xfs_sysctl.c delete mode 100644 fs/xfs/linux-2.6/xfs_sysctl.h delete mode 100644 fs/xfs/linux-2.6/xfs_trace.c delete mode 100644 fs/xfs/linux-2.6/xfs_trace.h delete mode 100644 fs/xfs/linux-2.6/xfs_vnode.h delete mode 100644 fs/xfs/linux-2.6/xfs_xattr.c create mode 100644 fs/xfs/mrlock.h delete mode 100644 fs/xfs/quota/xfs_dquot.c delete mode 100644 fs/xfs/quota/xfs_dquot.h delete mode 100644 fs/xfs/quota/xfs_dquot_item.c delete mode 100644 fs/xfs/quota/xfs_dquot_item.h delete mode 100644 fs/xfs/quota/xfs_qm.c delete mode 100644 fs/xfs/quota/xfs_qm.h delete mode 100644 fs/xfs/quota/xfs_qm_bhv.c delete mode 100644 fs/xfs/quota/xfs_qm_stats.c delete mode 100644 fs/xfs/quota/xfs_qm_stats.h delete mode 100644 fs/xfs/quota/xfs_qm_syscalls.c delete mode 100644 fs/xfs/quota/xfs_quota_priv.h delete mode 100644 fs/xfs/quota/xfs_trans_dquot.c delete mode 100644 fs/xfs/support/uuid.c delete mode 100644 fs/xfs/support/uuid.h create mode 100644 fs/xfs/time.h create mode 100644 fs/xfs/uuid.c create mode 100644 fs/xfs/uuid.h create mode 100644 fs/xfs/xfs_acl.c create mode 100644 fs/xfs/xfs_aops.c create mode 100644 fs/xfs/xfs_aops.h create mode 100644 fs/xfs/xfs_buf.c create mode 100644 fs/xfs/xfs_buf.h create mode 100644 fs/xfs/xfs_discard.c create mode 100644 fs/xfs/xfs_discard.h create mode 100644 fs/xfs/xfs_dquot.c create mode 100644 fs/xfs/xfs_dquot.h create mode 100644 fs/xfs/xfs_dquot_item.c create mode 100644 fs/xfs/xfs_dquot_item.h create mode 100644 fs/xfs/xfs_export.c create mode 100644 fs/xfs/xfs_export.h create mode 100644 fs/xfs/xfs_file.c create mode 100644 fs/xfs/xfs_fs_subr.c create mode 100644 fs/xfs/xfs_globals.c create mode 100644 fs/xfs/xfs_ioctl.c create mode 100644 fs/xfs/xfs_ioctl.h create mode 100644 fs/xfs/xfs_ioctl32.c create mode 100644 fs/xfs/xfs_ioctl32.h create mode 100644 fs/xfs/xfs_iops.c create mode 100644 fs/xfs/xfs_iops.h create mode 100644 fs/xfs/xfs_linux.h create mode 100644 fs/xfs/xfs_message.c create mode 100644 fs/xfs/xfs_message.h create mode 100644 fs/xfs/xfs_qm.c create mode 100644 fs/xfs/xfs_qm.h create mode 100644 fs/xfs/xfs_qm_bhv.c create mode 100644 fs/xfs/xfs_qm_stats.c create mode 100644 fs/xfs/xfs_qm_stats.h create mode 100644 fs/xfs/xfs_qm_syscalls.c create mode 100644 fs/xfs/xfs_quota_priv.h create mode 100644 fs/xfs/xfs_quotaops.c create mode 100644 fs/xfs/xfs_stats.c create mode 100644 fs/xfs/xfs_stats.h create mode 100644 fs/xfs/xfs_super.c create mode 100644 fs/xfs/xfs_super.h create mode 100644 fs/xfs/xfs_sync.c create mode 100644 fs/xfs/xfs_sync.h create mode 100644 fs/xfs/xfs_sysctl.c create mode 100644 fs/xfs/xfs_sysctl.h create mode 100644 fs/xfs/xfs_trace.c create mode 100644 fs/xfs/xfs_trace.h create mode 100644 fs/xfs/xfs_trans_dquot.c create mode 100644 fs/xfs/xfs_vnode.h create mode 100644 fs/xfs/xfs_xattr.c (limited to 'kernel') diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile index b100cf445880..ffce328309b8 100644 --- a/fs/xfs/Makefile +++ b/fs/xfs/Makefile @@ -16,44 +16,51 @@ # Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA # -ccflags-y := -I$(src) -I$(src)/linux-2.6 -I$(src)/quota -I$(src)/support ccflags-$(CONFIG_XFS_DEBUG) += -g -XFS_LINUX := linux-2.6 - obj-$(CONFIG_XFS_FS) += xfs.o -xfs-y += linux-2.6/xfs_trace.o - -xfs-$(CONFIG_XFS_QUOTA) += $(addprefix quota/, \ - xfs_dquot.o \ - xfs_dquot_item.o \ - xfs_trans_dquot.o \ - xfs_qm_syscalls.o \ - xfs_qm_bhv.o \ - xfs_qm.o) -xfs-$(CONFIG_XFS_QUOTA) += linux-2.6/xfs_quotaops.o - -ifeq ($(CONFIG_XFS_QUOTA),y) -xfs-$(CONFIG_PROC_FS) += quota/xfs_qm_stats.o -endif - -xfs-$(CONFIG_XFS_RT) += xfs_rtalloc.o -xfs-$(CONFIG_XFS_POSIX_ACL) += $(XFS_LINUX)/xfs_acl.o -xfs-$(CONFIG_PROC_FS) += $(XFS_LINUX)/xfs_stats.o -xfs-$(CONFIG_SYSCTL) += $(XFS_LINUX)/xfs_sysctl.o -xfs-$(CONFIG_COMPAT) += $(XFS_LINUX)/xfs_ioctl32.o +# this one should be compiled first, as the tracing macros can easily blow up +xfs-y += xfs_trace.o +# highlevel code +xfs-y += xfs_aops.o \ + xfs_bit.o \ + xfs_buf.o \ + xfs_dfrag.o \ + xfs_discard.o \ + xfs_error.o \ + xfs_export.o \ + xfs_file.o \ + xfs_filestream.o \ + xfs_fsops.o \ + xfs_fs_subr.o \ + xfs_globals.o \ + xfs_iget.o \ + xfs_ioctl.o \ + xfs_iomap.o \ + xfs_iops.o \ + xfs_itable.o \ + xfs_message.o \ + xfs_mru_cache.o \ + xfs_super.o \ + xfs_sync.o \ + xfs_xattr.o \ + xfs_rename.o \ + xfs_rw.o \ + xfs_utils.o \ + xfs_vnodeops.o \ + kmem.o \ + uuid.o +# code shared with libxfs xfs-y += xfs_alloc.o \ xfs_alloc_btree.o \ xfs_attr.o \ xfs_attr_leaf.o \ - xfs_bit.o \ xfs_bmap.o \ xfs_bmap_btree.o \ xfs_btree.o \ - xfs_buf_item.o \ xfs_da_btree.o \ xfs_dir2.o \ xfs_dir2_block.o \ @@ -61,49 +68,37 @@ xfs-y += xfs_alloc.o \ xfs_dir2_leaf.o \ xfs_dir2_node.o \ xfs_dir2_sf.o \ - xfs_error.o \ - xfs_extfree_item.o \ - xfs_filestream.o \ - xfs_fsops.o \ xfs_ialloc.o \ xfs_ialloc_btree.o \ - xfs_iget.o \ xfs_inode.o \ - xfs_inode_item.o \ - xfs_iomap.o \ - xfs_itable.o \ - xfs_dfrag.o \ - xfs_log.o \ - xfs_log_cil.o \ xfs_log_recover.o \ xfs_mount.o \ - xfs_mru_cache.o \ - xfs_rename.o \ - xfs_trans.o \ + xfs_trans.o + +# low-level transaction/log code +xfs-y += xfs_log.o \ + xfs_log_cil.o \ + xfs_buf_item.o \ + xfs_extfree_item.o \ + xfs_inode_item.o \ xfs_trans_ail.o \ xfs_trans_buf.o \ xfs_trans_extfree.o \ xfs_trans_inode.o \ - xfs_utils.o \ - xfs_vnodeops.o \ - xfs_rw.o - -# Objects in linux/ -xfs-y += $(addprefix $(XFS_LINUX)/, \ - kmem.o \ - xfs_aops.o \ - xfs_buf.o \ - xfs_discard.o \ - xfs_export.o \ - xfs_file.o \ - xfs_fs_subr.o \ - xfs_globals.o \ - xfs_ioctl.o \ - xfs_iops.o \ - xfs_message.o \ - xfs_super.o \ - xfs_sync.o \ - xfs_xattr.o) -# Objects in support/ -xfs-y += support/uuid.o +# optional features +xfs-$(CONFIG_XFS_QUOTA) += xfs_dquot.o \ + xfs_dquot_item.o \ + xfs_trans_dquot.o \ + xfs_qm_syscalls.o \ + xfs_qm_bhv.o \ + xfs_qm.o \ + xfs_quotaops.o +ifeq ($(CONFIG_XFS_QUOTA),y) +xfs-$(CONFIG_PROC_FS) += xfs_qm_stats.o +endif +xfs-$(CONFIG_XFS_RT) += xfs_rtalloc.o +xfs-$(CONFIG_XFS_POSIX_ACL) += xfs_acl.o +xfs-$(CONFIG_PROC_FS) += xfs_stats.o +xfs-$(CONFIG_SYSCTL) += xfs_sysctl.o +xfs-$(CONFIG_COMPAT) += xfs_ioctl32.o diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c new file mode 100644 index 000000000000..a907de565db3 --- /dev/null +++ b/fs/xfs/kmem.c @@ -0,0 +1,132 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include +#include +#include +#include +#include +#include +#include "time.h" +#include "kmem.h" +#include "xfs_message.h" + +/* + * Greedy allocation. May fail and may return vmalloced memory. + * + * Must be freed using kmem_free_large. + */ +void * +kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize) +{ + void *ptr; + size_t kmsize = maxsize; + + while (!(ptr = kmem_zalloc_large(kmsize))) { + if ((kmsize >>= 1) <= minsize) + kmsize = minsize; + } + if (ptr) + *size = kmsize; + return ptr; +} + +void * +kmem_alloc(size_t size, unsigned int __nocast flags) +{ + int retries = 0; + gfp_t lflags = kmem_flags_convert(flags); + void *ptr; + + do { + ptr = kmalloc(size, lflags); + if (ptr || (flags & (KM_MAYFAIL|KM_NOSLEEP))) + return ptr; + if (!(++retries % 100)) + xfs_err(NULL, + "possible memory allocation deadlock in %s (mode:0x%x)", + __func__, lflags); + congestion_wait(BLK_RW_ASYNC, HZ/50); + } while (1); +} + +void * +kmem_zalloc(size_t size, unsigned int __nocast flags) +{ + void *ptr; + + ptr = kmem_alloc(size, flags); + if (ptr) + memset((char *)ptr, 0, (int)size); + return ptr; +} + +void +kmem_free(const void *ptr) +{ + if (!is_vmalloc_addr(ptr)) { + kfree(ptr); + } else { + vfree(ptr); + } +} + +void * +kmem_realloc(const void *ptr, size_t newsize, size_t oldsize, + unsigned int __nocast flags) +{ + void *new; + + new = kmem_alloc(newsize, flags); + if (ptr) { + if (new) + memcpy(new, ptr, + ((oldsize < newsize) ? oldsize : newsize)); + kmem_free(ptr); + } + return new; +} + +void * +kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) +{ + int retries = 0; + gfp_t lflags = kmem_flags_convert(flags); + void *ptr; + + do { + ptr = kmem_cache_alloc(zone, lflags); + if (ptr || (flags & (KM_MAYFAIL|KM_NOSLEEP))) + return ptr; + if (!(++retries % 100)) + xfs_err(NULL, + "possible memory allocation deadlock in %s (mode:0x%x)", + __func__, lflags); + congestion_wait(BLK_RW_ASYNC, HZ/50); + } while (1); +} + +void * +kmem_zone_zalloc(kmem_zone_t *zone, unsigned int __nocast flags) +{ + void *ptr; + + ptr = kmem_zone_alloc(zone, flags); + if (ptr) + memset((char *)ptr, 0, kmem_cache_size(zone)); + return ptr; +} diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h new file mode 100644 index 000000000000..f7c8f7a9ea6d --- /dev/null +++ b/fs/xfs/kmem.h @@ -0,0 +1,124 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SUPPORT_KMEM_H__ +#define __XFS_SUPPORT_KMEM_H__ + +#include +#include +#include +#include + +/* + * General memory allocation interfaces + */ + +#define KM_SLEEP 0x0001u +#define KM_NOSLEEP 0x0002u +#define KM_NOFS 0x0004u +#define KM_MAYFAIL 0x0008u + +/* + * We use a special process flag to avoid recursive callbacks into + * the filesystem during transactions. We will also issue our own + * warnings, so we explicitly skip any generic ones (silly of us). + */ +static inline gfp_t +kmem_flags_convert(unsigned int __nocast flags) +{ + gfp_t lflags; + + BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL)); + + if (flags & KM_NOSLEEP) { + lflags = GFP_ATOMIC | __GFP_NOWARN; + } else { + lflags = GFP_KERNEL | __GFP_NOWARN; + if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS)) + lflags &= ~__GFP_FS; + } + return lflags; +} + +extern void *kmem_alloc(size_t, unsigned int __nocast); +extern void *kmem_zalloc(size_t, unsigned int __nocast); +extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast); +extern void kmem_free(const void *); + +static inline void *kmem_zalloc_large(size_t size) +{ + void *ptr; + + ptr = vmalloc(size); + if (ptr) + memset(ptr, 0, size); + return ptr; +} +static inline void kmem_free_large(void *ptr) +{ + vfree(ptr); +} + +extern void *kmem_zalloc_greedy(size_t *, size_t, size_t); + +/* + * Zone interfaces + */ + +#define KM_ZONE_HWALIGN SLAB_HWCACHE_ALIGN +#define KM_ZONE_RECLAIM SLAB_RECLAIM_ACCOUNT +#define KM_ZONE_SPREAD SLAB_MEM_SPREAD + +#define kmem_zone kmem_cache +#define kmem_zone_t struct kmem_cache + +static inline kmem_zone_t * +kmem_zone_init(int size, char *zone_name) +{ + return kmem_cache_create(zone_name, size, 0, 0, NULL); +} + +static inline kmem_zone_t * +kmem_zone_init_flags(int size, char *zone_name, unsigned long flags, + void (*construct)(void *)) +{ + return kmem_cache_create(zone_name, size, 0, flags, construct); +} + +static inline void +kmem_zone_free(kmem_zone_t *zone, void *ptr) +{ + kmem_cache_free(zone, ptr); +} + +static inline void +kmem_zone_destroy(kmem_zone_t *zone) +{ + if (zone) + kmem_cache_destroy(zone); +} + +extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast); +extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast); + +static inline int +kmem_shake_allow(gfp_t gfp_mask) +{ + return ((gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS)); +} + +#endif /* __XFS_SUPPORT_KMEM_H__ */ diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c deleted file mode 100644 index a907de565db3..000000000000 --- a/fs/xfs/linux-2.6/kmem.c +++ /dev/null @@ -1,132 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include -#include -#include -#include -#include -#include -#include "time.h" -#include "kmem.h" -#include "xfs_message.h" - -/* - * Greedy allocation. May fail and may return vmalloced memory. - * - * Must be freed using kmem_free_large. - */ -void * -kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize) -{ - void *ptr; - size_t kmsize = maxsize; - - while (!(ptr = kmem_zalloc_large(kmsize))) { - if ((kmsize >>= 1) <= minsize) - kmsize = minsize; - } - if (ptr) - *size = kmsize; - return ptr; -} - -void * -kmem_alloc(size_t size, unsigned int __nocast flags) -{ - int retries = 0; - gfp_t lflags = kmem_flags_convert(flags); - void *ptr; - - do { - ptr = kmalloc(size, lflags); - if (ptr || (flags & (KM_MAYFAIL|KM_NOSLEEP))) - return ptr; - if (!(++retries % 100)) - xfs_err(NULL, - "possible memory allocation deadlock in %s (mode:0x%x)", - __func__, lflags); - congestion_wait(BLK_RW_ASYNC, HZ/50); - } while (1); -} - -void * -kmem_zalloc(size_t size, unsigned int __nocast flags) -{ - void *ptr; - - ptr = kmem_alloc(size, flags); - if (ptr) - memset((char *)ptr, 0, (int)size); - return ptr; -} - -void -kmem_free(const void *ptr) -{ - if (!is_vmalloc_addr(ptr)) { - kfree(ptr); - } else { - vfree(ptr); - } -} - -void * -kmem_realloc(const void *ptr, size_t newsize, size_t oldsize, - unsigned int __nocast flags) -{ - void *new; - - new = kmem_alloc(newsize, flags); - if (ptr) { - if (new) - memcpy(new, ptr, - ((oldsize < newsize) ? oldsize : newsize)); - kmem_free(ptr); - } - return new; -} - -void * -kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) -{ - int retries = 0; - gfp_t lflags = kmem_flags_convert(flags); - void *ptr; - - do { - ptr = kmem_cache_alloc(zone, lflags); - if (ptr || (flags & (KM_MAYFAIL|KM_NOSLEEP))) - return ptr; - if (!(++retries % 100)) - xfs_err(NULL, - "possible memory allocation deadlock in %s (mode:0x%x)", - __func__, lflags); - congestion_wait(BLK_RW_ASYNC, HZ/50); - } while (1); -} - -void * -kmem_zone_zalloc(kmem_zone_t *zone, unsigned int __nocast flags) -{ - void *ptr; - - ptr = kmem_zone_alloc(zone, flags); - if (ptr) - memset((char *)ptr, 0, kmem_cache_size(zone)); - return ptr; -} diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h deleted file mode 100644 index f7c8f7a9ea6d..000000000000 --- a/fs/xfs/linux-2.6/kmem.h +++ /dev/null @@ -1,124 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_KMEM_H__ -#define __XFS_SUPPORT_KMEM_H__ - -#include -#include -#include -#include - -/* - * General memory allocation interfaces - */ - -#define KM_SLEEP 0x0001u -#define KM_NOSLEEP 0x0002u -#define KM_NOFS 0x0004u -#define KM_MAYFAIL 0x0008u - -/* - * We use a special process flag to avoid recursive callbacks into - * the filesystem during transactions. We will also issue our own - * warnings, so we explicitly skip any generic ones (silly of us). - */ -static inline gfp_t -kmem_flags_convert(unsigned int __nocast flags) -{ - gfp_t lflags; - - BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL)); - - if (flags & KM_NOSLEEP) { - lflags = GFP_ATOMIC | __GFP_NOWARN; - } else { - lflags = GFP_KERNEL | __GFP_NOWARN; - if ((current->flags & PF_FSTRANS) || (flags & KM_NOFS)) - lflags &= ~__GFP_FS; - } - return lflags; -} - -extern void *kmem_alloc(size_t, unsigned int __nocast); -extern void *kmem_zalloc(size_t, unsigned int __nocast); -extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast); -extern void kmem_free(const void *); - -static inline void *kmem_zalloc_large(size_t size) -{ - void *ptr; - - ptr = vmalloc(size); - if (ptr) - memset(ptr, 0, size); - return ptr; -} -static inline void kmem_free_large(void *ptr) -{ - vfree(ptr); -} - -extern void *kmem_zalloc_greedy(size_t *, size_t, size_t); - -/* - * Zone interfaces - */ - -#define KM_ZONE_HWALIGN SLAB_HWCACHE_ALIGN -#define KM_ZONE_RECLAIM SLAB_RECLAIM_ACCOUNT -#define KM_ZONE_SPREAD SLAB_MEM_SPREAD - -#define kmem_zone kmem_cache -#define kmem_zone_t struct kmem_cache - -static inline kmem_zone_t * -kmem_zone_init(int size, char *zone_name) -{ - return kmem_cache_create(zone_name, size, 0, 0, NULL); -} - -static inline kmem_zone_t * -kmem_zone_init_flags(int size, char *zone_name, unsigned long flags, - void (*construct)(void *)) -{ - return kmem_cache_create(zone_name, size, 0, flags, construct); -} - -static inline void -kmem_zone_free(kmem_zone_t *zone, void *ptr) -{ - kmem_cache_free(zone, ptr); -} - -static inline void -kmem_zone_destroy(kmem_zone_t *zone) -{ - if (zone) - kmem_cache_destroy(zone); -} - -extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast); -extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast); - -static inline int -kmem_shake_allow(gfp_t gfp_mask) -{ - return ((gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS)); -} - -#endif /* __XFS_SUPPORT_KMEM_H__ */ diff --git a/fs/xfs/linux-2.6/mrlock.h b/fs/xfs/linux-2.6/mrlock.h deleted file mode 100644 index ff6a19873e5c..000000000000 --- a/fs/xfs/linux-2.6/mrlock.h +++ /dev/null @@ -1,90 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_MRLOCK_H__ -#define __XFS_SUPPORT_MRLOCK_H__ - -#include - -typedef struct { - struct rw_semaphore mr_lock; -#ifdef DEBUG - int mr_writer; -#endif -} mrlock_t; - -#ifdef DEBUG -#define mrinit(mrp, name) \ - do { (mrp)->mr_writer = 0; init_rwsem(&(mrp)->mr_lock); } while (0) -#else -#define mrinit(mrp, name) \ - do { init_rwsem(&(mrp)->mr_lock); } while (0) -#endif - -#define mrlock_init(mrp, t,n,s) mrinit(mrp, n) -#define mrfree(mrp) do { } while (0) - -static inline void mraccess_nested(mrlock_t *mrp, int subclass) -{ - down_read_nested(&mrp->mr_lock, subclass); -} - -static inline void mrupdate_nested(mrlock_t *mrp, int subclass) -{ - down_write_nested(&mrp->mr_lock, subclass); -#ifdef DEBUG - mrp->mr_writer = 1; -#endif -} - -static inline int mrtryaccess(mrlock_t *mrp) -{ - return down_read_trylock(&mrp->mr_lock); -} - -static inline int mrtryupdate(mrlock_t *mrp) -{ - if (!down_write_trylock(&mrp->mr_lock)) - return 0; -#ifdef DEBUG - mrp->mr_writer = 1; -#endif - return 1; -} - -static inline void mrunlock_excl(mrlock_t *mrp) -{ -#ifdef DEBUG - mrp->mr_writer = 0; -#endif - up_write(&mrp->mr_lock); -} - -static inline void mrunlock_shared(mrlock_t *mrp) -{ - up_read(&mrp->mr_lock); -} - -static inline void mrdemote(mrlock_t *mrp) -{ -#ifdef DEBUG - mrp->mr_writer = 0; -#endif - downgrade_write(&mrp->mr_lock); -} - -#endif /* __XFS_SUPPORT_MRLOCK_H__ */ diff --git a/fs/xfs/linux-2.6/time.h b/fs/xfs/linux-2.6/time.h deleted file mode 100644 index 387e695a184c..000000000000 --- a/fs/xfs/linux-2.6/time.h +++ /dev/null @@ -1,36 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_TIME_H__ -#define __XFS_SUPPORT_TIME_H__ - -#include -#include - -typedef struct timespec timespec_t; - -static inline void delay(long ticks) -{ - schedule_timeout_uninterruptible(ticks); -} - -static inline void nanotime(struct timespec *tvp) -{ - *tvp = CURRENT_TIME; -} - -#endif /* __XFS_SUPPORT_TIME_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c deleted file mode 100644 index b6c4b3795c4a..000000000000 --- a/fs/xfs/linux-2.6/xfs_acl.c +++ /dev/null @@ -1,420 +0,0 @@ -/* - * Copyright (c) 2008, Christoph Hellwig - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_acl.h" -#include "xfs_attr.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" -#include -#include -#include - - -/* - * Locking scheme: - * - all ACL updates are protected by inode->i_mutex, which is taken before - * calling into this file. - */ - -STATIC struct posix_acl * -xfs_acl_from_disk(struct xfs_acl *aclp) -{ - struct posix_acl_entry *acl_e; - struct posix_acl *acl; - struct xfs_acl_entry *ace; - int count, i; - - count = be32_to_cpu(aclp->acl_cnt); - - acl = posix_acl_alloc(count, GFP_KERNEL); - if (!acl) - return ERR_PTR(-ENOMEM); - - for (i = 0; i < count; i++) { - acl_e = &acl->a_entries[i]; - ace = &aclp->acl_entry[i]; - - /* - * The tag is 32 bits on disk and 16 bits in core. - * - * Because every access to it goes through the core - * format first this is not a problem. - */ - acl_e->e_tag = be32_to_cpu(ace->ae_tag); - acl_e->e_perm = be16_to_cpu(ace->ae_perm); - - switch (acl_e->e_tag) { - case ACL_USER: - case ACL_GROUP: - acl_e->e_id = be32_to_cpu(ace->ae_id); - break; - case ACL_USER_OBJ: - case ACL_GROUP_OBJ: - case ACL_MASK: - case ACL_OTHER: - acl_e->e_id = ACL_UNDEFINED_ID; - break; - default: - goto fail; - } - } - return acl; - -fail: - posix_acl_release(acl); - return ERR_PTR(-EINVAL); -} - -STATIC void -xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl) -{ - const struct posix_acl_entry *acl_e; - struct xfs_acl_entry *ace; - int i; - - aclp->acl_cnt = cpu_to_be32(acl->a_count); - for (i = 0; i < acl->a_count; i++) { - ace = &aclp->acl_entry[i]; - acl_e = &acl->a_entries[i]; - - ace->ae_tag = cpu_to_be32(acl_e->e_tag); - ace->ae_id = cpu_to_be32(acl_e->e_id); - ace->ae_perm = cpu_to_be16(acl_e->e_perm); - } -} - -struct posix_acl * -xfs_get_acl(struct inode *inode, int type) -{ - struct xfs_inode *ip = XFS_I(inode); - struct posix_acl *acl; - struct xfs_acl *xfs_acl; - int len = sizeof(struct xfs_acl); - unsigned char *ea_name; - int error; - - acl = get_cached_acl(inode, type); - if (acl != ACL_NOT_CACHED) - return acl; - - trace_xfs_get_acl(ip); - - switch (type) { - case ACL_TYPE_ACCESS: - ea_name = SGI_ACL_FILE; - break; - case ACL_TYPE_DEFAULT: - ea_name = SGI_ACL_DEFAULT; - break; - default: - BUG(); - } - - /* - * If we have a cached ACLs value just return it, not need to - * go out to the disk. - */ - - xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); - if (!xfs_acl) - return ERR_PTR(-ENOMEM); - - error = -xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl, - &len, ATTR_ROOT); - if (error) { - /* - * If the attribute doesn't exist make sure we have a negative - * cache entry, for any other error assume it is transient and - * leave the cache entry as ACL_NOT_CACHED. - */ - if (error == -ENOATTR) { - acl = NULL; - goto out_update_cache; - } - goto out; - } - - acl = xfs_acl_from_disk(xfs_acl); - if (IS_ERR(acl)) - goto out; - - out_update_cache: - set_cached_acl(inode, type, acl); - out: - kfree(xfs_acl); - return acl; -} - -STATIC int -xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) -{ - struct xfs_inode *ip = XFS_I(inode); - unsigned char *ea_name; - int error; - - if (S_ISLNK(inode->i_mode)) - return -EOPNOTSUPP; - - switch (type) { - case ACL_TYPE_ACCESS: - ea_name = SGI_ACL_FILE; - break; - case ACL_TYPE_DEFAULT: - if (!S_ISDIR(inode->i_mode)) - return acl ? -EACCES : 0; - ea_name = SGI_ACL_DEFAULT; - break; - default: - return -EINVAL; - } - - if (acl) { - struct xfs_acl *xfs_acl; - int len; - - xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); - if (!xfs_acl) - return -ENOMEM; - - xfs_acl_to_disk(xfs_acl, acl); - len = sizeof(struct xfs_acl) - - (sizeof(struct xfs_acl_entry) * - (XFS_ACL_MAX_ENTRIES - acl->a_count)); - - error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl, - len, ATTR_ROOT); - - kfree(xfs_acl); - } else { - /* - * A NULL ACL argument means we want to remove the ACL. - */ - error = -xfs_attr_remove(ip, ea_name, ATTR_ROOT); - - /* - * If the attribute didn't exist to start with that's fine. - */ - if (error == -ENOATTR) - error = 0; - } - - if (!error) - set_cached_acl(inode, type, acl); - return error; -} - -static int -xfs_set_mode(struct inode *inode, umode_t mode) -{ - int error = 0; - - if (mode != inode->i_mode) { - struct iattr iattr; - - iattr.ia_valid = ATTR_MODE | ATTR_CTIME; - iattr.ia_mode = mode; - iattr.ia_ctime = current_fs_time(inode->i_sb); - - error = -xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL); - } - - return error; -} - -static int -xfs_acl_exists(struct inode *inode, unsigned char *name) -{ - int len = sizeof(struct xfs_acl); - - return (xfs_attr_get(XFS_I(inode), name, NULL, &len, - ATTR_ROOT|ATTR_KERNOVAL) == 0); -} - -int -posix_acl_access_exists(struct inode *inode) -{ - return xfs_acl_exists(inode, SGI_ACL_FILE); -} - -int -posix_acl_default_exists(struct inode *inode) -{ - if (!S_ISDIR(inode->i_mode)) - return 0; - return xfs_acl_exists(inode, SGI_ACL_DEFAULT); -} - -/* - * No need for i_mutex because the inode is not yet exposed to the VFS. - */ -int -xfs_inherit_acl(struct inode *inode, struct posix_acl *acl) -{ - umode_t mode = inode->i_mode; - int error = 0, inherit = 0; - - if (S_ISDIR(inode->i_mode)) { - error = xfs_set_acl(inode, ACL_TYPE_DEFAULT, acl); - if (error) - goto out; - } - - error = posix_acl_create(&acl, GFP_KERNEL, &mode); - if (error < 0) - return error; - - /* - * If posix_acl_create returns a positive value we need to - * inherit a permission that can't be represented using the Unix - * mode bits and we actually need to set an ACL. - */ - if (error > 0) - inherit = 1; - - error = xfs_set_mode(inode, mode); - if (error) - goto out; - - if (inherit) - error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl); - -out: - posix_acl_release(acl); - return error; -} - -int -xfs_acl_chmod(struct inode *inode) -{ - struct posix_acl *acl; - int error; - - if (S_ISLNK(inode->i_mode)) - return -EOPNOTSUPP; - - acl = xfs_get_acl(inode, ACL_TYPE_ACCESS); - if (IS_ERR(acl) || !acl) - return PTR_ERR(acl); - - error = posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode); - if (error) - return error; - - error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl); - posix_acl_release(acl); - return error; -} - -static int -xfs_xattr_acl_get(struct dentry *dentry, const char *name, - void *value, size_t size, int type) -{ - struct posix_acl *acl; - int error; - - acl = xfs_get_acl(dentry->d_inode, type); - if (IS_ERR(acl)) - return PTR_ERR(acl); - if (acl == NULL) - return -ENODATA; - - error = posix_acl_to_xattr(acl, value, size); - posix_acl_release(acl); - - return error; -} - -static int -xfs_xattr_acl_set(struct dentry *dentry, const char *name, - const void *value, size_t size, int flags, int type) -{ - struct inode *inode = dentry->d_inode; - struct posix_acl *acl = NULL; - int error = 0; - - if (flags & XATTR_CREATE) - return -EINVAL; - if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode)) - return value ? -EACCES : 0; - if ((current_fsuid() != inode->i_uid) && !capable(CAP_FOWNER)) - return -EPERM; - - if (!value) - goto set_acl; - - acl = posix_acl_from_xattr(value, size); - if (!acl) { - /* - * acl_set_file(3) may request that we set default ACLs with - * zero length -- defend (gracefully) against that here. - */ - goto out; - } - if (IS_ERR(acl)) { - error = PTR_ERR(acl); - goto out; - } - - error = posix_acl_valid(acl); - if (error) - goto out_release; - - error = -EINVAL; - if (acl->a_count > XFS_ACL_MAX_ENTRIES) - goto out_release; - - if (type == ACL_TYPE_ACCESS) { - umode_t mode = inode->i_mode; - error = posix_acl_equiv_mode(acl, &mode); - - if (error <= 0) { - posix_acl_release(acl); - acl = NULL; - - if (error < 0) - return error; - } - - error = xfs_set_mode(inode, mode); - if (error) - goto out_release; - } - - set_acl: - error = xfs_set_acl(inode, type, acl); - out_release: - posix_acl_release(acl); - out: - return error; -} - -const struct xattr_handler xfs_xattr_acl_access_handler = { - .prefix = POSIX_ACL_XATTR_ACCESS, - .flags = ACL_TYPE_ACCESS, - .get = xfs_xattr_acl_get, - .set = xfs_xattr_acl_set, -}; - -const struct xattr_handler xfs_xattr_acl_default_handler = { - .prefix = POSIX_ACL_XATTR_DEFAULT, - .flags = ACL_TYPE_DEFAULT, - .get = xfs_xattr_acl_get, - .set = xfs_xattr_acl_set, -}; diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c deleted file mode 100644 index 63e971e2b837..000000000000 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ /dev/null @@ -1,1499 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_trans.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_alloc.h" -#include "xfs_error.h" -#include "xfs_rw.h" -#include "xfs_iomap.h" -#include "xfs_vnodeops.h" -#include "xfs_trace.h" -#include "xfs_bmap.h" -#include -#include -#include -#include - - -/* - * Prime number of hash buckets since address is used as the key. - */ -#define NVSYNC 37 -#define to_ioend_wq(v) (&xfs_ioend_wq[((unsigned long)v) % NVSYNC]) -static wait_queue_head_t xfs_ioend_wq[NVSYNC]; - -void __init -xfs_ioend_init(void) -{ - int i; - - for (i = 0; i < NVSYNC; i++) - init_waitqueue_head(&xfs_ioend_wq[i]); -} - -void -xfs_ioend_wait( - xfs_inode_t *ip) -{ - wait_queue_head_t *wq = to_ioend_wq(ip); - - wait_event(*wq, (atomic_read(&ip->i_iocount) == 0)); -} - -STATIC void -xfs_ioend_wake( - xfs_inode_t *ip) -{ - if (atomic_dec_and_test(&ip->i_iocount)) - wake_up(to_ioend_wq(ip)); -} - -void -xfs_count_page_state( - struct page *page, - int *delalloc, - int *unwritten) -{ - struct buffer_head *bh, *head; - - *delalloc = *unwritten = 0; - - bh = head = page_buffers(page); - do { - if (buffer_unwritten(bh)) - (*unwritten) = 1; - else if (buffer_delay(bh)) - (*delalloc) = 1; - } while ((bh = bh->b_this_page) != head); -} - -STATIC struct block_device * -xfs_find_bdev_for_inode( - struct inode *inode) -{ - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - - if (XFS_IS_REALTIME_INODE(ip)) - return mp->m_rtdev_targp->bt_bdev; - else - return mp->m_ddev_targp->bt_bdev; -} - -/* - * We're now finished for good with this ioend structure. - * Update the page state via the associated buffer_heads, - * release holds on the inode and bio, and finally free - * up memory. Do not use the ioend after this. - */ -STATIC void -xfs_destroy_ioend( - xfs_ioend_t *ioend) -{ - struct buffer_head *bh, *next; - struct xfs_inode *ip = XFS_I(ioend->io_inode); - - for (bh = ioend->io_buffer_head; bh; bh = next) { - next = bh->b_private; - bh->b_end_io(bh, !ioend->io_error); - } - - /* - * Volume managers supporting multiple paths can send back ENODEV - * when the final path disappears. In this case continuing to fill - * the page cache with dirty data which cannot be written out is - * evil, so prevent that. - */ - if (unlikely(ioend->io_error == -ENODEV)) { - xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, - __FILE__, __LINE__); - } - - xfs_ioend_wake(ip); - mempool_free(ioend, xfs_ioend_pool); -} - -/* - * If the end of the current ioend is beyond the current EOF, - * return the new EOF value, otherwise zero. - */ -STATIC xfs_fsize_t -xfs_ioend_new_eof( - xfs_ioend_t *ioend) -{ - xfs_inode_t *ip = XFS_I(ioend->io_inode); - xfs_fsize_t isize; - xfs_fsize_t bsize; - - bsize = ioend->io_offset + ioend->io_size; - isize = MAX(ip->i_size, ip->i_new_size); - isize = MIN(isize, bsize); - return isize > ip->i_d.di_size ? isize : 0; -} - -/* - * Update on-disk file size now that data has been written to disk. The - * current in-memory file size is i_size. If a write is beyond eof i_new_size - * will be the intended file size until i_size is updated. If this write does - * not extend all the way to the valid file size then restrict this update to - * the end of the write. - * - * This function does not block as blocking on the inode lock in IO completion - * can lead to IO completion order dependency deadlocks.. If it can't get the - * inode ilock it will return EAGAIN. Callers must handle this. - */ -STATIC int -xfs_setfilesize( - xfs_ioend_t *ioend) -{ - xfs_inode_t *ip = XFS_I(ioend->io_inode); - xfs_fsize_t isize; - - if (unlikely(ioend->io_error)) - return 0; - - if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) - return EAGAIN; - - isize = xfs_ioend_new_eof(ioend); - if (isize) { - trace_xfs_setfilesize(ip, ioend->io_offset, ioend->io_size); - ip->i_d.di_size = isize; - xfs_mark_inode_dirty(ip); - } - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return 0; -} - -/* - * Schedule IO completion handling on the final put of an ioend. - */ -STATIC void -xfs_finish_ioend( - struct xfs_ioend *ioend) -{ - if (atomic_dec_and_test(&ioend->io_remaining)) { - if (ioend->io_type == IO_UNWRITTEN) - queue_work(xfsconvertd_workqueue, &ioend->io_work); - else - queue_work(xfsdatad_workqueue, &ioend->io_work); - } -} - -/* - * IO write completion. - */ -STATIC void -xfs_end_io( - struct work_struct *work) -{ - xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work); - struct xfs_inode *ip = XFS_I(ioend->io_inode); - int error = 0; - - /* - * For unwritten extents we need to issue transactions to convert a - * range to normal written extens after the data I/O has finished. - */ - if (ioend->io_type == IO_UNWRITTEN && - likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) { - - error = xfs_iomap_write_unwritten(ip, ioend->io_offset, - ioend->io_size); - if (error) - ioend->io_error = error; - } - - /* - * We might have to update the on-disk file size after extending - * writes. - */ - error = xfs_setfilesize(ioend); - ASSERT(!error || error == EAGAIN); - - /* - * If we didn't complete processing of the ioend, requeue it to the - * tail of the workqueue for another attempt later. Otherwise destroy - * it. - */ - if (error == EAGAIN) { - atomic_inc(&ioend->io_remaining); - xfs_finish_ioend(ioend); - /* ensure we don't spin on blocked ioends */ - delay(1); - } else { - if (ioend->io_iocb) - aio_complete(ioend->io_iocb, ioend->io_result, 0); - xfs_destroy_ioend(ioend); - } -} - -/* - * Call IO completion handling in caller context on the final put of an ioend. - */ -STATIC void -xfs_finish_ioend_sync( - struct xfs_ioend *ioend) -{ - if (atomic_dec_and_test(&ioend->io_remaining)) - xfs_end_io(&ioend->io_work); -} - -/* - * Allocate and initialise an IO completion structure. - * We need to track unwritten extent write completion here initially. - * We'll need to extend this for updating the ondisk inode size later - * (vs. incore size). - */ -STATIC xfs_ioend_t * -xfs_alloc_ioend( - struct inode *inode, - unsigned int type) -{ - xfs_ioend_t *ioend; - - ioend = mempool_alloc(xfs_ioend_pool, GFP_NOFS); - - /* - * Set the count to 1 initially, which will prevent an I/O - * completion callback from happening before we have started - * all the I/O from calling the completion routine too early. - */ - atomic_set(&ioend->io_remaining, 1); - ioend->io_error = 0; - ioend->io_list = NULL; - ioend->io_type = type; - ioend->io_inode = inode; - ioend->io_buffer_head = NULL; - ioend->io_buffer_tail = NULL; - atomic_inc(&XFS_I(ioend->io_inode)->i_iocount); - ioend->io_offset = 0; - ioend->io_size = 0; - ioend->io_iocb = NULL; - ioend->io_result = 0; - - INIT_WORK(&ioend->io_work, xfs_end_io); - return ioend; -} - -STATIC int -xfs_map_blocks( - struct inode *inode, - loff_t offset, - struct xfs_bmbt_irec *imap, - int type, - int nonblocking) -{ - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - ssize_t count = 1 << inode->i_blkbits; - xfs_fileoff_t offset_fsb, end_fsb; - int error = 0; - int bmapi_flags = XFS_BMAPI_ENTIRE; - int nimaps = 1; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); - - if (type == IO_UNWRITTEN) - bmapi_flags |= XFS_BMAPI_IGSTATE; - - if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) { - if (nonblocking) - return -XFS_ERROR(EAGAIN); - xfs_ilock(ip, XFS_ILOCK_SHARED); - } - - ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || - (ip->i_df.if_flags & XFS_IFEXTENTS)); - ASSERT(offset <= mp->m_maxioffset); - - if (offset + count > mp->m_maxioffset) - count = mp->m_maxioffset - offset; - end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count); - offset_fsb = XFS_B_TO_FSBT(mp, offset); - error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb, - bmapi_flags, NULL, 0, imap, &nimaps, NULL); - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - if (error) - return -XFS_ERROR(error); - - if (type == IO_DELALLOC && - (!nimaps || isnullstartblock(imap->br_startblock))) { - error = xfs_iomap_write_allocate(ip, offset, count, imap); - if (!error) - trace_xfs_map_blocks_alloc(ip, offset, count, type, imap); - return -XFS_ERROR(error); - } - -#ifdef DEBUG - if (type == IO_UNWRITTEN) { - ASSERT(nimaps); - ASSERT(imap->br_startblock != HOLESTARTBLOCK); - ASSERT(imap->br_startblock != DELAYSTARTBLOCK); - } -#endif - if (nimaps) - trace_xfs_map_blocks_found(ip, offset, count, type, imap); - return 0; -} - -STATIC int -xfs_imap_valid( - struct inode *inode, - struct xfs_bmbt_irec *imap, - xfs_off_t offset) -{ - offset >>= inode->i_blkbits; - - return offset >= imap->br_startoff && - offset < imap->br_startoff + imap->br_blockcount; -} - -/* - * BIO completion handler for buffered IO. - */ -STATIC void -xfs_end_bio( - struct bio *bio, - int error) -{ - xfs_ioend_t *ioend = bio->bi_private; - - ASSERT(atomic_read(&bio->bi_cnt) >= 1); - ioend->io_error = test_bit(BIO_UPTODATE, &bio->bi_flags) ? 0 : error; - - /* Toss bio and pass work off to an xfsdatad thread */ - bio->bi_private = NULL; - bio->bi_end_io = NULL; - bio_put(bio); - - xfs_finish_ioend(ioend); -} - -STATIC void -xfs_submit_ioend_bio( - struct writeback_control *wbc, - xfs_ioend_t *ioend, - struct bio *bio) -{ - atomic_inc(&ioend->io_remaining); - bio->bi_private = ioend; - bio->bi_end_io = xfs_end_bio; - - /* - * If the I/O is beyond EOF we mark the inode dirty immediately - * but don't update the inode size until I/O completion. - */ - if (xfs_ioend_new_eof(ioend)) - xfs_mark_inode_dirty(XFS_I(ioend->io_inode)); - - submit_bio(wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE, bio); -} - -STATIC struct bio * -xfs_alloc_ioend_bio( - struct buffer_head *bh) -{ - int nvecs = bio_get_nr_vecs(bh->b_bdev); - struct bio *bio = bio_alloc(GFP_NOIO, nvecs); - - ASSERT(bio->bi_private == NULL); - bio->bi_sector = bh->b_blocknr * (bh->b_size >> 9); - bio->bi_bdev = bh->b_bdev; - return bio; -} - -STATIC void -xfs_start_buffer_writeback( - struct buffer_head *bh) -{ - ASSERT(buffer_mapped(bh)); - ASSERT(buffer_locked(bh)); - ASSERT(!buffer_delay(bh)); - ASSERT(!buffer_unwritten(bh)); - - mark_buffer_async_write(bh); - set_buffer_uptodate(bh); - clear_buffer_dirty(bh); -} - -STATIC void -xfs_start_page_writeback( - struct page *page, - int clear_dirty, - int buffers) -{ - ASSERT(PageLocked(page)); - ASSERT(!PageWriteback(page)); - if (clear_dirty) - clear_page_dirty_for_io(page); - set_page_writeback(page); - unlock_page(page); - /* If no buffers on the page are to be written, finish it here */ - if (!buffers) - end_page_writeback(page); -} - -static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh) -{ - return bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh)); -} - -/* - * Submit all of the bios for all of the ioends we have saved up, covering the - * initial writepage page and also any probed pages. - * - * Because we may have multiple ioends spanning a page, we need to start - * writeback on all the buffers before we submit them for I/O. If we mark the - * buffers as we got, then we can end up with a page that only has buffers - * marked async write and I/O complete on can occur before we mark the other - * buffers async write. - * - * The end result of this is that we trip a bug in end_page_writeback() because - * we call it twice for the one page as the code in end_buffer_async_write() - * assumes that all buffers on the page are started at the same time. - * - * The fix is two passes across the ioend list - one to start writeback on the - * buffer_heads, and then submit them for I/O on the second pass. - */ -STATIC void -xfs_submit_ioend( - struct writeback_control *wbc, - xfs_ioend_t *ioend) -{ - xfs_ioend_t *head = ioend; - xfs_ioend_t *next; - struct buffer_head *bh; - struct bio *bio; - sector_t lastblock = 0; - - /* Pass 1 - start writeback */ - do { - next = ioend->io_list; - for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) - xfs_start_buffer_writeback(bh); - } while ((ioend = next) != NULL); - - /* Pass 2 - submit I/O */ - ioend = head; - do { - next = ioend->io_list; - bio = NULL; - - for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) { - - if (!bio) { - retry: - bio = xfs_alloc_ioend_bio(bh); - } else if (bh->b_blocknr != lastblock + 1) { - xfs_submit_ioend_bio(wbc, ioend, bio); - goto retry; - } - - if (bio_add_buffer(bio, bh) != bh->b_size) { - xfs_submit_ioend_bio(wbc, ioend, bio); - goto retry; - } - - lastblock = bh->b_blocknr; - } - if (bio) - xfs_submit_ioend_bio(wbc, ioend, bio); - xfs_finish_ioend(ioend); - } while ((ioend = next) != NULL); -} - -/* - * Cancel submission of all buffer_heads so far in this endio. - * Toss the endio too. Only ever called for the initial page - * in a writepage request, so only ever one page. - */ -STATIC void -xfs_cancel_ioend( - xfs_ioend_t *ioend) -{ - xfs_ioend_t *next; - struct buffer_head *bh, *next_bh; - - do { - next = ioend->io_list; - bh = ioend->io_buffer_head; - do { - next_bh = bh->b_private; - clear_buffer_async_write(bh); - unlock_buffer(bh); - } while ((bh = next_bh) != NULL); - - xfs_ioend_wake(XFS_I(ioend->io_inode)); - mempool_free(ioend, xfs_ioend_pool); - } while ((ioend = next) != NULL); -} - -/* - * Test to see if we've been building up a completion structure for - * earlier buffers -- if so, we try to append to this ioend if we - * can, otherwise we finish off any current ioend and start another. - * Return true if we've finished the given ioend. - */ -STATIC void -xfs_add_to_ioend( - struct inode *inode, - struct buffer_head *bh, - xfs_off_t offset, - unsigned int type, - xfs_ioend_t **result, - int need_ioend) -{ - xfs_ioend_t *ioend = *result; - - if (!ioend || need_ioend || type != ioend->io_type) { - xfs_ioend_t *previous = *result; - - ioend = xfs_alloc_ioend(inode, type); - ioend->io_offset = offset; - ioend->io_buffer_head = bh; - ioend->io_buffer_tail = bh; - if (previous) - previous->io_list = ioend; - *result = ioend; - } else { - ioend->io_buffer_tail->b_private = bh; - ioend->io_buffer_tail = bh; - } - - bh->b_private = NULL; - ioend->io_size += bh->b_size; -} - -STATIC void -xfs_map_buffer( - struct inode *inode, - struct buffer_head *bh, - struct xfs_bmbt_irec *imap, - xfs_off_t offset) -{ - sector_t bn; - struct xfs_mount *m = XFS_I(inode)->i_mount; - xfs_off_t iomap_offset = XFS_FSB_TO_B(m, imap->br_startoff); - xfs_daddr_t iomap_bn = xfs_fsb_to_db(XFS_I(inode), imap->br_startblock); - - ASSERT(imap->br_startblock != HOLESTARTBLOCK); - ASSERT(imap->br_startblock != DELAYSTARTBLOCK); - - bn = (iomap_bn >> (inode->i_blkbits - BBSHIFT)) + - ((offset - iomap_offset) >> inode->i_blkbits); - - ASSERT(bn || XFS_IS_REALTIME_INODE(XFS_I(inode))); - - bh->b_blocknr = bn; - set_buffer_mapped(bh); -} - -STATIC void -xfs_map_at_offset( - struct inode *inode, - struct buffer_head *bh, - struct xfs_bmbt_irec *imap, - xfs_off_t offset) -{ - ASSERT(imap->br_startblock != HOLESTARTBLOCK); - ASSERT(imap->br_startblock != DELAYSTARTBLOCK); - - xfs_map_buffer(inode, bh, imap, offset); - set_buffer_mapped(bh); - clear_buffer_delay(bh); - clear_buffer_unwritten(bh); -} - -/* - * Test if a given page is suitable for writing as part of an unwritten - * or delayed allocate extent. - */ -STATIC int -xfs_is_delayed_page( - struct page *page, - unsigned int type) -{ - if (PageWriteback(page)) - return 0; - - if (page->mapping && page_has_buffers(page)) { - struct buffer_head *bh, *head; - int acceptable = 0; - - bh = head = page_buffers(page); - do { - if (buffer_unwritten(bh)) - acceptable = (type == IO_UNWRITTEN); - else if (buffer_delay(bh)) - acceptable = (type == IO_DELALLOC); - else if (buffer_dirty(bh) && buffer_mapped(bh)) - acceptable = (type == IO_OVERWRITE); - else - break; - } while ((bh = bh->b_this_page) != head); - - if (acceptable) - return 1; - } - - return 0; -} - -/* - * Allocate & map buffers for page given the extent map. Write it out. - * except for the original page of a writepage, this is called on - * delalloc/unwritten pages only, for the original page it is possible - * that the page has no mapping at all. - */ -STATIC int -xfs_convert_page( - struct inode *inode, - struct page *page, - loff_t tindex, - struct xfs_bmbt_irec *imap, - xfs_ioend_t **ioendp, - struct writeback_control *wbc) -{ - struct buffer_head *bh, *head; - xfs_off_t end_offset; - unsigned long p_offset; - unsigned int type; - int len, page_dirty; - int count = 0, done = 0, uptodate = 1; - xfs_off_t offset = page_offset(page); - - if (page->index != tindex) - goto fail; - if (!trylock_page(page)) - goto fail; - if (PageWriteback(page)) - goto fail_unlock_page; - if (page->mapping != inode->i_mapping) - goto fail_unlock_page; - if (!xfs_is_delayed_page(page, (*ioendp)->io_type)) - goto fail_unlock_page; - - /* - * page_dirty is initially a count of buffers on the page before - * EOF and is decremented as we move each into a cleanable state. - * - * Derivation: - * - * End offset is the highest offset that this page should represent. - * If we are on the last page, (end_offset & (PAGE_CACHE_SIZE - 1)) - * will evaluate non-zero and be less than PAGE_CACHE_SIZE and - * hence give us the correct page_dirty count. On any other page, - * it will be zero and in that case we need page_dirty to be the - * count of buffers on the page. - */ - end_offset = min_t(unsigned long long, - (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, - i_size_read(inode)); - - len = 1 << inode->i_blkbits; - p_offset = min_t(unsigned long, end_offset & (PAGE_CACHE_SIZE - 1), - PAGE_CACHE_SIZE); - p_offset = p_offset ? roundup(p_offset, len) : PAGE_CACHE_SIZE; - page_dirty = p_offset / len; - - bh = head = page_buffers(page); - do { - if (offset >= end_offset) - break; - if (!buffer_uptodate(bh)) - uptodate = 0; - if (!(PageUptodate(page) || buffer_uptodate(bh))) { - done = 1; - continue; - } - - if (buffer_unwritten(bh) || buffer_delay(bh) || - buffer_mapped(bh)) { - if (buffer_unwritten(bh)) - type = IO_UNWRITTEN; - else if (buffer_delay(bh)) - type = IO_DELALLOC; - else - type = IO_OVERWRITE; - - if (!xfs_imap_valid(inode, imap, offset)) { - done = 1; - continue; - } - - lock_buffer(bh); - if (type != IO_OVERWRITE) - xfs_map_at_offset(inode, bh, imap, offset); - xfs_add_to_ioend(inode, bh, offset, type, - ioendp, done); - - page_dirty--; - count++; - } else { - done = 1; - } - } while (offset += len, (bh = bh->b_this_page) != head); - - if (uptodate && bh == head) - SetPageUptodate(page); - - if (count) { - if (--wbc->nr_to_write <= 0 && - wbc->sync_mode == WB_SYNC_NONE) - done = 1; - } - xfs_start_page_writeback(page, !page_dirty, count); - - return done; - fail_unlock_page: - unlock_page(page); - fail: - return 1; -} - -/* - * Convert & write out a cluster of pages in the same extent as defined - * by mp and following the start page. - */ -STATIC void -xfs_cluster_write( - struct inode *inode, - pgoff_t tindex, - struct xfs_bmbt_irec *imap, - xfs_ioend_t **ioendp, - struct writeback_control *wbc, - pgoff_t tlast) -{ - struct pagevec pvec; - int done = 0, i; - - pagevec_init(&pvec, 0); - while (!done && tindex <= tlast) { - unsigned len = min_t(pgoff_t, PAGEVEC_SIZE, tlast - tindex + 1); - - if (!pagevec_lookup(&pvec, inode->i_mapping, tindex, len)) - break; - - for (i = 0; i < pagevec_count(&pvec); i++) { - done = xfs_convert_page(inode, pvec.pages[i], tindex++, - imap, ioendp, wbc); - if (done) - break; - } - - pagevec_release(&pvec); - cond_resched(); - } -} - -STATIC void -xfs_vm_invalidatepage( - struct page *page, - unsigned long offset) -{ - trace_xfs_invalidatepage(page->mapping->host, page, offset); - block_invalidatepage(page, offset); -} - -/* - * If the page has delalloc buffers on it, we need to punch them out before we - * invalidate the page. If we don't, we leave a stale delalloc mapping on the - * inode that can trip a BUG() in xfs_get_blocks() later on if a direct IO read - * is done on that same region - the delalloc extent is returned when none is - * supposed to be there. - * - * We prevent this by truncating away the delalloc regions on the page before - * invalidating it. Because they are delalloc, we can do this without needing a - * transaction. Indeed - if we get ENOSPC errors, we have to be able to do this - * truncation without a transaction as there is no space left for block - * reservation (typically why we see a ENOSPC in writeback). - * - * This is not a performance critical path, so for now just do the punching a - * buffer head at a time. - */ -STATIC void -xfs_aops_discard_page( - struct page *page) -{ - struct inode *inode = page->mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct buffer_head *bh, *head; - loff_t offset = page_offset(page); - - if (!xfs_is_delayed_page(page, IO_DELALLOC)) - goto out_invalidate; - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - goto out_invalidate; - - xfs_alert(ip->i_mount, - "page discard on page %p, inode 0x%llx, offset %llu.", - page, ip->i_ino, offset); - - xfs_ilock(ip, XFS_ILOCK_EXCL); - bh = head = page_buffers(page); - do { - int error; - xfs_fileoff_t start_fsb; - - if (!buffer_delay(bh)) - goto next_buffer; - - start_fsb = XFS_B_TO_FSBT(ip->i_mount, offset); - error = xfs_bmap_punch_delalloc_range(ip, start_fsb, 1); - if (error) { - /* something screwed, just bail */ - if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_alert(ip->i_mount, - "page discard unable to remove delalloc mapping."); - } - break; - } -next_buffer: - offset += 1 << inode->i_blkbits; - - } while ((bh = bh->b_this_page) != head); - - xfs_iunlock(ip, XFS_ILOCK_EXCL); -out_invalidate: - xfs_vm_invalidatepage(page, 0); - return; -} - -/* - * Write out a dirty page. - * - * For delalloc space on the page we need to allocate space and flush it. - * For unwritten space on the page we need to start the conversion to - * regular allocated space. - * For any other dirty buffer heads on the page we should flush them. - */ -STATIC int -xfs_vm_writepage( - struct page *page, - struct writeback_control *wbc) -{ - struct inode *inode = page->mapping->host; - struct buffer_head *bh, *head; - struct xfs_bmbt_irec imap; - xfs_ioend_t *ioend = NULL, *iohead = NULL; - loff_t offset; - unsigned int type; - __uint64_t end_offset; - pgoff_t end_index, last_index; - ssize_t len; - int err, imap_valid = 0, uptodate = 1; - int count = 0; - int nonblocking = 0; - - trace_xfs_writepage(inode, page, 0); - - ASSERT(page_has_buffers(page)); - - /* - * Refuse to write the page out if we are called from reclaim context. - * - * This avoids stack overflows when called from deeply used stacks in - * random callers for direct reclaim or memcg reclaim. We explicitly - * allow reclaim from kswapd as the stack usage there is relatively low. - * - * This should really be done by the core VM, but until that happens - * filesystems like XFS, btrfs and ext4 have to take care of this - * by themselves. - */ - if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC) - goto redirty; - - /* - * Given that we do not allow direct reclaim to call us, we should - * never be called while in a filesystem transaction. - */ - if (WARN_ON(current->flags & PF_FSTRANS)) - goto redirty; - - /* Is this page beyond the end of the file? */ - offset = i_size_read(inode); - end_index = offset >> PAGE_CACHE_SHIFT; - last_index = (offset - 1) >> PAGE_CACHE_SHIFT; - if (page->index >= end_index) { - if ((page->index >= end_index + 1) || - !(i_size_read(inode) & (PAGE_CACHE_SIZE - 1))) { - unlock_page(page); - return 0; - } - } - - end_offset = min_t(unsigned long long, - (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, - offset); - len = 1 << inode->i_blkbits; - - bh = head = page_buffers(page); - offset = page_offset(page); - type = IO_OVERWRITE; - - if (wbc->sync_mode == WB_SYNC_NONE) - nonblocking = 1; - - do { - int new_ioend = 0; - - if (offset >= end_offset) - break; - if (!buffer_uptodate(bh)) - uptodate = 0; - - /* - * set_page_dirty dirties all buffers in a page, independent - * of their state. The dirty state however is entirely - * meaningless for holes (!mapped && uptodate), so skip - * buffers covering holes here. - */ - if (!buffer_mapped(bh) && buffer_uptodate(bh)) { - imap_valid = 0; - continue; - } - - if (buffer_unwritten(bh)) { - if (type != IO_UNWRITTEN) { - type = IO_UNWRITTEN; - imap_valid = 0; - } - } else if (buffer_delay(bh)) { - if (type != IO_DELALLOC) { - type = IO_DELALLOC; - imap_valid = 0; - } - } else if (buffer_uptodate(bh)) { - if (type != IO_OVERWRITE) { - type = IO_OVERWRITE; - imap_valid = 0; - } - } else { - if (PageUptodate(page)) { - ASSERT(buffer_mapped(bh)); - imap_valid = 0; - } - continue; - } - - if (imap_valid) - imap_valid = xfs_imap_valid(inode, &imap, offset); - if (!imap_valid) { - /* - * If we didn't have a valid mapping then we need to - * put the new mapping into a separate ioend structure. - * This ensures non-contiguous extents always have - * separate ioends, which is particularly important - * for unwritten extent conversion at I/O completion - * time. - */ - new_ioend = 1; - err = xfs_map_blocks(inode, offset, &imap, type, - nonblocking); - if (err) - goto error; - imap_valid = xfs_imap_valid(inode, &imap, offset); - } - if (imap_valid) { - lock_buffer(bh); - if (type != IO_OVERWRITE) - xfs_map_at_offset(inode, bh, &imap, offset); - xfs_add_to_ioend(inode, bh, offset, type, &ioend, - new_ioend); - count++; - } - - if (!iohead) - iohead = ioend; - - } while (offset += len, ((bh = bh->b_this_page) != head)); - - if (uptodate && bh == head) - SetPageUptodate(page); - - xfs_start_page_writeback(page, 1, count); - - if (ioend && imap_valid) { - xfs_off_t end_index; - - end_index = imap.br_startoff + imap.br_blockcount; - - /* to bytes */ - end_index <<= inode->i_blkbits; - - /* to pages */ - end_index = (end_index - 1) >> PAGE_CACHE_SHIFT; - - /* check against file size */ - if (end_index > last_index) - end_index = last_index; - - xfs_cluster_write(inode, page->index + 1, &imap, &ioend, - wbc, end_index); - } - - if (iohead) - xfs_submit_ioend(wbc, iohead); - - return 0; - -error: - if (iohead) - xfs_cancel_ioend(iohead); - - if (err == -EAGAIN) - goto redirty; - - xfs_aops_discard_page(page); - ClearPageUptodate(page); - unlock_page(page); - return err; - -redirty: - redirty_page_for_writepage(wbc, page); - unlock_page(page); - return 0; -} - -STATIC int -xfs_vm_writepages( - struct address_space *mapping, - struct writeback_control *wbc) -{ - xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); - return generic_writepages(mapping, wbc); -} - -/* - * Called to move a page into cleanable state - and from there - * to be released. The page should already be clean. We always - * have buffer heads in this call. - * - * Returns 1 if the page is ok to release, 0 otherwise. - */ -STATIC int -xfs_vm_releasepage( - struct page *page, - gfp_t gfp_mask) -{ - int delalloc, unwritten; - - trace_xfs_releasepage(page->mapping->host, page, 0); - - xfs_count_page_state(page, &delalloc, &unwritten); - - if (WARN_ON(delalloc)) - return 0; - if (WARN_ON(unwritten)) - return 0; - - return try_to_free_buffers(page); -} - -STATIC int -__xfs_get_blocks( - struct inode *inode, - sector_t iblock, - struct buffer_head *bh_result, - int create, - int direct) -{ - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - xfs_fileoff_t offset_fsb, end_fsb; - int error = 0; - int lockmode = 0; - struct xfs_bmbt_irec imap; - int nimaps = 1; - xfs_off_t offset; - ssize_t size; - int new = 0; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); - - offset = (xfs_off_t)iblock << inode->i_blkbits; - ASSERT(bh_result->b_size >= (1 << inode->i_blkbits)); - size = bh_result->b_size; - - if (!create && direct && offset >= i_size_read(inode)) - return 0; - - if (create) { - lockmode = XFS_ILOCK_EXCL; - xfs_ilock(ip, lockmode); - } else { - lockmode = xfs_ilock_map_shared(ip); - } - - ASSERT(offset <= mp->m_maxioffset); - if (offset + size > mp->m_maxioffset) - size = mp->m_maxioffset - offset; - end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + size); - offset_fsb = XFS_B_TO_FSBT(mp, offset); - - error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb, - XFS_BMAPI_ENTIRE, NULL, 0, &imap, &nimaps, NULL); - if (error) - goto out_unlock; - - if (create && - (!nimaps || - (imap.br_startblock == HOLESTARTBLOCK || - imap.br_startblock == DELAYSTARTBLOCK))) { - if (direct) { - error = xfs_iomap_write_direct(ip, offset, size, - &imap, nimaps); - } else { - error = xfs_iomap_write_delay(ip, offset, size, &imap); - } - if (error) - goto out_unlock; - - trace_xfs_get_blocks_alloc(ip, offset, size, 0, &imap); - } else if (nimaps) { - trace_xfs_get_blocks_found(ip, offset, size, 0, &imap); - } else { - trace_xfs_get_blocks_notfound(ip, offset, size); - goto out_unlock; - } - xfs_iunlock(ip, lockmode); - - if (imap.br_startblock != HOLESTARTBLOCK && - imap.br_startblock != DELAYSTARTBLOCK) { - /* - * For unwritten extents do not report a disk address on - * the read case (treat as if we're reading into a hole). - */ - if (create || !ISUNWRITTEN(&imap)) - xfs_map_buffer(inode, bh_result, &imap, offset); - if (create && ISUNWRITTEN(&imap)) { - if (direct) - bh_result->b_private = inode; - set_buffer_unwritten(bh_result); - } - } - - /* - * If this is a realtime file, data may be on a different device. - * to that pointed to from the buffer_head b_bdev currently. - */ - bh_result->b_bdev = xfs_find_bdev_for_inode(inode); - - /* - * If we previously allocated a block out beyond eof and we are now - * coming back to use it then we will need to flag it as new even if it - * has a disk address. - * - * With sub-block writes into unwritten extents we also need to mark - * the buffer as new so that the unwritten parts of the buffer gets - * correctly zeroed. - */ - if (create && - ((!buffer_mapped(bh_result) && !buffer_uptodate(bh_result)) || - (offset >= i_size_read(inode)) || - (new || ISUNWRITTEN(&imap)))) - set_buffer_new(bh_result); - - if (imap.br_startblock == DELAYSTARTBLOCK) { - BUG_ON(direct); - if (create) { - set_buffer_uptodate(bh_result); - set_buffer_mapped(bh_result); - set_buffer_delay(bh_result); - } - } - - /* - * If this is O_DIRECT or the mpage code calling tell them how large - * the mapping is, so that we can avoid repeated get_blocks calls. - */ - if (direct || size > (1 << inode->i_blkbits)) { - xfs_off_t mapping_size; - - mapping_size = imap.br_startoff + imap.br_blockcount - iblock; - mapping_size <<= inode->i_blkbits; - - ASSERT(mapping_size > 0); - if (mapping_size > size) - mapping_size = size; - if (mapping_size > LONG_MAX) - mapping_size = LONG_MAX; - - bh_result->b_size = mapping_size; - } - - return 0; - -out_unlock: - xfs_iunlock(ip, lockmode); - return -error; -} - -int -xfs_get_blocks( - struct inode *inode, - sector_t iblock, - struct buffer_head *bh_result, - int create) -{ - return __xfs_get_blocks(inode, iblock, bh_result, create, 0); -} - -STATIC int -xfs_get_blocks_direct( - struct inode *inode, - sector_t iblock, - struct buffer_head *bh_result, - int create) -{ - return __xfs_get_blocks(inode, iblock, bh_result, create, 1); -} - -/* - * Complete a direct I/O write request. - * - * If the private argument is non-NULL __xfs_get_blocks signals us that we - * need to issue a transaction to convert the range from unwritten to written - * extents. In case this is regular synchronous I/O we just call xfs_end_io - * to do this and we are done. But in case this was a successful AIO - * request this handler is called from interrupt context, from which we - * can't start transactions. In that case offload the I/O completion to - * the workqueues we also use for buffered I/O completion. - */ -STATIC void -xfs_end_io_direct_write( - struct kiocb *iocb, - loff_t offset, - ssize_t size, - void *private, - int ret, - bool is_async) -{ - struct xfs_ioend *ioend = iocb->private; - - /* - * blockdev_direct_IO can return an error even after the I/O - * completion handler was called. Thus we need to protect - * against double-freeing. - */ - iocb->private = NULL; - - ioend->io_offset = offset; - ioend->io_size = size; - if (private && size > 0) - ioend->io_type = IO_UNWRITTEN; - - if (is_async) { - /* - * If we are converting an unwritten extent we need to delay - * the AIO completion until after the unwrittent extent - * conversion has completed, otherwise do it ASAP. - */ - if (ioend->io_type == IO_UNWRITTEN) { - ioend->io_iocb = iocb; - ioend->io_result = ret; - } else { - aio_complete(iocb, ret, 0); - } - xfs_finish_ioend(ioend); - } else { - xfs_finish_ioend_sync(ioend); - } - - /* XXX: probably should move into the real I/O completion handler */ - inode_dio_done(ioend->io_inode); -} - -STATIC ssize_t -xfs_vm_direct_IO( - int rw, - struct kiocb *iocb, - const struct iovec *iov, - loff_t offset, - unsigned long nr_segs) -{ - struct inode *inode = iocb->ki_filp->f_mapping->host; - struct block_device *bdev = xfs_find_bdev_for_inode(inode); - ssize_t ret; - - if (rw & WRITE) { - iocb->private = xfs_alloc_ioend(inode, IO_DIRECT); - - ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov, - offset, nr_segs, - xfs_get_blocks_direct, - xfs_end_io_direct_write, NULL, 0); - if (ret != -EIOCBQUEUED && iocb->private) - xfs_destroy_ioend(iocb->private); - } else { - ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov, - offset, nr_segs, - xfs_get_blocks_direct, - NULL, NULL, 0); - } - - return ret; -} - -STATIC void -xfs_vm_write_failed( - struct address_space *mapping, - loff_t to) -{ - struct inode *inode = mapping->host; - - if (to > inode->i_size) { - /* - * punch out the delalloc blocks we have already allocated. We - * don't call xfs_setattr() to do this as we may be in the - * middle of a multi-iovec write and so the vfs inode->i_size - * will not match the xfs ip->i_size and so it will zero too - * much. Hence we jus truncate the page cache to zero what is - * necessary and punch the delalloc blocks directly. - */ - struct xfs_inode *ip = XFS_I(inode); - xfs_fileoff_t start_fsb; - xfs_fileoff_t end_fsb; - int error; - - truncate_pagecache(inode, to, inode->i_size); - - /* - * Check if there are any blocks that are outside of i_size - * that need to be trimmed back. - */ - start_fsb = XFS_B_TO_FSB(ip->i_mount, inode->i_size) + 1; - end_fsb = XFS_B_TO_FSB(ip->i_mount, to); - if (end_fsb <= start_fsb) - return; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_bmap_punch_delalloc_range(ip, start_fsb, - end_fsb - start_fsb); - if (error) { - /* something screwed, just bail */ - if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_alert(ip->i_mount, - "xfs_vm_write_failed: unable to clean up ino %lld", - ip->i_ino); - } - } - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } -} - -STATIC int -xfs_vm_write_begin( - struct file *file, - struct address_space *mapping, - loff_t pos, - unsigned len, - unsigned flags, - struct page **pagep, - void **fsdata) -{ - int ret; - - ret = block_write_begin(mapping, pos, len, flags | AOP_FLAG_NOFS, - pagep, xfs_get_blocks); - if (unlikely(ret)) - xfs_vm_write_failed(mapping, pos + len); - return ret; -} - -STATIC int -xfs_vm_write_end( - struct file *file, - struct address_space *mapping, - loff_t pos, - unsigned len, - unsigned copied, - struct page *page, - void *fsdata) -{ - int ret; - - ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); - if (unlikely(ret < len)) - xfs_vm_write_failed(mapping, pos + len); - return ret; -} - -STATIC sector_t -xfs_vm_bmap( - struct address_space *mapping, - sector_t block) -{ - struct inode *inode = (struct inode *)mapping->host; - struct xfs_inode *ip = XFS_I(inode); - - trace_xfs_vm_bmap(XFS_I(inode)); - xfs_ilock(ip, XFS_IOLOCK_SHARED); - xfs_flush_pages(ip, (xfs_off_t)0, -1, 0, FI_REMAPF); - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - return generic_block_bmap(mapping, block, xfs_get_blocks); -} - -STATIC int -xfs_vm_readpage( - struct file *unused, - struct page *page) -{ - return mpage_readpage(page, xfs_get_blocks); -} - -STATIC int -xfs_vm_readpages( - struct file *unused, - struct address_space *mapping, - struct list_head *pages, - unsigned nr_pages) -{ - return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks); -} - -const struct address_space_operations xfs_address_space_operations = { - .readpage = xfs_vm_readpage, - .readpages = xfs_vm_readpages, - .writepage = xfs_vm_writepage, - .writepages = xfs_vm_writepages, - .releasepage = xfs_vm_releasepage, - .invalidatepage = xfs_vm_invalidatepage, - .write_begin = xfs_vm_write_begin, - .write_end = xfs_vm_write_end, - .bmap = xfs_vm_bmap, - .direct_IO = xfs_vm_direct_IO, - .migratepage = buffer_migrate_page, - .is_partially_uptodate = block_is_partially_uptodate, - .error_remove_page = generic_error_remove_page, -}; diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h deleted file mode 100644 index 71f721e1a71f..000000000000 --- a/fs/xfs/linux-2.6/xfs_aops.h +++ /dev/null @@ -1,68 +0,0 @@ -/* - * Copyright (c) 2005-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_AOPS_H__ -#define __XFS_AOPS_H__ - -extern struct workqueue_struct *xfsdatad_workqueue; -extern struct workqueue_struct *xfsconvertd_workqueue; -extern mempool_t *xfs_ioend_pool; - -/* - * Types of I/O for bmap clustering and I/O completion tracking. - */ -enum { - IO_DIRECT = 0, /* special case for direct I/O ioends */ - IO_DELALLOC, /* mapping covers delalloc region */ - IO_UNWRITTEN, /* mapping covers allocated but uninitialized data */ - IO_OVERWRITE, /* mapping covers already allocated extent */ -}; - -#define XFS_IO_TYPES \ - { 0, "" }, \ - { IO_DELALLOC, "delalloc" }, \ - { IO_UNWRITTEN, "unwritten" }, \ - { IO_OVERWRITE, "overwrite" } - -/* - * xfs_ioend struct manages large extent writes for XFS. - * It can manage several multi-page bio's at once. - */ -typedef struct xfs_ioend { - struct xfs_ioend *io_list; /* next ioend in chain */ - unsigned int io_type; /* delalloc / unwritten */ - int io_error; /* I/O error code */ - atomic_t io_remaining; /* hold count */ - struct inode *io_inode; /* file being written to */ - struct buffer_head *io_buffer_head;/* buffer linked list head */ - struct buffer_head *io_buffer_tail;/* buffer linked list tail */ - size_t io_size; /* size of the extent */ - xfs_off_t io_offset; /* offset in the file */ - struct work_struct io_work; /* xfsdatad work queue */ - struct kiocb *io_iocb; - int io_result; -} xfs_ioend_t; - -extern const struct address_space_operations xfs_address_space_operations; -extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int); - -extern void xfs_ioend_init(void); -extern void xfs_ioend_wait(struct xfs_inode *); - -extern void xfs_count_page_state(struct page *, int *, int *); - -#endif /* __XFS_AOPS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c deleted file mode 100644 index c57836dc778f..000000000000 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ /dev/null @@ -1,1876 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "xfs_sb.h" -#include "xfs_inum.h" -#include "xfs_log.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_trace.h" - -static kmem_zone_t *xfs_buf_zone; -STATIC int xfsbufd(void *); -STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int); - -static struct workqueue_struct *xfslogd_workqueue; -struct workqueue_struct *xfsdatad_workqueue; -struct workqueue_struct *xfsconvertd_workqueue; - -#ifdef XFS_BUF_LOCK_TRACKING -# define XB_SET_OWNER(bp) ((bp)->b_last_holder = current->pid) -# define XB_CLEAR_OWNER(bp) ((bp)->b_last_holder = -1) -# define XB_GET_OWNER(bp) ((bp)->b_last_holder) -#else -# define XB_SET_OWNER(bp) do { } while (0) -# define XB_CLEAR_OWNER(bp) do { } while (0) -# define XB_GET_OWNER(bp) do { } while (0) -#endif - -#define xb_to_gfp(flags) \ - ((((flags) & XBF_READ_AHEAD) ? __GFP_NORETRY : \ - ((flags) & XBF_DONT_BLOCK) ? GFP_NOFS : GFP_KERNEL) | __GFP_NOWARN) - -#define xb_to_km(flags) \ - (((flags) & XBF_DONT_BLOCK) ? KM_NOFS : KM_SLEEP) - -#define xfs_buf_allocate(flags) \ - kmem_zone_alloc(xfs_buf_zone, xb_to_km(flags)) -#define xfs_buf_deallocate(bp) \ - kmem_zone_free(xfs_buf_zone, (bp)); - -static inline int -xfs_buf_is_vmapped( - struct xfs_buf *bp) -{ - /* - * Return true if the buffer is vmapped. - * - * The XBF_MAPPED flag is set if the buffer should be mapped, but the - * code is clever enough to know it doesn't have to map a single page, - * so the check has to be both for XBF_MAPPED and bp->b_page_count > 1. - */ - return (bp->b_flags & XBF_MAPPED) && bp->b_page_count > 1; -} - -static inline int -xfs_buf_vmap_len( - struct xfs_buf *bp) -{ - return (bp->b_page_count * PAGE_SIZE) - bp->b_offset; -} - -/* - * xfs_buf_lru_add - add a buffer to the LRU. - * - * The LRU takes a new reference to the buffer so that it will only be freed - * once the shrinker takes the buffer off the LRU. - */ -STATIC void -xfs_buf_lru_add( - struct xfs_buf *bp) -{ - struct xfs_buftarg *btp = bp->b_target; - - spin_lock(&btp->bt_lru_lock); - if (list_empty(&bp->b_lru)) { - atomic_inc(&bp->b_hold); - list_add_tail(&bp->b_lru, &btp->bt_lru); - btp->bt_lru_nr++; - } - spin_unlock(&btp->bt_lru_lock); -} - -/* - * xfs_buf_lru_del - remove a buffer from the LRU - * - * The unlocked check is safe here because it only occurs when there are not - * b_lru_ref counts left on the inode under the pag->pag_buf_lock. it is there - * to optimise the shrinker removing the buffer from the LRU and calling - * xfs_buf_free(). i.e. it removes an unnecessary round trip on the - * bt_lru_lock. - */ -STATIC void -xfs_buf_lru_del( - struct xfs_buf *bp) -{ - struct xfs_buftarg *btp = bp->b_target; - - if (list_empty(&bp->b_lru)) - return; - - spin_lock(&btp->bt_lru_lock); - if (!list_empty(&bp->b_lru)) { - list_del_init(&bp->b_lru); - btp->bt_lru_nr--; - } - spin_unlock(&btp->bt_lru_lock); -} - -/* - * When we mark a buffer stale, we remove the buffer from the LRU and clear the - * b_lru_ref count so that the buffer is freed immediately when the buffer - * reference count falls to zero. If the buffer is already on the LRU, we need - * to remove the reference that LRU holds on the buffer. - * - * This prevents build-up of stale buffers on the LRU. - */ -void -xfs_buf_stale( - struct xfs_buf *bp) -{ - bp->b_flags |= XBF_STALE; - atomic_set(&(bp)->b_lru_ref, 0); - if (!list_empty(&bp->b_lru)) { - struct xfs_buftarg *btp = bp->b_target; - - spin_lock(&btp->bt_lru_lock); - if (!list_empty(&bp->b_lru)) { - list_del_init(&bp->b_lru); - btp->bt_lru_nr--; - atomic_dec(&bp->b_hold); - } - spin_unlock(&btp->bt_lru_lock); - } - ASSERT(atomic_read(&bp->b_hold) >= 1); -} - -STATIC void -_xfs_buf_initialize( - xfs_buf_t *bp, - xfs_buftarg_t *target, - xfs_off_t range_base, - size_t range_length, - xfs_buf_flags_t flags) -{ - /* - * We don't want certain flags to appear in b_flags. - */ - flags &= ~(XBF_LOCK|XBF_MAPPED|XBF_DONT_BLOCK|XBF_READ_AHEAD); - - memset(bp, 0, sizeof(xfs_buf_t)); - atomic_set(&bp->b_hold, 1); - atomic_set(&bp->b_lru_ref, 1); - init_completion(&bp->b_iowait); - INIT_LIST_HEAD(&bp->b_lru); - INIT_LIST_HEAD(&bp->b_list); - RB_CLEAR_NODE(&bp->b_rbnode); - sema_init(&bp->b_sema, 0); /* held, no waiters */ - XB_SET_OWNER(bp); - bp->b_target = target; - bp->b_file_offset = range_base; - /* - * Set buffer_length and count_desired to the same value initially. - * I/O routines should use count_desired, which will be the same in - * most cases but may be reset (e.g. XFS recovery). - */ - bp->b_buffer_length = bp->b_count_desired = range_length; - bp->b_flags = flags; - bp->b_bn = XFS_BUF_DADDR_NULL; - atomic_set(&bp->b_pin_count, 0); - init_waitqueue_head(&bp->b_waiters); - - XFS_STATS_INC(xb_create); - - trace_xfs_buf_init(bp, _RET_IP_); -} - -/* - * Allocate a page array capable of holding a specified number - * of pages, and point the page buf at it. - */ -STATIC int -_xfs_buf_get_pages( - xfs_buf_t *bp, - int page_count, - xfs_buf_flags_t flags) -{ - /* Make sure that we have a page list */ - if (bp->b_pages == NULL) { - bp->b_offset = xfs_buf_poff(bp->b_file_offset); - bp->b_page_count = page_count; - if (page_count <= XB_PAGES) { - bp->b_pages = bp->b_page_array; - } else { - bp->b_pages = kmem_alloc(sizeof(struct page *) * - page_count, xb_to_km(flags)); - if (bp->b_pages == NULL) - return -ENOMEM; - } - memset(bp->b_pages, 0, sizeof(struct page *) * page_count); - } - return 0; -} - -/* - * Frees b_pages if it was allocated. - */ -STATIC void -_xfs_buf_free_pages( - xfs_buf_t *bp) -{ - if (bp->b_pages != bp->b_page_array) { - kmem_free(bp->b_pages); - bp->b_pages = NULL; - } -} - -/* - * Releases the specified buffer. - * - * The modification state of any associated pages is left unchanged. - * The buffer most not be on any hash - use xfs_buf_rele instead for - * hashed and refcounted buffers - */ -void -xfs_buf_free( - xfs_buf_t *bp) -{ - trace_xfs_buf_free(bp, _RET_IP_); - - ASSERT(list_empty(&bp->b_lru)); - - if (bp->b_flags & _XBF_PAGES) { - uint i; - - if (xfs_buf_is_vmapped(bp)) - vm_unmap_ram(bp->b_addr - bp->b_offset, - bp->b_page_count); - - for (i = 0; i < bp->b_page_count; i++) { - struct page *page = bp->b_pages[i]; - - __free_page(page); - } - } else if (bp->b_flags & _XBF_KMEM) - kmem_free(bp->b_addr); - _xfs_buf_free_pages(bp); - xfs_buf_deallocate(bp); -} - -/* - * Allocates all the pages for buffer in question and builds it's page list. - */ -STATIC int -xfs_buf_allocate_memory( - xfs_buf_t *bp, - uint flags) -{ - size_t size = bp->b_count_desired; - size_t nbytes, offset; - gfp_t gfp_mask = xb_to_gfp(flags); - unsigned short page_count, i; - xfs_off_t end; - int error; - - /* - * for buffers that are contained within a single page, just allocate - * the memory from the heap - there's no need for the complexity of - * page arrays to keep allocation down to order 0. - */ - if (bp->b_buffer_length < PAGE_SIZE) { - bp->b_addr = kmem_alloc(bp->b_buffer_length, xb_to_km(flags)); - if (!bp->b_addr) { - /* low memory - use alloc_page loop instead */ - goto use_alloc_page; - } - - if (((unsigned long)(bp->b_addr + bp->b_buffer_length - 1) & - PAGE_MASK) != - ((unsigned long)bp->b_addr & PAGE_MASK)) { - /* b_addr spans two pages - use alloc_page instead */ - kmem_free(bp->b_addr); - bp->b_addr = NULL; - goto use_alloc_page; - } - bp->b_offset = offset_in_page(bp->b_addr); - bp->b_pages = bp->b_page_array; - bp->b_pages[0] = virt_to_page(bp->b_addr); - bp->b_page_count = 1; - bp->b_flags |= XBF_MAPPED | _XBF_KMEM; - return 0; - } - -use_alloc_page: - end = bp->b_file_offset + bp->b_buffer_length; - page_count = xfs_buf_btoc(end) - xfs_buf_btoct(bp->b_file_offset); - error = _xfs_buf_get_pages(bp, page_count, flags); - if (unlikely(error)) - return error; - - offset = bp->b_offset; - bp->b_flags |= _XBF_PAGES; - - for (i = 0; i < bp->b_page_count; i++) { - struct page *page; - uint retries = 0; -retry: - page = alloc_page(gfp_mask); - if (unlikely(page == NULL)) { - if (flags & XBF_READ_AHEAD) { - bp->b_page_count = i; - error = ENOMEM; - goto out_free_pages; - } - - /* - * This could deadlock. - * - * But until all the XFS lowlevel code is revamped to - * handle buffer allocation failures we can't do much. - */ - if (!(++retries % 100)) - xfs_err(NULL, - "possible memory allocation deadlock in %s (mode:0x%x)", - __func__, gfp_mask); - - XFS_STATS_INC(xb_page_retries); - congestion_wait(BLK_RW_ASYNC, HZ/50); - goto retry; - } - - XFS_STATS_INC(xb_page_found); - - nbytes = min_t(size_t, size, PAGE_SIZE - offset); - size -= nbytes; - bp->b_pages[i] = page; - offset = 0; - } - return 0; - -out_free_pages: - for (i = 0; i < bp->b_page_count; i++) - __free_page(bp->b_pages[i]); - return error; -} - -/* - * Map buffer into kernel address-space if necessary. - */ -STATIC int -_xfs_buf_map_pages( - xfs_buf_t *bp, - uint flags) -{ - ASSERT(bp->b_flags & _XBF_PAGES); - if (bp->b_page_count == 1) { - /* A single page buffer is always mappable */ - bp->b_addr = page_address(bp->b_pages[0]) + bp->b_offset; - bp->b_flags |= XBF_MAPPED; - } else if (flags & XBF_MAPPED) { - int retried = 0; - - do { - bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count, - -1, PAGE_KERNEL); - if (bp->b_addr) - break; - vm_unmap_aliases(); - } while (retried++ <= 1); - - if (!bp->b_addr) - return -ENOMEM; - bp->b_addr += bp->b_offset; - bp->b_flags |= XBF_MAPPED; - } - - return 0; -} - -/* - * Finding and Reading Buffers - */ - -/* - * Look up, and creates if absent, a lockable buffer for - * a given range of an inode. The buffer is returned - * locked. If other overlapping buffers exist, they are - * released before the new buffer is created and locked, - * which may imply that this call will block until those buffers - * are unlocked. No I/O is implied by this call. - */ -xfs_buf_t * -_xfs_buf_find( - xfs_buftarg_t *btp, /* block device target */ - xfs_off_t ioff, /* starting offset of range */ - size_t isize, /* length of range */ - xfs_buf_flags_t flags, - xfs_buf_t *new_bp) -{ - xfs_off_t range_base; - size_t range_length; - struct xfs_perag *pag; - struct rb_node **rbp; - struct rb_node *parent; - xfs_buf_t *bp; - - range_base = (ioff << BBSHIFT); - range_length = (isize << BBSHIFT); - - /* Check for IOs smaller than the sector size / not sector aligned */ - ASSERT(!(range_length < (1 << btp->bt_sshift))); - ASSERT(!(range_base & (xfs_off_t)btp->bt_smask)); - - /* get tree root */ - pag = xfs_perag_get(btp->bt_mount, - xfs_daddr_to_agno(btp->bt_mount, ioff)); - - /* walk tree */ - spin_lock(&pag->pag_buf_lock); - rbp = &pag->pag_buf_tree.rb_node; - parent = NULL; - bp = NULL; - while (*rbp) { - parent = *rbp; - bp = rb_entry(parent, struct xfs_buf, b_rbnode); - - if (range_base < bp->b_file_offset) - rbp = &(*rbp)->rb_left; - else if (range_base > bp->b_file_offset) - rbp = &(*rbp)->rb_right; - else { - /* - * found a block offset match. If the range doesn't - * match, the only way this is allowed is if the buffer - * in the cache is stale and the transaction that made - * it stale has not yet committed. i.e. we are - * reallocating a busy extent. Skip this buffer and - * continue searching to the right for an exact match. - */ - if (bp->b_buffer_length != range_length) { - ASSERT(bp->b_flags & XBF_STALE); - rbp = &(*rbp)->rb_right; - continue; - } - atomic_inc(&bp->b_hold); - goto found; - } - } - - /* No match found */ - if (new_bp) { - _xfs_buf_initialize(new_bp, btp, range_base, - range_length, flags); - rb_link_node(&new_bp->b_rbnode, parent, rbp); - rb_insert_color(&new_bp->b_rbnode, &pag->pag_buf_tree); - /* the buffer keeps the perag reference until it is freed */ - new_bp->b_pag = pag; - spin_unlock(&pag->pag_buf_lock); - } else { - XFS_STATS_INC(xb_miss_locked); - spin_unlock(&pag->pag_buf_lock); - xfs_perag_put(pag); - } - return new_bp; - -found: - spin_unlock(&pag->pag_buf_lock); - xfs_perag_put(pag); - - if (!xfs_buf_trylock(bp)) { - if (flags & XBF_TRYLOCK) { - xfs_buf_rele(bp); - XFS_STATS_INC(xb_busy_locked); - return NULL; - } - xfs_buf_lock(bp); - XFS_STATS_INC(xb_get_locked_waited); - } - - /* - * if the buffer is stale, clear all the external state associated with - * it. We need to keep flags such as how we allocated the buffer memory - * intact here. - */ - if (bp->b_flags & XBF_STALE) { - ASSERT((bp->b_flags & _XBF_DELWRI_Q) == 0); - bp->b_flags &= XBF_MAPPED | _XBF_KMEM | _XBF_PAGES; - } - - trace_xfs_buf_find(bp, flags, _RET_IP_); - XFS_STATS_INC(xb_get_locked); - return bp; -} - -/* - * Assembles a buffer covering the specified range. - * Storage in memory for all portions of the buffer will be allocated, - * although backing storage may not be. - */ -xfs_buf_t * -xfs_buf_get( - xfs_buftarg_t *target,/* target for buffer */ - xfs_off_t ioff, /* starting offset of range */ - size_t isize, /* length of range */ - xfs_buf_flags_t flags) -{ - xfs_buf_t *bp, *new_bp; - int error = 0; - - new_bp = xfs_buf_allocate(flags); - if (unlikely(!new_bp)) - return NULL; - - bp = _xfs_buf_find(target, ioff, isize, flags, new_bp); - if (bp == new_bp) { - error = xfs_buf_allocate_memory(bp, flags); - if (error) - goto no_buffer; - } else { - xfs_buf_deallocate(new_bp); - if (unlikely(bp == NULL)) - return NULL; - } - - if (!(bp->b_flags & XBF_MAPPED)) { - error = _xfs_buf_map_pages(bp, flags); - if (unlikely(error)) { - xfs_warn(target->bt_mount, - "%s: failed to map pages\n", __func__); - goto no_buffer; - } - } - - XFS_STATS_INC(xb_get); - - /* - * Always fill in the block number now, the mapped cases can do - * their own overlay of this later. - */ - bp->b_bn = ioff; - bp->b_count_desired = bp->b_buffer_length; - - trace_xfs_buf_get(bp, flags, _RET_IP_); - return bp; - - no_buffer: - if (flags & (XBF_LOCK | XBF_TRYLOCK)) - xfs_buf_unlock(bp); - xfs_buf_rele(bp); - return NULL; -} - -STATIC int -_xfs_buf_read( - xfs_buf_t *bp, - xfs_buf_flags_t flags) -{ - int status; - - ASSERT(!(flags & (XBF_DELWRI|XBF_WRITE))); - ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL); - - bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_DELWRI | XBF_READ_AHEAD); - bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | XBF_READ_AHEAD); - - status = xfs_buf_iorequest(bp); - if (status || bp->b_error || (flags & XBF_ASYNC)) - return status; - return xfs_buf_iowait(bp); -} - -xfs_buf_t * -xfs_buf_read( - xfs_buftarg_t *target, - xfs_off_t ioff, - size_t isize, - xfs_buf_flags_t flags) -{ - xfs_buf_t *bp; - - flags |= XBF_READ; - - bp = xfs_buf_get(target, ioff, isize, flags); - if (bp) { - trace_xfs_buf_read(bp, flags, _RET_IP_); - - if (!XFS_BUF_ISDONE(bp)) { - XFS_STATS_INC(xb_get_read); - _xfs_buf_read(bp, flags); - } else if (flags & XBF_ASYNC) { - /* - * Read ahead call which is already satisfied, - * drop the buffer - */ - goto no_buffer; - } else { - /* We do not want read in the flags */ - bp->b_flags &= ~XBF_READ; - } - } - - return bp; - - no_buffer: - if (flags & (XBF_LOCK | XBF_TRYLOCK)) - xfs_buf_unlock(bp); - xfs_buf_rele(bp); - return NULL; -} - -/* - * If we are not low on memory then do the readahead in a deadlock - * safe manner. - */ -void -xfs_buf_readahead( - xfs_buftarg_t *target, - xfs_off_t ioff, - size_t isize) -{ - if (bdi_read_congested(target->bt_bdi)) - return; - - xfs_buf_read(target, ioff, isize, - XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD|XBF_DONT_BLOCK); -} - -/* - * Read an uncached buffer from disk. Allocates and returns a locked - * buffer containing the disk contents or nothing. - */ -struct xfs_buf * -xfs_buf_read_uncached( - struct xfs_mount *mp, - struct xfs_buftarg *target, - xfs_daddr_t daddr, - size_t length, - int flags) -{ - xfs_buf_t *bp; - int error; - - bp = xfs_buf_get_uncached(target, length, flags); - if (!bp) - return NULL; - - /* set up the buffer for a read IO */ - XFS_BUF_SET_ADDR(bp, daddr); - XFS_BUF_READ(bp); - - xfsbdstrat(mp, bp); - error = xfs_buf_iowait(bp); - if (error || bp->b_error) { - xfs_buf_relse(bp); - return NULL; - } - return bp; -} - -xfs_buf_t * -xfs_buf_get_empty( - size_t len, - xfs_buftarg_t *target) -{ - xfs_buf_t *bp; - - bp = xfs_buf_allocate(0); - if (bp) - _xfs_buf_initialize(bp, target, 0, len, 0); - return bp; -} - -/* - * Return a buffer allocated as an empty buffer and associated to external - * memory via xfs_buf_associate_memory() back to it's empty state. - */ -void -xfs_buf_set_empty( - struct xfs_buf *bp, - size_t len) -{ - if (bp->b_pages) - _xfs_buf_free_pages(bp); - - bp->b_pages = NULL; - bp->b_page_count = 0; - bp->b_addr = NULL; - bp->b_file_offset = 0; - bp->b_buffer_length = bp->b_count_desired = len; - bp->b_bn = XFS_BUF_DADDR_NULL; - bp->b_flags &= ~XBF_MAPPED; -} - -static inline struct page * -mem_to_page( - void *addr) -{ - if ((!is_vmalloc_addr(addr))) { - return virt_to_page(addr); - } else { - return vmalloc_to_page(addr); - } -} - -int -xfs_buf_associate_memory( - xfs_buf_t *bp, - void *mem, - size_t len) -{ - int rval; - int i = 0; - unsigned long pageaddr; - unsigned long offset; - size_t buflen; - int page_count; - - pageaddr = (unsigned long)mem & PAGE_MASK; - offset = (unsigned long)mem - pageaddr; - buflen = PAGE_ALIGN(len + offset); - page_count = buflen >> PAGE_SHIFT; - - /* Free any previous set of page pointers */ - if (bp->b_pages) - _xfs_buf_free_pages(bp); - - bp->b_pages = NULL; - bp->b_addr = mem; - - rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK); - if (rval) - return rval; - - bp->b_offset = offset; - - for (i = 0; i < bp->b_page_count; i++) { - bp->b_pages[i] = mem_to_page((void *)pageaddr); - pageaddr += PAGE_SIZE; - } - - bp->b_count_desired = len; - bp->b_buffer_length = buflen; - bp->b_flags |= XBF_MAPPED; - - return 0; -} - -xfs_buf_t * -xfs_buf_get_uncached( - struct xfs_buftarg *target, - size_t len, - int flags) -{ - unsigned long page_count = PAGE_ALIGN(len) >> PAGE_SHIFT; - int error, i; - xfs_buf_t *bp; - - bp = xfs_buf_allocate(0); - if (unlikely(bp == NULL)) - goto fail; - _xfs_buf_initialize(bp, target, 0, len, 0); - - error = _xfs_buf_get_pages(bp, page_count, 0); - if (error) - goto fail_free_buf; - - for (i = 0; i < page_count; i++) { - bp->b_pages[i] = alloc_page(xb_to_gfp(flags)); - if (!bp->b_pages[i]) - goto fail_free_mem; - } - bp->b_flags |= _XBF_PAGES; - - error = _xfs_buf_map_pages(bp, XBF_MAPPED); - if (unlikely(error)) { - xfs_warn(target->bt_mount, - "%s: failed to map pages\n", __func__); - goto fail_free_mem; - } - - trace_xfs_buf_get_uncached(bp, _RET_IP_); - return bp; - - fail_free_mem: - while (--i >= 0) - __free_page(bp->b_pages[i]); - _xfs_buf_free_pages(bp); - fail_free_buf: - xfs_buf_deallocate(bp); - fail: - return NULL; -} - -/* - * Increment reference count on buffer, to hold the buffer concurrently - * with another thread which may release (free) the buffer asynchronously. - * Must hold the buffer already to call this function. - */ -void -xfs_buf_hold( - xfs_buf_t *bp) -{ - trace_xfs_buf_hold(bp, _RET_IP_); - atomic_inc(&bp->b_hold); -} - -/* - * Releases a hold on the specified buffer. If the - * the hold count is 1, calls xfs_buf_free. - */ -void -xfs_buf_rele( - xfs_buf_t *bp) -{ - struct xfs_perag *pag = bp->b_pag; - - trace_xfs_buf_rele(bp, _RET_IP_); - - if (!pag) { - ASSERT(list_empty(&bp->b_lru)); - ASSERT(RB_EMPTY_NODE(&bp->b_rbnode)); - if (atomic_dec_and_test(&bp->b_hold)) - xfs_buf_free(bp); - return; - } - - ASSERT(!RB_EMPTY_NODE(&bp->b_rbnode)); - - ASSERT(atomic_read(&bp->b_hold) > 0); - if (atomic_dec_and_lock(&bp->b_hold, &pag->pag_buf_lock)) { - if (!(bp->b_flags & XBF_STALE) && - atomic_read(&bp->b_lru_ref)) { - xfs_buf_lru_add(bp); - spin_unlock(&pag->pag_buf_lock); - } else { - xfs_buf_lru_del(bp); - ASSERT(!(bp->b_flags & (XBF_DELWRI|_XBF_DELWRI_Q))); - rb_erase(&bp->b_rbnode, &pag->pag_buf_tree); - spin_unlock(&pag->pag_buf_lock); - xfs_perag_put(pag); - xfs_buf_free(bp); - } - } -} - - -/* - * Lock a buffer object, if it is not already locked. - * - * If we come across a stale, pinned, locked buffer, we know that we are - * being asked to lock a buffer that has been reallocated. Because it is - * pinned, we know that the log has not been pushed to disk and hence it - * will still be locked. Rather than continuing to have trylock attempts - * fail until someone else pushes the log, push it ourselves before - * returning. This means that the xfsaild will not get stuck trying - * to push on stale inode buffers. - */ -int -xfs_buf_trylock( - struct xfs_buf *bp) -{ - int locked; - - locked = down_trylock(&bp->b_sema) == 0; - if (locked) - XB_SET_OWNER(bp); - else if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) - xfs_log_force(bp->b_target->bt_mount, 0); - - trace_xfs_buf_trylock(bp, _RET_IP_); - return locked; -} - -/* - * Lock a buffer object. - * - * If we come across a stale, pinned, locked buffer, we know that we - * are being asked to lock a buffer that has been reallocated. Because - * it is pinned, we know that the log has not been pushed to disk and - * hence it will still be locked. Rather than sleeping until someone - * else pushes the log, push it ourselves before trying to get the lock. - */ -void -xfs_buf_lock( - struct xfs_buf *bp) -{ - trace_xfs_buf_lock(bp, _RET_IP_); - - if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) - xfs_log_force(bp->b_target->bt_mount, 0); - down(&bp->b_sema); - XB_SET_OWNER(bp); - - trace_xfs_buf_lock_done(bp, _RET_IP_); -} - -/* - * Releases the lock on the buffer object. - * If the buffer is marked delwri but is not queued, do so before we - * unlock the buffer as we need to set flags correctly. We also need to - * take a reference for the delwri queue because the unlocker is going to - * drop their's and they don't know we just queued it. - */ -void -xfs_buf_unlock( - struct xfs_buf *bp) -{ - if ((bp->b_flags & (XBF_DELWRI|_XBF_DELWRI_Q)) == XBF_DELWRI) { - atomic_inc(&bp->b_hold); - bp->b_flags |= XBF_ASYNC; - xfs_buf_delwri_queue(bp, 0); - } - - XB_CLEAR_OWNER(bp); - up(&bp->b_sema); - - trace_xfs_buf_unlock(bp, _RET_IP_); -} - -STATIC void -xfs_buf_wait_unpin( - xfs_buf_t *bp) -{ - DECLARE_WAITQUEUE (wait, current); - - if (atomic_read(&bp->b_pin_count) == 0) - return; - - add_wait_queue(&bp->b_waiters, &wait); - for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); - if (atomic_read(&bp->b_pin_count) == 0) - break; - io_schedule(); - } - remove_wait_queue(&bp->b_waiters, &wait); - set_current_state(TASK_RUNNING); -} - -/* - * Buffer Utility Routines - */ - -STATIC void -xfs_buf_iodone_work( - struct work_struct *work) -{ - xfs_buf_t *bp = - container_of(work, xfs_buf_t, b_iodone_work); - - if (bp->b_iodone) - (*(bp->b_iodone))(bp); - else if (bp->b_flags & XBF_ASYNC) - xfs_buf_relse(bp); -} - -void -xfs_buf_ioend( - xfs_buf_t *bp, - int schedule) -{ - trace_xfs_buf_iodone(bp, _RET_IP_); - - bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD); - if (bp->b_error == 0) - bp->b_flags |= XBF_DONE; - - if ((bp->b_iodone) || (bp->b_flags & XBF_ASYNC)) { - if (schedule) { - INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work); - queue_work(xfslogd_workqueue, &bp->b_iodone_work); - } else { - xfs_buf_iodone_work(&bp->b_iodone_work); - } - } else { - complete(&bp->b_iowait); - } -} - -void -xfs_buf_ioerror( - xfs_buf_t *bp, - int error) -{ - ASSERT(error >= 0 && error <= 0xffff); - bp->b_error = (unsigned short)error; - trace_xfs_buf_ioerror(bp, error, _RET_IP_); -} - -int -xfs_bwrite( - struct xfs_mount *mp, - struct xfs_buf *bp) -{ - int error; - - bp->b_flags |= XBF_WRITE; - bp->b_flags &= ~(XBF_ASYNC | XBF_READ); - - xfs_buf_delwri_dequeue(bp); - xfs_bdstrat_cb(bp); - - error = xfs_buf_iowait(bp); - if (error) - xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR); - xfs_buf_relse(bp); - return error; -} - -void -xfs_bdwrite( - void *mp, - struct xfs_buf *bp) -{ - trace_xfs_buf_bdwrite(bp, _RET_IP_); - - bp->b_flags &= ~XBF_READ; - bp->b_flags |= (XBF_DELWRI | XBF_ASYNC); - - xfs_buf_delwri_queue(bp, 1); -} - -/* - * Called when we want to stop a buffer from getting written or read. - * We attach the EIO error, muck with its flags, and call xfs_buf_ioend - * so that the proper iodone callbacks get called. - */ -STATIC int -xfs_bioerror( - xfs_buf_t *bp) -{ -#ifdef XFSERRORDEBUG - ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone); -#endif - - /* - * No need to wait until the buffer is unpinned, we aren't flushing it. - */ - xfs_buf_ioerror(bp, EIO); - - /* - * We're calling xfs_buf_ioend, so delete XBF_DONE flag. - */ - XFS_BUF_UNREAD(bp); - XFS_BUF_UNDELAYWRITE(bp); - XFS_BUF_UNDONE(bp); - XFS_BUF_STALE(bp); - - xfs_buf_ioend(bp, 0); - - return EIO; -} - -/* - * Same as xfs_bioerror, except that we are releasing the buffer - * here ourselves, and avoiding the xfs_buf_ioend call. - * This is meant for userdata errors; metadata bufs come with - * iodone functions attached, so that we can track down errors. - */ -STATIC int -xfs_bioerror_relse( - struct xfs_buf *bp) -{ - int64_t fl = bp->b_flags; - /* - * No need to wait until the buffer is unpinned. - * We aren't flushing it. - * - * chunkhold expects B_DONE to be set, whether - * we actually finish the I/O or not. We don't want to - * change that interface. - */ - XFS_BUF_UNREAD(bp); - XFS_BUF_UNDELAYWRITE(bp); - XFS_BUF_DONE(bp); - XFS_BUF_STALE(bp); - bp->b_iodone = NULL; - if (!(fl & XBF_ASYNC)) { - /* - * Mark b_error and B_ERROR _both_. - * Lot's of chunkcache code assumes that. - * There's no reason to mark error for - * ASYNC buffers. - */ - xfs_buf_ioerror(bp, EIO); - XFS_BUF_FINISH_IOWAIT(bp); - } else { - xfs_buf_relse(bp); - } - - return EIO; -} - - -/* - * All xfs metadata buffers except log state machine buffers - * get this attached as their b_bdstrat callback function. - * This is so that we can catch a buffer - * after prematurely unpinning it to forcibly shutdown the filesystem. - */ -int -xfs_bdstrat_cb( - struct xfs_buf *bp) -{ - if (XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) { - trace_xfs_bdstrat_shut(bp, _RET_IP_); - /* - * Metadata write that didn't get logged but - * written delayed anyway. These aren't associated - * with a transaction, and can be ignored. - */ - if (!bp->b_iodone && !XFS_BUF_ISREAD(bp)) - return xfs_bioerror_relse(bp); - else - return xfs_bioerror(bp); - } - - xfs_buf_iorequest(bp); - return 0; -} - -/* - * Wrapper around bdstrat so that we can stop data from going to disk in case - * we are shutting down the filesystem. Typically user data goes thru this - * path; one of the exceptions is the superblock. - */ -void -xfsbdstrat( - struct xfs_mount *mp, - struct xfs_buf *bp) -{ - if (XFS_FORCED_SHUTDOWN(mp)) { - trace_xfs_bdstrat_shut(bp, _RET_IP_); - xfs_bioerror_relse(bp); - return; - } - - xfs_buf_iorequest(bp); -} - -STATIC void -_xfs_buf_ioend( - xfs_buf_t *bp, - int schedule) -{ - if (atomic_dec_and_test(&bp->b_io_remaining) == 1) - xfs_buf_ioend(bp, schedule); -} - -STATIC void -xfs_buf_bio_end_io( - struct bio *bio, - int error) -{ - xfs_buf_t *bp = (xfs_buf_t *)bio->bi_private; - - xfs_buf_ioerror(bp, -error); - - if (!error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ)) - invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp)); - - _xfs_buf_ioend(bp, 1); - bio_put(bio); -} - -STATIC void -_xfs_buf_ioapply( - xfs_buf_t *bp) -{ - int rw, map_i, total_nr_pages, nr_pages; - struct bio *bio; - int offset = bp->b_offset; - int size = bp->b_count_desired; - sector_t sector = bp->b_bn; - - total_nr_pages = bp->b_page_count; - map_i = 0; - - if (bp->b_flags & XBF_WRITE) { - if (bp->b_flags & XBF_SYNCIO) - rw = WRITE_SYNC; - else - rw = WRITE; - if (bp->b_flags & XBF_FUA) - rw |= REQ_FUA; - if (bp->b_flags & XBF_FLUSH) - rw |= REQ_FLUSH; - } else if (bp->b_flags & XBF_READ_AHEAD) { - rw = READA; - } else { - rw = READ; - } - - /* we only use the buffer cache for meta-data */ - rw |= REQ_META; - -next_chunk: - atomic_inc(&bp->b_io_remaining); - nr_pages = BIO_MAX_SECTORS >> (PAGE_SHIFT - BBSHIFT); - if (nr_pages > total_nr_pages) - nr_pages = total_nr_pages; - - bio = bio_alloc(GFP_NOIO, nr_pages); - bio->bi_bdev = bp->b_target->bt_bdev; - bio->bi_sector = sector; - bio->bi_end_io = xfs_buf_bio_end_io; - bio->bi_private = bp; - - - for (; size && nr_pages; nr_pages--, map_i++) { - int rbytes, nbytes = PAGE_SIZE - offset; - - if (nbytes > size) - nbytes = size; - - rbytes = bio_add_page(bio, bp->b_pages[map_i], nbytes, offset); - if (rbytes < nbytes) - break; - - offset = 0; - sector += nbytes >> BBSHIFT; - size -= nbytes; - total_nr_pages--; - } - - if (likely(bio->bi_size)) { - if (xfs_buf_is_vmapped(bp)) { - flush_kernel_vmap_range(bp->b_addr, - xfs_buf_vmap_len(bp)); - } - submit_bio(rw, bio); - if (size) - goto next_chunk; - } else { - xfs_buf_ioerror(bp, EIO); - bio_put(bio); - } -} - -int -xfs_buf_iorequest( - xfs_buf_t *bp) -{ - trace_xfs_buf_iorequest(bp, _RET_IP_); - - if (bp->b_flags & XBF_DELWRI) { - xfs_buf_delwri_queue(bp, 1); - return 0; - } - - if (bp->b_flags & XBF_WRITE) { - xfs_buf_wait_unpin(bp); - } - - xfs_buf_hold(bp); - - /* Set the count to 1 initially, this will stop an I/O - * completion callout which happens before we have started - * all the I/O from calling xfs_buf_ioend too early. - */ - atomic_set(&bp->b_io_remaining, 1); - _xfs_buf_ioapply(bp); - _xfs_buf_ioend(bp, 0); - - xfs_buf_rele(bp); - return 0; -} - -/* - * Waits for I/O to complete on the buffer supplied. - * It returns immediately if no I/O is pending. - * It returns the I/O error code, if any, or 0 if there was no error. - */ -int -xfs_buf_iowait( - xfs_buf_t *bp) -{ - trace_xfs_buf_iowait(bp, _RET_IP_); - - wait_for_completion(&bp->b_iowait); - - trace_xfs_buf_iowait_done(bp, _RET_IP_); - return bp->b_error; -} - -xfs_caddr_t -xfs_buf_offset( - xfs_buf_t *bp, - size_t offset) -{ - struct page *page; - - if (bp->b_flags & XBF_MAPPED) - return bp->b_addr + offset; - - offset += bp->b_offset; - page = bp->b_pages[offset >> PAGE_SHIFT]; - return (xfs_caddr_t)page_address(page) + (offset & (PAGE_SIZE-1)); -} - -/* - * Move data into or out of a buffer. - */ -void -xfs_buf_iomove( - xfs_buf_t *bp, /* buffer to process */ - size_t boff, /* starting buffer offset */ - size_t bsize, /* length to copy */ - void *data, /* data address */ - xfs_buf_rw_t mode) /* read/write/zero flag */ -{ - size_t bend, cpoff, csize; - struct page *page; - - bend = boff + bsize; - while (boff < bend) { - page = bp->b_pages[xfs_buf_btoct(boff + bp->b_offset)]; - cpoff = xfs_buf_poff(boff + bp->b_offset); - csize = min_t(size_t, - PAGE_SIZE-cpoff, bp->b_count_desired-boff); - - ASSERT(((csize + cpoff) <= PAGE_SIZE)); - - switch (mode) { - case XBRW_ZERO: - memset(page_address(page) + cpoff, 0, csize); - break; - case XBRW_READ: - memcpy(data, page_address(page) + cpoff, csize); - break; - case XBRW_WRITE: - memcpy(page_address(page) + cpoff, data, csize); - } - - boff += csize; - data += csize; - } -} - -/* - * Handling of buffer targets (buftargs). - */ - -/* - * Wait for any bufs with callbacks that have been submitted but have not yet - * returned. These buffers will have an elevated hold count, so wait on those - * while freeing all the buffers only held by the LRU. - */ -void -xfs_wait_buftarg( - struct xfs_buftarg *btp) -{ - struct xfs_buf *bp; - -restart: - spin_lock(&btp->bt_lru_lock); - while (!list_empty(&btp->bt_lru)) { - bp = list_first_entry(&btp->bt_lru, struct xfs_buf, b_lru); - if (atomic_read(&bp->b_hold) > 1) { - spin_unlock(&btp->bt_lru_lock); - delay(100); - goto restart; - } - /* - * clear the LRU reference count so the bufer doesn't get - * ignored in xfs_buf_rele(). - */ - atomic_set(&bp->b_lru_ref, 0); - spin_unlock(&btp->bt_lru_lock); - xfs_buf_rele(bp); - spin_lock(&btp->bt_lru_lock); - } - spin_unlock(&btp->bt_lru_lock); -} - -int -xfs_buftarg_shrink( - struct shrinker *shrink, - struct shrink_control *sc) -{ - struct xfs_buftarg *btp = container_of(shrink, - struct xfs_buftarg, bt_shrinker); - struct xfs_buf *bp; - int nr_to_scan = sc->nr_to_scan; - LIST_HEAD(dispose); - - if (!nr_to_scan) - return btp->bt_lru_nr; - - spin_lock(&btp->bt_lru_lock); - while (!list_empty(&btp->bt_lru)) { - if (nr_to_scan-- <= 0) - break; - - bp = list_first_entry(&btp->bt_lru, struct xfs_buf, b_lru); - - /* - * Decrement the b_lru_ref count unless the value is already - * zero. If the value is already zero, we need to reclaim the - * buffer, otherwise it gets another trip through the LRU. - */ - if (!atomic_add_unless(&bp->b_lru_ref, -1, 0)) { - list_move_tail(&bp->b_lru, &btp->bt_lru); - continue; - } - - /* - * remove the buffer from the LRU now to avoid needing another - * lock round trip inside xfs_buf_rele(). - */ - list_move(&bp->b_lru, &dispose); - btp->bt_lru_nr--; - } - spin_unlock(&btp->bt_lru_lock); - - while (!list_empty(&dispose)) { - bp = list_first_entry(&dispose, struct xfs_buf, b_lru); - list_del_init(&bp->b_lru); - xfs_buf_rele(bp); - } - - return btp->bt_lru_nr; -} - -void -xfs_free_buftarg( - struct xfs_mount *mp, - struct xfs_buftarg *btp) -{ - unregister_shrinker(&btp->bt_shrinker); - - xfs_flush_buftarg(btp, 1); - if (mp->m_flags & XFS_MOUNT_BARRIER) - xfs_blkdev_issue_flush(btp); - - kthread_stop(btp->bt_task); - kmem_free(btp); -} - -STATIC int -xfs_setsize_buftarg_flags( - xfs_buftarg_t *btp, - unsigned int blocksize, - unsigned int sectorsize, - int verbose) -{ - btp->bt_bsize = blocksize; - btp->bt_sshift = ffs(sectorsize) - 1; - btp->bt_smask = sectorsize - 1; - - if (set_blocksize(btp->bt_bdev, sectorsize)) { - xfs_warn(btp->bt_mount, - "Cannot set_blocksize to %u on device %s\n", - sectorsize, xfs_buf_target_name(btp)); - return EINVAL; - } - - return 0; -} - -/* - * When allocating the initial buffer target we have not yet - * read in the superblock, so don't know what sized sectors - * are being used is at this early stage. Play safe. - */ -STATIC int -xfs_setsize_buftarg_early( - xfs_buftarg_t *btp, - struct block_device *bdev) -{ - return xfs_setsize_buftarg_flags(btp, - PAGE_SIZE, bdev_logical_block_size(bdev), 0); -} - -int -xfs_setsize_buftarg( - xfs_buftarg_t *btp, - unsigned int blocksize, - unsigned int sectorsize) -{ - return xfs_setsize_buftarg_flags(btp, blocksize, sectorsize, 1); -} - -STATIC int -xfs_alloc_delwrite_queue( - xfs_buftarg_t *btp, - const char *fsname) -{ - INIT_LIST_HEAD(&btp->bt_delwrite_queue); - spin_lock_init(&btp->bt_delwrite_lock); - btp->bt_flags = 0; - btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname); - if (IS_ERR(btp->bt_task)) - return PTR_ERR(btp->bt_task); - return 0; -} - -xfs_buftarg_t * -xfs_alloc_buftarg( - struct xfs_mount *mp, - struct block_device *bdev, - int external, - const char *fsname) -{ - xfs_buftarg_t *btp; - - btp = kmem_zalloc(sizeof(*btp), KM_SLEEP); - - btp->bt_mount = mp; - btp->bt_dev = bdev->bd_dev; - btp->bt_bdev = bdev; - btp->bt_bdi = blk_get_backing_dev_info(bdev); - if (!btp->bt_bdi) - goto error; - - INIT_LIST_HEAD(&btp->bt_lru); - spin_lock_init(&btp->bt_lru_lock); - if (xfs_setsize_buftarg_early(btp, bdev)) - goto error; - if (xfs_alloc_delwrite_queue(btp, fsname)) - goto error; - btp->bt_shrinker.shrink = xfs_buftarg_shrink; - btp->bt_shrinker.seeks = DEFAULT_SEEKS; - register_shrinker(&btp->bt_shrinker); - return btp; - -error: - kmem_free(btp); - return NULL; -} - - -/* - * Delayed write buffer handling - */ -STATIC void -xfs_buf_delwri_queue( - xfs_buf_t *bp, - int unlock) -{ - struct list_head *dwq = &bp->b_target->bt_delwrite_queue; - spinlock_t *dwlk = &bp->b_target->bt_delwrite_lock; - - trace_xfs_buf_delwri_queue(bp, _RET_IP_); - - ASSERT((bp->b_flags&(XBF_DELWRI|XBF_ASYNC)) == (XBF_DELWRI|XBF_ASYNC)); - - spin_lock(dwlk); - /* If already in the queue, dequeue and place at tail */ - if (!list_empty(&bp->b_list)) { - ASSERT(bp->b_flags & _XBF_DELWRI_Q); - if (unlock) - atomic_dec(&bp->b_hold); - list_del(&bp->b_list); - } - - if (list_empty(dwq)) { - /* start xfsbufd as it is about to have something to do */ - wake_up_process(bp->b_target->bt_task); - } - - bp->b_flags |= _XBF_DELWRI_Q; - list_add_tail(&bp->b_list, dwq); - bp->b_queuetime = jiffies; - spin_unlock(dwlk); - - if (unlock) - xfs_buf_unlock(bp); -} - -void -xfs_buf_delwri_dequeue( - xfs_buf_t *bp) -{ - spinlock_t *dwlk = &bp->b_target->bt_delwrite_lock; - int dequeued = 0; - - spin_lock(dwlk); - if ((bp->b_flags & XBF_DELWRI) && !list_empty(&bp->b_list)) { - ASSERT(bp->b_flags & _XBF_DELWRI_Q); - list_del_init(&bp->b_list); - dequeued = 1; - } - bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q); - spin_unlock(dwlk); - - if (dequeued) - xfs_buf_rele(bp); - - trace_xfs_buf_delwri_dequeue(bp, _RET_IP_); -} - -/* - * If a delwri buffer needs to be pushed before it has aged out, then promote - * it to the head of the delwri queue so that it will be flushed on the next - * xfsbufd run. We do this by resetting the queuetime of the buffer to be older - * than the age currently needed to flush the buffer. Hence the next time the - * xfsbufd sees it is guaranteed to be considered old enough to flush. - */ -void -xfs_buf_delwri_promote( - struct xfs_buf *bp) -{ - struct xfs_buftarg *btp = bp->b_target; - long age = xfs_buf_age_centisecs * msecs_to_jiffies(10) + 1; - - ASSERT(bp->b_flags & XBF_DELWRI); - ASSERT(bp->b_flags & _XBF_DELWRI_Q); - - /* - * Check the buffer age before locking the delayed write queue as we - * don't need to promote buffers that are already past the flush age. - */ - if (bp->b_queuetime < jiffies - age) - return; - bp->b_queuetime = jiffies - age; - spin_lock(&btp->bt_delwrite_lock); - list_move(&bp->b_list, &btp->bt_delwrite_queue); - spin_unlock(&btp->bt_delwrite_lock); -} - -STATIC void -xfs_buf_runall_queues( - struct workqueue_struct *queue) -{ - flush_workqueue(queue); -} - -/* - * Move as many buffers as specified to the supplied list - * idicating if we skipped any buffers to prevent deadlocks. - */ -STATIC int -xfs_buf_delwri_split( - xfs_buftarg_t *target, - struct list_head *list, - unsigned long age) -{ - xfs_buf_t *bp, *n; - struct list_head *dwq = &target->bt_delwrite_queue; - spinlock_t *dwlk = &target->bt_delwrite_lock; - int skipped = 0; - int force; - - force = test_and_clear_bit(XBT_FORCE_FLUSH, &target->bt_flags); - INIT_LIST_HEAD(list); - spin_lock(dwlk); - list_for_each_entry_safe(bp, n, dwq, b_list) { - ASSERT(bp->b_flags & XBF_DELWRI); - - if (!xfs_buf_ispinned(bp) && xfs_buf_trylock(bp)) { - if (!force && - time_before(jiffies, bp->b_queuetime + age)) { - xfs_buf_unlock(bp); - break; - } - - bp->b_flags &= ~(XBF_DELWRI | _XBF_DELWRI_Q); - bp->b_flags |= XBF_WRITE; - list_move_tail(&bp->b_list, list); - trace_xfs_buf_delwri_split(bp, _RET_IP_); - } else - skipped++; - } - spin_unlock(dwlk); - - return skipped; - -} - -/* - * Compare function is more complex than it needs to be because - * the return value is only 32 bits and we are doing comparisons - * on 64 bit values - */ -static int -xfs_buf_cmp( - void *priv, - struct list_head *a, - struct list_head *b) -{ - struct xfs_buf *ap = container_of(a, struct xfs_buf, b_list); - struct xfs_buf *bp = container_of(b, struct xfs_buf, b_list); - xfs_daddr_t diff; - - diff = ap->b_bn - bp->b_bn; - if (diff < 0) - return -1; - if (diff > 0) - return 1; - return 0; -} - -STATIC int -xfsbufd( - void *data) -{ - xfs_buftarg_t *target = (xfs_buftarg_t *)data; - - current->flags |= PF_MEMALLOC; - - set_freezable(); - - do { - long age = xfs_buf_age_centisecs * msecs_to_jiffies(10); - long tout = xfs_buf_timer_centisecs * msecs_to_jiffies(10); - struct list_head tmp; - struct blk_plug plug; - - if (unlikely(freezing(current))) { - set_bit(XBT_FORCE_SLEEP, &target->bt_flags); - refrigerator(); - } else { - clear_bit(XBT_FORCE_SLEEP, &target->bt_flags); - } - - /* sleep for a long time if there is nothing to do. */ - if (list_empty(&target->bt_delwrite_queue)) - tout = MAX_SCHEDULE_TIMEOUT; - schedule_timeout_interruptible(tout); - - xfs_buf_delwri_split(target, &tmp, age); - list_sort(NULL, &tmp, xfs_buf_cmp); - - blk_start_plug(&plug); - while (!list_empty(&tmp)) { - struct xfs_buf *bp; - bp = list_first_entry(&tmp, struct xfs_buf, b_list); - list_del_init(&bp->b_list); - xfs_bdstrat_cb(bp); - } - blk_finish_plug(&plug); - } while (!kthread_should_stop()); - - return 0; -} - -/* - * Go through all incore buffers, and release buffers if they belong to - * the given device. This is used in filesystem error handling to - * preserve the consistency of its metadata. - */ -int -xfs_flush_buftarg( - xfs_buftarg_t *target, - int wait) -{ - xfs_buf_t *bp; - int pincount = 0; - LIST_HEAD(tmp_list); - LIST_HEAD(wait_list); - struct blk_plug plug; - - xfs_buf_runall_queues(xfsconvertd_workqueue); - xfs_buf_runall_queues(xfsdatad_workqueue); - xfs_buf_runall_queues(xfslogd_workqueue); - - set_bit(XBT_FORCE_FLUSH, &target->bt_flags); - pincount = xfs_buf_delwri_split(target, &tmp_list, 0); - - /* - * Dropped the delayed write list lock, now walk the temporary list. - * All I/O is issued async and then if we need to wait for completion - * we do that after issuing all the IO. - */ - list_sort(NULL, &tmp_list, xfs_buf_cmp); - - blk_start_plug(&plug); - while (!list_empty(&tmp_list)) { - bp = list_first_entry(&tmp_list, struct xfs_buf, b_list); - ASSERT(target == bp->b_target); - list_del_init(&bp->b_list); - if (wait) { - bp->b_flags &= ~XBF_ASYNC; - list_add(&bp->b_list, &wait_list); - } - xfs_bdstrat_cb(bp); - } - blk_finish_plug(&plug); - - if (wait) { - /* Wait for IO to complete. */ - while (!list_empty(&wait_list)) { - bp = list_first_entry(&wait_list, struct xfs_buf, b_list); - - list_del_init(&bp->b_list); - xfs_buf_iowait(bp); - xfs_buf_relse(bp); - } - } - - return pincount; -} - -int __init -xfs_buf_init(void) -{ - xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf", - KM_ZONE_HWALIGN, NULL); - if (!xfs_buf_zone) - goto out; - - xfslogd_workqueue = alloc_workqueue("xfslogd", - WQ_MEM_RECLAIM | WQ_HIGHPRI, 1); - if (!xfslogd_workqueue) - goto out_free_buf_zone; - - xfsdatad_workqueue = alloc_workqueue("xfsdatad", WQ_MEM_RECLAIM, 1); - if (!xfsdatad_workqueue) - goto out_destroy_xfslogd_workqueue; - - xfsconvertd_workqueue = alloc_workqueue("xfsconvertd", - WQ_MEM_RECLAIM, 1); - if (!xfsconvertd_workqueue) - goto out_destroy_xfsdatad_workqueue; - - return 0; - - out_destroy_xfsdatad_workqueue: - destroy_workqueue(xfsdatad_workqueue); - out_destroy_xfslogd_workqueue: - destroy_workqueue(xfslogd_workqueue); - out_free_buf_zone: - kmem_zone_destroy(xfs_buf_zone); - out: - return -ENOMEM; -} - -void -xfs_buf_terminate(void) -{ - destroy_workqueue(xfsconvertd_workqueue); - destroy_workqueue(xfsdatad_workqueue); - destroy_workqueue(xfslogd_workqueue); - kmem_zone_destroy(xfs_buf_zone); -} - -#ifdef CONFIG_KDB_MODULES -struct list_head * -xfs_get_buftarg_list(void) -{ - return &xfs_buftarg_list; -} -#endif diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h deleted file mode 100644 index 620972b8094d..000000000000 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ /dev/null @@ -1,326 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_BUF_H__ -#define __XFS_BUF_H__ - -#include -#include -#include -#include -#include -#include -#include -#include - -/* - * Base types - */ - -#define XFS_BUF_DADDR_NULL ((xfs_daddr_t) (-1LL)) - -#define xfs_buf_ctob(pp) ((pp) * PAGE_CACHE_SIZE) -#define xfs_buf_btoc(dd) (((dd) + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -#define xfs_buf_btoct(dd) ((dd) >> PAGE_CACHE_SHIFT) -#define xfs_buf_poff(aa) ((aa) & ~PAGE_CACHE_MASK) - -typedef enum { - XBRW_READ = 1, /* transfer into target memory */ - XBRW_WRITE = 2, /* transfer from target memory */ - XBRW_ZERO = 3, /* Zero target memory */ -} xfs_buf_rw_t; - -#define XBF_READ (1 << 0) /* buffer intended for reading from device */ -#define XBF_WRITE (1 << 1) /* buffer intended for writing to device */ -#define XBF_READ_AHEAD (1 << 2) /* asynchronous read-ahead */ -#define XBF_MAPPED (1 << 3) /* buffer mapped (b_addr valid) */ -#define XBF_ASYNC (1 << 4) /* initiator will not wait for completion */ -#define XBF_DONE (1 << 5) /* all pages in the buffer uptodate */ -#define XBF_DELWRI (1 << 6) /* buffer has dirty pages */ -#define XBF_STALE (1 << 7) /* buffer has been staled, do not find it */ - -/* I/O hints for the BIO layer */ -#define XBF_SYNCIO (1 << 10)/* treat this buffer as synchronous I/O */ -#define XBF_FUA (1 << 11)/* force cache write through mode */ -#define XBF_FLUSH (1 << 12)/* flush the disk cache before a write */ - -/* flags used only as arguments to access routines */ -#define XBF_LOCK (1 << 15)/* lock requested */ -#define XBF_TRYLOCK (1 << 16)/* lock requested, but do not wait */ -#define XBF_DONT_BLOCK (1 << 17)/* do not block in current thread */ - -/* flags used only internally */ -#define _XBF_PAGES (1 << 20)/* backed by refcounted pages */ -#define _XBF_KMEM (1 << 21)/* backed by heap memory */ -#define _XBF_DELWRI_Q (1 << 22)/* buffer on delwri queue */ - -typedef unsigned int xfs_buf_flags_t; - -#define XFS_BUF_FLAGS \ - { XBF_READ, "READ" }, \ - { XBF_WRITE, "WRITE" }, \ - { XBF_READ_AHEAD, "READ_AHEAD" }, \ - { XBF_MAPPED, "MAPPED" }, \ - { XBF_ASYNC, "ASYNC" }, \ - { XBF_DONE, "DONE" }, \ - { XBF_DELWRI, "DELWRI" }, \ - { XBF_STALE, "STALE" }, \ - { XBF_SYNCIO, "SYNCIO" }, \ - { XBF_FUA, "FUA" }, \ - { XBF_FLUSH, "FLUSH" }, \ - { XBF_LOCK, "LOCK" }, /* should never be set */\ - { XBF_TRYLOCK, "TRYLOCK" }, /* ditto */\ - { XBF_DONT_BLOCK, "DONT_BLOCK" }, /* ditto */\ - { _XBF_PAGES, "PAGES" }, \ - { _XBF_KMEM, "KMEM" }, \ - { _XBF_DELWRI_Q, "DELWRI_Q" } - -typedef enum { - XBT_FORCE_SLEEP = 0, - XBT_FORCE_FLUSH = 1, -} xfs_buftarg_flags_t; - -typedef struct xfs_buftarg { - dev_t bt_dev; - struct block_device *bt_bdev; - struct backing_dev_info *bt_bdi; - struct xfs_mount *bt_mount; - unsigned int bt_bsize; - unsigned int bt_sshift; - size_t bt_smask; - - /* per device delwri queue */ - struct task_struct *bt_task; - struct list_head bt_delwrite_queue; - spinlock_t bt_delwrite_lock; - unsigned long bt_flags; - - /* LRU control structures */ - struct shrinker bt_shrinker; - struct list_head bt_lru; - spinlock_t bt_lru_lock; - unsigned int bt_lru_nr; -} xfs_buftarg_t; - -struct xfs_buf; -typedef void (*xfs_buf_iodone_t)(struct xfs_buf *); - -#define XB_PAGES 2 - -typedef struct xfs_buf { - /* - * first cacheline holds all the fields needed for an uncontended cache - * hit to be fully processed. The semaphore straddles the cacheline - * boundary, but the counter and lock sits on the first cacheline, - * which is the only bit that is touched if we hit the semaphore - * fast-path on locking. - */ - struct rb_node b_rbnode; /* rbtree node */ - xfs_off_t b_file_offset; /* offset in file */ - size_t b_buffer_length;/* size of buffer in bytes */ - atomic_t b_hold; /* reference count */ - atomic_t b_lru_ref; /* lru reclaim ref count */ - xfs_buf_flags_t b_flags; /* status flags */ - struct semaphore b_sema; /* semaphore for lockables */ - - struct list_head b_lru; /* lru list */ - wait_queue_head_t b_waiters; /* unpin waiters */ - struct list_head b_list; - struct xfs_perag *b_pag; /* contains rbtree root */ - xfs_buftarg_t *b_target; /* buffer target (device) */ - xfs_daddr_t b_bn; /* block number for I/O */ - size_t b_count_desired;/* desired transfer size */ - void *b_addr; /* virtual address of buffer */ - struct work_struct b_iodone_work; - xfs_buf_iodone_t b_iodone; /* I/O completion function */ - struct completion b_iowait; /* queue for I/O waiters */ - void *b_fspriv; - struct xfs_trans *b_transp; - struct page **b_pages; /* array of page pointers */ - struct page *b_page_array[XB_PAGES]; /* inline pages */ - unsigned long b_queuetime; /* time buffer was queued */ - atomic_t b_pin_count; /* pin count */ - atomic_t b_io_remaining; /* #outstanding I/O requests */ - unsigned int b_page_count; /* size of page array */ - unsigned int b_offset; /* page offset in first page */ - unsigned short b_error; /* error code on I/O */ -#ifdef XFS_BUF_LOCK_TRACKING - int b_last_holder; -#endif -} xfs_buf_t; - - -/* Finding and Reading Buffers */ -extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t, - xfs_buf_flags_t, xfs_buf_t *); -#define xfs_incore(buftarg,blkno,len,lockit) \ - _xfs_buf_find(buftarg, blkno ,len, lockit, NULL) - -extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t, - xfs_buf_flags_t); -extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t, - xfs_buf_flags_t); - -extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *); -extern void xfs_buf_set_empty(struct xfs_buf *bp, size_t len); -extern xfs_buf_t *xfs_buf_get_uncached(struct xfs_buftarg *, size_t, int); -extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t); -extern void xfs_buf_hold(xfs_buf_t *); -extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t); -struct xfs_buf *xfs_buf_read_uncached(struct xfs_mount *mp, - struct xfs_buftarg *target, - xfs_daddr_t daddr, size_t length, int flags); - -/* Releasing Buffers */ -extern void xfs_buf_free(xfs_buf_t *); -extern void xfs_buf_rele(xfs_buf_t *); - -/* Locking and Unlocking Buffers */ -extern int xfs_buf_trylock(xfs_buf_t *); -extern void xfs_buf_lock(xfs_buf_t *); -extern void xfs_buf_unlock(xfs_buf_t *); -#define xfs_buf_islocked(bp) \ - ((bp)->b_sema.count <= 0) - -/* Buffer Read and Write Routines */ -extern int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp); -extern void xfs_bdwrite(void *mp, xfs_buf_t *bp); - -extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *); -extern int xfs_bdstrat_cb(struct xfs_buf *); - -extern void xfs_buf_ioend(xfs_buf_t *, int); -extern void xfs_buf_ioerror(xfs_buf_t *, int); -extern int xfs_buf_iorequest(xfs_buf_t *); -extern int xfs_buf_iowait(xfs_buf_t *); -extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *, - xfs_buf_rw_t); -#define xfs_buf_zero(bp, off, len) \ - xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO) - -static inline int xfs_buf_geterror(xfs_buf_t *bp) -{ - return bp ? bp->b_error : ENOMEM; -} - -/* Buffer Utility Routines */ -extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t); - -/* Delayed Write Buffer Routines */ -extern void xfs_buf_delwri_dequeue(xfs_buf_t *); -extern void xfs_buf_delwri_promote(xfs_buf_t *); - -/* Buffer Daemon Setup Routines */ -extern int xfs_buf_init(void); -extern void xfs_buf_terminate(void); - -static inline const char * -xfs_buf_target_name(struct xfs_buftarg *target) -{ - static char __b[BDEVNAME_SIZE]; - - return bdevname(target->bt_bdev, __b); -} - - -#define XFS_BUF_ZEROFLAGS(bp) \ - ((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI| \ - XBF_SYNCIO|XBF_FUA|XBF_FLUSH)) - -void xfs_buf_stale(struct xfs_buf *bp); -#define XFS_BUF_STALE(bp) xfs_buf_stale(bp); -#define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XBF_STALE) -#define XFS_BUF_ISSTALE(bp) ((bp)->b_flags & XBF_STALE) -#define XFS_BUF_SUPER_STALE(bp) do { \ - XFS_BUF_STALE(bp); \ - xfs_buf_delwri_dequeue(bp); \ - XFS_BUF_DONE(bp); \ - } while (0) - -#define XFS_BUF_DELAYWRITE(bp) ((bp)->b_flags |= XBF_DELWRI) -#define XFS_BUF_UNDELAYWRITE(bp) xfs_buf_delwri_dequeue(bp) -#define XFS_BUF_ISDELAYWRITE(bp) ((bp)->b_flags & XBF_DELWRI) - -#define XFS_BUF_DONE(bp) ((bp)->b_flags |= XBF_DONE) -#define XFS_BUF_UNDONE(bp) ((bp)->b_flags &= ~XBF_DONE) -#define XFS_BUF_ISDONE(bp) ((bp)->b_flags & XBF_DONE) - -#define XFS_BUF_ASYNC(bp) ((bp)->b_flags |= XBF_ASYNC) -#define XFS_BUF_UNASYNC(bp) ((bp)->b_flags &= ~XBF_ASYNC) -#define XFS_BUF_ISASYNC(bp) ((bp)->b_flags & XBF_ASYNC) - -#define XFS_BUF_READ(bp) ((bp)->b_flags |= XBF_READ) -#define XFS_BUF_UNREAD(bp) ((bp)->b_flags &= ~XBF_READ) -#define XFS_BUF_ISREAD(bp) ((bp)->b_flags & XBF_READ) - -#define XFS_BUF_WRITE(bp) ((bp)->b_flags |= XBF_WRITE) -#define XFS_BUF_UNWRITE(bp) ((bp)->b_flags &= ~XBF_WRITE) -#define XFS_BUF_ISWRITE(bp) ((bp)->b_flags & XBF_WRITE) - -#define XFS_BUF_ADDR(bp) ((bp)->b_bn) -#define XFS_BUF_SET_ADDR(bp, bno) ((bp)->b_bn = (xfs_daddr_t)(bno)) -#define XFS_BUF_OFFSET(bp) ((bp)->b_file_offset) -#define XFS_BUF_SET_OFFSET(bp, off) ((bp)->b_file_offset = (off)) -#define XFS_BUF_COUNT(bp) ((bp)->b_count_desired) -#define XFS_BUF_SET_COUNT(bp, cnt) ((bp)->b_count_desired = (cnt)) -#define XFS_BUF_SIZE(bp) ((bp)->b_buffer_length) -#define XFS_BUF_SET_SIZE(bp, cnt) ((bp)->b_buffer_length = (cnt)) - -static inline void -xfs_buf_set_ref( - struct xfs_buf *bp, - int lru_ref) -{ - atomic_set(&bp->b_lru_ref, lru_ref); -} -#define XFS_BUF_SET_VTYPE_REF(bp, type, ref) xfs_buf_set_ref(bp, ref) -#define XFS_BUF_SET_VTYPE(bp, type) do { } while (0) - -static inline int xfs_buf_ispinned(struct xfs_buf *bp) -{ - return atomic_read(&bp->b_pin_count); -} - -#define XFS_BUF_FINISH_IOWAIT(bp) complete(&bp->b_iowait); - -static inline void xfs_buf_relse(xfs_buf_t *bp) -{ - xfs_buf_unlock(bp); - xfs_buf_rele(bp); -} - -/* - * Handling of buftargs. - */ -extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *, - struct block_device *, int, const char *); -extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *); -extern void xfs_wait_buftarg(xfs_buftarg_t *); -extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int); -extern int xfs_flush_buftarg(xfs_buftarg_t *, int); - -#ifdef CONFIG_KDB_MODULES -extern struct list_head *xfs_get_buftarg_list(void); -#endif - -#define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev) -#define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev) - -#define xfs_binval(buftarg) xfs_flush_buftarg(buftarg, 1) -#define XFS_bflush(buftarg) xfs_flush_buftarg(buftarg, 1) - -#endif /* __XFS_BUF_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_discard.c b/fs/xfs/linux-2.6/xfs_discard.c deleted file mode 100644 index 244e797dae32..000000000000 --- a/fs/xfs/linux-2.6/xfs_discard.c +++ /dev/null @@ -1,222 +0,0 @@ -/* - * Copyright (C) 2010 Red Hat, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_sb.h" -#include "xfs_inum.h" -#include "xfs_log.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_quota.h" -#include "xfs_trans.h" -#include "xfs_alloc_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_inode.h" -#include "xfs_alloc.h" -#include "xfs_error.h" -#include "xfs_discard.h" -#include "xfs_trace.h" - -STATIC int -xfs_trim_extents( - struct xfs_mount *mp, - xfs_agnumber_t agno, - xfs_fsblock_t start, - xfs_fsblock_t len, - xfs_fsblock_t minlen, - __uint64_t *blocks_trimmed) -{ - struct block_device *bdev = mp->m_ddev_targp->bt_bdev; - struct xfs_btree_cur *cur; - struct xfs_buf *agbp; - struct xfs_perag *pag; - int error; - int i; - - pag = xfs_perag_get(mp, agno); - - error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp); - if (error || !agbp) - goto out_put_perag; - - cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT); - - /* - * Force out the log. This means any transactions that might have freed - * space before we took the AGF buffer lock are now on disk, and the - * volatile disk cache is flushed. - */ - xfs_log_force(mp, XFS_LOG_SYNC); - - /* - * Look up the longest btree in the AGF and start with it. - */ - error = xfs_alloc_lookup_le(cur, 0, - XFS_BUF_TO_AGF(agbp)->agf_longest, &i); - if (error) - goto out_del_cursor; - - /* - * Loop until we are done with all extents that are large - * enough to be worth discarding. - */ - while (i) { - xfs_agblock_t fbno; - xfs_extlen_t flen; - - error = xfs_alloc_get_rec(cur, &fbno, &flen, &i); - if (error) - goto out_del_cursor; - XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor); - ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest); - - /* - * Too small? Give up. - */ - if (flen < minlen) { - trace_xfs_discard_toosmall(mp, agno, fbno, flen); - goto out_del_cursor; - } - - /* - * If the extent is entirely outside of the range we are - * supposed to discard skip it. Do not bother to trim - * down partially overlapping ranges for now. - */ - if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start || - XFS_AGB_TO_FSB(mp, agno, fbno) >= start + len) { - trace_xfs_discard_exclude(mp, agno, fbno, flen); - goto next_extent; - } - - /* - * If any blocks in the range are still busy, skip the - * discard and try again the next time. - */ - if (xfs_alloc_busy_search(mp, agno, fbno, flen)) { - trace_xfs_discard_busy(mp, agno, fbno, flen); - goto next_extent; - } - - trace_xfs_discard_extent(mp, agno, fbno, flen); - error = -blkdev_issue_discard(bdev, - XFS_AGB_TO_DADDR(mp, agno, fbno), - XFS_FSB_TO_BB(mp, flen), - GFP_NOFS, 0); - if (error) - goto out_del_cursor; - *blocks_trimmed += flen; - -next_extent: - error = xfs_btree_decrement(cur, 0, &i); - if (error) - goto out_del_cursor; - } - -out_del_cursor: - xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); - xfs_buf_relse(agbp); -out_put_perag: - xfs_perag_put(pag); - return error; -} - -int -xfs_ioc_trim( - struct xfs_mount *mp, - struct fstrim_range __user *urange) -{ - struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue; - unsigned int granularity = q->limits.discard_granularity; - struct fstrim_range range; - xfs_fsblock_t start, len, minlen; - xfs_agnumber_t start_agno, end_agno, agno; - __uint64_t blocks_trimmed = 0; - int error, last_error = 0; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (!blk_queue_discard(q)) - return -XFS_ERROR(EOPNOTSUPP); - if (copy_from_user(&range, urange, sizeof(range))) - return -XFS_ERROR(EFAULT); - - /* - * Truncating down the len isn't actually quite correct, but using - * XFS_B_TO_FSB would mean we trivially get overflows for values - * of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default - * used by the fstrim application. In the end it really doesn't - * matter as trimming blocks is an advisory interface. - */ - start = XFS_B_TO_FSBT(mp, range.start); - len = XFS_B_TO_FSBT(mp, range.len); - minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen)); - - start_agno = XFS_FSB_TO_AGNO(mp, start); - if (start_agno >= mp->m_sb.sb_agcount) - return -XFS_ERROR(EINVAL); - - end_agno = XFS_FSB_TO_AGNO(mp, start + len); - if (end_agno >= mp->m_sb.sb_agcount) - end_agno = mp->m_sb.sb_agcount - 1; - - for (agno = start_agno; agno <= end_agno; agno++) { - error = -xfs_trim_extents(mp, agno, start, len, minlen, - &blocks_trimmed); - if (error) - last_error = error; - } - - if (last_error) - return last_error; - - range.len = XFS_FSB_TO_B(mp, blocks_trimmed); - if (copy_to_user(urange, &range, sizeof(range))) - return -XFS_ERROR(EFAULT); - return 0; -} - -int -xfs_discard_extents( - struct xfs_mount *mp, - struct list_head *list) -{ - struct xfs_busy_extent *busyp; - int error = 0; - - list_for_each_entry(busyp, list, list) { - trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, - busyp->length); - - error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, - XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), - XFS_FSB_TO_BB(mp, busyp->length), - GFP_NOFS, 0); - if (error && error != EOPNOTSUPP) { - xfs_info(mp, - "discard failed for extent [0x%llu,%u], error %d", - (unsigned long long)busyp->bno, - busyp->length, - error); - return error; - } - } - - return 0; -} diff --git a/fs/xfs/linux-2.6/xfs_discard.h b/fs/xfs/linux-2.6/xfs_discard.h deleted file mode 100644 index 344879aea646..000000000000 --- a/fs/xfs/linux-2.6/xfs_discard.h +++ /dev/null @@ -1,10 +0,0 @@ -#ifndef XFS_DISCARD_H -#define XFS_DISCARD_H 1 - -struct fstrim_range; -struct list_head; - -extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *); -extern int xfs_discard_extents(struct xfs_mount *, struct list_head *); - -#endif /* XFS_DISCARD_H */ diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c deleted file mode 100644 index 75e5d322e48f..000000000000 --- a/fs/xfs/linux-2.6/xfs_export.c +++ /dev/null @@ -1,250 +0,0 @@ -/* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_types.h" -#include "xfs_inum.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_mount.h" -#include "xfs_export.h" -#include "xfs_vnodeops.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_trace.h" - -/* - * Note that we only accept fileids which are long enough rather than allow - * the parent generation number to default to zero. XFS considers zero a - * valid generation number not an invalid/wildcard value. - */ -static int xfs_fileid_length(int fileid_type) -{ - switch (fileid_type) { - case FILEID_INO32_GEN: - return 2; - case FILEID_INO32_GEN_PARENT: - return 4; - case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: - return 3; - case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: - return 6; - } - return 255; /* invalid */ -} - -STATIC int -xfs_fs_encode_fh( - struct dentry *dentry, - __u32 *fh, - int *max_len, - int connectable) -{ - struct fid *fid = (struct fid *)fh; - struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fh; - struct inode *inode = dentry->d_inode; - int fileid_type; - int len; - - /* Directories don't need their parent encoded, they have ".." */ - if (S_ISDIR(inode->i_mode) || !connectable) - fileid_type = FILEID_INO32_GEN; - else - fileid_type = FILEID_INO32_GEN_PARENT; - - /* - * If the the filesystem may contain 64bit inode numbers, we need - * to use larger file handles that can represent them. - * - * While we only allocate inodes that do not fit into 32 bits any - * large enough filesystem may contain them, thus the slightly - * confusing looking conditional below. - */ - if (!(XFS_M(inode->i_sb)->m_flags & XFS_MOUNT_SMALL_INUMS) || - (XFS_M(inode->i_sb)->m_flags & XFS_MOUNT_32BITINODES)) - fileid_type |= XFS_FILEID_TYPE_64FLAG; - - /* - * Only encode if there is enough space given. In practice - * this means we can't export a filesystem with 64bit inodes - * over NFSv2 with the subtree_check export option; the other - * seven combinations work. The real answer is "don't use v2". - */ - len = xfs_fileid_length(fileid_type); - if (*max_len < len) { - *max_len = len; - return 255; - } - *max_len = len; - - switch (fileid_type) { - case FILEID_INO32_GEN_PARENT: - spin_lock(&dentry->d_lock); - fid->i32.parent_ino = dentry->d_parent->d_inode->i_ino; - fid->i32.parent_gen = dentry->d_parent->d_inode->i_generation; - spin_unlock(&dentry->d_lock); - /*FALLTHRU*/ - case FILEID_INO32_GEN: - fid->i32.ino = inode->i_ino; - fid->i32.gen = inode->i_generation; - break; - case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: - spin_lock(&dentry->d_lock); - fid64->parent_ino = dentry->d_parent->d_inode->i_ino; - fid64->parent_gen = dentry->d_parent->d_inode->i_generation; - spin_unlock(&dentry->d_lock); - /*FALLTHRU*/ - case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: - fid64->ino = inode->i_ino; - fid64->gen = inode->i_generation; - break; - } - - return fileid_type; -} - -STATIC struct inode * -xfs_nfs_get_inode( - struct super_block *sb, - u64 ino, - u32 generation) - { - xfs_mount_t *mp = XFS_M(sb); - xfs_inode_t *ip; - int error; - - /* - * NFS can sometimes send requests for ino 0. Fail them gracefully. - */ - if (ino == 0) - return ERR_PTR(-ESTALE); - - /* - * The XFS_IGET_UNTRUSTED means that an invalid inode number is just - * fine and not an indication of a corrupted filesystem as clients can - * send invalid file handles and we have to handle it gracefully.. - */ - error = xfs_iget(mp, NULL, ino, XFS_IGET_UNTRUSTED, 0, &ip); - if (error) { - /* - * EINVAL means the inode cluster doesn't exist anymore. - * This implies the filehandle is stale, so we should - * translate it here. - * We don't use ESTALE directly down the chain to not - * confuse applications using bulkstat that expect EINVAL. - */ - if (error == EINVAL || error == ENOENT) - error = ESTALE; - return ERR_PTR(-error); - } - - if (ip->i_d.di_gen != generation) { - IRELE(ip); - return ERR_PTR(-ESTALE); - } - - return VFS_I(ip); -} - -STATIC struct dentry * -xfs_fs_fh_to_dentry(struct super_block *sb, struct fid *fid, - int fh_len, int fileid_type) -{ - struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fid; - struct inode *inode = NULL; - - if (fh_len < xfs_fileid_length(fileid_type)) - return NULL; - - switch (fileid_type) { - case FILEID_INO32_GEN_PARENT: - case FILEID_INO32_GEN: - inode = xfs_nfs_get_inode(sb, fid->i32.ino, fid->i32.gen); - break; - case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: - case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: - inode = xfs_nfs_get_inode(sb, fid64->ino, fid64->gen); - break; - } - - return d_obtain_alias(inode); -} - -STATIC struct dentry * -xfs_fs_fh_to_parent(struct super_block *sb, struct fid *fid, - int fh_len, int fileid_type) -{ - struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fid; - struct inode *inode = NULL; - - switch (fileid_type) { - case FILEID_INO32_GEN_PARENT: - inode = xfs_nfs_get_inode(sb, fid->i32.parent_ino, - fid->i32.parent_gen); - break; - case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: - inode = xfs_nfs_get_inode(sb, fid64->parent_ino, - fid64->parent_gen); - break; - } - - return d_obtain_alias(inode); -} - -STATIC struct dentry * -xfs_fs_get_parent( - struct dentry *child) -{ - int error; - struct xfs_inode *cip; - - error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL); - if (unlikely(error)) - return ERR_PTR(-error); - - return d_obtain_alias(VFS_I(cip)); -} - -STATIC int -xfs_fs_nfs_commit_metadata( - struct inode *inode) -{ - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - int error = 0; - - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (xfs_ipincount(ip)) { - error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn, - XFS_LOG_SYNC, NULL); - } - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - return error; -} - -const struct export_operations xfs_export_operations = { - .encode_fh = xfs_fs_encode_fh, - .fh_to_dentry = xfs_fs_fh_to_dentry, - .fh_to_parent = xfs_fs_fh_to_parent, - .get_parent = xfs_fs_get_parent, - .commit_metadata = xfs_fs_nfs_commit_metadata, -}; diff --git a/fs/xfs/linux-2.6/xfs_export.h b/fs/xfs/linux-2.6/xfs_export.h deleted file mode 100644 index 3272b6ae7a35..000000000000 --- a/fs/xfs/linux-2.6/xfs_export.h +++ /dev/null @@ -1,72 +0,0 @@ -/* - * Copyright (c) 2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_EXPORT_H__ -#define __XFS_EXPORT_H__ - -/* - * Common defines for code related to exporting XFS filesystems over NFS. - * - * The NFS fileid goes out on the wire as an array of - * 32bit unsigned ints in host order. There are 5 possible - * formats. - * - * (1) fileid_type=0x00 - * (no fileid data; handled by the generic code) - * - * (2) fileid_type=0x01 - * inode-num - * generation - * - * (3) fileid_type=0x02 - * inode-num - * generation - * parent-inode-num - * parent-generation - * - * (4) fileid_type=0x81 - * inode-num-lo32 - * inode-num-hi32 - * generation - * - * (5) fileid_type=0x82 - * inode-num-lo32 - * inode-num-hi32 - * generation - * parent-inode-num-lo32 - * parent-inode-num-hi32 - * parent-generation - * - * Note, the NFS filehandle also includes an fsid portion which - * may have an inode number in it. That number is hardcoded to - * 32bits and there is no way for XFS to intercept it. In - * practice this means when exporting an XFS filesystem with 64bit - * inodes you should either export the mountpoint (rather than - * a subdirectory) or use the "fsid" export option. - */ - -struct xfs_fid64 { - u64 ino; - u32 gen; - u64 parent_ino; - u32 parent_gen; -} __attribute__((packed)); - -/* This flag goes on the wire. Don't play with it. */ -#define XFS_FILEID_TYPE_64FLAG 0x80 /* NFS fileid has 64bit inodes */ - -#endif /* __XFS_EXPORT_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c deleted file mode 100644 index 7f7b42469ea7..000000000000 --- a/fs/xfs/linux-2.6/xfs_file.c +++ /dev/null @@ -1,1096 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_trans.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" -#include "xfs_bmap.h" -#include "xfs_error.h" -#include "xfs_vnodeops.h" -#include "xfs_da_btree.h" -#include "xfs_ioctl.h" -#include "xfs_trace.h" - -#include -#include - -static const struct vm_operations_struct xfs_file_vm_ops; - -/* - * Locking primitives for read and write IO paths to ensure we consistently use - * and order the inode->i_mutex, ip->i_lock and ip->i_iolock. - */ -static inline void -xfs_rw_ilock( - struct xfs_inode *ip, - int type) -{ - if (type & XFS_IOLOCK_EXCL) - mutex_lock(&VFS_I(ip)->i_mutex); - xfs_ilock(ip, type); -} - -static inline void -xfs_rw_iunlock( - struct xfs_inode *ip, - int type) -{ - xfs_iunlock(ip, type); - if (type & XFS_IOLOCK_EXCL) - mutex_unlock(&VFS_I(ip)->i_mutex); -} - -static inline void -xfs_rw_ilock_demote( - struct xfs_inode *ip, - int type) -{ - xfs_ilock_demote(ip, type); - if (type & XFS_IOLOCK_EXCL) - mutex_unlock(&VFS_I(ip)->i_mutex); -} - -/* - * xfs_iozero - * - * xfs_iozero clears the specified range of buffer supplied, - * and marks all the affected blocks as valid and modified. If - * an affected block is not allocated, it will be allocated. If - * an affected block is not completely overwritten, and is not - * valid before the operation, it will be read from disk before - * being partially zeroed. - */ -STATIC int -xfs_iozero( - struct xfs_inode *ip, /* inode */ - loff_t pos, /* offset in file */ - size_t count) /* size of data to zero */ -{ - struct page *page; - struct address_space *mapping; - int status; - - mapping = VFS_I(ip)->i_mapping; - do { - unsigned offset, bytes; - void *fsdata; - - offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ - bytes = PAGE_CACHE_SIZE - offset; - if (bytes > count) - bytes = count; - - status = pagecache_write_begin(NULL, mapping, pos, bytes, - AOP_FLAG_UNINTERRUPTIBLE, - &page, &fsdata); - if (status) - break; - - zero_user(page, offset, bytes); - - status = pagecache_write_end(NULL, mapping, pos, bytes, bytes, - page, fsdata); - WARN_ON(status <= 0); /* can't return less than zero! */ - pos += bytes; - count -= bytes; - status = 0; - } while (count); - - return (-status); -} - -STATIC int -xfs_file_fsync( - struct file *file, - loff_t start, - loff_t end, - int datasync) -{ - struct inode *inode = file->f_mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - struct xfs_trans *tp; - int error = 0; - int log_flushed = 0; - - trace_xfs_file_fsync(ip); - - error = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (error) - return error; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); - - xfs_iflags_clear(ip, XFS_ITRUNCATED); - - xfs_ilock(ip, XFS_IOLOCK_SHARED); - xfs_ioend_wait(ip); - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - - if (mp->m_flags & XFS_MOUNT_BARRIER) { - /* - * If we have an RT and/or log subvolume we need to make sure - * to flush the write cache the device used for file data - * first. This is to ensure newly written file data make - * it to disk before logging the new inode size in case of - * an extending write. - */ - if (XFS_IS_REALTIME_INODE(ip)) - xfs_blkdev_issue_flush(mp->m_rtdev_targp); - else if (mp->m_logdev_targp != mp->m_ddev_targp) - xfs_blkdev_issue_flush(mp->m_ddev_targp); - } - - /* - * We always need to make sure that the required inode state is safe on - * disk. The inode might be clean but we still might need to force the - * log because of committed transactions that haven't hit the disk yet. - * Likewise, there could be unflushed non-transactional changes to the - * inode core that have to go to disk and this requires us to issue - * a synchronous transaction to capture these changes correctly. - * - * This code relies on the assumption that if the i_update_core field - * of the inode is clear and the inode is unpinned then it is clean - * and no action is required. - */ - xfs_ilock(ip, XFS_ILOCK_SHARED); - - /* - * First check if the VFS inode is marked dirty. All the dirtying - * of non-transactional updates no goes through mark_inode_dirty*, - * which allows us to distinguish beteeen pure timestamp updates - * and i_size updates which need to be caught for fdatasync. - * After that also theck for the dirty state in the XFS inode, which - * might gets cleared when the inode gets written out via the AIL - * or xfs_iflush_cluster. - */ - if (((inode->i_state & I_DIRTY_DATASYNC) || - ((inode->i_state & I_DIRTY_SYNC) && !datasync)) && - ip->i_update_core) { - /* - * Kick off a transaction to log the inode core to get the - * updates. The sync transaction will also force the log. - */ - xfs_iunlock(ip, XFS_ILOCK_SHARED); - tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, - XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); - if (error) { - xfs_trans_cancel(tp, 0); - return -error; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * Note - it's possible that we might have pushed ourselves out - * of the way during trans_reserve which would flush the inode. - * But there's no guarantee that the inode buffer has actually - * gone out yet (it's delwri). Plus the buffer could be pinned - * anyway if it's part of an inode in another recent - * transaction. So we play it safe and fire off the - * transaction anyway. - */ - xfs_trans_ijoin(tp, ip); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - xfs_trans_set_sync(tp); - error = _xfs_trans_commit(tp, 0, &log_flushed); - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - } else { - /* - * Timestamps/size haven't changed since last inode flush or - * inode transaction commit. That means either nothing got - * written or a transaction committed which caught the updates. - * If the latter happened and the transaction hasn't hit the - * disk yet, the inode will be still be pinned. If it is, - * force the log. - */ - if (xfs_ipincount(ip)) { - error = _xfs_log_force_lsn(mp, - ip->i_itemp->ili_last_lsn, - XFS_LOG_SYNC, &log_flushed); - } - xfs_iunlock(ip, XFS_ILOCK_SHARED); - } - - /* - * If we only have a single device, and the log force about was - * a no-op we might have to flush the data device cache here. - * This can only happen for fdatasync/O_DSYNC if we were overwriting - * an already allocated file and thus do not have any metadata to - * commit. - */ - if ((mp->m_flags & XFS_MOUNT_BARRIER) && - mp->m_logdev_targp == mp->m_ddev_targp && - !XFS_IS_REALTIME_INODE(ip) && - !log_flushed) - xfs_blkdev_issue_flush(mp->m_ddev_targp); - - return -error; -} - -STATIC ssize_t -xfs_file_aio_read( - struct kiocb *iocb, - const struct iovec *iovp, - unsigned long nr_segs, - loff_t pos) -{ - struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - size_t size = 0; - ssize_t ret = 0; - int ioflags = 0; - xfs_fsize_t n; - unsigned long seg; - - XFS_STATS_INC(xs_read_calls); - - BUG_ON(iocb->ki_pos != pos); - - if (unlikely(file->f_flags & O_DIRECT)) - ioflags |= IO_ISDIRECT; - if (file->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - - /* START copy & waste from filemap.c */ - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *iv = &iovp[seg]; - - /* - * If any segment has a negative length, or the cumulative - * length ever wraps negative then return -EINVAL. - */ - size += iv->iov_len; - if (unlikely((ssize_t)(size|iv->iov_len) < 0)) - return XFS_ERROR(-EINVAL); - } - /* END copy & waste from filemap.c */ - - if (unlikely(ioflags & IO_ISDIRECT)) { - xfs_buftarg_t *target = - XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; - if ((iocb->ki_pos & target->bt_smask) || - (size & target->bt_smask)) { - if (iocb->ki_pos == ip->i_size) - return 0; - return -XFS_ERROR(EINVAL); - } - } - - n = XFS_MAXIOFFSET(mp) - iocb->ki_pos; - if (n <= 0 || size == 0) - return 0; - - if (n < size) - size = n; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -EIO; - - if (unlikely(ioflags & IO_ISDIRECT)) { - xfs_rw_ilock(ip, XFS_IOLOCK_EXCL); - - if (inode->i_mapping->nrpages) { - ret = -xfs_flushinval_pages(ip, - (iocb->ki_pos & PAGE_CACHE_MASK), - -1, FI_REMAPF_LOCKED); - if (ret) { - xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL); - return ret; - } - } - xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL); - } else - xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); - - trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags); - - ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos); - if (ret > 0) - XFS_STATS_ADD(xs_read_bytes, ret); - - xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED); - return ret; -} - -STATIC ssize_t -xfs_file_splice_read( - struct file *infilp, - loff_t *ppos, - struct pipe_inode_info *pipe, - size_t count, - unsigned int flags) -{ - struct xfs_inode *ip = XFS_I(infilp->f_mapping->host); - int ioflags = 0; - ssize_t ret; - - XFS_STATS_INC(xs_read_calls); - - if (infilp->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return -EIO; - - xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); - - trace_xfs_file_splice_read(ip, count, *ppos, ioflags); - - ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); - if (ret > 0) - XFS_STATS_ADD(xs_read_bytes, ret); - - xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED); - return ret; -} - -STATIC void -xfs_aio_write_isize_update( - struct inode *inode, - loff_t *ppos, - ssize_t bytes_written) -{ - struct xfs_inode *ip = XFS_I(inode); - xfs_fsize_t isize = i_size_read(inode); - - if (bytes_written > 0) - XFS_STATS_ADD(xs_write_bytes, bytes_written); - - if (unlikely(bytes_written < 0 && bytes_written != -EFAULT && - *ppos > isize)) - *ppos = isize; - - if (*ppos > ip->i_size) { - xfs_rw_ilock(ip, XFS_ILOCK_EXCL); - if (*ppos > ip->i_size) - ip->i_size = *ppos; - xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); - } -} - -/* - * If this was a direct or synchronous I/O that failed (such as ENOSPC) then - * part of the I/O may have been written to disk before the error occurred. In - * this case the on-disk file size may have been adjusted beyond the in-memory - * file size and now needs to be truncated back. - */ -STATIC void -xfs_aio_write_newsize_update( - struct xfs_inode *ip) -{ - if (ip->i_new_size) { - xfs_rw_ilock(ip, XFS_ILOCK_EXCL); - ip->i_new_size = 0; - if (ip->i_d.di_size > ip->i_size) - ip->i_d.di_size = ip->i_size; - xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); - } -} - -/* - * xfs_file_splice_write() does not use xfs_rw_ilock() because - * generic_file_splice_write() takes the i_mutex itself. This, in theory, - * couuld cause lock inversions between the aio_write path and the splice path - * if someone is doing concurrent splice(2) based writes and write(2) based - * writes to the same inode. The only real way to fix this is to re-implement - * the generic code here with correct locking orders. - */ -STATIC ssize_t -xfs_file_splice_write( - struct pipe_inode_info *pipe, - struct file *outfilp, - loff_t *ppos, - size_t count, - unsigned int flags) -{ - struct inode *inode = outfilp->f_mapping->host; - struct xfs_inode *ip = XFS_I(inode); - xfs_fsize_t new_size; - int ioflags = 0; - ssize_t ret; - - XFS_STATS_INC(xs_write_calls); - - if (outfilp->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return -EIO; - - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - new_size = *ppos + count; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - if (new_size > ip->i_size) - ip->i_new_size = new_size; - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - trace_xfs_file_splice_write(ip, count, *ppos, ioflags); - - ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags); - - xfs_aio_write_isize_update(inode, ppos, ret); - xfs_aio_write_newsize_update(ip); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return ret; -} - -/* - * This routine is called to handle zeroing any space in the last - * block of the file that is beyond the EOF. We do this since the - * size is being increased without writing anything to that block - * and we don't want anyone to read the garbage on the disk. - */ -STATIC int /* error (positive) */ -xfs_zero_last_block( - xfs_inode_t *ip, - xfs_fsize_t offset, - xfs_fsize_t isize) -{ - xfs_fileoff_t last_fsb; - xfs_mount_t *mp = ip->i_mount; - int nimaps; - int zero_offset; - int zero_len; - int error = 0; - xfs_bmbt_irec_t imap; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - - zero_offset = XFS_B_FSB_OFFSET(mp, isize); - if (zero_offset == 0) { - /* - * There are no extra bytes in the last block on disk to - * zero, so return. - */ - return 0; - } - - last_fsb = XFS_B_TO_FSBT(mp, isize); - nimaps = 1; - error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap, - &nimaps, NULL); - if (error) { - return error; - } - ASSERT(nimaps > 0); - /* - * If the block underlying isize is just a hole, then there - * is nothing to zero. - */ - if (imap.br_startblock == HOLESTARTBLOCK) { - return 0; - } - /* - * Zero the part of the last block beyond the EOF, and write it - * out sync. We need to drop the ilock while we do this so we - * don't deadlock when the buffer cache calls back to us. - */ - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - zero_len = mp->m_sb.sb_blocksize - zero_offset; - if (isize + zero_len > offset) - zero_len = offset - isize; - error = xfs_iozero(ip, isize, zero_len); - - xfs_ilock(ip, XFS_ILOCK_EXCL); - ASSERT(error >= 0); - return error; -} - -/* - * Zero any on disk space between the current EOF and the new, - * larger EOF. This handles the normal case of zeroing the remainder - * of the last block in the file and the unusual case of zeroing blocks - * out beyond the size of the file. This second case only happens - * with fixed size extents and when the system crashes before the inode - * size was updated but after blocks were allocated. If fill is set, - * then any holes in the range are filled and zeroed. If not, the holes - * are left alone as holes. - */ - -int /* error (positive) */ -xfs_zero_eof( - xfs_inode_t *ip, - xfs_off_t offset, /* starting I/O offset */ - xfs_fsize_t isize) /* current inode size */ -{ - xfs_mount_t *mp = ip->i_mount; - xfs_fileoff_t start_zero_fsb; - xfs_fileoff_t end_zero_fsb; - xfs_fileoff_t zero_count_fsb; - xfs_fileoff_t last_fsb; - xfs_fileoff_t zero_off; - xfs_fsize_t zero_len; - int nimaps; - int error = 0; - xfs_bmbt_irec_t imap; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - ASSERT(offset > isize); - - /* - * First handle zeroing the block on which isize resides. - * We only zero a part of that block so it is handled specially. - */ - error = xfs_zero_last_block(ip, offset, isize); - if (error) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - return error; - } - - /* - * Calculate the range between the new size and the old - * where blocks needing to be zeroed may exist. To get the - * block where the last byte in the file currently resides, - * we need to subtract one from the size and truncate back - * to a block boundary. We subtract 1 in case the size is - * exactly on a block boundary. - */ - last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1; - start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize); - end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1); - ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb); - if (last_fsb == end_zero_fsb) { - /* - * The size was only incremented on its last block. - * We took care of that above, so just return. - */ - return 0; - } - - ASSERT(start_zero_fsb <= end_zero_fsb); - while (start_zero_fsb <= end_zero_fsb) { - nimaps = 1; - zero_count_fsb = end_zero_fsb - start_zero_fsb + 1; - error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, - 0, NULL, 0, &imap, &nimaps, NULL); - if (error) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); - return error; - } - ASSERT(nimaps > 0); - - if (imap.br_state == XFS_EXT_UNWRITTEN || - imap.br_startblock == HOLESTARTBLOCK) { - /* - * This loop handles initializing pages that were - * partially initialized by the code below this - * loop. It basically zeroes the part of the page - * that sits on a hole and sets the page as P_HOLE - * and calls remapf if it is a mapped file. - */ - start_zero_fsb = imap.br_startoff + imap.br_blockcount; - ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - continue; - } - - /* - * There are blocks we need to zero. - * Drop the inode lock while we're doing the I/O. - * We'll still have the iolock to protect us. - */ - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); - zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); - - if ((zero_off + zero_len) > offset) - zero_len = offset - zero_off; - - error = xfs_iozero(ip, zero_off, zero_len); - if (error) { - goto out_lock; - } - - start_zero_fsb = imap.br_startoff + imap.br_blockcount; - ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); - - xfs_ilock(ip, XFS_ILOCK_EXCL); - } - - return 0; - -out_lock: - xfs_ilock(ip, XFS_ILOCK_EXCL); - ASSERT(error >= 0); - return error; -} - -/* - * Common pre-write limit and setup checks. - * - * Returns with iolock held according to @iolock. - */ -STATIC ssize_t -xfs_file_aio_write_checks( - struct file *file, - loff_t *pos, - size_t *count, - int *iolock) -{ - struct inode *inode = file->f_mapping->host; - struct xfs_inode *ip = XFS_I(inode); - xfs_fsize_t new_size; - int error = 0; - - error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode)); - if (error) { - xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock); - *iolock = 0; - return error; - } - - new_size = *pos + *count; - if (new_size > ip->i_size) - ip->i_new_size = new_size; - - if (likely(!(file->f_mode & FMODE_NOCMTIME))) - file_update_time(file); - - /* - * If the offset is beyond the size of the file, we need to zero any - * blocks that fall between the existing EOF and the start of this - * write. - */ - if (*pos > ip->i_size) - error = -xfs_zero_eof(ip, *pos, ip->i_size); - - xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); - if (error) - return error; - - /* - * If we're writing the file then make sure to clear the setuid and - * setgid bits if the process is not being run by root. This keeps - * people from modifying setuid and setgid binaries. - */ - return file_remove_suid(file); - -} - -/* - * xfs_file_dio_aio_write - handle direct IO writes - * - * Lock the inode appropriately to prepare for and issue a direct IO write. - * By separating it from the buffered write path we remove all the tricky to - * follow locking changes and looping. - * - * If there are cached pages or we're extending the file, we need IOLOCK_EXCL - * until we're sure the bytes at the new EOF have been zeroed and/or the cached - * pages are flushed out. - * - * In most cases the direct IO writes will be done holding IOLOCK_SHARED - * allowing them to be done in parallel with reads and other direct IO writes. - * However, if the IO is not aligned to filesystem blocks, the direct IO layer - * needs to do sub-block zeroing and that requires serialisation against other - * direct IOs to the same block. In this case we need to serialise the - * submission of the unaligned IOs so that we don't get racing block zeroing in - * the dio layer. To avoid the problem with aio, we also need to wait for - * outstanding IOs to complete so that unwritten extent conversion is completed - * before we try to map the overlapping block. This is currently implemented by - * hitting it with a big hammer (i.e. xfs_ioend_wait()). - * - * Returns with locks held indicated by @iolock and errors indicated by - * negative return values. - */ -STATIC ssize_t -xfs_file_dio_aio_write( - struct kiocb *iocb, - const struct iovec *iovp, - unsigned long nr_segs, - loff_t pos, - size_t ocount, - int *iolock) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - ssize_t ret = 0; - size_t count = ocount; - int unaligned_io = 0; - struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; - - *iolock = 0; - if ((pos & target->bt_smask) || (count & target->bt_smask)) - return -XFS_ERROR(EINVAL); - - if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask)) - unaligned_io = 1; - - if (unaligned_io || mapping->nrpages || pos > ip->i_size) - *iolock = XFS_IOLOCK_EXCL; - else - *iolock = XFS_IOLOCK_SHARED; - xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock); - - ret = xfs_file_aio_write_checks(file, &pos, &count, iolock); - if (ret) - return ret; - - if (mapping->nrpages) { - WARN_ON(*iolock != XFS_IOLOCK_EXCL); - ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1, - FI_REMAPF_LOCKED); - if (ret) - return ret; - } - - /* - * If we are doing unaligned IO, wait for all other IO to drain, - * otherwise demote the lock if we had to flush cached pages - */ - if (unaligned_io) - xfs_ioend_wait(ip); - else if (*iolock == XFS_IOLOCK_EXCL) { - xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL); - *iolock = XFS_IOLOCK_SHARED; - } - - trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0); - ret = generic_file_direct_write(iocb, iovp, - &nr_segs, pos, &iocb->ki_pos, count, ocount); - - /* No fallback to buffered IO on errors for XFS. */ - ASSERT(ret < 0 || ret == count); - return ret; -} - -STATIC ssize_t -xfs_file_buffered_aio_write( - struct kiocb *iocb, - const struct iovec *iovp, - unsigned long nr_segs, - loff_t pos, - size_t ocount, - int *iolock) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - struct xfs_inode *ip = XFS_I(inode); - ssize_t ret; - int enospc = 0; - size_t count = ocount; - - *iolock = XFS_IOLOCK_EXCL; - xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock); - - ret = xfs_file_aio_write_checks(file, &pos, &count, iolock); - if (ret) - return ret; - - /* We can write back this queue in page reclaim */ - current->backing_dev_info = mapping->backing_dev_info; - -write_retry: - trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0); - ret = generic_file_buffered_write(iocb, iovp, nr_segs, - pos, &iocb->ki_pos, count, ret); - /* - * if we just got an ENOSPC, flush the inode now we aren't holding any - * page locks and retry *once* - */ - if (ret == -ENOSPC && !enospc) { - ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE); - if (ret) - return ret; - enospc = 1; - goto write_retry; - } - current->backing_dev_info = NULL; - return ret; -} - -STATIC ssize_t -xfs_file_aio_write( - struct kiocb *iocb, - const struct iovec *iovp, - unsigned long nr_segs, - loff_t pos) -{ - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - struct xfs_inode *ip = XFS_I(inode); - ssize_t ret; - int iolock; - size_t ocount = 0; - - XFS_STATS_INC(xs_write_calls); - - BUG_ON(iocb->ki_pos != pos); - - ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ); - if (ret) - return ret; - - if (ocount == 0) - return 0; - - xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE); - - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return -EIO; - - if (unlikely(file->f_flags & O_DIRECT)) - ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, - ocount, &iolock); - else - ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos, - ocount, &iolock); - - xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret); - - if (ret <= 0) - goto out_unlock; - - /* Handle various SYNC-type writes */ - if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { - loff_t end = pos + ret - 1; - int error; - - xfs_rw_iunlock(ip, iolock); - error = xfs_file_fsync(file, pos, end, - (file->f_flags & __O_SYNC) ? 0 : 1); - xfs_rw_ilock(ip, iolock); - if (error) - ret = error; - } - -out_unlock: - xfs_aio_write_newsize_update(ip); - xfs_rw_iunlock(ip, iolock); - return ret; -} - -STATIC long -xfs_file_fallocate( - struct file *file, - int mode, - loff_t offset, - loff_t len) -{ - struct inode *inode = file->f_path.dentry->d_inode; - long error; - loff_t new_size = 0; - xfs_flock64_t bf; - xfs_inode_t *ip = XFS_I(inode); - int cmd = XFS_IOC_RESVSP; - int attr_flags = XFS_ATTR_NOLOCK; - - if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) - return -EOPNOTSUPP; - - bf.l_whence = 0; - bf.l_start = offset; - bf.l_len = len; - - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - if (mode & FALLOC_FL_PUNCH_HOLE) - cmd = XFS_IOC_UNRESVSP; - - /* check the new inode size is valid before allocating */ - if (!(mode & FALLOC_FL_KEEP_SIZE) && - offset + len > i_size_read(inode)) { - new_size = offset + len; - error = inode_newsize_ok(inode, new_size); - if (error) - goto out_unlock; - } - - if (file->f_flags & O_DSYNC) - attr_flags |= XFS_ATTR_SYNC; - - error = -xfs_change_file_space(ip, cmd, &bf, 0, attr_flags); - if (error) - goto out_unlock; - - /* Change file size if needed */ - if (new_size) { - struct iattr iattr; - - iattr.ia_valid = ATTR_SIZE; - iattr.ia_size = new_size; - error = -xfs_setattr_size(ip, &iattr, XFS_ATTR_NOLOCK); - } - -out_unlock: - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return error; -} - - -STATIC int -xfs_file_open( - struct inode *inode, - struct file *file) -{ - if (!(file->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS) - return -EFBIG; - if (XFS_FORCED_SHUTDOWN(XFS_M(inode->i_sb))) - return -EIO; - return 0; -} - -STATIC int -xfs_dir_open( - struct inode *inode, - struct file *file) -{ - struct xfs_inode *ip = XFS_I(inode); - int mode; - int error; - - error = xfs_file_open(inode, file); - if (error) - return error; - - /* - * If there are any blocks, read-ahead block 0 as we're almost - * certain to have the next operation be a read there. - */ - mode = xfs_ilock_map_shared(ip); - if (ip->i_d.di_nextents > 0) - xfs_da_reada_buf(NULL, ip, 0, XFS_DATA_FORK); - xfs_iunlock(ip, mode); - return 0; -} - -STATIC int -xfs_file_release( - struct inode *inode, - struct file *filp) -{ - return -xfs_release(XFS_I(inode)); -} - -STATIC int -xfs_file_readdir( - struct file *filp, - void *dirent, - filldir_t filldir) -{ - struct inode *inode = filp->f_path.dentry->d_inode; - xfs_inode_t *ip = XFS_I(inode); - int error; - size_t bufsize; - - /* - * The Linux API doesn't pass down the total size of the buffer - * we read into down to the filesystem. With the filldir concept - * it's not needed for correct information, but the XFS dir2 leaf - * code wants an estimate of the buffer size to calculate it's - * readahead window and size the buffers used for mapping to - * physical blocks. - * - * Try to give it an estimate that's good enough, maybe at some - * point we can change the ->readdir prototype to include the - * buffer size. For now we use the current glibc buffer size. - */ - bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size); - - error = xfs_readdir(ip, dirent, bufsize, - (xfs_off_t *)&filp->f_pos, filldir); - if (error) - return -error; - return 0; -} - -STATIC int -xfs_file_mmap( - struct file *filp, - struct vm_area_struct *vma) -{ - vma->vm_ops = &xfs_file_vm_ops; - vma->vm_flags |= VM_CAN_NONLINEAR; - - file_accessed(filp); - return 0; -} - -/* - * mmap()d file has taken write protection fault and is being made - * writable. We can set the page state up correctly for a writable - * page, which means we can do correct delalloc accounting (ENOSPC - * checking!) and unwritten extent mapping. - */ -STATIC int -xfs_vm_page_mkwrite( - struct vm_area_struct *vma, - struct vm_fault *vmf) -{ - return block_page_mkwrite(vma, vmf, xfs_get_blocks); -} - -const struct file_operations xfs_file_operations = { - .llseek = generic_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = xfs_file_aio_read, - .aio_write = xfs_file_aio_write, - .splice_read = xfs_file_splice_read, - .splice_write = xfs_file_splice_write, - .unlocked_ioctl = xfs_file_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = xfs_file_compat_ioctl, -#endif - .mmap = xfs_file_mmap, - .open = xfs_file_open, - .release = xfs_file_release, - .fsync = xfs_file_fsync, - .fallocate = xfs_file_fallocate, -}; - -const struct file_operations xfs_dir_file_operations = { - .open = xfs_dir_open, - .read = generic_read_dir, - .readdir = xfs_file_readdir, - .llseek = generic_file_llseek, - .unlocked_ioctl = xfs_file_ioctl, -#ifdef CONFIG_COMPAT - .compat_ioctl = xfs_file_compat_ioctl, -#endif - .fsync = xfs_file_fsync, -}; - -static const struct vm_operations_struct xfs_file_vm_ops = { - .fault = filemap_fault, - .page_mkwrite = xfs_vm_page_mkwrite, -}; diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c deleted file mode 100644 index ed88ed16811c..000000000000 --- a/fs/xfs/linux-2.6/xfs_fs_subr.c +++ /dev/null @@ -1,96 +0,0 @@ -/* - * Copyright (c) 2000-2002,2005-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_vnodeops.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_trace.h" - -/* - * note: all filemap functions return negative error codes. These - * need to be inverted before returning to the xfs core functions. - */ -void -xfs_tosspages( - xfs_inode_t *ip, - xfs_off_t first, - xfs_off_t last, - int fiopt) -{ - /* can't toss partial tail pages, so mask them out */ - last &= ~(PAGE_SIZE - 1); - truncate_inode_pages_range(VFS_I(ip)->i_mapping, first, last - 1); -} - -int -xfs_flushinval_pages( - xfs_inode_t *ip, - xfs_off_t first, - xfs_off_t last, - int fiopt) -{ - struct address_space *mapping = VFS_I(ip)->i_mapping; - int ret = 0; - - trace_xfs_pagecache_inval(ip, first, last); - - xfs_iflags_clear(ip, XFS_ITRUNCATED); - ret = filemap_write_and_wait_range(mapping, first, - last == -1 ? LLONG_MAX : last); - if (!ret) - truncate_inode_pages_range(mapping, first, last); - return -ret; -} - -int -xfs_flush_pages( - xfs_inode_t *ip, - xfs_off_t first, - xfs_off_t last, - uint64_t flags, - int fiopt) -{ - struct address_space *mapping = VFS_I(ip)->i_mapping; - int ret = 0; - int ret2; - - xfs_iflags_clear(ip, XFS_ITRUNCATED); - ret = -filemap_fdatawrite_range(mapping, first, - last == -1 ? LLONG_MAX : last); - if (flags & XBF_ASYNC) - return ret; - ret2 = xfs_wait_on_pages(ip, first, last); - if (!ret) - ret = ret2; - return ret; -} - -int -xfs_wait_on_pages( - xfs_inode_t *ip, - xfs_off_t first, - xfs_off_t last) -{ - struct address_space *mapping = VFS_I(ip)->i_mapping; - - if (mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK)) { - return -filemap_fdatawait_range(mapping, first, - last == -1 ? ip->i_size - 1 : last); - } - return 0; -} diff --git a/fs/xfs/linux-2.6/xfs_globals.c b/fs/xfs/linux-2.6/xfs_globals.c deleted file mode 100644 index 76e81cff70b9..000000000000 --- a/fs/xfs/linux-2.6/xfs_globals.c +++ /dev/null @@ -1,43 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_sysctl.h" - -/* - * Tunable XFS parameters. xfs_params is required even when CONFIG_SYSCTL=n, - * other XFS code uses these values. Times are measured in centisecs (i.e. - * 100ths of a second). - */ -xfs_param_t xfs_params = { - /* MIN DFLT MAX */ - .sgid_inherit = { 0, 0, 1 }, - .symlink_mode = { 0, 0, 1 }, - .panic_mask = { 0, 0, 255 }, - .error_level = { 0, 3, 11 }, - .syncd_timer = { 1*100, 30*100, 7200*100}, - .stats_clear = { 0, 0, 1 }, - .inherit_sync = { 0, 1, 1 }, - .inherit_nodump = { 0, 1, 1 }, - .inherit_noatim = { 0, 1, 1 }, - .xfs_buf_timer = { 100/2, 1*100, 30*100 }, - .xfs_buf_age = { 1*100, 15*100, 7200*100}, - .inherit_nosym = { 0, 0, 1 }, - .rotorstep = { 1, 1, 255 }, - .inherit_nodfrg = { 0, 1, 1 }, - .fstrm_timer = { 1, 30*100, 3600*100}, -}; diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c deleted file mode 100644 index f7ce7debe14c..000000000000 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ /dev/null @@ -1,1556 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_ioctl.h" -#include "xfs_rtalloc.h" -#include "xfs_itable.h" -#include "xfs_error.h" -#include "xfs_attr.h" -#include "xfs_bmap.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_dfrag.h" -#include "xfs_fsops.h" -#include "xfs_vnodeops.h" -#include "xfs_discard.h" -#include "xfs_quota.h" -#include "xfs_inode_item.h" -#include "xfs_export.h" -#include "xfs_trace.h" - -#include -#include -#include -#include -#include -#include -#include - -/* - * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to - * a file or fs handle. - * - * XFS_IOC_PATH_TO_FSHANDLE - * returns fs handle for a mount point or path within that mount point - * XFS_IOC_FD_TO_HANDLE - * returns full handle for a FD opened in user space - * XFS_IOC_PATH_TO_HANDLE - * returns full handle for a path - */ -int -xfs_find_handle( - unsigned int cmd, - xfs_fsop_handlereq_t *hreq) -{ - int hsize; - xfs_handle_t handle; - struct inode *inode; - struct file *file = NULL; - struct path path; - int error; - struct xfs_inode *ip; - - if (cmd == XFS_IOC_FD_TO_HANDLE) { - file = fget(hreq->fd); - if (!file) - return -EBADF; - inode = file->f_path.dentry->d_inode; - } else { - error = user_lpath((const char __user *)hreq->path, &path); - if (error) - return error; - inode = path.dentry->d_inode; - } - ip = XFS_I(inode); - - /* - * We can only generate handles for inodes residing on a XFS filesystem, - * and only for regular files, directories or symbolic links. - */ - error = -EINVAL; - if (inode->i_sb->s_magic != XFS_SB_MAGIC) - goto out_put; - - error = -EBADF; - if (!S_ISREG(inode->i_mode) && - !S_ISDIR(inode->i_mode) && - !S_ISLNK(inode->i_mode)) - goto out_put; - - - memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); - - if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { - /* - * This handle only contains an fsid, zero the rest. - */ - memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); - hsize = sizeof(xfs_fsid_t); - } else { - int lock_mode; - - lock_mode = xfs_ilock_map_shared(ip); - handle.ha_fid.fid_len = sizeof(xfs_fid_t) - - sizeof(handle.ha_fid.fid_len); - handle.ha_fid.fid_pad = 0; - handle.ha_fid.fid_gen = ip->i_d.di_gen; - handle.ha_fid.fid_ino = ip->i_ino; - xfs_iunlock_map_shared(ip, lock_mode); - - hsize = XFS_HSIZE(handle); - } - - error = -EFAULT; - if (copy_to_user(hreq->ohandle, &handle, hsize) || - copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) - goto out_put; - - error = 0; - - out_put: - if (cmd == XFS_IOC_FD_TO_HANDLE) - fput(file); - else - path_put(&path); - return error; -} - -/* - * No need to do permission checks on the various pathname components - * as the handle operations are privileged. - */ -STATIC int -xfs_handle_acceptable( - void *context, - struct dentry *dentry) -{ - return 1; -} - -/* - * Convert userspace handle data into a dentry. - */ -struct dentry * -xfs_handle_to_dentry( - struct file *parfilp, - void __user *uhandle, - u32 hlen) -{ - xfs_handle_t handle; - struct xfs_fid64 fid; - - /* - * Only allow handle opens under a directory. - */ - if (!S_ISDIR(parfilp->f_path.dentry->d_inode->i_mode)) - return ERR_PTR(-ENOTDIR); - - if (hlen != sizeof(xfs_handle_t)) - return ERR_PTR(-EINVAL); - if (copy_from_user(&handle, uhandle, hlen)) - return ERR_PTR(-EFAULT); - if (handle.ha_fid.fid_len != - sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) - return ERR_PTR(-EINVAL); - - memset(&fid, 0, sizeof(struct fid)); - fid.ino = handle.ha_fid.fid_ino; - fid.gen = handle.ha_fid.fid_gen; - - return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, - FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, - xfs_handle_acceptable, NULL); -} - -STATIC struct dentry * -xfs_handlereq_to_dentry( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - return xfs_handle_to_dentry(parfilp, hreq->ihandle, hreq->ihandlen); -} - -int -xfs_open_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - const struct cred *cred = current_cred(); - int error; - int fd; - int permflag; - struct file *filp; - struct inode *inode; - struct dentry *dentry; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - inode = dentry->d_inode; - - /* Restrict xfs_open_by_handle to directories & regular files. */ - if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) { - error = -XFS_ERROR(EPERM); - goto out_dput; - } - -#if BITS_PER_LONG != 32 - hreq->oflags |= O_LARGEFILE; -#endif - - /* Put open permission in namei format. */ - permflag = hreq->oflags; - if ((permflag+1) & O_ACCMODE) - permflag++; - if (permflag & O_TRUNC) - permflag |= 2; - - if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) && - (permflag & FMODE_WRITE) && IS_APPEND(inode)) { - error = -XFS_ERROR(EPERM); - goto out_dput; - } - - if ((permflag & FMODE_WRITE) && IS_IMMUTABLE(inode)) { - error = -XFS_ERROR(EACCES); - goto out_dput; - } - - /* Can't write directories. */ - if (S_ISDIR(inode->i_mode) && (permflag & FMODE_WRITE)) { - error = -XFS_ERROR(EISDIR); - goto out_dput; - } - - fd = get_unused_fd(); - if (fd < 0) { - error = fd; - goto out_dput; - } - - filp = dentry_open(dentry, mntget(parfilp->f_path.mnt), - hreq->oflags, cred); - if (IS_ERR(filp)) { - put_unused_fd(fd); - return PTR_ERR(filp); - } - - if (S_ISREG(inode->i_mode)) { - filp->f_flags |= O_NOATIME; - filp->f_mode |= FMODE_NOCMTIME; - } - - fd_install(fd, filp); - return fd; - - out_dput: - dput(dentry); - return error; -} - -/* - * This is a copy from fs/namei.c:vfs_readlink(), except for removing it's - * unused first argument. - */ -STATIC int -do_readlink( - char __user *buffer, - int buflen, - const char *link) -{ - int len; - - len = PTR_ERR(link); - if (IS_ERR(link)) - goto out; - - len = strlen(link); - if (len > (unsigned) buflen) - len = buflen; - if (copy_to_user(buffer, link, len)) - len = -EFAULT; - out: - return len; -} - - -int -xfs_readlink_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq) -{ - struct dentry *dentry; - __u32 olen; - void *link; - int error; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - dentry = xfs_handlereq_to_dentry(parfilp, hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - /* Restrict this handle operation to symlinks only. */ - if (!S_ISLNK(dentry->d_inode->i_mode)) { - error = -XFS_ERROR(EINVAL); - goto out_dput; - } - - if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) { - error = -XFS_ERROR(EFAULT); - goto out_dput; - } - - link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); - if (!link) { - error = -XFS_ERROR(ENOMEM); - goto out_dput; - } - - error = -xfs_readlink(XFS_I(dentry->d_inode), link); - if (error) - goto out_kfree; - error = do_readlink(hreq->ohandle, olen, link); - if (error) - goto out_kfree; - - out_kfree: - kfree(link); - out_dput: - dput(dentry); - return error; -} - -STATIC int -xfs_fssetdm_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - struct fsdmidata fsd; - xfs_fsop_setdm_handlereq_t dmhreq; - struct dentry *dentry; - - if (!capable(CAP_MKNOD)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t))) - return -XFS_ERROR(EFAULT); - - dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) { - error = -XFS_ERROR(EPERM); - goto out; - } - - if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) { - error = -XFS_ERROR(EFAULT); - goto out; - } - - error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask, - fsd.fsd_dmstate); - - out: - dput(dentry); - return error; -} - -STATIC int -xfs_attrlist_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error = -ENOMEM; - attrlist_cursor_kern_t *cursor; - xfs_fsop_attrlist_handlereq_t al_hreq; - struct dentry *dentry; - char *kbuf; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t))) - return -XFS_ERROR(EFAULT); - if (al_hreq.buflen > XATTR_LIST_MAX) - return -XFS_ERROR(EINVAL); - - /* - * Reject flags, only allow namespaces. - */ - if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) - return -XFS_ERROR(EINVAL); - - dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - kbuf = kzalloc(al_hreq.buflen, GFP_KERNEL); - if (!kbuf) - goto out_dput; - - cursor = (attrlist_cursor_kern_t *)&al_hreq.pos; - error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen, - al_hreq.flags, cursor); - if (error) - goto out_kfree; - - if (copy_to_user(al_hreq.buffer, kbuf, al_hreq.buflen)) - error = -EFAULT; - - out_kfree: - kfree(kbuf); - out_dput: - dput(dentry); - return error; -} - -int -xfs_attrmulti_attr_get( - struct inode *inode, - unsigned char *name, - unsigned char __user *ubuf, - __uint32_t *len, - __uint32_t flags) -{ - unsigned char *kbuf; - int error = EFAULT; - - if (*len > XATTR_SIZE_MAX) - return EINVAL; - kbuf = kmalloc(*len, GFP_KERNEL); - if (!kbuf) - return ENOMEM; - - error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags); - if (error) - goto out_kfree; - - if (copy_to_user(ubuf, kbuf, *len)) - error = EFAULT; - - out_kfree: - kfree(kbuf); - return error; -} - -int -xfs_attrmulti_attr_set( - struct inode *inode, - unsigned char *name, - const unsigned char __user *ubuf, - __uint32_t len, - __uint32_t flags) -{ - unsigned char *kbuf; - int error = EFAULT; - - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return EPERM; - if (len > XATTR_SIZE_MAX) - return EINVAL; - - kbuf = memdup_user(ubuf, len); - if (IS_ERR(kbuf)) - return PTR_ERR(kbuf); - - error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); - - return error; -} - -int -xfs_attrmulti_attr_remove( - struct inode *inode, - unsigned char *name, - __uint32_t flags) -{ - if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - return EPERM; - return xfs_attr_remove(XFS_I(inode), name, flags); -} - -STATIC int -xfs_attrmulti_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - xfs_attr_multiop_t *ops; - xfs_fsop_attrmulti_handlereq_t am_hreq; - struct dentry *dentry; - unsigned int i, size; - unsigned char *attr_name; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) - return -XFS_ERROR(EFAULT); - - /* overflow check */ - if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t)) - return -E2BIG; - - dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = E2BIG; - size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); - if (!size || size > 16 * PAGE_SIZE) - goto out_dput; - - ops = memdup_user(am_hreq.ops, size); - if (IS_ERR(ops)) { - error = PTR_ERR(ops); - goto out_dput; - } - - attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); - if (!attr_name) - goto out_kfree_ops; - - error = 0; - for (i = 0; i < am_hreq.opcount; i++) { - ops[i].am_error = strncpy_from_user((char *)attr_name, - ops[i].am_attrname, MAXNAMELEN); - if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) - error = -ERANGE; - if (ops[i].am_error < 0) - break; - - switch (ops[i].am_opcode) { - case ATTR_OP_GET: - ops[i].am_error = xfs_attrmulti_attr_get( - dentry->d_inode, attr_name, - ops[i].am_attrvalue, &ops[i].am_length, - ops[i].am_flags); - break; - case ATTR_OP_SET: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_set( - dentry->d_inode, attr_name, - ops[i].am_attrvalue, ops[i].am_length, - ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); - break; - case ATTR_OP_REMOVE: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_remove( - dentry->d_inode, attr_name, - ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); - break; - default: - ops[i].am_error = EINVAL; - } - } - - if (copy_to_user(am_hreq.ops, ops, size)) - error = XFS_ERROR(EFAULT); - - kfree(attr_name); - out_kfree_ops: - kfree(ops); - out_dput: - dput(dentry); - return -error; -} - -int -xfs_ioc_space( - struct xfs_inode *ip, - struct inode *inode, - struct file *filp, - int ioflags, - unsigned int cmd, - xfs_flock64_t *bf) -{ - int attr_flags = 0; - int error; - - /* - * Only allow the sys admin to reserve space unless - * unwritten extents are enabled. - */ - if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) && - !capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) - return -XFS_ERROR(EPERM); - - if (!(filp->f_mode & FMODE_WRITE)) - return -XFS_ERROR(EBADF); - - if (!S_ISREG(inode->i_mode)) - return -XFS_ERROR(EINVAL); - - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - attr_flags |= XFS_ATTR_NONBLOCK; - - if (filp->f_flags & O_DSYNC) - attr_flags |= XFS_ATTR_SYNC; - - if (ioflags & IO_INVIS) - attr_flags |= XFS_ATTR_DMI; - - error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags); - return -error; -} - -STATIC int -xfs_ioc_bulkstat( - xfs_mount_t *mp, - unsigned int cmd, - void __user *arg) -{ - xfs_fsop_bulkreq_t bulkreq; - int count; /* # of records returned */ - xfs_ino_t inlast; /* last inode number */ - int done; - int error; - - /* done = 1 if there are more stats to get and if bulkstat */ - /* should be called again (unused here, but used in dmapi) */ - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); - - if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t))) - return -XFS_ERROR(EFAULT); - - if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64))) - return -XFS_ERROR(EFAULT); - - if ((count = bulkreq.icount) <= 0) - return -XFS_ERROR(EINVAL); - - if (bulkreq.ubuffer == NULL) - return -XFS_ERROR(EINVAL); - - if (cmd == XFS_IOC_FSINUMBERS) - error = xfs_inumbers(mp, &inlast, &count, - bulkreq.ubuffer, xfs_inumbers_fmt); - else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE) - error = xfs_bulkstat_single(mp, &inlast, - bulkreq.ubuffer, &done); - else /* XFS_IOC_FSBULKSTAT */ - error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one, - sizeof(xfs_bstat_t), bulkreq.ubuffer, - &done); - - if (error) - return -error; - - if (bulkreq.ocount != NULL) { - if (copy_to_user(bulkreq.lastip, &inlast, - sizeof(xfs_ino_t))) - return -XFS_ERROR(EFAULT); - - if (copy_to_user(bulkreq.ocount, &count, sizeof(count))) - return -XFS_ERROR(EFAULT); - } - - return 0; -} - -STATIC int -xfs_ioc_fsgeometry_v1( - xfs_mount_t *mp, - void __user *arg) -{ - xfs_fsop_geom_t fsgeo; - int error; - - error = xfs_fs_geometry(mp, &fsgeo, 3); - if (error) - return -error; - - /* - * Caller should have passed an argument of type - * xfs_fsop_geom_v1_t. This is a proper subset of the - * xfs_fsop_geom_t that xfs_fs_geometry() fills in. - */ - if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t))) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_ioc_fsgeometry( - xfs_mount_t *mp, - void __user *arg) -{ - xfs_fsop_geom_t fsgeo; - int error; - - error = xfs_fs_geometry(mp, &fsgeo, 4); - if (error) - return -error; - - if (copy_to_user(arg, &fsgeo, sizeof(fsgeo))) - return -XFS_ERROR(EFAULT); - return 0; -} - -/* - * Linux extended inode flags interface. - */ - -STATIC unsigned int -xfs_merge_ioc_xflags( - unsigned int flags, - unsigned int start) -{ - unsigned int xflags = start; - - if (flags & FS_IMMUTABLE_FL) - xflags |= XFS_XFLAG_IMMUTABLE; - else - xflags &= ~XFS_XFLAG_IMMUTABLE; - if (flags & FS_APPEND_FL) - xflags |= XFS_XFLAG_APPEND; - else - xflags &= ~XFS_XFLAG_APPEND; - if (flags & FS_SYNC_FL) - xflags |= XFS_XFLAG_SYNC; - else - xflags &= ~XFS_XFLAG_SYNC; - if (flags & FS_NOATIME_FL) - xflags |= XFS_XFLAG_NOATIME; - else - xflags &= ~XFS_XFLAG_NOATIME; - if (flags & FS_NODUMP_FL) - xflags |= XFS_XFLAG_NODUMP; - else - xflags &= ~XFS_XFLAG_NODUMP; - - return xflags; -} - -STATIC unsigned int -xfs_di2lxflags( - __uint16_t di_flags) -{ - unsigned int flags = 0; - - if (di_flags & XFS_DIFLAG_IMMUTABLE) - flags |= FS_IMMUTABLE_FL; - if (di_flags & XFS_DIFLAG_APPEND) - flags |= FS_APPEND_FL; - if (di_flags & XFS_DIFLAG_SYNC) - flags |= FS_SYNC_FL; - if (di_flags & XFS_DIFLAG_NOATIME) - flags |= FS_NOATIME_FL; - if (di_flags & XFS_DIFLAG_NODUMP) - flags |= FS_NODUMP_FL; - return flags; -} - -STATIC int -xfs_ioc_fsgetxattr( - xfs_inode_t *ip, - int attr, - void __user *arg) -{ - struct fsxattr fa; - - memset(&fa, 0, sizeof(struct fsxattr)); - - xfs_ilock(ip, XFS_ILOCK_SHARED); - fa.fsx_xflags = xfs_ip2xflags(ip); - fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog; - fa.fsx_projid = xfs_get_projid(ip); - - if (attr) { - if (ip->i_afp) { - if (ip->i_afp->if_flags & XFS_IFEXTENTS) - fa.fsx_nextents = ip->i_afp->if_bytes / - sizeof(xfs_bmbt_rec_t); - else - fa.fsx_nextents = ip->i_d.di_anextents; - } else - fa.fsx_nextents = 0; - } else { - if (ip->i_df.if_flags & XFS_IFEXTENTS) - fa.fsx_nextents = ip->i_df.if_bytes / - sizeof(xfs_bmbt_rec_t); - else - fa.fsx_nextents = ip->i_d.di_nextents; - } - xfs_iunlock(ip, XFS_ILOCK_SHARED); - - if (copy_to_user(arg, &fa, sizeof(fa))) - return -EFAULT; - return 0; -} - -STATIC void -xfs_set_diflags( - struct xfs_inode *ip, - unsigned int xflags) -{ - unsigned int di_flags; - - /* can't set PREALLOC this way, just preserve it */ - di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); - if (xflags & XFS_XFLAG_IMMUTABLE) - di_flags |= XFS_DIFLAG_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) - di_flags |= XFS_DIFLAG_APPEND; - if (xflags & XFS_XFLAG_SYNC) - di_flags |= XFS_DIFLAG_SYNC; - if (xflags & XFS_XFLAG_NOATIME) - di_flags |= XFS_DIFLAG_NOATIME; - if (xflags & XFS_XFLAG_NODUMP) - di_flags |= XFS_DIFLAG_NODUMP; - if (xflags & XFS_XFLAG_PROJINHERIT) - di_flags |= XFS_DIFLAG_PROJINHERIT; - if (xflags & XFS_XFLAG_NODEFRAG) - di_flags |= XFS_DIFLAG_NODEFRAG; - if (xflags & XFS_XFLAG_FILESTREAM) - di_flags |= XFS_DIFLAG_FILESTREAM; - if (S_ISDIR(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_RTINHERIT) - di_flags |= XFS_DIFLAG_RTINHERIT; - if (xflags & XFS_XFLAG_NOSYMLINKS) - di_flags |= XFS_DIFLAG_NOSYMLINKS; - if (xflags & XFS_XFLAG_EXTSZINHERIT) - di_flags |= XFS_DIFLAG_EXTSZINHERIT; - } else if (S_ISREG(ip->i_d.di_mode)) { - if (xflags & XFS_XFLAG_REALTIME) - di_flags |= XFS_DIFLAG_REALTIME; - if (xflags & XFS_XFLAG_EXTSIZE) - di_flags |= XFS_DIFLAG_EXTSIZE; - } - - ip->i_d.di_flags = di_flags; -} - -STATIC void -xfs_diflags_to_linux( - struct xfs_inode *ip) -{ - struct inode *inode = VFS_I(ip); - unsigned int xflags = xfs_ip2xflags(ip); - - if (xflags & XFS_XFLAG_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - if (xflags & XFS_XFLAG_APPEND) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; - if (xflags & XFS_XFLAG_SYNC) - inode->i_flags |= S_SYNC; - else - inode->i_flags &= ~S_SYNC; - if (xflags & XFS_XFLAG_NOATIME) - inode->i_flags |= S_NOATIME; - else - inode->i_flags &= ~S_NOATIME; -} - -#define FSX_PROJID 1 -#define FSX_EXTSIZE 2 -#define FSX_XFLAGS 4 -#define FSX_NONBLOCK 8 - -STATIC int -xfs_ioctl_setattr( - xfs_inode_t *ip, - struct fsxattr *fa, - int mask) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_trans *tp; - unsigned int lock_flags = 0; - struct xfs_dquot *udqp = NULL; - struct xfs_dquot *gdqp = NULL; - struct xfs_dquot *olddquot = NULL; - int code; - - trace_xfs_ioctl_setattr(ip); - - if (mp->m_flags & XFS_MOUNT_RDONLY) - return XFS_ERROR(EROFS); - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - /* - * Disallow 32bit project ids when projid32bit feature is not enabled. - */ - if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) && - !xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb)) - return XFS_ERROR(EINVAL); - - /* - * If disk quotas is on, we make sure that the dquots do exist on disk, - * before we start any other transactions. Trying to do this later - * is messy. We don't care to take a readlock to look at the ids - * in inode here, because we can't hold it across the trans_reserve. - * If the IDs do change before we take the ilock, we're covered - * because the i_*dquot fields will get updated anyway. - */ - if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) { - code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid, - ip->i_d.di_gid, fa->fsx_projid, - XFS_QMOPT_PQUOTA, &udqp, &gdqp); - if (code) - return code; - } - - /* - * For the other attributes, we acquire the inode lock and - * first do an error checking pass. - */ - tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); - code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); - if (code) - goto error_return; - - lock_flags = XFS_ILOCK_EXCL; - xfs_ilock(ip, lock_flags); - - /* - * CAP_FOWNER overrides the following restrictions: - * - * The user ID of the calling process must be equal - * to the file owner ID, except in cases where the - * CAP_FSETID capability is applicable. - */ - if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) { - code = XFS_ERROR(EPERM); - goto error_return; - } - - /* - * Do a quota reservation only if projid is actually going to change. - */ - if (mask & FSX_PROJID) { - if (XFS_IS_QUOTA_RUNNING(mp) && - XFS_IS_PQUOTA_ON(mp) && - xfs_get_projid(ip) != fa->fsx_projid) { - ASSERT(tp); - code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, - capable(CAP_FOWNER) ? - XFS_QMOPT_FORCE_RES : 0); - if (code) /* out of quota */ - goto error_return; - } - } - - if (mask & FSX_EXTSIZE) { - /* - * Can't change extent size if any extents are allocated. - */ - if (ip->i_d.di_nextents && - ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != - fa->fsx_extsize)) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } - - /* - * Extent size must be a multiple of the appropriate block - * size, if set at all. It must also be smaller than the - * maximum extent size supported by the filesystem. - * - * Also, for non-realtime files, limit the extent size hint to - * half the size of the AGs in the filesystem so alignment - * doesn't result in extents larger than an AG. - */ - if (fa->fsx_extsize != 0) { - xfs_extlen_t size; - xfs_fsblock_t extsize_fsb; - - extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize); - if (extsize_fsb > MAXEXTLEN) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - - if (XFS_IS_REALTIME_INODE(ip) || - ((mask & FSX_XFLAGS) && - (fa->fsx_xflags & XFS_XFLAG_REALTIME))) { - size = mp->m_sb.sb_rextsize << - mp->m_sb.sb_blocklog; - } else { - size = mp->m_sb.sb_blocksize; - if (extsize_fsb > mp->m_sb.sb_agblocks / 2) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - - if (fa->fsx_extsize % size) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - } - - - if (mask & FSX_XFLAGS) { - /* - * Can't change realtime flag if any extents are allocated. - */ - if ((ip->i_d.di_nextents || ip->i_delayed_blks) && - (XFS_IS_REALTIME_INODE(ip)) != - (fa->fsx_xflags & XFS_XFLAG_REALTIME)) { - code = XFS_ERROR(EINVAL); /* EFBIG? */ - goto error_return; - } - - /* - * If realtime flag is set then must have realtime data. - */ - if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) { - if ((mp->m_sb.sb_rblocks == 0) || - (mp->m_sb.sb_rextsize == 0) || - (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) { - code = XFS_ERROR(EINVAL); - goto error_return; - } - } - - /* - * Can't modify an immutable/append-only file unless - * we have appropriate permission. - */ - if ((ip->i_d.di_flags & - (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) || - (fa->fsx_xflags & - (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && - !capable(CAP_LINUX_IMMUTABLE)) { - code = XFS_ERROR(EPERM); - goto error_return; - } - } - - xfs_trans_ijoin(tp, ip); - - /* - * Change file ownership. Must be the owner or privileged. - */ - if (mask & FSX_PROJID) { - /* - * CAP_FSETID overrides the following restrictions: - * - * The set-user-ID and set-group-ID bits of a file will be - * cleared upon successful return from chown() - */ - if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && - !capable(CAP_FSETID)) - ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); - - /* - * Change the ownerships and register quota modifications - * in the transaction. - */ - if (xfs_get_projid(ip) != fa->fsx_projid) { - if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) { - olddquot = xfs_qm_vop_chown(tp, ip, - &ip->i_gdquot, gdqp); - } - xfs_set_projid(ip, fa->fsx_projid); - - /* - * We may have to rev the inode as well as - * the superblock version number since projids didn't - * exist before DINODE_VERSION_2 and SB_VERSION_NLINK. - */ - if (ip->i_d.di_version == 1) - xfs_bump_ino_vers2(tp, ip); - } - - } - - if (mask & FSX_EXTSIZE) - ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog; - if (mask & FSX_XFLAGS) { - xfs_set_diflags(ip, fa->fsx_xflags); - xfs_diflags_to_linux(ip); - } - - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - XFS_STATS_INC(xs_ig_attrchg); - - /* - * If this is a synchronous mount, make sure that the - * transaction goes to disk before returning to the user. - * This is slightly sub-optimal in that truncates require - * two sync transactions instead of one for wsync filesystems. - * One for the truncate and one for the timestamps since we - * don't want to change the timestamps unless we're sure the - * truncate worked. Truncates are less than 1% of the laddis - * mix so this probably isn't worth the trouble to optimize. - */ - if (mp->m_flags & XFS_MOUNT_WSYNC) - xfs_trans_set_sync(tp); - code = xfs_trans_commit(tp, 0); - xfs_iunlock(ip, lock_flags); - - /* - * Release any dquot(s) the inode had kept before chown. - */ - xfs_qm_dqrele(olddquot); - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - - return code; - - error_return: - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - xfs_trans_cancel(tp, 0); - if (lock_flags) - xfs_iunlock(ip, lock_flags); - return code; -} - -STATIC int -xfs_ioc_fssetxattr( - xfs_inode_t *ip, - struct file *filp, - void __user *arg) -{ - struct fsxattr fa; - unsigned int mask; - - if (copy_from_user(&fa, arg, sizeof(fa))) - return -EFAULT; - - mask = FSX_XFLAGS | FSX_EXTSIZE | FSX_PROJID; - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - mask |= FSX_NONBLOCK; - - return -xfs_ioctl_setattr(ip, &fa, mask); -} - -STATIC int -xfs_ioc_getxflags( - xfs_inode_t *ip, - void __user *arg) -{ - unsigned int flags; - - flags = xfs_di2lxflags(ip->i_d.di_flags); - if (copy_to_user(arg, &flags, sizeof(flags))) - return -EFAULT; - return 0; -} - -STATIC int -xfs_ioc_setxflags( - xfs_inode_t *ip, - struct file *filp, - void __user *arg) -{ - struct fsxattr fa; - unsigned int flags; - unsigned int mask; - - if (copy_from_user(&flags, arg, sizeof(flags))) - return -EFAULT; - - if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ - FS_NOATIME_FL | FS_NODUMP_FL | \ - FS_SYNC_FL)) - return -EOPNOTSUPP; - - mask = FSX_XFLAGS; - if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) - mask |= FSX_NONBLOCK; - fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip)); - - return -xfs_ioctl_setattr(ip, &fa, mask); -} - -STATIC int -xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full) -{ - struct getbmap __user *base = *ap; - - /* copy only getbmap portion (not getbmapx) */ - if (copy_to_user(base, bmv, sizeof(struct getbmap))) - return XFS_ERROR(EFAULT); - - *ap += sizeof(struct getbmap); - return 0; -} - -STATIC int -xfs_ioc_getbmap( - struct xfs_inode *ip, - int ioflags, - unsigned int cmd, - void __user *arg) -{ - struct getbmapx bmx; - int error; - - if (copy_from_user(&bmx, arg, sizeof(struct getbmapx))) - return -XFS_ERROR(EFAULT); - - if (bmx.bmv_count < 2) - return -XFS_ERROR(EINVAL); - - bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0); - if (ioflags & IO_INVIS) - bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ; - - error = xfs_getbmap(ip, &bmx, xfs_getbmap_format, - (struct getbmap *)arg+1); - if (error) - return -error; - - /* copy back header - only size of getbmap */ - if (copy_to_user(arg, &bmx, sizeof(struct getbmap))) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full) -{ - struct getbmapx __user *base = *ap; - - if (copy_to_user(base, bmv, sizeof(struct getbmapx))) - return XFS_ERROR(EFAULT); - - *ap += sizeof(struct getbmapx); - return 0; -} - -STATIC int -xfs_ioc_getbmapx( - struct xfs_inode *ip, - void __user *arg) -{ - struct getbmapx bmx; - int error; - - if (copy_from_user(&bmx, arg, sizeof(bmx))) - return -XFS_ERROR(EFAULT); - - if (bmx.bmv_count < 2) - return -XFS_ERROR(EINVAL); - - if (bmx.bmv_iflags & (~BMV_IF_VALID)) - return -XFS_ERROR(EINVAL); - - error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format, - (struct getbmapx *)arg+1); - if (error) - return -error; - - /* copy back header */ - if (copy_to_user(arg, &bmx, sizeof(struct getbmapx))) - return -XFS_ERROR(EFAULT); - - return 0; -} - -/* - * Note: some of the ioctl's return positive numbers as a - * byte count indicating success, such as readlink_by_handle. - * So we don't "sign flip" like most other routines. This means - * true errors need to be returned as a negative value. - */ -long -xfs_file_ioctl( - struct file *filp, - unsigned int cmd, - unsigned long p) -{ - struct inode *inode = filp->f_path.dentry->d_inode; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - void __user *arg = (void __user *)p; - int ioflags = 0; - int error; - - if (filp->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - - trace_xfs_file_ioctl(ip); - - switch (cmd) { - case FITRIM: - return xfs_ioc_trim(mp, arg); - case XFS_IOC_ALLOCSP: - case XFS_IOC_FREESP: - case XFS_IOC_RESVSP: - case XFS_IOC_UNRESVSP: - case XFS_IOC_ALLOCSP64: - case XFS_IOC_FREESP64: - case XFS_IOC_RESVSP64: - case XFS_IOC_UNRESVSP64: - case XFS_IOC_ZERO_RANGE: { - xfs_flock64_t bf; - - if (copy_from_user(&bf, arg, sizeof(bf))) - return -XFS_ERROR(EFAULT); - return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf); - } - case XFS_IOC_DIOINFO: { - struct dioattr da; - xfs_buftarg_t *target = - XFS_IS_REALTIME_INODE(ip) ? - mp->m_rtdev_targp : mp->m_ddev_targp; - - da.d_mem = da.d_miniosz = 1 << target->bt_sshift; - da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1); - - if (copy_to_user(arg, &da, sizeof(da))) - return -XFS_ERROR(EFAULT); - return 0; - } - - case XFS_IOC_FSBULKSTAT_SINGLE: - case XFS_IOC_FSBULKSTAT: - case XFS_IOC_FSINUMBERS: - return xfs_ioc_bulkstat(mp, cmd, arg); - - case XFS_IOC_FSGEOMETRY_V1: - return xfs_ioc_fsgeometry_v1(mp, arg); - - case XFS_IOC_FSGEOMETRY: - return xfs_ioc_fsgeometry(mp, arg); - - case XFS_IOC_GETVERSION: - return put_user(inode->i_generation, (int __user *)arg); - - case XFS_IOC_FSGETXATTR: - return xfs_ioc_fsgetxattr(ip, 0, arg); - case XFS_IOC_FSGETXATTRA: - return xfs_ioc_fsgetxattr(ip, 1, arg); - case XFS_IOC_FSSETXATTR: - return xfs_ioc_fssetxattr(ip, filp, arg); - case XFS_IOC_GETXFLAGS: - return xfs_ioc_getxflags(ip, arg); - case XFS_IOC_SETXFLAGS: - return xfs_ioc_setxflags(ip, filp, arg); - - case XFS_IOC_FSSETDM: { - struct fsdmidata dmi; - - if (copy_from_user(&dmi, arg, sizeof(dmi))) - return -XFS_ERROR(EFAULT); - - error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask, - dmi.fsd_dmstate); - return -error; - } - - case XFS_IOC_GETBMAP: - case XFS_IOC_GETBMAPA: - return xfs_ioc_getbmap(ip, ioflags, cmd, arg); - - case XFS_IOC_GETBMAPX: - return xfs_ioc_getbmapx(ip, arg); - - case XFS_IOC_FD_TO_HANDLE: - case XFS_IOC_PATH_TO_HANDLE: - case XFS_IOC_PATH_TO_FSHANDLE: { - xfs_fsop_handlereq_t hreq; - - if (copy_from_user(&hreq, arg, sizeof(hreq))) - return -XFS_ERROR(EFAULT); - return xfs_find_handle(cmd, &hreq); - } - case XFS_IOC_OPEN_BY_HANDLE: { - xfs_fsop_handlereq_t hreq; - - if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) - return -XFS_ERROR(EFAULT); - return xfs_open_by_handle(filp, &hreq); - } - case XFS_IOC_FSSETDM_BY_HANDLE: - return xfs_fssetdm_by_handle(filp, arg); - - case XFS_IOC_READLINK_BY_HANDLE: { - xfs_fsop_handlereq_t hreq; - - if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) - return -XFS_ERROR(EFAULT); - return xfs_readlink_by_handle(filp, &hreq); - } - case XFS_IOC_ATTRLIST_BY_HANDLE: - return xfs_attrlist_by_handle(filp, arg); - - case XFS_IOC_ATTRMULTI_BY_HANDLE: - return xfs_attrmulti_by_handle(filp, arg); - - case XFS_IOC_SWAPEXT: { - struct xfs_swapext sxp; - - if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t))) - return -XFS_ERROR(EFAULT); - error = xfs_swapext(&sxp); - return -error; - } - - case XFS_IOC_FSCOUNTS: { - xfs_fsop_counts_t out; - - error = xfs_fs_counts(mp, &out); - if (error) - return -error; - - if (copy_to_user(arg, &out, sizeof(out))) - return -XFS_ERROR(EFAULT); - return 0; - } - - case XFS_IOC_SET_RESBLKS: { - xfs_fsop_resblks_t inout; - __uint64_t in; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (mp->m_flags & XFS_MOUNT_RDONLY) - return -XFS_ERROR(EROFS); - - if (copy_from_user(&inout, arg, sizeof(inout))) - return -XFS_ERROR(EFAULT); - - /* input parameter is passed in resblks field of structure */ - in = inout.resblks; - error = xfs_reserve_blocks(mp, &in, &inout); - if (error) - return -error; - - if (copy_to_user(arg, &inout, sizeof(inout))) - return -XFS_ERROR(EFAULT); - return 0; - } - - case XFS_IOC_GET_RESBLKS: { - xfs_fsop_resblks_t out; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - error = xfs_reserve_blocks(mp, NULL, &out); - if (error) - return -error; - - if (copy_to_user(arg, &out, sizeof(out))) - return -XFS_ERROR(EFAULT); - - return 0; - } - - case XFS_IOC_FSGROWFSDATA: { - xfs_growfs_data_t in; - - if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); - - error = xfs_growfs_data(mp, &in); - return -error; - } - - case XFS_IOC_FSGROWFSLOG: { - xfs_growfs_log_t in; - - if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); - - error = xfs_growfs_log(mp, &in); - return -error; - } - - case XFS_IOC_FSGROWFSRT: { - xfs_growfs_rt_t in; - - if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); - - error = xfs_growfs_rt(mp, &in); - return -error; - } - - case XFS_IOC_GOINGDOWN: { - __uint32_t in; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (get_user(in, (__uint32_t __user *)arg)) - return -XFS_ERROR(EFAULT); - - error = xfs_fs_goingdown(mp, in); - return -error; - } - - case XFS_IOC_ERROR_INJECTION: { - xfs_error_injection_t in; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - if (copy_from_user(&in, arg, sizeof(in))) - return -XFS_ERROR(EFAULT); - - error = xfs_errortag_add(in.errtag, mp); - return -error; - } - - case XFS_IOC_ERROR_CLEARALL: - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - error = xfs_errortag_clearall(mp, 1); - return -error; - - default: - return -ENOTTY; - } -} diff --git a/fs/xfs/linux-2.6/xfs_ioctl.h b/fs/xfs/linux-2.6/xfs_ioctl.h deleted file mode 100644 index d56173b34a2a..000000000000 --- a/fs/xfs/linux-2.6/xfs_ioctl.h +++ /dev/null @@ -1,85 +0,0 @@ -/* - * Copyright (c) 2008 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_IOCTL_H__ -#define __XFS_IOCTL_H__ - -extern int -xfs_ioc_space( - struct xfs_inode *ip, - struct inode *inode, - struct file *filp, - int ioflags, - unsigned int cmd, - xfs_flock64_t *bf); - -extern int -xfs_find_handle( - unsigned int cmd, - xfs_fsop_handlereq_t *hreq); - -extern int -xfs_open_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq); - -extern int -xfs_readlink_by_handle( - struct file *parfilp, - xfs_fsop_handlereq_t *hreq); - -extern int -xfs_attrmulti_attr_get( - struct inode *inode, - unsigned char *name, - unsigned char __user *ubuf, - __uint32_t *len, - __uint32_t flags); - -extern int -xfs_attrmulti_attr_set( - struct inode *inode, - unsigned char *name, - const unsigned char __user *ubuf, - __uint32_t len, - __uint32_t flags); - -extern int -xfs_attrmulti_attr_remove( - struct inode *inode, - unsigned char *name, - __uint32_t flags); - -extern struct dentry * -xfs_handle_to_dentry( - struct file *parfilp, - void __user *uhandle, - u32 hlen); - -extern long -xfs_file_ioctl( - struct file *filp, - unsigned int cmd, - unsigned long p); - -extern long -xfs_file_compat_ioctl( - struct file *file, - unsigned int cmd, - unsigned long arg); - -#endif diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c deleted file mode 100644 index 54e623bfbb85..000000000000 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ /dev/null @@ -1,672 +0,0 @@ -/* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include -#include -#include -#include -#include -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_vnode.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_itable.h" -#include "xfs_error.h" -#include "xfs_dfrag.h" -#include "xfs_vnodeops.h" -#include "xfs_fsops.h" -#include "xfs_alloc.h" -#include "xfs_rtalloc.h" -#include "xfs_attr.h" -#include "xfs_ioctl.h" -#include "xfs_ioctl32.h" -#include "xfs_trace.h" - -#define _NATIVE_IOC(cmd, type) \ - _IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type)) - -#ifdef BROKEN_X86_ALIGNMENT -STATIC int -xfs_compat_flock64_copyin( - xfs_flock64_t *bf, - compat_xfs_flock64_t __user *arg32) -{ - if (get_user(bf->l_type, &arg32->l_type) || - get_user(bf->l_whence, &arg32->l_whence) || - get_user(bf->l_start, &arg32->l_start) || - get_user(bf->l_len, &arg32->l_len) || - get_user(bf->l_sysid, &arg32->l_sysid) || - get_user(bf->l_pid, &arg32->l_pid) || - copy_from_user(bf->l_pad, &arg32->l_pad, 4*sizeof(u32))) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_compat_ioc_fsgeometry_v1( - struct xfs_mount *mp, - compat_xfs_fsop_geom_v1_t __user *arg32) -{ - xfs_fsop_geom_t fsgeo; - int error; - - error = xfs_fs_geometry(mp, &fsgeo, 3); - if (error) - return -error; - /* The 32-bit variant simply has some padding at the end */ - if (copy_to_user(arg32, &fsgeo, sizeof(struct compat_xfs_fsop_geom_v1))) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_compat_growfs_data_copyin( - struct xfs_growfs_data *in, - compat_xfs_growfs_data_t __user *arg32) -{ - if (get_user(in->newblocks, &arg32->newblocks) || - get_user(in->imaxpct, &arg32->imaxpct)) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_compat_growfs_rt_copyin( - struct xfs_growfs_rt *in, - compat_xfs_growfs_rt_t __user *arg32) -{ - if (get_user(in->newblocks, &arg32->newblocks) || - get_user(in->extsize, &arg32->extsize)) - return -XFS_ERROR(EFAULT); - return 0; -} - -STATIC int -xfs_inumbers_fmt_compat( - void __user *ubuffer, - const xfs_inogrp_t *buffer, - long count, - long *written) -{ - compat_xfs_inogrp_t __user *p32 = ubuffer; - long i; - - for (i = 0; i < count; i++) { - if (put_user(buffer[i].xi_startino, &p32[i].xi_startino) || - put_user(buffer[i].xi_alloccount, &p32[i].xi_alloccount) || - put_user(buffer[i].xi_allocmask, &p32[i].xi_allocmask)) - return -XFS_ERROR(EFAULT); - } - *written = count * sizeof(*p32); - return 0; -} - -#else -#define xfs_inumbers_fmt_compat xfs_inumbers_fmt -#endif /* BROKEN_X86_ALIGNMENT */ - -STATIC int -xfs_ioctl32_bstime_copyin( - xfs_bstime_t *bstime, - compat_xfs_bstime_t __user *bstime32) -{ - compat_time_t sec32; /* tv_sec differs on 64 vs. 32 */ - - if (get_user(sec32, &bstime32->tv_sec) || - get_user(bstime->tv_nsec, &bstime32->tv_nsec)) - return -XFS_ERROR(EFAULT); - bstime->tv_sec = sec32; - return 0; -} - -/* xfs_bstat_t has differing alignment on intel, & bstime_t sizes everywhere */ -STATIC int -xfs_ioctl32_bstat_copyin( - xfs_bstat_t *bstat, - compat_xfs_bstat_t __user *bstat32) -{ - if (get_user(bstat->bs_ino, &bstat32->bs_ino) || - get_user(bstat->bs_mode, &bstat32->bs_mode) || - get_user(bstat->bs_nlink, &bstat32->bs_nlink) || - get_user(bstat->bs_uid, &bstat32->bs_uid) || - get_user(bstat->bs_gid, &bstat32->bs_gid) || - get_user(bstat->bs_rdev, &bstat32->bs_rdev) || - get_user(bstat->bs_blksize, &bstat32->bs_blksize) || - get_user(bstat->bs_size, &bstat32->bs_size) || - xfs_ioctl32_bstime_copyin(&bstat->bs_atime, &bstat32->bs_atime) || - xfs_ioctl32_bstime_copyin(&bstat->bs_mtime, &bstat32->bs_mtime) || - xfs_ioctl32_bstime_copyin(&bstat->bs_ctime, &bstat32->bs_ctime) || - get_user(bstat->bs_blocks, &bstat32->bs_size) || - get_user(bstat->bs_xflags, &bstat32->bs_size) || - get_user(bstat->bs_extsize, &bstat32->bs_extsize) || - get_user(bstat->bs_extents, &bstat32->bs_extents) || - get_user(bstat->bs_gen, &bstat32->bs_gen) || - get_user(bstat->bs_projid_lo, &bstat32->bs_projid_lo) || - get_user(bstat->bs_projid_hi, &bstat32->bs_projid_hi) || - get_user(bstat->bs_dmevmask, &bstat32->bs_dmevmask) || - get_user(bstat->bs_dmstate, &bstat32->bs_dmstate) || - get_user(bstat->bs_aextents, &bstat32->bs_aextents)) - return -XFS_ERROR(EFAULT); - return 0; -} - -/* XFS_IOC_FSBULKSTAT and friends */ - -STATIC int -xfs_bstime_store_compat( - compat_xfs_bstime_t __user *p32, - const xfs_bstime_t *p) -{ - __s32 sec32; - - sec32 = p->tv_sec; - if (put_user(sec32, &p32->tv_sec) || - put_user(p->tv_nsec, &p32->tv_nsec)) - return -XFS_ERROR(EFAULT); - return 0; -} - -/* Return 0 on success or positive error (to xfs_bulkstat()) */ -STATIC int -xfs_bulkstat_one_fmt_compat( - void __user *ubuffer, - int ubsize, - int *ubused, - const xfs_bstat_t *buffer) -{ - compat_xfs_bstat_t __user *p32 = ubuffer; - - if (ubsize < sizeof(*p32)) - return XFS_ERROR(ENOMEM); - - if (put_user(buffer->bs_ino, &p32->bs_ino) || - put_user(buffer->bs_mode, &p32->bs_mode) || - put_user(buffer->bs_nlink, &p32->bs_nlink) || - put_user(buffer->bs_uid, &p32->bs_uid) || - put_user(buffer->bs_gid, &p32->bs_gid) || - put_user(buffer->bs_rdev, &p32->bs_rdev) || - put_user(buffer->bs_blksize, &p32->bs_blksize) || - put_user(buffer->bs_size, &p32->bs_size) || - xfs_bstime_store_compat(&p32->bs_atime, &buffer->bs_atime) || - xfs_bstime_store_compat(&p32->bs_mtime, &buffer->bs_mtime) || - xfs_bstime_store_compat(&p32->bs_ctime, &buffer->bs_ctime) || - put_user(buffer->bs_blocks, &p32->bs_blocks) || - put_user(buffer->bs_xflags, &p32->bs_xflags) || - put_user(buffer->bs_extsize, &p32->bs_extsize) || - put_user(buffer->bs_extents, &p32->bs_extents) || - put_user(buffer->bs_gen, &p32->bs_gen) || - put_user(buffer->bs_projid, &p32->bs_projid) || - put_user(buffer->bs_projid_hi, &p32->bs_projid_hi) || - put_user(buffer->bs_dmevmask, &p32->bs_dmevmask) || - put_user(buffer->bs_dmstate, &p32->bs_dmstate) || - put_user(buffer->bs_aextents, &p32->bs_aextents)) - return XFS_ERROR(EFAULT); - if (ubused) - *ubused = sizeof(*p32); - return 0; -} - -STATIC int -xfs_bulkstat_one_compat( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t ino, /* inode number to get data for */ - void __user *buffer, /* buffer to place output in */ - int ubsize, /* size of buffer */ - int *ubused, /* bytes used by me */ - int *stat) /* BULKSTAT_RV_... */ -{ - return xfs_bulkstat_one_int(mp, ino, buffer, ubsize, - xfs_bulkstat_one_fmt_compat, - ubused, stat); -} - -/* copied from xfs_ioctl.c */ -STATIC int -xfs_compat_ioc_bulkstat( - xfs_mount_t *mp, - unsigned int cmd, - compat_xfs_fsop_bulkreq_t __user *p32) -{ - u32 addr; - xfs_fsop_bulkreq_t bulkreq; - int count; /* # of records returned */ - xfs_ino_t inlast; /* last inode number */ - int done; - int error; - - /* done = 1 if there are more stats to get and if bulkstat */ - /* should be called again (unused here, but used in dmapi) */ - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - - if (XFS_FORCED_SHUTDOWN(mp)) - return -XFS_ERROR(EIO); - - if (get_user(addr, &p32->lastip)) - return -XFS_ERROR(EFAULT); - bulkreq.lastip = compat_ptr(addr); - if (get_user(bulkreq.icount, &p32->icount) || - get_user(addr, &p32->ubuffer)) - return -XFS_ERROR(EFAULT); - bulkreq.ubuffer = compat_ptr(addr); - if (get_user(addr, &p32->ocount)) - return -XFS_ERROR(EFAULT); - bulkreq.ocount = compat_ptr(addr); - - if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64))) - return -XFS_ERROR(EFAULT); - - if ((count = bulkreq.icount) <= 0) - return -XFS_ERROR(EINVAL); - - if (bulkreq.ubuffer == NULL) - return -XFS_ERROR(EINVAL); - - if (cmd == XFS_IOC_FSINUMBERS_32) { - error = xfs_inumbers(mp, &inlast, &count, - bulkreq.ubuffer, xfs_inumbers_fmt_compat); - } else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE_32) { - int res; - - error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer, - sizeof(compat_xfs_bstat_t), 0, &res); - } else if (cmd == XFS_IOC_FSBULKSTAT_32) { - error = xfs_bulkstat(mp, &inlast, &count, - xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t), - bulkreq.ubuffer, &done); - } else - error = XFS_ERROR(EINVAL); - if (error) - return -error; - - if (bulkreq.ocount != NULL) { - if (copy_to_user(bulkreq.lastip, &inlast, - sizeof(xfs_ino_t))) - return -XFS_ERROR(EFAULT); - - if (copy_to_user(bulkreq.ocount, &count, sizeof(count))) - return -XFS_ERROR(EFAULT); - } - - return 0; -} - -STATIC int -xfs_compat_handlereq_copyin( - xfs_fsop_handlereq_t *hreq, - compat_xfs_fsop_handlereq_t __user *arg32) -{ - compat_xfs_fsop_handlereq_t hreq32; - - if (copy_from_user(&hreq32, arg32, sizeof(compat_xfs_fsop_handlereq_t))) - return -XFS_ERROR(EFAULT); - - hreq->fd = hreq32.fd; - hreq->path = compat_ptr(hreq32.path); - hreq->oflags = hreq32.oflags; - hreq->ihandle = compat_ptr(hreq32.ihandle); - hreq->ihandlen = hreq32.ihandlen; - hreq->ohandle = compat_ptr(hreq32.ohandle); - hreq->ohandlen = compat_ptr(hreq32.ohandlen); - - return 0; -} - -STATIC struct dentry * -xfs_compat_handlereq_to_dentry( - struct file *parfilp, - compat_xfs_fsop_handlereq_t *hreq) -{ - return xfs_handle_to_dentry(parfilp, - compat_ptr(hreq->ihandle), hreq->ihandlen); -} - -STATIC int -xfs_compat_attrlist_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - attrlist_cursor_kern_t *cursor; - compat_xfs_fsop_attrlist_handlereq_t al_hreq; - struct dentry *dentry; - char *kbuf; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&al_hreq, arg, - sizeof(compat_xfs_fsop_attrlist_handlereq_t))) - return -XFS_ERROR(EFAULT); - if (al_hreq.buflen > XATTR_LIST_MAX) - return -XFS_ERROR(EINVAL); - - /* - * Reject flags, only allow namespaces. - */ - if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) - return -XFS_ERROR(EINVAL); - - dentry = xfs_compat_handlereq_to_dentry(parfilp, &al_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = -ENOMEM; - kbuf = kmalloc(al_hreq.buflen, GFP_KERNEL); - if (!kbuf) - goto out_dput; - - cursor = (attrlist_cursor_kern_t *)&al_hreq.pos; - error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen, - al_hreq.flags, cursor); - if (error) - goto out_kfree; - - if (copy_to_user(compat_ptr(al_hreq.buffer), kbuf, al_hreq.buflen)) - error = -EFAULT; - - out_kfree: - kfree(kbuf); - out_dput: - dput(dentry); - return error; -} - -STATIC int -xfs_compat_attrmulti_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - compat_xfs_attr_multiop_t *ops; - compat_xfs_fsop_attrmulti_handlereq_t am_hreq; - struct dentry *dentry; - unsigned int i, size; - unsigned char *attr_name; - - if (!capable(CAP_SYS_ADMIN)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&am_hreq, arg, - sizeof(compat_xfs_fsop_attrmulti_handlereq_t))) - return -XFS_ERROR(EFAULT); - - /* overflow check */ - if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t)) - return -E2BIG; - - dentry = xfs_compat_handlereq_to_dentry(parfilp, &am_hreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - error = E2BIG; - size = am_hreq.opcount * sizeof(compat_xfs_attr_multiop_t); - if (!size || size > 16 * PAGE_SIZE) - goto out_dput; - - ops = memdup_user(compat_ptr(am_hreq.ops), size); - if (IS_ERR(ops)) { - error = PTR_ERR(ops); - goto out_dput; - } - - attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); - if (!attr_name) - goto out_kfree_ops; - - error = 0; - for (i = 0; i < am_hreq.opcount; i++) { - ops[i].am_error = strncpy_from_user((char *)attr_name, - compat_ptr(ops[i].am_attrname), - MAXNAMELEN); - if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) - error = -ERANGE; - if (ops[i].am_error < 0) - break; - - switch (ops[i].am_opcode) { - case ATTR_OP_GET: - ops[i].am_error = xfs_attrmulti_attr_get( - dentry->d_inode, attr_name, - compat_ptr(ops[i].am_attrvalue), - &ops[i].am_length, ops[i].am_flags); - break; - case ATTR_OP_SET: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_set( - dentry->d_inode, attr_name, - compat_ptr(ops[i].am_attrvalue), - ops[i].am_length, ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); - break; - case ATTR_OP_REMOVE: - ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); - if (ops[i].am_error) - break; - ops[i].am_error = xfs_attrmulti_attr_remove( - dentry->d_inode, attr_name, - ops[i].am_flags); - mnt_drop_write(parfilp->f_path.mnt); - break; - default: - ops[i].am_error = EINVAL; - } - } - - if (copy_to_user(compat_ptr(am_hreq.ops), ops, size)) - error = XFS_ERROR(EFAULT); - - kfree(attr_name); - out_kfree_ops: - kfree(ops); - out_dput: - dput(dentry); - return -error; -} - -STATIC int -xfs_compat_fssetdm_by_handle( - struct file *parfilp, - void __user *arg) -{ - int error; - struct fsdmidata fsd; - compat_xfs_fsop_setdm_handlereq_t dmhreq; - struct dentry *dentry; - - if (!capable(CAP_MKNOD)) - return -XFS_ERROR(EPERM); - if (copy_from_user(&dmhreq, arg, - sizeof(compat_xfs_fsop_setdm_handlereq_t))) - return -XFS_ERROR(EFAULT); - - dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq); - if (IS_ERR(dentry)) - return PTR_ERR(dentry); - - if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) { - error = -XFS_ERROR(EPERM); - goto out; - } - - if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) { - error = -XFS_ERROR(EFAULT); - goto out; - } - - error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask, - fsd.fsd_dmstate); - -out: - dput(dentry); - return error; -} - -long -xfs_file_compat_ioctl( - struct file *filp, - unsigned cmd, - unsigned long p) -{ - struct inode *inode = filp->f_path.dentry->d_inode; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - void __user *arg = (void __user *)p; - int ioflags = 0; - int error; - - if (filp->f_mode & FMODE_NOCMTIME) - ioflags |= IO_INVIS; - - trace_xfs_file_compat_ioctl(ip); - - switch (cmd) { - /* No size or alignment issues on any arch */ - case XFS_IOC_DIOINFO: - case XFS_IOC_FSGEOMETRY: - case XFS_IOC_FSGETXATTR: - case XFS_IOC_FSSETXATTR: - case XFS_IOC_FSGETXATTRA: - case XFS_IOC_FSSETDM: - case XFS_IOC_GETBMAP: - case XFS_IOC_GETBMAPA: - case XFS_IOC_GETBMAPX: - case XFS_IOC_FSCOUNTS: - case XFS_IOC_SET_RESBLKS: - case XFS_IOC_GET_RESBLKS: - case XFS_IOC_FSGROWFSLOG: - case XFS_IOC_GOINGDOWN: - case XFS_IOC_ERROR_INJECTION: - case XFS_IOC_ERROR_CLEARALL: - return xfs_file_ioctl(filp, cmd, p); -#ifndef BROKEN_X86_ALIGNMENT - /* These are handled fine if no alignment issues */ - case XFS_IOC_ALLOCSP: - case XFS_IOC_FREESP: - case XFS_IOC_RESVSP: - case XFS_IOC_UNRESVSP: - case XFS_IOC_ALLOCSP64: - case XFS_IOC_FREESP64: - case XFS_IOC_RESVSP64: - case XFS_IOC_UNRESVSP64: - case XFS_IOC_FSGEOMETRY_V1: - case XFS_IOC_FSGROWFSDATA: - case XFS_IOC_FSGROWFSRT: - case XFS_IOC_ZERO_RANGE: - return xfs_file_ioctl(filp, cmd, p); -#else - case XFS_IOC_ALLOCSP_32: - case XFS_IOC_FREESP_32: - case XFS_IOC_ALLOCSP64_32: - case XFS_IOC_FREESP64_32: - case XFS_IOC_RESVSP_32: - case XFS_IOC_UNRESVSP_32: - case XFS_IOC_RESVSP64_32: - case XFS_IOC_UNRESVSP64_32: - case XFS_IOC_ZERO_RANGE_32: { - struct xfs_flock64 bf; - - if (xfs_compat_flock64_copyin(&bf, arg)) - return -XFS_ERROR(EFAULT); - cmd = _NATIVE_IOC(cmd, struct xfs_flock64); - return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf); - } - case XFS_IOC_FSGEOMETRY_V1_32: - return xfs_compat_ioc_fsgeometry_v1(mp, arg); - case XFS_IOC_FSGROWFSDATA_32: { - struct xfs_growfs_data in; - - if (xfs_compat_growfs_data_copyin(&in, arg)) - return -XFS_ERROR(EFAULT); - error = xfs_growfs_data(mp, &in); - return -error; - } - case XFS_IOC_FSGROWFSRT_32: { - struct xfs_growfs_rt in; - - if (xfs_compat_growfs_rt_copyin(&in, arg)) - return -XFS_ERROR(EFAULT); - error = xfs_growfs_rt(mp, &in); - return -error; - } -#endif - /* long changes size, but xfs only copiese out 32 bits */ - case XFS_IOC_GETXFLAGS_32: - case XFS_IOC_SETXFLAGS_32: - case XFS_IOC_GETVERSION_32: - cmd = _NATIVE_IOC(cmd, long); - return xfs_file_ioctl(filp, cmd, p); - case XFS_IOC_SWAPEXT_32: { - struct xfs_swapext sxp; - struct compat_xfs_swapext __user *sxu = arg; - - /* Bulk copy in up to the sx_stat field, then copy bstat */ - if (copy_from_user(&sxp, sxu, - offsetof(struct xfs_swapext, sx_stat)) || - xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat)) - return -XFS_ERROR(EFAULT); - error = xfs_swapext(&sxp); - return -error; - } - case XFS_IOC_FSBULKSTAT_32: - case XFS_IOC_FSBULKSTAT_SINGLE_32: - case XFS_IOC_FSINUMBERS_32: - return xfs_compat_ioc_bulkstat(mp, cmd, arg); - case XFS_IOC_FD_TO_HANDLE_32: - case XFS_IOC_PATH_TO_HANDLE_32: - case XFS_IOC_PATH_TO_FSHANDLE_32: { - struct xfs_fsop_handlereq hreq; - - if (xfs_compat_handlereq_copyin(&hreq, arg)) - return -XFS_ERROR(EFAULT); - cmd = _NATIVE_IOC(cmd, struct xfs_fsop_handlereq); - return xfs_find_handle(cmd, &hreq); - } - case XFS_IOC_OPEN_BY_HANDLE_32: { - struct xfs_fsop_handlereq hreq; - - if (xfs_compat_handlereq_copyin(&hreq, arg)) - return -XFS_ERROR(EFAULT); - return xfs_open_by_handle(filp, &hreq); - } - case XFS_IOC_READLINK_BY_HANDLE_32: { - struct xfs_fsop_handlereq hreq; - - if (xfs_compat_handlereq_copyin(&hreq, arg)) - return -XFS_ERROR(EFAULT); - return xfs_readlink_by_handle(filp, &hreq); - } - case XFS_IOC_ATTRLIST_BY_HANDLE_32: - return xfs_compat_attrlist_by_handle(filp, arg); - case XFS_IOC_ATTRMULTI_BY_HANDLE_32: - return xfs_compat_attrmulti_by_handle(filp, arg); - case XFS_IOC_FSSETDM_BY_HANDLE_32: - return xfs_compat_fssetdm_by_handle(filp, arg); - default: - return -XFS_ERROR(ENOIOCTLCMD); - } -} diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.h b/fs/xfs/linux-2.6/xfs_ioctl32.h deleted file mode 100644 index 80f4060e8970..000000000000 --- a/fs/xfs/linux-2.6/xfs_ioctl32.h +++ /dev/null @@ -1,237 +0,0 @@ -/* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_IOCTL32_H__ -#define __XFS_IOCTL32_H__ - -#include - -/* - * on 32-bit arches, ioctl argument structures may have different sizes - * and/or alignment. We define compat structures which match the - * 32-bit sizes/alignments here, and their associated ioctl numbers. - * - * xfs_ioctl32.c contains routines to copy these structures in and out. - */ - -/* stock kernel-level ioctls we support */ -#define XFS_IOC_GETXFLAGS_32 FS_IOC32_GETFLAGS -#define XFS_IOC_SETXFLAGS_32 FS_IOC32_SETFLAGS -#define XFS_IOC_GETVERSION_32 FS_IOC32_GETVERSION - -/* - * On intel, even if sizes match, alignment and/or padding may differ. - */ -#if defined(CONFIG_IA64) || defined(CONFIG_X86_64) -#define BROKEN_X86_ALIGNMENT -#define __compat_packed __attribute__((packed)) -#else -#define __compat_packed -#endif - -typedef struct compat_xfs_bstime { - compat_time_t tv_sec; /* seconds */ - __s32 tv_nsec; /* and nanoseconds */ -} compat_xfs_bstime_t; - -typedef struct compat_xfs_bstat { - __u64 bs_ino; /* inode number */ - __u16 bs_mode; /* type and mode */ - __u16 bs_nlink; /* number of links */ - __u32 bs_uid; /* user id */ - __u32 bs_gid; /* group id */ - __u32 bs_rdev; /* device value */ - __s32 bs_blksize; /* block size */ - __s64 bs_size; /* file size */ - compat_xfs_bstime_t bs_atime; /* access time */ - compat_xfs_bstime_t bs_mtime; /* modify time */ - compat_xfs_bstime_t bs_ctime; /* inode change time */ - int64_t bs_blocks; /* number of blocks */ - __u32 bs_xflags; /* extended flags */ - __s32 bs_extsize; /* extent size */ - __s32 bs_extents; /* number of extents */ - __u32 bs_gen; /* generation count */ - __u16 bs_projid_lo; /* lower part of project id */ -#define bs_projid bs_projid_lo /* (previously just bs_projid) */ - __u16 bs_projid_hi; /* high part of project id */ - unsigned char bs_pad[12]; /* pad space, unused */ - __u32 bs_dmevmask; /* DMIG event mask */ - __u16 bs_dmstate; /* DMIG state info */ - __u16 bs_aextents; /* attribute number of extents */ -} __compat_packed compat_xfs_bstat_t; - -typedef struct compat_xfs_fsop_bulkreq { - compat_uptr_t lastip; /* last inode # pointer */ - __s32 icount; /* count of entries in buffer */ - compat_uptr_t ubuffer; /* user buffer for inode desc. */ - compat_uptr_t ocount; /* output count pointer */ -} compat_xfs_fsop_bulkreq_t; - -#define XFS_IOC_FSBULKSTAT_32 \ - _IOWR('X', 101, struct compat_xfs_fsop_bulkreq) -#define XFS_IOC_FSBULKSTAT_SINGLE_32 \ - _IOWR('X', 102, struct compat_xfs_fsop_bulkreq) -#define XFS_IOC_FSINUMBERS_32 \ - _IOWR('X', 103, struct compat_xfs_fsop_bulkreq) - -typedef struct compat_xfs_fsop_handlereq { - __u32 fd; /* fd for FD_TO_HANDLE */ - compat_uptr_t path; /* user pathname */ - __u32 oflags; /* open flags */ - compat_uptr_t ihandle; /* user supplied handle */ - __u32 ihandlen; /* user supplied length */ - compat_uptr_t ohandle; /* user buffer for handle */ - compat_uptr_t ohandlen; /* user buffer length */ -} compat_xfs_fsop_handlereq_t; - -#define XFS_IOC_PATH_TO_FSHANDLE_32 \ - _IOWR('X', 104, struct compat_xfs_fsop_handlereq) -#define XFS_IOC_PATH_TO_HANDLE_32 \ - _IOWR('X', 105, struct compat_xfs_fsop_handlereq) -#define XFS_IOC_FD_TO_HANDLE_32 \ - _IOWR('X', 106, struct compat_xfs_fsop_handlereq) -#define XFS_IOC_OPEN_BY_HANDLE_32 \ - _IOWR('X', 107, struct compat_xfs_fsop_handlereq) -#define XFS_IOC_READLINK_BY_HANDLE_32 \ - _IOWR('X', 108, struct compat_xfs_fsop_handlereq) - -/* The bstat field in the swapext struct needs translation */ -typedef struct compat_xfs_swapext { - __int64_t sx_version; /* version */ - __int64_t sx_fdtarget; /* fd of target file */ - __int64_t sx_fdtmp; /* fd of tmp file */ - xfs_off_t sx_offset; /* offset into file */ - xfs_off_t sx_length; /* leng from offset */ - char sx_pad[16]; /* pad space, unused */ - compat_xfs_bstat_t sx_stat; /* stat of target b4 copy */ -} __compat_packed compat_xfs_swapext_t; - -#define XFS_IOC_SWAPEXT_32 _IOWR('X', 109, struct compat_xfs_swapext) - -typedef struct compat_xfs_fsop_attrlist_handlereq { - struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */ - struct xfs_attrlist_cursor pos; /* opaque cookie, list offset */ - __u32 flags; /* which namespace to use */ - __u32 buflen; /* length of buffer supplied */ - compat_uptr_t buffer; /* returned names */ -} __compat_packed compat_xfs_fsop_attrlist_handlereq_t; - -/* Note: actually this is read/write */ -#define XFS_IOC_ATTRLIST_BY_HANDLE_32 \ - _IOW('X', 122, struct compat_xfs_fsop_attrlist_handlereq) - -/* am_opcodes defined in xfs_fs.h */ -typedef struct compat_xfs_attr_multiop { - __u32 am_opcode; - __s32 am_error; - compat_uptr_t am_attrname; - compat_uptr_t am_attrvalue; - __u32 am_length; - __u32 am_flags; -} compat_xfs_attr_multiop_t; - -typedef struct compat_xfs_fsop_attrmulti_handlereq { - struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */ - __u32 opcount;/* count of following multiop */ - /* ptr to compat_xfs_attr_multiop */ - compat_uptr_t ops; /* attr_multi data */ -} compat_xfs_fsop_attrmulti_handlereq_t; - -#define XFS_IOC_ATTRMULTI_BY_HANDLE_32 \ - _IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq) - -typedef struct compat_xfs_fsop_setdm_handlereq { - struct compat_xfs_fsop_handlereq hreq; /* handle information */ - /* ptr to struct fsdmidata */ - compat_uptr_t data; /* DMAPI data */ -} compat_xfs_fsop_setdm_handlereq_t; - -#define XFS_IOC_FSSETDM_BY_HANDLE_32 \ - _IOW('X', 121, struct compat_xfs_fsop_setdm_handlereq) - -#ifdef BROKEN_X86_ALIGNMENT -/* on ia32 l_start is on a 32-bit boundary */ -typedef struct compat_xfs_flock64 { - __s16 l_type; - __s16 l_whence; - __s64 l_start __attribute__((packed)); - /* len == 0 means until end of file */ - __s64 l_len __attribute__((packed)); - __s32 l_sysid; - __u32 l_pid; - __s32 l_pad[4]; /* reserve area */ -} compat_xfs_flock64_t; - -#define XFS_IOC_ALLOCSP_32 _IOW('X', 10, struct compat_xfs_flock64) -#define XFS_IOC_FREESP_32 _IOW('X', 11, struct compat_xfs_flock64) -#define XFS_IOC_ALLOCSP64_32 _IOW('X', 36, struct compat_xfs_flock64) -#define XFS_IOC_FREESP64_32 _IOW('X', 37, struct compat_xfs_flock64) -#define XFS_IOC_RESVSP_32 _IOW('X', 40, struct compat_xfs_flock64) -#define XFS_IOC_UNRESVSP_32 _IOW('X', 41, struct compat_xfs_flock64) -#define XFS_IOC_RESVSP64_32 _IOW('X', 42, struct compat_xfs_flock64) -#define XFS_IOC_UNRESVSP64_32 _IOW('X', 43, struct compat_xfs_flock64) -#define XFS_IOC_ZERO_RANGE_32 _IOW('X', 57, struct compat_xfs_flock64) - -typedef struct compat_xfs_fsop_geom_v1 { - __u32 blocksize; /* filesystem (data) block size */ - __u32 rtextsize; /* realtime extent size */ - __u32 agblocks; /* fsblocks in an AG */ - __u32 agcount; /* number of allocation groups */ - __u32 logblocks; /* fsblocks in the log */ - __u32 sectsize; /* (data) sector size, bytes */ - __u32 inodesize; /* inode size in bytes */ - __u32 imaxpct; /* max allowed inode space(%) */ - __u64 datablocks; /* fsblocks in data subvolume */ - __u64 rtblocks; /* fsblocks in realtime subvol */ - __u64 rtextents; /* rt extents in realtime subvol*/ - __u64 logstart; /* starting fsblock of the log */ - unsigned char uuid[16]; /* unique id of the filesystem */ - __u32 sunit; /* stripe unit, fsblocks */ - __u32 swidth; /* stripe width, fsblocks */ - __s32 version; /* structure version */ - __u32 flags; /* superblock version flags */ - __u32 logsectsize; /* log sector size, bytes */ - __u32 rtsectsize; /* realtime sector size, bytes */ - __u32 dirblocksize; /* directory block size, bytes */ -} __attribute__((packed)) compat_xfs_fsop_geom_v1_t; - -#define XFS_IOC_FSGEOMETRY_V1_32 \ - _IOR('X', 100, struct compat_xfs_fsop_geom_v1) - -typedef struct compat_xfs_inogrp { - __u64 xi_startino; /* starting inode number */ - __s32 xi_alloccount; /* # bits set in allocmask */ - __u64 xi_allocmask; /* mask of allocated inodes */ -} __attribute__((packed)) compat_xfs_inogrp_t; - -/* These growfs input structures have padding on the end, so must translate */ -typedef struct compat_xfs_growfs_data { - __u64 newblocks; /* new data subvol size, fsblocks */ - __u32 imaxpct; /* new inode space percentage limit */ -} __attribute__((packed)) compat_xfs_growfs_data_t; - -typedef struct compat_xfs_growfs_rt { - __u64 newblocks; /* new realtime size, fsblocks */ - __u32 extsize; /* new realtime extent size, fsblocks */ -} __attribute__((packed)) compat_xfs_growfs_rt_t; - -#define XFS_IOC_FSGROWFSDATA_32 _IOW('X', 110, struct compat_xfs_growfs_data) -#define XFS_IOC_FSGROWFSRT_32 _IOW('X', 112, struct compat_xfs_growfs_rt) - -#endif /* BROKEN_X86_ALIGNMENT */ - -#endif /* __XFS_IOCTL32_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c deleted file mode 100644 index b9c172b3fbbe..000000000000 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ /dev/null @@ -1,1210 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_acl.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_itable.h" -#include "xfs_rw.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_vnodeops.h" -#include "xfs_inode_item.h" -#include "xfs_trace.h" - -#include -#include -#include -#include -#include -#include -#include - -/* - * Bring the timestamps in the XFS inode uptodate. - * - * Used before writing the inode to disk. - */ -void -xfs_synchronize_times( - xfs_inode_t *ip) -{ - struct inode *inode = VFS_I(ip); - - ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec; - ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec; - ip->i_d.di_ctime.t_sec = (__int32_t)inode->i_ctime.tv_sec; - ip->i_d.di_ctime.t_nsec = (__int32_t)inode->i_ctime.tv_nsec; - ip->i_d.di_mtime.t_sec = (__int32_t)inode->i_mtime.tv_sec; - ip->i_d.di_mtime.t_nsec = (__int32_t)inode->i_mtime.tv_nsec; -} - -/* - * If the linux inode is valid, mark it dirty. - * Used when committing a dirty inode into a transaction so that - * the inode will get written back by the linux code - */ -void -xfs_mark_inode_dirty_sync( - xfs_inode_t *ip) -{ - struct inode *inode = VFS_I(ip); - - if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) - mark_inode_dirty_sync(inode); -} - -void -xfs_mark_inode_dirty( - xfs_inode_t *ip) -{ - struct inode *inode = VFS_I(ip); - - if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) - mark_inode_dirty(inode); -} - -/* - * Hook in SELinux. This is not quite correct yet, what we really need - * here (as we do for default ACLs) is a mechanism by which creation of - * these attrs can be journalled at inode creation time (along with the - * inode, of course, such that log replay can't cause these to be lost). - */ -STATIC int -xfs_init_security( - struct inode *inode, - struct inode *dir, - const struct qstr *qstr) -{ - struct xfs_inode *ip = XFS_I(inode); - size_t length; - void *value; - unsigned char *name; - int error; - - error = security_inode_init_security(inode, dir, qstr, (char **)&name, - &value, &length); - if (error) { - if (error == -EOPNOTSUPP) - return 0; - return -error; - } - - error = xfs_attr_set(ip, name, value, length, ATTR_SECURE); - - kfree(name); - kfree(value); - return error; -} - -static void -xfs_dentry_to_name( - struct xfs_name *namep, - struct dentry *dentry) -{ - namep->name = dentry->d_name.name; - namep->len = dentry->d_name.len; -} - -STATIC void -xfs_cleanup_inode( - struct inode *dir, - struct inode *inode, - struct dentry *dentry) -{ - struct xfs_name teardown; - - /* Oh, the horror. - * If we can't add the ACL or we fail in - * xfs_init_security we must back out. - * ENOSPC can hit here, among other things. - */ - xfs_dentry_to_name(&teardown, dentry); - - xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); - iput(inode); -} - -STATIC int -xfs_vn_mknod( - struct inode *dir, - struct dentry *dentry, - int mode, - dev_t rdev) -{ - struct inode *inode; - struct xfs_inode *ip = NULL; - struct posix_acl *default_acl = NULL; - struct xfs_name name; - int error; - - /* - * Irix uses Missed'em'V split, but doesn't want to see - * the upper 5 bits of (14bit) major. - */ - if (S_ISCHR(mode) || S_ISBLK(mode)) { - if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff)) - return -EINVAL; - rdev = sysv_encode_dev(rdev); - } else { - rdev = 0; - } - - if (IS_POSIXACL(dir)) { - default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT); - if (IS_ERR(default_acl)) - return PTR_ERR(default_acl); - - if (!default_acl) - mode &= ~current_umask(); - } - - xfs_dentry_to_name(&name, dentry); - error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); - if (unlikely(error)) - goto out_free_acl; - - inode = VFS_I(ip); - - error = xfs_init_security(inode, dir, &dentry->d_name); - if (unlikely(error)) - goto out_cleanup_inode; - - if (default_acl) { - error = -xfs_inherit_acl(inode, default_acl); - default_acl = NULL; - if (unlikely(error)) - goto out_cleanup_inode; - } - - - d_instantiate(dentry, inode); - return -error; - - out_cleanup_inode: - xfs_cleanup_inode(dir, inode, dentry); - out_free_acl: - posix_acl_release(default_acl); - return -error; -} - -STATIC int -xfs_vn_create( - struct inode *dir, - struct dentry *dentry, - int mode, - struct nameidata *nd) -{ - return xfs_vn_mknod(dir, dentry, mode, 0); -} - -STATIC int -xfs_vn_mkdir( - struct inode *dir, - struct dentry *dentry, - int mode) -{ - return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0); -} - -STATIC struct dentry * -xfs_vn_lookup( - struct inode *dir, - struct dentry *dentry, - struct nameidata *nd) -{ - struct xfs_inode *cip; - struct xfs_name name; - int error; - - if (dentry->d_name.len >= MAXNAMELEN) - return ERR_PTR(-ENAMETOOLONG); - - xfs_dentry_to_name(&name, dentry); - error = xfs_lookup(XFS_I(dir), &name, &cip, NULL); - if (unlikely(error)) { - if (unlikely(error != ENOENT)) - return ERR_PTR(-error); - d_add(dentry, NULL); - return NULL; - } - - return d_splice_alias(VFS_I(cip), dentry); -} - -STATIC struct dentry * -xfs_vn_ci_lookup( - struct inode *dir, - struct dentry *dentry, - struct nameidata *nd) -{ - struct xfs_inode *ip; - struct xfs_name xname; - struct xfs_name ci_name; - struct qstr dname; - int error; - - if (dentry->d_name.len >= MAXNAMELEN) - return ERR_PTR(-ENAMETOOLONG); - - xfs_dentry_to_name(&xname, dentry); - error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name); - if (unlikely(error)) { - if (unlikely(error != ENOENT)) - return ERR_PTR(-error); - /* - * call d_add(dentry, NULL) here when d_drop_negative_children - * is called in xfs_vn_mknod (ie. allow negative dentries - * with CI filesystems). - */ - return NULL; - } - - /* if exact match, just splice and exit */ - if (!ci_name.name) - return d_splice_alias(VFS_I(ip), dentry); - - /* else case-insensitive match... */ - dname.name = ci_name.name; - dname.len = ci_name.len; - dentry = d_add_ci(dentry, VFS_I(ip), &dname); - kmem_free(ci_name.name); - return dentry; -} - -STATIC int -xfs_vn_link( - struct dentry *old_dentry, - struct inode *dir, - struct dentry *dentry) -{ - struct inode *inode = old_dentry->d_inode; - struct xfs_name name; - int error; - - xfs_dentry_to_name(&name, dentry); - - error = xfs_link(XFS_I(dir), XFS_I(inode), &name); - if (unlikely(error)) - return -error; - - ihold(inode); - d_instantiate(dentry, inode); - return 0; -} - -STATIC int -xfs_vn_unlink( - struct inode *dir, - struct dentry *dentry) -{ - struct xfs_name name; - int error; - - xfs_dentry_to_name(&name, dentry); - - error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode)); - if (error) - return error; - - /* - * With unlink, the VFS makes the dentry "negative": no inode, - * but still hashed. This is incompatible with case-insensitive - * mode, so invalidate (unhash) the dentry in CI-mode. - */ - if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb)) - d_invalidate(dentry); - return 0; -} - -STATIC int -xfs_vn_symlink( - struct inode *dir, - struct dentry *dentry, - const char *symname) -{ - struct inode *inode; - struct xfs_inode *cip = NULL; - struct xfs_name name; - int error; - mode_t mode; - - mode = S_IFLNK | - (irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO); - xfs_dentry_to_name(&name, dentry); - - error = xfs_symlink(XFS_I(dir), &name, symname, mode, &cip); - if (unlikely(error)) - goto out; - - inode = VFS_I(cip); - - error = xfs_init_security(inode, dir, &dentry->d_name); - if (unlikely(error)) - goto out_cleanup_inode; - - d_instantiate(dentry, inode); - return 0; - - out_cleanup_inode: - xfs_cleanup_inode(dir, inode, dentry); - out: - return -error; -} - -STATIC int -xfs_vn_rename( - struct inode *odir, - struct dentry *odentry, - struct inode *ndir, - struct dentry *ndentry) -{ - struct inode *new_inode = ndentry->d_inode; - struct xfs_name oname; - struct xfs_name nname; - - xfs_dentry_to_name(&oname, odentry); - xfs_dentry_to_name(&nname, ndentry); - - return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), - XFS_I(ndir), &nname, new_inode ? - XFS_I(new_inode) : NULL); -} - -/* - * careful here - this function can get called recursively, so - * we need to be very careful about how much stack we use. - * uio is kmalloced for this reason... - */ -STATIC void * -xfs_vn_follow_link( - struct dentry *dentry, - struct nameidata *nd) -{ - char *link; - int error = -ENOMEM; - - link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); - if (!link) - goto out_err; - - error = -xfs_readlink(XFS_I(dentry->d_inode), link); - if (unlikely(error)) - goto out_kfree; - - nd_set_link(nd, link); - return NULL; - - out_kfree: - kfree(link); - out_err: - nd_set_link(nd, ERR_PTR(error)); - return NULL; -} - -STATIC void -xfs_vn_put_link( - struct dentry *dentry, - struct nameidata *nd, - void *p) -{ - char *s = nd_get_link(nd); - - if (!IS_ERR(s)) - kfree(s); -} - -STATIC int -xfs_vn_getattr( - struct vfsmount *mnt, - struct dentry *dentry, - struct kstat *stat) -{ - struct inode *inode = dentry->d_inode; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - - trace_xfs_getattr(ip); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - stat->size = XFS_ISIZE(ip); - stat->dev = inode->i_sb->s_dev; - stat->mode = ip->i_d.di_mode; - stat->nlink = ip->i_d.di_nlink; - stat->uid = ip->i_d.di_uid; - stat->gid = ip->i_d.di_gid; - stat->ino = ip->i_ino; - stat->atime = inode->i_atime; - stat->mtime = inode->i_mtime; - stat->ctime = inode->i_ctime; - stat->blocks = - XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks); - - - switch (inode->i_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - stat->blksize = BLKDEV_IOSIZE; - stat->rdev = MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, - sysv_minor(ip->i_df.if_u2.if_rdev)); - break; - default: - if (XFS_IS_REALTIME_INODE(ip)) { - /* - * If the file blocks are being allocated from a - * realtime volume, then return the inode's realtime - * extent size or the realtime volume's extent size. - */ - stat->blksize = - xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog; - } else - stat->blksize = xfs_preferred_iosize(mp); - stat->rdev = 0; - break; - } - - return 0; -} - -int -xfs_setattr_nonsize( - struct xfs_inode *ip, - struct iattr *iattr, - int flags) -{ - xfs_mount_t *mp = ip->i_mount; - struct inode *inode = VFS_I(ip); - int mask = iattr->ia_valid; - xfs_trans_t *tp; - int error; - uid_t uid = 0, iuid = 0; - gid_t gid = 0, igid = 0; - struct xfs_dquot *udqp = NULL, *gdqp = NULL; - struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL; - - trace_xfs_setattr(ip); - - if (mp->m_flags & XFS_MOUNT_RDONLY) - return XFS_ERROR(EROFS); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - error = -inode_change_ok(inode, iattr); - if (error) - return XFS_ERROR(error); - - ASSERT((mask & ATTR_SIZE) == 0); - - /* - * If disk quotas is on, we make sure that the dquots do exist on disk, - * before we start any other transactions. Trying to do this later - * is messy. We don't care to take a readlock to look at the ids - * in inode here, because we can't hold it across the trans_reserve. - * If the IDs do change before we take the ilock, we're covered - * because the i_*dquot fields will get updated anyway. - */ - if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) { - uint qflags = 0; - - if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) { - uid = iattr->ia_uid; - qflags |= XFS_QMOPT_UQUOTA; - } else { - uid = ip->i_d.di_uid; - } - if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) { - gid = iattr->ia_gid; - qflags |= XFS_QMOPT_GQUOTA; - } else { - gid = ip->i_d.di_gid; - } - - /* - * We take a reference when we initialize udqp and gdqp, - * so it is important that we never blindly double trip on - * the same variable. See xfs_create() for an example. - */ - ASSERT(udqp == NULL); - ASSERT(gdqp == NULL); - error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip), - qflags, &udqp, &gdqp); - if (error) - return error; - } - - tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); - error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); - if (error) - goto out_dqrele; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * Change file ownership. Must be the owner or privileged. - */ - if (mask & (ATTR_UID|ATTR_GID)) { - /* - * These IDs could have changed since we last looked at them. - * But, we're assured that if the ownership did change - * while we didn't have the inode locked, inode's dquot(s) - * would have changed also. - */ - iuid = ip->i_d.di_uid; - igid = ip->i_d.di_gid; - gid = (mask & ATTR_GID) ? iattr->ia_gid : igid; - uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid; - - /* - * Do a quota reservation only if uid/gid is actually - * going to change. - */ - if (XFS_IS_QUOTA_RUNNING(mp) && - ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) || - (XFS_IS_GQUOTA_ON(mp) && igid != gid))) { - ASSERT(tp); - error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, - capable(CAP_FOWNER) ? - XFS_QMOPT_FORCE_RES : 0); - if (error) /* out of quota */ - goto out_trans_cancel; - } - } - - xfs_trans_ijoin(tp, ip); - - /* - * Change file ownership. Must be the owner or privileged. - */ - if (mask & (ATTR_UID|ATTR_GID)) { - /* - * CAP_FSETID overrides the following restrictions: - * - * The set-user-ID and set-group-ID bits of a file will be - * cleared upon successful return from chown() - */ - if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && - !capable(CAP_FSETID)) - ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); - - /* - * Change the ownerships and register quota modifications - * in the transaction. - */ - if (iuid != uid) { - if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) { - ASSERT(mask & ATTR_UID); - ASSERT(udqp); - olddquot1 = xfs_qm_vop_chown(tp, ip, - &ip->i_udquot, udqp); - } - ip->i_d.di_uid = uid; - inode->i_uid = uid; - } - if (igid != gid) { - if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) { - ASSERT(!XFS_IS_PQUOTA_ON(mp)); - ASSERT(mask & ATTR_GID); - ASSERT(gdqp); - olddquot2 = xfs_qm_vop_chown(tp, ip, - &ip->i_gdquot, gdqp); - } - ip->i_d.di_gid = gid; - inode->i_gid = gid; - } - } - - /* - * Change file access modes. - */ - if (mask & ATTR_MODE) { - umode_t mode = iattr->ia_mode; - - if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) - mode &= ~S_ISGID; - - ip->i_d.di_mode &= S_IFMT; - ip->i_d.di_mode |= mode & ~S_IFMT; - - inode->i_mode &= S_IFMT; - inode->i_mode |= mode & ~S_IFMT; - } - - /* - * Change file access or modified times. - */ - if (mask & ATTR_ATIME) { - inode->i_atime = iattr->ia_atime; - ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec; - ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec; - ip->i_update_core = 1; - } - if (mask & ATTR_CTIME) { - inode->i_ctime = iattr->ia_ctime; - ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; - ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec; - ip->i_update_core = 1; - } - if (mask & ATTR_MTIME) { - inode->i_mtime = iattr->ia_mtime; - ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec; - ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec; - ip->i_update_core = 1; - } - - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - XFS_STATS_INC(xs_ig_attrchg); - - if (mp->m_flags & XFS_MOUNT_WSYNC) - xfs_trans_set_sync(tp); - error = xfs_trans_commit(tp, 0); - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - /* - * Release any dquot(s) the inode had kept before chown. - */ - xfs_qm_dqrele(olddquot1); - xfs_qm_dqrele(olddquot2); - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - - if (error) - return XFS_ERROR(error); - - /* - * XXX(hch): Updating the ACL entries is not atomic vs the i_mode - * update. We could avoid this with linked transactions - * and passing down the transaction pointer all the way - * to attr_set. No previous user of the generic - * Posix ACL code seems to care about this issue either. - */ - if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) { - error = -xfs_acl_chmod(inode); - if (error) - return XFS_ERROR(error); - } - - return 0; - -out_trans_cancel: - xfs_trans_cancel(tp, 0); - xfs_iunlock(ip, XFS_ILOCK_EXCL); -out_dqrele: - xfs_qm_dqrele(udqp); - xfs_qm_dqrele(gdqp); - return error; -} - -/* - * Truncate file. Must have write permission and not be a directory. - */ -int -xfs_setattr_size( - struct xfs_inode *ip, - struct iattr *iattr, - int flags) -{ - struct xfs_mount *mp = ip->i_mount; - struct inode *inode = VFS_I(ip); - int mask = iattr->ia_valid; - struct xfs_trans *tp; - int error; - uint lock_flags; - uint commit_flags = 0; - - trace_xfs_setattr(ip); - - if (mp->m_flags & XFS_MOUNT_RDONLY) - return XFS_ERROR(EROFS); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - error = -inode_change_ok(inode, iattr); - if (error) - return XFS_ERROR(error); - - ASSERT(S_ISREG(ip->i_d.di_mode)); - ASSERT((mask & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET| - ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID| - ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0); - - lock_flags = XFS_ILOCK_EXCL; - if (!(flags & XFS_ATTR_NOLOCK)) - lock_flags |= XFS_IOLOCK_EXCL; - xfs_ilock(ip, lock_flags); - - /* - * Short circuit the truncate case for zero length files. - */ - if (iattr->ia_size == 0 && - ip->i_size == 0 && ip->i_d.di_nextents == 0) { - if (!(mask & (ATTR_CTIME|ATTR_MTIME))) - goto out_unlock; - - /* - * Use the regular setattr path to update the timestamps. - */ - xfs_iunlock(ip, lock_flags); - iattr->ia_valid &= ~ATTR_SIZE; - return xfs_setattr_nonsize(ip, iattr, 0); - } - - /* - * Make sure that the dquots are attached to the inode. - */ - error = xfs_qm_dqattach_locked(ip, 0); - if (error) - goto out_unlock; - - /* - * Now we can make the changes. Before we join the inode to the - * transaction, take care of the part of the truncation that must be - * done without the inode lock. This needs to be done before joining - * the inode to the transaction, because the inode cannot be unlocked - * once it is a part of the transaction. - */ - if (iattr->ia_size > ip->i_size) { - /* - * Do the first part of growing a file: zero any data in the - * last block that is beyond the old EOF. We need to do this - * before the inode is joined to the transaction to modify - * i_size. - */ - error = xfs_zero_eof(ip, iattr->ia_size, ip->i_size); - if (error) - goto out_unlock; - } - xfs_iunlock(ip, XFS_ILOCK_EXCL); - lock_flags &= ~XFS_ILOCK_EXCL; - - /* - * We are going to log the inode size change in this transaction so - * any previous writes that are beyond the on disk EOF and the new - * EOF that have not been written out need to be written here. If we - * do not write the data out, we expose ourselves to the null files - * problem. - * - * Only flush from the on disk size to the smaller of the in memory - * file size or the new size as that's the range we really care about - * here and prevents waiting for other data not within the range we - * care about here. - */ - if (ip->i_size != ip->i_d.di_size && iattr->ia_size > ip->i_d.di_size) { - error = xfs_flush_pages(ip, ip->i_d.di_size, iattr->ia_size, - XBF_ASYNC, FI_NONE); - if (error) - goto out_unlock; - } - - /* - * Wait for all I/O to complete. - */ - xfs_ioend_wait(ip); - - error = -block_truncate_page(inode->i_mapping, iattr->ia_size, - xfs_get_blocks); - if (error) - goto out_unlock; - - tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE); - error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); - if (error) - goto out_trans_cancel; - - truncate_setsize(inode, iattr->ia_size); - - commit_flags = XFS_TRANS_RELEASE_LOG_RES; - lock_flags |= XFS_ILOCK_EXCL; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - - xfs_trans_ijoin(tp, ip); - - /* - * Only change the c/mtime if we are changing the size or we are - * explicitly asked to change it. This handles the semantic difference - * between truncate() and ftruncate() as implemented in the VFS. - * - * The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a - * special case where we need to update the times despite not having - * these flags set. For all other operations the VFS set these flags - * explicitly if it wants a timestamp update. - */ - if (iattr->ia_size != ip->i_size && - (!(mask & (ATTR_CTIME | ATTR_MTIME)))) { - iattr->ia_ctime = iattr->ia_mtime = - current_fs_time(inode->i_sb); - mask |= ATTR_CTIME | ATTR_MTIME; - } - - if (iattr->ia_size > ip->i_size) { - ip->i_d.di_size = iattr->ia_size; - ip->i_size = iattr->ia_size; - } else if (iattr->ia_size <= ip->i_size || - (iattr->ia_size == 0 && ip->i_d.di_nextents)) { - error = xfs_itruncate_data(&tp, ip, iattr->ia_size); - if (error) - goto out_trans_abort; - - /* - * Truncated "down", so we're removing references to old data - * here - if we delay flushing for a long time, we expose - * ourselves unduly to the notorious NULL files problem. So, - * we mark this inode and flush it when the file is closed, - * and do not wait the usual (long) time for writeout. - */ - xfs_iflags_set(ip, XFS_ITRUNCATED); - } - - if (mask & ATTR_CTIME) { - inode->i_ctime = iattr->ia_ctime; - ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; - ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec; - ip->i_update_core = 1; - } - if (mask & ATTR_MTIME) { - inode->i_mtime = iattr->ia_mtime; - ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec; - ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec; - ip->i_update_core = 1; - } - - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - - XFS_STATS_INC(xs_ig_attrchg); - - if (mp->m_flags & XFS_MOUNT_WSYNC) - xfs_trans_set_sync(tp); - - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); -out_unlock: - if (lock_flags) - xfs_iunlock(ip, lock_flags); - return error; - -out_trans_abort: - commit_flags |= XFS_TRANS_ABORT; -out_trans_cancel: - xfs_trans_cancel(tp, commit_flags); - goto out_unlock; -} - -STATIC int -xfs_vn_setattr( - struct dentry *dentry, - struct iattr *iattr) -{ - if (iattr->ia_valid & ATTR_SIZE) - return -xfs_setattr_size(XFS_I(dentry->d_inode), iattr, 0); - return -xfs_setattr_nonsize(XFS_I(dentry->d_inode), iattr, 0); -} - -#define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) - -/* - * Call fiemap helper to fill in user data. - * Returns positive errors to xfs_getbmap. - */ -STATIC int -xfs_fiemap_format( - void **arg, - struct getbmapx *bmv, - int *full) -{ - int error; - struct fiemap_extent_info *fieinfo = *arg; - u32 fiemap_flags = 0; - u64 logical, physical, length; - - /* Do nothing for a hole */ - if (bmv->bmv_block == -1LL) - return 0; - - logical = BBTOB(bmv->bmv_offset); - physical = BBTOB(bmv->bmv_block); - length = BBTOB(bmv->bmv_length); - - if (bmv->bmv_oflags & BMV_OF_PREALLOC) - fiemap_flags |= FIEMAP_EXTENT_UNWRITTEN; - else if (bmv->bmv_oflags & BMV_OF_DELALLOC) { - fiemap_flags |= FIEMAP_EXTENT_DELALLOC; - physical = 0; /* no block yet */ - } - if (bmv->bmv_oflags & BMV_OF_LAST) - fiemap_flags |= FIEMAP_EXTENT_LAST; - - error = fiemap_fill_next_extent(fieinfo, logical, physical, - length, fiemap_flags); - if (error > 0) { - error = 0; - *full = 1; /* user array now full */ - } - - return -error; -} - -STATIC int -xfs_vn_fiemap( - struct inode *inode, - struct fiemap_extent_info *fieinfo, - u64 start, - u64 length) -{ - xfs_inode_t *ip = XFS_I(inode); - struct getbmapx bm; - int error; - - error = fiemap_check_flags(fieinfo, XFS_FIEMAP_FLAGS); - if (error) - return error; - - /* Set up bmap header for xfs internal routine */ - bm.bmv_offset = BTOBB(start); - /* Special case for whole file */ - if (length == FIEMAP_MAX_OFFSET) - bm.bmv_length = -1LL; - else - bm.bmv_length = BTOBB(length); - - /* We add one because in getbmap world count includes the header */ - bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM : - fieinfo->fi_extents_max + 1; - bm.bmv_count = min_t(__s32, bm.bmv_count, - (PAGE_SIZE * 16 / sizeof(struct getbmapx))); - bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES; - if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) - bm.bmv_iflags |= BMV_IF_ATTRFORK; - if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC)) - bm.bmv_iflags |= BMV_IF_DELALLOC; - - error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo); - if (error) - return -error; - - return 0; -} - -static const struct inode_operations xfs_inode_operations = { - .get_acl = xfs_get_acl, - .getattr = xfs_vn_getattr, - .setattr = xfs_vn_setattr, - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .removexattr = generic_removexattr, - .listxattr = xfs_vn_listxattr, - .fiemap = xfs_vn_fiemap, -}; - -static const struct inode_operations xfs_dir_inode_operations = { - .create = xfs_vn_create, - .lookup = xfs_vn_lookup, - .link = xfs_vn_link, - .unlink = xfs_vn_unlink, - .symlink = xfs_vn_symlink, - .mkdir = xfs_vn_mkdir, - /* - * Yes, XFS uses the same method for rmdir and unlink. - * - * There are some subtile differences deeper in the code, - * but we use S_ISDIR to check for those. - */ - .rmdir = xfs_vn_unlink, - .mknod = xfs_vn_mknod, - .rename = xfs_vn_rename, - .get_acl = xfs_get_acl, - .getattr = xfs_vn_getattr, - .setattr = xfs_vn_setattr, - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .removexattr = generic_removexattr, - .listxattr = xfs_vn_listxattr, -}; - -static const struct inode_operations xfs_dir_ci_inode_operations = { - .create = xfs_vn_create, - .lookup = xfs_vn_ci_lookup, - .link = xfs_vn_link, - .unlink = xfs_vn_unlink, - .symlink = xfs_vn_symlink, - .mkdir = xfs_vn_mkdir, - /* - * Yes, XFS uses the same method for rmdir and unlink. - * - * There are some subtile differences deeper in the code, - * but we use S_ISDIR to check for those. - */ - .rmdir = xfs_vn_unlink, - .mknod = xfs_vn_mknod, - .rename = xfs_vn_rename, - .get_acl = xfs_get_acl, - .getattr = xfs_vn_getattr, - .setattr = xfs_vn_setattr, - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .removexattr = generic_removexattr, - .listxattr = xfs_vn_listxattr, -}; - -static const struct inode_operations xfs_symlink_inode_operations = { - .readlink = generic_readlink, - .follow_link = xfs_vn_follow_link, - .put_link = xfs_vn_put_link, - .get_acl = xfs_get_acl, - .getattr = xfs_vn_getattr, - .setattr = xfs_vn_setattr, - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, - .removexattr = generic_removexattr, - .listxattr = xfs_vn_listxattr, -}; - -STATIC void -xfs_diflags_to_iflags( - struct inode *inode, - struct xfs_inode *ip) -{ - if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) - inode->i_flags |= S_IMMUTABLE; - else - inode->i_flags &= ~S_IMMUTABLE; - if (ip->i_d.di_flags & XFS_DIFLAG_APPEND) - inode->i_flags |= S_APPEND; - else - inode->i_flags &= ~S_APPEND; - if (ip->i_d.di_flags & XFS_DIFLAG_SYNC) - inode->i_flags |= S_SYNC; - else - inode->i_flags &= ~S_SYNC; - if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME) - inode->i_flags |= S_NOATIME; - else - inode->i_flags &= ~S_NOATIME; -} - -/* - * Initialize the Linux inode, set up the operation vectors and - * unlock the inode. - * - * When reading existing inodes from disk this is called directly - * from xfs_iget, when creating a new inode it is called from - * xfs_ialloc after setting up the inode. - * - * We are always called with an uninitialised linux inode here. - * We need to initialise the necessary fields and take a reference - * on it. - */ -void -xfs_setup_inode( - struct xfs_inode *ip) -{ - struct inode *inode = &ip->i_vnode; - - inode->i_ino = ip->i_ino; - inode->i_state = I_NEW; - - inode_sb_list_add(inode); - /* make the inode look hashed for the writeback code */ - hlist_add_fake(&inode->i_hash); - - inode->i_mode = ip->i_d.di_mode; - inode->i_nlink = ip->i_d.di_nlink; - inode->i_uid = ip->i_d.di_uid; - inode->i_gid = ip->i_d.di_gid; - - switch (inode->i_mode & S_IFMT) { - case S_IFBLK: - case S_IFCHR: - inode->i_rdev = - MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, - sysv_minor(ip->i_df.if_u2.if_rdev)); - break; - default: - inode->i_rdev = 0; - break; - } - - inode->i_generation = ip->i_d.di_gen; - i_size_write(inode, ip->i_d.di_size); - inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec; - inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec; - inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; - inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; - inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; - inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; - xfs_diflags_to_iflags(inode, ip); - - switch (inode->i_mode & S_IFMT) { - case S_IFREG: - inode->i_op = &xfs_inode_operations; - inode->i_fop = &xfs_file_operations; - inode->i_mapping->a_ops = &xfs_address_space_operations; - break; - case S_IFDIR: - if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb)) - inode->i_op = &xfs_dir_ci_inode_operations; - else - inode->i_op = &xfs_dir_inode_operations; - inode->i_fop = &xfs_dir_file_operations; - break; - case S_IFLNK: - inode->i_op = &xfs_symlink_inode_operations; - if (!(ip->i_df.if_flags & XFS_IFINLINE)) - inode->i_mapping->a_ops = &xfs_address_space_operations; - break; - default: - inode->i_op = &xfs_inode_operations; - init_special_inode(inode, inode->i_mode, inode->i_rdev); - break; - } - - /* - * If there is no attribute fork no ACL can exist on this inode, - * and it can't have any file capabilities attached to it either. - */ - if (!XFS_IFORK_Q(ip)) { - inode_has_no_xattr(inode); - cache_no_acl(inode); - } - - xfs_iflags_clear(ip, XFS_INEW); - barrier(); - - unlock_new_inode(inode); -} diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h deleted file mode 100644 index ef41c92ce66e..000000000000 --- a/fs/xfs/linux-2.6/xfs_iops.h +++ /dev/null @@ -1,30 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_IOPS_H__ -#define __XFS_IOPS_H__ - -struct xfs_inode; - -extern const struct file_operations xfs_file_operations; -extern const struct file_operations xfs_dir_file_operations; - -extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); - -extern void xfs_setup_inode(struct xfs_inode *); - -#endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h deleted file mode 100644 index 1e8a45e74c3e..000000000000 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ /dev/null @@ -1,309 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_LINUX__ -#define __XFS_LINUX__ - -#include - -/* - * XFS_BIG_BLKNOS needs block layer disk addresses to be 64 bits. - * XFS_BIG_INUMS requires XFS_BIG_BLKNOS to be set. - */ -#if defined(CONFIG_LBDAF) || (BITS_PER_LONG == 64) -# define XFS_BIG_BLKNOS 1 -# define XFS_BIG_INUMS 1 -#else -# define XFS_BIG_BLKNOS 0 -# define XFS_BIG_INUMS 0 -#endif - -#include "xfs_types.h" - -#include "kmem.h" -#include "mrlock.h" -#include "time.h" -#include "uuid.h" - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include - -#include "xfs_vnode.h" -#include "xfs_stats.h" -#include "xfs_sysctl.h" -#include "xfs_iops.h" -#include "xfs_aops.h" -#include "xfs_super.h" -#include "xfs_buf.h" -#include "xfs_message.h" - -#ifdef __BIG_ENDIAN -#define XFS_NATIVE_HOST 1 -#else -#undef XFS_NATIVE_HOST -#endif - -/* - * Feature macros (disable/enable) - */ -#ifdef CONFIG_SMP -#define HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */ -#else -#undef HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */ -#endif - -#define irix_sgid_inherit xfs_params.sgid_inherit.val -#define irix_symlink_mode xfs_params.symlink_mode.val -#define xfs_panic_mask xfs_params.panic_mask.val -#define xfs_error_level xfs_params.error_level.val -#define xfs_syncd_centisecs xfs_params.syncd_timer.val -#define xfs_stats_clear xfs_params.stats_clear.val -#define xfs_inherit_sync xfs_params.inherit_sync.val -#define xfs_inherit_nodump xfs_params.inherit_nodump.val -#define xfs_inherit_noatime xfs_params.inherit_noatim.val -#define xfs_buf_timer_centisecs xfs_params.xfs_buf_timer.val -#define xfs_buf_age_centisecs xfs_params.xfs_buf_age.val -#define xfs_inherit_nosymlinks xfs_params.inherit_nosym.val -#define xfs_rotorstep xfs_params.rotorstep.val -#define xfs_inherit_nodefrag xfs_params.inherit_nodfrg.val -#define xfs_fstrm_centisecs xfs_params.fstrm_timer.val - -#define current_cpu() (raw_smp_processor_id()) -#define current_pid() (current->pid) -#define current_test_flags(f) (current->flags & (f)) -#define current_set_flags_nested(sp, f) \ - (*(sp) = current->flags, current->flags |= (f)) -#define current_clear_flags_nested(sp, f) \ - (*(sp) = current->flags, current->flags &= ~(f)) -#define current_restore_flags_nested(sp, f) \ - (current->flags = ((current->flags & ~(f)) | (*(sp) & (f)))) - -#define spinlock_destroy(lock) - -#define NBBY 8 /* number of bits per byte */ - -/* - * Size of block device i/o is parameterized here. - * Currently the system supports page-sized i/o. - */ -#define BLKDEV_IOSHIFT PAGE_CACHE_SHIFT -#define BLKDEV_IOSIZE (1<> 32; - __low = c; - if (__high) { - __upper = __high % (b); - __high = __high / (b); - } - asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper)); - asm("":"=A" (c):"a" (__low),"d" (__high)); - *(__u64 *)a = c; - return __mod; - } - } - - /* NOTREACHED */ - return 0; -} - -/* Side effect free 64 bit mod operation */ -static inline __u32 xfs_do_mod(void *a, __u32 b, int n) -{ - switch (n) { - case 4: - return *(__u32 *)a % b; - case 8: - { - unsigned long __upper, __low, __high, __mod; - __u64 c = *(__u64 *)a; - __upper = __high = c >> 32; - __low = c; - if (__high) { - __upper = __high % (b); - __high = __high / (b); - } - asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper)); - asm("":"=A" (c):"a" (__low),"d" (__high)); - return __mod; - } - } - - /* NOTREACHED */ - return 0; -} -#else -static inline __u32 xfs_do_div(void *a, __u32 b, int n) -{ - __u32 mod; - - switch (n) { - case 4: - mod = *(__u32 *)a % b; - *(__u32 *)a = *(__u32 *)a / b; - return mod; - case 8: - mod = do_div(*(__u64 *)a, b); - return mod; - } - - /* NOTREACHED */ - return 0; -} - -/* Side effect free 64 bit mod operation */ -static inline __u32 xfs_do_mod(void *a, __u32 b, int n) -{ - switch (n) { - case 4: - return *(__u32 *)a % b; - case 8: - { - __u64 c = *(__u64 *)a; - return do_div(c, b); - } - } - - /* NOTREACHED */ - return 0; -} -#endif - -#undef do_div -#define do_div(a, b) xfs_do_div(&(a), (b), sizeof(a)) -#define do_mod(a, b) xfs_do_mod(&(a), (b), sizeof(a)) - -static inline __uint64_t roundup_64(__uint64_t x, __uint32_t y) -{ - x += y - 1; - do_div(x, y); - return(x * y); -} - -static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y) -{ - x += y - 1; - do_div(x, y); - return x; -} - -/* ARM old ABI has some weird alignment/padding */ -#if defined(__arm__) && !defined(__ARM_EABI__) -#define __arch_pack __attribute__((packed)) -#else -#define __arch_pack -#endif - -#define ASSERT_ALWAYS(expr) \ - (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) - -#ifndef DEBUG -#define ASSERT(expr) ((void)0) - -#ifndef STATIC -# define STATIC static noinline -#endif - -#else /* DEBUG */ - -#define ASSERT(expr) \ - (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) - -#ifndef STATIC -# define STATIC noinline -#endif - -#endif /* DEBUG */ - -#endif /* __XFS_LINUX__ */ diff --git a/fs/xfs/linux-2.6/xfs_message.c b/fs/xfs/linux-2.6/xfs_message.c deleted file mode 100644 index bd672def95ac..000000000000 --- a/fs/xfs/linux-2.6/xfs_message.c +++ /dev/null @@ -1,108 +0,0 @@ -/* - * Copyright (c) 2011 Red Hat, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ - -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_mount.h" - -/* - * XFS logging functions - */ -static void -__xfs_printk( - const char *level, - const struct xfs_mount *mp, - struct va_format *vaf) -{ - if (mp && mp->m_fsname) { - printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf); - return; - } - printk("%sXFS: %pV\n", level, vaf); -} - -#define define_xfs_printk_level(func, kern_level) \ -void func(const struct xfs_mount *mp, const char *fmt, ...) \ -{ \ - struct va_format vaf; \ - va_list args; \ - \ - va_start(args, fmt); \ - \ - vaf.fmt = fmt; \ - vaf.va = &args; \ - \ - __xfs_printk(kern_level, mp, &vaf); \ - va_end(args); \ -} \ - -define_xfs_printk_level(xfs_emerg, KERN_EMERG); -define_xfs_printk_level(xfs_alert, KERN_ALERT); -define_xfs_printk_level(xfs_crit, KERN_CRIT); -define_xfs_printk_level(xfs_err, KERN_ERR); -define_xfs_printk_level(xfs_warn, KERN_WARNING); -define_xfs_printk_level(xfs_notice, KERN_NOTICE); -define_xfs_printk_level(xfs_info, KERN_INFO); -#ifdef DEBUG -define_xfs_printk_level(xfs_debug, KERN_DEBUG); -#endif - -void -xfs_alert_tag( - const struct xfs_mount *mp, - int panic_tag, - const char *fmt, ...) -{ - struct va_format vaf; - va_list args; - int do_panic = 0; - - if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) { - xfs_alert(mp, "Transforming an alert into a BUG."); - do_panic = 1; - } - - va_start(args, fmt); - - vaf.fmt = fmt; - vaf.va = &args; - - __xfs_printk(KERN_ALERT, mp, &vaf); - va_end(args); - - BUG_ON(do_panic); -} - -void -assfail(char *expr, char *file, int line) -{ - xfs_emerg(NULL, "Assertion failed: %s, file: %s, line: %d", - expr, file, line); - BUG(); -} - -void -xfs_hex_dump(void *p, int length) -{ - print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_ADDRESS, 16, 1, p, length, 1); -} diff --git a/fs/xfs/linux-2.6/xfs_message.h b/fs/xfs/linux-2.6/xfs_message.h deleted file mode 100644 index 7fb7ea007672..000000000000 --- a/fs/xfs/linux-2.6/xfs_message.h +++ /dev/null @@ -1,39 +0,0 @@ -#ifndef __XFS_MESSAGE_H -#define __XFS_MESSAGE_H 1 - -struct xfs_mount; - -extern void xfs_emerg(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_alert(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_alert_tag(const struct xfs_mount *mp, int tag, - const char *fmt, ...) - __attribute__ ((format (printf, 3, 4))); -extern void xfs_crit(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_err(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_warn(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_notice(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -extern void xfs_info(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); - -#ifdef DEBUG -extern void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) - __attribute__ ((format (printf, 2, 3))); -#else -static inline void -__attribute__ ((format (printf, 2, 3))) -xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) -{ -} -#endif - -extern void assfail(char *expr, char *f, int l); - -extern void xfs_hex_dump(void *p, int length); - -#endif /* __XFS_MESSAGE_H */ diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c deleted file mode 100644 index 7e76f537abb7..000000000000 --- a/fs/xfs/linux-2.6/xfs_quotaops.c +++ /dev/null @@ -1,139 +0,0 @@ -/* - * Copyright (c) 2008, Christoph Hellwig - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_sb.h" -#include "xfs_inum.h" -#include "xfs_log.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_quota.h" -#include "xfs_trans.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_qm.h" -#include - - -STATIC int -xfs_quota_type(int type) -{ - switch (type) { - case USRQUOTA: - return XFS_DQ_USER; - case GRPQUOTA: - return XFS_DQ_GROUP; - default: - return XFS_DQ_PROJ; - } -} - -STATIC int -xfs_fs_get_xstate( - struct super_block *sb, - struct fs_quota_stat *fqs) -{ - struct xfs_mount *mp = XFS_M(sb); - - if (!XFS_IS_QUOTA_RUNNING(mp)) - return -ENOSYS; - return -xfs_qm_scall_getqstat(mp, fqs); -} - -STATIC int -xfs_fs_set_xstate( - struct super_block *sb, - unsigned int uflags, - int op) -{ - struct xfs_mount *mp = XFS_M(sb); - unsigned int flags = 0; - - if (sb->s_flags & MS_RDONLY) - return -EROFS; - if (op != Q_XQUOTARM && !XFS_IS_QUOTA_RUNNING(mp)) - return -ENOSYS; - - if (uflags & FS_QUOTA_UDQ_ACCT) - flags |= XFS_UQUOTA_ACCT; - if (uflags & FS_QUOTA_PDQ_ACCT) - flags |= XFS_PQUOTA_ACCT; - if (uflags & FS_QUOTA_GDQ_ACCT) - flags |= XFS_GQUOTA_ACCT; - if (uflags & FS_QUOTA_UDQ_ENFD) - flags |= XFS_UQUOTA_ENFD; - if (uflags & (FS_QUOTA_PDQ_ENFD|FS_QUOTA_GDQ_ENFD)) - flags |= XFS_OQUOTA_ENFD; - - switch (op) { - case Q_XQUOTAON: - return -xfs_qm_scall_quotaon(mp, flags); - case Q_XQUOTAOFF: - if (!XFS_IS_QUOTA_ON(mp)) - return -EINVAL; - return -xfs_qm_scall_quotaoff(mp, flags); - case Q_XQUOTARM: - if (XFS_IS_QUOTA_ON(mp)) - return -EINVAL; - return -xfs_qm_scall_trunc_qfiles(mp, flags); - } - - return -EINVAL; -} - -STATIC int -xfs_fs_get_dqblk( - struct super_block *sb, - int type, - qid_t id, - struct fs_disk_quota *fdq) -{ - struct xfs_mount *mp = XFS_M(sb); - - if (!XFS_IS_QUOTA_RUNNING(mp)) - return -ENOSYS; - if (!XFS_IS_QUOTA_ON(mp)) - return -ESRCH; - - return -xfs_qm_scall_getquota(mp, id, xfs_quota_type(type), fdq); -} - -STATIC int -xfs_fs_set_dqblk( - struct super_block *sb, - int type, - qid_t id, - struct fs_disk_quota *fdq) -{ - struct xfs_mount *mp = XFS_M(sb); - - if (sb->s_flags & MS_RDONLY) - return -EROFS; - if (!XFS_IS_QUOTA_RUNNING(mp)) - return -ENOSYS; - if (!XFS_IS_QUOTA_ON(mp)) - return -ESRCH; - - return -xfs_qm_scall_setqlim(mp, id, xfs_quota_type(type), fdq); -} - -const struct quotactl_ops xfs_quotactl_operations = { - .get_xstate = xfs_fs_get_xstate, - .set_xstate = xfs_fs_set_xstate, - .get_dqblk = xfs_fs_get_dqblk, - .set_dqblk = xfs_fs_set_dqblk, -}; diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c deleted file mode 100644 index 76fdc5861932..000000000000 --- a/fs/xfs/linux-2.6/xfs_stats.c +++ /dev/null @@ -1,122 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include - -DEFINE_PER_CPU(struct xfsstats, xfsstats); - -static int xfs_stat_proc_show(struct seq_file *m, void *v) -{ - int c, i, j, val; - __uint64_t xs_xstrat_bytes = 0; - __uint64_t xs_write_bytes = 0; - __uint64_t xs_read_bytes = 0; - - static const struct xstats_entry { - char *desc; - int endpoint; - } xstats[] = { - { "extent_alloc", XFSSTAT_END_EXTENT_ALLOC }, - { "abt", XFSSTAT_END_ALLOC_BTREE }, - { "blk_map", XFSSTAT_END_BLOCK_MAPPING }, - { "bmbt", XFSSTAT_END_BLOCK_MAP_BTREE }, - { "dir", XFSSTAT_END_DIRECTORY_OPS }, - { "trans", XFSSTAT_END_TRANSACTIONS }, - { "ig", XFSSTAT_END_INODE_OPS }, - { "log", XFSSTAT_END_LOG_OPS }, - { "push_ail", XFSSTAT_END_TAIL_PUSHING }, - { "xstrat", XFSSTAT_END_WRITE_CONVERT }, - { "rw", XFSSTAT_END_READ_WRITE_OPS }, - { "attr", XFSSTAT_END_ATTRIBUTE_OPS }, - { "icluster", XFSSTAT_END_INODE_CLUSTER }, - { "vnodes", XFSSTAT_END_VNODE_OPS }, - { "buf", XFSSTAT_END_BUF }, - { "abtb2", XFSSTAT_END_ABTB_V2 }, - { "abtc2", XFSSTAT_END_ABTC_V2 }, - { "bmbt2", XFSSTAT_END_BMBT_V2 }, - { "ibt2", XFSSTAT_END_IBT_V2 }, - }; - - /* Loop over all stats groups */ - for (i=j = 0; i < ARRAY_SIZE(xstats); i++) { - seq_printf(m, "%s", xstats[i].desc); - /* inner loop does each group */ - while (j < xstats[i].endpoint) { - val = 0; - /* sum over all cpus */ - for_each_possible_cpu(c) - val += *(((__u32*)&per_cpu(xfsstats, c) + j)); - seq_printf(m, " %u", val); - j++; - } - seq_putc(m, '\n'); - } - /* extra precision counters */ - for_each_possible_cpu(i) { - xs_xstrat_bytes += per_cpu(xfsstats, i).xs_xstrat_bytes; - xs_write_bytes += per_cpu(xfsstats, i).xs_write_bytes; - xs_read_bytes += per_cpu(xfsstats, i).xs_read_bytes; - } - - seq_printf(m, "xpc %Lu %Lu %Lu\n", - xs_xstrat_bytes, xs_write_bytes, xs_read_bytes); - seq_printf(m, "debug %u\n", -#if defined(DEBUG) - 1); -#else - 0); -#endif - return 0; -} - -static int xfs_stat_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, xfs_stat_proc_show, NULL); -} - -static const struct file_operations xfs_stat_proc_fops = { - .owner = THIS_MODULE, - .open = xfs_stat_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -int -xfs_init_procfs(void) -{ - if (!proc_mkdir("fs/xfs", NULL)) - goto out; - - if (!proc_create("fs/xfs/stat", 0, NULL, - &xfs_stat_proc_fops)) - goto out_remove_entry; - return 0; - - out_remove_entry: - remove_proc_entry("fs/xfs", NULL); - out: - return -ENOMEM; -} - -void -xfs_cleanup_procfs(void) -{ - remove_proc_entry("fs/xfs/stat", NULL); - remove_proc_entry("fs/xfs", NULL); -} diff --git a/fs/xfs/linux-2.6/xfs_stats.h b/fs/xfs/linux-2.6/xfs_stats.h deleted file mode 100644 index 736854b1ca1a..000000000000 --- a/fs/xfs/linux-2.6/xfs_stats.h +++ /dev/null @@ -1,223 +0,0 @@ -/* - * Copyright (c) 2000,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_STATS_H__ -#define __XFS_STATS_H__ - - -#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF) - -#include - -/* - * XFS global statistics - */ -struct xfsstats { -# define XFSSTAT_END_EXTENT_ALLOC 4 - __uint32_t xs_allocx; - __uint32_t xs_allocb; - __uint32_t xs_freex; - __uint32_t xs_freeb; -# define XFSSTAT_END_ALLOC_BTREE (XFSSTAT_END_EXTENT_ALLOC+4) - __uint32_t xs_abt_lookup; - __uint32_t xs_abt_compare; - __uint32_t xs_abt_insrec; - __uint32_t xs_abt_delrec; -# define XFSSTAT_END_BLOCK_MAPPING (XFSSTAT_END_ALLOC_BTREE+7) - __uint32_t xs_blk_mapr; - __uint32_t xs_blk_mapw; - __uint32_t xs_blk_unmap; - __uint32_t xs_add_exlist; - __uint32_t xs_del_exlist; - __uint32_t xs_look_exlist; - __uint32_t xs_cmp_exlist; -# define XFSSTAT_END_BLOCK_MAP_BTREE (XFSSTAT_END_BLOCK_MAPPING+4) - __uint32_t xs_bmbt_lookup; - __uint32_t xs_bmbt_compare; - __uint32_t xs_bmbt_insrec; - __uint32_t xs_bmbt_delrec; -# define XFSSTAT_END_DIRECTORY_OPS (XFSSTAT_END_BLOCK_MAP_BTREE+4) - __uint32_t xs_dir_lookup; - __uint32_t xs_dir_create; - __uint32_t xs_dir_remove; - __uint32_t xs_dir_getdents; -# define XFSSTAT_END_TRANSACTIONS (XFSSTAT_END_DIRECTORY_OPS+3) - __uint32_t xs_trans_sync; - __uint32_t xs_trans_async; - __uint32_t xs_trans_empty; -# define XFSSTAT_END_INODE_OPS (XFSSTAT_END_TRANSACTIONS+7) - __uint32_t xs_ig_attempts; - __uint32_t xs_ig_found; - __uint32_t xs_ig_frecycle; - __uint32_t xs_ig_missed; - __uint32_t xs_ig_dup; - __uint32_t xs_ig_reclaims; - __uint32_t xs_ig_attrchg; -# define XFSSTAT_END_LOG_OPS (XFSSTAT_END_INODE_OPS+5) - __uint32_t xs_log_writes; - __uint32_t xs_log_blocks; - __uint32_t xs_log_noiclogs; - __uint32_t xs_log_force; - __uint32_t xs_log_force_sleep; -# define XFSSTAT_END_TAIL_PUSHING (XFSSTAT_END_LOG_OPS+10) - __uint32_t xs_try_logspace; - __uint32_t xs_sleep_logspace; - __uint32_t xs_push_ail; - __uint32_t xs_push_ail_success; - __uint32_t xs_push_ail_pushbuf; - __uint32_t xs_push_ail_pinned; - __uint32_t xs_push_ail_locked; - __uint32_t xs_push_ail_flushing; - __uint32_t xs_push_ail_restarts; - __uint32_t xs_push_ail_flush; -# define XFSSTAT_END_WRITE_CONVERT (XFSSTAT_END_TAIL_PUSHING+2) - __uint32_t xs_xstrat_quick; - __uint32_t xs_xstrat_split; -# define XFSSTAT_END_READ_WRITE_OPS (XFSSTAT_END_WRITE_CONVERT+2) - __uint32_t xs_write_calls; - __uint32_t xs_read_calls; -# define XFSSTAT_END_ATTRIBUTE_OPS (XFSSTAT_END_READ_WRITE_OPS+4) - __uint32_t xs_attr_get; - __uint32_t xs_attr_set; - __uint32_t xs_attr_remove; - __uint32_t xs_attr_list; -# define XFSSTAT_END_INODE_CLUSTER (XFSSTAT_END_ATTRIBUTE_OPS+3) - __uint32_t xs_iflush_count; - __uint32_t xs_icluster_flushcnt; - __uint32_t xs_icluster_flushinode; -# define XFSSTAT_END_VNODE_OPS (XFSSTAT_END_INODE_CLUSTER+8) - __uint32_t vn_active; /* # vnodes not on free lists */ - __uint32_t vn_alloc; /* # times vn_alloc called */ - __uint32_t vn_get; /* # times vn_get called */ - __uint32_t vn_hold; /* # times vn_hold called */ - __uint32_t vn_rele; /* # times vn_rele called */ - __uint32_t vn_reclaim; /* # times vn_reclaim called */ - __uint32_t vn_remove; /* # times vn_remove called */ - __uint32_t vn_free; /* # times vn_free called */ -#define XFSSTAT_END_BUF (XFSSTAT_END_VNODE_OPS+9) - __uint32_t xb_get; - __uint32_t xb_create; - __uint32_t xb_get_locked; - __uint32_t xb_get_locked_waited; - __uint32_t xb_busy_locked; - __uint32_t xb_miss_locked; - __uint32_t xb_page_retries; - __uint32_t xb_page_found; - __uint32_t xb_get_read; -/* Version 2 btree counters */ -#define XFSSTAT_END_ABTB_V2 (XFSSTAT_END_BUF+15) - __uint32_t xs_abtb_2_lookup; - __uint32_t xs_abtb_2_compare; - __uint32_t xs_abtb_2_insrec; - __uint32_t xs_abtb_2_delrec; - __uint32_t xs_abtb_2_newroot; - __uint32_t xs_abtb_2_killroot; - __uint32_t xs_abtb_2_increment; - __uint32_t xs_abtb_2_decrement; - __uint32_t xs_abtb_2_lshift; - __uint32_t xs_abtb_2_rshift; - __uint32_t xs_abtb_2_split; - __uint32_t xs_abtb_2_join; - __uint32_t xs_abtb_2_alloc; - __uint32_t xs_abtb_2_free; - __uint32_t xs_abtb_2_moves; -#define XFSSTAT_END_ABTC_V2 (XFSSTAT_END_ABTB_V2+15) - __uint32_t xs_abtc_2_lookup; - __uint32_t xs_abtc_2_compare; - __uint32_t xs_abtc_2_insrec; - __uint32_t xs_abtc_2_delrec; - __uint32_t xs_abtc_2_newroot; - __uint32_t xs_abtc_2_killroot; - __uint32_t xs_abtc_2_increment; - __uint32_t xs_abtc_2_decrement; - __uint32_t xs_abtc_2_lshift; - __uint32_t xs_abtc_2_rshift; - __uint32_t xs_abtc_2_split; - __uint32_t xs_abtc_2_join; - __uint32_t xs_abtc_2_alloc; - __uint32_t xs_abtc_2_free; - __uint32_t xs_abtc_2_moves; -#define XFSSTAT_END_BMBT_V2 (XFSSTAT_END_ABTC_V2+15) - __uint32_t xs_bmbt_2_lookup; - __uint32_t xs_bmbt_2_compare; - __uint32_t xs_bmbt_2_insrec; - __uint32_t xs_bmbt_2_delrec; - __uint32_t xs_bmbt_2_newroot; - __uint32_t xs_bmbt_2_killroot; - __uint32_t xs_bmbt_2_increment; - __uint32_t xs_bmbt_2_decrement; - __uint32_t xs_bmbt_2_lshift; - __uint32_t xs_bmbt_2_rshift; - __uint32_t xs_bmbt_2_split; - __uint32_t xs_bmbt_2_join; - __uint32_t xs_bmbt_2_alloc; - __uint32_t xs_bmbt_2_free; - __uint32_t xs_bmbt_2_moves; -#define XFSSTAT_END_IBT_V2 (XFSSTAT_END_BMBT_V2+15) - __uint32_t xs_ibt_2_lookup; - __uint32_t xs_ibt_2_compare; - __uint32_t xs_ibt_2_insrec; - __uint32_t xs_ibt_2_delrec; - __uint32_t xs_ibt_2_newroot; - __uint32_t xs_ibt_2_killroot; - __uint32_t xs_ibt_2_increment; - __uint32_t xs_ibt_2_decrement; - __uint32_t xs_ibt_2_lshift; - __uint32_t xs_ibt_2_rshift; - __uint32_t xs_ibt_2_split; - __uint32_t xs_ibt_2_join; - __uint32_t xs_ibt_2_alloc; - __uint32_t xs_ibt_2_free; - __uint32_t xs_ibt_2_moves; -/* Extra precision counters */ - __uint64_t xs_xstrat_bytes; - __uint64_t xs_write_bytes; - __uint64_t xs_read_bytes; -}; - -DECLARE_PER_CPU(struct xfsstats, xfsstats); - -/* - * We don't disable preempt, not too worried about poking the - * wrong CPU's stat for now (also aggregated before reporting). - */ -#define XFS_STATS_INC(v) (per_cpu(xfsstats, current_cpu()).v++) -#define XFS_STATS_DEC(v) (per_cpu(xfsstats, current_cpu()).v--) -#define XFS_STATS_ADD(v, inc) (per_cpu(xfsstats, current_cpu()).v += (inc)) - -extern int xfs_init_procfs(void); -extern void xfs_cleanup_procfs(void); - - -#else /* !CONFIG_PROC_FS */ - -# define XFS_STATS_INC(count) -# define XFS_STATS_DEC(count) -# define XFS_STATS_ADD(count, inc) - -static inline int xfs_init_procfs(void) -{ - return 0; -} - -static inline void xfs_cleanup_procfs(void) -{ -} - -#endif /* !CONFIG_PROC_FS */ - -#endif /* __XFS_STATS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c deleted file mode 100644 index 9a72dda58bd0..000000000000 --- a/fs/xfs/linux-2.6/xfs_super.c +++ /dev/null @@ -1,1773 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ - -#include "xfs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir2.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_itable.h" -#include "xfs_fsops.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_vnodeops.h" -#include "xfs_log_priv.h" -#include "xfs_trans_priv.h" -#include "xfs_filestream.h" -#include "xfs_da_btree.h" -#include "xfs_extfree_item.h" -#include "xfs_mru_cache.h" -#include "xfs_inode_item.h" -#include "xfs_sync.h" -#include "xfs_trace.h" - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -static const struct super_operations xfs_super_operations; -static kmem_zone_t *xfs_ioend_zone; -mempool_t *xfs_ioend_pool; - -#define MNTOPT_LOGBUFS "logbufs" /* number of XFS log buffers */ -#define MNTOPT_LOGBSIZE "logbsize" /* size of XFS log buffers */ -#define MNTOPT_LOGDEV "logdev" /* log device */ -#define MNTOPT_RTDEV "rtdev" /* realtime I/O device */ -#define MNTOPT_BIOSIZE "biosize" /* log2 of preferred buffered io size */ -#define MNTOPT_WSYNC "wsync" /* safe-mode nfs compatible mount */ -#define MNTOPT_NOALIGN "noalign" /* turn off stripe alignment */ -#define MNTOPT_SWALLOC "swalloc" /* turn on stripe width allocation */ -#define MNTOPT_SUNIT "sunit" /* data volume stripe unit */ -#define MNTOPT_SWIDTH "swidth" /* data volume stripe width */ -#define MNTOPT_NOUUID "nouuid" /* ignore filesystem UUID */ -#define MNTOPT_MTPT "mtpt" /* filesystem mount point */ -#define MNTOPT_GRPID "grpid" /* group-ID from parent directory */ -#define MNTOPT_NOGRPID "nogrpid" /* group-ID from current process */ -#define MNTOPT_BSDGROUPS "bsdgroups" /* group-ID from parent directory */ -#define MNTOPT_SYSVGROUPS "sysvgroups" /* group-ID from current process */ -#define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ -#define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ -#define MNTOPT_BARRIER "barrier" /* use writer barriers for log write and - * unwritten extent conversion */ -#define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ -#define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ -#define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ -#define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ -#define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ -#define MNTOPT_NOLARGEIO "nolargeio" /* do not report large I/O sizes - * in stat(). */ -#define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */ -#define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */ -#define MNTOPT_FILESTREAM "filestreams" /* use filestreams allocator */ -#define MNTOPT_QUOTA "quota" /* disk quotas (user) */ -#define MNTOPT_NOQUOTA "noquota" /* no quotas */ -#define MNTOPT_USRQUOTA "usrquota" /* user quota enabled */ -#define MNTOPT_GRPQUOTA "grpquota" /* group quota enabled */ -#define MNTOPT_PRJQUOTA "prjquota" /* project quota enabled */ -#define MNTOPT_UQUOTA "uquota" /* user quota (IRIX variant) */ -#define MNTOPT_GQUOTA "gquota" /* group quota (IRIX variant) */ -#define MNTOPT_PQUOTA "pquota" /* project quota (IRIX variant) */ -#define MNTOPT_UQUOTANOENF "uqnoenforce"/* user quota limit enforcement */ -#define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */ -#define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */ -#define MNTOPT_QUOTANOENF "qnoenforce" /* same as uqnoenforce */ -#define MNTOPT_DELAYLOG "delaylog" /* Delayed logging enabled */ -#define MNTOPT_NODELAYLOG "nodelaylog" /* Delayed logging disabled */ -#define MNTOPT_DISCARD "discard" /* Discard unused blocks */ -#define MNTOPT_NODISCARD "nodiscard" /* Do not discard unused blocks */ - -/* - * Table driven mount option parser. - * - * Currently only used for remount, but it will be used for mount - * in the future, too. - */ -enum { - Opt_barrier, Opt_nobarrier, Opt_err -}; - -static const match_table_t tokens = { - {Opt_barrier, "barrier"}, - {Opt_nobarrier, "nobarrier"}, - {Opt_err, NULL} -}; - - -STATIC unsigned long -suffix_strtoul(char *s, char **endp, unsigned int base) -{ - int last, shift_left_factor = 0; - char *value = s; - - last = strlen(value) - 1; - if (value[last] == 'K' || value[last] == 'k') { - shift_left_factor = 10; - value[last] = '\0'; - } - if (value[last] == 'M' || value[last] == 'm') { - shift_left_factor = 20; - value[last] = '\0'; - } - if (value[last] == 'G' || value[last] == 'g') { - shift_left_factor = 30; - value[last] = '\0'; - } - - return simple_strtoul((const char *)s, endp, base) << shift_left_factor; -} - -/* - * This function fills in xfs_mount_t fields based on mount args. - * Note: the superblock has _not_ yet been read in. - * - * Note that this function leaks the various device name allocations on - * failure. The caller takes care of them. - */ -STATIC int -xfs_parseargs( - struct xfs_mount *mp, - char *options) -{ - struct super_block *sb = mp->m_super; - char *this_char, *value, *eov; - int dsunit = 0; - int dswidth = 0; - int iosize = 0; - __uint8_t iosizelog = 0; - - /* - * set up the mount name first so all the errors will refer to the - * correct device. - */ - mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL); - if (!mp->m_fsname) - return ENOMEM; - mp->m_fsname_len = strlen(mp->m_fsname) + 1; - - /* - * Copy binary VFS mount flags we are interested in. - */ - if (sb->s_flags & MS_RDONLY) - mp->m_flags |= XFS_MOUNT_RDONLY; - if (sb->s_flags & MS_DIRSYNC) - mp->m_flags |= XFS_MOUNT_DIRSYNC; - if (sb->s_flags & MS_SYNCHRONOUS) - mp->m_flags |= XFS_MOUNT_WSYNC; - - /* - * Set some default flags that could be cleared by the mount option - * parsing. - */ - mp->m_flags |= XFS_MOUNT_BARRIER; - mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; - mp->m_flags |= XFS_MOUNT_SMALL_INUMS; - mp->m_flags |= XFS_MOUNT_DELAYLOG; - - /* - * These can be overridden by the mount option parsing. - */ - mp->m_logbufs = -1; - mp->m_logbsize = -1; - - if (!options) - goto done; - - while ((this_char = strsep(&options, ",")) != NULL) { - if (!*this_char) - continue; - if ((value = strchr(this_char, '=')) != NULL) - *value++ = 0; - - if (!strcmp(this_char, MNTOPT_LOGBUFS)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - mp->m_logbufs = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - mp->m_logbsize = suffix_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_LOGDEV)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL); - if (!mp->m_logname) - return ENOMEM; - } else if (!strcmp(this_char, MNTOPT_MTPT)) { - xfs_warn(mp, "%s option not allowed on this system", - this_char); - return EINVAL; - } else if (!strcmp(this_char, MNTOPT_RTDEV)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL); - if (!mp->m_rtname) - return ENOMEM; - } else if (!strcmp(this_char, MNTOPT_BIOSIZE)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - iosize = simple_strtoul(value, &eov, 10); - iosizelog = ffs(iosize) - 1; - } else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - iosize = suffix_strtoul(value, &eov, 10); - iosizelog = ffs(iosize) - 1; - } else if (!strcmp(this_char, MNTOPT_GRPID) || - !strcmp(this_char, MNTOPT_BSDGROUPS)) { - mp->m_flags |= XFS_MOUNT_GRPID; - } else if (!strcmp(this_char, MNTOPT_NOGRPID) || - !strcmp(this_char, MNTOPT_SYSVGROUPS)) { - mp->m_flags &= ~XFS_MOUNT_GRPID; - } else if (!strcmp(this_char, MNTOPT_WSYNC)) { - mp->m_flags |= XFS_MOUNT_WSYNC; - } else if (!strcmp(this_char, MNTOPT_NORECOVERY)) { - mp->m_flags |= XFS_MOUNT_NORECOVERY; - } else if (!strcmp(this_char, MNTOPT_NOALIGN)) { - mp->m_flags |= XFS_MOUNT_NOALIGN; - } else if (!strcmp(this_char, MNTOPT_SWALLOC)) { - mp->m_flags |= XFS_MOUNT_SWALLOC; - } else if (!strcmp(this_char, MNTOPT_SUNIT)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - dsunit = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_SWIDTH)) { - if (!value || !*value) { - xfs_warn(mp, "%s option requires an argument", - this_char); - return EINVAL; - } - dswidth = simple_strtoul(value, &eov, 10); - } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { - mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS; -#if !XFS_BIG_INUMS - xfs_warn(mp, "%s option not allowed on this system", - this_char); - return EINVAL; -#endif - } else if (!strcmp(this_char, MNTOPT_NOUUID)) { - mp->m_flags |= XFS_MOUNT_NOUUID; - } else if (!strcmp(this_char, MNTOPT_BARRIER)) { - mp->m_flags |= XFS_MOUNT_BARRIER; - } else if (!strcmp(this_char, MNTOPT_NOBARRIER)) { - mp->m_flags &= ~XFS_MOUNT_BARRIER; - } else if (!strcmp(this_char, MNTOPT_IKEEP)) { - mp->m_flags |= XFS_MOUNT_IKEEP; - } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) { - mp->m_flags &= ~XFS_MOUNT_IKEEP; - } else if (!strcmp(this_char, MNTOPT_LARGEIO)) { - mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE; - } else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) { - mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; - } else if (!strcmp(this_char, MNTOPT_ATTR2)) { - mp->m_flags |= XFS_MOUNT_ATTR2; - } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { - mp->m_flags &= ~XFS_MOUNT_ATTR2; - mp->m_flags |= XFS_MOUNT_NOATTR2; - } else if (!strcmp(this_char, MNTOPT_FILESTREAM)) { - mp->m_flags |= XFS_MOUNT_FILESTREAMS; - } else if (!strcmp(this_char, MNTOPT_NOQUOTA)) { - mp->m_qflags &= ~(XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE | - XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE | - XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE | - XFS_UQUOTA_ENFD | XFS_OQUOTA_ENFD); - } else if (!strcmp(this_char, MNTOPT_QUOTA) || - !strcmp(this_char, MNTOPT_UQUOTA) || - !strcmp(this_char, MNTOPT_USRQUOTA)) { - mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE | - XFS_UQUOTA_ENFD); - } else if (!strcmp(this_char, MNTOPT_QUOTANOENF) || - !strcmp(this_char, MNTOPT_UQUOTANOENF)) { - mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE); - mp->m_qflags &= ~XFS_UQUOTA_ENFD; - } else if (!strcmp(this_char, MNTOPT_PQUOTA) || - !strcmp(this_char, MNTOPT_PRJQUOTA)) { - mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE | - XFS_OQUOTA_ENFD); - } else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) { - mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE); - mp->m_qflags &= ~XFS_OQUOTA_ENFD; - } else if (!strcmp(this_char, MNTOPT_GQUOTA) || - !strcmp(this_char, MNTOPT_GRPQUOTA)) { - mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE | - XFS_OQUOTA_ENFD); - } else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) { - mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE); - mp->m_qflags &= ~XFS_OQUOTA_ENFD; - } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) { - mp->m_flags |= XFS_MOUNT_DELAYLOG; - } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { - mp->m_flags &= ~XFS_MOUNT_DELAYLOG; - } else if (!strcmp(this_char, MNTOPT_DISCARD)) { - mp->m_flags |= XFS_MOUNT_DISCARD; - } else if (!strcmp(this_char, MNTOPT_NODISCARD)) { - mp->m_flags &= ~XFS_MOUNT_DISCARD; - } else if (!strcmp(this_char, "ihashsize")) { - xfs_warn(mp, - "ihashsize no longer used, option is deprecated."); - } else if (!strcmp(this_char, "osyncisdsync")) { - xfs_warn(mp, - "osyncisdsync has no effect, option is deprecated."); - } else if (!strcmp(this_char, "osyncisosync")) { - xfs_warn(mp, - "osyncisosync has no effect, option is deprecated."); - } else if (!strcmp(this_char, "irixsgid")) { - xfs_warn(mp, - "irixsgid is now a sysctl(2) variable, option is deprecated."); - } else { - xfs_warn(mp, "unknown mount option [%s].", this_char); - return EINVAL; - } - } - - /* - * no recovery flag requires a read-only mount - */ - if ((mp->m_flags & XFS_MOUNT_NORECOVERY) && - !(mp->m_flags & XFS_MOUNT_RDONLY)) { - xfs_warn(mp, "no-recovery mounts must be read-only."); - return EINVAL; - } - - if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) { - xfs_warn(mp, - "sunit and swidth options incompatible with the noalign option"); - return EINVAL; - } - - if ((mp->m_flags & XFS_MOUNT_DISCARD) && - !(mp->m_flags & XFS_MOUNT_DELAYLOG)) { - xfs_warn(mp, - "the discard option is incompatible with the nodelaylog option"); - return EINVAL; - } - -#ifndef CONFIG_XFS_QUOTA - if (XFS_IS_QUOTA_RUNNING(mp)) { - xfs_warn(mp, "quota support not available in this kernel."); - return EINVAL; - } -#endif - - if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) && - (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE))) { - xfs_warn(mp, "cannot mount with both project and group quota"); - return EINVAL; - } - - if ((dsunit && !dswidth) || (!dsunit && dswidth)) { - xfs_warn(mp, "sunit and swidth must be specified together"); - return EINVAL; - } - - if (dsunit && (dswidth % dsunit != 0)) { - xfs_warn(mp, - "stripe width (%d) must be a multiple of the stripe unit (%d)", - dswidth, dsunit); - return EINVAL; - } - -done: - if (!(mp->m_flags & XFS_MOUNT_NOALIGN)) { - /* - * At this point the superblock has not been read - * in, therefore we do not know the block size. - * Before the mount call ends we will convert - * these to FSBs. - */ - if (dsunit) { - mp->m_dalign = dsunit; - mp->m_flags |= XFS_MOUNT_RETERR; - } - - if (dswidth) - mp->m_swidth = dswidth; - } - - if (mp->m_logbufs != -1 && - mp->m_logbufs != 0 && - (mp->m_logbufs < XLOG_MIN_ICLOGS || - mp->m_logbufs > XLOG_MAX_ICLOGS)) { - xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]", - mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS); - return XFS_ERROR(EINVAL); - } - if (mp->m_logbsize != -1 && - mp->m_logbsize != 0 && - (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE || - mp->m_logbsize > XLOG_MAX_RECORD_BSIZE || - !is_power_of_2(mp->m_logbsize))) { - xfs_warn(mp, - "invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]", - mp->m_logbsize); - return XFS_ERROR(EINVAL); - } - - if (iosizelog) { - if (iosizelog > XFS_MAX_IO_LOG || - iosizelog < XFS_MIN_IO_LOG) { - xfs_warn(mp, "invalid log iosize: %d [not %d-%d]", - iosizelog, XFS_MIN_IO_LOG, - XFS_MAX_IO_LOG); - return XFS_ERROR(EINVAL); - } - - mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE; - mp->m_readio_log = iosizelog; - mp->m_writeio_log = iosizelog; - } - - return 0; -} - -struct proc_xfs_info { - int flag; - char *str; -}; - -STATIC int -xfs_showargs( - struct xfs_mount *mp, - struct seq_file *m) -{ - static struct proc_xfs_info xfs_info_set[] = { - /* the few simple ones we can get from the mount struct */ - { XFS_MOUNT_IKEEP, "," MNTOPT_IKEEP }, - { XFS_MOUNT_WSYNC, "," MNTOPT_WSYNC }, - { XFS_MOUNT_NOALIGN, "," MNTOPT_NOALIGN }, - { XFS_MOUNT_SWALLOC, "," MNTOPT_SWALLOC }, - { XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID }, - { XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY }, - { XFS_MOUNT_ATTR2, "," MNTOPT_ATTR2 }, - { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, - { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, - { XFS_MOUNT_DELAYLOG, "," MNTOPT_DELAYLOG }, - { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, - { 0, NULL } - }; - static struct proc_xfs_info xfs_info_unset[] = { - /* the few simple ones we can get from the mount struct */ - { XFS_MOUNT_COMPAT_IOSIZE, "," MNTOPT_LARGEIO }, - { XFS_MOUNT_BARRIER, "," MNTOPT_NOBARRIER }, - { XFS_MOUNT_SMALL_INUMS, "," MNTOPT_64BITINODE }, - { 0, NULL } - }; - struct proc_xfs_info *xfs_infop; - - for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) { - if (mp->m_flags & xfs_infop->flag) - seq_puts(m, xfs_infop->str); - } - for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) { - if (!(mp->m_flags & xfs_infop->flag)) - seq_puts(m, xfs_infop->str); - } - - if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) - seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk", - (int)(1 << mp->m_writeio_log) >> 10); - - if (mp->m_logbufs > 0) - seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); - if (mp->m_logbsize > 0) - seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10); - - if (mp->m_logname) - seq_printf(m, "," MNTOPT_LOGDEV "=%s", mp->m_logname); - if (mp->m_rtname) - seq_printf(m, "," MNTOPT_RTDEV "=%s", mp->m_rtname); - - if (mp->m_dalign > 0) - seq_printf(m, "," MNTOPT_SUNIT "=%d", - (int)XFS_FSB_TO_BB(mp, mp->m_dalign)); - if (mp->m_swidth > 0) - seq_printf(m, "," MNTOPT_SWIDTH "=%d", - (int)XFS_FSB_TO_BB(mp, mp->m_swidth)); - - if (mp->m_qflags & (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD)) - seq_puts(m, "," MNTOPT_USRQUOTA); - else if (mp->m_qflags & XFS_UQUOTA_ACCT) - seq_puts(m, "," MNTOPT_UQUOTANOENF); - - /* Either project or group quotas can be active, not both */ - - if (mp->m_qflags & XFS_PQUOTA_ACCT) { - if (mp->m_qflags & XFS_OQUOTA_ENFD) - seq_puts(m, "," MNTOPT_PRJQUOTA); - else - seq_puts(m, "," MNTOPT_PQUOTANOENF); - } else if (mp->m_qflags & XFS_GQUOTA_ACCT) { - if (mp->m_qflags & XFS_OQUOTA_ENFD) - seq_puts(m, "," MNTOPT_GRPQUOTA); - else - seq_puts(m, "," MNTOPT_GQUOTANOENF); - } - - if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT)) - seq_puts(m, "," MNTOPT_NOQUOTA); - - return 0; -} -__uint64_t -xfs_max_file_offset( - unsigned int blockshift) -{ - unsigned int pagefactor = 1; - unsigned int bitshift = BITS_PER_LONG - 1; - - /* Figure out maximum filesize, on Linux this can depend on - * the filesystem blocksize (on 32 bit platforms). - * __block_write_begin does this in an [unsigned] long... - * page->index << (PAGE_CACHE_SHIFT - bbits) - * So, for page sized blocks (4K on 32 bit platforms), - * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is - * (((u64)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1) - * but for smaller blocksizes it is less (bbits = log2 bsize). - * Note1: get_block_t takes a long (implicit cast from above) - * Note2: The Large Block Device (LBD and HAVE_SECTOR_T) patch - * can optionally convert the [unsigned] long from above into - * an [unsigned] long long. - */ - -#if BITS_PER_LONG == 32 -# if defined(CONFIG_LBDAF) - ASSERT(sizeof(sector_t) == 8); - pagefactor = PAGE_CACHE_SIZE; - bitshift = BITS_PER_LONG; -# else - pagefactor = PAGE_CACHE_SIZE >> (PAGE_CACHE_SHIFT - blockshift); -# endif -#endif - - return (((__uint64_t)pagefactor) << bitshift) - 1; -} - -STATIC int -xfs_blkdev_get( - xfs_mount_t *mp, - const char *name, - struct block_device **bdevp) -{ - int error = 0; - - *bdevp = blkdev_get_by_path(name, FMODE_READ|FMODE_WRITE|FMODE_EXCL, - mp); - if (IS_ERR(*bdevp)) { - error = PTR_ERR(*bdevp); - xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error); - } - - return -error; -} - -STATIC void -xfs_blkdev_put( - struct block_device *bdev) -{ - if (bdev) - blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); -} - -void -xfs_blkdev_issue_flush( - xfs_buftarg_t *buftarg) -{ - blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL); -} - -STATIC void -xfs_close_devices( - struct xfs_mount *mp) -{ - if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { - struct block_device *logdev = mp->m_logdev_targp->bt_bdev; - xfs_free_buftarg(mp, mp->m_logdev_targp); - xfs_blkdev_put(logdev); - } - if (mp->m_rtdev_targp) { - struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev; - xfs_free_buftarg(mp, mp->m_rtdev_targp); - xfs_blkdev_put(rtdev); - } - xfs_free_buftarg(mp, mp->m_ddev_targp); -} - -/* - * The file system configurations are: - * (1) device (partition) with data and internal log - * (2) logical volume with data and log subvolumes. - * (3) logical volume with data, log, and realtime subvolumes. - * - * We only have to handle opening the log and realtime volumes here if - * they are present. The data subvolume has already been opened by - * get_sb_bdev() and is stored in sb->s_bdev. - */ -STATIC int -xfs_open_devices( - struct xfs_mount *mp) -{ - struct block_device *ddev = mp->m_super->s_bdev; - struct block_device *logdev = NULL, *rtdev = NULL; - int error; - - /* - * Open real time and log devices - order is important. - */ - if (mp->m_logname) { - error = xfs_blkdev_get(mp, mp->m_logname, &logdev); - if (error) - goto out; - } - - if (mp->m_rtname) { - error = xfs_blkdev_get(mp, mp->m_rtname, &rtdev); - if (error) - goto out_close_logdev; - - if (rtdev == ddev || rtdev == logdev) { - xfs_warn(mp, - "Cannot mount filesystem with identical rtdev and ddev/logdev."); - error = EINVAL; - goto out_close_rtdev; - } - } - - /* - * Setup xfs_mount buffer target pointers - */ - error = ENOMEM; - mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev, 0, mp->m_fsname); - if (!mp->m_ddev_targp) - goto out_close_rtdev; - - if (rtdev) { - mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev, 1, - mp->m_fsname); - if (!mp->m_rtdev_targp) - goto out_free_ddev_targ; - } - - if (logdev && logdev != ddev) { - mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev, 1, - mp->m_fsname); - if (!mp->m_logdev_targp) - goto out_free_rtdev_targ; - } else { - mp->m_logdev_targp = mp->m_ddev_targp; - } - - return 0; - - out_free_rtdev_targ: - if (mp->m_rtdev_targp) - xfs_free_buftarg(mp, mp->m_rtdev_targp); - out_free_ddev_targ: - xfs_free_buftarg(mp, mp->m_ddev_targp); - out_close_rtdev: - if (rtdev) - xfs_blkdev_put(rtdev); - out_close_logdev: - if (logdev && logdev != ddev) - xfs_blkdev_put(logdev); - out: - return error; -} - -/* - * Setup xfs_mount buffer target pointers based on superblock - */ -STATIC int -xfs_setup_devices( - struct xfs_mount *mp) -{ - int error; - - error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize, - mp->m_sb.sb_sectsize); - if (error) - return error; - - if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { - unsigned int log_sector_size = BBSIZE; - - if (xfs_sb_version_hassector(&mp->m_sb)) - log_sector_size = mp->m_sb.sb_logsectsize; - error = xfs_setsize_buftarg(mp->m_logdev_targp, - mp->m_sb.sb_blocksize, - log_sector_size); - if (error) - return error; - } - if (mp->m_rtdev_targp) { - error = xfs_setsize_buftarg(mp->m_rtdev_targp, - mp->m_sb.sb_blocksize, - mp->m_sb.sb_sectsize); - if (error) - return error; - } - - return 0; -} - -/* Catch misguided souls that try to use this interface on XFS */ -STATIC struct inode * -xfs_fs_alloc_inode( - struct super_block *sb) -{ - BUG(); - return NULL; -} - -/* - * Now that the generic code is guaranteed not to be accessing - * the linux inode, we can reclaim the inode. - */ -STATIC void -xfs_fs_destroy_inode( - struct inode *inode) -{ - struct xfs_inode *ip = XFS_I(inode); - - trace_xfs_destroy_inode(ip); - - XFS_STATS_INC(vn_reclaim); - - /* bad inode, get out here ASAP */ - if (is_bad_inode(inode)) - goto out_reclaim; - - xfs_ioend_wait(ip); - - ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0); - - /* - * We should never get here with one of the reclaim flags already set. - */ - ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIMABLE)); - ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM)); - - /* - * We always use background reclaim here because even if the - * inode is clean, it still may be under IO and hence we have - * to take the flush lock. The background reclaim path handles - * this more efficiently than we can here, so simply let background - * reclaim tear down all inodes. - */ -out_reclaim: - xfs_inode_set_reclaim_tag(ip); -} - -/* - * Slab object creation initialisation for the XFS inode. - * This covers only the idempotent fields in the XFS inode; - * all other fields need to be initialised on allocation - * from the slab. This avoids the need to repeatedly initialise - * fields in the xfs inode that left in the initialise state - * when freeing the inode. - */ -STATIC void -xfs_fs_inode_init_once( - void *inode) -{ - struct xfs_inode *ip = inode; - - memset(ip, 0, sizeof(struct xfs_inode)); - - /* vfs inode */ - inode_init_once(VFS_I(ip)); - - /* xfs inode */ - atomic_set(&ip->i_iocount, 0); - atomic_set(&ip->i_pincount, 0); - spin_lock_init(&ip->i_flags_lock); - init_waitqueue_head(&ip->i_ipin_wait); - /* - * Because we want to use a counting completion, complete - * the flush completion once to allow a single access to - * the flush completion without blocking. - */ - init_completion(&ip->i_flush); - complete(&ip->i_flush); - - mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER, - "xfsino", ip->i_ino); -} - -/* - * Dirty the XFS inode when mark_inode_dirty_sync() is called so that - * we catch unlogged VFS level updates to the inode. - * - * We need the barrier() to maintain correct ordering between unlogged - * updates and the transaction commit code that clears the i_update_core - * field. This requires all updates to be completed before marking the - * inode dirty. - */ -STATIC void -xfs_fs_dirty_inode( - struct inode *inode, - int flags) -{ - barrier(); - XFS_I(inode)->i_update_core = 1; -} - -STATIC int -xfs_log_inode( - struct xfs_inode *ip) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_trans *tp; - int error; - - xfs_iunlock(ip, XFS_ILOCK_SHARED); - tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); - error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); - - if (error) { - xfs_trans_cancel(tp, 0); - /* we need to return with the lock hold shared */ - xfs_ilock(ip, XFS_ILOCK_SHARED); - return error; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * Note - it's possible that we might have pushed ourselves out of the - * way during trans_reserve which would flush the inode. But there's - * no guarantee that the inode buffer has actually gone out yet (it's - * delwri). Plus the buffer could be pinned anyway if it's part of - * an inode in another recent transaction. So we play it safe and - * fire off the transaction anyway. - */ - xfs_trans_ijoin(tp, ip); - xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); - error = xfs_trans_commit(tp, 0); - xfs_ilock_demote(ip, XFS_ILOCK_EXCL); - - return error; -} - -STATIC int -xfs_fs_write_inode( - struct inode *inode, - struct writeback_control *wbc) -{ - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - int error = EAGAIN; - - trace_xfs_write_inode(ip); - - if (XFS_FORCED_SHUTDOWN(mp)) - return XFS_ERROR(EIO); - - if (wbc->sync_mode == WB_SYNC_ALL) { - /* - * Make sure the inode has made it it into the log. Instead - * of forcing it all the way to stable storage using a - * synchronous transaction we let the log force inside the - * ->sync_fs call do that for thus, which reduces the number - * of synchronous log foces dramatically. - */ - xfs_ioend_wait(ip); - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (ip->i_update_core) { - error = xfs_log_inode(ip); - if (error) - goto out_unlock; - } - } else { - /* - * We make this non-blocking if the inode is contended, return - * EAGAIN to indicate to the caller that they did not succeed. - * This prevents the flush path from blocking on inodes inside - * another operation right now, they get caught later by - * xfs_sync. - */ - if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) - goto out; - - if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) - goto out_unlock; - - /* - * Now we have the flush lock and the inode is not pinned, we - * can check if the inode is really clean as we know that - * there are no pending transaction completions, it is not - * waiting on the delayed write queue and there is no IO in - * progress. - */ - if (xfs_inode_clean(ip)) { - xfs_ifunlock(ip); - error = 0; - goto out_unlock; - } - error = xfs_iflush(ip, SYNC_TRYLOCK); - } - - out_unlock: - xfs_iunlock(ip, XFS_ILOCK_SHARED); - out: - /* - * if we failed to write out the inode then mark - * it dirty again so we'll try again later. - */ - if (error) - xfs_mark_inode_dirty_sync(ip); - return -error; -} - -STATIC void -xfs_fs_evict_inode( - struct inode *inode) -{ - xfs_inode_t *ip = XFS_I(inode); - - trace_xfs_evict_inode(ip); - - truncate_inode_pages(&inode->i_data, 0); - end_writeback(inode); - XFS_STATS_INC(vn_rele); - XFS_STATS_INC(vn_remove); - XFS_STATS_DEC(vn_active); - - /* - * The iolock is used by the file system to coordinate reads, - * writes, and block truncates. Up to this point the lock - * protected concurrent accesses by users of the inode. But - * from here forward we're doing some final processing of the - * inode because we're done with it, and although we reuse the - * iolock for protection it is really a distinct lock class - * (in the lockdep sense) from before. To keep lockdep happy - * (and basically indicate what we are doing), we explicitly - * re-init the iolock here. - */ - ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock)); - mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino); - lockdep_set_class_and_name(&ip->i_iolock.mr_lock, - &xfs_iolock_reclaimable, "xfs_iolock_reclaimable"); - - xfs_inactive(ip); -} - -STATIC void -xfs_free_fsname( - struct xfs_mount *mp) -{ - kfree(mp->m_fsname); - kfree(mp->m_rtname); - kfree(mp->m_logname); -} - -STATIC void -xfs_fs_put_super( - struct super_block *sb) -{ - struct xfs_mount *mp = XFS_M(sb); - - xfs_syncd_stop(mp); - - /* - * Blow away any referenced inode in the filestreams cache. - * This can and will cause log traffic as inodes go inactive - * here. - */ - xfs_filestream_unmount(mp); - - XFS_bflush(mp->m_ddev_targp); - - xfs_unmountfs(mp); - xfs_freesb(mp); - xfs_icsb_destroy_counters(mp); - xfs_close_devices(mp); - xfs_free_fsname(mp); - kfree(mp); -} - -STATIC int -xfs_fs_sync_fs( - struct super_block *sb, - int wait) -{ - struct xfs_mount *mp = XFS_M(sb); - int error; - - /* - * Not much we can do for the first async pass. Writing out the - * superblock would be counter-productive as we are going to redirty - * when writing out other data and metadata (and writing out a single - * block is quite fast anyway). - * - * Try to asynchronously kick off quota syncing at least. - */ - if (!wait) { - xfs_qm_sync(mp, SYNC_TRYLOCK); - return 0; - } - - error = xfs_quiesce_data(mp); - if (error) - return -error; - - if (laptop_mode) { - /* - * The disk must be active because we're syncing. - * We schedule xfssyncd now (now that the disk is - * active) instead of later (when it might not be). - */ - flush_delayed_work_sync(&mp->m_sync_work); - } - - return 0; -} - -STATIC int -xfs_fs_statfs( - struct dentry *dentry, - struct kstatfs *statp) -{ - struct xfs_mount *mp = XFS_M(dentry->d_sb); - xfs_sb_t *sbp = &mp->m_sb; - struct xfs_inode *ip = XFS_I(dentry->d_inode); - __uint64_t fakeinos, id; - xfs_extlen_t lsize; - __int64_t ffree; - - statp->f_type = XFS_SB_MAGIC; - statp->f_namelen = MAXNAMELEN - 1; - - id = huge_encode_dev(mp->m_ddev_targp->bt_dev); - statp->f_fsid.val[0] = (u32)id; - statp->f_fsid.val[1] = (u32)(id >> 32); - - xfs_icsb_sync_counters(mp, XFS_ICSB_LAZY_COUNT); - - spin_lock(&mp->m_sb_lock); - statp->f_bsize = sbp->sb_blocksize; - lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; - statp->f_blocks = sbp->sb_dblocks - lsize; - statp->f_bfree = statp->f_bavail = - sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); - fakeinos = statp->f_bfree << sbp->sb_inopblog; - statp->f_files = - MIN(sbp->sb_icount + fakeinos, (__uint64_t)XFS_MAXINUMBER); - if (mp->m_maxicount) - statp->f_files = min_t(typeof(statp->f_files), - statp->f_files, - mp->m_maxicount); - - /* make sure statp->f_ffree does not underflow */ - ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); - statp->f_ffree = max_t(__int64_t, ffree, 0); - - spin_unlock(&mp->m_sb_lock); - - if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) || - ((mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))) == - (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD)) - xfs_qm_statvfs(ip, statp); - return 0; -} - -STATIC void -xfs_save_resvblks(struct xfs_mount *mp) -{ - __uint64_t resblks = 0; - - mp->m_resblks_save = mp->m_resblks; - xfs_reserve_blocks(mp, &resblks, NULL); -} - -STATIC void -xfs_restore_resvblks(struct xfs_mount *mp) -{ - __uint64_t resblks; - - if (mp->m_resblks_save) { - resblks = mp->m_resblks_save; - mp->m_resblks_save = 0; - } else - resblks = xfs_default_resblks(mp); - - xfs_reserve_blocks(mp, &resblks, NULL); -} - -STATIC int -xfs_fs_remount( - struct super_block *sb, - int *flags, - char *options) -{ - struct xfs_mount *mp = XFS_M(sb); - substring_t args[MAX_OPT_ARGS]; - char *p; - int error; - - while ((p = strsep(&options, ",")) != NULL) { - int token; - - if (!*p) - continue; - - token = match_token(p, tokens, args); - switch (token) { - case Opt_barrier: - mp->m_flags |= XFS_MOUNT_BARRIER; - break; - case Opt_nobarrier: - mp->m_flags &= ~XFS_MOUNT_BARRIER; - break; - default: - /* - * Logically we would return an error here to prevent - * users from believing they might have changed - * mount options using remount which can't be changed. - * - * But unfortunately mount(8) adds all options from - * mtab and fstab to the mount arguments in some cases - * so we can't blindly reject options, but have to - * check for each specified option if it actually - * differs from the currently set option and only - * reject it if that's the case. - * - * Until that is implemented we return success for - * every remount request, and silently ignore all - * options that we can't actually change. - */ -#if 0 - xfs_info(mp, - "mount option \"%s\" not supported for remount\n", p); - return -EINVAL; -#else - break; -#endif - } - } - - /* ro -> rw */ - if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) { - mp->m_flags &= ~XFS_MOUNT_RDONLY; - - /* - * If this is the first remount to writeable state we - * might have some superblock changes to update. - */ - if (mp->m_update_flags) { - error = xfs_mount_log_sb(mp, mp->m_update_flags); - if (error) { - xfs_warn(mp, "failed to write sb changes"); - return error; - } - mp->m_update_flags = 0; - } - - /* - * Fill out the reserve pool if it is empty. Use the stashed - * value if it is non-zero, otherwise go with the default. - */ - xfs_restore_resvblks(mp); - } - - /* rw -> ro */ - if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) { - /* - * After we have synced the data but before we sync the - * metadata, we need to free up the reserve block pool so that - * the used block count in the superblock on disk is correct at - * the end of the remount. Stash the current reserve pool size - * so that if we get remounted rw, we can return it to the same - * size. - */ - - xfs_quiesce_data(mp); - xfs_save_resvblks(mp); - xfs_quiesce_attr(mp); - mp->m_flags |= XFS_MOUNT_RDONLY; - } - - return 0; -} - -/* - * Second stage of a freeze. The data is already frozen so we only - * need to take care of the metadata. Once that's done write a dummy - * record to dirty the log in case of a crash while frozen. - */ -STATIC int -xfs_fs_freeze( - struct super_block *sb) -{ - struct xfs_mount *mp = XFS_M(sb); - - xfs_save_resvblks(mp); - xfs_quiesce_attr(mp); - return -xfs_fs_log_dummy(mp); -} - -STATIC int -xfs_fs_unfreeze( - struct super_block *sb) -{ - struct xfs_mount *mp = XFS_M(sb); - - xfs_restore_resvblks(mp); - return 0; -} - -STATIC int -xfs_fs_show_options( - struct seq_file *m, - struct vfsmount *mnt) -{ - return -xfs_showargs(XFS_M(mnt->mnt_sb), m); -} - -/* - * This function fills in xfs_mount_t fields based on mount args. - * Note: the superblock _has_ now been read in. - */ -STATIC int -xfs_finish_flags( - struct xfs_mount *mp) -{ - int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); - - /* Fail a mount where the logbuf is smaller than the log stripe */ - if (xfs_sb_version_haslogv2(&mp->m_sb)) { - if (mp->m_logbsize <= 0 && - mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) { - mp->m_logbsize = mp->m_sb.sb_logsunit; - } else if (mp->m_logbsize > 0 && - mp->m_logbsize < mp->m_sb.sb_logsunit) { - xfs_warn(mp, - "logbuf size must be greater than or equal to log stripe size"); - return XFS_ERROR(EINVAL); - } - } else { - /* Fail a mount if the logbuf is larger than 32K */ - if (mp->m_logbsize > XLOG_BIG_RECORD_BSIZE) { - xfs_warn(mp, - "logbuf size for version 1 logs must be 16K or 32K"); - return XFS_ERROR(EINVAL); - } - } - - /* - * mkfs'ed attr2 will turn on attr2 mount unless explicitly - * told by noattr2 to turn it off - */ - if (xfs_sb_version_hasattr2(&mp->m_sb) && - !(mp->m_flags & XFS_MOUNT_NOATTR2)) - mp->m_flags |= XFS_MOUNT_ATTR2; - - /* - * prohibit r/w mounts of read-only filesystems - */ - if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) { - xfs_warn(mp, - "cannot mount a read-only filesystem as read-write"); - return XFS_ERROR(EROFS); - } - - return 0; -} - -STATIC int -xfs_fs_fill_super( - struct super_block *sb, - void *data, - int silent) -{ - struct inode *root; - struct xfs_mount *mp = NULL; - int flags = 0, error = ENOMEM; - - mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL); - if (!mp) - goto out; - - spin_lock_init(&mp->m_sb_lock); - mutex_init(&mp->m_growlock); - atomic_set(&mp->m_active_trans, 0); - - mp->m_super = sb; - sb->s_fs_info = mp; - - error = xfs_parseargs(mp, (char *)data); - if (error) - goto out_free_fsname; - - sb_min_blocksize(sb, BBSIZE); - sb->s_xattr = xfs_xattr_handlers; - sb->s_export_op = &xfs_export_operations; -#ifdef CONFIG_XFS_QUOTA - sb->s_qcop = &xfs_quotactl_operations; -#endif - sb->s_op = &xfs_super_operations; - - if (silent) - flags |= XFS_MFSI_QUIET; - - error = xfs_open_devices(mp); - if (error) - goto out_free_fsname; - - error = xfs_icsb_init_counters(mp); - if (error) - goto out_close_devices; - - error = xfs_readsb(mp, flags); - if (error) - goto out_destroy_counters; - - error = xfs_finish_flags(mp); - if (error) - goto out_free_sb; - - error = xfs_setup_devices(mp); - if (error) - goto out_free_sb; - - error = xfs_filestream_mount(mp); - if (error) - goto out_free_sb; - - /* - * we must configure the block size in the superblock before we run the - * full mount process as the mount process can lookup and cache inodes. - * For the same reason we must also initialise the syncd and register - * the inode cache shrinker so that inodes can be reclaimed during - * operations like a quotacheck that iterate all inodes in the - * filesystem. - */ - sb->s_magic = XFS_SB_MAGIC; - sb->s_blocksize = mp->m_sb.sb_blocksize; - sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1; - sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits); - sb->s_time_gran = 1; - set_posix_acl_flag(sb); - - error = xfs_mountfs(mp); - if (error) - goto out_filestream_unmount; - - error = xfs_syncd_init(mp); - if (error) - goto out_unmount; - - root = igrab(VFS_I(mp->m_rootip)); - if (!root) { - error = ENOENT; - goto out_syncd_stop; - } - if (is_bad_inode(root)) { - error = EINVAL; - goto out_syncd_stop; - } - sb->s_root = d_alloc_root(root); - if (!sb->s_root) { - error = ENOMEM; - goto out_iput; - } - - return 0; - - out_filestream_unmount: - xfs_filestream_unmount(mp); - out_free_sb: - xfs_freesb(mp); - out_destroy_counters: - xfs_icsb_destroy_counters(mp); - out_close_devices: - xfs_close_devices(mp); - out_free_fsname: - xfs_free_fsname(mp); - kfree(mp); - out: - return -error; - - out_iput: - iput(root); - out_syncd_stop: - xfs_syncd_stop(mp); - out_unmount: - /* - * Blow away any referenced inode in the filestreams cache. - * This can and will cause log traffic as inodes go inactive - * here. - */ - xfs_filestream_unmount(mp); - - XFS_bflush(mp->m_ddev_targp); - - xfs_unmountfs(mp); - goto out_free_sb; -} - -STATIC struct dentry * -xfs_fs_mount( - struct file_system_type *fs_type, - int flags, - const char *dev_name, - void *data) -{ - return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super); -} - -static int -xfs_fs_nr_cached_objects( - struct super_block *sb) -{ - return xfs_reclaim_inodes_count(XFS_M(sb)); -} - -static void -xfs_fs_free_cached_objects( - struct super_block *sb, - int nr_to_scan) -{ - xfs_reclaim_inodes_nr(XFS_M(sb), nr_to_scan); -} - -static const struct super_operations xfs_super_operations = { - .alloc_inode = xfs_fs_alloc_inode, - .destroy_inode = xfs_fs_destroy_inode, - .dirty_inode = xfs_fs_dirty_inode, - .write_inode = xfs_fs_write_inode, - .evict_inode = xfs_fs_evict_inode, - .put_super = xfs_fs_put_super, - .sync_fs = xfs_fs_sync_fs, - .freeze_fs = xfs_fs_freeze, - .unfreeze_fs = xfs_fs_unfreeze, - .statfs = xfs_fs_statfs, - .remount_fs = xfs_fs_remount, - .show_options = xfs_fs_show_options, - .nr_cached_objects = xfs_fs_nr_cached_objects, - .free_cached_objects = xfs_fs_free_cached_objects, -}; - -static struct file_system_type xfs_fs_type = { - .owner = THIS_MODULE, - .name = "xfs", - .mount = xfs_fs_mount, - .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, -}; - -STATIC int __init -xfs_init_zones(void) -{ - - xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend"); - if (!xfs_ioend_zone) - goto out; - - xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE, - xfs_ioend_zone); - if (!xfs_ioend_pool) - goto out_destroy_ioend_zone; - - xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t), - "xfs_log_ticket"); - if (!xfs_log_ticket_zone) - goto out_destroy_ioend_pool; - - xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t), - "xfs_bmap_free_item"); - if (!xfs_bmap_free_item_zone) - goto out_destroy_log_ticket_zone; - - xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t), - "xfs_btree_cur"); - if (!xfs_btree_cur_zone) - goto out_destroy_bmap_free_item_zone; - - xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t), - "xfs_da_state"); - if (!xfs_da_state_zone) - goto out_destroy_btree_cur_zone; - - xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf"); - if (!xfs_dabuf_zone) - goto out_destroy_da_state_zone; - - xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork"); - if (!xfs_ifork_zone) - goto out_destroy_dabuf_zone; - - xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans"); - if (!xfs_trans_zone) - goto out_destroy_ifork_zone; - - xfs_log_item_desc_zone = - kmem_zone_init(sizeof(struct xfs_log_item_desc), - "xfs_log_item_desc"); - if (!xfs_log_item_desc_zone) - goto out_destroy_trans_zone; - - /* - * The size of the zone allocated buf log item is the maximum - * size possible under XFS. This wastes a little bit of memory, - * but it is much faster. - */ - xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) + - (((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / - NBWORD) * sizeof(int))), "xfs_buf_item"); - if (!xfs_buf_item_zone) - goto out_destroy_log_item_desc_zone; - - xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) + - ((XFS_EFD_MAX_FAST_EXTENTS - 1) * - sizeof(xfs_extent_t))), "xfs_efd_item"); - if (!xfs_efd_zone) - goto out_destroy_buf_item_zone; - - xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) + - ((XFS_EFI_MAX_FAST_EXTENTS - 1) * - sizeof(xfs_extent_t))), "xfs_efi_item"); - if (!xfs_efi_zone) - goto out_destroy_efd_zone; - - xfs_inode_zone = - kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode", - KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD, - xfs_fs_inode_init_once); - if (!xfs_inode_zone) - goto out_destroy_efi_zone; - - xfs_ili_zone = - kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili", - KM_ZONE_SPREAD, NULL); - if (!xfs_ili_zone) - goto out_destroy_inode_zone; - - return 0; - - out_destroy_inode_zone: - kmem_zone_destroy(xfs_inode_zone); - out_destroy_efi_zone: - kmem_zone_destroy(xfs_efi_zone); - out_destroy_efd_zone: - kmem_zone_destroy(xfs_efd_zone); - out_destroy_buf_item_zone: - kmem_zone_destroy(xfs_buf_item_zone); - out_destroy_log_item_desc_zone: - kmem_zone_destroy(xfs_log_item_desc_zone); - out_destroy_trans_zone: - kmem_zone_destroy(xfs_trans_zone); - out_destroy_ifork_zone: - kmem_zone_destroy(xfs_ifork_zone); - out_destroy_dabuf_zone: - kmem_zone_destroy(xfs_dabuf_zone); - out_destroy_da_state_zone: - kmem_zone_destroy(xfs_da_state_zone); - out_destroy_btree_cur_zone: - kmem_zone_destroy(xfs_btree_cur_zone); - out_destroy_bmap_free_item_zone: - kmem_zone_destroy(xfs_bmap_free_item_zone); - out_destroy_log_ticket_zone: - kmem_zone_destroy(xfs_log_ticket_zone); - out_destroy_ioend_pool: - mempool_destroy(xfs_ioend_pool); - out_destroy_ioend_zone: - kmem_zone_destroy(xfs_ioend_zone); - out: - return -ENOMEM; -} - -STATIC void -xfs_destroy_zones(void) -{ - kmem_zone_destroy(xfs_ili_zone); - kmem_zone_destroy(xfs_inode_zone); - kmem_zone_destroy(xfs_efi_zone); - kmem_zone_destroy(xfs_efd_zone); - kmem_zone_destroy(xfs_buf_item_zone); - kmem_zone_destroy(xfs_log_item_desc_zone); - kmem_zone_destroy(xfs_trans_zone); - kmem_zone_destroy(xfs_ifork_zone); - kmem_zone_destroy(xfs_dabuf_zone); - kmem_zone_destroy(xfs_da_state_zone); - kmem_zone_destroy(xfs_btree_cur_zone); - kmem_zone_destroy(xfs_bmap_free_item_zone); - kmem_zone_destroy(xfs_log_ticket_zone); - mempool_destroy(xfs_ioend_pool); - kmem_zone_destroy(xfs_ioend_zone); - -} - -STATIC int __init -xfs_init_workqueues(void) -{ - /* - * max_active is set to 8 to give enough concurency to allow - * multiple work operations on each CPU to run. This allows multiple - * filesystems to be running sync work concurrently, and scales with - * the number of CPUs in the system. - */ - xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8); - if (!xfs_syncd_wq) - goto out; - - xfs_ail_wq = alloc_workqueue("xfsail", WQ_CPU_INTENSIVE, 8); - if (!xfs_ail_wq) - goto out_destroy_syncd; - - return 0; - -out_destroy_syncd: - destroy_workqueue(xfs_syncd_wq); -out: - return -ENOMEM; -} - -STATIC void -xfs_destroy_workqueues(void) -{ - destroy_workqueue(xfs_ail_wq); - destroy_workqueue(xfs_syncd_wq); -} - -STATIC int __init -init_xfs_fs(void) -{ - int error; - - printk(KERN_INFO XFS_VERSION_STRING " with " - XFS_BUILD_OPTIONS " enabled\n"); - - xfs_ioend_init(); - xfs_dir_startup(); - - error = xfs_init_zones(); - if (error) - goto out; - - error = xfs_init_workqueues(); - if (error) - goto out_destroy_zones; - - error = xfs_mru_cache_init(); - if (error) - goto out_destroy_wq; - - error = xfs_filestream_init(); - if (error) - goto out_mru_cache_uninit; - - error = xfs_buf_init(); - if (error) - goto out_filestream_uninit; - - error = xfs_init_procfs(); - if (error) - goto out_buf_terminate; - - error = xfs_sysctl_register(); - if (error) - goto out_cleanup_procfs; - - vfs_initquota(); - - error = register_filesystem(&xfs_fs_type); - if (error) - goto out_sysctl_unregister; - return 0; - - out_sysctl_unregister: - xfs_sysctl_unregister(); - out_cleanup_procfs: - xfs_cleanup_procfs(); - out_buf_terminate: - xfs_buf_terminate(); - out_filestream_uninit: - xfs_filestream_uninit(); - out_mru_cache_uninit: - xfs_mru_cache_uninit(); - out_destroy_wq: - xfs_destroy_workqueues(); - out_destroy_zones: - xfs_destroy_zones(); - out: - return error; -} - -STATIC void __exit -exit_xfs_fs(void) -{ - vfs_exitquota(); - unregister_filesystem(&xfs_fs_type); - xfs_sysctl_unregister(); - xfs_cleanup_procfs(); - xfs_buf_terminate(); - xfs_filestream_uninit(); - xfs_mru_cache_uninit(); - xfs_destroy_workqueues(); - xfs_destroy_zones(); -} - -module_init(init_xfs_fs); -module_exit(exit_xfs_fs); - -MODULE_AUTHOR("Silicon Graphics, Inc."); -MODULE_DESCRIPTION(XFS_VERSION_STRING " with " XFS_BUILD_OPTIONS " enabled"); -MODULE_LICENSE("GPL"); diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h deleted file mode 100644 index 50a3266c999e..000000000000 --- a/fs/xfs/linux-2.6/xfs_super.h +++ /dev/null @@ -1,87 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPER_H__ -#define __XFS_SUPER_H__ - -#include - -#ifdef CONFIG_XFS_QUOTA -extern void xfs_qm_init(void); -extern void xfs_qm_exit(void); -# define vfs_initquota() xfs_qm_init() -# define vfs_exitquota() xfs_qm_exit() -#else -# define vfs_initquota() do { } while (0) -# define vfs_exitquota() do { } while (0) -#endif - -#ifdef CONFIG_XFS_POSIX_ACL -# define XFS_ACL_STRING "ACLs, " -# define set_posix_acl_flag(sb) ((sb)->s_flags |= MS_POSIXACL) -#else -# define XFS_ACL_STRING -# define set_posix_acl_flag(sb) do { } while (0) -#endif - -#define XFS_SECURITY_STRING "security attributes, " - -#ifdef CONFIG_XFS_RT -# define XFS_REALTIME_STRING "realtime, " -#else -# define XFS_REALTIME_STRING -#endif - -#if XFS_BIG_BLKNOS -# if XFS_BIG_INUMS -# define XFS_BIGFS_STRING "large block/inode numbers, " -# else -# define XFS_BIGFS_STRING "large block numbers, " -# endif -#else -# define XFS_BIGFS_STRING -#endif - -#ifdef DEBUG -# define XFS_DBG_STRING "debug" -#else -# define XFS_DBG_STRING "no debug" -#endif - -#define XFS_VERSION_STRING "SGI XFS" -#define XFS_BUILD_OPTIONS XFS_ACL_STRING \ - XFS_SECURITY_STRING \ - XFS_REALTIME_STRING \ - XFS_BIGFS_STRING \ - XFS_DBG_STRING /* DBG must be last */ - -struct xfs_inode; -struct xfs_mount; -struct xfs_buftarg; -struct block_device; - -extern __uint64_t xfs_max_file_offset(unsigned int); - -extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); - -extern const struct export_operations xfs_export_operations; -extern const struct xattr_handler *xfs_xattr_handlers[]; -extern const struct quotactl_ops xfs_quotactl_operations; - -#define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info)) - -#endif /* __XFS_SUPER_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c deleted file mode 100644 index 4604f90f86a3..000000000000 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ /dev/null @@ -1,1065 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_trans_priv.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_dinode.h" -#include "xfs_error.h" -#include "xfs_filestream.h" -#include "xfs_vnodeops.h" -#include "xfs_inode_item.h" -#include "xfs_quota.h" -#include "xfs_trace.h" -#include "xfs_fsops.h" - -#include -#include - -struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ - -/* - * The inode lookup is done in batches to keep the amount of lock traffic and - * radix tree lookups to a minimum. The batch size is a trade off between - * lookup reduction and stack usage. This is in the reclaim path, so we can't - * be too greedy. - */ -#define XFS_LOOKUP_BATCH 32 - -STATIC int -xfs_inode_ag_walk_grab( - struct xfs_inode *ip) -{ - struct inode *inode = VFS_I(ip); - - ASSERT(rcu_read_lock_held()); - - /* - * check for stale RCU freed inode - * - * If the inode has been reallocated, it doesn't matter if it's not in - * the AG we are walking - we are walking for writeback, so if it - * passes all the "valid inode" checks and is dirty, then we'll write - * it back anyway. If it has been reallocated and still being - * initialised, the XFS_INEW check below will catch it. - */ - spin_lock(&ip->i_flags_lock); - if (!ip->i_ino) - goto out_unlock_noent; - - /* avoid new or reclaimable inodes. Leave for reclaim code to flush */ - if (__xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM)) - goto out_unlock_noent; - spin_unlock(&ip->i_flags_lock); - - /* nothing to sync during shutdown */ - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) - return EFSCORRUPTED; - - /* If we can't grab the inode, it must on it's way to reclaim. */ - if (!igrab(inode)) - return ENOENT; - - if (is_bad_inode(inode)) { - IRELE(ip); - return ENOENT; - } - - /* inode is valid */ - return 0; - -out_unlock_noent: - spin_unlock(&ip->i_flags_lock); - return ENOENT; -} - -STATIC int -xfs_inode_ag_walk( - struct xfs_mount *mp, - struct xfs_perag *pag, - int (*execute)(struct xfs_inode *ip, - struct xfs_perag *pag, int flags), - int flags) -{ - uint32_t first_index; - int last_error = 0; - int skipped; - int done; - int nr_found; - -restart: - done = 0; - skipped = 0; - first_index = 0; - nr_found = 0; - do { - struct xfs_inode *batch[XFS_LOOKUP_BATCH]; - int error = 0; - int i; - - rcu_read_lock(); - nr_found = radix_tree_gang_lookup(&pag->pag_ici_root, - (void **)batch, first_index, - XFS_LOOKUP_BATCH); - if (!nr_found) { - rcu_read_unlock(); - break; - } - - /* - * Grab the inodes before we drop the lock. if we found - * nothing, nr == 0 and the loop will be skipped. - */ - for (i = 0; i < nr_found; i++) { - struct xfs_inode *ip = batch[i]; - - if (done || xfs_inode_ag_walk_grab(ip)) - batch[i] = NULL; - - /* - * Update the index for the next lookup. Catch - * overflows into the next AG range which can occur if - * we have inodes in the last block of the AG and we - * are currently pointing to the last inode. - * - * Because we may see inodes that are from the wrong AG - * due to RCU freeing and reallocation, only update the - * index if it lies in this AG. It was a race that lead - * us to see this inode, so another lookup from the - * same index will not find it again. - */ - if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag->pag_agno) - continue; - first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); - if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) - done = 1; - } - - /* unlock now we've grabbed the inodes. */ - rcu_read_unlock(); - - for (i = 0; i < nr_found; i++) { - if (!batch[i]) - continue; - error = execute(batch[i], pag, flags); - IRELE(batch[i]); - if (error == EAGAIN) { - skipped++; - continue; - } - if (error && last_error != EFSCORRUPTED) - last_error = error; - } - - /* bail out if the filesystem is corrupted. */ - if (error == EFSCORRUPTED) - break; - - cond_resched(); - - } while (nr_found && !done); - - if (skipped) { - delay(1); - goto restart; - } - return last_error; -} - -int -xfs_inode_ag_iterator( - struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, - struct xfs_perag *pag, int flags), - int flags) -{ - struct xfs_perag *pag; - int error = 0; - int last_error = 0; - xfs_agnumber_t ag; - - ag = 0; - while ((pag = xfs_perag_get(mp, ag))) { - ag = pag->pag_agno + 1; - error = xfs_inode_ag_walk(mp, pag, execute, flags); - xfs_perag_put(pag); - if (error) { - last_error = error; - if (error == EFSCORRUPTED) - break; - } - } - return XFS_ERROR(last_error); -} - -STATIC int -xfs_sync_inode_data( - struct xfs_inode *ip, - struct xfs_perag *pag, - int flags) -{ - struct inode *inode = VFS_I(ip); - struct address_space *mapping = inode->i_mapping; - int error = 0; - - if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) - goto out_wait; - - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) { - if (flags & SYNC_TRYLOCK) - goto out_wait; - xfs_ilock(ip, XFS_IOLOCK_SHARED); - } - - error = xfs_flush_pages(ip, 0, -1, (flags & SYNC_WAIT) ? - 0 : XBF_ASYNC, FI_NONE); - xfs_iunlock(ip, XFS_IOLOCK_SHARED); - - out_wait: - if (flags & SYNC_WAIT) - xfs_ioend_wait(ip); - return error; -} - -STATIC int -xfs_sync_inode_attr( - struct xfs_inode *ip, - struct xfs_perag *pag, - int flags) -{ - int error = 0; - - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (xfs_inode_clean(ip)) - goto out_unlock; - if (!xfs_iflock_nowait(ip)) { - if (!(flags & SYNC_WAIT)) - goto out_unlock; - xfs_iflock(ip); - } - - if (xfs_inode_clean(ip)) { - xfs_ifunlock(ip); - goto out_unlock; - } - - error = xfs_iflush(ip, flags); - - /* - * We don't want to try again on non-blocking flushes that can't run - * again immediately. If an inode really must be written, then that's - * what the SYNC_WAIT flag is for. - */ - if (error == EAGAIN) { - ASSERT(!(flags & SYNC_WAIT)); - error = 0; - } - - out_unlock: - xfs_iunlock(ip, XFS_ILOCK_SHARED); - return error; -} - -/* - * Write out pagecache data for the whole filesystem. - */ -STATIC int -xfs_sync_data( - struct xfs_mount *mp, - int flags) -{ - int error; - - ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0); - - error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags); - if (error) - return XFS_ERROR(error); - - xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0); - return 0; -} - -/* - * Write out inode metadata (attributes) for the whole filesystem. - */ -STATIC int -xfs_sync_attr( - struct xfs_mount *mp, - int flags) -{ - ASSERT((flags & ~SYNC_WAIT) == 0); - - return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags); -} - -STATIC int -xfs_sync_fsdata( - struct xfs_mount *mp) -{ - struct xfs_buf *bp; - - /* - * If the buffer is pinned then push on the log so we won't get stuck - * waiting in the write for someone, maybe ourselves, to flush the log. - * - * Even though we just pushed the log above, we did not have the - * superblock buffer locked at that point so it can become pinned in - * between there and here. - */ - bp = xfs_getsb(mp, 0); - if (xfs_buf_ispinned(bp)) - xfs_log_force(mp, 0); - - return xfs_bwrite(mp, bp); -} - -/* - * When remounting a filesystem read-only or freezing the filesystem, we have - * two phases to execute. This first phase is syncing the data before we - * quiesce the filesystem, and the second is flushing all the inodes out after - * we've waited for all the transactions created by the first phase to - * complete. The second phase ensures that the inodes are written to their - * location on disk rather than just existing in transactions in the log. This - * means after a quiesce there is no log replay required to write the inodes to - * disk (this is the main difference between a sync and a quiesce). - */ -/* - * First stage of freeze - no writers will make progress now we are here, - * so we flush delwri and delalloc buffers here, then wait for all I/O to - * complete. Data is frozen at that point. Metadata is not frozen, - * transactions can still occur here so don't bother flushing the buftarg - * because it'll just get dirty again. - */ -int -xfs_quiesce_data( - struct xfs_mount *mp) -{ - int error, error2 = 0; - - xfs_qm_sync(mp, SYNC_TRYLOCK); - xfs_qm_sync(mp, SYNC_WAIT); - - /* force out the newly dirtied log buffers */ - xfs_log_force(mp, XFS_LOG_SYNC); - - /* write superblock and hoover up shutdown errors */ - error = xfs_sync_fsdata(mp); - - /* make sure all delwri buffers are written out */ - xfs_flush_buftarg(mp->m_ddev_targp, 1); - - /* mark the log as covered if needed */ - if (xfs_log_need_covered(mp)) - error2 = xfs_fs_log_dummy(mp); - - /* flush data-only devices */ - if (mp->m_rtdev_targp) - XFS_bflush(mp->m_rtdev_targp); - - return error ? error : error2; -} - -STATIC void -xfs_quiesce_fs( - struct xfs_mount *mp) -{ - int count = 0, pincount; - - xfs_reclaim_inodes(mp, 0); - xfs_flush_buftarg(mp->m_ddev_targp, 0); - - /* - * This loop must run at least twice. The first instance of the loop - * will flush most meta data but that will generate more meta data - * (typically directory updates). Which then must be flushed and - * logged before we can write the unmount record. We also so sync - * reclaim of inodes to catch any that the above delwri flush skipped. - */ - do { - xfs_reclaim_inodes(mp, SYNC_WAIT); - xfs_sync_attr(mp, SYNC_WAIT); - pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1); - if (!pincount) { - delay(50); - count++; - } - } while (count < 2); -} - -/* - * Second stage of a quiesce. The data is already synced, now we have to take - * care of the metadata. New transactions are already blocked, so we need to - * wait for any remaining transactions to drain out before proceeding. - */ -void -xfs_quiesce_attr( - struct xfs_mount *mp) -{ - int error = 0; - - /* wait for all modifications to complete */ - while (atomic_read(&mp->m_active_trans) > 0) - delay(100); - - /* flush inodes and push all remaining buffers out to disk */ - xfs_quiesce_fs(mp); - - /* - * Just warn here till VFS can correctly support - * read-only remount without racing. - */ - WARN_ON(atomic_read(&mp->m_active_trans) != 0); - - /* Push the superblock and write an unmount record */ - error = xfs_log_sbcount(mp); - if (error) - xfs_warn(mp, "xfs_attr_quiesce: failed to log sb changes. " - "Frozen image may not be consistent."); - xfs_log_unmount_write(mp); - xfs_unmountfs_writesb(mp); -} - -static void -xfs_syncd_queue_sync( - struct xfs_mount *mp) -{ - queue_delayed_work(xfs_syncd_wq, &mp->m_sync_work, - msecs_to_jiffies(xfs_syncd_centisecs * 10)); -} - -/* - * Every sync period we need to unpin all items, reclaim inodes and sync - * disk quotas. We might need to cover the log to indicate that the - * filesystem is idle and not frozen. - */ -STATIC void -xfs_sync_worker( - struct work_struct *work) -{ - struct xfs_mount *mp = container_of(to_delayed_work(work), - struct xfs_mount, m_sync_work); - int error; - - if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { - /* dgc: errors ignored here */ - if (mp->m_super->s_frozen == SB_UNFROZEN && - xfs_log_need_covered(mp)) - error = xfs_fs_log_dummy(mp); - else - xfs_log_force(mp, 0); - error = xfs_qm_sync(mp, SYNC_TRYLOCK); - - /* start pushing all the metadata that is currently dirty */ - xfs_ail_push_all(mp->m_ail); - } - - /* queue us up again */ - xfs_syncd_queue_sync(mp); -} - -/* - * Queue a new inode reclaim pass if there are reclaimable inodes and there - * isn't a reclaim pass already in progress. By default it runs every 5s based - * on the xfs syncd work default of 30s. Perhaps this should have it's own - * tunable, but that can be done if this method proves to be ineffective or too - * aggressive. - */ -static void -xfs_syncd_queue_reclaim( - struct xfs_mount *mp) -{ - - /* - * We can have inodes enter reclaim after we've shut down the syncd - * workqueue during unmount, so don't allow reclaim work to be queued - * during unmount. - */ - if (!(mp->m_super->s_flags & MS_ACTIVE)) - return; - - rcu_read_lock(); - if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) { - queue_delayed_work(xfs_syncd_wq, &mp->m_reclaim_work, - msecs_to_jiffies(xfs_syncd_centisecs / 6 * 10)); - } - rcu_read_unlock(); -} - -/* - * This is a fast pass over the inode cache to try to get reclaim moving on as - * many inodes as possible in a short period of time. It kicks itself every few - * seconds, as well as being kicked by the inode cache shrinker when memory - * goes low. It scans as quickly as possible avoiding locked inodes or those - * already being flushed, and once done schedules a future pass. - */ -STATIC void -xfs_reclaim_worker( - struct work_struct *work) -{ - struct xfs_mount *mp = container_of(to_delayed_work(work), - struct xfs_mount, m_reclaim_work); - - xfs_reclaim_inodes(mp, SYNC_TRYLOCK); - xfs_syncd_queue_reclaim(mp); -} - -/* - * Flush delayed allocate data, attempting to free up reserved space - * from existing allocations. At this point a new allocation attempt - * has failed with ENOSPC and we are in the process of scratching our - * heads, looking about for more room. - * - * Queue a new data flush if there isn't one already in progress and - * wait for completion of the flush. This means that we only ever have one - * inode flush in progress no matter how many ENOSPC events are occurring and - * so will prevent the system from bogging down due to every concurrent - * ENOSPC event scanning all the active inodes in the system for writeback. - */ -void -xfs_flush_inodes( - struct xfs_inode *ip) -{ - struct xfs_mount *mp = ip->i_mount; - - queue_work(xfs_syncd_wq, &mp->m_flush_work); - flush_work_sync(&mp->m_flush_work); -} - -STATIC void -xfs_flush_worker( - struct work_struct *work) -{ - struct xfs_mount *mp = container_of(work, - struct xfs_mount, m_flush_work); - - xfs_sync_data(mp, SYNC_TRYLOCK); - xfs_sync_data(mp, SYNC_TRYLOCK | SYNC_WAIT); -} - -int -xfs_syncd_init( - struct xfs_mount *mp) -{ - INIT_WORK(&mp->m_flush_work, xfs_flush_worker); - INIT_DELAYED_WORK(&mp->m_sync_work, xfs_sync_worker); - INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker); - - xfs_syncd_queue_sync(mp); - xfs_syncd_queue_reclaim(mp); - - return 0; -} - -void -xfs_syncd_stop( - struct xfs_mount *mp) -{ - cancel_delayed_work_sync(&mp->m_sync_work); - cancel_delayed_work_sync(&mp->m_reclaim_work); - cancel_work_sync(&mp->m_flush_work); -} - -void -__xfs_inode_set_reclaim_tag( - struct xfs_perag *pag, - struct xfs_inode *ip) -{ - radix_tree_tag_set(&pag->pag_ici_root, - XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), - XFS_ICI_RECLAIM_TAG); - - if (!pag->pag_ici_reclaimable) { - /* propagate the reclaim tag up into the perag radix tree */ - spin_lock(&ip->i_mount->m_perag_lock); - radix_tree_tag_set(&ip->i_mount->m_perag_tree, - XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), - XFS_ICI_RECLAIM_TAG); - spin_unlock(&ip->i_mount->m_perag_lock); - - /* schedule periodic background inode reclaim */ - xfs_syncd_queue_reclaim(ip->i_mount); - - trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno, - -1, _RET_IP_); - } - pag->pag_ici_reclaimable++; -} - -/* - * We set the inode flag atomically with the radix tree tag. - * Once we get tag lookups on the radix tree, this inode flag - * can go away. - */ -void -xfs_inode_set_reclaim_tag( - xfs_inode_t *ip) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_perag *pag; - - pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); - spin_lock(&pag->pag_ici_lock); - spin_lock(&ip->i_flags_lock); - __xfs_inode_set_reclaim_tag(pag, ip); - __xfs_iflags_set(ip, XFS_IRECLAIMABLE); - spin_unlock(&ip->i_flags_lock); - spin_unlock(&pag->pag_ici_lock); - xfs_perag_put(pag); -} - -STATIC void -__xfs_inode_clear_reclaim( - xfs_perag_t *pag, - xfs_inode_t *ip) -{ - pag->pag_ici_reclaimable--; - if (!pag->pag_ici_reclaimable) { - /* clear the reclaim tag from the perag radix tree */ - spin_lock(&ip->i_mount->m_perag_lock); - radix_tree_tag_clear(&ip->i_mount->m_perag_tree, - XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), - XFS_ICI_RECLAIM_TAG); - spin_unlock(&ip->i_mount->m_perag_lock); - trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno, - -1, _RET_IP_); - } -} - -void -__xfs_inode_clear_reclaim_tag( - xfs_mount_t *mp, - xfs_perag_t *pag, - xfs_inode_t *ip) -{ - radix_tree_tag_clear(&pag->pag_ici_root, - XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); - __xfs_inode_clear_reclaim(pag, ip); -} - -/* - * Grab the inode for reclaim exclusively. - * Return 0 if we grabbed it, non-zero otherwise. - */ -STATIC int -xfs_reclaim_inode_grab( - struct xfs_inode *ip, - int flags) -{ - ASSERT(rcu_read_lock_held()); - - /* quick check for stale RCU freed inode */ - if (!ip->i_ino) - return 1; - - /* - * do some unlocked checks first to avoid unnecessary lock traffic. - * The first is a flush lock check, the second is a already in reclaim - * check. Only do these checks if we are not going to block on locks. - */ - if ((flags & SYNC_TRYLOCK) && - (!ip->i_flush.done || __xfs_iflags_test(ip, XFS_IRECLAIM))) { - return 1; - } - - /* - * The radix tree lock here protects a thread in xfs_iget from racing - * with us starting reclaim on the inode. Once we have the - * XFS_IRECLAIM flag set it will not touch us. - * - * Due to RCU lookup, we may find inodes that have been freed and only - * have XFS_IRECLAIM set. Indeed, we may see reallocated inodes that - * aren't candidates for reclaim at all, so we must check the - * XFS_IRECLAIMABLE is set first before proceeding to reclaim. - */ - spin_lock(&ip->i_flags_lock); - if (!__xfs_iflags_test(ip, XFS_IRECLAIMABLE) || - __xfs_iflags_test(ip, XFS_IRECLAIM)) { - /* not a reclaim candidate. */ - spin_unlock(&ip->i_flags_lock); - return 1; - } - __xfs_iflags_set(ip, XFS_IRECLAIM); - spin_unlock(&ip->i_flags_lock); - return 0; -} - -/* - * Inodes in different states need to be treated differently, and the return - * value of xfs_iflush is not sufficient to get this right. The following table - * lists the inode states and the reclaim actions necessary for non-blocking - * reclaim: - * - * - * inode state iflush ret required action - * --------------- ---------- --------------- - * bad - reclaim - * shutdown EIO unpin and reclaim - * clean, unpinned 0 reclaim - * stale, unpinned 0 reclaim - * clean, pinned(*) 0 requeue - * stale, pinned EAGAIN requeue - * dirty, delwri ok 0 requeue - * dirty, delwri blocked EAGAIN requeue - * dirty, sync flush 0 reclaim - * - * (*) dgc: I don't think the clean, pinned state is possible but it gets - * handled anyway given the order of checks implemented. - * - * As can be seen from the table, the return value of xfs_iflush() is not - * sufficient to correctly decide the reclaim action here. The checks in - * xfs_iflush() might look like duplicates, but they are not. - * - * Also, because we get the flush lock first, we know that any inode that has - * been flushed delwri has had the flush completed by the time we check that - * the inode is clean. The clean inode check needs to be done before flushing - * the inode delwri otherwise we would loop forever requeuing clean inodes as - * we cannot tell apart a successful delwri flush and a clean inode from the - * return value of xfs_iflush(). - * - * Note that because the inode is flushed delayed write by background - * writeback, the flush lock may already be held here and waiting on it can - * result in very long latencies. Hence for sync reclaims, where we wait on the - * flush lock, the caller should push out delayed write inodes first before - * trying to reclaim them to minimise the amount of time spent waiting. For - * background relaim, we just requeue the inode for the next pass. - * - * Hence the order of actions after gaining the locks should be: - * bad => reclaim - * shutdown => unpin and reclaim - * pinned, delwri => requeue - * pinned, sync => unpin - * stale => reclaim - * clean => reclaim - * dirty, delwri => flush and requeue - * dirty, sync => flush, wait and reclaim - */ -STATIC int -xfs_reclaim_inode( - struct xfs_inode *ip, - struct xfs_perag *pag, - int sync_mode) -{ - int error; - -restart: - error = 0; - xfs_ilock(ip, XFS_ILOCK_EXCL); - if (!xfs_iflock_nowait(ip)) { - if (!(sync_mode & SYNC_WAIT)) - goto out; - xfs_iflock(ip); - } - - if (is_bad_inode(VFS_I(ip))) - goto reclaim; - if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_iunpin_wait(ip); - goto reclaim; - } - if (xfs_ipincount(ip)) { - if (!(sync_mode & SYNC_WAIT)) { - xfs_ifunlock(ip); - goto out; - } - xfs_iunpin_wait(ip); - } - if (xfs_iflags_test(ip, XFS_ISTALE)) - goto reclaim; - if (xfs_inode_clean(ip)) - goto reclaim; - - /* - * Now we have an inode that needs flushing. - * - * We do a nonblocking flush here even if we are doing a SYNC_WAIT - * reclaim as we can deadlock with inode cluster removal. - * xfs_ifree_cluster() can lock the inode buffer before it locks the - * ip->i_lock, and we are doing the exact opposite here. As a result, - * doing a blocking xfs_itobp() to get the cluster buffer will result - * in an ABBA deadlock with xfs_ifree_cluster(). - * - * As xfs_ifree_cluser() must gather all inodes that are active in the - * cache to mark them stale, if we hit this case we don't actually want - * to do IO here - we want the inode marked stale so we can simply - * reclaim it. Hence if we get an EAGAIN error on a SYNC_WAIT flush, - * just unlock the inode, back off and try again. Hopefully the next - * pass through will see the stale flag set on the inode. - */ - error = xfs_iflush(ip, SYNC_TRYLOCK | sync_mode); - if (sync_mode & SYNC_WAIT) { - if (error == EAGAIN) { - xfs_iunlock(ip, XFS_ILOCK_EXCL); - /* backoff longer than in xfs_ifree_cluster */ - delay(2); - goto restart; - } - xfs_iflock(ip); - goto reclaim; - } - - /* - * When we have to flush an inode but don't have SYNC_WAIT set, we - * flush the inode out using a delwri buffer and wait for the next - * call into reclaim to find it in a clean state instead of waiting for - * it now. We also don't return errors here - if the error is transient - * then the next reclaim pass will flush the inode, and if the error - * is permanent then the next sync reclaim will reclaim the inode and - * pass on the error. - */ - if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_warn(ip->i_mount, - "inode 0x%llx background reclaim flush failed with %d", - (long long)ip->i_ino, error); - } -out: - xfs_iflags_clear(ip, XFS_IRECLAIM); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - /* - * We could return EAGAIN here to make reclaim rescan the inode tree in - * a short while. However, this just burns CPU time scanning the tree - * waiting for IO to complete and xfssyncd never goes back to the idle - * state. Instead, return 0 to let the next scheduled background reclaim - * attempt to reclaim the inode again. - */ - return 0; - -reclaim: - xfs_ifunlock(ip); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - XFS_STATS_INC(xs_ig_reclaims); - /* - * Remove the inode from the per-AG radix tree. - * - * Because radix_tree_delete won't complain even if the item was never - * added to the tree assert that it's been there before to catch - * problems with the inode life time early on. - */ - spin_lock(&pag->pag_ici_lock); - if (!radix_tree_delete(&pag->pag_ici_root, - XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino))) - ASSERT(0); - __xfs_inode_clear_reclaim(pag, ip); - spin_unlock(&pag->pag_ici_lock); - - /* - * Here we do an (almost) spurious inode lock in order to coordinate - * with inode cache radix tree lookups. This is because the lookup - * can reference the inodes in the cache without taking references. - * - * We make that OK here by ensuring that we wait until the inode is - * unlocked after the lookup before we go ahead and free it. We get - * both the ilock and the iolock because the code may need to drop the - * ilock one but will still hold the iolock. - */ - xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - xfs_qm_dqdetach(ip); - xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); - - xfs_inode_free(ip); - return error; - -} - -/* - * Walk the AGs and reclaim the inodes in them. Even if the filesystem is - * corrupted, we still want to try to reclaim all the inodes. If we don't, - * then a shut down during filesystem unmount reclaim walk leak all the - * unreclaimed inodes. - */ -int -xfs_reclaim_inodes_ag( - struct xfs_mount *mp, - int flags, - int *nr_to_scan) -{ - struct xfs_perag *pag; - int error = 0; - int last_error = 0; - xfs_agnumber_t ag; - int trylock = flags & SYNC_TRYLOCK; - int skipped; - -restart: - ag = 0; - skipped = 0; - while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) { - unsigned long first_index = 0; - int done = 0; - int nr_found = 0; - - ag = pag->pag_agno + 1; - - if (trylock) { - if (!mutex_trylock(&pag->pag_ici_reclaim_lock)) { - skipped++; - xfs_perag_put(pag); - continue; - } - first_index = pag->pag_ici_reclaim_cursor; - } else - mutex_lock(&pag->pag_ici_reclaim_lock); - - do { - struct xfs_inode *batch[XFS_LOOKUP_BATCH]; - int i; - - rcu_read_lock(); - nr_found = radix_tree_gang_lookup_tag( - &pag->pag_ici_root, - (void **)batch, first_index, - XFS_LOOKUP_BATCH, - XFS_ICI_RECLAIM_TAG); - if (!nr_found) { - done = 1; - rcu_read_unlock(); - break; - } - - /* - * Grab the inodes before we drop the lock. if we found - * nothing, nr == 0 and the loop will be skipped. - */ - for (i = 0; i < nr_found; i++) { - struct xfs_inode *ip = batch[i]; - - if (done || xfs_reclaim_inode_grab(ip, flags)) - batch[i] = NULL; - - /* - * Update the index for the next lookup. Catch - * overflows into the next AG range which can - * occur if we have inodes in the last block of - * the AG and we are currently pointing to the - * last inode. - * - * Because we may see inodes that are from the - * wrong AG due to RCU freeing and - * reallocation, only update the index if it - * lies in this AG. It was a race that lead us - * to see this inode, so another lookup from - * the same index will not find it again. - */ - if (XFS_INO_TO_AGNO(mp, ip->i_ino) != - pag->pag_agno) - continue; - first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); - if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) - done = 1; - } - - /* unlock now we've grabbed the inodes. */ - rcu_read_unlock(); - - for (i = 0; i < nr_found; i++) { - if (!batch[i]) - continue; - error = xfs_reclaim_inode(batch[i], pag, flags); - if (error && last_error != EFSCORRUPTED) - last_error = error; - } - - *nr_to_scan -= XFS_LOOKUP_BATCH; - - cond_resched(); - - } while (nr_found && !done && *nr_to_scan > 0); - - if (trylock && !done) - pag->pag_ici_reclaim_cursor = first_index; - else - pag->pag_ici_reclaim_cursor = 0; - mutex_unlock(&pag->pag_ici_reclaim_lock); - xfs_perag_put(pag); - } - - /* - * if we skipped any AG, and we still have scan count remaining, do - * another pass this time using blocking reclaim semantics (i.e - * waiting on the reclaim locks and ignoring the reclaim cursors). This - * ensure that when we get more reclaimers than AGs we block rather - * than spin trying to execute reclaim. - */ - if (skipped && (flags & SYNC_WAIT) && *nr_to_scan > 0) { - trylock = 0; - goto restart; - } - return XFS_ERROR(last_error); -} - -int -xfs_reclaim_inodes( - xfs_mount_t *mp, - int mode) -{ - int nr_to_scan = INT_MAX; - - return xfs_reclaim_inodes_ag(mp, mode, &nr_to_scan); -} - -/* - * Scan a certain number of inodes for reclaim. - * - * When called we make sure that there is a background (fast) inode reclaim in - * progress, while we will throttle the speed of reclaim via doing synchronous - * reclaim of inodes. That means if we come across dirty inodes, we wait for - * them to be cleaned, which we hope will not be very long due to the - * background walker having already kicked the IO off on those dirty inodes. - */ -void -xfs_reclaim_inodes_nr( - struct xfs_mount *mp, - int nr_to_scan) -{ - /* kick background reclaimer and push the AIL */ - xfs_syncd_queue_reclaim(mp); - xfs_ail_push_all(mp->m_ail); - - xfs_reclaim_inodes_ag(mp, SYNC_TRYLOCK | SYNC_WAIT, &nr_to_scan); -} - -/* - * Return the number of reclaimable inodes in the filesystem for - * the shrinker to determine how much to reclaim. - */ -int -xfs_reclaim_inodes_count( - struct xfs_mount *mp) -{ - struct xfs_perag *pag; - xfs_agnumber_t ag = 0; - int reclaimable = 0; - - while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) { - ag = pag->pag_agno + 1; - reclaimable += pag->pag_ici_reclaimable; - xfs_perag_put(pag); - } - return reclaimable; -} - diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h deleted file mode 100644 index 941202e7ac6e..000000000000 --- a/fs/xfs/linux-2.6/xfs_sync.h +++ /dev/null @@ -1,51 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef XFS_SYNC_H -#define XFS_SYNC_H 1 - -struct xfs_mount; -struct xfs_perag; - -#define SYNC_WAIT 0x0001 /* wait for i/o to complete */ -#define SYNC_TRYLOCK 0x0002 /* only try to lock inodes */ - -extern struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ - -int xfs_syncd_init(struct xfs_mount *mp); -void xfs_syncd_stop(struct xfs_mount *mp); - -int xfs_quiesce_data(struct xfs_mount *mp); -void xfs_quiesce_attr(struct xfs_mount *mp); - -void xfs_flush_inodes(struct xfs_inode *ip); - -int xfs_reclaim_inodes(struct xfs_mount *mp, int mode); -int xfs_reclaim_inodes_count(struct xfs_mount *mp); -void xfs_reclaim_inodes_nr(struct xfs_mount *mp, int nr_to_scan); - -void xfs_inode_set_reclaim_tag(struct xfs_inode *ip); -void __xfs_inode_set_reclaim_tag(struct xfs_perag *pag, struct xfs_inode *ip); -void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag, - struct xfs_inode *ip); - -int xfs_sync_inode_grab(struct xfs_inode *ip); -int xfs_inode_ag_iterator(struct xfs_mount *mp, - int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags), - int flags); - -#endif diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c deleted file mode 100644 index ee2d2adaa438..000000000000 --- a/fs/xfs/linux-2.6/xfs_sysctl.c +++ /dev/null @@ -1,252 +0,0 @@ -/* - * Copyright (c) 2001-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include -#include -#include "xfs_error.h" - -static struct ctl_table_header *xfs_table_header; - -#ifdef CONFIG_PROC_FS -STATIC int -xfs_stats_clear_proc_handler( - ctl_table *ctl, - int write, - void __user *buffer, - size_t *lenp, - loff_t *ppos) -{ - int c, ret, *valp = ctl->data; - __uint32_t vn_active; - - ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); - - if (!ret && write && *valp) { - xfs_notice(NULL, "Clearing xfsstats"); - for_each_possible_cpu(c) { - preempt_disable(); - /* save vn_active, it's a universal truth! */ - vn_active = per_cpu(xfsstats, c).vn_active; - memset(&per_cpu(xfsstats, c), 0, - sizeof(struct xfsstats)); - per_cpu(xfsstats, c).vn_active = vn_active; - preempt_enable(); - } - xfs_stats_clear = 0; - } - - return ret; -} - -STATIC int -xfs_panic_mask_proc_handler( - ctl_table *ctl, - int write, - void __user *buffer, - size_t *lenp, - loff_t *ppos) -{ - int ret, *valp = ctl->data; - - ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); - if (!ret && write) { - xfs_panic_mask = *valp; -#ifdef DEBUG - xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES); -#endif - } - return ret; -} -#endif /* CONFIG_PROC_FS */ - -static ctl_table xfs_table[] = { - { - .procname = "irix_sgid_inherit", - .data = &xfs_params.sgid_inherit.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.sgid_inherit.min, - .extra2 = &xfs_params.sgid_inherit.max - }, - { - .procname = "irix_symlink_mode", - .data = &xfs_params.symlink_mode.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.symlink_mode.min, - .extra2 = &xfs_params.symlink_mode.max - }, - { - .procname = "panic_mask", - .data = &xfs_params.panic_mask.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = xfs_panic_mask_proc_handler, - .extra1 = &xfs_params.panic_mask.min, - .extra2 = &xfs_params.panic_mask.max - }, - - { - .procname = "error_level", - .data = &xfs_params.error_level.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.error_level.min, - .extra2 = &xfs_params.error_level.max - }, - { - .procname = "xfssyncd_centisecs", - .data = &xfs_params.syncd_timer.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.syncd_timer.min, - .extra2 = &xfs_params.syncd_timer.max - }, - { - .procname = "inherit_sync", - .data = &xfs_params.inherit_sync.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.inherit_sync.min, - .extra2 = &xfs_params.inherit_sync.max - }, - { - .procname = "inherit_nodump", - .data = &xfs_params.inherit_nodump.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.inherit_nodump.min, - .extra2 = &xfs_params.inherit_nodump.max - }, - { - .procname = "inherit_noatime", - .data = &xfs_params.inherit_noatim.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.inherit_noatim.min, - .extra2 = &xfs_params.inherit_noatim.max - }, - { - .procname = "xfsbufd_centisecs", - .data = &xfs_params.xfs_buf_timer.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.xfs_buf_timer.min, - .extra2 = &xfs_params.xfs_buf_timer.max - }, - { - .procname = "age_buffer_centisecs", - .data = &xfs_params.xfs_buf_age.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.xfs_buf_age.min, - .extra2 = &xfs_params.xfs_buf_age.max - }, - { - .procname = "inherit_nosymlinks", - .data = &xfs_params.inherit_nosym.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.inherit_nosym.min, - .extra2 = &xfs_params.inherit_nosym.max - }, - { - .procname = "rotorstep", - .data = &xfs_params.rotorstep.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.rotorstep.min, - .extra2 = &xfs_params.rotorstep.max - }, - { - .procname = "inherit_nodefrag", - .data = &xfs_params.inherit_nodfrg.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.inherit_nodfrg.min, - .extra2 = &xfs_params.inherit_nodfrg.max - }, - { - .procname = "filestream_centisecs", - .data = &xfs_params.fstrm_timer.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.fstrm_timer.min, - .extra2 = &xfs_params.fstrm_timer.max, - }, - /* please keep this the last entry */ -#ifdef CONFIG_PROC_FS - { - .procname = "stats_clear", - .data = &xfs_params.stats_clear.val, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = xfs_stats_clear_proc_handler, - .extra1 = &xfs_params.stats_clear.min, - .extra2 = &xfs_params.stats_clear.max - }, -#endif /* CONFIG_PROC_FS */ - - {} -}; - -static ctl_table xfs_dir_table[] = { - { - .procname = "xfs", - .mode = 0555, - .child = xfs_table - }, - {} -}; - -static ctl_table xfs_root_table[] = { - { - .procname = "fs", - .mode = 0555, - .child = xfs_dir_table - }, - {} -}; - -int -xfs_sysctl_register(void) -{ - xfs_table_header = register_sysctl_table(xfs_root_table); - if (!xfs_table_header) - return -ENOMEM; - return 0; -} - -void -xfs_sysctl_unregister(void) -{ - unregister_sysctl_table(xfs_table_header); -} diff --git a/fs/xfs/linux-2.6/xfs_sysctl.h b/fs/xfs/linux-2.6/xfs_sysctl.h deleted file mode 100644 index b9937d450f8e..000000000000 --- a/fs/xfs/linux-2.6/xfs_sysctl.h +++ /dev/null @@ -1,102 +0,0 @@ -/* - * Copyright (c) 2001-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SYSCTL_H__ -#define __XFS_SYSCTL_H__ - -#include - -/* - * Tunable xfs parameters - */ - -typedef struct xfs_sysctl_val { - int min; - int val; - int max; -} xfs_sysctl_val_t; - -typedef struct xfs_param { - xfs_sysctl_val_t sgid_inherit; /* Inherit S_ISGID if process' GID is - * not a member of parent dir GID. */ - xfs_sysctl_val_t symlink_mode; /* Link creat mode affected by umask */ - xfs_sysctl_val_t panic_mask; /* bitmask to cause panic on errors. */ - xfs_sysctl_val_t error_level; /* Degree of reporting for problems */ - xfs_sysctl_val_t syncd_timer; /* Interval between xfssyncd wakeups */ - xfs_sysctl_val_t stats_clear; /* Reset all XFS statistics to zero. */ - xfs_sysctl_val_t inherit_sync; /* Inherit the "sync" inode flag. */ - xfs_sysctl_val_t inherit_nodump;/* Inherit the "nodump" inode flag. */ - xfs_sysctl_val_t inherit_noatim;/* Inherit the "noatime" inode flag. */ - xfs_sysctl_val_t xfs_buf_timer; /* Interval between xfsbufd wakeups. */ - xfs_sysctl_val_t xfs_buf_age; /* Metadata buffer age before flush. */ - xfs_sysctl_val_t inherit_nosym; /* Inherit the "nosymlinks" flag. */ - xfs_sysctl_val_t rotorstep; /* inode32 AG rotoring control knob */ - xfs_sysctl_val_t inherit_nodfrg;/* Inherit the "nodefrag" inode flag. */ - xfs_sysctl_val_t fstrm_timer; /* Filestream dir-AG assoc'n timeout. */ -} xfs_param_t; - -/* - * xfs_error_level: - * - * How much error reporting will be done when internal problems are - * encountered. These problems normally return an EFSCORRUPTED to their - * caller, with no other information reported. - * - * 0 No error reports - * 1 Report EFSCORRUPTED errors that will cause a filesystem shutdown - * 5 Report all EFSCORRUPTED errors (all of the above errors, plus any - * additional errors that are known to not cause shutdowns) - * - * xfs_panic_mask bit 0x8 turns the error reports into panics - */ - -enum { - /* XFS_REFCACHE_SIZE = 1 */ - /* XFS_REFCACHE_PURGE = 2 */ - /* XFS_RESTRICT_CHOWN = 3 */ - XFS_SGID_INHERIT = 4, - XFS_SYMLINK_MODE = 5, - XFS_PANIC_MASK = 6, - XFS_ERRLEVEL = 7, - XFS_SYNCD_TIMER = 8, - /* XFS_PROBE_DMAPI = 9 */ - /* XFS_PROBE_IOOPS = 10 */ - /* XFS_PROBE_QUOTA = 11 */ - XFS_STATS_CLEAR = 12, - XFS_INHERIT_SYNC = 13, - XFS_INHERIT_NODUMP = 14, - XFS_INHERIT_NOATIME = 15, - XFS_BUF_TIMER = 16, - XFS_BUF_AGE = 17, - /* XFS_IO_BYPASS = 18 */ - XFS_INHERIT_NOSYM = 19, - XFS_ROTORSTEP = 20, - XFS_INHERIT_NODFRG = 21, - XFS_FILESTREAM_TIMER = 22, -}; - -extern xfs_param_t xfs_params; - -#ifdef CONFIG_SYSCTL -extern int xfs_sysctl_register(void); -extern void xfs_sysctl_unregister(void); -#else -# define xfs_sysctl_register() (0) -# define xfs_sysctl_unregister() do { } while (0) -#endif /* CONFIG_SYSCTL */ - -#endif /* __XFS_SYSCTL_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c deleted file mode 100644 index 9010ce885e6a..000000000000 --- a/fs/xfs/linux-2.6/xfs_trace.c +++ /dev/null @@ -1,56 +0,0 @@ -/* - * Copyright (c) 2009, Christoph Hellwig - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_types.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_da_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_btree.h" -#include "xfs_mount.h" -#include "xfs_ialloc.h" -#include "xfs_itable.h" -#include "xfs_alloc.h" -#include "xfs_bmap.h" -#include "xfs_attr.h" -#include "xfs_attr_leaf.h" -#include "xfs_log_priv.h" -#include "xfs_buf_item.h" -#include "xfs_quota.h" -#include "xfs_iomap.h" -#include "xfs_aops.h" -#include "xfs_dquot_item.h" -#include "xfs_dquot.h" -#include "xfs_log_recover.h" -#include "xfs_inode_item.h" - -/* - * We include this last to have the helpers above available for the trace - * event implementations. - */ -#define CREATE_TRACE_POINTS -#include "xfs_trace.h" diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h deleted file mode 100644 index 690fc7a7bd72..000000000000 --- a/fs/xfs/linux-2.6/xfs_trace.h +++ /dev/null @@ -1,1746 +0,0 @@ -/* - * Copyright (c) 2009, Christoph Hellwig - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#undef TRACE_SYSTEM -#define TRACE_SYSTEM xfs - -#if !defined(_TRACE_XFS_H) || defined(TRACE_HEADER_MULTI_READ) -#define _TRACE_XFS_H - -#include - -struct xfs_agf; -struct xfs_alloc_arg; -struct xfs_attr_list_context; -struct xfs_buf_log_item; -struct xfs_da_args; -struct xfs_da_node_entry; -struct xfs_dquot; -struct xlog_ticket; -struct log; -struct xlog_recover; -struct xlog_recover_item; -struct xfs_buf_log_format; -struct xfs_inode_log_format; - -DECLARE_EVENT_CLASS(xfs_attr_list_class, - TP_PROTO(struct xfs_attr_list_context *ctx), - TP_ARGS(ctx), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(u32, hashval) - __field(u32, blkno) - __field(u32, offset) - __field(void *, alist) - __field(int, bufsize) - __field(int, count) - __field(int, firstu) - __field(int, dupcnt) - __field(int, flags) - ), - TP_fast_assign( - __entry->dev = VFS_I(ctx->dp)->i_sb->s_dev; - __entry->ino = ctx->dp->i_ino; - __entry->hashval = ctx->cursor->hashval; - __entry->blkno = ctx->cursor->blkno; - __entry->offset = ctx->cursor->offset; - __entry->alist = ctx->alist; - __entry->bufsize = ctx->bufsize; - __entry->count = ctx->count; - __entry->firstu = ctx->firstu; - __entry->flags = ctx->flags; - ), - TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u " - "alist 0x%p size %u count %u firstu %u flags %d %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->hashval, - __entry->blkno, - __entry->offset, - __entry->dupcnt, - __entry->alist, - __entry->bufsize, - __entry->count, - __entry->firstu, - __entry->flags, - __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS) - ) -) - -#define DEFINE_ATTR_LIST_EVENT(name) \ -DEFINE_EVENT(xfs_attr_list_class, name, \ - TP_PROTO(struct xfs_attr_list_context *ctx), \ - TP_ARGS(ctx)) -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf_all); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf_end); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_full); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_add); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_wrong_blk); -DEFINE_ATTR_LIST_EVENT(xfs_attr_list_notfound); - -DECLARE_EVENT_CLASS(xfs_perag_class, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, - unsigned long caller_ip), - TP_ARGS(mp, agno, refcount, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(int, refcount) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->refcount = refcount; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d agno %u refcount %d caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->refcount, - (char *)__entry->caller_ip) -); - -#define DEFINE_PERAG_REF_EVENT(name) \ -DEFINE_EVENT(xfs_perag_class, name, \ - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, \ - unsigned long caller_ip), \ - TP_ARGS(mp, agno, refcount, caller_ip)) -DEFINE_PERAG_REF_EVENT(xfs_perag_get); -DEFINE_PERAG_REF_EVENT(xfs_perag_get_tag); -DEFINE_PERAG_REF_EVENT(xfs_perag_put); -DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim); -DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim); - -TRACE_EVENT(xfs_attr_list_node_descend, - TP_PROTO(struct xfs_attr_list_context *ctx, - struct xfs_da_node_entry *btree), - TP_ARGS(ctx, btree), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(u32, hashval) - __field(u32, blkno) - __field(u32, offset) - __field(void *, alist) - __field(int, bufsize) - __field(int, count) - __field(int, firstu) - __field(int, dupcnt) - __field(int, flags) - __field(u32, bt_hashval) - __field(u32, bt_before) - ), - TP_fast_assign( - __entry->dev = VFS_I(ctx->dp)->i_sb->s_dev; - __entry->ino = ctx->dp->i_ino; - __entry->hashval = ctx->cursor->hashval; - __entry->blkno = ctx->cursor->blkno; - __entry->offset = ctx->cursor->offset; - __entry->alist = ctx->alist; - __entry->bufsize = ctx->bufsize; - __entry->count = ctx->count; - __entry->firstu = ctx->firstu; - __entry->flags = ctx->flags; - __entry->bt_hashval = be32_to_cpu(btree->hashval); - __entry->bt_before = be32_to_cpu(btree->before); - ), - TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u " - "alist 0x%p size %u count %u firstu %u flags %d %s " - "node hashval %u, node before %u", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->hashval, - __entry->blkno, - __entry->offset, - __entry->dupcnt, - __entry->alist, - __entry->bufsize, - __entry->count, - __entry->firstu, - __entry->flags, - __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS), - __entry->bt_hashval, - __entry->bt_before) -); - -TRACE_EVENT(xfs_iext_insert, - TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, - struct xfs_bmbt_irec *r, int state, unsigned long caller_ip), - TP_ARGS(ip, idx, r, state, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_extnum_t, idx) - __field(xfs_fileoff_t, startoff) - __field(xfs_fsblock_t, startblock) - __field(xfs_filblks_t, blockcount) - __field(xfs_exntst_t, state) - __field(int, bmap_state) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->idx = idx; - __entry->startoff = r->br_startoff; - __entry->startblock = r->br_startblock; - __entry->blockcount = r->br_blockcount; - __entry->state = r->br_state; - __entry->bmap_state = state; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " - "offset %lld block %lld count %lld flag %d caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), - (long)__entry->idx, - __entry->startoff, - (__int64_t)__entry->startblock, - __entry->blockcount, - __entry->state, - (char *)__entry->caller_ip) -); - -DECLARE_EVENT_CLASS(xfs_bmap_class, - TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, - unsigned long caller_ip), - TP_ARGS(ip, idx, state, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_extnum_t, idx) - __field(xfs_fileoff_t, startoff) - __field(xfs_fsblock_t, startblock) - __field(xfs_filblks_t, blockcount) - __field(xfs_exntst_t, state) - __field(int, bmap_state) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - struct xfs_ifork *ifp = (state & BMAP_ATTRFORK) ? - ip->i_afp : &ip->i_df; - struct xfs_bmbt_irec r; - - xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &r); - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->idx = idx; - __entry->startoff = r.br_startoff; - __entry->startblock = r.br_startblock; - __entry->blockcount = r.br_blockcount; - __entry->state = r.br_state; - __entry->bmap_state = state; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " - "offset %lld block %lld count %lld flag %d caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), - (long)__entry->idx, - __entry->startoff, - (__int64_t)__entry->startblock, - __entry->blockcount, - __entry->state, - (char *)__entry->caller_ip) -) - -#define DEFINE_BMAP_EVENT(name) \ -DEFINE_EVENT(xfs_bmap_class, name, \ - TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, \ - unsigned long caller_ip), \ - TP_ARGS(ip, idx, state, caller_ip)) -DEFINE_BMAP_EVENT(xfs_iext_remove); -DEFINE_BMAP_EVENT(xfs_bmap_pre_update); -DEFINE_BMAP_EVENT(xfs_bmap_post_update); -DEFINE_BMAP_EVENT(xfs_extlist); - -DECLARE_EVENT_CLASS(xfs_buf_class, - TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip), - TP_ARGS(bp, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_daddr_t, bno) - __field(size_t, buffer_length) - __field(int, hold) - __field(int, pincount) - __field(unsigned, lockval) - __field(unsigned, flags) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = bp->b_target->bt_dev; - __entry->bno = bp->b_bn; - __entry->buffer_length = bp->b_buffer_length; - __entry->hold = atomic_read(&bp->b_hold); - __entry->pincount = atomic_read(&bp->b_pin_count); - __entry->lockval = bp->b_sema.count; - __entry->flags = bp->b_flags; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " - "lock %d flags %s caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long long)__entry->bno, - __entry->buffer_length, - __entry->hold, - __entry->pincount, - __entry->lockval, - __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), - (void *)__entry->caller_ip) -) - -#define DEFINE_BUF_EVENT(name) \ -DEFINE_EVENT(xfs_buf_class, name, \ - TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip), \ - TP_ARGS(bp, caller_ip)) -DEFINE_BUF_EVENT(xfs_buf_init); -DEFINE_BUF_EVENT(xfs_buf_free); -DEFINE_BUF_EVENT(xfs_buf_hold); -DEFINE_BUF_EVENT(xfs_buf_rele); -DEFINE_BUF_EVENT(xfs_buf_iodone); -DEFINE_BUF_EVENT(xfs_buf_iorequest); -DEFINE_BUF_EVENT(xfs_buf_bawrite); -DEFINE_BUF_EVENT(xfs_buf_bdwrite); -DEFINE_BUF_EVENT(xfs_buf_lock); -DEFINE_BUF_EVENT(xfs_buf_lock_done); -DEFINE_BUF_EVENT(xfs_buf_trylock); -DEFINE_BUF_EVENT(xfs_buf_unlock); -DEFINE_BUF_EVENT(xfs_buf_iowait); -DEFINE_BUF_EVENT(xfs_buf_iowait_done); -DEFINE_BUF_EVENT(xfs_buf_delwri_queue); -DEFINE_BUF_EVENT(xfs_buf_delwri_dequeue); -DEFINE_BUF_EVENT(xfs_buf_delwri_split); -DEFINE_BUF_EVENT(xfs_buf_get_uncached); -DEFINE_BUF_EVENT(xfs_bdstrat_shut); -DEFINE_BUF_EVENT(xfs_buf_item_relse); -DEFINE_BUF_EVENT(xfs_buf_item_iodone); -DEFINE_BUF_EVENT(xfs_buf_item_iodone_async); -DEFINE_BUF_EVENT(xfs_buf_error_relse); -DEFINE_BUF_EVENT(xfs_trans_read_buf_io); -DEFINE_BUF_EVENT(xfs_trans_read_buf_shut); - -/* not really buffer traces, but the buf provides useful information */ -DEFINE_BUF_EVENT(xfs_btree_corrupt); -DEFINE_BUF_EVENT(xfs_da_btree_corrupt); -DEFINE_BUF_EVENT(xfs_reset_dqcounts); -DEFINE_BUF_EVENT(xfs_inode_item_push); - -/* pass flags explicitly */ -DECLARE_EVENT_CLASS(xfs_buf_flags_class, - TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip), - TP_ARGS(bp, flags, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_daddr_t, bno) - __field(size_t, buffer_length) - __field(int, hold) - __field(int, pincount) - __field(unsigned, lockval) - __field(unsigned, flags) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = bp->b_target->bt_dev; - __entry->bno = bp->b_bn; - __entry->buffer_length = bp->b_buffer_length; - __entry->flags = flags; - __entry->hold = atomic_read(&bp->b_hold); - __entry->pincount = atomic_read(&bp->b_pin_count); - __entry->lockval = bp->b_sema.count; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " - "lock %d flags %s caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long long)__entry->bno, - __entry->buffer_length, - __entry->hold, - __entry->pincount, - __entry->lockval, - __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), - (void *)__entry->caller_ip) -) - -#define DEFINE_BUF_FLAGS_EVENT(name) \ -DEFINE_EVENT(xfs_buf_flags_class, name, \ - TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip), \ - TP_ARGS(bp, flags, caller_ip)) -DEFINE_BUF_FLAGS_EVENT(xfs_buf_find); -DEFINE_BUF_FLAGS_EVENT(xfs_buf_get); -DEFINE_BUF_FLAGS_EVENT(xfs_buf_read); - -TRACE_EVENT(xfs_buf_ioerror, - TP_PROTO(struct xfs_buf *bp, int error, unsigned long caller_ip), - TP_ARGS(bp, error, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_daddr_t, bno) - __field(size_t, buffer_length) - __field(unsigned, flags) - __field(int, hold) - __field(int, pincount) - __field(unsigned, lockval) - __field(int, error) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = bp->b_target->bt_dev; - __entry->bno = bp->b_bn; - __entry->buffer_length = bp->b_buffer_length; - __entry->hold = atomic_read(&bp->b_hold); - __entry->pincount = atomic_read(&bp->b_pin_count); - __entry->lockval = bp->b_sema.count; - __entry->error = error; - __entry->flags = bp->b_flags; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " - "lock %d error %d flags %s caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long long)__entry->bno, - __entry->buffer_length, - __entry->hold, - __entry->pincount, - __entry->lockval, - __entry->error, - __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), - (void *)__entry->caller_ip) -); - -DECLARE_EVENT_CLASS(xfs_buf_item_class, - TP_PROTO(struct xfs_buf_log_item *bip), - TP_ARGS(bip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_daddr_t, buf_bno) - __field(size_t, buf_len) - __field(int, buf_hold) - __field(int, buf_pincount) - __field(int, buf_lockval) - __field(unsigned, buf_flags) - __field(unsigned, bli_recur) - __field(int, bli_refcount) - __field(unsigned, bli_flags) - __field(void *, li_desc) - __field(unsigned, li_flags) - ), - TP_fast_assign( - __entry->dev = bip->bli_buf->b_target->bt_dev; - __entry->bli_flags = bip->bli_flags; - __entry->bli_recur = bip->bli_recur; - __entry->bli_refcount = atomic_read(&bip->bli_refcount); - __entry->buf_bno = bip->bli_buf->b_bn; - __entry->buf_len = bip->bli_buf->b_buffer_length; - __entry->buf_flags = bip->bli_buf->b_flags; - __entry->buf_hold = atomic_read(&bip->bli_buf->b_hold); - __entry->buf_pincount = atomic_read(&bip->bli_buf->b_pin_count); - __entry->buf_lockval = bip->bli_buf->b_sema.count; - __entry->li_desc = bip->bli_item.li_desc; - __entry->li_flags = bip->bli_item.li_flags; - ), - TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " - "lock %d flags %s recur %d refcount %d bliflags %s " - "lidesc 0x%p liflags %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - (unsigned long long)__entry->buf_bno, - __entry->buf_len, - __entry->buf_hold, - __entry->buf_pincount, - __entry->buf_lockval, - __print_flags(__entry->buf_flags, "|", XFS_BUF_FLAGS), - __entry->bli_recur, - __entry->bli_refcount, - __print_flags(__entry->bli_flags, "|", XFS_BLI_FLAGS), - __entry->li_desc, - __print_flags(__entry->li_flags, "|", XFS_LI_FLAGS)) -) - -#define DEFINE_BUF_ITEM_EVENT(name) \ -DEFINE_EVENT(xfs_buf_item_class, name, \ - TP_PROTO(struct xfs_buf_log_item *bip), \ - TP_ARGS(bip)) -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size_stale); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format_stale); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pin); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin_stale); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_trylock); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock_stale); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_committed); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_push); -DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pushbuf); -DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf); -DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf_recur); -DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb); -DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb_recur); -DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf); -DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf_recur); -DEFINE_BUF_ITEM_EVENT(xfs_trans_log_buf); -DEFINE_BUF_ITEM_EVENT(xfs_trans_brelse); -DEFINE_BUF_ITEM_EVENT(xfs_trans_bjoin); -DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold); -DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release); -DEFINE_BUF_ITEM_EVENT(xfs_trans_binval); - -DECLARE_EVENT_CLASS(xfs_lock_class, - TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, - unsigned long caller_ip), - TP_ARGS(ip, lock_flags, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(int, lock_flags) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->lock_flags = lock_flags; - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d ino 0x%llx flags %s caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_flags(__entry->lock_flags, "|", XFS_LOCK_FLAGS), - (void *)__entry->caller_ip) -) - -#define DEFINE_LOCK_EVENT(name) \ -DEFINE_EVENT(xfs_lock_class, name, \ - TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, \ - unsigned long caller_ip), \ - TP_ARGS(ip, lock_flags, caller_ip)) -DEFINE_LOCK_EVENT(xfs_ilock); -DEFINE_LOCK_EVENT(xfs_ilock_nowait); -DEFINE_LOCK_EVENT(xfs_ilock_demote); -DEFINE_LOCK_EVENT(xfs_iunlock); - -DECLARE_EVENT_CLASS(xfs_inode_class, - TP_PROTO(struct xfs_inode *ip), - TP_ARGS(ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - ), - TP_printk("dev %d:%d ino 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino) -) - -#define DEFINE_INODE_EVENT(name) \ -DEFINE_EVENT(xfs_inode_class, name, \ - TP_PROTO(struct xfs_inode *ip), \ - TP_ARGS(ip)) -DEFINE_INODE_EVENT(xfs_iget_skip); -DEFINE_INODE_EVENT(xfs_iget_reclaim); -DEFINE_INODE_EVENT(xfs_iget_reclaim_fail); -DEFINE_INODE_EVENT(xfs_iget_hit); -DEFINE_INODE_EVENT(xfs_iget_miss); - -DEFINE_INODE_EVENT(xfs_getattr); -DEFINE_INODE_EVENT(xfs_setattr); -DEFINE_INODE_EVENT(xfs_readlink); -DEFINE_INODE_EVENT(xfs_alloc_file_space); -DEFINE_INODE_EVENT(xfs_free_file_space); -DEFINE_INODE_EVENT(xfs_readdir); -#ifdef CONFIG_XFS_POSIX_ACL -DEFINE_INODE_EVENT(xfs_get_acl); -#endif -DEFINE_INODE_EVENT(xfs_vm_bmap); -DEFINE_INODE_EVENT(xfs_file_ioctl); -DEFINE_INODE_EVENT(xfs_file_compat_ioctl); -DEFINE_INODE_EVENT(xfs_ioctl_setattr); -DEFINE_INODE_EVENT(xfs_file_fsync); -DEFINE_INODE_EVENT(xfs_destroy_inode); -DEFINE_INODE_EVENT(xfs_write_inode); -DEFINE_INODE_EVENT(xfs_evict_inode); - -DEFINE_INODE_EVENT(xfs_dquot_dqalloc); -DEFINE_INODE_EVENT(xfs_dquot_dqdetach); - -DECLARE_EVENT_CLASS(xfs_iref_class, - TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), - TP_ARGS(ip, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(int, count) - __field(int, pincount) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->count = atomic_read(&VFS_I(ip)->i_count); - __entry->pincount = atomic_read(&ip->i_pincount); - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d ino 0x%llx count %d pincount %d caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->count, - __entry->pincount, - (char *)__entry->caller_ip) -) - -#define DEFINE_IREF_EVENT(name) \ -DEFINE_EVENT(xfs_iref_class, name, \ - TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), \ - TP_ARGS(ip, caller_ip)) -DEFINE_IREF_EVENT(xfs_ihold); -DEFINE_IREF_EVENT(xfs_irele); -DEFINE_IREF_EVENT(xfs_inode_pin); -DEFINE_IREF_EVENT(xfs_inode_unpin); -DEFINE_IREF_EVENT(xfs_inode_unpin_nowait); - -DECLARE_EVENT_CLASS(xfs_namespace_class, - TP_PROTO(struct xfs_inode *dp, struct xfs_name *name), - TP_ARGS(dp, name), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, dp_ino) - __dynamic_array(char, name, name->len) - ), - TP_fast_assign( - __entry->dev = VFS_I(dp)->i_sb->s_dev; - __entry->dp_ino = dp->i_ino; - memcpy(__get_str(name), name->name, name->len); - ), - TP_printk("dev %d:%d dp ino 0x%llx name %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->dp_ino, - __get_str(name)) -) - -#define DEFINE_NAMESPACE_EVENT(name) \ -DEFINE_EVENT(xfs_namespace_class, name, \ - TP_PROTO(struct xfs_inode *dp, struct xfs_name *name), \ - TP_ARGS(dp, name)) -DEFINE_NAMESPACE_EVENT(xfs_remove); -DEFINE_NAMESPACE_EVENT(xfs_link); -DEFINE_NAMESPACE_EVENT(xfs_lookup); -DEFINE_NAMESPACE_EVENT(xfs_create); -DEFINE_NAMESPACE_EVENT(xfs_symlink); - -TRACE_EVENT(xfs_rename, - TP_PROTO(struct xfs_inode *src_dp, struct xfs_inode *target_dp, - struct xfs_name *src_name, struct xfs_name *target_name), - TP_ARGS(src_dp, target_dp, src_name, target_name), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, src_dp_ino) - __field(xfs_ino_t, target_dp_ino) - __dynamic_array(char, src_name, src_name->len) - __dynamic_array(char, target_name, target_name->len) - ), - TP_fast_assign( - __entry->dev = VFS_I(src_dp)->i_sb->s_dev; - __entry->src_dp_ino = src_dp->i_ino; - __entry->target_dp_ino = target_dp->i_ino; - memcpy(__get_str(src_name), src_name->name, src_name->len); - memcpy(__get_str(target_name), target_name->name, target_name->len); - ), - TP_printk("dev %d:%d src dp ino 0x%llx target dp ino 0x%llx" - " src name %s target name %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->src_dp_ino, - __entry->target_dp_ino, - __get_str(src_name), - __get_str(target_name)) -) - -DECLARE_EVENT_CLASS(xfs_dquot_class, - TP_PROTO(struct xfs_dquot *dqp), - TP_ARGS(dqp), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(u32, id) - __field(unsigned, flags) - __field(unsigned, nrefs) - __field(unsigned long long, res_bcount) - __field(unsigned long long, bcount) - __field(unsigned long long, icount) - __field(unsigned long long, blk_hardlimit) - __field(unsigned long long, blk_softlimit) - __field(unsigned long long, ino_hardlimit) - __field(unsigned long long, ino_softlimit) - ), \ - TP_fast_assign( - __entry->dev = dqp->q_mount->m_super->s_dev; - __entry->id = be32_to_cpu(dqp->q_core.d_id); - __entry->flags = dqp->dq_flags; - __entry->nrefs = dqp->q_nrefs; - __entry->res_bcount = dqp->q_res_bcount; - __entry->bcount = be64_to_cpu(dqp->q_core.d_bcount); - __entry->icount = be64_to_cpu(dqp->q_core.d_icount); - __entry->blk_hardlimit = - be64_to_cpu(dqp->q_core.d_blk_hardlimit); - __entry->blk_softlimit = - be64_to_cpu(dqp->q_core.d_blk_softlimit); - __entry->ino_hardlimit = - be64_to_cpu(dqp->q_core.d_ino_hardlimit); - __entry->ino_softlimit = - be64_to_cpu(dqp->q_core.d_ino_softlimit); - ), - TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx " - "bcnt 0x%llx bhardlimit 0x%llx bsoftlimit 0x%llx " - "icnt 0x%llx ihardlimit 0x%llx isoftlimit 0x%llx]", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->id, - __print_flags(__entry->flags, "|", XFS_DQ_FLAGS), - __entry->nrefs, - __entry->res_bcount, - __entry->bcount, - __entry->blk_hardlimit, - __entry->blk_softlimit, - __entry->icount, - __entry->ino_hardlimit, - __entry->ino_softlimit) -) - -#define DEFINE_DQUOT_EVENT(name) \ -DEFINE_EVENT(xfs_dquot_class, name, \ - TP_PROTO(struct xfs_dquot *dqp), \ - TP_ARGS(dqp)) -DEFINE_DQUOT_EVENT(xfs_dqadjust); -DEFINE_DQUOT_EVENT(xfs_dqreclaim_want); -DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty); -DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink); -DEFINE_DQUOT_EVENT(xfs_dqattach_found); -DEFINE_DQUOT_EVENT(xfs_dqattach_get); -DEFINE_DQUOT_EVENT(xfs_dqinit); -DEFINE_DQUOT_EVENT(xfs_dqreuse); -DEFINE_DQUOT_EVENT(xfs_dqalloc); -DEFINE_DQUOT_EVENT(xfs_dqtobp_read); -DEFINE_DQUOT_EVENT(xfs_dqread); -DEFINE_DQUOT_EVENT(xfs_dqread_fail); -DEFINE_DQUOT_EVENT(xfs_dqlookup_found); -DEFINE_DQUOT_EVENT(xfs_dqlookup_want); -DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist); -DEFINE_DQUOT_EVENT(xfs_dqlookup_done); -DEFINE_DQUOT_EVENT(xfs_dqget_hit); -DEFINE_DQUOT_EVENT(xfs_dqget_miss); -DEFINE_DQUOT_EVENT(xfs_dqput); -DEFINE_DQUOT_EVENT(xfs_dqput_wait); -DEFINE_DQUOT_EVENT(xfs_dqput_free); -DEFINE_DQUOT_EVENT(xfs_dqrele); -DEFINE_DQUOT_EVENT(xfs_dqflush); -DEFINE_DQUOT_EVENT(xfs_dqflush_force); -DEFINE_DQUOT_EVENT(xfs_dqflush_done); - -DECLARE_EVENT_CLASS(xfs_loggrant_class, - TP_PROTO(struct log *log, struct xlog_ticket *tic), - TP_ARGS(log, tic), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(unsigned, trans_type) - __field(char, ocnt) - __field(char, cnt) - __field(int, curr_res) - __field(int, unit_res) - __field(unsigned int, flags) - __field(int, reserveq) - __field(int, writeq) - __field(int, grant_reserve_cycle) - __field(int, grant_reserve_bytes) - __field(int, grant_write_cycle) - __field(int, grant_write_bytes) - __field(int, curr_cycle) - __field(int, curr_block) - __field(xfs_lsn_t, tail_lsn) - ), - TP_fast_assign( - __entry->dev = log->l_mp->m_super->s_dev; - __entry->trans_type = tic->t_trans_type; - __entry->ocnt = tic->t_ocnt; - __entry->cnt = tic->t_cnt; - __entry->curr_res = tic->t_curr_res; - __entry->unit_res = tic->t_unit_res; - __entry->flags = tic->t_flags; - __entry->reserveq = list_empty(&log->l_reserveq); - __entry->writeq = list_empty(&log->l_writeq); - xlog_crack_grant_head(&log->l_grant_reserve_head, - &__entry->grant_reserve_cycle, - &__entry->grant_reserve_bytes); - xlog_crack_grant_head(&log->l_grant_write_head, - &__entry->grant_write_cycle, - &__entry->grant_write_bytes); - __entry->curr_cycle = log->l_curr_cycle; - __entry->curr_block = log->l_curr_block; - __entry->tail_lsn = atomic64_read(&log->l_tail_lsn); - ), - TP_printk("dev %d:%d type %s t_ocnt %u t_cnt %u t_curr_res %u " - "t_unit_res %u t_flags %s reserveq %s " - "writeq %s grant_reserve_cycle %d " - "grant_reserve_bytes %d grant_write_cycle %d " - "grant_write_bytes %d curr_cycle %d curr_block %d " - "tail_cycle %d tail_block %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __print_symbolic(__entry->trans_type, XFS_TRANS_TYPES), - __entry->ocnt, - __entry->cnt, - __entry->curr_res, - __entry->unit_res, - __print_flags(__entry->flags, "|", XLOG_TIC_FLAGS), - __entry->reserveq ? "empty" : "active", - __entry->writeq ? "empty" : "active", - __entry->grant_reserve_cycle, - __entry->grant_reserve_bytes, - __entry->grant_write_cycle, - __entry->grant_write_bytes, - __entry->curr_cycle, - __entry->curr_block, - CYCLE_LSN(__entry->tail_lsn), - BLOCK_LSN(__entry->tail_lsn) - ) -) - -#define DEFINE_LOGGRANT_EVENT(name) \ -DEFINE_EVENT(xfs_loggrant_class, name, \ - TP_PROTO(struct log *log, struct xlog_ticket *tic), \ - TP_ARGS(log, tic)) -DEFINE_LOGGRANT_EVENT(xfs_log_done_nonperm); -DEFINE_LOGGRANT_EVENT(xfs_log_done_perm); -DEFINE_LOGGRANT_EVENT(xfs_log_reserve); -DEFINE_LOGGRANT_EVENT(xfs_log_umount_write); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_error); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep1); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake1); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep2); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake2); -DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake_up); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep1); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake1); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep2); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake2); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake_up); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit); -DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_sub); -DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_enter); -DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_exit); -DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_sub); - -DECLARE_EVENT_CLASS(xfs_file_class, - TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), - TP_ARGS(ip, count, offset, flags), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_fsize_t, size) - __field(xfs_fsize_t, new_size) - __field(loff_t, offset) - __field(size_t, count) - __field(int, flags) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->new_size = ip->i_new_size; - __entry->offset = offset; - __entry->count = count; - __entry->flags = flags; - ), - TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " - "offset 0x%llx count 0x%zx ioflags %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->new_size, - __entry->offset, - __entry->count, - __print_flags(__entry->flags, "|", XFS_IO_FLAGS)) -) - -#define DEFINE_RW_EVENT(name) \ -DEFINE_EVENT(xfs_file_class, name, \ - TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), \ - TP_ARGS(ip, count, offset, flags)) -DEFINE_RW_EVENT(xfs_file_read); -DEFINE_RW_EVENT(xfs_file_buffered_write); -DEFINE_RW_EVENT(xfs_file_direct_write); -DEFINE_RW_EVENT(xfs_file_splice_read); -DEFINE_RW_EVENT(xfs_file_splice_write); - -DECLARE_EVENT_CLASS(xfs_page_class, - TP_PROTO(struct inode *inode, struct page *page, unsigned long off), - TP_ARGS(inode, page, off), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(pgoff_t, pgoff) - __field(loff_t, size) - __field(unsigned long, offset) - __field(int, delalloc) - __field(int, unwritten) - ), - TP_fast_assign( - int delalloc = -1, unwritten = -1; - - if (page_has_buffers(page)) - xfs_count_page_state(page, &delalloc, &unwritten); - __entry->dev = inode->i_sb->s_dev; - __entry->ino = XFS_I(inode)->i_ino; - __entry->pgoff = page_offset(page); - __entry->size = i_size_read(inode); - __entry->offset = off; - __entry->delalloc = delalloc; - __entry->unwritten = unwritten; - ), - TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx " - "delalloc %d unwritten %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->pgoff, - __entry->size, - __entry->offset, - __entry->delalloc, - __entry->unwritten) -) - -#define DEFINE_PAGE_EVENT(name) \ -DEFINE_EVENT(xfs_page_class, name, \ - TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \ - TP_ARGS(inode, page, off)) -DEFINE_PAGE_EVENT(xfs_writepage); -DEFINE_PAGE_EVENT(xfs_releasepage); -DEFINE_PAGE_EVENT(xfs_invalidatepage); - -DECLARE_EVENT_CLASS(xfs_imap_class, - TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, - int type, struct xfs_bmbt_irec *irec), - TP_ARGS(ip, offset, count, type, irec), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(loff_t, size) - __field(loff_t, new_size) - __field(loff_t, offset) - __field(size_t, count) - __field(int, type) - __field(xfs_fileoff_t, startoff) - __field(xfs_fsblock_t, startblock) - __field(xfs_filblks_t, blockcount) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->new_size = ip->i_new_size; - __entry->offset = offset; - __entry->count = count; - __entry->type = type; - __entry->startoff = irec ? irec->br_startoff : 0; - __entry->startblock = irec ? irec->br_startblock : 0; - __entry->blockcount = irec ? irec->br_blockcount : 0; - ), - TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " - "offset 0x%llx count %zd type %s " - "startoff 0x%llx startblock %lld blockcount 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->new_size, - __entry->offset, - __entry->count, - __print_symbolic(__entry->type, XFS_IO_TYPES), - __entry->startoff, - (__int64_t)__entry->startblock, - __entry->blockcount) -) - -#define DEFINE_IOMAP_EVENT(name) \ -DEFINE_EVENT(xfs_imap_class, name, \ - TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, \ - int type, struct xfs_bmbt_irec *irec), \ - TP_ARGS(ip, offset, count, type, irec)) -DEFINE_IOMAP_EVENT(xfs_map_blocks_found); -DEFINE_IOMAP_EVENT(xfs_map_blocks_alloc); -DEFINE_IOMAP_EVENT(xfs_get_blocks_found); -DEFINE_IOMAP_EVENT(xfs_get_blocks_alloc); - -DECLARE_EVENT_CLASS(xfs_simple_io_class, - TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), - TP_ARGS(ip, offset, count), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(loff_t, isize) - __field(loff_t, disize) - __field(loff_t, new_size) - __field(loff_t, offset) - __field(size_t, count) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->isize = ip->i_size; - __entry->disize = ip->i_d.di_size; - __entry->new_size = ip->i_new_size; - __entry->offset = offset; - __entry->count = count; - ), - TP_printk("dev %d:%d ino 0x%llx isize 0x%llx disize 0x%llx new_size 0x%llx " - "offset 0x%llx count %zd", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->isize, - __entry->disize, - __entry->new_size, - __entry->offset, - __entry->count) -); - -#define DEFINE_SIMPLE_IO_EVENT(name) \ -DEFINE_EVENT(xfs_simple_io_class, name, \ - TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), \ - TP_ARGS(ip, offset, count)) -DEFINE_SIMPLE_IO_EVENT(xfs_delalloc_enospc); -DEFINE_SIMPLE_IO_EVENT(xfs_unwritten_convert); -DEFINE_SIMPLE_IO_EVENT(xfs_get_blocks_notfound); -DEFINE_SIMPLE_IO_EVENT(xfs_setfilesize); - -DECLARE_EVENT_CLASS(xfs_itrunc_class, - TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), - TP_ARGS(ip, new_size), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_fsize_t, size) - __field(xfs_fsize_t, new_size) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->new_size = new_size; - ), - TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->new_size) -) - -#define DEFINE_ITRUNC_EVENT(name) \ -DEFINE_EVENT(xfs_itrunc_class, name, \ - TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), \ - TP_ARGS(ip, new_size)) -DEFINE_ITRUNC_EVENT(xfs_itruncate_data_start); -DEFINE_ITRUNC_EVENT(xfs_itruncate_data_end); - -TRACE_EVENT(xfs_pagecache_inval, - TP_PROTO(struct xfs_inode *ip, xfs_off_t start, xfs_off_t finish), - TP_ARGS(ip, start, finish), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_fsize_t, size) - __field(xfs_off_t, start) - __field(xfs_off_t, finish) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->start = start; - __entry->finish = finish; - ), - TP_printk("dev %d:%d ino 0x%llx size 0x%llx start 0x%llx finish 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->start, - __entry->finish) -); - -TRACE_EVENT(xfs_bunmap, - TP_PROTO(struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len, - int flags, unsigned long caller_ip), - TP_ARGS(ip, bno, len, flags, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(xfs_fsize_t, size) - __field(xfs_fileoff_t, bno) - __field(xfs_filblks_t, len) - __field(unsigned long, caller_ip) - __field(int, flags) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->bno = bno; - __entry->len = len; - __entry->caller_ip = caller_ip; - __entry->flags = flags; - ), - TP_printk("dev %d:%d ino 0x%llx size 0x%llx bno 0x%llx len 0x%llx" - "flags %s caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->bno, - __entry->len, - __print_flags(__entry->flags, "|", XFS_BMAPI_FLAGS), - (void *)__entry->caller_ip) - -); - -DECLARE_EVENT_CLASS(xfs_busy_class, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno, xfs_extlen_t len), - TP_ARGS(mp, agno, agbno, len), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, len) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->agbno = agbno; - __entry->len = len; - ), - TP_printk("dev %d:%d agno %u agbno %u len %u", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->agbno, - __entry->len) -); -#define DEFINE_BUSY_EVENT(name) \ -DEFINE_EVENT(xfs_busy_class, name, \ - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ - xfs_agblock_t agbno, xfs_extlen_t len), \ - TP_ARGS(mp, agno, agbno, len)) -DEFINE_BUSY_EVENT(xfs_alloc_busy); -DEFINE_BUSY_EVENT(xfs_alloc_busy_enomem); -DEFINE_BUSY_EVENT(xfs_alloc_busy_force); -DEFINE_BUSY_EVENT(xfs_alloc_busy_reuse); -DEFINE_BUSY_EVENT(xfs_alloc_busy_clear); - -TRACE_EVENT(xfs_alloc_busy_trim, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno, xfs_extlen_t len, - xfs_agblock_t tbno, xfs_extlen_t tlen), - TP_ARGS(mp, agno, agbno, len, tbno, tlen), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, len) - __field(xfs_agblock_t, tbno) - __field(xfs_extlen_t, tlen) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->agbno = agbno; - __entry->len = len; - __entry->tbno = tbno; - __entry->tlen = tlen; - ), - TP_printk("dev %d:%d agno %u agbno %u len %u tbno %u tlen %u", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->agbno, - __entry->len, - __entry->tbno, - __entry->tlen) -); - -TRACE_EVENT(xfs_trans_commit_lsn, - TP_PROTO(struct xfs_trans *trans), - TP_ARGS(trans), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(struct xfs_trans *, tp) - __field(xfs_lsn_t, lsn) - ), - TP_fast_assign( - __entry->dev = trans->t_mountp->m_super->s_dev; - __entry->tp = trans; - __entry->lsn = trans->t_commit_lsn; - ), - TP_printk("dev %d:%d trans 0x%p commit_lsn 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->tp, - __entry->lsn) -); - -TRACE_EVENT(xfs_agf, - TP_PROTO(struct xfs_mount *mp, struct xfs_agf *agf, int flags, - unsigned long caller_ip), - TP_ARGS(mp, agf, flags, caller_ip), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(int, flags) - __field(__u32, length) - __field(__u32, bno_root) - __field(__u32, cnt_root) - __field(__u32, bno_level) - __field(__u32, cnt_level) - __field(__u32, flfirst) - __field(__u32, fllast) - __field(__u32, flcount) - __field(__u32, freeblks) - __field(__u32, longest) - __field(unsigned long, caller_ip) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = be32_to_cpu(agf->agf_seqno), - __entry->flags = flags; - __entry->length = be32_to_cpu(agf->agf_length), - __entry->bno_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]), - __entry->cnt_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]), - __entry->bno_level = - be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]), - __entry->cnt_level = - be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]), - __entry->flfirst = be32_to_cpu(agf->agf_flfirst), - __entry->fllast = be32_to_cpu(agf->agf_fllast), - __entry->flcount = be32_to_cpu(agf->agf_flcount), - __entry->freeblks = be32_to_cpu(agf->agf_freeblks), - __entry->longest = be32_to_cpu(agf->agf_longest); - __entry->caller_ip = caller_ip; - ), - TP_printk("dev %d:%d agno %u flags %s length %u roots b %u c %u " - "levels b %u c %u flfirst %u fllast %u flcount %u " - "freeblks %u longest %u caller %pf", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __print_flags(__entry->flags, "|", XFS_AGF_FLAGS), - __entry->length, - __entry->bno_root, - __entry->cnt_root, - __entry->bno_level, - __entry->cnt_level, - __entry->flfirst, - __entry->fllast, - __entry->flcount, - __entry->freeblks, - __entry->longest, - (void *)__entry->caller_ip) -); - -TRACE_EVENT(xfs_free_extent, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, - xfs_extlen_t len, bool isfl, int haveleft, int haveright), - TP_ARGS(mp, agno, agbno, len, isfl, haveleft, haveright), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, len) - __field(int, isfl) - __field(int, haveleft) - __field(int, haveright) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->agbno = agbno; - __entry->len = len; - __entry->isfl = isfl; - __entry->haveleft = haveleft; - __entry->haveright = haveright; - ), - TP_printk("dev %d:%d agno %u agbno %u len %u isfl %d %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->agbno, - __entry->len, - __entry->isfl, - __entry->haveleft ? - (__entry->haveright ? "both" : "left") : - (__entry->haveright ? "right" : "none")) - -); - -DECLARE_EVENT_CLASS(xfs_alloc_class, - TP_PROTO(struct xfs_alloc_arg *args), - TP_ARGS(args), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, minlen) - __field(xfs_extlen_t, maxlen) - __field(xfs_extlen_t, mod) - __field(xfs_extlen_t, prod) - __field(xfs_extlen_t, minleft) - __field(xfs_extlen_t, total) - __field(xfs_extlen_t, alignment) - __field(xfs_extlen_t, minalignslop) - __field(xfs_extlen_t, len) - __field(short, type) - __field(short, otype) - __field(char, wasdel) - __field(char, wasfromfl) - __field(char, isfl) - __field(char, userdata) - __field(xfs_fsblock_t, firstblock) - ), - TP_fast_assign( - __entry->dev = args->mp->m_super->s_dev; - __entry->agno = args->agno; - __entry->agbno = args->agbno; - __entry->minlen = args->minlen; - __entry->maxlen = args->maxlen; - __entry->mod = args->mod; - __entry->prod = args->prod; - __entry->minleft = args->minleft; - __entry->total = args->total; - __entry->alignment = args->alignment; - __entry->minalignslop = args->minalignslop; - __entry->len = args->len; - __entry->type = args->type; - __entry->otype = args->otype; - __entry->wasdel = args->wasdel; - __entry->wasfromfl = args->wasfromfl; - __entry->isfl = args->isfl; - __entry->userdata = args->userdata; - __entry->firstblock = args->firstblock; - ), - TP_printk("dev %d:%d agno %u agbno %u minlen %u maxlen %u mod %u " - "prod %u minleft %u total %u alignment %u minalignslop %u " - "len %u type %s otype %s wasdel %d wasfromfl %d isfl %d " - "userdata %d firstblock 0x%llx", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->agbno, - __entry->minlen, - __entry->maxlen, - __entry->mod, - __entry->prod, - __entry->minleft, - __entry->total, - __entry->alignment, - __entry->minalignslop, - __entry->len, - __print_symbolic(__entry->type, XFS_ALLOC_TYPES), - __print_symbolic(__entry->otype, XFS_ALLOC_TYPES), - __entry->wasdel, - __entry->wasfromfl, - __entry->isfl, - __entry->userdata, - (unsigned long long)__entry->firstblock) -) - -#define DEFINE_ALLOC_EVENT(name) \ -DEFINE_EVENT(xfs_alloc_class, name, \ - TP_PROTO(struct xfs_alloc_arg *args), \ - TP_ARGS(args)) -DEFINE_ALLOC_EVENT(xfs_alloc_exact_done); -DEFINE_ALLOC_EVENT(xfs_alloc_exact_notfound); -DEFINE_ALLOC_EVENT(xfs_alloc_exact_error); -DEFINE_ALLOC_EVENT(xfs_alloc_near_nominleft); -DEFINE_ALLOC_EVENT(xfs_alloc_near_first); -DEFINE_ALLOC_EVENT(xfs_alloc_near_greater); -DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser); -DEFINE_ALLOC_EVENT(xfs_alloc_near_error); -DEFINE_ALLOC_EVENT(xfs_alloc_near_noentry); -DEFINE_ALLOC_EVENT(xfs_alloc_near_busy); -DEFINE_ALLOC_EVENT(xfs_alloc_size_neither); -DEFINE_ALLOC_EVENT(xfs_alloc_size_noentry); -DEFINE_ALLOC_EVENT(xfs_alloc_size_nominleft); -DEFINE_ALLOC_EVENT(xfs_alloc_size_done); -DEFINE_ALLOC_EVENT(xfs_alloc_size_error); -DEFINE_ALLOC_EVENT(xfs_alloc_size_busy); -DEFINE_ALLOC_EVENT(xfs_alloc_small_freelist); -DEFINE_ALLOC_EVENT(xfs_alloc_small_notenough); -DEFINE_ALLOC_EVENT(xfs_alloc_small_done); -DEFINE_ALLOC_EVENT(xfs_alloc_small_error); -DEFINE_ALLOC_EVENT(xfs_alloc_vextent_badargs); -DEFINE_ALLOC_EVENT(xfs_alloc_vextent_nofix); -DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp); -DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed); -DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed); - -DECLARE_EVENT_CLASS(xfs_dir2_class, - TP_PROTO(struct xfs_da_args *args), - TP_ARGS(args), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __dynamic_array(char, name, args->namelen) - __field(int, namelen) - __field(xfs_dahash_t, hashval) - __field(xfs_ino_t, inumber) - __field(int, op_flags) - ), - TP_fast_assign( - __entry->dev = VFS_I(args->dp)->i_sb->s_dev; - __entry->ino = args->dp->i_ino; - if (args->namelen) - memcpy(__get_str(name), args->name, args->namelen); - __entry->namelen = args->namelen; - __entry->hashval = args->hashval; - __entry->inumber = args->inumber; - __entry->op_flags = args->op_flags; - ), - TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d hashval 0x%x " - "inumber 0x%llx op_flags %s", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->namelen, - __entry->namelen ? __get_str(name) : NULL, - __entry->namelen, - __entry->hashval, - __entry->inumber, - __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS)) -) - -#define DEFINE_DIR2_EVENT(name) \ -DEFINE_EVENT(xfs_dir2_class, name, \ - TP_PROTO(struct xfs_da_args *args), \ - TP_ARGS(args)) -DEFINE_DIR2_EVENT(xfs_dir2_sf_addname); -DEFINE_DIR2_EVENT(xfs_dir2_sf_create); -DEFINE_DIR2_EVENT(xfs_dir2_sf_lookup); -DEFINE_DIR2_EVENT(xfs_dir2_sf_replace); -DEFINE_DIR2_EVENT(xfs_dir2_sf_removename); -DEFINE_DIR2_EVENT(xfs_dir2_sf_toino4); -DEFINE_DIR2_EVENT(xfs_dir2_sf_toino8); -DEFINE_DIR2_EVENT(xfs_dir2_sf_to_block); -DEFINE_DIR2_EVENT(xfs_dir2_block_addname); -DEFINE_DIR2_EVENT(xfs_dir2_block_lookup); -DEFINE_DIR2_EVENT(xfs_dir2_block_replace); -DEFINE_DIR2_EVENT(xfs_dir2_block_removename); -DEFINE_DIR2_EVENT(xfs_dir2_block_to_sf); -DEFINE_DIR2_EVENT(xfs_dir2_block_to_leaf); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_addname); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_lookup); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_replace); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_removename); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_block); -DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_node); -DEFINE_DIR2_EVENT(xfs_dir2_node_addname); -DEFINE_DIR2_EVENT(xfs_dir2_node_lookup); -DEFINE_DIR2_EVENT(xfs_dir2_node_replace); -DEFINE_DIR2_EVENT(xfs_dir2_node_removename); -DEFINE_DIR2_EVENT(xfs_dir2_node_to_leaf); - -DECLARE_EVENT_CLASS(xfs_dir2_space_class, - TP_PROTO(struct xfs_da_args *args, int idx), - TP_ARGS(args, idx), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(int, op_flags) - __field(int, idx) - ), - TP_fast_assign( - __entry->dev = VFS_I(args->dp)->i_sb->s_dev; - __entry->ino = args->dp->i_ino; - __entry->op_flags = args->op_flags; - __entry->idx = idx; - ), - TP_printk("dev %d:%d ino 0x%llx op_flags %s index %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS), - __entry->idx) -) - -#define DEFINE_DIR2_SPACE_EVENT(name) \ -DEFINE_EVENT(xfs_dir2_space_class, name, \ - TP_PROTO(struct xfs_da_args *args, int idx), \ - TP_ARGS(args, idx)) -DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_add); -DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_remove); -DEFINE_DIR2_SPACE_EVENT(xfs_dir2_grow_inode); -DEFINE_DIR2_SPACE_EVENT(xfs_dir2_shrink_inode); - -TRACE_EVENT(xfs_dir2_leafn_moveents, - TP_PROTO(struct xfs_da_args *args, int src_idx, int dst_idx, int count), - TP_ARGS(args, src_idx, dst_idx, count), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(int, op_flags) - __field(int, src_idx) - __field(int, dst_idx) - __field(int, count) - ), - TP_fast_assign( - __entry->dev = VFS_I(args->dp)->i_sb->s_dev; - __entry->ino = args->dp->i_ino; - __entry->op_flags = args->op_flags; - __entry->src_idx = src_idx; - __entry->dst_idx = dst_idx; - __entry->count = count; - ), - TP_printk("dev %d:%d ino 0x%llx op_flags %s " - "src_idx %d dst_idx %d count %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS), - __entry->src_idx, - __entry->dst_idx, - __entry->count) -); - -#define XFS_SWAPEXT_INODES \ - { 0, "target" }, \ - { 1, "temp" } - -#define XFS_INODE_FORMAT_STR \ - { 0, "invalid" }, \ - { 1, "local" }, \ - { 2, "extent" }, \ - { 3, "btree" } - -DECLARE_EVENT_CLASS(xfs_swap_extent_class, - TP_PROTO(struct xfs_inode *ip, int which), - TP_ARGS(ip, which), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(int, which) - __field(xfs_ino_t, ino) - __field(int, format) - __field(int, nex) - __field(int, max_nex) - __field(int, broot_size) - __field(int, fork_off) - ), - TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->which = which; - __entry->ino = ip->i_ino; - __entry->format = ip->i_d.di_format; - __entry->nex = ip->i_d.di_nextents; - __entry->max_nex = ip->i_df.if_ext_max; - __entry->broot_size = ip->i_df.if_broot_bytes; - __entry->fork_off = XFS_IFORK_BOFF(ip); - ), - TP_printk("dev %d:%d ino 0x%llx (%s), %s format, num_extents %d, " - "Max in-fork extents %d, broot size %d, fork offset %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __print_symbolic(__entry->which, XFS_SWAPEXT_INODES), - __print_symbolic(__entry->format, XFS_INODE_FORMAT_STR), - __entry->nex, - __entry->max_nex, - __entry->broot_size, - __entry->fork_off) -) - -#define DEFINE_SWAPEXT_EVENT(name) \ -DEFINE_EVENT(xfs_swap_extent_class, name, \ - TP_PROTO(struct xfs_inode *ip, int which), \ - TP_ARGS(ip, which)) - -DEFINE_SWAPEXT_EVENT(xfs_swap_extent_before); -DEFINE_SWAPEXT_EVENT(xfs_swap_extent_after); - -DECLARE_EVENT_CLASS(xfs_log_recover_item_class, - TP_PROTO(struct log *log, struct xlog_recover *trans, - struct xlog_recover_item *item, int pass), - TP_ARGS(log, trans, item, pass), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(unsigned long, item) - __field(xlog_tid_t, tid) - __field(int, type) - __field(int, pass) - __field(int, count) - __field(int, total) - ), - TP_fast_assign( - __entry->dev = log->l_mp->m_super->s_dev; - __entry->item = (unsigned long)item; - __entry->tid = trans->r_log_tid; - __entry->type = ITEM_TYPE(item); - __entry->pass = pass; - __entry->count = item->ri_cnt; - __entry->total = item->ri_total; - ), - TP_printk("dev %d:%d trans 0x%x, pass %d, item 0x%p, item type %s " - "item region count/total %d/%d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->tid, - __entry->pass, - (void *)__entry->item, - __print_symbolic(__entry->type, XFS_LI_TYPE_DESC), - __entry->count, - __entry->total) -) - -#define DEFINE_LOG_RECOVER_ITEM(name) \ -DEFINE_EVENT(xfs_log_recover_item_class, name, \ - TP_PROTO(struct log *log, struct xlog_recover *trans, \ - struct xlog_recover_item *item, int pass), \ - TP_ARGS(log, trans, item, pass)) - -DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add); -DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add_cont); -DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_head); -DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_tail); -DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_recover); - -DECLARE_EVENT_CLASS(xfs_log_recover_buf_item_class, - TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), - TP_ARGS(log, buf_f), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(__int64_t, blkno) - __field(unsigned short, len) - __field(unsigned short, flags) - __field(unsigned short, size) - __field(unsigned int, map_size) - ), - TP_fast_assign( - __entry->dev = log->l_mp->m_super->s_dev; - __entry->blkno = buf_f->blf_blkno; - __entry->len = buf_f->blf_len; - __entry->flags = buf_f->blf_flags; - __entry->size = buf_f->blf_size; - __entry->map_size = buf_f->blf_map_size; - ), - TP_printk("dev %d:%d blkno 0x%llx, len %u, flags 0x%x, size %d, " - "map_size %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->blkno, - __entry->len, - __entry->flags, - __entry->size, - __entry->map_size) -) - -#define DEFINE_LOG_RECOVER_BUF_ITEM(name) \ -DEFINE_EVENT(xfs_log_recover_buf_item_class, name, \ - TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), \ - TP_ARGS(log, buf_f)) - -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_not_cancel); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_add); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_ref_inc); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_recover); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_inode_buf); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_reg_buf); -DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_dquot_buf); - -DECLARE_EVENT_CLASS(xfs_log_recover_ino_item_class, - TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), - TP_ARGS(log, in_f), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_ino_t, ino) - __field(unsigned short, size) - __field(int, fields) - __field(unsigned short, asize) - __field(unsigned short, dsize) - __field(__int64_t, blkno) - __field(int, len) - __field(int, boffset) - ), - TP_fast_assign( - __entry->dev = log->l_mp->m_super->s_dev; - __entry->ino = in_f->ilf_ino; - __entry->size = in_f->ilf_size; - __entry->fields = in_f->ilf_fields; - __entry->asize = in_f->ilf_asize; - __entry->dsize = in_f->ilf_dsize; - __entry->blkno = in_f->ilf_blkno; - __entry->len = in_f->ilf_len; - __entry->boffset = in_f->ilf_boffset; - ), - TP_printk("dev %d:%d ino 0x%llx, size %u, fields 0x%x, asize %d, " - "dsize %d, blkno 0x%llx, len %d, boffset %d", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->ino, - __entry->size, - __entry->fields, - __entry->asize, - __entry->dsize, - __entry->blkno, - __entry->len, - __entry->boffset) -) -#define DEFINE_LOG_RECOVER_INO_ITEM(name) \ -DEFINE_EVENT(xfs_log_recover_ino_item_class, name, \ - TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), \ - TP_ARGS(log, in_f)) - -DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover); -DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel); -DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip); - -DECLARE_EVENT_CLASS(xfs_discard_class, - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno, xfs_extlen_t len), - TP_ARGS(mp, agno, agbno, len), - TP_STRUCT__entry( - __field(dev_t, dev) - __field(xfs_agnumber_t, agno) - __field(xfs_agblock_t, agbno) - __field(xfs_extlen_t, len) - ), - TP_fast_assign( - __entry->dev = mp->m_super->s_dev; - __entry->agno = agno; - __entry->agbno = agbno; - __entry->len = len; - ), - TP_printk("dev %d:%d agno %u agbno %u len %u\n", - MAJOR(__entry->dev), MINOR(__entry->dev), - __entry->agno, - __entry->agbno, - __entry->len) -) - -#define DEFINE_DISCARD_EVENT(name) \ -DEFINE_EVENT(xfs_discard_class, name, \ - TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ - xfs_agblock_t agbno, xfs_extlen_t len), \ - TP_ARGS(mp, agno, agbno, len)) -DEFINE_DISCARD_EVENT(xfs_discard_extent); -DEFINE_DISCARD_EVENT(xfs_discard_toosmall); -DEFINE_DISCARD_EVENT(xfs_discard_exclude); -DEFINE_DISCARD_EVENT(xfs_discard_busy); - -#endif /* _TRACE_XFS_H */ - -#undef TRACE_INCLUDE_PATH -#define TRACE_INCLUDE_PATH . -#define TRACE_INCLUDE_FILE xfs_trace -#include diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h deleted file mode 100644 index 7c220b4227bc..000000000000 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ /dev/null @@ -1,64 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_VNODE_H__ -#define __XFS_VNODE_H__ - -#include "xfs_fs.h" - -struct file; -struct xfs_inode; -struct xfs_iomap; -struct attrlist_cursor_kern; - -/* - * Return values for xfs_inactive. A return value of - * VN_INACTIVE_NOCACHE implies that the file system behavior - * has disassociated its state and bhv_desc_t from the vnode. - */ -#define VN_INACTIVE_CACHE 0 -#define VN_INACTIVE_NOCACHE 1 - -/* - * Flags for read/write calls - same values as IRIX - */ -#define IO_ISDIRECT 0x00004 /* bypass page cache */ -#define IO_INVIS 0x00020 /* don't update inode timestamps */ - -#define XFS_IO_FLAGS \ - { IO_ISDIRECT, "DIRECT" }, \ - { IO_INVIS, "INVIS"} - -/* - * Flush/Invalidate options for vop_toss/flush/flushinval_pages. - */ -#define FI_NONE 0 /* none */ -#define FI_REMAPF 1 /* Do a remapf prior to the operation */ -#define FI_REMAPF_LOCKED 2 /* Do a remapf prior to the operation. - Prevent VM access to the pages until - the operation completes. */ - -/* - * Some useful predicates. - */ -#define VN_MAPPED(vp) mapping_mapped(vp->i_mapping) -#define VN_CACHED(vp) (vp->i_mapping->nrpages) -#define VN_DIRTY(vp) mapping_tagged(vp->i_mapping, \ - PAGECACHE_TAG_DIRTY) - - -#endif /* __XFS_VNODE_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c deleted file mode 100644 index 87d3e03878c8..000000000000 --- a/fs/xfs/linux-2.6/xfs_xattr.c +++ /dev/null @@ -1,241 +0,0 @@ -/* - * Copyright (C) 2008 Christoph Hellwig. - * Portions Copyright (C) 2000-2008 Silicon Graphics, Inc. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ - -#include "xfs.h" -#include "xfs_da_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_attr.h" -#include "xfs_attr_leaf.h" -#include "xfs_acl.h" -#include "xfs_vnodeops.h" - -#include -#include - - -static int -xfs_xattr_get(struct dentry *dentry, const char *name, - void *value, size_t size, int xflags) -{ - struct xfs_inode *ip = XFS_I(dentry->d_inode); - int error, asize = size; - - if (strcmp(name, "") == 0) - return -EINVAL; - - /* Convert Linux syscall to XFS internal ATTR flags */ - if (!size) { - xflags |= ATTR_KERNOVAL; - value = NULL; - } - - error = -xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags); - if (error) - return error; - return asize; -} - -static int -xfs_xattr_set(struct dentry *dentry, const char *name, const void *value, - size_t size, int flags, int xflags) -{ - struct xfs_inode *ip = XFS_I(dentry->d_inode); - - if (strcmp(name, "") == 0) - return -EINVAL; - - /* Convert Linux syscall to XFS internal ATTR flags */ - if (flags & XATTR_CREATE) - xflags |= ATTR_CREATE; - if (flags & XATTR_REPLACE) - xflags |= ATTR_REPLACE; - - if (!value) - return -xfs_attr_remove(ip, (unsigned char *)name, xflags); - return -xfs_attr_set(ip, (unsigned char *)name, - (void *)value, size, xflags); -} - -static const struct xattr_handler xfs_xattr_user_handler = { - .prefix = XATTR_USER_PREFIX, - .flags = 0, /* no flags implies user namespace */ - .get = xfs_xattr_get, - .set = xfs_xattr_set, -}; - -static const struct xattr_handler xfs_xattr_trusted_handler = { - .prefix = XATTR_TRUSTED_PREFIX, - .flags = ATTR_ROOT, - .get = xfs_xattr_get, - .set = xfs_xattr_set, -}; - -static const struct xattr_handler xfs_xattr_security_handler = { - .prefix = XATTR_SECURITY_PREFIX, - .flags = ATTR_SECURE, - .get = xfs_xattr_get, - .set = xfs_xattr_set, -}; - -const struct xattr_handler *xfs_xattr_handlers[] = { - &xfs_xattr_user_handler, - &xfs_xattr_trusted_handler, - &xfs_xattr_security_handler, -#ifdef CONFIG_XFS_POSIX_ACL - &xfs_xattr_acl_access_handler, - &xfs_xattr_acl_default_handler, -#endif - NULL -}; - -static unsigned int xfs_xattr_prefix_len(int flags) -{ - if (flags & XFS_ATTR_SECURE) - return sizeof("security"); - else if (flags & XFS_ATTR_ROOT) - return sizeof("trusted"); - else - return sizeof("user"); -} - -static const char *xfs_xattr_prefix(int flags) -{ - if (flags & XFS_ATTR_SECURE) - return xfs_xattr_security_handler.prefix; - else if (flags & XFS_ATTR_ROOT) - return xfs_xattr_trusted_handler.prefix; - else - return xfs_xattr_user_handler.prefix; -} - -static int -xfs_xattr_put_listent( - struct xfs_attr_list_context *context, - int flags, - unsigned char *name, - int namelen, - int valuelen, - unsigned char *value) -{ - unsigned int prefix_len = xfs_xattr_prefix_len(flags); - char *offset; - int arraytop; - - ASSERT(context->count >= 0); - - /* - * Only show root namespace entries if we are actually allowed to - * see them. - */ - if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN)) - return 0; - - arraytop = context->count + prefix_len + namelen + 1; - if (arraytop > context->firstu) { - context->count = -1; /* insufficient space */ - return 1; - } - offset = (char *)context->alist + context->count; - strncpy(offset, xfs_xattr_prefix(flags), prefix_len); - offset += prefix_len; - strncpy(offset, (char *)name, namelen); /* real name */ - offset += namelen; - *offset = '\0'; - context->count += prefix_len + namelen + 1; - return 0; -} - -static int -xfs_xattr_put_listent_sizes( - struct xfs_attr_list_context *context, - int flags, - unsigned char *name, - int namelen, - int valuelen, - unsigned char *value) -{ - context->count += xfs_xattr_prefix_len(flags) + namelen + 1; - return 0; -} - -static int -list_one_attr(const char *name, const size_t len, void *data, - size_t size, ssize_t *result) -{ - char *p = data + *result; - - *result += len; - if (!size) - return 0; - if (*result > size) - return -ERANGE; - - strcpy(p, name); - return 0; -} - -ssize_t -xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size) -{ - struct xfs_attr_list_context context; - struct attrlist_cursor_kern cursor = { 0 }; - struct inode *inode = dentry->d_inode; - int error; - - /* - * First read the regular on-disk attributes. - */ - memset(&context, 0, sizeof(context)); - context.dp = XFS_I(inode); - context.cursor = &cursor; - context.resynch = 1; - context.alist = data; - context.bufsize = size; - context.firstu = context.bufsize; - - if (size) - context.put_listent = xfs_xattr_put_listent; - else - context.put_listent = xfs_xattr_put_listent_sizes; - - xfs_attr_list_int(&context); - if (context.count < 0) - return -ERANGE; - - /* - * Then add the two synthetic ACL attributes. - */ - if (posix_acl_access_exists(inode)) { - error = list_one_attr(POSIX_ACL_XATTR_ACCESS, - strlen(POSIX_ACL_XATTR_ACCESS) + 1, - data, size, &context.count); - if (error) - return error; - } - - if (posix_acl_default_exists(inode)) { - error = list_one_attr(POSIX_ACL_XATTR_DEFAULT, - strlen(POSIX_ACL_XATTR_DEFAULT) + 1, - data, size, &context.count); - if (error) - return error; - } - - return context.count; -} diff --git a/fs/xfs/mrlock.h b/fs/xfs/mrlock.h new file mode 100644 index 000000000000..ff6a19873e5c --- /dev/null +++ b/fs/xfs/mrlock.h @@ -0,0 +1,90 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SUPPORT_MRLOCK_H__ +#define __XFS_SUPPORT_MRLOCK_H__ + +#include + +typedef struct { + struct rw_semaphore mr_lock; +#ifdef DEBUG + int mr_writer; +#endif +} mrlock_t; + +#ifdef DEBUG +#define mrinit(mrp, name) \ + do { (mrp)->mr_writer = 0; init_rwsem(&(mrp)->mr_lock); } while (0) +#else +#define mrinit(mrp, name) \ + do { init_rwsem(&(mrp)->mr_lock); } while (0) +#endif + +#define mrlock_init(mrp, t,n,s) mrinit(mrp, n) +#define mrfree(mrp) do { } while (0) + +static inline void mraccess_nested(mrlock_t *mrp, int subclass) +{ + down_read_nested(&mrp->mr_lock, subclass); +} + +static inline void mrupdate_nested(mrlock_t *mrp, int subclass) +{ + down_write_nested(&mrp->mr_lock, subclass); +#ifdef DEBUG + mrp->mr_writer = 1; +#endif +} + +static inline int mrtryaccess(mrlock_t *mrp) +{ + return down_read_trylock(&mrp->mr_lock); +} + +static inline int mrtryupdate(mrlock_t *mrp) +{ + if (!down_write_trylock(&mrp->mr_lock)) + return 0; +#ifdef DEBUG + mrp->mr_writer = 1; +#endif + return 1; +} + +static inline void mrunlock_excl(mrlock_t *mrp) +{ +#ifdef DEBUG + mrp->mr_writer = 0; +#endif + up_write(&mrp->mr_lock); +} + +static inline void mrunlock_shared(mrlock_t *mrp) +{ + up_read(&mrp->mr_lock); +} + +static inline void mrdemote(mrlock_t *mrp) +{ +#ifdef DEBUG + mrp->mr_writer = 0; +#endif + downgrade_write(&mrp->mr_lock); +} + +#endif /* __XFS_SUPPORT_MRLOCK_H__ */ diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c deleted file mode 100644 index db62959bed13..000000000000 --- a/fs/xfs/quota/xfs_dquot.c +++ /dev/null @@ -1,1454 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_itable.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_trans_space.h" -#include "xfs_trans_priv.h" -#include "xfs_qm.h" -#include "xfs_trace.h" - - -/* - LOCK ORDER - - inode lock (ilock) - dquot hash-chain lock (hashlock) - xqm dquot freelist lock (freelistlock - mount's dquot list lock (mplistlock) - user dquot lock - lock ordering among dquots is based on the uid or gid - group dquot lock - similar to udquots. Between the two dquots, the udquot - has to be locked first. - pin lock - the dquot lock must be held to take this lock. - flush lock - ditto. -*/ - -#ifdef DEBUG -xfs_buftarg_t *xfs_dqerror_target; -int xfs_do_dqerror; -int xfs_dqreq_num; -int xfs_dqerror_mod = 33; -#endif - -static struct lock_class_key xfs_dquot_other_class; - -/* - * Allocate and initialize a dquot. We don't always allocate fresh memory; - * we try to reclaim a free dquot if the number of incore dquots are above - * a threshold. - * The only field inside the core that gets initialized at this point - * is the d_id field. The idea is to fill in the entire q_core - * when we read in the on disk dquot. - */ -STATIC xfs_dquot_t * -xfs_qm_dqinit( - xfs_mount_t *mp, - xfs_dqid_t id, - uint type) -{ - xfs_dquot_t *dqp; - boolean_t brandnewdquot; - - brandnewdquot = xfs_qm_dqalloc_incore(&dqp); - dqp->dq_flags = type; - dqp->q_core.d_id = cpu_to_be32(id); - dqp->q_mount = mp; - - /* - * No need to re-initialize these if this is a reclaimed dquot. - */ - if (brandnewdquot) { - INIT_LIST_HEAD(&dqp->q_freelist); - mutex_init(&dqp->q_qlock); - init_waitqueue_head(&dqp->q_pinwait); - - /* - * Because we want to use a counting completion, complete - * the flush completion once to allow a single access to - * the flush completion without blocking. - */ - init_completion(&dqp->q_flush); - complete(&dqp->q_flush); - - trace_xfs_dqinit(dqp); - } else { - /* - * Only the q_core portion was zeroed in dqreclaim_one(). - * So, we need to reset others. - */ - dqp->q_nrefs = 0; - dqp->q_blkno = 0; - INIT_LIST_HEAD(&dqp->q_mplist); - INIT_LIST_HEAD(&dqp->q_hashlist); - dqp->q_bufoffset = 0; - dqp->q_fileoffset = 0; - dqp->q_transp = NULL; - dqp->q_gdquot = NULL; - dqp->q_res_bcount = 0; - dqp->q_res_icount = 0; - dqp->q_res_rtbcount = 0; - atomic_set(&dqp->q_pincount, 0); - dqp->q_hash = NULL; - ASSERT(list_empty(&dqp->q_freelist)); - - trace_xfs_dqreuse(dqp); - } - - /* - * In either case we need to make sure group quotas have a different - * lock class than user quotas, to make sure lockdep knows we can - * locks of one of each at the same time. - */ - if (!(type & XFS_DQ_USER)) - lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class); - - /* - * log item gets initialized later - */ - return (dqp); -} - -/* - * This is called to free all the memory associated with a dquot - */ -void -xfs_qm_dqdestroy( - xfs_dquot_t *dqp) -{ - ASSERT(list_empty(&dqp->q_freelist)); - - mutex_destroy(&dqp->q_qlock); - kmem_zone_free(xfs_Gqm->qm_dqzone, dqp); - - atomic_dec(&xfs_Gqm->qm_totaldquots); -} - -/* - * This is what a 'fresh' dquot inside a dquot chunk looks like on disk. - */ -STATIC void -xfs_qm_dqinit_core( - xfs_dqid_t id, - uint type, - xfs_dqblk_t *d) -{ - /* - * Caller has zero'd the entire dquot 'chunk' already. - */ - d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); - d->dd_diskdq.d_version = XFS_DQUOT_VERSION; - d->dd_diskdq.d_id = cpu_to_be32(id); - d->dd_diskdq.d_flags = type; -} - -/* - * If default limits are in force, push them into the dquot now. - * We overwrite the dquot limits only if they are zero and this - * is not the root dquot. - */ -void -xfs_qm_adjust_dqlimits( - xfs_mount_t *mp, - xfs_disk_dquot_t *d) -{ - xfs_quotainfo_t *q = mp->m_quotainfo; - - ASSERT(d->d_id); - - if (q->qi_bsoftlimit && !d->d_blk_softlimit) - d->d_blk_softlimit = cpu_to_be64(q->qi_bsoftlimit); - if (q->qi_bhardlimit && !d->d_blk_hardlimit) - d->d_blk_hardlimit = cpu_to_be64(q->qi_bhardlimit); - if (q->qi_isoftlimit && !d->d_ino_softlimit) - d->d_ino_softlimit = cpu_to_be64(q->qi_isoftlimit); - if (q->qi_ihardlimit && !d->d_ino_hardlimit) - d->d_ino_hardlimit = cpu_to_be64(q->qi_ihardlimit); - if (q->qi_rtbsoftlimit && !d->d_rtb_softlimit) - d->d_rtb_softlimit = cpu_to_be64(q->qi_rtbsoftlimit); - if (q->qi_rtbhardlimit && !d->d_rtb_hardlimit) - d->d_rtb_hardlimit = cpu_to_be64(q->qi_rtbhardlimit); -} - -/* - * Check the limits and timers of a dquot and start or reset timers - * if necessary. - * This gets called even when quota enforcement is OFF, which makes our - * life a little less complicated. (We just don't reject any quota - * reservations in that case, when enforcement is off). - * We also return 0 as the values of the timers in Q_GETQUOTA calls, when - * enforcement's off. - * In contrast, warnings are a little different in that they don't - * 'automatically' get started when limits get exceeded. They do - * get reset to zero, however, when we find the count to be under - * the soft limit (they are only ever set non-zero via userspace). - */ -void -xfs_qm_adjust_dqtimers( - xfs_mount_t *mp, - xfs_disk_dquot_t *d) -{ - ASSERT(d->d_id); - -#ifdef DEBUG - if (d->d_blk_hardlimit) - ASSERT(be64_to_cpu(d->d_blk_softlimit) <= - be64_to_cpu(d->d_blk_hardlimit)); - if (d->d_ino_hardlimit) - ASSERT(be64_to_cpu(d->d_ino_softlimit) <= - be64_to_cpu(d->d_ino_hardlimit)); - if (d->d_rtb_hardlimit) - ASSERT(be64_to_cpu(d->d_rtb_softlimit) <= - be64_to_cpu(d->d_rtb_hardlimit)); -#endif - - if (!d->d_btimer) { - if ((d->d_blk_softlimit && - (be64_to_cpu(d->d_bcount) >= - be64_to_cpu(d->d_blk_softlimit))) || - (d->d_blk_hardlimit && - (be64_to_cpu(d->d_bcount) >= - be64_to_cpu(d->d_blk_hardlimit)))) { - d->d_btimer = cpu_to_be32(get_seconds() + - mp->m_quotainfo->qi_btimelimit); - } else { - d->d_bwarns = 0; - } - } else { - if ((!d->d_blk_softlimit || - (be64_to_cpu(d->d_bcount) < - be64_to_cpu(d->d_blk_softlimit))) && - (!d->d_blk_hardlimit || - (be64_to_cpu(d->d_bcount) < - be64_to_cpu(d->d_blk_hardlimit)))) { - d->d_btimer = 0; - } - } - - if (!d->d_itimer) { - if ((d->d_ino_softlimit && - (be64_to_cpu(d->d_icount) >= - be64_to_cpu(d->d_ino_softlimit))) || - (d->d_ino_hardlimit && - (be64_to_cpu(d->d_icount) >= - be64_to_cpu(d->d_ino_hardlimit)))) { - d->d_itimer = cpu_to_be32(get_seconds() + - mp->m_quotainfo->qi_itimelimit); - } else { - d->d_iwarns = 0; - } - } else { - if ((!d->d_ino_softlimit || - (be64_to_cpu(d->d_icount) < - be64_to_cpu(d->d_ino_softlimit))) && - (!d->d_ino_hardlimit || - (be64_to_cpu(d->d_icount) < - be64_to_cpu(d->d_ino_hardlimit)))) { - d->d_itimer = 0; - } - } - - if (!d->d_rtbtimer) { - if ((d->d_rtb_softlimit && - (be64_to_cpu(d->d_rtbcount) >= - be64_to_cpu(d->d_rtb_softlimit))) || - (d->d_rtb_hardlimit && - (be64_to_cpu(d->d_rtbcount) >= - be64_to_cpu(d->d_rtb_hardlimit)))) { - d->d_rtbtimer = cpu_to_be32(get_seconds() + - mp->m_quotainfo->qi_rtbtimelimit); - } else { - d->d_rtbwarns = 0; - } - } else { - if ((!d->d_rtb_softlimit || - (be64_to_cpu(d->d_rtbcount) < - be64_to_cpu(d->d_rtb_softlimit))) && - (!d->d_rtb_hardlimit || - (be64_to_cpu(d->d_rtbcount) < - be64_to_cpu(d->d_rtb_hardlimit)))) { - d->d_rtbtimer = 0; - } - } -} - -/* - * initialize a buffer full of dquots and log the whole thing - */ -STATIC void -xfs_qm_init_dquot_blk( - xfs_trans_t *tp, - xfs_mount_t *mp, - xfs_dqid_t id, - uint type, - xfs_buf_t *bp) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - xfs_dqblk_t *d; - int curid, i; - - ASSERT(tp); - ASSERT(xfs_buf_islocked(bp)); - - d = bp->b_addr; - - /* - * ID of the first dquot in the block - id's are zero based. - */ - curid = id - (id % q->qi_dqperchunk); - ASSERT(curid >= 0); - memset(d, 0, BBTOB(q->qi_dqchunklen)); - for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) - xfs_qm_dqinit_core(curid, type, d); - xfs_trans_dquot_buf(tp, bp, - (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF : - ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF : - XFS_BLF_GDQUOT_BUF))); - xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1); -} - - - -/* - * Allocate a block and fill it with dquots. - * This is called when the bmapi finds a hole. - */ -STATIC int -xfs_qm_dqalloc( - xfs_trans_t **tpp, - xfs_mount_t *mp, - xfs_dquot_t *dqp, - xfs_inode_t *quotip, - xfs_fileoff_t offset_fsb, - xfs_buf_t **O_bpp) -{ - xfs_fsblock_t firstblock; - xfs_bmap_free_t flist; - xfs_bmbt_irec_t map; - int nmaps, error, committed; - xfs_buf_t *bp; - xfs_trans_t *tp = *tpp; - - ASSERT(tp != NULL); - - trace_xfs_dqalloc(dqp); - - /* - * Initialize the bmap freelist prior to calling bmapi code. - */ - xfs_bmap_init(&flist, &firstblock); - xfs_ilock(quotip, XFS_ILOCK_EXCL); - /* - * Return if this type of quotas is turned off while we didn't - * have an inode lock - */ - if (XFS_IS_THIS_QUOTA_OFF(dqp)) { - xfs_iunlock(quotip, XFS_ILOCK_EXCL); - return (ESRCH); - } - - xfs_trans_ijoin_ref(tp, quotip, XFS_ILOCK_EXCL); - nmaps = 1; - if ((error = xfs_bmapi(tp, quotip, - offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB, - XFS_BMAPI_METADATA | XFS_BMAPI_WRITE, - &firstblock, - XFS_QM_DQALLOC_SPACE_RES(mp), - &map, &nmaps, &flist))) { - goto error0; - } - ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB); - ASSERT(nmaps == 1); - ASSERT((map.br_startblock != DELAYSTARTBLOCK) && - (map.br_startblock != HOLESTARTBLOCK)); - - /* - * Keep track of the blkno to save a lookup later - */ - dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); - - /* now we can just get the buffer (there's nothing to read yet) */ - bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, - dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, - 0); - if (!bp || (error = xfs_buf_geterror(bp))) - goto error1; - /* - * Make a chunk of dquots out of this buffer and log - * the entire thing. - */ - xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id), - dqp->dq_flags & XFS_DQ_ALLTYPES, bp); - - /* - * xfs_bmap_finish() may commit the current transaction and - * start a second transaction if the freelist is not empty. - * - * Since we still want to modify this buffer, we need to - * ensure that the buffer is not released on commit of - * the first transaction and ensure the buffer is added to the - * second transaction. - * - * If there is only one transaction then don't stop the buffer - * from being released when it commits later on. - */ - - xfs_trans_bhold(tp, bp); - - if ((error = xfs_bmap_finish(tpp, &flist, &committed))) { - goto error1; - } - - if (committed) { - tp = *tpp; - xfs_trans_bjoin(tp, bp); - } else { - xfs_trans_bhold_release(tp, bp); - } - - *O_bpp = bp; - return 0; - - error1: - xfs_bmap_cancel(&flist); - error0: - xfs_iunlock(quotip, XFS_ILOCK_EXCL); - - return (error); -} - -/* - * Maps a dquot to the buffer containing its on-disk version. - * This returns a ptr to the buffer containing the on-disk dquot - * in the bpp param, and a ptr to the on-disk dquot within that buffer - */ -STATIC int -xfs_qm_dqtobp( - xfs_trans_t **tpp, - xfs_dquot_t *dqp, - xfs_disk_dquot_t **O_ddpp, - xfs_buf_t **O_bpp, - uint flags) -{ - xfs_bmbt_irec_t map; - int nmaps = 1, error; - xfs_buf_t *bp; - xfs_inode_t *quotip = XFS_DQ_TO_QIP(dqp); - xfs_mount_t *mp = dqp->q_mount; - xfs_disk_dquot_t *ddq; - xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id); - xfs_trans_t *tp = (tpp ? *tpp : NULL); - - dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk; - - xfs_ilock(quotip, XFS_ILOCK_SHARED); - if (XFS_IS_THIS_QUOTA_OFF(dqp)) { - /* - * Return if this type of quotas is turned off while we - * didn't have the quota inode lock. - */ - xfs_iunlock(quotip, XFS_ILOCK_SHARED); - return ESRCH; - } - - /* - * Find the block map; no allocations yet - */ - error = xfs_bmapi(NULL, quotip, dqp->q_fileoffset, - XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, - NULL, 0, &map, &nmaps, NULL); - - xfs_iunlock(quotip, XFS_ILOCK_SHARED); - if (error) - return error; - - ASSERT(nmaps == 1); - ASSERT(map.br_blockcount == 1); - - /* - * Offset of dquot in the (fixed sized) dquot chunk. - */ - dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) * - sizeof(xfs_dqblk_t); - - ASSERT(map.br_startblock != DELAYSTARTBLOCK); - if (map.br_startblock == HOLESTARTBLOCK) { - /* - * We don't allocate unless we're asked to - */ - if (!(flags & XFS_QMOPT_DQALLOC)) - return ENOENT; - - ASSERT(tp); - error = xfs_qm_dqalloc(tpp, mp, dqp, quotip, - dqp->q_fileoffset, &bp); - if (error) - return error; - tp = *tpp; - } else { - trace_xfs_dqtobp_read(dqp); - - /* - * store the blkno etc so that we don't have to do the - * mapping all the time - */ - dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); - - error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, - dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, - 0, &bp); - if (error || !bp) - return XFS_ERROR(error); - } - - ASSERT(xfs_buf_islocked(bp)); - - /* - * calculate the location of the dquot inside the buffer. - */ - ddq = bp->b_addr + dqp->q_bufoffset; - - /* - * A simple sanity check in case we got a corrupted dquot... - */ - error = xfs_qm_dqcheck(mp, ddq, id, dqp->dq_flags & XFS_DQ_ALLTYPES, - flags & (XFS_QMOPT_DQREPAIR|XFS_QMOPT_DOWARN), - "dqtobp"); - if (error) { - if (!(flags & XFS_QMOPT_DQREPAIR)) { - xfs_trans_brelse(tp, bp); - return XFS_ERROR(EIO); - } - } - - *O_bpp = bp; - *O_ddpp = ddq; - - return (0); -} - - -/* - * Read in the ondisk dquot using dqtobp() then copy it to an incore version, - * and release the buffer immediately. - * - */ -/* ARGSUSED */ -STATIC int -xfs_qm_dqread( - xfs_trans_t **tpp, - xfs_dqid_t id, - xfs_dquot_t *dqp, /* dquot to get filled in */ - uint flags) -{ - xfs_disk_dquot_t *ddqp; - xfs_buf_t *bp; - int error; - xfs_trans_t *tp; - - ASSERT(tpp); - - trace_xfs_dqread(dqp); - - /* - * get a pointer to the on-disk dquot and the buffer containing it - * dqp already knows its own type (GROUP/USER). - */ - if ((error = xfs_qm_dqtobp(tpp, dqp, &ddqp, &bp, flags))) { - return (error); - } - tp = *tpp; - - /* copy everything from disk dquot to the incore dquot */ - memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t)); - ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); - xfs_qm_dquot_logitem_init(dqp); - - /* - * Reservation counters are defined as reservation plus current usage - * to avoid having to add every time. - */ - dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount); - dqp->q_res_icount = be64_to_cpu(ddqp->d_icount); - dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount); - - /* Mark the buf so that this will stay incore a little longer */ - XFS_BUF_SET_VTYPE_REF(bp, B_FS_DQUOT, XFS_DQUOT_REF); - - /* - * We got the buffer with a xfs_trans_read_buf() (in dqtobp()) - * So we need to release with xfs_trans_brelse(). - * The strategy here is identical to that of inodes; we lock - * the dquot in xfs_qm_dqget() before making it accessible to - * others. This is because dquots, like inodes, need a good level of - * concurrency, and we don't want to take locks on the entire buffers - * for dquot accesses. - * Note also that the dquot buffer may even be dirty at this point, if - * this particular dquot was repaired. We still aren't afraid to - * brelse it because we have the changes incore. - */ - ASSERT(xfs_buf_islocked(bp)); - xfs_trans_brelse(tp, bp); - - return (error); -} - - -/* - * allocate an incore dquot from the kernel heap, - * and fill its core with quota information kept on disk. - * If XFS_QMOPT_DQALLOC is set, it'll allocate a dquot on disk - * if it wasn't already allocated. - */ -STATIC int -xfs_qm_idtodq( - xfs_mount_t *mp, - xfs_dqid_t id, /* gid or uid, depending on type */ - uint type, /* UDQUOT or GDQUOT */ - uint flags, /* DQALLOC, DQREPAIR */ - xfs_dquot_t **O_dqpp)/* OUT : incore dquot, not locked */ -{ - xfs_dquot_t *dqp; - int error; - xfs_trans_t *tp; - int cancelflags=0; - - dqp = xfs_qm_dqinit(mp, id, type); - tp = NULL; - if (flags & XFS_QMOPT_DQALLOC) { - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC); - error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp), - XFS_WRITE_LOG_RES(mp) + - BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 + - 128, - 0, - XFS_TRANS_PERM_LOG_RES, - XFS_WRITE_LOG_COUNT); - if (error) { - cancelflags = 0; - goto error0; - } - cancelflags = XFS_TRANS_RELEASE_LOG_RES; - } - - /* - * Read it from disk; xfs_dqread() takes care of - * all the necessary initialization of dquot's fields (locks, etc) - */ - if ((error = xfs_qm_dqread(&tp, id, dqp, flags))) { - /* - * This can happen if quotas got turned off (ESRCH), - * or if the dquot didn't exist on disk and we ask to - * allocate (ENOENT). - */ - trace_xfs_dqread_fail(dqp); - cancelflags |= XFS_TRANS_ABORT; - goto error0; - } - if (tp) { - if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) - goto error1; - } - - *O_dqpp = dqp; - return (0); - - error0: - ASSERT(error); - if (tp) - xfs_trans_cancel(tp, cancelflags); - error1: - xfs_qm_dqdestroy(dqp); - *O_dqpp = NULL; - return (error); -} - -/* - * Lookup a dquot in the incore dquot hashtable. We keep two separate - * hashtables for user and group dquots; and, these are global tables - * inside the XQM, not per-filesystem tables. - * The hash chain must be locked by caller, and it is left locked - * on return. Returning dquot is locked. - */ -STATIC int -xfs_qm_dqlookup( - xfs_mount_t *mp, - xfs_dqid_t id, - xfs_dqhash_t *qh, - xfs_dquot_t **O_dqpp) -{ - xfs_dquot_t *dqp; - uint flist_locked; - - ASSERT(mutex_is_locked(&qh->qh_lock)); - - flist_locked = B_FALSE; - - /* - * Traverse the hashchain looking for a match - */ - list_for_each_entry(dqp, &qh->qh_list, q_hashlist) { - /* - * We already have the hashlock. We don't need the - * dqlock to look at the id field of the dquot, since the - * id can't be modified without the hashlock anyway. - */ - if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) { - trace_xfs_dqlookup_found(dqp); - - /* - * All in core dquots must be on the dqlist of mp - */ - ASSERT(!list_empty(&dqp->q_mplist)); - - xfs_dqlock(dqp); - if (dqp->q_nrefs == 0) { - ASSERT(!list_empty(&dqp->q_freelist)); - if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { - trace_xfs_dqlookup_want(dqp); - - /* - * We may have raced with dqreclaim_one() - * (and lost). So, flag that we don't - * want the dquot to be reclaimed. - */ - dqp->dq_flags |= XFS_DQ_WANT; - xfs_dqunlock(dqp); - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - xfs_dqlock(dqp); - dqp->dq_flags &= ~(XFS_DQ_WANT); - } - flist_locked = B_TRUE; - } - - /* - * id couldn't have changed; we had the hashlock all - * along - */ - ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); - - if (flist_locked) { - if (dqp->q_nrefs != 0) { - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - flist_locked = B_FALSE; - } else { - /* take it off the freelist */ - trace_xfs_dqlookup_freelist(dqp); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - } - } - - XFS_DQHOLD(dqp); - - if (flist_locked) - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - /* - * move the dquot to the front of the hashchain - */ - ASSERT(mutex_is_locked(&qh->qh_lock)); - list_move(&dqp->q_hashlist, &qh->qh_list); - trace_xfs_dqlookup_done(dqp); - *O_dqpp = dqp; - return 0; - } - } - - *O_dqpp = NULL; - ASSERT(mutex_is_locked(&qh->qh_lock)); - return (1); -} - -/* - * Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a - * a locked dquot, doing an allocation (if requested) as needed. - * When both an inode and an id are given, the inode's id takes precedence. - * That is, if the id changes while we don't hold the ilock inside this - * function, the new dquot is returned, not necessarily the one requested - * in the id argument. - */ -int -xfs_qm_dqget( - xfs_mount_t *mp, - xfs_inode_t *ip, /* locked inode (optional) */ - xfs_dqid_t id, /* uid/projid/gid depending on type */ - uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */ - uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */ - xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */ -{ - xfs_dquot_t *dqp; - xfs_dqhash_t *h; - uint version; - int error; - - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) || - (! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) || - (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) { - return (ESRCH); - } - h = XFS_DQ_HASH(mp, id, type); - -#ifdef DEBUG - if (xfs_do_dqerror) { - if ((xfs_dqerror_target == mp->m_ddev_targp) && - (xfs_dqreq_num++ % xfs_dqerror_mod) == 0) { - xfs_debug(mp, "Returning error in dqget"); - return (EIO); - } - } -#endif - - again: - -#ifdef DEBUG - ASSERT(type == XFS_DQ_USER || - type == XFS_DQ_PROJ || - type == XFS_DQ_GROUP); - if (ip) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (type == XFS_DQ_USER) - ASSERT(ip->i_udquot == NULL); - else - ASSERT(ip->i_gdquot == NULL); - } -#endif - mutex_lock(&h->qh_lock); - - /* - * Look in the cache (hashtable). - * The chain is kept locked during lookup. - */ - if (xfs_qm_dqlookup(mp, id, h, O_dqpp) == 0) { - XQM_STATS_INC(xqmstats.xs_qm_dqcachehits); - /* - * The dquot was found, moved to the front of the chain, - * taken off the freelist if it was on it, and locked - * at this point. Just unlock the hashchain and return. - */ - ASSERT(*O_dqpp); - ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp)); - mutex_unlock(&h->qh_lock); - trace_xfs_dqget_hit(*O_dqpp); - return (0); /* success */ - } - XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses); - - /* - * Dquot cache miss. We don't want to keep the inode lock across - * a (potential) disk read. Also we don't want to deal with the lock - * ordering between quotainode and this inode. OTOH, dropping the inode - * lock here means dealing with a chown that can happen before - * we re-acquire the lock. - */ - if (ip) - xfs_iunlock(ip, XFS_ILOCK_EXCL); - /* - * Save the hashchain version stamp, and unlock the chain, so that - * we don't keep the lock across a disk read - */ - version = h->qh_version; - mutex_unlock(&h->qh_lock); - - /* - * Allocate the dquot on the kernel heap, and read the ondisk - * portion off the disk. Also, do all the necessary initialization - * This can return ENOENT if dquot didn't exist on disk and we didn't - * ask it to allocate; ESRCH if quotas got turned off suddenly. - */ - if ((error = xfs_qm_idtodq(mp, id, type, - flags & (XFS_QMOPT_DQALLOC|XFS_QMOPT_DQREPAIR| - XFS_QMOPT_DOWARN), - &dqp))) { - if (ip) - xfs_ilock(ip, XFS_ILOCK_EXCL); - return (error); - } - - /* - * See if this is mount code calling to look at the overall quota limits - * which are stored in the id == 0 user or group's dquot. - * Since we may not have done a quotacheck by this point, just return - * the dquot without attaching it to any hashtables, lists, etc, or even - * taking a reference. - * The caller must dqdestroy this once done. - */ - if (flags & XFS_QMOPT_DQSUSER) { - ASSERT(id == 0); - ASSERT(! ip); - goto dqret; - } - - /* - * Dquot lock comes after hashlock in the lock ordering - */ - if (ip) { - xfs_ilock(ip, XFS_ILOCK_EXCL); - - /* - * A dquot could be attached to this inode by now, since - * we had dropped the ilock. - */ - if (type == XFS_DQ_USER) { - if (!XFS_IS_UQUOTA_ON(mp)) { - /* inode stays locked on return */ - xfs_qm_dqdestroy(dqp); - return XFS_ERROR(ESRCH); - } - if (ip->i_udquot) { - xfs_qm_dqdestroy(dqp); - dqp = ip->i_udquot; - xfs_dqlock(dqp); - goto dqret; - } - } else { - if (!XFS_IS_OQUOTA_ON(mp)) { - /* inode stays locked on return */ - xfs_qm_dqdestroy(dqp); - return XFS_ERROR(ESRCH); - } - if (ip->i_gdquot) { - xfs_qm_dqdestroy(dqp); - dqp = ip->i_gdquot; - xfs_dqlock(dqp); - goto dqret; - } - } - } - - /* - * Hashlock comes after ilock in lock order - */ - mutex_lock(&h->qh_lock); - if (version != h->qh_version) { - xfs_dquot_t *tmpdqp; - /* - * Now, see if somebody else put the dquot in the - * hashtable before us. This can happen because we didn't - * keep the hashchain lock. We don't have to worry about - * lock order between the two dquots here since dqp isn't - * on any findable lists yet. - */ - if (xfs_qm_dqlookup(mp, id, h, &tmpdqp) == 0) { - /* - * Duplicate found. Just throw away the new dquot - * and start over. - */ - xfs_qm_dqput(tmpdqp); - mutex_unlock(&h->qh_lock); - xfs_qm_dqdestroy(dqp); - XQM_STATS_INC(xqmstats.xs_qm_dquot_dups); - goto again; - } - } - - /* - * Put the dquot at the beginning of the hash-chain and mp's list - * LOCK ORDER: hashlock, freelistlock, mplistlock, udqlock, gdqlock .. - */ - ASSERT(mutex_is_locked(&h->qh_lock)); - dqp->q_hash = h; - list_add(&dqp->q_hashlist, &h->qh_list); - h->qh_version++; - - /* - * Attach this dquot to this filesystem's list of all dquots, - * kept inside the mount structure in m_quotainfo field - */ - mutex_lock(&mp->m_quotainfo->qi_dqlist_lock); - - /* - * We return a locked dquot to the caller, with a reference taken - */ - xfs_dqlock(dqp); - dqp->q_nrefs = 1; - - list_add(&dqp->q_mplist, &mp->m_quotainfo->qi_dqlist); - mp->m_quotainfo->qi_dquots++; - mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); - mutex_unlock(&h->qh_lock); - dqret: - ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL)); - trace_xfs_dqget_miss(dqp); - *O_dqpp = dqp; - return (0); -} - - -/* - * Release a reference to the dquot (decrement ref-count) - * and unlock it. If there is a group quota attached to this - * dquot, carefully release that too without tripping over - * deadlocks'n'stuff. - */ -void -xfs_qm_dqput( - xfs_dquot_t *dqp) -{ - xfs_dquot_t *gdqp; - - ASSERT(dqp->q_nrefs > 0); - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - - trace_xfs_dqput(dqp); - - if (dqp->q_nrefs != 1) { - dqp->q_nrefs--; - xfs_dqunlock(dqp); - return; - } - - /* - * drop the dqlock and acquire the freelist and dqlock - * in the right order; but try to get it out-of-order first - */ - if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { - trace_xfs_dqput_wait(dqp); - xfs_dqunlock(dqp); - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - xfs_dqlock(dqp); - } - - while (1) { - gdqp = NULL; - - /* We can't depend on nrefs being == 1 here */ - if (--dqp->q_nrefs == 0) { - trace_xfs_dqput_free(dqp); - - list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist); - xfs_Gqm->qm_dqfrlist_cnt++; - - /* - * If we just added a udquot to the freelist, then - * we want to release the gdquot reference that - * it (probably) has. Otherwise it'll keep the - * gdquot from getting reclaimed. - */ - if ((gdqp = dqp->q_gdquot)) { - /* - * Avoid a recursive dqput call - */ - xfs_dqlock(gdqp); - dqp->q_gdquot = NULL; - } - } - xfs_dqunlock(dqp); - - /* - * If we had a group quota inside the user quota as a hint, - * release it now. - */ - if (! gdqp) - break; - dqp = gdqp; - } - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); -} - -/* - * Release a dquot. Flush it if dirty, then dqput() it. - * dquot must not be locked. - */ -void -xfs_qm_dqrele( - xfs_dquot_t *dqp) -{ - if (!dqp) - return; - - trace_xfs_dqrele(dqp); - - xfs_dqlock(dqp); - /* - * We don't care to flush it if the dquot is dirty here. - * That will create stutters that we want to avoid. - * Instead we do a delayed write when we try to reclaim - * a dirty dquot. Also xfs_sync will take part of the burden... - */ - xfs_qm_dqput(dqp); -} - -/* - * This is the dquot flushing I/O completion routine. It is called - * from interrupt level when the buffer containing the dquot is - * flushed to disk. It is responsible for removing the dquot logitem - * from the AIL if it has not been re-logged, and unlocking the dquot's - * flush lock. This behavior is very similar to that of inodes.. - */ -STATIC void -xfs_qm_dqflush_done( - struct xfs_buf *bp, - struct xfs_log_item *lip) -{ - xfs_dq_logitem_t *qip = (struct xfs_dq_logitem *)lip; - xfs_dquot_t *dqp = qip->qli_dquot; - struct xfs_ail *ailp = lip->li_ailp; - - /* - * We only want to pull the item from the AIL if its - * location in the log has not changed since we started the flush. - * Thus, we only bother if the dquot's lsn has - * not changed. First we check the lsn outside the lock - * since it's cheaper, and then we recheck while - * holding the lock before removing the dquot from the AIL. - */ - if ((lip->li_flags & XFS_LI_IN_AIL) && - lip->li_lsn == qip->qli_flush_lsn) { - - /* xfs_trans_ail_delete() drops the AIL lock. */ - spin_lock(&ailp->xa_lock); - if (lip->li_lsn == qip->qli_flush_lsn) - xfs_trans_ail_delete(ailp, lip); - else - spin_unlock(&ailp->xa_lock); - } - - /* - * Release the dq's flush lock since we're done with it. - */ - xfs_dqfunlock(dqp); -} - -/* - * Write a modified dquot to disk. - * The dquot must be locked and the flush lock too taken by caller. - * The flush lock will not be unlocked until the dquot reaches the disk, - * but the dquot is free to be unlocked and modified by the caller - * in the interim. Dquot is still locked on return. This behavior is - * identical to that of inodes. - */ -int -xfs_qm_dqflush( - xfs_dquot_t *dqp, - uint flags) -{ - struct xfs_mount *mp = dqp->q_mount; - struct xfs_buf *bp; - struct xfs_disk_dquot *ddqp; - int error; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(!completion_done(&dqp->q_flush)); - - trace_xfs_dqflush(dqp); - - /* - * If not dirty, or it's pinned and we are not supposed to block, nada. - */ - if (!XFS_DQ_IS_DIRTY(dqp) || - (!(flags & SYNC_WAIT) && atomic_read(&dqp->q_pincount) > 0)) { - xfs_dqfunlock(dqp); - return 0; - } - xfs_qm_dqunpin_wait(dqp); - - /* - * This may have been unpinned because the filesystem is shutting - * down forcibly. If that's the case we must not write this dquot - * to disk, because the log record didn't make it to disk! - */ - if (XFS_FORCED_SHUTDOWN(mp)) { - dqp->dq_flags &= ~XFS_DQ_DIRTY; - xfs_dqfunlock(dqp); - return XFS_ERROR(EIO); - } - - /* - * Get the buffer containing the on-disk dquot - */ - error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, 0, &bp); - if (error) { - ASSERT(error != ENOENT); - xfs_dqfunlock(dqp); - return error; - } - - /* - * Calculate the location of the dquot inside the buffer. - */ - ddqp = bp->b_addr + dqp->q_bufoffset; - - /* - * A simple sanity check in case we got a corrupted dquot.. - */ - error = xfs_qm_dqcheck(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0, - XFS_QMOPT_DOWARN, "dqflush (incore copy)"); - if (error) { - xfs_buf_relse(bp); - xfs_dqfunlock(dqp); - xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); - return XFS_ERROR(EIO); - } - - /* This is the only portion of data that needs to persist */ - memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t)); - - /* - * Clear the dirty field and remember the flush lsn for later use. - */ - dqp->dq_flags &= ~XFS_DQ_DIRTY; - - xfs_trans_ail_copy_lsn(mp->m_ail, &dqp->q_logitem.qli_flush_lsn, - &dqp->q_logitem.qli_item.li_lsn); - - /* - * Attach an iodone routine so that we can remove this dquot from the - * AIL and release the flush lock once the dquot is synced to disk. - */ - xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done, - &dqp->q_logitem.qli_item); - - /* - * If the buffer is pinned then push on the log so we won't - * get stuck waiting in the write for too long. - */ - if (xfs_buf_ispinned(bp)) { - trace_xfs_dqflush_force(dqp); - xfs_log_force(mp, 0); - } - - if (flags & SYNC_WAIT) - error = xfs_bwrite(mp, bp); - else - xfs_bdwrite(mp, bp); - - trace_xfs_dqflush_done(dqp); - - /* - * dqp is still locked, but caller is free to unlock it now. - */ - return error; - -} - -int -xfs_qm_dqlock_nowait( - xfs_dquot_t *dqp) -{ - return mutex_trylock(&dqp->q_qlock); -} - -void -xfs_dqlock( - xfs_dquot_t *dqp) -{ - mutex_lock(&dqp->q_qlock); -} - -void -xfs_dqunlock( - xfs_dquot_t *dqp) -{ - mutex_unlock(&(dqp->q_qlock)); - if (dqp->q_logitem.qli_dquot == dqp) { - /* Once was dqp->q_mount, but might just have been cleared */ - xfs_trans_unlocked_item(dqp->q_logitem.qli_item.li_ailp, - (xfs_log_item_t*)&(dqp->q_logitem)); - } -} - - -void -xfs_dqunlock_nonotify( - xfs_dquot_t *dqp) -{ - mutex_unlock(&(dqp->q_qlock)); -} - -/* - * Lock two xfs_dquot structures. - * - * To avoid deadlocks we always lock the quota structure with - * the lowerd id first. - */ -void -xfs_dqlock2( - xfs_dquot_t *d1, - xfs_dquot_t *d2) -{ - if (d1 && d2) { - ASSERT(d1 != d2); - if (be32_to_cpu(d1->q_core.d_id) > - be32_to_cpu(d2->q_core.d_id)) { - mutex_lock(&d2->q_qlock); - mutex_lock_nested(&d1->q_qlock, XFS_QLOCK_NESTED); - } else { - mutex_lock(&d1->q_qlock); - mutex_lock_nested(&d2->q_qlock, XFS_QLOCK_NESTED); - } - } else if (d1) { - mutex_lock(&d1->q_qlock); - } else if (d2) { - mutex_lock(&d2->q_qlock); - } -} - - -/* - * Take a dquot out of the mount's dqlist as well as the hashlist. - * This is called via unmount as well as quotaoff, and the purge - * will always succeed unless there are soft (temp) references - * outstanding. - * - * This returns 0 if it was purged, 1 if it wasn't. It's not an error code - * that we're returning! XXXsup - not cool. - */ -/* ARGSUSED */ -int -xfs_qm_dqpurge( - xfs_dquot_t *dqp) -{ - xfs_dqhash_t *qh = dqp->q_hash; - xfs_mount_t *mp = dqp->q_mount; - - ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock)); - ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock)); - - xfs_dqlock(dqp); - /* - * We really can't afford to purge a dquot that is - * referenced, because these are hard refs. - * It shouldn't happen in general because we went thru _all_ inodes in - * dqrele_all_inodes before calling this and didn't let the mountlock go. - * However it is possible that we have dquots with temporary - * references that are not attached to an inode. e.g. see xfs_setattr(). - */ - if (dqp->q_nrefs != 0) { - xfs_dqunlock(dqp); - mutex_unlock(&dqp->q_hash->qh_lock); - return (1); - } - - ASSERT(!list_empty(&dqp->q_freelist)); - - /* - * If we're turning off quotas, we have to make sure that, for - * example, we don't delete quota disk blocks while dquots are - * in the process of getting written to those disk blocks. - * This dquot might well be on AIL, and we can't leave it there - * if we're turning off quotas. Basically, we need this flush - * lock, and are willing to block on it. - */ - if (!xfs_dqflock_nowait(dqp)) { - /* - * Block on the flush lock after nudging dquot buffer, - * if it is incore. - */ - xfs_qm_dqflock_pushbuf_wait(dqp); - } - - /* - * XXXIf we're turning this type of quotas off, we don't care - * about the dirty metadata sitting in this dquot. OTOH, if - * we're unmounting, we do care, so we flush it and wait. - */ - if (XFS_DQ_IS_DIRTY(dqp)) { - int error; - - /* dqflush unlocks dqflock */ - /* - * Given that dqpurge is a very rare occurrence, it is OK - * that we're holding the hashlist and mplist locks - * across the disk write. But, ... XXXsup - * - * We don't care about getting disk errors here. We need - * to purge this dquot anyway, so we go ahead regardless. - */ - error = xfs_qm_dqflush(dqp, SYNC_WAIT); - if (error) - xfs_warn(mp, "%s: dquot %p flush failed", - __func__, dqp); - xfs_dqflock(dqp); - } - ASSERT(atomic_read(&dqp->q_pincount) == 0); - ASSERT(XFS_FORCED_SHUTDOWN(mp) || - !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL)); - - list_del_init(&dqp->q_hashlist); - qh->qh_version++; - list_del_init(&dqp->q_mplist); - mp->m_quotainfo->qi_dqreclaims++; - mp->m_quotainfo->qi_dquots--; - /* - * XXX Move this to the front of the freelist, if we can get the - * freelist lock. - */ - ASSERT(!list_empty(&dqp->q_freelist)); - - dqp->q_mount = NULL; - dqp->q_hash = NULL; - dqp->dq_flags = XFS_DQ_INACTIVE; - memset(&dqp->q_core, 0, sizeof(dqp->q_core)); - xfs_dqfunlock(dqp); - xfs_dqunlock(dqp); - mutex_unlock(&qh->qh_lock); - return (0); -} - - -/* - * Give the buffer a little push if it is incore and - * wait on the flush lock. - */ -void -xfs_qm_dqflock_pushbuf_wait( - xfs_dquot_t *dqp) -{ - xfs_mount_t *mp = dqp->q_mount; - xfs_buf_t *bp; - - /* - * Check to see if the dquot has been flushed delayed - * write. If so, grab its buffer and send it - * out immediately. We'll be able to acquire - * the flush lock when the I/O completes. - */ - bp = xfs_incore(mp->m_ddev_targp, dqp->q_blkno, - mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK); - if (!bp) - goto out_lock; - - if (XFS_BUF_ISDELAYWRITE(bp)) { - if (xfs_buf_ispinned(bp)) - xfs_log_force(mp, 0); - xfs_buf_delwri_promote(bp); - wake_up_process(bp->b_target->bt_task); - } - xfs_buf_relse(bp); -out_lock: - xfs_dqflock(dqp); -} diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h deleted file mode 100644 index 34b7e945dbfa..000000000000 --- a/fs/xfs/quota/xfs_dquot.h +++ /dev/null @@ -1,137 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_DQUOT_H__ -#define __XFS_DQUOT_H__ - -/* - * Dquots are structures that hold quota information about a user or a group, - * much like inodes are for files. In fact, dquots share many characteristics - * with inodes. However, dquots can also be a centralized resource, relative - * to a collection of inodes. In this respect, dquots share some characteristics - * of the superblock. - * XFS dquots exploit both those in its algorithms. They make every attempt - * to not be a bottleneck when quotas are on and have minimal impact, if any, - * when quotas are off. - */ - -/* - * The hash chain headers (hash buckets) - */ -typedef struct xfs_dqhash { - struct list_head qh_list; - struct mutex qh_lock; - uint qh_version; /* ever increasing version */ - uint qh_nelems; /* number of dquots on the list */ -} xfs_dqhash_t; - -struct xfs_mount; -struct xfs_trans; - -/* - * The incore dquot structure - */ -typedef struct xfs_dquot { - uint dq_flags; /* various flags (XFS_DQ_*) */ - struct list_head q_freelist; /* global free list of dquots */ - struct list_head q_mplist; /* mount's list of dquots */ - struct list_head q_hashlist; /* gloabl hash list of dquots */ - xfs_dqhash_t *q_hash; /* the hashchain header */ - struct xfs_mount*q_mount; /* filesystem this relates to */ - struct xfs_trans*q_transp; /* trans this belongs to currently */ - uint q_nrefs; /* # active refs from inodes */ - xfs_daddr_t q_blkno; /* blkno of dquot buffer */ - int q_bufoffset; /* off of dq in buffer (# dquots) */ - xfs_fileoff_t q_fileoffset; /* offset in quotas file */ - - struct xfs_dquot*q_gdquot; /* group dquot, hint only */ - xfs_disk_dquot_t q_core; /* actual usage & quotas */ - xfs_dq_logitem_t q_logitem; /* dquot log item */ - xfs_qcnt_t q_res_bcount; /* total regular nblks used+reserved */ - xfs_qcnt_t q_res_icount; /* total inos allocd+reserved */ - xfs_qcnt_t q_res_rtbcount;/* total realtime blks used+reserved */ - struct mutex q_qlock; /* quota lock */ - struct completion q_flush; /* flush completion queue */ - atomic_t q_pincount; /* dquot pin count */ - wait_queue_head_t q_pinwait; /* dquot pinning wait queue */ -} xfs_dquot_t; - -/* - * Lock hierarchy for q_qlock: - * XFS_QLOCK_NORMAL is the implicit default, - * XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2 - */ -enum { - XFS_QLOCK_NORMAL = 0, - XFS_QLOCK_NESTED, -}; - -#define XFS_DQHOLD(dqp) ((dqp)->q_nrefs++) - -/* - * Manage the q_flush completion queue embedded in the dquot. This completion - * queue synchronizes processes attempting to flush the in-core dquot back to - * disk. - */ -static inline void xfs_dqflock(xfs_dquot_t *dqp) -{ - wait_for_completion(&dqp->q_flush); -} - -static inline int xfs_dqflock_nowait(xfs_dquot_t *dqp) -{ - return try_wait_for_completion(&dqp->q_flush); -} - -static inline void xfs_dqfunlock(xfs_dquot_t *dqp) -{ - complete(&dqp->q_flush); -} - -#define XFS_DQ_IS_LOCKED(dqp) (mutex_is_locked(&((dqp)->q_qlock))) -#define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY) -#define XFS_QM_ISUDQ(dqp) ((dqp)->dq_flags & XFS_DQ_USER) -#define XFS_QM_ISPDQ(dqp) ((dqp)->dq_flags & XFS_DQ_PROJ) -#define XFS_QM_ISGDQ(dqp) ((dqp)->dq_flags & XFS_DQ_GROUP) -#define XFS_DQ_TO_QINF(dqp) ((dqp)->q_mount->m_quotainfo) -#define XFS_DQ_TO_QIP(dqp) (XFS_QM_ISUDQ(dqp) ? \ - XFS_DQ_TO_QINF(dqp)->qi_uquotaip : \ - XFS_DQ_TO_QINF(dqp)->qi_gquotaip) - -#define XFS_IS_THIS_QUOTA_OFF(d) (! (XFS_QM_ISUDQ(d) ? \ - (XFS_IS_UQUOTA_ON((d)->q_mount)) : \ - (XFS_IS_OQUOTA_ON((d)->q_mount)))) - -extern void xfs_qm_dqdestroy(xfs_dquot_t *); -extern int xfs_qm_dqflush(xfs_dquot_t *, uint); -extern int xfs_qm_dqpurge(xfs_dquot_t *); -extern void xfs_qm_dqunpin_wait(xfs_dquot_t *); -extern int xfs_qm_dqlock_nowait(xfs_dquot_t *); -extern void xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp); -extern void xfs_qm_adjust_dqtimers(xfs_mount_t *, - xfs_disk_dquot_t *); -extern void xfs_qm_adjust_dqlimits(xfs_mount_t *, - xfs_disk_dquot_t *); -extern int xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *, - xfs_dqid_t, uint, uint, xfs_dquot_t **); -extern void xfs_qm_dqput(xfs_dquot_t *); -extern void xfs_dqlock(xfs_dquot_t *); -extern void xfs_dqlock2(xfs_dquot_t *, xfs_dquot_t *); -extern void xfs_dqunlock(xfs_dquot_t *); -extern void xfs_dqunlock_nonotify(xfs_dquot_t *); - -#endif /* __XFS_DQUOT_H__ */ diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c deleted file mode 100644 index 9e0e2fa3f2c8..000000000000 --- a/fs/xfs/quota/xfs_dquot_item.c +++ /dev/null @@ -1,529 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_itable.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_trans_priv.h" -#include "xfs_qm.h" - -static inline struct xfs_dq_logitem *DQUOT_ITEM(struct xfs_log_item *lip) -{ - return container_of(lip, struct xfs_dq_logitem, qli_item); -} - -/* - * returns the number of iovecs needed to log the given dquot item. - */ -STATIC uint -xfs_qm_dquot_logitem_size( - struct xfs_log_item *lip) -{ - /* - * we need only two iovecs, one for the format, one for the real thing - */ - return 2; -} - -/* - * fills in the vector of log iovecs for the given dquot log item. - */ -STATIC void -xfs_qm_dquot_logitem_format( - struct xfs_log_item *lip, - struct xfs_log_iovec *logvec) -{ - struct xfs_dq_logitem *qlip = DQUOT_ITEM(lip); - - logvec->i_addr = &qlip->qli_format; - logvec->i_len = sizeof(xfs_dq_logformat_t); - logvec->i_type = XLOG_REG_TYPE_QFORMAT; - logvec++; - logvec->i_addr = &qlip->qli_dquot->q_core; - logvec->i_len = sizeof(xfs_disk_dquot_t); - logvec->i_type = XLOG_REG_TYPE_DQUOT; - - ASSERT(2 == lip->li_desc->lid_size); - qlip->qli_format.qlf_size = 2; - -} - -/* - * Increment the pin count of the given dquot. - */ -STATIC void -xfs_qm_dquot_logitem_pin( - struct xfs_log_item *lip) -{ - struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - atomic_inc(&dqp->q_pincount); -} - -/* - * Decrement the pin count of the given dquot, and wake up - * anyone in xfs_dqwait_unpin() if the count goes to 0. The - * dquot must have been previously pinned with a call to - * xfs_qm_dquot_logitem_pin(). - */ -STATIC void -xfs_qm_dquot_logitem_unpin( - struct xfs_log_item *lip, - int remove) -{ - struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; - - ASSERT(atomic_read(&dqp->q_pincount) > 0); - if (atomic_dec_and_test(&dqp->q_pincount)) - wake_up(&dqp->q_pinwait); -} - -/* - * Given the logitem, this writes the corresponding dquot entry to disk - * asynchronously. This is called with the dquot entry securely locked; - * we simply get xfs_qm_dqflush() to do the work, and unlock the dquot - * at the end. - */ -STATIC void -xfs_qm_dquot_logitem_push( - struct xfs_log_item *lip) -{ - struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; - int error; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(!completion_done(&dqp->q_flush)); - - /* - * Since we were able to lock the dquot's flush lock and - * we found it on the AIL, the dquot must be dirty. This - * is because the dquot is removed from the AIL while still - * holding the flush lock in xfs_dqflush_done(). Thus, if - * we found it in the AIL and were able to obtain the flush - * lock without sleeping, then there must not have been - * anyone in the process of flushing the dquot. - */ - error = xfs_qm_dqflush(dqp, 0); - if (error) - xfs_warn(dqp->q_mount, "%s: push error %d on dqp %p", - __func__, error, dqp); - xfs_dqunlock(dqp); -} - -STATIC xfs_lsn_t -xfs_qm_dquot_logitem_committed( - struct xfs_log_item *lip, - xfs_lsn_t lsn) -{ - /* - * We always re-log the entire dquot when it becomes dirty, - * so, the latest copy _is_ the only one that matters. - */ - return lsn; -} - -/* - * This is called to wait for the given dquot to be unpinned. - * Most of these pin/unpin routines are plagiarized from inode code. - */ -void -xfs_qm_dqunpin_wait( - struct xfs_dquot *dqp) -{ - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - if (atomic_read(&dqp->q_pincount) == 0) - return; - - /* - * Give the log a push so we don't wait here too long. - */ - xfs_log_force(dqp->q_mount, 0); - wait_event(dqp->q_pinwait, (atomic_read(&dqp->q_pincount) == 0)); -} - -/* - * This is called when IOP_TRYLOCK returns XFS_ITEM_PUSHBUF to indicate that - * the dquot is locked by us, but the flush lock isn't. So, here we are - * going to see if the relevant dquot buffer is incore, waiting on DELWRI. - * If so, we want to push it out to help us take this item off the AIL as soon - * as possible. - * - * We must not be holding the AIL lock at this point. Calling incore() to - * search the buffer cache can be a time consuming thing, and AIL lock is a - * spinlock. - */ -STATIC void -xfs_qm_dquot_logitem_pushbuf( - struct xfs_log_item *lip) -{ - struct xfs_dq_logitem *qlip = DQUOT_ITEM(lip); - struct xfs_dquot *dqp = qlip->qli_dquot; - struct xfs_buf *bp; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - - /* - * If flushlock isn't locked anymore, chances are that the - * inode flush completed and the inode was taken off the AIL. - * So, just get out. - */ - if (completion_done(&dqp->q_flush) || - !(lip->li_flags & XFS_LI_IN_AIL)) { - xfs_dqunlock(dqp); - return; - } - - bp = xfs_incore(dqp->q_mount->m_ddev_targp, qlip->qli_format.qlf_blkno, - dqp->q_mount->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK); - xfs_dqunlock(dqp); - if (!bp) - return; - if (XFS_BUF_ISDELAYWRITE(bp)) - xfs_buf_delwri_promote(bp); - xfs_buf_relse(bp); -} - -/* - * This is called to attempt to lock the dquot associated with this - * dquot log item. Don't sleep on the dquot lock or the flush lock. - * If the flush lock is already held, indicating that the dquot has - * been or is in the process of being flushed, then see if we can - * find the dquot's buffer in the buffer cache without sleeping. If - * we can and it is marked delayed write, then we want to send it out. - * We delay doing so until the push routine, though, to avoid sleeping - * in any device strategy routines. - */ -STATIC uint -xfs_qm_dquot_logitem_trylock( - struct xfs_log_item *lip) -{ - struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; - - if (atomic_read(&dqp->q_pincount) > 0) - return XFS_ITEM_PINNED; - - if (!xfs_qm_dqlock_nowait(dqp)) - return XFS_ITEM_LOCKED; - - if (!xfs_dqflock_nowait(dqp)) { - /* - * dquot has already been flushed to the backing buffer, - * leave it locked, pushbuf routine will unlock it. - */ - return XFS_ITEM_PUSHBUF; - } - - ASSERT(lip->li_flags & XFS_LI_IN_AIL); - return XFS_ITEM_SUCCESS; -} - -/* - * Unlock the dquot associated with the log item. - * Clear the fields of the dquot and dquot log item that - * are specific to the current transaction. If the - * hold flags is set, do not unlock the dquot. - */ -STATIC void -xfs_qm_dquot_logitem_unlock( - struct xfs_log_item *lip) -{ - struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - - /* - * Clear the transaction pointer in the dquot - */ - dqp->q_transp = NULL; - - /* - * dquots are never 'held' from getting unlocked at the end of - * a transaction. Their locking and unlocking is hidden inside the - * transaction layer, within trans_commit. Hence, no LI_HOLD flag - * for the logitem. - */ - xfs_dqunlock(dqp); -} - -/* - * this needs to stamp an lsn into the dquot, I think. - * rpc's that look at user dquot's would then have to - * push on the dependency recorded in the dquot - */ -STATIC void -xfs_qm_dquot_logitem_committing( - struct xfs_log_item *lip, - xfs_lsn_t lsn) -{ -} - -/* - * This is the ops vector for dquots - */ -static struct xfs_item_ops xfs_dquot_item_ops = { - .iop_size = xfs_qm_dquot_logitem_size, - .iop_format = xfs_qm_dquot_logitem_format, - .iop_pin = xfs_qm_dquot_logitem_pin, - .iop_unpin = xfs_qm_dquot_logitem_unpin, - .iop_trylock = xfs_qm_dquot_logitem_trylock, - .iop_unlock = xfs_qm_dquot_logitem_unlock, - .iop_committed = xfs_qm_dquot_logitem_committed, - .iop_push = xfs_qm_dquot_logitem_push, - .iop_pushbuf = xfs_qm_dquot_logitem_pushbuf, - .iop_committing = xfs_qm_dquot_logitem_committing -}; - -/* - * Initialize the dquot log item for a newly allocated dquot. - * The dquot isn't locked at this point, but it isn't on any of the lists - * either, so we don't care. - */ -void -xfs_qm_dquot_logitem_init( - struct xfs_dquot *dqp) -{ - struct xfs_dq_logitem *lp = &dqp->q_logitem; - - xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT, - &xfs_dquot_item_ops); - lp->qli_dquot = dqp; - lp->qli_format.qlf_type = XFS_LI_DQUOT; - lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id); - lp->qli_format.qlf_blkno = dqp->q_blkno; - lp->qli_format.qlf_len = 1; - /* - * This is just the offset of this dquot within its buffer - * (which is currently 1 FSB and probably won't change). - * Hence 32 bits for this offset should be just fine. - * Alternatively, we can store (bufoffset / sizeof(xfs_dqblk_t)) - * here, and recompute it at recovery time. - */ - lp->qli_format.qlf_boffset = (__uint32_t)dqp->q_bufoffset; -} - -/*------------------ QUOTAOFF LOG ITEMS -------------------*/ - -static inline struct xfs_qoff_logitem *QOFF_ITEM(struct xfs_log_item *lip) -{ - return container_of(lip, struct xfs_qoff_logitem, qql_item); -} - - -/* - * This returns the number of iovecs needed to log the given quotaoff item. - * We only need 1 iovec for an quotaoff item. It just logs the - * quotaoff_log_format structure. - */ -STATIC uint -xfs_qm_qoff_logitem_size( - struct xfs_log_item *lip) -{ - return 1; -} - -/* - * This is called to fill in the vector of log iovecs for the - * given quotaoff log item. We use only 1 iovec, and we point that - * at the quotaoff_log_format structure embedded in the quotaoff item. - * It is at this point that we assert that all of the extent - * slots in the quotaoff item have been filled. - */ -STATIC void -xfs_qm_qoff_logitem_format( - struct xfs_log_item *lip, - struct xfs_log_iovec *log_vector) -{ - struct xfs_qoff_logitem *qflip = QOFF_ITEM(lip); - - ASSERT(qflip->qql_format.qf_type == XFS_LI_QUOTAOFF); - - log_vector->i_addr = &qflip->qql_format; - log_vector->i_len = sizeof(xfs_qoff_logitem_t); - log_vector->i_type = XLOG_REG_TYPE_QUOTAOFF; - qflip->qql_format.qf_size = 1; -} - -/* - * Pinning has no meaning for an quotaoff item, so just return. - */ -STATIC void -xfs_qm_qoff_logitem_pin( - struct xfs_log_item *lip) -{ -} - -/* - * Since pinning has no meaning for an quotaoff item, unpinning does - * not either. - */ -STATIC void -xfs_qm_qoff_logitem_unpin( - struct xfs_log_item *lip, - int remove) -{ -} - -/* - * Quotaoff items have no locking, so just return success. - */ -STATIC uint -xfs_qm_qoff_logitem_trylock( - struct xfs_log_item *lip) -{ - return XFS_ITEM_LOCKED; -} - -/* - * Quotaoff items have no locking or pushing, so return failure - * so that the caller doesn't bother with us. - */ -STATIC void -xfs_qm_qoff_logitem_unlock( - struct xfs_log_item *lip) -{ -} - -/* - * The quotaoff-start-item is logged only once and cannot be moved in the log, - * so simply return the lsn at which it's been logged. - */ -STATIC xfs_lsn_t -xfs_qm_qoff_logitem_committed( - struct xfs_log_item *lip, - xfs_lsn_t lsn) -{ - return lsn; -} - -/* - * There isn't much you can do to push on an quotaoff item. It is simply - * stuck waiting for the log to be flushed to disk. - */ -STATIC void -xfs_qm_qoff_logitem_push( - struct xfs_log_item *lip) -{ -} - - -STATIC xfs_lsn_t -xfs_qm_qoffend_logitem_committed( - struct xfs_log_item *lip, - xfs_lsn_t lsn) -{ - struct xfs_qoff_logitem *qfe = QOFF_ITEM(lip); - struct xfs_qoff_logitem *qfs = qfe->qql_start_lip; - struct xfs_ail *ailp = qfs->qql_item.li_ailp; - - /* - * Delete the qoff-start logitem from the AIL. - * xfs_trans_ail_delete() drops the AIL lock. - */ - spin_lock(&ailp->xa_lock); - xfs_trans_ail_delete(ailp, (xfs_log_item_t *)qfs); - - kmem_free(qfs); - kmem_free(qfe); - return (xfs_lsn_t)-1; -} - -/* - * XXX rcc - don't know quite what to do with this. I think we can - * just ignore it. The only time that isn't the case is if we allow - * the client to somehow see that quotas have been turned off in which - * we can't allow that to get back until the quotaoff hits the disk. - * So how would that happen? Also, do we need different routines for - * quotaoff start and quotaoff end? I suspect the answer is yes but - * to be sure, I need to look at the recovery code and see how quota off - * recovery is handled (do we roll forward or back or do something else). - * If we roll forwards or backwards, then we need two separate routines, - * one that does nothing and one that stamps in the lsn that matters - * (truly makes the quotaoff irrevocable). If we do something else, - * then maybe we don't need two. - */ -STATIC void -xfs_qm_qoff_logitem_committing( - struct xfs_log_item *lip, - xfs_lsn_t commit_lsn) -{ -} - -static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = { - .iop_size = xfs_qm_qoff_logitem_size, - .iop_format = xfs_qm_qoff_logitem_format, - .iop_pin = xfs_qm_qoff_logitem_pin, - .iop_unpin = xfs_qm_qoff_logitem_unpin, - .iop_trylock = xfs_qm_qoff_logitem_trylock, - .iop_unlock = xfs_qm_qoff_logitem_unlock, - .iop_committed = xfs_qm_qoffend_logitem_committed, - .iop_push = xfs_qm_qoff_logitem_push, - .iop_committing = xfs_qm_qoff_logitem_committing -}; - -/* - * This is the ops vector shared by all quotaoff-start log items. - */ -static struct xfs_item_ops xfs_qm_qoff_logitem_ops = { - .iop_size = xfs_qm_qoff_logitem_size, - .iop_format = xfs_qm_qoff_logitem_format, - .iop_pin = xfs_qm_qoff_logitem_pin, - .iop_unpin = xfs_qm_qoff_logitem_unpin, - .iop_trylock = xfs_qm_qoff_logitem_trylock, - .iop_unlock = xfs_qm_qoff_logitem_unlock, - .iop_committed = xfs_qm_qoff_logitem_committed, - .iop_push = xfs_qm_qoff_logitem_push, - .iop_committing = xfs_qm_qoff_logitem_committing -}; - -/* - * Allocate and initialize an quotaoff item of the correct quota type(s). - */ -struct xfs_qoff_logitem * -xfs_qm_qoff_logitem_init( - struct xfs_mount *mp, - struct xfs_qoff_logitem *start, - uint flags) -{ - struct xfs_qoff_logitem *qf; - - qf = kmem_zalloc(sizeof(struct xfs_qoff_logitem), KM_SLEEP); - - xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ? - &xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops); - qf->qql_item.li_mountp = mp; - qf->qql_format.qf_type = XFS_LI_QUOTAOFF; - qf->qql_format.qf_flags = flags; - qf->qql_start_lip = start; - return qf; -} diff --git a/fs/xfs/quota/xfs_dquot_item.h b/fs/xfs/quota/xfs_dquot_item.h deleted file mode 100644 index 5acae2ada70b..000000000000 --- a/fs/xfs/quota/xfs_dquot_item.h +++ /dev/null @@ -1,48 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_DQUOT_ITEM_H__ -#define __XFS_DQUOT_ITEM_H__ - -struct xfs_dquot; -struct xfs_trans; -struct xfs_mount; -struct xfs_qoff_logitem; - -typedef struct xfs_dq_logitem { - xfs_log_item_t qli_item; /* common portion */ - struct xfs_dquot *qli_dquot; /* dquot ptr */ - xfs_lsn_t qli_flush_lsn; /* lsn at last flush */ - xfs_dq_logformat_t qli_format; /* logged structure */ -} xfs_dq_logitem_t; - -typedef struct xfs_qoff_logitem { - xfs_log_item_t qql_item; /* common portion */ - struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */ - xfs_qoff_logformat_t qql_format; /* logged structure */ -} xfs_qoff_logitem_t; - - -extern void xfs_qm_dquot_logitem_init(struct xfs_dquot *); -extern xfs_qoff_logitem_t *xfs_qm_qoff_logitem_init(struct xfs_mount *, - struct xfs_qoff_logitem *, uint); -extern xfs_qoff_logitem_t *xfs_trans_get_qoff_item(struct xfs_trans *, - struct xfs_qoff_logitem *, uint); -extern void xfs_trans_log_quotaoff_item(struct xfs_trans *, - struct xfs_qoff_logitem *); - -#endif /* __XFS_DQUOT_ITEM_H__ */ diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c deleted file mode 100644 index 9a0aa76facdf..000000000000 --- a/fs/xfs/quota/xfs_qm.c +++ /dev/null @@ -1,2416 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_ialloc.h" -#include "xfs_itable.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_bmap.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_trans_space.h" -#include "xfs_utils.h" -#include "xfs_qm.h" -#include "xfs_trace.h" - -/* - * The global quota manager. There is only one of these for the entire - * system, _not_ one per file system. XQM keeps track of the overall - * quota functionality, including maintaining the freelist and hash - * tables of dquots. - */ -struct mutex xfs_Gqm_lock; -struct xfs_qm *xfs_Gqm; -uint ndquot; - -kmem_zone_t *qm_dqzone; -kmem_zone_t *qm_dqtrxzone; - -STATIC void xfs_qm_list_init(xfs_dqlist_t *, char *, int); -STATIC void xfs_qm_list_destroy(xfs_dqlist_t *); - -STATIC int xfs_qm_init_quotainos(xfs_mount_t *); -STATIC int xfs_qm_init_quotainfo(xfs_mount_t *); -STATIC int xfs_qm_shake(struct shrinker *, struct shrink_control *); - -static struct shrinker xfs_qm_shaker = { - .shrink = xfs_qm_shake, - .seeks = DEFAULT_SEEKS, -}; - -/* - * Initialize the XQM structure. - * Note that there is not one quota manager per file system. - */ -STATIC struct xfs_qm * -xfs_Gqm_init(void) -{ - xfs_dqhash_t *udqhash, *gdqhash; - xfs_qm_t *xqm; - size_t hsize; - uint i; - - /* - * Initialize the dquot hash tables. - */ - udqhash = kmem_zalloc_greedy(&hsize, - XFS_QM_HASHSIZE_LOW * sizeof(xfs_dqhash_t), - XFS_QM_HASHSIZE_HIGH * sizeof(xfs_dqhash_t)); - if (!udqhash) - goto out; - - gdqhash = kmem_zalloc_large(hsize); - if (!gdqhash) - goto out_free_udqhash; - - hsize /= sizeof(xfs_dqhash_t); - ndquot = hsize << 8; - - xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP); - xqm->qm_dqhashmask = hsize - 1; - xqm->qm_usr_dqhtable = udqhash; - xqm->qm_grp_dqhtable = gdqhash; - ASSERT(xqm->qm_usr_dqhtable != NULL); - ASSERT(xqm->qm_grp_dqhtable != NULL); - - for (i = 0; i < hsize; i++) { - xfs_qm_list_init(&(xqm->qm_usr_dqhtable[i]), "uxdqh", i); - xfs_qm_list_init(&(xqm->qm_grp_dqhtable[i]), "gxdqh", i); - } - - /* - * Freelist of all dquots of all file systems - */ - INIT_LIST_HEAD(&xqm->qm_dqfrlist); - xqm->qm_dqfrlist_cnt = 0; - mutex_init(&xqm->qm_dqfrlist_lock); - - /* - * dquot zone. we register our own low-memory callback. - */ - if (!qm_dqzone) { - xqm->qm_dqzone = kmem_zone_init(sizeof(xfs_dquot_t), - "xfs_dquots"); - qm_dqzone = xqm->qm_dqzone; - } else - xqm->qm_dqzone = qm_dqzone; - - register_shrinker(&xfs_qm_shaker); - - /* - * The t_dqinfo portion of transactions. - */ - if (!qm_dqtrxzone) { - xqm->qm_dqtrxzone = kmem_zone_init(sizeof(xfs_dquot_acct_t), - "xfs_dqtrx"); - qm_dqtrxzone = xqm->qm_dqtrxzone; - } else - xqm->qm_dqtrxzone = qm_dqtrxzone; - - atomic_set(&xqm->qm_totaldquots, 0); - xqm->qm_dqfree_ratio = XFS_QM_DQFREE_RATIO; - xqm->qm_nrefs = 0; - return xqm; - - out_free_udqhash: - kmem_free_large(udqhash); - out: - return NULL; -} - -/* - * Destroy the global quota manager when its reference count goes to zero. - */ -STATIC void -xfs_qm_destroy( - struct xfs_qm *xqm) -{ - struct xfs_dquot *dqp, *n; - int hsize, i; - - ASSERT(xqm != NULL); - ASSERT(xqm->qm_nrefs == 0); - unregister_shrinker(&xfs_qm_shaker); - hsize = xqm->qm_dqhashmask + 1; - for (i = 0; i < hsize; i++) { - xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i])); - xfs_qm_list_destroy(&(xqm->qm_grp_dqhtable[i])); - } - kmem_free_large(xqm->qm_usr_dqhtable); - kmem_free_large(xqm->qm_grp_dqhtable); - xqm->qm_usr_dqhtable = NULL; - xqm->qm_grp_dqhtable = NULL; - xqm->qm_dqhashmask = 0; - - /* frlist cleanup */ - mutex_lock(&xqm->qm_dqfrlist_lock); - list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) { - xfs_dqlock(dqp); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - xfs_dqunlock(dqp); - xfs_qm_dqdestroy(dqp); - } - mutex_unlock(&xqm->qm_dqfrlist_lock); - mutex_destroy(&xqm->qm_dqfrlist_lock); - kmem_free(xqm); -} - -/* - * Called at mount time to let XQM know that another file system is - * starting quotas. This isn't crucial information as the individual mount - * structures are pretty independent, but it helps the XQM keep a - * global view of what's going on. - */ -/* ARGSUSED */ -STATIC int -xfs_qm_hold_quotafs_ref( - struct xfs_mount *mp) -{ - /* - * Need to lock the xfs_Gqm structure for things like this. For example, - * the structure could disappear between the entry to this routine and - * a HOLD operation if not locked. - */ - mutex_lock(&xfs_Gqm_lock); - - if (!xfs_Gqm) { - xfs_Gqm = xfs_Gqm_init(); - if (!xfs_Gqm) { - mutex_unlock(&xfs_Gqm_lock); - return ENOMEM; - } - } - - /* - * We can keep a list of all filesystems with quotas mounted for - * debugging and statistical purposes, but ... - * Just take a reference and get out. - */ - xfs_Gqm->qm_nrefs++; - mutex_unlock(&xfs_Gqm_lock); - - return 0; -} - - -/* - * Release the reference that a filesystem took at mount time, - * so that we know when we need to destroy the entire quota manager. - */ -/* ARGSUSED */ -STATIC void -xfs_qm_rele_quotafs_ref( - struct xfs_mount *mp) -{ - xfs_dquot_t *dqp, *n; - - ASSERT(xfs_Gqm); - ASSERT(xfs_Gqm->qm_nrefs > 0); - - /* - * Go thru the freelist and destroy all inactive dquots. - */ - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - - list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) { - xfs_dqlock(dqp); - if (dqp->dq_flags & XFS_DQ_INACTIVE) { - ASSERT(dqp->q_mount == NULL); - ASSERT(! XFS_DQ_IS_DIRTY(dqp)); - ASSERT(list_empty(&dqp->q_hashlist)); - ASSERT(list_empty(&dqp->q_mplist)); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - xfs_dqunlock(dqp); - xfs_qm_dqdestroy(dqp); - } else { - xfs_dqunlock(dqp); - } - } - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - - /* - * Destroy the entire XQM. If somebody mounts with quotaon, this'll - * be restarted. - */ - mutex_lock(&xfs_Gqm_lock); - if (--xfs_Gqm->qm_nrefs == 0) { - xfs_qm_destroy(xfs_Gqm); - xfs_Gqm = NULL; - } - mutex_unlock(&xfs_Gqm_lock); -} - -/* - * Just destroy the quotainfo structure. - */ -void -xfs_qm_unmount( - struct xfs_mount *mp) -{ - if (mp->m_quotainfo) { - xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL); - xfs_qm_destroy_quotainfo(mp); - } -} - - -/* - * This is called from xfs_mountfs to start quotas and initialize all - * necessary data structures like quotainfo. This is also responsible for - * running a quotacheck as necessary. We are guaranteed that the superblock - * is consistently read in at this point. - * - * If we fail here, the mount will continue with quota turned off. We don't - * need to inidicate success or failure at all. - */ -void -xfs_qm_mount_quotas( - xfs_mount_t *mp) -{ - int error = 0; - uint sbf; - - /* - * If quotas on realtime volumes is not supported, we disable - * quotas immediately. - */ - if (mp->m_sb.sb_rextents) { - xfs_notice(mp, "Cannot turn on quotas for realtime filesystem"); - mp->m_qflags = 0; - goto write_changes; - } - - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - /* - * Allocate the quotainfo structure inside the mount struct, and - * create quotainode(s), and change/rev superblock if necessary. - */ - error = xfs_qm_init_quotainfo(mp); - if (error) { - /* - * We must turn off quotas. - */ - ASSERT(mp->m_quotainfo == NULL); - mp->m_qflags = 0; - goto write_changes; - } - /* - * If any of the quotas are not consistent, do a quotacheck. - */ - if (XFS_QM_NEED_QUOTACHECK(mp)) { - error = xfs_qm_quotacheck(mp); - if (error) { - /* Quotacheck failed and disabled quotas. */ - return; - } - } - /* - * If one type of quotas is off, then it will lose its - * quotachecked status, since we won't be doing accounting for - * that type anymore. - */ - if (!XFS_IS_UQUOTA_ON(mp)) - mp->m_qflags &= ~XFS_UQUOTA_CHKD; - if (!(XFS_IS_GQUOTA_ON(mp) || XFS_IS_PQUOTA_ON(mp))) - mp->m_qflags &= ~XFS_OQUOTA_CHKD; - - write_changes: - /* - * We actually don't have to acquire the m_sb_lock at all. - * This can only be called from mount, and that's single threaded. XXX - */ - spin_lock(&mp->m_sb_lock); - sbf = mp->m_sb.sb_qflags; - mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL; - spin_unlock(&mp->m_sb_lock); - - if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) { - if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) { - /* - * We could only have been turning quotas off. - * We aren't in very good shape actually because - * the incore structures are convinced that quotas are - * off, but the on disk superblock doesn't know that ! - */ - ASSERT(!(XFS_IS_QUOTA_RUNNING(mp))); - xfs_alert(mp, "%s: Superblock update failed!", - __func__); - } - } - - if (error) { - xfs_warn(mp, "Failed to initialize disk quotas."); - return; - } -} - -/* - * Called from the vfsops layer. - */ -void -xfs_qm_unmount_quotas( - xfs_mount_t *mp) -{ - /* - * Release the dquots that root inode, et al might be holding, - * before we flush quotas and blow away the quotainfo structure. - */ - ASSERT(mp->m_rootip); - xfs_qm_dqdetach(mp->m_rootip); - if (mp->m_rbmip) - xfs_qm_dqdetach(mp->m_rbmip); - if (mp->m_rsumip) - xfs_qm_dqdetach(mp->m_rsumip); - - /* - * Release the quota inodes. - */ - if (mp->m_quotainfo) { - if (mp->m_quotainfo->qi_uquotaip) { - IRELE(mp->m_quotainfo->qi_uquotaip); - mp->m_quotainfo->qi_uquotaip = NULL; - } - if (mp->m_quotainfo->qi_gquotaip) { - IRELE(mp->m_quotainfo->qi_gquotaip); - mp->m_quotainfo->qi_gquotaip = NULL; - } - } -} - -/* - * Flush all dquots of the given file system to disk. The dquots are - * _not_ purged from memory here, just their data written to disk. - */ -STATIC int -xfs_qm_dqflush_all( - struct xfs_mount *mp, - int sync_mode) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - int recl; - struct xfs_dquot *dqp; - int error; - - if (!q) - return 0; -again: - mutex_lock(&q->qi_dqlist_lock); - list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { - xfs_dqlock(dqp); - if (! XFS_DQ_IS_DIRTY(dqp)) { - xfs_dqunlock(dqp); - continue; - } - - /* XXX a sentinel would be better */ - recl = q->qi_dqreclaims; - if (!xfs_dqflock_nowait(dqp)) { - /* - * If we can't grab the flush lock then check - * to see if the dquot has been flushed delayed - * write. If so, grab its buffer and send it - * out immediately. We'll be able to acquire - * the flush lock when the I/O completes. - */ - xfs_qm_dqflock_pushbuf_wait(dqp); - } - /* - * Let go of the mplist lock. We don't want to hold it - * across a disk write. - */ - mutex_unlock(&q->qi_dqlist_lock); - error = xfs_qm_dqflush(dqp, sync_mode); - xfs_dqunlock(dqp); - if (error) - return error; - - mutex_lock(&q->qi_dqlist_lock); - if (recl != q->qi_dqreclaims) { - mutex_unlock(&q->qi_dqlist_lock); - /* XXX restart limit */ - goto again; - } - } - - mutex_unlock(&q->qi_dqlist_lock); - /* return ! busy */ - return 0; -} -/* - * Release the group dquot pointers the user dquots may be - * carrying around as a hint. mplist is locked on entry and exit. - */ -STATIC void -xfs_qm_detach_gdquots( - struct xfs_mount *mp) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - struct xfs_dquot *dqp, *gdqp; - int nrecl; - - again: - ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); - list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { - xfs_dqlock(dqp); - if ((gdqp = dqp->q_gdquot)) { - xfs_dqlock(gdqp); - dqp->q_gdquot = NULL; - } - xfs_dqunlock(dqp); - - if (gdqp) { - /* - * Can't hold the mplist lock across a dqput. - * XXXmust convert to marker based iterations here. - */ - nrecl = q->qi_dqreclaims; - mutex_unlock(&q->qi_dqlist_lock); - xfs_qm_dqput(gdqp); - - mutex_lock(&q->qi_dqlist_lock); - if (nrecl != q->qi_dqreclaims) - goto again; - } - } -} - -/* - * Go through all the incore dquots of this file system and take them - * off the mplist and hashlist, if the dquot type matches the dqtype - * parameter. This is used when turning off quota accounting for - * users and/or groups, as well as when the filesystem is unmounting. - */ -STATIC int -xfs_qm_dqpurge_int( - struct xfs_mount *mp, - uint flags) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - struct xfs_dquot *dqp, *n; - uint dqtype; - int nrecl; - int nmisses; - - if (!q) - return 0; - - dqtype = (flags & XFS_QMOPT_UQUOTA) ? XFS_DQ_USER : 0; - dqtype |= (flags & XFS_QMOPT_PQUOTA) ? XFS_DQ_PROJ : 0; - dqtype |= (flags & XFS_QMOPT_GQUOTA) ? XFS_DQ_GROUP : 0; - - mutex_lock(&q->qi_dqlist_lock); - - /* - * In the first pass through all incore dquots of this filesystem, - * we release the group dquot pointers the user dquots may be - * carrying around as a hint. We need to do this irrespective of - * what's being turned off. - */ - xfs_qm_detach_gdquots(mp); - - again: - nmisses = 0; - ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); - /* - * Try to get rid of all of the unwanted dquots. The idea is to - * get them off mplist and hashlist, but leave them on freelist. - */ - list_for_each_entry_safe(dqp, n, &q->qi_dqlist, q_mplist) { - /* - * It's OK to look at the type without taking dqlock here. - * We're holding the mplist lock here, and that's needed for - * a dqreclaim. - */ - if ((dqp->dq_flags & dqtype) == 0) - continue; - - if (!mutex_trylock(&dqp->q_hash->qh_lock)) { - nrecl = q->qi_dqreclaims; - mutex_unlock(&q->qi_dqlist_lock); - mutex_lock(&dqp->q_hash->qh_lock); - mutex_lock(&q->qi_dqlist_lock); - - /* - * XXXTheoretically, we can get into a very long - * ping pong game here. - * No one can be adding dquots to the mplist at - * this point, but somebody might be taking things off. - */ - if (nrecl != q->qi_dqreclaims) { - mutex_unlock(&dqp->q_hash->qh_lock); - goto again; - } - } - - /* - * Take the dquot off the mplist and hashlist. It may remain on - * freelist in INACTIVE state. - */ - nmisses += xfs_qm_dqpurge(dqp); - } - mutex_unlock(&q->qi_dqlist_lock); - return nmisses; -} - -int -xfs_qm_dqpurge_all( - xfs_mount_t *mp, - uint flags) -{ - int ndquots; - - /* - * Purge the dquot cache. - * None of the dquots should really be busy at this point. - */ - if (mp->m_quotainfo) { - while ((ndquots = xfs_qm_dqpurge_int(mp, flags))) { - delay(ndquots * 10); - } - } - return 0; -} - -STATIC int -xfs_qm_dqattach_one( - xfs_inode_t *ip, - xfs_dqid_t id, - uint type, - uint doalloc, - xfs_dquot_t *udqhint, /* hint */ - xfs_dquot_t **IO_idqpp) -{ - xfs_dquot_t *dqp; - int error; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - error = 0; - - /* - * See if we already have it in the inode itself. IO_idqpp is - * &i_udquot or &i_gdquot. This made the code look weird, but - * made the logic a lot simpler. - */ - dqp = *IO_idqpp; - if (dqp) { - trace_xfs_dqattach_found(dqp); - return 0; - } - - /* - * udqhint is the i_udquot field in inode, and is non-NULL only - * when the type arg is group/project. Its purpose is to save a - * lookup by dqid (xfs_qm_dqget) by caching a group dquot inside - * the user dquot. - */ - if (udqhint) { - ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ); - xfs_dqlock(udqhint); - - /* - * No need to take dqlock to look at the id. - * - * The ID can't change until it gets reclaimed, and it won't - * be reclaimed as long as we have a ref from inode and we - * hold the ilock. - */ - dqp = udqhint->q_gdquot; - if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) { - xfs_dqlock(dqp); - XFS_DQHOLD(dqp); - ASSERT(*IO_idqpp == NULL); - *IO_idqpp = dqp; - - xfs_dqunlock(dqp); - xfs_dqunlock(udqhint); - return 0; - } - - /* - * We can't hold a dquot lock when we call the dqget code. - * We'll deadlock in no time, because of (not conforming to) - * lock ordering - the inodelock comes before any dquot lock, - * and we may drop and reacquire the ilock in xfs_qm_dqget(). - */ - xfs_dqunlock(udqhint); - } - - /* - * Find the dquot from somewhere. This bumps the - * reference count of dquot and returns it locked. - * This can return ENOENT if dquot didn't exist on - * disk and we didn't ask it to allocate; - * ESRCH if quotas got turned off suddenly. - */ - error = xfs_qm_dqget(ip->i_mount, ip, id, type, XFS_QMOPT_DOWARN, &dqp); - if (error) - return error; - - trace_xfs_dqattach_get(dqp); - - /* - * dqget may have dropped and re-acquired the ilock, but it guarantees - * that the dquot returned is the one that should go in the inode. - */ - *IO_idqpp = dqp; - xfs_dqunlock(dqp); - return 0; -} - - -/* - * Given a udquot and gdquot, attach a ptr to the group dquot in the - * udquot as a hint for future lookups. The idea sounds simple, but the - * execution isn't, because the udquot might have a group dquot attached - * already and getting rid of that gets us into lock ordering constraints. - * The process is complicated more by the fact that the dquots may or may not - * be locked on entry. - */ -STATIC void -xfs_qm_dqattach_grouphint( - xfs_dquot_t *udq, - xfs_dquot_t *gdq) -{ - xfs_dquot_t *tmp; - - xfs_dqlock(udq); - - if ((tmp = udq->q_gdquot)) { - if (tmp == gdq) { - xfs_dqunlock(udq); - return; - } - - udq->q_gdquot = NULL; - /* - * We can't keep any dqlocks when calling dqrele, - * because the freelist lock comes before dqlocks. - */ - xfs_dqunlock(udq); - /* - * we took a hard reference once upon a time in dqget, - * so give it back when the udquot no longer points at it - * dqput() does the unlocking of the dquot. - */ - xfs_qm_dqrele(tmp); - - xfs_dqlock(udq); - xfs_dqlock(gdq); - - } else { - ASSERT(XFS_DQ_IS_LOCKED(udq)); - xfs_dqlock(gdq); - } - - ASSERT(XFS_DQ_IS_LOCKED(udq)); - ASSERT(XFS_DQ_IS_LOCKED(gdq)); - /* - * Somebody could have attached a gdquot here, - * when we dropped the uqlock. If so, just do nothing. - */ - if (udq->q_gdquot == NULL) { - XFS_DQHOLD(gdq); - udq->q_gdquot = gdq; - } - - xfs_dqunlock(gdq); - xfs_dqunlock(udq); -} - - -/* - * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON - * into account. - * If XFS_QMOPT_DQALLOC, the dquot(s) will be allocated if needed. - * Inode may get unlocked and relocked in here, and the caller must deal with - * the consequences. - */ -int -xfs_qm_dqattach_locked( - xfs_inode_t *ip, - uint flags) -{ - xfs_mount_t *mp = ip->i_mount; - uint nquotas = 0; - int error = 0; - - if (!XFS_IS_QUOTA_RUNNING(mp) || - !XFS_IS_QUOTA_ON(mp) || - !XFS_NOT_DQATTACHED(mp, ip) || - ip->i_ino == mp->m_sb.sb_uquotino || - ip->i_ino == mp->m_sb.sb_gquotino) - return 0; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - - if (XFS_IS_UQUOTA_ON(mp)) { - error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER, - flags & XFS_QMOPT_DQALLOC, - NULL, &ip->i_udquot); - if (error) - goto done; - nquotas++; - } - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - if (XFS_IS_OQUOTA_ON(mp)) { - error = XFS_IS_GQUOTA_ON(mp) ? - xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, - flags & XFS_QMOPT_DQALLOC, - ip->i_udquot, &ip->i_gdquot) : - xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, - flags & XFS_QMOPT_DQALLOC, - ip->i_udquot, &ip->i_gdquot); - /* - * Don't worry about the udquot that we may have - * attached above. It'll get detached, if not already. - */ - if (error) - goto done; - nquotas++; - } - - /* - * Attach this group quota to the user quota as a hint. - * This WON'T, in general, result in a thrash. - */ - if (nquotas == 2) { - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(ip->i_udquot); - ASSERT(ip->i_gdquot); - - /* - * We may or may not have the i_udquot locked at this point, - * but this check is OK since we don't depend on the i_gdquot to - * be accurate 100% all the time. It is just a hint, and this - * will succeed in general. - */ - if (ip->i_udquot->q_gdquot == ip->i_gdquot) - goto done; - /* - * Attach i_gdquot to the gdquot hint inside the i_udquot. - */ - xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot); - } - - done: -#ifdef DEBUG - if (!error) { - if (XFS_IS_UQUOTA_ON(mp)) - ASSERT(ip->i_udquot); - if (XFS_IS_OQUOTA_ON(mp)) - ASSERT(ip->i_gdquot); - } - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); -#endif - return error; -} - -int -xfs_qm_dqattach( - struct xfs_inode *ip, - uint flags) -{ - int error; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_qm_dqattach_locked(ip, flags); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - - return error; -} - -/* - * Release dquots (and their references) if any. - * The inode should be locked EXCL except when this's called by - * xfs_ireclaim. - */ -void -xfs_qm_dqdetach( - xfs_inode_t *ip) -{ - if (!(ip->i_udquot || ip->i_gdquot)) - return; - - trace_xfs_dquot_dqdetach(ip); - - ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_uquotino); - ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_gquotino); - if (ip->i_udquot) { - xfs_qm_dqrele(ip->i_udquot); - ip->i_udquot = NULL; - } - if (ip->i_gdquot) { - xfs_qm_dqrele(ip->i_gdquot); - ip->i_gdquot = NULL; - } -} - -int -xfs_qm_sync( - struct xfs_mount *mp, - int flags) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - int recl, restarts; - struct xfs_dquot *dqp; - int error; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - - restarts = 0; - - again: - mutex_lock(&q->qi_dqlist_lock); - /* - * dqpurge_all() also takes the mplist lock and iterate thru all dquots - * in quotaoff. However, if the QUOTA_ACTIVE bits are not cleared - * when we have the mplist lock, we know that dquots will be consistent - * as long as we have it locked. - */ - if (!XFS_IS_QUOTA_ON(mp)) { - mutex_unlock(&q->qi_dqlist_lock); - return 0; - } - ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); - list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { - /* - * If this is vfs_sync calling, then skip the dquots that - * don't 'seem' to be dirty. ie. don't acquire dqlock. - * This is very similar to what xfs_sync does with inodes. - */ - if (flags & SYNC_TRYLOCK) { - if (!XFS_DQ_IS_DIRTY(dqp)) - continue; - if (!xfs_qm_dqlock_nowait(dqp)) - continue; - } else { - xfs_dqlock(dqp); - } - - /* - * Now, find out for sure if this dquot is dirty or not. - */ - if (! XFS_DQ_IS_DIRTY(dqp)) { - xfs_dqunlock(dqp); - continue; - } - - /* XXX a sentinel would be better */ - recl = q->qi_dqreclaims; - if (!xfs_dqflock_nowait(dqp)) { - if (flags & SYNC_TRYLOCK) { - xfs_dqunlock(dqp); - continue; - } - /* - * If we can't grab the flush lock then if the caller - * really wanted us to give this our best shot, so - * see if we can give a push to the buffer before we wait - * on the flush lock. At this point, we know that - * even though the dquot is being flushed, - * it has (new) dirty data. - */ - xfs_qm_dqflock_pushbuf_wait(dqp); - } - /* - * Let go of the mplist lock. We don't want to hold it - * across a disk write - */ - mutex_unlock(&q->qi_dqlist_lock); - error = xfs_qm_dqflush(dqp, flags); - xfs_dqunlock(dqp); - if (error && XFS_FORCED_SHUTDOWN(mp)) - return 0; /* Need to prevent umount failure */ - else if (error) - return error; - - mutex_lock(&q->qi_dqlist_lock); - if (recl != q->qi_dqreclaims) { - if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS) - break; - - mutex_unlock(&q->qi_dqlist_lock); - goto again; - } - } - - mutex_unlock(&q->qi_dqlist_lock); - return 0; -} - -/* - * The hash chains and the mplist use the same xfs_dqhash structure as - * their list head, but we can take the mplist qh_lock and one of the - * hash qh_locks at the same time without any problem as they aren't - * related. - */ -static struct lock_class_key xfs_quota_mplist_class; - -/* - * This initializes all the quota information that's kept in the - * mount structure - */ -STATIC int -xfs_qm_init_quotainfo( - xfs_mount_t *mp) -{ - xfs_quotainfo_t *qinf; - int error; - xfs_dquot_t *dqp; - - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - /* - * Tell XQM that we exist as soon as possible. - */ - if ((error = xfs_qm_hold_quotafs_ref(mp))) { - return error; - } - - qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP); - - /* - * See if quotainodes are setup, and if not, allocate them, - * and change the superblock accordingly. - */ - if ((error = xfs_qm_init_quotainos(mp))) { - kmem_free(qinf); - mp->m_quotainfo = NULL; - return error; - } - - INIT_LIST_HEAD(&qinf->qi_dqlist); - mutex_init(&qinf->qi_dqlist_lock); - lockdep_set_class(&qinf->qi_dqlist_lock, &xfs_quota_mplist_class); - - qinf->qi_dqreclaims = 0; - - /* mutex used to serialize quotaoffs */ - mutex_init(&qinf->qi_quotaofflock); - - /* Precalc some constants */ - qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); - ASSERT(qinf->qi_dqchunklen); - qinf->qi_dqperchunk = BBTOB(qinf->qi_dqchunklen); - do_div(qinf->qi_dqperchunk, sizeof(xfs_dqblk_t)); - - mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD); - - /* - * We try to get the limits from the superuser's limits fields. - * This is quite hacky, but it is standard quota practice. - * We look at the USR dquot with id == 0 first, but if user quotas - * are not enabled we goto the GRP dquot with id == 0. - * We don't really care to keep separate default limits for user - * and group quotas, at least not at this point. - */ - error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)0, - XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : - (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP : - XFS_DQ_PROJ), - XFS_QMOPT_DQSUSER|XFS_QMOPT_DOWARN, - &dqp); - if (! error) { - xfs_disk_dquot_t *ddqp = &dqp->q_core; - - /* - * The warnings and timers set the grace period given to - * a user or group before he or she can not perform any - * more writing. If it is zero, a default is used. - */ - qinf->qi_btimelimit = ddqp->d_btimer ? - be32_to_cpu(ddqp->d_btimer) : XFS_QM_BTIMELIMIT; - qinf->qi_itimelimit = ddqp->d_itimer ? - be32_to_cpu(ddqp->d_itimer) : XFS_QM_ITIMELIMIT; - qinf->qi_rtbtimelimit = ddqp->d_rtbtimer ? - be32_to_cpu(ddqp->d_rtbtimer) : XFS_QM_RTBTIMELIMIT; - qinf->qi_bwarnlimit = ddqp->d_bwarns ? - be16_to_cpu(ddqp->d_bwarns) : XFS_QM_BWARNLIMIT; - qinf->qi_iwarnlimit = ddqp->d_iwarns ? - be16_to_cpu(ddqp->d_iwarns) : XFS_QM_IWARNLIMIT; - qinf->qi_rtbwarnlimit = ddqp->d_rtbwarns ? - be16_to_cpu(ddqp->d_rtbwarns) : XFS_QM_RTBWARNLIMIT; - qinf->qi_bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); - qinf->qi_bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit); - qinf->qi_ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); - qinf->qi_isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); - qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); - qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); - - /* - * We sent the XFS_QMOPT_DQSUSER flag to dqget because - * we don't want this dquot cached. We haven't done a - * quotacheck yet, and quotacheck doesn't like incore dquots. - */ - xfs_qm_dqdestroy(dqp); - } else { - qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; - qinf->qi_itimelimit = XFS_QM_ITIMELIMIT; - qinf->qi_rtbtimelimit = XFS_QM_RTBTIMELIMIT; - qinf->qi_bwarnlimit = XFS_QM_BWARNLIMIT; - qinf->qi_iwarnlimit = XFS_QM_IWARNLIMIT; - qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT; - } - - return 0; -} - - -/* - * Gets called when unmounting a filesystem or when all quotas get - * turned off. - * This purges the quota inodes, destroys locks and frees itself. - */ -void -xfs_qm_destroy_quotainfo( - xfs_mount_t *mp) -{ - xfs_quotainfo_t *qi; - - qi = mp->m_quotainfo; - ASSERT(qi != NULL); - ASSERT(xfs_Gqm != NULL); - - /* - * Release the reference that XQM kept, so that we know - * when the XQM structure should be freed. We cannot assume - * that xfs_Gqm is non-null after this point. - */ - xfs_qm_rele_quotafs_ref(mp); - - ASSERT(list_empty(&qi->qi_dqlist)); - mutex_destroy(&qi->qi_dqlist_lock); - - if (qi->qi_uquotaip) { - IRELE(qi->qi_uquotaip); - qi->qi_uquotaip = NULL; /* paranoia */ - } - if (qi->qi_gquotaip) { - IRELE(qi->qi_gquotaip); - qi->qi_gquotaip = NULL; - } - mutex_destroy(&qi->qi_quotaofflock); - kmem_free(qi); - mp->m_quotainfo = NULL; -} - - - -/* ------------------- PRIVATE STATIC FUNCTIONS ----------------------- */ - -/* ARGSUSED */ -STATIC void -xfs_qm_list_init( - xfs_dqlist_t *list, - char *str, - int n) -{ - mutex_init(&list->qh_lock); - INIT_LIST_HEAD(&list->qh_list); - list->qh_version = 0; - list->qh_nelems = 0; -} - -STATIC void -xfs_qm_list_destroy( - xfs_dqlist_t *list) -{ - mutex_destroy(&(list->qh_lock)); -} - -/* - * Create an inode and return with a reference already taken, but unlocked - * This is how we create quota inodes - */ -STATIC int -xfs_qm_qino_alloc( - xfs_mount_t *mp, - xfs_inode_t **ip, - __int64_t sbfields, - uint flags) -{ - xfs_trans_t *tp; - int error; - int committed; - - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE); - if ((error = xfs_trans_reserve(tp, - XFS_QM_QINOCREATE_SPACE_RES(mp), - XFS_CREATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_CREATE_LOG_COUNT))) { - xfs_trans_cancel(tp, 0); - return error; - } - - error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip, &committed); - if (error) { - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT); - return error; - } - - /* - * Make the changes in the superblock, and log those too. - * sbfields arg may contain fields other than *QUOTINO; - * VERSIONNUM for example. - */ - spin_lock(&mp->m_sb_lock); - if (flags & XFS_QMOPT_SBVERSION) { - ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); - ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == - (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); - - xfs_sb_version_addquota(&mp->m_sb); - mp->m_sb.sb_uquotino = NULLFSINO; - mp->m_sb.sb_gquotino = NULLFSINO; - - /* qflags will get updated _after_ quotacheck */ - mp->m_sb.sb_qflags = 0; - } - if (flags & XFS_QMOPT_UQUOTA) - mp->m_sb.sb_uquotino = (*ip)->i_ino; - else - mp->m_sb.sb_gquotino = (*ip)->i_ino; - spin_unlock(&mp->m_sb_lock); - xfs_mod_sb(tp, sbfields); - - if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) { - xfs_alert(mp, "%s failed (error %d)!", __func__, error); - return error; - } - return 0; -} - - -STATIC void -xfs_qm_reset_dqcounts( - xfs_mount_t *mp, - xfs_buf_t *bp, - xfs_dqid_t id, - uint type) -{ - xfs_disk_dquot_t *ddq; - int j; - - trace_xfs_reset_dqcounts(bp, _RET_IP_); - - /* - * Reset all counters and timers. They'll be - * started afresh by xfs_qm_quotacheck. - */ -#ifdef DEBUG - j = XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); - do_div(j, sizeof(xfs_dqblk_t)); - ASSERT(mp->m_quotainfo->qi_dqperchunk == j); -#endif - ddq = bp->b_addr; - for (j = 0; j < mp->m_quotainfo->qi_dqperchunk; j++) { - /* - * Do a sanity check, and if needed, repair the dqblk. Don't - * output any warnings because it's perfectly possible to - * find uninitialised dquot blks. See comment in xfs_qm_dqcheck. - */ - (void) xfs_qm_dqcheck(mp, ddq, id+j, type, XFS_QMOPT_DQREPAIR, - "xfs_quotacheck"); - ddq->d_bcount = 0; - ddq->d_icount = 0; - ddq->d_rtbcount = 0; - ddq->d_btimer = 0; - ddq->d_itimer = 0; - ddq->d_rtbtimer = 0; - ddq->d_bwarns = 0; - ddq->d_iwarns = 0; - ddq->d_rtbwarns = 0; - ddq = (xfs_disk_dquot_t *) ((xfs_dqblk_t *)ddq + 1); - } -} - -STATIC int -xfs_qm_dqiter_bufs( - xfs_mount_t *mp, - xfs_dqid_t firstid, - xfs_fsblock_t bno, - xfs_filblks_t blkcnt, - uint flags) -{ - xfs_buf_t *bp; - int error; - int type; - - ASSERT(blkcnt > 0); - type = flags & XFS_QMOPT_UQUOTA ? XFS_DQ_USER : - (flags & XFS_QMOPT_PQUOTA ? XFS_DQ_PROJ : XFS_DQ_GROUP); - error = 0; - - /* - * Blkcnt arg can be a very big number, and might even be - * larger than the log itself. So, we have to break it up into - * manageable-sized transactions. - * Note that we don't start a permanent transaction here; we might - * not be able to get a log reservation for the whole thing up front, - * and we don't really care to either, because we just discard - * everything if we were to crash in the middle of this loop. - */ - while (blkcnt--) { - error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, - XFS_FSB_TO_DADDR(mp, bno), - mp->m_quotainfo->qi_dqchunklen, 0, &bp); - if (error) - break; - - xfs_qm_reset_dqcounts(mp, bp, firstid, type); - xfs_bdwrite(mp, bp); - /* - * goto the next block. - */ - bno++; - firstid += mp->m_quotainfo->qi_dqperchunk; - } - return error; -} - -/* - * Iterate over all allocated USR/GRP/PRJ dquots in the system, calling a - * caller supplied function for every chunk of dquots that we find. - */ -STATIC int -xfs_qm_dqiterate( - xfs_mount_t *mp, - xfs_inode_t *qip, - uint flags) -{ - xfs_bmbt_irec_t *map; - int i, nmaps; /* number of map entries */ - int error; /* return value */ - xfs_fileoff_t lblkno; - xfs_filblks_t maxlblkcnt; - xfs_dqid_t firstid; - xfs_fsblock_t rablkno; - xfs_filblks_t rablkcnt; - - error = 0; - /* - * This looks racy, but we can't keep an inode lock across a - * trans_reserve. But, this gets called during quotacheck, and that - * happens only at mount time which is single threaded. - */ - if (qip->i_d.di_nblocks == 0) - return 0; - - map = kmem_alloc(XFS_DQITER_MAP_SIZE * sizeof(*map), KM_SLEEP); - - lblkno = 0; - maxlblkcnt = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp)); - do { - nmaps = XFS_DQITER_MAP_SIZE; - /* - * We aren't changing the inode itself. Just changing - * some of its data. No new blocks are added here, and - * the inode is never added to the transaction. - */ - xfs_ilock(qip, XFS_ILOCK_SHARED); - error = xfs_bmapi(NULL, qip, lblkno, - maxlblkcnt - lblkno, - XFS_BMAPI_METADATA, - NULL, - 0, map, &nmaps, NULL); - xfs_iunlock(qip, XFS_ILOCK_SHARED); - if (error) - break; - - ASSERT(nmaps <= XFS_DQITER_MAP_SIZE); - for (i = 0; i < nmaps; i++) { - ASSERT(map[i].br_startblock != DELAYSTARTBLOCK); - ASSERT(map[i].br_blockcount); - - - lblkno += map[i].br_blockcount; - - if (map[i].br_startblock == HOLESTARTBLOCK) - continue; - - firstid = (xfs_dqid_t) map[i].br_startoff * - mp->m_quotainfo->qi_dqperchunk; - /* - * Do a read-ahead on the next extent. - */ - if ((i+1 < nmaps) && - (map[i+1].br_startblock != HOLESTARTBLOCK)) { - rablkcnt = map[i+1].br_blockcount; - rablkno = map[i+1].br_startblock; - while (rablkcnt--) { - xfs_buf_readahead(mp->m_ddev_targp, - XFS_FSB_TO_DADDR(mp, rablkno), - mp->m_quotainfo->qi_dqchunklen); - rablkno++; - } - } - /* - * Iterate thru all the blks in the extent and - * reset the counters of all the dquots inside them. - */ - if ((error = xfs_qm_dqiter_bufs(mp, - firstid, - map[i].br_startblock, - map[i].br_blockcount, - flags))) { - break; - } - } - - if (error) - break; - } while (nmaps > 0); - - kmem_free(map); - - return error; -} - -/* - * Called by dqusage_adjust in doing a quotacheck. - * - * Given the inode, and a dquot id this updates both the incore dqout as well - * as the buffer copy. This is so that once the quotacheck is done, we can - * just log all the buffers, as opposed to logging numerous updates to - * individual dquots. - */ -STATIC int -xfs_qm_quotacheck_dqadjust( - struct xfs_inode *ip, - xfs_dqid_t id, - uint type, - xfs_qcnt_t nblks, - xfs_qcnt_t rtblks) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_dquot *dqp; - int error; - - error = xfs_qm_dqget(mp, ip, id, type, - XFS_QMOPT_DQALLOC | XFS_QMOPT_DOWARN, &dqp); - if (error) { - /* - * Shouldn't be able to turn off quotas here. - */ - ASSERT(error != ESRCH); - ASSERT(error != ENOENT); - return error; - } - - trace_xfs_dqadjust(dqp); - - /* - * Adjust the inode count and the block count to reflect this inode's - * resource usage. - */ - be64_add_cpu(&dqp->q_core.d_icount, 1); - dqp->q_res_icount++; - if (nblks) { - be64_add_cpu(&dqp->q_core.d_bcount, nblks); - dqp->q_res_bcount += nblks; - } - if (rtblks) { - be64_add_cpu(&dqp->q_core.d_rtbcount, rtblks); - dqp->q_res_rtbcount += rtblks; - } - - /* - * Set default limits, adjust timers (since we changed usages) - * - * There are no timers for the default values set in the root dquot. - */ - if (dqp->q_core.d_id) { - xfs_qm_adjust_dqlimits(mp, &dqp->q_core); - xfs_qm_adjust_dqtimers(mp, &dqp->q_core); - } - - dqp->dq_flags |= XFS_DQ_DIRTY; - xfs_qm_dqput(dqp); - return 0; -} - -STATIC int -xfs_qm_get_rtblks( - xfs_inode_t *ip, - xfs_qcnt_t *O_rtblks) -{ - xfs_filblks_t rtblks; /* total rt blks */ - xfs_extnum_t idx; /* extent record index */ - xfs_ifork_t *ifp; /* inode fork pointer */ - xfs_extnum_t nextents; /* number of extent entries */ - int error; - - ASSERT(XFS_IS_REALTIME_INODE(ip)); - ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); - if (!(ifp->if_flags & XFS_IFEXTENTS)) { - if ((error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK))) - return error; - } - rtblks = 0; - nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); - for (idx = 0; idx < nextents; idx++) - rtblks += xfs_bmbt_get_blockcount(xfs_iext_get_ext(ifp, idx)); - *O_rtblks = (xfs_qcnt_t)rtblks; - return 0; -} - -/* - * callback routine supplied to bulkstat(). Given an inumber, find its - * dquots and update them to account for resources taken by that inode. - */ -/* ARGSUSED */ -STATIC int -xfs_qm_dqusage_adjust( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_ino_t ino, /* inode number to get data for */ - void __user *buffer, /* not used */ - int ubsize, /* not used */ - int *ubused, /* not used */ - int *res) /* result code value */ -{ - xfs_inode_t *ip; - xfs_qcnt_t nblks, rtblks = 0; - int error; - - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - /* - * rootino must have its resources accounted for, not so with the quota - * inodes. - */ - if (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino) { - *res = BULKSTAT_RV_NOTHING; - return XFS_ERROR(EINVAL); - } - - /* - * We don't _need_ to take the ilock EXCL. However, the xfs_qm_dqget - * interface expects the inode to be exclusively locked because that's - * the case in all other instances. It's OK that we do this because - * quotacheck is done only at mount time. - */ - error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip); - if (error) { - *res = BULKSTAT_RV_NOTHING; - return error; - } - - ASSERT(ip->i_delayed_blks == 0); - - if (XFS_IS_REALTIME_INODE(ip)) { - /* - * Walk thru the extent list and count the realtime blocks. - */ - error = xfs_qm_get_rtblks(ip, &rtblks); - if (error) - goto error0; - } - - nblks = (xfs_qcnt_t)ip->i_d.di_nblocks - rtblks; - - /* - * Add the (disk blocks and inode) resources occupied by this - * inode to its dquots. We do this adjustment in the incore dquot, - * and also copy the changes to its buffer. - * We don't care about putting these changes in a transaction - * envelope because if we crash in the middle of a 'quotacheck' - * we have to start from the beginning anyway. - * Once we're done, we'll log all the dquot bufs. - * - * The *QUOTA_ON checks below may look pretty racy, but quotachecks - * and quotaoffs don't race. (Quotachecks happen at mount time only). - */ - if (XFS_IS_UQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_uid, - XFS_DQ_USER, nblks, rtblks); - if (error) - goto error0; - } - - if (XFS_IS_GQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_gid, - XFS_DQ_GROUP, nblks, rtblks); - if (error) - goto error0; - } - - if (XFS_IS_PQUOTA_ON(mp)) { - error = xfs_qm_quotacheck_dqadjust(ip, xfs_get_projid(ip), - XFS_DQ_PROJ, nblks, rtblks); - if (error) - goto error0; - } - - xfs_iunlock(ip, XFS_ILOCK_EXCL); - IRELE(ip); - *res = BULKSTAT_RV_DIDONE; - return 0; - -error0: - xfs_iunlock(ip, XFS_ILOCK_EXCL); - IRELE(ip); - *res = BULKSTAT_RV_GIVEUP; - return error; -} - -/* - * Walk thru all the filesystem inodes and construct a consistent view - * of the disk quota world. If the quotacheck fails, disable quotas. - */ -int -xfs_qm_quotacheck( - xfs_mount_t *mp) -{ - int done, count, error; - xfs_ino_t lastino; - size_t structsz; - xfs_inode_t *uip, *gip; - uint flags; - - count = INT_MAX; - structsz = 1; - lastino = 0; - flags = 0; - - ASSERT(mp->m_quotainfo->qi_uquotaip || mp->m_quotainfo->qi_gquotaip); - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - /* - * There should be no cached dquots. The (simplistic) quotacheck - * algorithm doesn't like that. - */ - ASSERT(list_empty(&mp->m_quotainfo->qi_dqlist)); - - xfs_notice(mp, "Quotacheck needed: Please wait."); - - /* - * First we go thru all the dquots on disk, USR and GRP/PRJ, and reset - * their counters to zero. We need a clean slate. - * We don't log our changes till later. - */ - uip = mp->m_quotainfo->qi_uquotaip; - if (uip) { - error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA); - if (error) - goto error_return; - flags |= XFS_UQUOTA_CHKD; - } - - gip = mp->m_quotainfo->qi_gquotaip; - if (gip) { - error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ? - XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA); - if (error) - goto error_return; - flags |= XFS_OQUOTA_CHKD; - } - - do { - /* - * Iterate thru all the inodes in the file system, - * adjusting the corresponding dquot counters in core. - */ - error = xfs_bulkstat(mp, &lastino, &count, - xfs_qm_dqusage_adjust, - structsz, NULL, &done); - if (error) - break; - - } while (!done); - - /* - * We've made all the changes that we need to make incore. - * Flush them down to disk buffers if everything was updated - * successfully. - */ - if (!error) - error = xfs_qm_dqflush_all(mp, 0); - - /* - * We can get this error if we couldn't do a dquot allocation inside - * xfs_qm_dqusage_adjust (via bulkstat). We don't care about the - * dirty dquots that might be cached, we just want to get rid of them - * and turn quotaoff. The dquots won't be attached to any of the inodes - * at this point (because we intentionally didn't in dqget_noattach). - */ - if (error) { - xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL); - goto error_return; - } - - /* - * We didn't log anything, because if we crashed, we'll have to - * start the quotacheck from scratch anyway. However, we must make - * sure that our dquot changes are secure before we put the - * quotacheck'd stamp on the superblock. So, here we do a synchronous - * flush. - */ - XFS_bflush(mp->m_ddev_targp); - - /* - * If one type of quotas is off, then it will lose its - * quotachecked status, since we won't be doing accounting for - * that type anymore. - */ - mp->m_qflags &= ~(XFS_OQUOTA_CHKD | XFS_UQUOTA_CHKD); - mp->m_qflags |= flags; - - error_return: - if (error) { - xfs_warn(mp, - "Quotacheck: Unsuccessful (Error %d): Disabling quotas.", - error); - /* - * We must turn off quotas. - */ - ASSERT(mp->m_quotainfo != NULL); - ASSERT(xfs_Gqm != NULL); - xfs_qm_destroy_quotainfo(mp); - if (xfs_mount_reset_sbqflags(mp)) { - xfs_warn(mp, - "Quotacheck: Failed to reset quota flags."); - } - } else - xfs_notice(mp, "Quotacheck: Done."); - return (error); -} - -/* - * This is called after the superblock has been read in and we're ready to - * iget the quota inodes. - */ -STATIC int -xfs_qm_init_quotainos( - xfs_mount_t *mp) -{ - xfs_inode_t *uip, *gip; - int error; - __int64_t sbflags; - uint flags; - - ASSERT(mp->m_quotainfo); - uip = gip = NULL; - sbflags = 0; - flags = 0; - - /* - * Get the uquota and gquota inodes - */ - if (xfs_sb_version_hasquota(&mp->m_sb)) { - if (XFS_IS_UQUOTA_ON(mp) && - mp->m_sb.sb_uquotino != NULLFSINO) { - ASSERT(mp->m_sb.sb_uquotino > 0); - if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, - 0, 0, &uip))) - return XFS_ERROR(error); - } - if (XFS_IS_OQUOTA_ON(mp) && - mp->m_sb.sb_gquotino != NULLFSINO) { - ASSERT(mp->m_sb.sb_gquotino > 0); - if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, - 0, 0, &gip))) { - if (uip) - IRELE(uip); - return XFS_ERROR(error); - } - } - } else { - flags |= XFS_QMOPT_SBVERSION; - sbflags |= (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | - XFS_SB_GQUOTINO | XFS_SB_QFLAGS); - } - - /* - * Create the two inodes, if they don't exist already. The changes - * made above will get added to a transaction and logged in one of - * the qino_alloc calls below. If the device is readonly, - * temporarily switch to read-write to do this. - */ - if (XFS_IS_UQUOTA_ON(mp) && uip == NULL) { - if ((error = xfs_qm_qino_alloc(mp, &uip, - sbflags | XFS_SB_UQUOTINO, - flags | XFS_QMOPT_UQUOTA))) - return XFS_ERROR(error); - - flags &= ~XFS_QMOPT_SBVERSION; - } - if (XFS_IS_OQUOTA_ON(mp) && gip == NULL) { - flags |= (XFS_IS_GQUOTA_ON(mp) ? - XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA); - error = xfs_qm_qino_alloc(mp, &gip, - sbflags | XFS_SB_GQUOTINO, flags); - if (error) { - if (uip) - IRELE(uip); - - return XFS_ERROR(error); - } - } - - mp->m_quotainfo->qi_uquotaip = uip; - mp->m_quotainfo->qi_gquotaip = gip; - - return 0; -} - - - -/* - * Just pop the least recently used dquot off the freelist and - * recycle it. The returned dquot is locked. - */ -STATIC xfs_dquot_t * -xfs_qm_dqreclaim_one(void) -{ - xfs_dquot_t *dqpout; - xfs_dquot_t *dqp; - int restarts; - int startagain; - - restarts = 0; - dqpout = NULL; - - /* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */ -again: - startagain = 0; - mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); - - list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) { - struct xfs_mount *mp = dqp->q_mount; - xfs_dqlock(dqp); - - /* - * We are racing with dqlookup here. Naturally we don't - * want to reclaim a dquot that lookup wants. We release the - * freelist lock and start over, so that lookup will grab - * both the dquot and the freelistlock. - */ - if (dqp->dq_flags & XFS_DQ_WANT) { - ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE)); - - trace_xfs_dqreclaim_want(dqp); - XQM_STATS_INC(xqmstats.xs_qm_dqwants); - restarts++; - startagain = 1; - goto dqunlock; - } - - /* - * If the dquot is inactive, we are assured that it is - * not on the mplist or the hashlist, and that makes our - * life easier. - */ - if (dqp->dq_flags & XFS_DQ_INACTIVE) { - ASSERT(mp == NULL); - ASSERT(! XFS_DQ_IS_DIRTY(dqp)); - ASSERT(list_empty(&dqp->q_hashlist)); - ASSERT(list_empty(&dqp->q_mplist)); - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - dqpout = dqp; - XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims); - goto dqunlock; - } - - ASSERT(dqp->q_hash); - ASSERT(!list_empty(&dqp->q_mplist)); - - /* - * Try to grab the flush lock. If this dquot is in the process - * of getting flushed to disk, we don't want to reclaim it. - */ - if (!xfs_dqflock_nowait(dqp)) - goto dqunlock; - - /* - * We have the flush lock so we know that this is not in the - * process of being flushed. So, if this is dirty, flush it - * DELWRI so that we don't get a freelist infested with - * dirty dquots. - */ - if (XFS_DQ_IS_DIRTY(dqp)) { - int error; - - trace_xfs_dqreclaim_dirty(dqp); - - /* - * We flush it delayed write, so don't bother - * releasing the freelist lock. - */ - error = xfs_qm_dqflush(dqp, 0); - if (error) { - xfs_warn(mp, "%s: dquot %p flush failed", - __func__, dqp); - } - goto dqunlock; - } - - /* - * We're trying to get the hashlock out of order. This races - * with dqlookup; so, we giveup and goto the next dquot if - * we couldn't get the hashlock. This way, we won't starve - * a dqlookup process that holds the hashlock that is - * waiting for the freelist lock. - */ - if (!mutex_trylock(&dqp->q_hash->qh_lock)) { - restarts++; - goto dqfunlock; - } - - /* - * This races with dquot allocation code as well as dqflush_all - * and reclaim code. So, if we failed to grab the mplist lock, - * giveup everything and start over. - */ - if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) { - restarts++; - startagain = 1; - goto qhunlock; - } - - ASSERT(dqp->q_nrefs == 0); - list_del_init(&dqp->q_mplist); - mp->m_quotainfo->qi_dquots--; - mp->m_quotainfo->qi_dqreclaims++; - list_del_init(&dqp->q_hashlist); - dqp->q_hash->qh_version++; - list_del_init(&dqp->q_freelist); - xfs_Gqm->qm_dqfrlist_cnt--; - dqpout = dqp; - mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); -qhunlock: - mutex_unlock(&dqp->q_hash->qh_lock); -dqfunlock: - xfs_dqfunlock(dqp); -dqunlock: - xfs_dqunlock(dqp); - if (dqpout) - break; - if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS) - break; - if (startagain) { - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - goto again; - } - } - mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); - return dqpout; -} - -/* - * Traverse the freelist of dquots and attempt to reclaim a maximum of - * 'howmany' dquots. This operation races with dqlookup(), and attempts to - * favor the lookup function ... - */ -STATIC int -xfs_qm_shake_freelist( - int howmany) -{ - int nreclaimed = 0; - xfs_dquot_t *dqp; - - if (howmany <= 0) - return 0; - - while (nreclaimed < howmany) { - dqp = xfs_qm_dqreclaim_one(); - if (!dqp) - return nreclaimed; - xfs_qm_dqdestroy(dqp); - nreclaimed++; - } - return nreclaimed; -} - -/* - * The kmem_shake interface is invoked when memory is running low. - */ -/* ARGSUSED */ -STATIC int -xfs_qm_shake( - struct shrinker *shrink, - struct shrink_control *sc) -{ - int ndqused, nfree, n; - gfp_t gfp_mask = sc->gfp_mask; - - if (!kmem_shake_allow(gfp_mask)) - return 0; - if (!xfs_Gqm) - return 0; - - nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */ - /* incore dquots in all f/s's */ - ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree; - - ASSERT(ndqused >= 0); - - if (nfree <= ndqused && nfree < ndquot) - return 0; - - ndqused *= xfs_Gqm->qm_dqfree_ratio; /* target # of free dquots */ - n = nfree - ndqused - ndquot; /* # over target */ - - return xfs_qm_shake_freelist(MAX(nfree, n)); -} - - -/*------------------------------------------------------------------*/ - -/* - * Return a new incore dquot. Depending on the number of - * dquots in the system, we either allocate a new one on the kernel heap, - * or reclaim a free one. - * Return value is B_TRUE if we allocated a new dquot, B_FALSE if we managed - * to reclaim an existing one from the freelist. - */ -boolean_t -xfs_qm_dqalloc_incore( - xfs_dquot_t **O_dqpp) -{ - xfs_dquot_t *dqp; - - /* - * Check against high water mark to see if we want to pop - * a nincompoop dquot off the freelist. - */ - if (atomic_read(&xfs_Gqm->qm_totaldquots) >= ndquot) { - /* - * Try to recycle a dquot from the freelist. - */ - if ((dqp = xfs_qm_dqreclaim_one())) { - XQM_STATS_INC(xqmstats.xs_qm_dqreclaims); - /* - * Just zero the core here. The rest will get - * reinitialized by caller. XXX we shouldn't even - * do this zero ... - */ - memset(&dqp->q_core, 0, sizeof(dqp->q_core)); - *O_dqpp = dqp; - return B_FALSE; - } - XQM_STATS_INC(xqmstats.xs_qm_dqreclaim_misses); - } - - /* - * Allocate a brand new dquot on the kernel heap and return it - * to the caller to initialize. - */ - ASSERT(xfs_Gqm->qm_dqzone != NULL); - *O_dqpp = kmem_zone_zalloc(xfs_Gqm->qm_dqzone, KM_SLEEP); - atomic_inc(&xfs_Gqm->qm_totaldquots); - - return B_TRUE; -} - - -/* - * Start a transaction and write the incore superblock changes to - * disk. flags parameter indicates which fields have changed. - */ -int -xfs_qm_write_sb_changes( - xfs_mount_t *mp, - __int64_t flags) -{ - xfs_trans_t *tp; - int error; - - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE); - if ((error = xfs_trans_reserve(tp, 0, - mp->m_sb.sb_sectsize + 128, 0, - 0, - XFS_DEFAULT_LOG_COUNT))) { - xfs_trans_cancel(tp, 0); - return error; - } - - xfs_mod_sb(tp, flags); - error = xfs_trans_commit(tp, 0); - - return error; -} - - -/* --------------- utility functions for vnodeops ---------------- */ - - -/* - * Given an inode, a uid, gid and prid make sure that we have - * allocated relevant dquot(s) on disk, and that we won't exceed inode - * quotas by creating this file. - * This also attaches dquot(s) to the given inode after locking it, - * and returns the dquots corresponding to the uid and/or gid. - * - * in : inode (unlocked) - * out : udquot, gdquot with references taken and unlocked - */ -int -xfs_qm_vop_dqalloc( - struct xfs_inode *ip, - uid_t uid, - gid_t gid, - prid_t prid, - uint flags, - struct xfs_dquot **O_udqpp, - struct xfs_dquot **O_gdqpp) -{ - struct xfs_mount *mp = ip->i_mount; - struct xfs_dquot *uq, *gq; - int error; - uint lockflags; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - - lockflags = XFS_ILOCK_EXCL; - xfs_ilock(ip, lockflags); - - if ((flags & XFS_QMOPT_INHERIT) && XFS_INHERIT_GID(ip)) - gid = ip->i_d.di_gid; - - /* - * Attach the dquot(s) to this inode, doing a dquot allocation - * if necessary. The dquot(s) will not be locked. - */ - if (XFS_NOT_DQATTACHED(mp, ip)) { - error = xfs_qm_dqattach_locked(ip, XFS_QMOPT_DQALLOC); - if (error) { - xfs_iunlock(ip, lockflags); - return error; - } - } - - uq = gq = NULL; - if ((flags & XFS_QMOPT_UQUOTA) && XFS_IS_UQUOTA_ON(mp)) { - if (ip->i_d.di_uid != uid) { - /* - * What we need is the dquot that has this uid, and - * if we send the inode to dqget, the uid of the inode - * takes priority over what's sent in the uid argument. - * We must unlock inode here before calling dqget if - * we're not sending the inode, because otherwise - * we'll deadlock by doing trans_reserve while - * holding ilock. - */ - xfs_iunlock(ip, lockflags); - if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t) uid, - XFS_DQ_USER, - XFS_QMOPT_DQALLOC | - XFS_QMOPT_DOWARN, - &uq))) { - ASSERT(error != ENOENT); - return error; - } - /* - * Get the ilock in the right order. - */ - xfs_dqunlock(uq); - lockflags = XFS_ILOCK_SHARED; - xfs_ilock(ip, lockflags); - } else { - /* - * Take an extra reference, because we'll return - * this to caller - */ - ASSERT(ip->i_udquot); - uq = ip->i_udquot; - xfs_dqlock(uq); - XFS_DQHOLD(uq); - xfs_dqunlock(uq); - } - } - if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) { - if (ip->i_d.di_gid != gid) { - xfs_iunlock(ip, lockflags); - if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)gid, - XFS_DQ_GROUP, - XFS_QMOPT_DQALLOC | - XFS_QMOPT_DOWARN, - &gq))) { - if (uq) - xfs_qm_dqrele(uq); - ASSERT(error != ENOENT); - return error; - } - xfs_dqunlock(gq); - lockflags = XFS_ILOCK_SHARED; - xfs_ilock(ip, lockflags); - } else { - ASSERT(ip->i_gdquot); - gq = ip->i_gdquot; - xfs_dqlock(gq); - XFS_DQHOLD(gq); - xfs_dqunlock(gq); - } - } else if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { - if (xfs_get_projid(ip) != prid) { - xfs_iunlock(ip, lockflags); - if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)prid, - XFS_DQ_PROJ, - XFS_QMOPT_DQALLOC | - XFS_QMOPT_DOWARN, - &gq))) { - if (uq) - xfs_qm_dqrele(uq); - ASSERT(error != ENOENT); - return (error); - } - xfs_dqunlock(gq); - lockflags = XFS_ILOCK_SHARED; - xfs_ilock(ip, lockflags); - } else { - ASSERT(ip->i_gdquot); - gq = ip->i_gdquot; - xfs_dqlock(gq); - XFS_DQHOLD(gq); - xfs_dqunlock(gq); - } - } - if (uq) - trace_xfs_dquot_dqalloc(ip); - - xfs_iunlock(ip, lockflags); - if (O_udqpp) - *O_udqpp = uq; - else if (uq) - xfs_qm_dqrele(uq); - if (O_gdqpp) - *O_gdqpp = gq; - else if (gq) - xfs_qm_dqrele(gq); - return 0; -} - -/* - * Actually transfer ownership, and do dquot modifications. - * These were already reserved. - */ -xfs_dquot_t * -xfs_qm_vop_chown( - xfs_trans_t *tp, - xfs_inode_t *ip, - xfs_dquot_t **IO_olddq, - xfs_dquot_t *newdq) -{ - xfs_dquot_t *prevdq; - uint bfield = XFS_IS_REALTIME_INODE(ip) ? - XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT; - - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount)); - - /* old dquot */ - prevdq = *IO_olddq; - ASSERT(prevdq); - ASSERT(prevdq != newdq); - - xfs_trans_mod_dquot(tp, prevdq, bfield, -(ip->i_d.di_nblocks)); - xfs_trans_mod_dquot(tp, prevdq, XFS_TRANS_DQ_ICOUNT, -1); - - /* the sparkling new dquot */ - xfs_trans_mod_dquot(tp, newdq, bfield, ip->i_d.di_nblocks); - xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_ICOUNT, 1); - - /* - * Take an extra reference, because the inode - * is going to keep this dquot pointer even - * after the trans_commit. - */ - xfs_dqlock(newdq); - XFS_DQHOLD(newdq); - xfs_dqunlock(newdq); - *IO_olddq = newdq; - - return prevdq; -} - -/* - * Quota reservations for setattr(AT_UID|AT_GID|AT_PROJID). - */ -int -xfs_qm_vop_chown_reserve( - xfs_trans_t *tp, - xfs_inode_t *ip, - xfs_dquot_t *udqp, - xfs_dquot_t *gdqp, - uint flags) -{ - xfs_mount_t *mp = ip->i_mount; - uint delblks, blkflags, prjflags = 0; - xfs_dquot_t *unresudq, *unresgdq, *delblksudq, *delblksgdq; - int error; - - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - delblks = ip->i_delayed_blks; - delblksudq = delblksgdq = unresudq = unresgdq = NULL; - blkflags = XFS_IS_REALTIME_INODE(ip) ? - XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS; - - if (XFS_IS_UQUOTA_ON(mp) && udqp && - ip->i_d.di_uid != (uid_t)be32_to_cpu(udqp->q_core.d_id)) { - delblksudq = udqp; - /* - * If there are delayed allocation blocks, then we have to - * unreserve those from the old dquot, and add them to the - * new dquot. - */ - if (delblks) { - ASSERT(ip->i_udquot); - unresudq = ip->i_udquot; - } - } - if (XFS_IS_OQUOTA_ON(ip->i_mount) && gdqp) { - if (XFS_IS_PQUOTA_ON(ip->i_mount) && - xfs_get_projid(ip) != be32_to_cpu(gdqp->q_core.d_id)) - prjflags = XFS_QMOPT_ENOSPC; - - if (prjflags || - (XFS_IS_GQUOTA_ON(ip->i_mount) && - ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id))) { - delblksgdq = gdqp; - if (delblks) { - ASSERT(ip->i_gdquot); - unresgdq = ip->i_gdquot; - } - } - } - - if ((error = xfs_trans_reserve_quota_bydquots(tp, ip->i_mount, - delblksudq, delblksgdq, ip->i_d.di_nblocks, 1, - flags | blkflags | prjflags))) - return (error); - - /* - * Do the delayed blks reservations/unreservations now. Since, these - * are done without the help of a transaction, if a reservation fails - * its previous reservations won't be automatically undone by trans - * code. So, we have to do it manually here. - */ - if (delblks) { - /* - * Do the reservations first. Unreservation can't fail. - */ - ASSERT(delblksudq || delblksgdq); - ASSERT(unresudq || unresgdq); - if ((error = xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - delblksudq, delblksgdq, (xfs_qcnt_t)delblks, 0, - flags | blkflags | prjflags))) - return (error); - xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - unresudq, unresgdq, -((xfs_qcnt_t)delblks), 0, - blkflags); - } - - return (0); -} - -int -xfs_qm_vop_rename_dqattach( - struct xfs_inode **i_tab) -{ - struct xfs_mount *mp = i_tab[0]->i_mount; - int i; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - - for (i = 0; (i < 4 && i_tab[i]); i++) { - struct xfs_inode *ip = i_tab[i]; - int error; - - /* - * Watch out for duplicate entries in the table. - */ - if (i == 0 || ip != i_tab[i-1]) { - if (XFS_NOT_DQATTACHED(mp, ip)) { - error = xfs_qm_dqattach(ip, 0); - if (error) - return error; - } - } - } - return 0; -} - -void -xfs_qm_vop_create_dqattach( - struct xfs_trans *tp, - struct xfs_inode *ip, - struct xfs_dquot *udqp, - struct xfs_dquot *gdqp) -{ - struct xfs_mount *mp = tp->t_mountp; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return; - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - if (udqp) { - xfs_dqlock(udqp); - XFS_DQHOLD(udqp); - xfs_dqunlock(udqp); - ASSERT(ip->i_udquot == NULL); - ip->i_udquot = udqp; - ASSERT(XFS_IS_UQUOTA_ON(mp)); - ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id)); - xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1); - } - if (gdqp) { - xfs_dqlock(gdqp); - XFS_DQHOLD(gdqp); - xfs_dqunlock(gdqp); - ASSERT(ip->i_gdquot == NULL); - ip->i_gdquot = gdqp; - ASSERT(XFS_IS_OQUOTA_ON(mp)); - ASSERT((XFS_IS_GQUOTA_ON(mp) ? - ip->i_d.di_gid : xfs_get_projid(ip)) == - be32_to_cpu(gdqp->q_core.d_id)); - xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1); - } -} - diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h deleted file mode 100644 index 43b9abe1052c..000000000000 --- a/fs/xfs/quota/xfs_qm.h +++ /dev/null @@ -1,166 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_QM_H__ -#define __XFS_QM_H__ - -#include "xfs_dquot_item.h" -#include "xfs_dquot.h" -#include "xfs_quota_priv.h" -#include "xfs_qm_stats.h" - -struct xfs_qm; -struct xfs_inode; - -extern uint ndquot; -extern struct mutex xfs_Gqm_lock; -extern struct xfs_qm *xfs_Gqm; -extern kmem_zone_t *qm_dqzone; -extern kmem_zone_t *qm_dqtrxzone; - -/* - * Used in xfs_qm_sync called by xfs_sync to count the max times that it can - * iterate over the mountpt's dquot list in one call. - */ -#define XFS_QM_SYNC_MAX_RESTARTS 7 - -/* - * Ditto, for xfs_qm_dqreclaim_one. - */ -#define XFS_QM_RECLAIM_MAX_RESTARTS 4 - -/* - * Ideal ratio of free to in use dquots. Quota manager makes an attempt - * to keep this balance. - */ -#define XFS_QM_DQFREE_RATIO 2 - -/* - * Dquot hashtable constants/threshold values. - */ -#define XFS_QM_HASHSIZE_LOW (PAGE_SIZE / sizeof(xfs_dqhash_t)) -#define XFS_QM_HASHSIZE_HIGH ((PAGE_SIZE * 4) / sizeof(xfs_dqhash_t)) - -/* - * This defines the unit of allocation of dquots. - * Currently, it is just one file system block, and a 4K blk contains 30 - * (136 * 30 = 4080) dquots. It's probably not worth trying to make - * this more dynamic. - * XXXsup However, if this number is changed, we have to make sure that we don't - * implicitly assume that we do allocations in chunks of a single filesystem - * block in the dquot/xqm code. - */ -#define XFS_DQUOT_CLUSTER_SIZE_FSB (xfs_filblks_t)1 - -typedef xfs_dqhash_t xfs_dqlist_t; - -/* - * Quota Manager (global) structure. Lives only in core. - */ -typedef struct xfs_qm { - xfs_dqlist_t *qm_usr_dqhtable;/* udquot hash table */ - xfs_dqlist_t *qm_grp_dqhtable;/* gdquot hash table */ - uint qm_dqhashmask; /* # buckets in dq hashtab - 1 */ - struct list_head qm_dqfrlist; /* freelist of dquots */ - struct mutex qm_dqfrlist_lock; - int qm_dqfrlist_cnt; - atomic_t qm_totaldquots; /* total incore dquots */ - uint qm_nrefs; /* file systems with quota on */ - int qm_dqfree_ratio;/* ratio of free to inuse dquots */ - kmem_zone_t *qm_dqzone; /* dquot mem-alloc zone */ - kmem_zone_t *qm_dqtrxzone; /* t_dqinfo of transactions */ -} xfs_qm_t; - -/* - * Various quota information for individual filesystems. - * The mount structure keeps a pointer to this. - */ -typedef struct xfs_quotainfo { - xfs_inode_t *qi_uquotaip; /* user quota inode */ - xfs_inode_t *qi_gquotaip; /* group quota inode */ - struct list_head qi_dqlist; /* all dquots in filesys */ - struct mutex qi_dqlist_lock; - int qi_dquots; - int qi_dqreclaims; /* a change here indicates - a removal in the dqlist */ - time_t qi_btimelimit; /* limit for blks timer */ - time_t qi_itimelimit; /* limit for inodes timer */ - time_t qi_rtbtimelimit;/* limit for rt blks timer */ - xfs_qwarncnt_t qi_bwarnlimit; /* limit for blks warnings */ - xfs_qwarncnt_t qi_iwarnlimit; /* limit for inodes warnings */ - xfs_qwarncnt_t qi_rtbwarnlimit;/* limit for rt blks warnings */ - struct mutex qi_quotaofflock;/* to serialize quotaoff */ - xfs_filblks_t qi_dqchunklen; /* # BBs in a chunk of dqs */ - uint qi_dqperchunk; /* # ondisk dqs in above chunk */ - xfs_qcnt_t qi_bhardlimit; /* default data blk hard limit */ - xfs_qcnt_t qi_bsoftlimit; /* default data blk soft limit */ - xfs_qcnt_t qi_ihardlimit; /* default inode count hard limit */ - xfs_qcnt_t qi_isoftlimit; /* default inode count soft limit */ - xfs_qcnt_t qi_rtbhardlimit;/* default realtime blk hard limit */ - xfs_qcnt_t qi_rtbsoftlimit;/* default realtime blk soft limit */ -} xfs_quotainfo_t; - - -extern void xfs_trans_mod_dquot(xfs_trans_t *, xfs_dquot_t *, uint, long); -extern int xfs_trans_reserve_quota_bydquots(xfs_trans_t *, xfs_mount_t *, - xfs_dquot_t *, xfs_dquot_t *, long, long, uint); -extern void xfs_trans_dqjoin(xfs_trans_t *, xfs_dquot_t *); -extern void xfs_trans_log_dquot(xfs_trans_t *, xfs_dquot_t *); - -/* - * We keep the usr and grp dquots separately so that locking will be easier - * to do at commit time. All transactions that we know of at this point - * affect no more than two dquots of one type. Hence, the TRANS_MAXDQS value. - */ -#define XFS_QM_TRANS_MAXDQS 2 -typedef struct xfs_dquot_acct { - xfs_dqtrx_t dqa_usrdquots[XFS_QM_TRANS_MAXDQS]; - xfs_dqtrx_t dqa_grpdquots[XFS_QM_TRANS_MAXDQS]; -} xfs_dquot_acct_t; - -/* - * Users are allowed to have a usage exceeding their softlimit for - * a period this long. - */ -#define XFS_QM_BTIMELIMIT (7 * 24*60*60) /* 1 week */ -#define XFS_QM_RTBTIMELIMIT (7 * 24*60*60) /* 1 week */ -#define XFS_QM_ITIMELIMIT (7 * 24*60*60) /* 1 week */ - -#define XFS_QM_BWARNLIMIT 5 -#define XFS_QM_IWARNLIMIT 5 -#define XFS_QM_RTBWARNLIMIT 5 - -extern void xfs_qm_destroy_quotainfo(xfs_mount_t *); -extern int xfs_qm_quotacheck(xfs_mount_t *); -extern int xfs_qm_write_sb_changes(xfs_mount_t *, __int64_t); - -/* dquot stuff */ -extern boolean_t xfs_qm_dqalloc_incore(xfs_dquot_t **); -extern int xfs_qm_dqpurge_all(xfs_mount_t *, uint); -extern void xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint); - -/* quota ops */ -extern int xfs_qm_scall_trunc_qfiles(xfs_mount_t *, uint); -extern int xfs_qm_scall_getquota(xfs_mount_t *, xfs_dqid_t, uint, - fs_disk_quota_t *); -extern int xfs_qm_scall_setqlim(xfs_mount_t *, xfs_dqid_t, uint, - fs_disk_quota_t *); -extern int xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *); -extern int xfs_qm_scall_quotaon(xfs_mount_t *, uint); -extern int xfs_qm_scall_quotaoff(xfs_mount_t *, uint); - -#endif /* __XFS_QM_H__ */ diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c deleted file mode 100644 index a0a829addca9..000000000000 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ /dev/null @@ -1,176 +0,0 @@ -/* - * Copyright (c) 2000-2006 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_itable.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_qm.h" - - -STATIC void -xfs_fill_statvfs_from_dquot( - struct kstatfs *statp, - xfs_disk_dquot_t *dp) -{ - __uint64_t limit; - - limit = dp->d_blk_softlimit ? - be64_to_cpu(dp->d_blk_softlimit) : - be64_to_cpu(dp->d_blk_hardlimit); - if (limit && statp->f_blocks > limit) { - statp->f_blocks = limit; - statp->f_bfree = statp->f_bavail = - (statp->f_blocks > be64_to_cpu(dp->d_bcount)) ? - (statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0; - } - - limit = dp->d_ino_softlimit ? - be64_to_cpu(dp->d_ino_softlimit) : - be64_to_cpu(dp->d_ino_hardlimit); - if (limit && statp->f_files > limit) { - statp->f_files = limit; - statp->f_ffree = - (statp->f_files > be64_to_cpu(dp->d_icount)) ? - (statp->f_ffree - be64_to_cpu(dp->d_icount)) : 0; - } -} - - -/* - * Directory tree accounting is implemented using project quotas, where - * the project identifier is inherited from parent directories. - * A statvfs (df, etc.) of a directory that is using project quota should - * return a statvfs of the project, not the entire filesystem. - * This makes such trees appear as if they are filesystems in themselves. - */ -void -xfs_qm_statvfs( - xfs_inode_t *ip, - struct kstatfs *statp) -{ - xfs_mount_t *mp = ip->i_mount; - xfs_dquot_t *dqp; - - if (!xfs_qm_dqget(mp, NULL, xfs_get_projid(ip), XFS_DQ_PROJ, 0, &dqp)) { - xfs_fill_statvfs_from_dquot(statp, &dqp->q_core); - xfs_qm_dqput(dqp); - } -} - -int -xfs_qm_newmount( - xfs_mount_t *mp, - uint *needquotamount, - uint *quotaflags) -{ - uint quotaondisk; - uint uquotaondisk = 0, gquotaondisk = 0, pquotaondisk = 0; - - quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) && - (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT); - - if (quotaondisk) { - uquotaondisk = mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT; - pquotaondisk = mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT; - gquotaondisk = mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT; - } - - /* - * If the device itself is read-only, we can't allow - * the user to change the state of quota on the mount - - * this would generate a transaction on the ro device, - * which would lead to an I/O error and shutdown - */ - - if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) || - (!uquotaondisk && XFS_IS_UQUOTA_ON(mp)) || - (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) || - (!pquotaondisk && XFS_IS_PQUOTA_ON(mp)) || - (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) || - (!gquotaondisk && XFS_IS_OQUOTA_ON(mp))) && - xfs_dev_is_read_only(mp, "changing quota state")) { - xfs_warn(mp, "please mount with%s%s%s%s.", - (!quotaondisk ? "out quota" : ""), - (uquotaondisk ? " usrquota" : ""), - (pquotaondisk ? " prjquota" : ""), - (gquotaondisk ? " grpquota" : "")); - return XFS_ERROR(EPERM); - } - - if (XFS_IS_QUOTA_ON(mp) || quotaondisk) { - /* - * Call mount_quotas at this point only if we won't have to do - * a quotacheck. - */ - if (quotaondisk && !XFS_QM_NEED_QUOTACHECK(mp)) { - /* - * If an error occurred, qm_mount_quotas code - * has already disabled quotas. So, just finish - * mounting, and get on with the boring life - * without disk quotas. - */ - xfs_qm_mount_quotas(mp); - } else { - /* - * Clear the quota flags, but remember them. This - * is so that the quota code doesn't get invoked - * before we're ready. This can happen when an - * inode goes inactive and wants to free blocks, - * or via xfs_log_mount_finish. - */ - *needquotamount = B_TRUE; - *quotaflags = mp->m_qflags; - mp->m_qflags = 0; - } - } - - return 0; -} - -void __init -xfs_qm_init(void) -{ - printk(KERN_INFO "SGI XFS Quota Management subsystem\n"); - mutex_init(&xfs_Gqm_lock); - xfs_qm_init_procfs(); -} - -void __exit -xfs_qm_exit(void) -{ - xfs_qm_cleanup_procfs(); - if (qm_dqzone) - kmem_zone_destroy(qm_dqzone); - if (qm_dqtrxzone) - kmem_zone_destroy(qm_dqtrxzone); -} diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c deleted file mode 100644 index 8671a0b32644..000000000000 --- a/fs/xfs/quota/xfs_qm_stats.c +++ /dev/null @@ -1,105 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_itable.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_qm.h" - -struct xqmstats xqmstats; - -static int xqm_proc_show(struct seq_file *m, void *v) -{ - /* maximum; incore; ratio free to inuse; freelist */ - seq_printf(m, "%d\t%d\t%d\t%u\n", - ndquot, - xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0, - xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0, - xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0); - return 0; -} - -static int xqm_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, xqm_proc_show, NULL); -} - -static const struct file_operations xqm_proc_fops = { - .owner = THIS_MODULE, - .open = xqm_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int xqmstat_proc_show(struct seq_file *m, void *v) -{ - /* quota performance statistics */ - seq_printf(m, "qm %u %u %u %u %u %u %u %u\n", - xqmstats.xs_qm_dqreclaims, - xqmstats.xs_qm_dqreclaim_misses, - xqmstats.xs_qm_dquot_dups, - xqmstats.xs_qm_dqcachemisses, - xqmstats.xs_qm_dqcachehits, - xqmstats.xs_qm_dqwants, - xqmstats.xs_qm_dqshake_reclaims, - xqmstats.xs_qm_dqinact_reclaims); - return 0; -} - -static int xqmstat_proc_open(struct inode *inode, struct file *file) -{ - return single_open(file, xqmstat_proc_show, NULL); -} - -static const struct file_operations xqmstat_proc_fops = { - .owner = THIS_MODULE, - .open = xqmstat_proc_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -void -xfs_qm_init_procfs(void) -{ - proc_create("fs/xfs/xqmstat", 0, NULL, &xqmstat_proc_fops); - proc_create("fs/xfs/xqm", 0, NULL, &xqm_proc_fops); -} - -void -xfs_qm_cleanup_procfs(void) -{ - remove_proc_entry("fs/xfs/xqm", NULL); - remove_proc_entry("fs/xfs/xqmstat", NULL); -} diff --git a/fs/xfs/quota/xfs_qm_stats.h b/fs/xfs/quota/xfs_qm_stats.h deleted file mode 100644 index 5b964fc0dc09..000000000000 --- a/fs/xfs/quota/xfs_qm_stats.h +++ /dev/null @@ -1,53 +0,0 @@ -/* - * Copyright (c) 2002 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_QM_STATS_H__ -#define __XFS_QM_STATS_H__ - -#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF) - -/* - * XQM global statistics - */ -struct xqmstats { - __uint32_t xs_qm_dqreclaims; - __uint32_t xs_qm_dqreclaim_misses; - __uint32_t xs_qm_dquot_dups; - __uint32_t xs_qm_dqcachemisses; - __uint32_t xs_qm_dqcachehits; - __uint32_t xs_qm_dqwants; - __uint32_t xs_qm_dqshake_reclaims; - __uint32_t xs_qm_dqinact_reclaims; -}; - -extern struct xqmstats xqmstats; - -# define XQM_STATS_INC(count) ( (count)++ ) - -extern void xfs_qm_init_procfs(void); -extern void xfs_qm_cleanup_procfs(void); - -#else - -# define XQM_STATS_INC(count) do { } while (0) - -static inline void xfs_qm_init_procfs(void) { }; -static inline void xfs_qm_cleanup_procfs(void) { }; - -#endif - -#endif /* __XFS_QM_STATS_H__ */ diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c deleted file mode 100644 index 609246f42e6c..000000000000 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ /dev/null @@ -1,906 +0,0 @@ -/* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ - -#include - -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_itable.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_utils.h" -#include "xfs_qm.h" -#include "xfs_trace.h" - -STATIC int xfs_qm_log_quotaoff(xfs_mount_t *, xfs_qoff_logitem_t **, uint); -STATIC int xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *, - uint); -STATIC uint xfs_qm_export_flags(uint); -STATIC uint xfs_qm_export_qtype_flags(uint); -STATIC void xfs_qm_export_dquot(xfs_mount_t *, xfs_disk_dquot_t *, - fs_disk_quota_t *); - - -/* - * Turn off quota accounting and/or enforcement for all udquots and/or - * gdquots. Called only at unmount time. - * - * This assumes that there are no dquots of this file system cached - * incore, and modifies the ondisk dquot directly. Therefore, for example, - * it is an error to call this twice, without purging the cache. - */ -int -xfs_qm_scall_quotaoff( - xfs_mount_t *mp, - uint flags) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - uint dqtype; - int error; - uint inactivate_flags; - xfs_qoff_logitem_t *qoffstart; - int nculprits; - - /* - * No file system can have quotas enabled on disk but not in core. - * Note that quota utilities (like quotaoff) _expect_ - * errno == EEXIST here. - */ - if ((mp->m_qflags & flags) == 0) - return XFS_ERROR(EEXIST); - error = 0; - - flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD); - - /* - * We don't want to deal with two quotaoffs messing up each other, - * so we're going to serialize it. quotaoff isn't exactly a performance - * critical thing. - * If quotaoff, then we must be dealing with the root filesystem. - */ - ASSERT(q); - mutex_lock(&q->qi_quotaofflock); - - /* - * If we're just turning off quota enforcement, change mp and go. - */ - if ((flags & XFS_ALL_QUOTA_ACCT) == 0) { - mp->m_qflags &= ~(flags); - - spin_lock(&mp->m_sb_lock); - mp->m_sb.sb_qflags = mp->m_qflags; - spin_unlock(&mp->m_sb_lock); - mutex_unlock(&q->qi_quotaofflock); - - /* XXX what to do if error ? Revert back to old vals incore ? */ - error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS); - return (error); - } - - dqtype = 0; - inactivate_flags = 0; - /* - * If accounting is off, we must turn enforcement off, clear the - * quota 'CHKD' certificate to make it known that we have to - * do a quotacheck the next time this quota is turned on. - */ - if (flags & XFS_UQUOTA_ACCT) { - dqtype |= XFS_QMOPT_UQUOTA; - flags |= (XFS_UQUOTA_CHKD | XFS_UQUOTA_ENFD); - inactivate_flags |= XFS_UQUOTA_ACTIVE; - } - if (flags & XFS_GQUOTA_ACCT) { - dqtype |= XFS_QMOPT_GQUOTA; - flags |= (XFS_OQUOTA_CHKD | XFS_OQUOTA_ENFD); - inactivate_flags |= XFS_GQUOTA_ACTIVE; - } else if (flags & XFS_PQUOTA_ACCT) { - dqtype |= XFS_QMOPT_PQUOTA; - flags |= (XFS_OQUOTA_CHKD | XFS_OQUOTA_ENFD); - inactivate_flags |= XFS_PQUOTA_ACTIVE; - } - - /* - * Nothing to do? Don't complain. This happens when we're just - * turning off quota enforcement. - */ - if ((mp->m_qflags & flags) == 0) - goto out_unlock; - - /* - * Write the LI_QUOTAOFF log record, and do SB changes atomically, - * and synchronously. If we fail to write, we should abort the - * operation as it cannot be recovered safely if we crash. - */ - error = xfs_qm_log_quotaoff(mp, &qoffstart, flags); - if (error) - goto out_unlock; - - /* - * Next we clear the XFS_MOUNT_*DQ_ACTIVE bit(s) in the mount struct - * to take care of the race between dqget and quotaoff. We don't take - * any special locks to reset these bits. All processes need to check - * these bits *after* taking inode lock(s) to see if the particular - * quota type is in the process of being turned off. If *ACTIVE, it is - * guaranteed that all dquot structures and all quotainode ptrs will all - * stay valid as long as that inode is kept locked. - * - * There is no turning back after this. - */ - mp->m_qflags &= ~inactivate_flags; - - /* - * Give back all the dquot reference(s) held by inodes. - * Here we go thru every single incore inode in this file system, and - * do a dqrele on the i_udquot/i_gdquot that it may have. - * Essentially, as long as somebody has an inode locked, this guarantees - * that quotas will not be turned off. This is handy because in a - * transaction once we lock the inode(s) and check for quotaon, we can - * depend on the quota inodes (and other things) being valid as long as - * we keep the lock(s). - */ - xfs_qm_dqrele_all_inodes(mp, flags); - - /* - * Next we make the changes in the quota flag in the mount struct. - * This isn't protected by a particular lock directly, because we - * don't want to take a mrlock every time we depend on quotas being on. - */ - mp->m_qflags &= ~(flags); - - /* - * Go through all the dquots of this file system and purge them, - * according to what was turned off. We may not be able to get rid - * of all dquots, because dquots can have temporary references that - * are not attached to inodes. eg. xfs_setattr, xfs_create. - * So, if we couldn't purge all the dquots from the filesystem, - * we can't get rid of the incore data structures. - */ - while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype))) - delay(10 * nculprits); - - /* - * Transactions that had started before ACTIVE state bit was cleared - * could have logged many dquots, so they'd have higher LSNs than - * the first QUOTAOFF log record does. If we happen to crash when - * the tail of the log has gone past the QUOTAOFF record, but - * before the last dquot modification, those dquots __will__ - * recover, and that's not good. - * - * So, we have QUOTAOFF start and end logitems; the start - * logitem won't get overwritten until the end logitem appears... - */ - error = xfs_qm_log_quotaoff_end(mp, qoffstart, flags); - if (error) { - /* We're screwed now. Shutdown is the only option. */ - xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); - goto out_unlock; - } - - /* - * If quotas is completely disabled, close shop. - */ - if (((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET1) || - ((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET2)) { - mutex_unlock(&q->qi_quotaofflock); - xfs_qm_destroy_quotainfo(mp); - return (0); - } - - /* - * Release our quotainode references if we don't need them anymore. - */ - if ((dqtype & XFS_QMOPT_UQUOTA) && q->qi_uquotaip) { - IRELE(q->qi_uquotaip); - q->qi_uquotaip = NULL; - } - if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && q->qi_gquotaip) { - IRELE(q->qi_gquotaip); - q->qi_gquotaip = NULL; - } - -out_unlock: - mutex_unlock(&q->qi_quotaofflock); - return error; -} - -STATIC int -xfs_qm_scall_trunc_qfile( - struct xfs_mount *mp, - xfs_ino_t ino) -{ - struct xfs_inode *ip; - struct xfs_trans *tp; - int error; - - if (ino == NULLFSINO) - return 0; - - error = xfs_iget(mp, NULL, ino, 0, 0, &ip); - if (error) - return error; - - xfs_ilock(ip, XFS_IOLOCK_EXCL); - - tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE); - error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, - XFS_TRANS_PERM_LOG_RES, - XFS_ITRUNCATE_LOG_COUNT); - if (error) { - xfs_trans_cancel(tp, 0); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - goto out_put; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip); - - error = xfs_itruncate_data(&tp, ip, 0); - if (error) { - xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | - XFS_TRANS_ABORT); - goto out_unlock; - } - - xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); - error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); - -out_unlock: - xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); -out_put: - IRELE(ip); - return error; -} - -int -xfs_qm_scall_trunc_qfiles( - xfs_mount_t *mp, - uint flags) -{ - int error = 0, error2 = 0; - - if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { - xfs_debug(mp, "%s: flags=%x m_qflags=%x\n", - __func__, flags, mp->m_qflags); - return XFS_ERROR(EINVAL); - } - - if (flags & XFS_DQ_USER) - error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_uquotino); - if (flags & (XFS_DQ_GROUP|XFS_DQ_PROJ)) - error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino); - - return error ? error : error2; -} - -/* - * Switch on (a given) quota enforcement for a filesystem. This takes - * effect immediately. - * (Switching on quota accounting must be done at mount time.) - */ -int -xfs_qm_scall_quotaon( - xfs_mount_t *mp, - uint flags) -{ - int error; - uint qf; - __int64_t sbflags; - - flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD); - /* - * Switching on quota accounting must be done at mount time. - */ - flags &= ~(XFS_ALL_QUOTA_ACCT); - - sbflags = 0; - - if (flags == 0) { - xfs_debug(mp, "%s: zero flags, m_qflags=%x\n", - __func__, mp->m_qflags); - return XFS_ERROR(EINVAL); - } - - /* No fs can turn on quotas with a delayed effect */ - ASSERT((flags & XFS_ALL_QUOTA_ACCT) == 0); - - /* - * Can't enforce without accounting. We check the superblock - * qflags here instead of m_qflags because rootfs can have - * quota acct on ondisk without m_qflags' knowing. - */ - if (((flags & XFS_UQUOTA_ACCT) == 0 && - (mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT) == 0 && - (flags & XFS_UQUOTA_ENFD)) - || - ((flags & XFS_PQUOTA_ACCT) == 0 && - (mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT) == 0 && - (flags & XFS_GQUOTA_ACCT) == 0 && - (mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) == 0 && - (flags & XFS_OQUOTA_ENFD))) { - xfs_debug(mp, - "%s: Can't enforce without acct, flags=%x sbflags=%x\n", - __func__, flags, mp->m_sb.sb_qflags); - return XFS_ERROR(EINVAL); - } - /* - * If everything's up to-date incore, then don't waste time. - */ - if ((mp->m_qflags & flags) == flags) - return XFS_ERROR(EEXIST); - - /* - * Change sb_qflags on disk but not incore mp->qflags - * if this is the root filesystem. - */ - spin_lock(&mp->m_sb_lock); - qf = mp->m_sb.sb_qflags; - mp->m_sb.sb_qflags = qf | flags; - spin_unlock(&mp->m_sb_lock); - - /* - * There's nothing to change if it's the same. - */ - if ((qf & flags) == flags && sbflags == 0) - return XFS_ERROR(EEXIST); - sbflags |= XFS_SB_QFLAGS; - - if ((error = xfs_qm_write_sb_changes(mp, sbflags))) - return (error); - /* - * If we aren't trying to switch on quota enforcement, we are done. - */ - if (((mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT) != - (mp->m_qflags & XFS_UQUOTA_ACCT)) || - ((mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT) != - (mp->m_qflags & XFS_PQUOTA_ACCT)) || - ((mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) != - (mp->m_qflags & XFS_GQUOTA_ACCT)) || - (flags & XFS_ALL_QUOTA_ENFD) == 0) - return (0); - - if (! XFS_IS_QUOTA_RUNNING(mp)) - return XFS_ERROR(ESRCH); - - /* - * Switch on quota enforcement in core. - */ - mutex_lock(&mp->m_quotainfo->qi_quotaofflock); - mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD); - mutex_unlock(&mp->m_quotainfo->qi_quotaofflock); - - return (0); -} - - -/* - * Return quota status information, such as uquota-off, enforcements, etc. - */ -int -xfs_qm_scall_getqstat( - struct xfs_mount *mp, - struct fs_quota_stat *out) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - struct xfs_inode *uip, *gip; - boolean_t tempuqip, tempgqip; - - uip = gip = NULL; - tempuqip = tempgqip = B_FALSE; - memset(out, 0, sizeof(fs_quota_stat_t)); - - out->qs_version = FS_QSTAT_VERSION; - if (!xfs_sb_version_hasquota(&mp->m_sb)) { - out->qs_uquota.qfs_ino = NULLFSINO; - out->qs_gquota.qfs_ino = NULLFSINO; - return (0); - } - out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags & - (XFS_ALL_QUOTA_ACCT| - XFS_ALL_QUOTA_ENFD)); - out->qs_pad = 0; - out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino; - out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino; - - if (q) { - uip = q->qi_uquotaip; - gip = q->qi_gquotaip; - } - if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) { - if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, - 0, 0, &uip) == 0) - tempuqip = B_TRUE; - } - if (!gip && mp->m_sb.sb_gquotino != NULLFSINO) { - if (xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, - 0, 0, &gip) == 0) - tempgqip = B_TRUE; - } - if (uip) { - out->qs_uquota.qfs_nblks = uip->i_d.di_nblocks; - out->qs_uquota.qfs_nextents = uip->i_d.di_nextents; - if (tempuqip) - IRELE(uip); - } - if (gip) { - out->qs_gquota.qfs_nblks = gip->i_d.di_nblocks; - out->qs_gquota.qfs_nextents = gip->i_d.di_nextents; - if (tempgqip) - IRELE(gip); - } - if (q) { - out->qs_incoredqs = q->qi_dquots; - out->qs_btimelimit = q->qi_btimelimit; - out->qs_itimelimit = q->qi_itimelimit; - out->qs_rtbtimelimit = q->qi_rtbtimelimit; - out->qs_bwarnlimit = q->qi_bwarnlimit; - out->qs_iwarnlimit = q->qi_iwarnlimit; - } - return 0; -} - -#define XFS_DQ_MASK \ - (FS_DQ_LIMIT_MASK | FS_DQ_TIMER_MASK | FS_DQ_WARNS_MASK) - -/* - * Adjust quota limits, and start/stop timers accordingly. - */ -int -xfs_qm_scall_setqlim( - xfs_mount_t *mp, - xfs_dqid_t id, - uint type, - fs_disk_quota_t *newlim) -{ - struct xfs_quotainfo *q = mp->m_quotainfo; - xfs_disk_dquot_t *ddq; - xfs_dquot_t *dqp; - xfs_trans_t *tp; - int error; - xfs_qcnt_t hard, soft; - - if (newlim->d_fieldmask & ~XFS_DQ_MASK) - return EINVAL; - if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0) - return 0; - - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM); - if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128, - 0, 0, XFS_DEFAULT_LOG_COUNT))) { - xfs_trans_cancel(tp, 0); - return (error); - } - - /* - * We don't want to race with a quotaoff so take the quotaoff lock. - * (We don't hold an inode lock, so there's nothing else to stop - * a quotaoff from happening). (XXXThis doesn't currently happen - * because we take the vfslock before calling xfs_qm_sysent). - */ - mutex_lock(&q->qi_quotaofflock); - - /* - * Get the dquot (locked), and join it to the transaction. - * Allocate the dquot if this doesn't exist. - */ - if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) { - xfs_trans_cancel(tp, XFS_TRANS_ABORT); - ASSERT(error != ENOENT); - goto out_unlock; - } - xfs_trans_dqjoin(tp, dqp); - ddq = &dqp->q_core; - - /* - * Make sure that hardlimits are >= soft limits before changing. - */ - hard = (newlim->d_fieldmask & FS_DQ_BHARD) ? - (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_hardlimit) : - be64_to_cpu(ddq->d_blk_hardlimit); - soft = (newlim->d_fieldmask & FS_DQ_BSOFT) ? - (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_softlimit) : - be64_to_cpu(ddq->d_blk_softlimit); - if (hard == 0 || hard >= soft) { - ddq->d_blk_hardlimit = cpu_to_be64(hard); - ddq->d_blk_softlimit = cpu_to_be64(soft); - if (id == 0) { - q->qi_bhardlimit = hard; - q->qi_bsoftlimit = soft; - } - } else { - xfs_debug(mp, "blkhard %Ld < blksoft %Ld\n", hard, soft); - } - hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ? - (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) : - be64_to_cpu(ddq->d_rtb_hardlimit); - soft = (newlim->d_fieldmask & FS_DQ_RTBSOFT) ? - (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_softlimit) : - be64_to_cpu(ddq->d_rtb_softlimit); - if (hard == 0 || hard >= soft) { - ddq->d_rtb_hardlimit = cpu_to_be64(hard); - ddq->d_rtb_softlimit = cpu_to_be64(soft); - if (id == 0) { - q->qi_rtbhardlimit = hard; - q->qi_rtbsoftlimit = soft; - } - } else { - xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld\n", hard, soft); - } - - hard = (newlim->d_fieldmask & FS_DQ_IHARD) ? - (xfs_qcnt_t) newlim->d_ino_hardlimit : - be64_to_cpu(ddq->d_ino_hardlimit); - soft = (newlim->d_fieldmask & FS_DQ_ISOFT) ? - (xfs_qcnt_t) newlim->d_ino_softlimit : - be64_to_cpu(ddq->d_ino_softlimit); - if (hard == 0 || hard >= soft) { - ddq->d_ino_hardlimit = cpu_to_be64(hard); - ddq->d_ino_softlimit = cpu_to_be64(soft); - if (id == 0) { - q->qi_ihardlimit = hard; - q->qi_isoftlimit = soft; - } - } else { - xfs_debug(mp, "ihard %Ld < isoft %Ld\n", hard, soft); - } - - /* - * Update warnings counter(s) if requested - */ - if (newlim->d_fieldmask & FS_DQ_BWARNS) - ddq->d_bwarns = cpu_to_be16(newlim->d_bwarns); - if (newlim->d_fieldmask & FS_DQ_IWARNS) - ddq->d_iwarns = cpu_to_be16(newlim->d_iwarns); - if (newlim->d_fieldmask & FS_DQ_RTBWARNS) - ddq->d_rtbwarns = cpu_to_be16(newlim->d_rtbwarns); - - if (id == 0) { - /* - * Timelimits for the super user set the relative time - * the other users can be over quota for this file system. - * If it is zero a default is used. Ditto for the default - * soft and hard limit values (already done, above), and - * for warnings. - */ - if (newlim->d_fieldmask & FS_DQ_BTIMER) { - q->qi_btimelimit = newlim->d_btimer; - ddq->d_btimer = cpu_to_be32(newlim->d_btimer); - } - if (newlim->d_fieldmask & FS_DQ_ITIMER) { - q->qi_itimelimit = newlim->d_itimer; - ddq->d_itimer = cpu_to_be32(newlim->d_itimer); - } - if (newlim->d_fieldmask & FS_DQ_RTBTIMER) { - q->qi_rtbtimelimit = newlim->d_rtbtimer; - ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer); - } - if (newlim->d_fieldmask & FS_DQ_BWARNS) - q->qi_bwarnlimit = newlim->d_bwarns; - if (newlim->d_fieldmask & FS_DQ_IWARNS) - q->qi_iwarnlimit = newlim->d_iwarns; - if (newlim->d_fieldmask & FS_DQ_RTBWARNS) - q->qi_rtbwarnlimit = newlim->d_rtbwarns; - } else { - /* - * If the user is now over quota, start the timelimit. - * The user will not be 'warned'. - * Note that we keep the timers ticking, whether enforcement - * is on or off. We don't really want to bother with iterating - * over all ondisk dquots and turning the timers on/off. - */ - xfs_qm_adjust_dqtimers(mp, ddq); - } - dqp->dq_flags |= XFS_DQ_DIRTY; - xfs_trans_log_dquot(tp, dqp); - - error = xfs_trans_commit(tp, 0); - xfs_qm_dqrele(dqp); - - out_unlock: - mutex_unlock(&q->qi_quotaofflock); - return error; -} - -int -xfs_qm_scall_getquota( - xfs_mount_t *mp, - xfs_dqid_t id, - uint type, - fs_disk_quota_t *out) -{ - xfs_dquot_t *dqp; - int error; - - /* - * Try to get the dquot. We don't want it allocated on disk, so - * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't - * exist, we'll get ENOENT back. - */ - if ((error = xfs_qm_dqget(mp, NULL, id, type, 0, &dqp))) { - return (error); - } - - /* - * If everything's NULL, this dquot doesn't quite exist as far as - * our utility programs are concerned. - */ - if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { - xfs_qm_dqput(dqp); - return XFS_ERROR(ENOENT); - } - /* - * Convert the disk dquot to the exportable format - */ - xfs_qm_export_dquot(mp, &dqp->q_core, out); - xfs_qm_dqput(dqp); - return (error ? XFS_ERROR(EFAULT) : 0); -} - - -STATIC int -xfs_qm_log_quotaoff_end( - xfs_mount_t *mp, - xfs_qoff_logitem_t *startqoff, - uint flags) -{ - xfs_trans_t *tp; - int error; - xfs_qoff_logitem_t *qoffi; - - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF_END); - - if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_qoff_logitem_t) * 2, - 0, 0, XFS_DEFAULT_LOG_COUNT))) { - xfs_trans_cancel(tp, 0); - return (error); - } - - qoffi = xfs_trans_get_qoff_item(tp, startqoff, - flags & XFS_ALL_QUOTA_ACCT); - xfs_trans_log_quotaoff_item(tp, qoffi); - - /* - * We have to make sure that the transaction is secure on disk before we - * return and actually stop quota accounting. So, make it synchronous. - * We don't care about quotoff's performance. - */ - xfs_trans_set_sync(tp); - error = xfs_trans_commit(tp, 0); - return (error); -} - - -STATIC int -xfs_qm_log_quotaoff( - xfs_mount_t *mp, - xfs_qoff_logitem_t **qoffstartp, - uint flags) -{ - xfs_trans_t *tp; - int error; - xfs_qoff_logitem_t *qoffi=NULL; - uint oldsbqflag=0; - - tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF); - if ((error = xfs_trans_reserve(tp, 0, - sizeof(xfs_qoff_logitem_t) * 2 + - mp->m_sb.sb_sectsize + 128, - 0, - 0, - XFS_DEFAULT_LOG_COUNT))) { - goto error0; - } - - qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT); - xfs_trans_log_quotaoff_item(tp, qoffi); - - spin_lock(&mp->m_sb_lock); - oldsbqflag = mp->m_sb.sb_qflags; - mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL; - spin_unlock(&mp->m_sb_lock); - - xfs_mod_sb(tp, XFS_SB_QFLAGS); - - /* - * We have to make sure that the transaction is secure on disk before we - * return and actually stop quota accounting. So, make it synchronous. - * We don't care about quotoff's performance. - */ - xfs_trans_set_sync(tp); - error = xfs_trans_commit(tp, 0); - -error0: - if (error) { - xfs_trans_cancel(tp, 0); - /* - * No one else is modifying sb_qflags, so this is OK. - * We still hold the quotaofflock. - */ - spin_lock(&mp->m_sb_lock); - mp->m_sb.sb_qflags = oldsbqflag; - spin_unlock(&mp->m_sb_lock); - } - *qoffstartp = qoffi; - return (error); -} - - -/* - * Translate an internal style on-disk-dquot to the exportable format. - * The main differences are that the counters/limits are all in Basic - * Blocks (BBs) instead of the internal FSBs, and all on-disk data has - * to be converted to the native endianness. - */ -STATIC void -xfs_qm_export_dquot( - xfs_mount_t *mp, - xfs_disk_dquot_t *src, - struct fs_disk_quota *dst) -{ - memset(dst, 0, sizeof(*dst)); - dst->d_version = FS_DQUOT_VERSION; /* different from src->d_version */ - dst->d_flags = xfs_qm_export_qtype_flags(src->d_flags); - dst->d_id = be32_to_cpu(src->d_id); - dst->d_blk_hardlimit = - XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_hardlimit)); - dst->d_blk_softlimit = - XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_softlimit)); - dst->d_ino_hardlimit = be64_to_cpu(src->d_ino_hardlimit); - dst->d_ino_softlimit = be64_to_cpu(src->d_ino_softlimit); - dst->d_bcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_bcount)); - dst->d_icount = be64_to_cpu(src->d_icount); - dst->d_btimer = be32_to_cpu(src->d_btimer); - dst->d_itimer = be32_to_cpu(src->d_itimer); - dst->d_iwarns = be16_to_cpu(src->d_iwarns); - dst->d_bwarns = be16_to_cpu(src->d_bwarns); - dst->d_rtb_hardlimit = - XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_hardlimit)); - dst->d_rtb_softlimit = - XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_softlimit)); - dst->d_rtbcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtbcount)); - dst->d_rtbtimer = be32_to_cpu(src->d_rtbtimer); - dst->d_rtbwarns = be16_to_cpu(src->d_rtbwarns); - - /* - * Internally, we don't reset all the timers when quota enforcement - * gets turned off. No need to confuse the user level code, - * so return zeroes in that case. - */ - if ((!XFS_IS_UQUOTA_ENFORCED(mp) && src->d_flags == XFS_DQ_USER) || - (!XFS_IS_OQUOTA_ENFORCED(mp) && - (src->d_flags & (XFS_DQ_PROJ | XFS_DQ_GROUP)))) { - dst->d_btimer = 0; - dst->d_itimer = 0; - dst->d_rtbtimer = 0; - } - -#ifdef DEBUG - if (((XFS_IS_UQUOTA_ENFORCED(mp) && dst->d_flags == FS_USER_QUOTA) || - (XFS_IS_OQUOTA_ENFORCED(mp) && - (dst->d_flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)))) && - dst->d_id != 0) { - if (((int) dst->d_bcount >= (int) dst->d_blk_softlimit) && - (dst->d_blk_softlimit > 0)) { - ASSERT(dst->d_btimer != 0); - } - if (((int) dst->d_icount >= (int) dst->d_ino_softlimit) && - (dst->d_ino_softlimit > 0)) { - ASSERT(dst->d_itimer != 0); - } - } -#endif -} - -STATIC uint -xfs_qm_export_qtype_flags( - uint flags) -{ - /* - * Can't be more than one, or none. - */ - ASSERT((flags & (FS_PROJ_QUOTA | FS_USER_QUOTA)) != - (FS_PROJ_QUOTA | FS_USER_QUOTA)); - ASSERT((flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)) != - (FS_PROJ_QUOTA | FS_GROUP_QUOTA)); - ASSERT((flags & (FS_USER_QUOTA | FS_GROUP_QUOTA)) != - (FS_USER_QUOTA | FS_GROUP_QUOTA)); - ASSERT((flags & (FS_PROJ_QUOTA|FS_USER_QUOTA|FS_GROUP_QUOTA)) != 0); - - return (flags & XFS_DQ_USER) ? - FS_USER_QUOTA : (flags & XFS_DQ_PROJ) ? - FS_PROJ_QUOTA : FS_GROUP_QUOTA; -} - -STATIC uint -xfs_qm_export_flags( - uint flags) -{ - uint uflags; - - uflags = 0; - if (flags & XFS_UQUOTA_ACCT) - uflags |= FS_QUOTA_UDQ_ACCT; - if (flags & XFS_PQUOTA_ACCT) - uflags |= FS_QUOTA_PDQ_ACCT; - if (flags & XFS_GQUOTA_ACCT) - uflags |= FS_QUOTA_GDQ_ACCT; - if (flags & XFS_UQUOTA_ENFD) - uflags |= FS_QUOTA_UDQ_ENFD; - if (flags & (XFS_OQUOTA_ENFD)) { - uflags |= (flags & XFS_GQUOTA_ACCT) ? - FS_QUOTA_GDQ_ENFD : FS_QUOTA_PDQ_ENFD; - } - return (uflags); -} - - -STATIC int -xfs_dqrele_inode( - struct xfs_inode *ip, - struct xfs_perag *pag, - int flags) -{ - /* skip quota inodes */ - if (ip == ip->i_mount->m_quotainfo->qi_uquotaip || - ip == ip->i_mount->m_quotainfo->qi_gquotaip) { - ASSERT(ip->i_udquot == NULL); - ASSERT(ip->i_gdquot == NULL); - return 0; - } - - xfs_ilock(ip, XFS_ILOCK_EXCL); - if ((flags & XFS_UQUOTA_ACCT) && ip->i_udquot) { - xfs_qm_dqrele(ip->i_udquot); - ip->i_udquot = NULL; - } - if (flags & (XFS_PQUOTA_ACCT|XFS_GQUOTA_ACCT) && ip->i_gdquot) { - xfs_qm_dqrele(ip->i_gdquot); - ip->i_gdquot = NULL; - } - xfs_iunlock(ip, XFS_ILOCK_EXCL); - return 0; -} - - -/* - * Go thru all the inodes in the file system, releasing their dquots. - * - * Note that the mount structure gets modified to indicate that quotas are off - * AFTER this, in the case of quotaoff. - */ -void -xfs_qm_dqrele_all_inodes( - struct xfs_mount *mp, - uint flags) -{ - ASSERT(mp->m_quotainfo); - xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags); -} diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h deleted file mode 100644 index 94a3d927d716..000000000000 --- a/fs/xfs/quota/xfs_quota_priv.h +++ /dev/null @@ -1,53 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_QUOTA_PRIV_H__ -#define __XFS_QUOTA_PRIV_H__ - -/* - * Number of bmaps that we ask from bmapi when doing a quotacheck. - * We make this restriction to keep the memory usage to a minimum. - */ -#define XFS_DQITER_MAP_SIZE 10 - -/* - * Hash into a bucket in the dquot hash table, based on . - */ -#define XFS_DQ_HASHVAL(mp, id) (((__psunsigned_t)(mp) + \ - (__psunsigned_t)(id)) & \ - (xfs_Gqm->qm_dqhashmask - 1)) -#define XFS_DQ_HASH(mp, id, type) (type == XFS_DQ_USER ? \ - (xfs_Gqm->qm_usr_dqhtable + \ - XFS_DQ_HASHVAL(mp, id)) : \ - (xfs_Gqm->qm_grp_dqhtable + \ - XFS_DQ_HASHVAL(mp, id))) -#define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \ - !dqp->q_core.d_blk_hardlimit && \ - !dqp->q_core.d_blk_softlimit && \ - !dqp->q_core.d_rtb_hardlimit && \ - !dqp->q_core.d_rtb_softlimit && \ - !dqp->q_core.d_ino_hardlimit && \ - !dqp->q_core.d_ino_softlimit && \ - !dqp->q_core.d_bcount && \ - !dqp->q_core.d_rtbcount && \ - !dqp->q_core.d_icount) - -#define DQFLAGTO_TYPESTR(d) (((d)->dq_flags & XFS_DQ_USER) ? "USR" : \ - (((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \ - (((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???"))) - -#endif /* __XFS_QUOTA_PRIV_H__ */ diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c deleted file mode 100644 index 4d00ee67792d..000000000000 --- a/fs/xfs/quota/xfs_trans_dquot.c +++ /dev/null @@ -1,890 +0,0 @@ -/* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include "xfs.h" -#include "xfs_fs.h" -#include "xfs_bit.h" -#include "xfs_log.h" -#include "xfs_inum.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_alloc.h" -#include "xfs_quota.h" -#include "xfs_mount.h" -#include "xfs_bmap_btree.h" -#include "xfs_inode.h" -#include "xfs_itable.h" -#include "xfs_bmap.h" -#include "xfs_rtalloc.h" -#include "xfs_error.h" -#include "xfs_attr.h" -#include "xfs_buf_item.h" -#include "xfs_trans_priv.h" -#include "xfs_qm.h" - -STATIC void xfs_trans_alloc_dqinfo(xfs_trans_t *); - -/* - * Add the locked dquot to the transaction. - * The dquot must be locked, and it cannot be associated with any - * transaction. - */ -void -xfs_trans_dqjoin( - xfs_trans_t *tp, - xfs_dquot_t *dqp) -{ - ASSERT(dqp->q_transp != tp); - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(dqp->q_logitem.qli_dquot == dqp); - - /* - * Get a log_item_desc to point at the new item. - */ - xfs_trans_add_item(tp, &dqp->q_logitem.qli_item); - - /* - * Initialize d_transp so we can later determine if this dquot is - * associated with this transaction. - */ - dqp->q_transp = tp; -} - - -/* - * This is called to mark the dquot as needing - * to be logged when the transaction is committed. The dquot must - * already be associated with the given transaction. - * Note that it marks the entire transaction as dirty. In the ordinary - * case, this gets called via xfs_trans_commit, after the transaction - * is already dirty. However, there's nothing stop this from getting - * called directly, as done by xfs_qm_scall_setqlim. Hence, the TRANS_DIRTY - * flag. - */ -void -xfs_trans_log_dquot( - xfs_trans_t *tp, - xfs_dquot_t *dqp) -{ - ASSERT(dqp->q_transp == tp); - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - - tp->t_flags |= XFS_TRANS_DIRTY; - dqp->q_logitem.qli_item.li_desc->lid_flags |= XFS_LID_DIRTY; -} - -/* - * Carry forward whatever is left of the quota blk reservation to - * the spanky new transaction - */ -void -xfs_trans_dup_dqinfo( - xfs_trans_t *otp, - xfs_trans_t *ntp) -{ - xfs_dqtrx_t *oq, *nq; - int i,j; - xfs_dqtrx_t *oqa, *nqa; - - if (!otp->t_dqinfo) - return; - - xfs_trans_alloc_dqinfo(ntp); - oqa = otp->t_dqinfo->dqa_usrdquots; - nqa = ntp->t_dqinfo->dqa_usrdquots; - - /* - * Because the quota blk reservation is carried forward, - * it is also necessary to carry forward the DQ_DIRTY flag. - */ - if(otp->t_flags & XFS_TRANS_DQ_DIRTY) - ntp->t_flags |= XFS_TRANS_DQ_DIRTY; - - for (j = 0; j < 2; j++) { - for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { - if (oqa[i].qt_dquot == NULL) - break; - oq = &oqa[i]; - nq = &nqa[i]; - - nq->qt_dquot = oq->qt_dquot; - nq->qt_bcount_delta = nq->qt_icount_delta = 0; - nq->qt_rtbcount_delta = 0; - - /* - * Transfer whatever is left of the reservations. - */ - nq->qt_blk_res = oq->qt_blk_res - oq->qt_blk_res_used; - oq->qt_blk_res = oq->qt_blk_res_used; - - nq->qt_rtblk_res = oq->qt_rtblk_res - - oq->qt_rtblk_res_used; - oq->qt_rtblk_res = oq->qt_rtblk_res_used; - - nq->qt_ino_res = oq->qt_ino_res - oq->qt_ino_res_used; - oq->qt_ino_res = oq->qt_ino_res_used; - - } - oqa = otp->t_dqinfo->dqa_grpdquots; - nqa = ntp->t_dqinfo->dqa_grpdquots; - } -} - -/* - * Wrap around mod_dquot to account for both user and group quotas. - */ -void -xfs_trans_mod_dquot_byino( - xfs_trans_t *tp, - xfs_inode_t *ip, - uint field, - long delta) -{ - xfs_mount_t *mp = tp->t_mountp; - - if (!XFS_IS_QUOTA_RUNNING(mp) || - !XFS_IS_QUOTA_ON(mp) || - ip->i_ino == mp->m_sb.sb_uquotino || - ip->i_ino == mp->m_sb.sb_gquotino) - return; - - if (tp->t_dqinfo == NULL) - xfs_trans_alloc_dqinfo(tp); - - if (XFS_IS_UQUOTA_ON(mp) && ip->i_udquot) - (void) xfs_trans_mod_dquot(tp, ip->i_udquot, field, delta); - if (XFS_IS_OQUOTA_ON(mp) && ip->i_gdquot) - (void) xfs_trans_mod_dquot(tp, ip->i_gdquot, field, delta); -} - -STATIC xfs_dqtrx_t * -xfs_trans_get_dqtrx( - xfs_trans_t *tp, - xfs_dquot_t *dqp) -{ - int i; - xfs_dqtrx_t *qa; - - qa = XFS_QM_ISUDQ(dqp) ? - tp->t_dqinfo->dqa_usrdquots : tp->t_dqinfo->dqa_grpdquots; - - for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { - if (qa[i].qt_dquot == NULL || - qa[i].qt_dquot == dqp) - return &qa[i]; - } - - return NULL; -} - -/* - * Make the changes in the transaction structure. - * The moral equivalent to xfs_trans_mod_sb(). - * We don't touch any fields in the dquot, so we don't care - * if it's locked or not (most of the time it won't be). - */ -void -xfs_trans_mod_dquot( - xfs_trans_t *tp, - xfs_dquot_t *dqp, - uint field, - long delta) -{ - xfs_dqtrx_t *qtrx; - - ASSERT(tp); - ASSERT(XFS_IS_QUOTA_RUNNING(tp->t_mountp)); - qtrx = NULL; - - if (tp->t_dqinfo == NULL) - xfs_trans_alloc_dqinfo(tp); - /* - * Find either the first free slot or the slot that belongs - * to this dquot. - */ - qtrx = xfs_trans_get_dqtrx(tp, dqp); - ASSERT(qtrx); - if (qtrx->qt_dquot == NULL) - qtrx->qt_dquot = dqp; - - switch (field) { - - /* - * regular disk blk reservation - */ - case XFS_TRANS_DQ_RES_BLKS: - qtrx->qt_blk_res += (ulong)delta; - break; - - /* - * inode reservation - */ - case XFS_TRANS_DQ_RES_INOS: - qtrx->qt_ino_res += (ulong)delta; - break; - - /* - * disk blocks used. - */ - case XFS_TRANS_DQ_BCOUNT: - if (qtrx->qt_blk_res && delta > 0) { - qtrx->qt_blk_res_used += (ulong)delta; - ASSERT(qtrx->qt_blk_res >= qtrx->qt_blk_res_used); - } - qtrx->qt_bcount_delta += delta; - break; - - case XFS_TRANS_DQ_DELBCOUNT: - qtrx->qt_delbcnt_delta += delta; - break; - - /* - * Inode Count - */ - case XFS_TRANS_DQ_ICOUNT: - if (qtrx->qt_ino_res && delta > 0) { - qtrx->qt_ino_res_used += (ulong)delta; - ASSERT(qtrx->qt_ino_res >= qtrx->qt_ino_res_used); - } - qtrx->qt_icount_delta += delta; - break; - - /* - * rtblk reservation - */ - case XFS_TRANS_DQ_RES_RTBLKS: - qtrx->qt_rtblk_res += (ulong)delta; - break; - - /* - * rtblk count - */ - case XFS_TRANS_DQ_RTBCOUNT: - if (qtrx->qt_rtblk_res && delta > 0) { - qtrx->qt_rtblk_res_used += (ulong)delta; - ASSERT(qtrx->qt_rtblk_res >= qtrx->qt_rtblk_res_used); - } - qtrx->qt_rtbcount_delta += delta; - break; - - case XFS_TRANS_DQ_DELRTBCOUNT: - qtrx->qt_delrtb_delta += delta; - break; - - default: - ASSERT(0); - } - tp->t_flags |= XFS_TRANS_DQ_DIRTY; -} - - -/* - * Given an array of dqtrx structures, lock all the dquots associated - * and join them to the transaction, provided they have been modified. - * We know that the highest number of dquots (of one type - usr OR grp), - * involved in a transaction is 2 and that both usr and grp combined - 3. - * So, we don't attempt to make this very generic. - */ -STATIC void -xfs_trans_dqlockedjoin( - xfs_trans_t *tp, - xfs_dqtrx_t *q) -{ - ASSERT(q[0].qt_dquot != NULL); - if (q[1].qt_dquot == NULL) { - xfs_dqlock(q[0].qt_dquot); - xfs_trans_dqjoin(tp, q[0].qt_dquot); - } else { - ASSERT(XFS_QM_TRANS_MAXDQS == 2); - xfs_dqlock2(q[0].qt_dquot, q[1].qt_dquot); - xfs_trans_dqjoin(tp, q[0].qt_dquot); - xfs_trans_dqjoin(tp, q[1].qt_dquot); - } -} - - -/* - * Called by xfs_trans_commit() and similar in spirit to - * xfs_trans_apply_sb_deltas(). - * Go thru all the dquots belonging to this transaction and modify the - * INCORE dquot to reflect the actual usages. - * Unreserve just the reservations done by this transaction. - * dquot is still left locked at exit. - */ -void -xfs_trans_apply_dquot_deltas( - xfs_trans_t *tp) -{ - int i, j; - xfs_dquot_t *dqp; - xfs_dqtrx_t *qtrx, *qa; - xfs_disk_dquot_t *d; - long totalbdelta; - long totalrtbdelta; - - if (!(tp->t_flags & XFS_TRANS_DQ_DIRTY)) - return; - - ASSERT(tp->t_dqinfo); - qa = tp->t_dqinfo->dqa_usrdquots; - for (j = 0; j < 2; j++) { - if (qa[0].qt_dquot == NULL) { - qa = tp->t_dqinfo->dqa_grpdquots; - continue; - } - - /* - * Lock all of the dquots and join them to the transaction. - */ - xfs_trans_dqlockedjoin(tp, qa); - - for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { - qtrx = &qa[i]; - /* - * The array of dquots is filled - * sequentially, not sparsely. - */ - if ((dqp = qtrx->qt_dquot) == NULL) - break; - - ASSERT(XFS_DQ_IS_LOCKED(dqp)); - ASSERT(dqp->q_transp == tp); - - /* - * adjust the actual number of blocks used - */ - d = &dqp->q_core; - - /* - * The issue here is - sometimes we don't make a blkquota - * reservation intentionally to be fair to users - * (when the amount is small). On the other hand, - * delayed allocs do make reservations, but that's - * outside of a transaction, so we have no - * idea how much was really reserved. - * So, here we've accumulated delayed allocation blks and - * non-delay blks. The assumption is that the - * delayed ones are always reserved (outside of a - * transaction), and the others may or may not have - * quota reservations. - */ - totalbdelta = qtrx->qt_bcount_delta + - qtrx->qt_delbcnt_delta; - totalrtbdelta = qtrx->qt_rtbcount_delta + - qtrx->qt_delrtb_delta; -#ifdef DEBUG - if (totalbdelta < 0) - ASSERT(be64_to_cpu(d->d_bcount) >= - -totalbdelta); - - if (totalrtbdelta < 0) - ASSERT(be64_to_cpu(d->d_rtbcount) >= - -totalrtbdelta); - - if (qtrx->qt_icount_delta < 0) - ASSERT(be64_to_cpu(d->d_icount) >= - -qtrx->qt_icount_delta); -#endif - if (totalbdelta) - be64_add_cpu(&d->d_bcount, (xfs_qcnt_t)totalbdelta); - - if (qtrx->qt_icount_delta) - be64_add_cpu(&d->d_icount, (xfs_qcnt_t)qtrx->qt_icount_delta); - - if (totalrtbdelta) - be64_add_cpu(&d->d_rtbcount, (xfs_qcnt_t)totalrtbdelta); - - /* - * Get any default limits in use. - * Start/reset the timer(s) if needed. - */ - if (d->d_id) { - xfs_qm_adjust_dqlimits(tp->t_mountp, d); - xfs_qm_adjust_dqtimers(tp->t_mountp, d); - } - - dqp->dq_flags |= XFS_DQ_DIRTY; - /* - * add this to the list of items to get logged - */ - xfs_trans_log_dquot(tp, dqp); - /* - * Take off what's left of the original reservation. - * In case of delayed allocations, there's no - * reservation that a transaction structure knows of. - */ - if (qtrx->qt_blk_res != 0) { - if (qtrx->qt_blk_res != qtrx->qt_blk_res_used) { - if (qtrx->qt_blk_res > - qtrx->qt_blk_res_used) - dqp->q_res_bcount -= (xfs_qcnt_t) - (qtrx->qt_blk_res - - qtrx->qt_blk_res_used); - else - dqp->q_res_bcount -= (xfs_qcnt_t) - (qtrx->qt_blk_res_used - - qtrx->qt_blk_res); - } - } else { - /* - * These blks were never reserved, either inside - * a transaction or outside one (in a delayed - * allocation). Also, this isn't always a - * negative number since we sometimes - * deliberately skip quota reservations. - */ - if (qtrx->qt_bcount_delta) { - dqp->q_res_bcount += - (xfs_qcnt_t)qtrx->qt_bcount_delta; - } - } - /* - * Adjust the RT reservation. - */ - if (qtrx->qt_rtblk_res != 0) { - if (qtrx->qt_rtblk_res != qtrx->qt_rtblk_res_used) { - if (qtrx->qt_rtblk_res > - qtrx->qt_rtblk_res_used) - dqp->q_res_rtbcount -= (xfs_qcnt_t) - (qtrx->qt_rtblk_res - - qtrx->qt_rtblk_res_used); - else - dqp->q_res_rtbcount -= (xfs_qcnt_t) - (qtrx->qt_rtblk_res_used - - qtrx->qt_rtblk_res); - } - } else { - if (qtrx->qt_rtbcount_delta) - dqp->q_res_rtbcount += - (xfs_qcnt_t)qtrx->qt_rtbcount_delta; - } - - /* - * Adjust the inode reservation. - */ - if (qtrx->qt_ino_res != 0) { - ASSERT(qtrx->qt_ino_res >= - qtrx->qt_ino_res_used); - if (qtrx->qt_ino_res > qtrx->qt_ino_res_used) - dqp->q_res_icount -= (xfs_qcnt_t) - (qtrx->qt_ino_res - - qtrx->qt_ino_res_used); - } else { - if (qtrx->qt_icount_delta) - dqp->q_res_icount += - (xfs_qcnt_t)qtrx->qt_icount_delta; - } - - ASSERT(dqp->q_res_bcount >= - be64_to_cpu(dqp->q_core.d_bcount)); - ASSERT(dqp->q_res_icount >= - be64_to_cpu(dqp->q_core.d_icount)); - ASSERT(dqp->q_res_rtbcount >= - be64_to_cpu(dqp->q_core.d_rtbcount)); - } - /* - * Do the group quotas next - */ - qa = tp->t_dqinfo->dqa_grpdquots; - } -} - -/* - * Release the reservations, and adjust the dquots accordingly. - * This is called only when the transaction is being aborted. If by - * any chance we have done dquot modifications incore (ie. deltas) already, - * we simply throw those away, since that's the expected behavior - * when a transaction is curtailed without a commit. - */ -void -xfs_trans_unreserve_and_mod_dquots( - xfs_trans_t *tp) -{ - int i, j; - xfs_dquot_t *dqp; - xfs_dqtrx_t *qtrx, *qa; - boolean_t locked; - - if (!tp->t_dqinfo || !(tp->t_flags & XFS_TRANS_DQ_DIRTY)) - return; - - qa = tp->t_dqinfo->dqa_usrdquots; - - for (j = 0; j < 2; j++) { - for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { - qtrx = &qa[i]; - /* - * We assume that the array of dquots is filled - * sequentially, not sparsely. - */ - if ((dqp = qtrx->qt_dquot) == NULL) - break; - /* - * Unreserve the original reservation. We don't care - * about the number of blocks used field, or deltas. - * Also we don't bother to zero the fields. - */ - locked = B_FALSE; - if (qtrx->qt_blk_res) { - xfs_dqlock(dqp); - locked = B_TRUE; - dqp->q_res_bcount -= - (xfs_qcnt_t)qtrx->qt_blk_res; - } - if (qtrx->qt_ino_res) { - if (!locked) { - xfs_dqlock(dqp); - locked = B_TRUE; - } - dqp->q_res_icount -= - (xfs_qcnt_t)qtrx->qt_ino_res; - } - - if (qtrx->qt_rtblk_res) { - if (!locked) { - xfs_dqlock(dqp); - locked = B_TRUE; - } - dqp->q_res_rtbcount -= - (xfs_qcnt_t)qtrx->qt_rtblk_res; - } - if (locked) - xfs_dqunlock(dqp); - - } - qa = tp->t_dqinfo->dqa_grpdquots; - } -} - -STATIC void -xfs_quota_warn( - struct xfs_mount *mp, - struct xfs_dquot *dqp, - int type) -{ - /* no warnings for project quotas - we just return ENOSPC later */ - if (dqp->dq_flags & XFS_DQ_PROJ) - return; - quota_send_warning((dqp->dq_flags & XFS_DQ_USER) ? USRQUOTA : GRPQUOTA, - be32_to_cpu(dqp->q_core.d_id), mp->m_super->s_dev, - type); -} - -/* - * This reserves disk blocks and inodes against a dquot. - * Flags indicate if the dquot is to be locked here and also - * if the blk reservation is for RT or regular blocks. - * Sending in XFS_QMOPT_FORCE_RES flag skips the quota check. - */ -STATIC int -xfs_trans_dqresv( - xfs_trans_t *tp, - xfs_mount_t *mp, - xfs_dquot_t *dqp, - long nblks, - long ninos, - uint flags) -{ - xfs_qcnt_t hardlimit; - xfs_qcnt_t softlimit; - time_t timer; - xfs_qwarncnt_t warns; - xfs_qwarncnt_t warnlimit; - xfs_qcnt_t count; - xfs_qcnt_t *resbcountp; - xfs_quotainfo_t *q = mp->m_quotainfo; - - - xfs_dqlock(dqp); - - if (flags & XFS_TRANS_DQ_RES_BLKS) { - hardlimit = be64_to_cpu(dqp->q_core.d_blk_hardlimit); - if (!hardlimit) - hardlimit = q->qi_bhardlimit; - softlimit = be64_to_cpu(dqp->q_core.d_blk_softlimit); - if (!softlimit) - softlimit = q->qi_bsoftlimit; - timer = be32_to_cpu(dqp->q_core.d_btimer); - warns = be16_to_cpu(dqp->q_core.d_bwarns); - warnlimit = dqp->q_mount->m_quotainfo->qi_bwarnlimit; - resbcountp = &dqp->q_res_bcount; - } else { - ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS); - hardlimit = be64_to_cpu(dqp->q_core.d_rtb_hardlimit); - if (!hardlimit) - hardlimit = q->qi_rtbhardlimit; - softlimit = be64_to_cpu(dqp->q_core.d_rtb_softlimit); - if (!softlimit) - softlimit = q->qi_rtbsoftlimit; - timer = be32_to_cpu(dqp->q_core.d_rtbtimer); - warns = be16_to_cpu(dqp->q_core.d_rtbwarns); - warnlimit = dqp->q_mount->m_quotainfo->qi_rtbwarnlimit; - resbcountp = &dqp->q_res_rtbcount; - } - - if ((flags & XFS_QMOPT_FORCE_RES) == 0 && - dqp->q_core.d_id && - ((XFS_IS_UQUOTA_ENFORCED(dqp->q_mount) && XFS_QM_ISUDQ(dqp)) || - (XFS_IS_OQUOTA_ENFORCED(dqp->q_mount) && - (XFS_QM_ISPDQ(dqp) || XFS_QM_ISGDQ(dqp))))) { - if (nblks > 0) { - /* - * dquot is locked already. See if we'd go over the - * hardlimit or exceed the timelimit if we allocate - * nblks. - */ - if (hardlimit > 0ULL && - hardlimit <= nblks + *resbcountp) { - xfs_quota_warn(mp, dqp, QUOTA_NL_BHARDWARN); - goto error_return; - } - if (softlimit > 0ULL && - softlimit <= nblks + *resbcountp) { - if ((timer != 0 && get_seconds() > timer) || - (warns != 0 && warns >= warnlimit)) { - xfs_quota_warn(mp, dqp, - QUOTA_NL_BSOFTLONGWARN); - goto error_return; - } - - xfs_quota_warn(mp, dqp, QUOTA_NL_BSOFTWARN); - } - } - if (ninos > 0) { - count = be64_to_cpu(dqp->q_core.d_icount); - timer = be32_to_cpu(dqp->q_core.d_itimer); - warns = be16_to_cpu(dqp->q_core.d_iwarns); - warnlimit = dqp->q_mount->m_quotainfo->qi_iwarnlimit; - hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit); - if (!hardlimit) - hardlimit = q->qi_ihardlimit; - softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit); - if (!softlimit) - softlimit = q->qi_isoftlimit; - - if (hardlimit > 0ULL && count >= hardlimit) { - xfs_quota_warn(mp, dqp, QUOTA_NL_IHARDWARN); - goto error_return; - } - if (softlimit > 0ULL && count >= softlimit) { - if ((timer != 0 && get_seconds() > timer) || - (warns != 0 && warns >= warnlimit)) { - xfs_quota_warn(mp, dqp, - QUOTA_NL_ISOFTLONGWARN); - goto error_return; - } - xfs_quota_warn(mp, dqp, QUOTA_NL_ISOFTWARN); - } - } - } - - /* - * Change the reservation, but not the actual usage. - * Note that q_res_bcount = q_core.d_bcount + resv - */ - (*resbcountp) += (xfs_qcnt_t)nblks; - if (ninos != 0) - dqp->q_res_icount += (xfs_qcnt_t)ninos; - - /* - * note the reservation amt in the trans struct too, - * so that the transaction knows how much was reserved by - * it against this particular dquot. - * We don't do this when we are reserving for a delayed allocation, - * because we don't have the luxury of a transaction envelope then. - */ - if (tp) { - ASSERT(tp->t_dqinfo); - ASSERT(flags & XFS_QMOPT_RESBLK_MASK); - if (nblks != 0) - xfs_trans_mod_dquot(tp, dqp, - flags & XFS_QMOPT_RESBLK_MASK, - nblks); - if (ninos != 0) - xfs_trans_mod_dquot(tp, dqp, - XFS_TRANS_DQ_RES_INOS, - ninos); - } - ASSERT(dqp->q_res_bcount >= be64_to_cpu(dqp->q_core.d_bcount)); - ASSERT(dqp->q_res_rtbcount >= be64_to_cpu(dqp->q_core.d_rtbcount)); - ASSERT(dqp->q_res_icount >= be64_to_cpu(dqp->q_core.d_icount)); - - xfs_dqunlock(dqp); - return 0; - -error_return: - xfs_dqunlock(dqp); - if (flags & XFS_QMOPT_ENOSPC) - return ENOSPC; - return EDQUOT; -} - - -/* - * Given dquot(s), make disk block and/or inode reservations against them. - * The fact that this does the reservation against both the usr and - * grp/prj quotas is important, because this follows a both-or-nothing - * approach. - * - * flags = XFS_QMOPT_FORCE_RES evades limit enforcement. Used by chown. - * XFS_QMOPT_ENOSPC returns ENOSPC not EDQUOT. Used by pquota. - * XFS_TRANS_DQ_RES_BLKS reserves regular disk blocks - * XFS_TRANS_DQ_RES_RTBLKS reserves realtime disk blocks - * dquots are unlocked on return, if they were not locked by caller. - */ -int -xfs_trans_reserve_quota_bydquots( - xfs_trans_t *tp, - xfs_mount_t *mp, - xfs_dquot_t *udqp, - xfs_dquot_t *gdqp, - long nblks, - long ninos, - uint flags) -{ - int resvd = 0, error; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - - if (tp && tp->t_dqinfo == NULL) - xfs_trans_alloc_dqinfo(tp); - - ASSERT(flags & XFS_QMOPT_RESBLK_MASK); - - if (udqp) { - error = xfs_trans_dqresv(tp, mp, udqp, nblks, ninos, - (flags & ~XFS_QMOPT_ENOSPC)); - if (error) - return error; - resvd = 1; - } - - if (gdqp) { - error = xfs_trans_dqresv(tp, mp, gdqp, nblks, ninos, flags); - if (error) { - /* - * can't do it, so backout previous reservation - */ - if (resvd) { - flags |= XFS_QMOPT_FORCE_RES; - xfs_trans_dqresv(tp, mp, udqp, - -nblks, -ninos, flags); - } - return error; - } - } - - /* - * Didn't change anything critical, so, no need to log - */ - return 0; -} - - -/* - * Lock the dquot and change the reservation if we can. - * This doesn't change the actual usage, just the reservation. - * The inode sent in is locked. - */ -int -xfs_trans_reserve_quota_nblks( - struct xfs_trans *tp, - struct xfs_inode *ip, - long nblks, - long ninos, - uint flags) -{ - struct xfs_mount *mp = ip->i_mount; - - if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) - return 0; - if (XFS_IS_PQUOTA_ON(mp)) - flags |= XFS_QMOPT_ENOSPC; - - ASSERT(ip->i_ino != mp->m_sb.sb_uquotino); - ASSERT(ip->i_ino != mp->m_sb.sb_gquotino); - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT((flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) == - XFS_TRANS_DQ_RES_RTBLKS || - (flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) == - XFS_TRANS_DQ_RES_BLKS); - - /* - * Reserve nblks against these dquots, with trans as the mediator. - */ - return xfs_trans_reserve_quota_bydquots(tp, mp, - ip->i_udquot, ip->i_gdquot, - nblks, ninos, flags); -} - -/* - * This routine is called to allocate a quotaoff log item. - */ -xfs_qoff_logitem_t * -xfs_trans_get_qoff_item( - xfs_trans_t *tp, - xfs_qoff_logitem_t *startqoff, - uint flags) -{ - xfs_qoff_logitem_t *q; - - ASSERT(tp != NULL); - - q = xfs_qm_qoff_logitem_init(tp->t_mountp, startqoff, flags); - ASSERT(q != NULL); - - /* - * Get a log_item_desc to point at the new item. - */ - xfs_trans_add_item(tp, &q->qql_item); - return q; -} - - -/* - * This is called to mark the quotaoff logitem as needing - * to be logged when the transaction is committed. The logitem must - * already be associated with the given transaction. - */ -void -xfs_trans_log_quotaoff_item( - xfs_trans_t *tp, - xfs_qoff_logitem_t *qlp) -{ - tp->t_flags |= XFS_TRANS_DIRTY; - qlp->qql_item.li_desc->lid_flags |= XFS_LID_DIRTY; -} - -STATIC void -xfs_trans_alloc_dqinfo( - xfs_trans_t *tp) -{ - tp->t_dqinfo = kmem_zone_zalloc(xfs_Gqm->qm_dqtrxzone, KM_SLEEP); -} - -void -xfs_trans_free_dqinfo( - xfs_trans_t *tp) -{ - if (!tp->t_dqinfo) - return; - kmem_zone_free(xfs_Gqm->qm_dqtrxzone, tp->t_dqinfo); - tp->t_dqinfo = NULL; -} diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c deleted file mode 100644 index b83f76b6d410..000000000000 --- a/fs/xfs/support/uuid.c +++ /dev/null @@ -1,63 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#include - -/* IRIX interpretation of an uuid_t */ -typedef struct { - __be32 uu_timelow; - __be16 uu_timemid; - __be16 uu_timehi; - __be16 uu_clockseq; - __be16 uu_node[3]; -} xfs_uu_t; - -/* - * uuid_getnodeuniq - obtain the node unique fields of a UUID. - * - * This is not in any way a standard or condoned UUID function; - * it just something that's needed for user-level file handles. - */ -void -uuid_getnodeuniq(uuid_t *uuid, int fsid [2]) -{ - xfs_uu_t *uup = (xfs_uu_t *)uuid; - - fsid[0] = (be16_to_cpu(uup->uu_clockseq) << 16) | - be16_to_cpu(uup->uu_timemid); - fsid[1] = be32_to_cpu(uup->uu_timelow); -} - -int -uuid_is_nil(uuid_t *uuid) -{ - int i; - char *cp = (char *)uuid; - - if (uuid == NULL) - return 0; - /* implied check of version number here... */ - for (i = 0; i < sizeof *uuid; i++) - if (*cp++) return 0; /* not nil */ - return 1; /* is nil */ -} - -int -uuid_equal(uuid_t *uuid1, uuid_t *uuid2) -{ - return memcmp(uuid1, uuid2, sizeof(uuid_t)) ? 0 : 1; -} diff --git a/fs/xfs/support/uuid.h b/fs/xfs/support/uuid.h deleted file mode 100644 index 4732d71262cc..000000000000 --- a/fs/xfs/support/uuid.h +++ /dev/null @@ -1,29 +0,0 @@ -/* - * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write the Free Software Foundation, - * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - */ -#ifndef __XFS_SUPPORT_UUID_H__ -#define __XFS_SUPPORT_UUID_H__ - -typedef struct { - unsigned char __u_bits[16]; -} uuid_t; - -extern int uuid_is_nil(uuid_t *uuid); -extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2); -extern void uuid_getnodeuniq(uuid_t *uuid, int fsid [2]); - -#endif /* __XFS_SUPPORT_UUID_H__ */ diff --git a/fs/xfs/time.h b/fs/xfs/time.h new file mode 100644 index 000000000000..387e695a184c --- /dev/null +++ b/fs/xfs/time.h @@ -0,0 +1,36 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SUPPORT_TIME_H__ +#define __XFS_SUPPORT_TIME_H__ + +#include +#include + +typedef struct timespec timespec_t; + +static inline void delay(long ticks) +{ + schedule_timeout_uninterruptible(ticks); +} + +static inline void nanotime(struct timespec *tvp) +{ + *tvp = CURRENT_TIME; +} + +#endif /* __XFS_SUPPORT_TIME_H__ */ diff --git a/fs/xfs/uuid.c b/fs/xfs/uuid.c new file mode 100644 index 000000000000..b83f76b6d410 --- /dev/null +++ b/fs/xfs/uuid.c @@ -0,0 +1,63 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include + +/* IRIX interpretation of an uuid_t */ +typedef struct { + __be32 uu_timelow; + __be16 uu_timemid; + __be16 uu_timehi; + __be16 uu_clockseq; + __be16 uu_node[3]; +} xfs_uu_t; + +/* + * uuid_getnodeuniq - obtain the node unique fields of a UUID. + * + * This is not in any way a standard or condoned UUID function; + * it just something that's needed for user-level file handles. + */ +void +uuid_getnodeuniq(uuid_t *uuid, int fsid [2]) +{ + xfs_uu_t *uup = (xfs_uu_t *)uuid; + + fsid[0] = (be16_to_cpu(uup->uu_clockseq) << 16) | + be16_to_cpu(uup->uu_timemid); + fsid[1] = be32_to_cpu(uup->uu_timelow); +} + +int +uuid_is_nil(uuid_t *uuid) +{ + int i; + char *cp = (char *)uuid; + + if (uuid == NULL) + return 0; + /* implied check of version number here... */ + for (i = 0; i < sizeof *uuid; i++) + if (*cp++) return 0; /* not nil */ + return 1; /* is nil */ +} + +int +uuid_equal(uuid_t *uuid1, uuid_t *uuid2) +{ + return memcmp(uuid1, uuid2, sizeof(uuid_t)) ? 0 : 1; +} diff --git a/fs/xfs/uuid.h b/fs/xfs/uuid.h new file mode 100644 index 000000000000..4732d71262cc --- /dev/null +++ b/fs/xfs/uuid.h @@ -0,0 +1,29 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SUPPORT_UUID_H__ +#define __XFS_SUPPORT_UUID_H__ + +typedef struct { + unsigned char __u_bits[16]; +} uuid_t; + +extern int uuid_is_nil(uuid_t *uuid); +extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2); +extern void uuid_getnodeuniq(uuid_t *uuid, int fsid [2]); + +#endif /* __XFS_SUPPORT_UUID_H__ */ diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c new file mode 100644 index 000000000000..b6c4b3795c4a --- /dev/null +++ b/fs/xfs/xfs_acl.c @@ -0,0 +1,420 @@ +/* + * Copyright (c) 2008, Christoph Hellwig + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_acl.h" +#include "xfs_attr.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_vnodeops.h" +#include "xfs_trace.h" +#include +#include +#include + + +/* + * Locking scheme: + * - all ACL updates are protected by inode->i_mutex, which is taken before + * calling into this file. + */ + +STATIC struct posix_acl * +xfs_acl_from_disk(struct xfs_acl *aclp) +{ + struct posix_acl_entry *acl_e; + struct posix_acl *acl; + struct xfs_acl_entry *ace; + int count, i; + + count = be32_to_cpu(aclp->acl_cnt); + + acl = posix_acl_alloc(count, GFP_KERNEL); + if (!acl) + return ERR_PTR(-ENOMEM); + + for (i = 0; i < count; i++) { + acl_e = &acl->a_entries[i]; + ace = &aclp->acl_entry[i]; + + /* + * The tag is 32 bits on disk and 16 bits in core. + * + * Because every access to it goes through the core + * format first this is not a problem. + */ + acl_e->e_tag = be32_to_cpu(ace->ae_tag); + acl_e->e_perm = be16_to_cpu(ace->ae_perm); + + switch (acl_e->e_tag) { + case ACL_USER: + case ACL_GROUP: + acl_e->e_id = be32_to_cpu(ace->ae_id); + break; + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + acl_e->e_id = ACL_UNDEFINED_ID; + break; + default: + goto fail; + } + } + return acl; + +fail: + posix_acl_release(acl); + return ERR_PTR(-EINVAL); +} + +STATIC void +xfs_acl_to_disk(struct xfs_acl *aclp, const struct posix_acl *acl) +{ + const struct posix_acl_entry *acl_e; + struct xfs_acl_entry *ace; + int i; + + aclp->acl_cnt = cpu_to_be32(acl->a_count); + for (i = 0; i < acl->a_count; i++) { + ace = &aclp->acl_entry[i]; + acl_e = &acl->a_entries[i]; + + ace->ae_tag = cpu_to_be32(acl_e->e_tag); + ace->ae_id = cpu_to_be32(acl_e->e_id); + ace->ae_perm = cpu_to_be16(acl_e->e_perm); + } +} + +struct posix_acl * +xfs_get_acl(struct inode *inode, int type) +{ + struct xfs_inode *ip = XFS_I(inode); + struct posix_acl *acl; + struct xfs_acl *xfs_acl; + int len = sizeof(struct xfs_acl); + unsigned char *ea_name; + int error; + + acl = get_cached_acl(inode, type); + if (acl != ACL_NOT_CACHED) + return acl; + + trace_xfs_get_acl(ip); + + switch (type) { + case ACL_TYPE_ACCESS: + ea_name = SGI_ACL_FILE; + break; + case ACL_TYPE_DEFAULT: + ea_name = SGI_ACL_DEFAULT; + break; + default: + BUG(); + } + + /* + * If we have a cached ACLs value just return it, not need to + * go out to the disk. + */ + + xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); + if (!xfs_acl) + return ERR_PTR(-ENOMEM); + + error = -xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl, + &len, ATTR_ROOT); + if (error) { + /* + * If the attribute doesn't exist make sure we have a negative + * cache entry, for any other error assume it is transient and + * leave the cache entry as ACL_NOT_CACHED. + */ + if (error == -ENOATTR) { + acl = NULL; + goto out_update_cache; + } + goto out; + } + + acl = xfs_acl_from_disk(xfs_acl); + if (IS_ERR(acl)) + goto out; + + out_update_cache: + set_cached_acl(inode, type, acl); + out: + kfree(xfs_acl); + return acl; +} + +STATIC int +xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) +{ + struct xfs_inode *ip = XFS_I(inode); + unsigned char *ea_name; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + switch (type) { + case ACL_TYPE_ACCESS: + ea_name = SGI_ACL_FILE; + break; + case ACL_TYPE_DEFAULT: + if (!S_ISDIR(inode->i_mode)) + return acl ? -EACCES : 0; + ea_name = SGI_ACL_DEFAULT; + break; + default: + return -EINVAL; + } + + if (acl) { + struct xfs_acl *xfs_acl; + int len; + + xfs_acl = kzalloc(sizeof(struct xfs_acl), GFP_KERNEL); + if (!xfs_acl) + return -ENOMEM; + + xfs_acl_to_disk(xfs_acl, acl); + len = sizeof(struct xfs_acl) - + (sizeof(struct xfs_acl_entry) * + (XFS_ACL_MAX_ENTRIES - acl->a_count)); + + error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl, + len, ATTR_ROOT); + + kfree(xfs_acl); + } else { + /* + * A NULL ACL argument means we want to remove the ACL. + */ + error = -xfs_attr_remove(ip, ea_name, ATTR_ROOT); + + /* + * If the attribute didn't exist to start with that's fine. + */ + if (error == -ENOATTR) + error = 0; + } + + if (!error) + set_cached_acl(inode, type, acl); + return error; +} + +static int +xfs_set_mode(struct inode *inode, umode_t mode) +{ + int error = 0; + + if (mode != inode->i_mode) { + struct iattr iattr; + + iattr.ia_valid = ATTR_MODE | ATTR_CTIME; + iattr.ia_mode = mode; + iattr.ia_ctime = current_fs_time(inode->i_sb); + + error = -xfs_setattr_nonsize(XFS_I(inode), &iattr, XFS_ATTR_NOACL); + } + + return error; +} + +static int +xfs_acl_exists(struct inode *inode, unsigned char *name) +{ + int len = sizeof(struct xfs_acl); + + return (xfs_attr_get(XFS_I(inode), name, NULL, &len, + ATTR_ROOT|ATTR_KERNOVAL) == 0); +} + +int +posix_acl_access_exists(struct inode *inode) +{ + return xfs_acl_exists(inode, SGI_ACL_FILE); +} + +int +posix_acl_default_exists(struct inode *inode) +{ + if (!S_ISDIR(inode->i_mode)) + return 0; + return xfs_acl_exists(inode, SGI_ACL_DEFAULT); +} + +/* + * No need for i_mutex because the inode is not yet exposed to the VFS. + */ +int +xfs_inherit_acl(struct inode *inode, struct posix_acl *acl) +{ + umode_t mode = inode->i_mode; + int error = 0, inherit = 0; + + if (S_ISDIR(inode->i_mode)) { + error = xfs_set_acl(inode, ACL_TYPE_DEFAULT, acl); + if (error) + goto out; + } + + error = posix_acl_create(&acl, GFP_KERNEL, &mode); + if (error < 0) + return error; + + /* + * If posix_acl_create returns a positive value we need to + * inherit a permission that can't be represented using the Unix + * mode bits and we actually need to set an ACL. + */ + if (error > 0) + inherit = 1; + + error = xfs_set_mode(inode, mode); + if (error) + goto out; + + if (inherit) + error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl); + +out: + posix_acl_release(acl); + return error; +} + +int +xfs_acl_chmod(struct inode *inode) +{ + struct posix_acl *acl; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + acl = xfs_get_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl) || !acl) + return PTR_ERR(acl); + + error = posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode); + if (error) + return error; + + error = xfs_set_acl(inode, ACL_TYPE_ACCESS, acl); + posix_acl_release(acl); + return error; +} + +static int +xfs_xattr_acl_get(struct dentry *dentry, const char *name, + void *value, size_t size, int type) +{ + struct posix_acl *acl; + int error; + + acl = xfs_get_acl(dentry->d_inode, type); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl == NULL) + return -ENODATA; + + error = posix_acl_to_xattr(acl, value, size); + posix_acl_release(acl); + + return error; +} + +static int +xfs_xattr_acl_set(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags, int type) +{ + struct inode *inode = dentry->d_inode; + struct posix_acl *acl = NULL; + int error = 0; + + if (flags & XATTR_CREATE) + return -EINVAL; + if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode)) + return value ? -EACCES : 0; + if ((current_fsuid() != inode->i_uid) && !capable(CAP_FOWNER)) + return -EPERM; + + if (!value) + goto set_acl; + + acl = posix_acl_from_xattr(value, size); + if (!acl) { + /* + * acl_set_file(3) may request that we set default ACLs with + * zero length -- defend (gracefully) against that here. + */ + goto out; + } + if (IS_ERR(acl)) { + error = PTR_ERR(acl); + goto out; + } + + error = posix_acl_valid(acl); + if (error) + goto out_release; + + error = -EINVAL; + if (acl->a_count > XFS_ACL_MAX_ENTRIES) + goto out_release; + + if (type == ACL_TYPE_ACCESS) { + umode_t mode = inode->i_mode; + error = posix_acl_equiv_mode(acl, &mode); + + if (error <= 0) { + posix_acl_release(acl); + acl = NULL; + + if (error < 0) + return error; + } + + error = xfs_set_mode(inode, mode); + if (error) + goto out_release; + } + + set_acl: + error = xfs_set_acl(inode, type, acl); + out_release: + posix_acl_release(acl); + out: + return error; +} + +const struct xattr_handler xfs_xattr_acl_access_handler = { + .prefix = POSIX_ACL_XATTR_ACCESS, + .flags = ACL_TYPE_ACCESS, + .get = xfs_xattr_acl_get, + .set = xfs_xattr_acl_set, +}; + +const struct xattr_handler xfs_xattr_acl_default_handler = { + .prefix = POSIX_ACL_XATTR_DEFAULT, + .flags = ACL_TYPE_DEFAULT, + .get = xfs_xattr_acl_get, + .set = xfs_xattr_acl_set, +}; diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c new file mode 100644 index 000000000000..63e971e2b837 --- /dev/null +++ b/fs/xfs/xfs_aops.c @@ -0,0 +1,1499 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_trans.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_alloc.h" +#include "xfs_error.h" +#include "xfs_rw.h" +#include "xfs_iomap.h" +#include "xfs_vnodeops.h" +#include "xfs_trace.h" +#include "xfs_bmap.h" +#include +#include +#include +#include + + +/* + * Prime number of hash buckets since address is used as the key. + */ +#define NVSYNC 37 +#define to_ioend_wq(v) (&xfs_ioend_wq[((unsigned long)v) % NVSYNC]) +static wait_queue_head_t xfs_ioend_wq[NVSYNC]; + +void __init +xfs_ioend_init(void) +{ + int i; + + for (i = 0; i < NVSYNC; i++) + init_waitqueue_head(&xfs_ioend_wq[i]); +} + +void +xfs_ioend_wait( + xfs_inode_t *ip) +{ + wait_queue_head_t *wq = to_ioend_wq(ip); + + wait_event(*wq, (atomic_read(&ip->i_iocount) == 0)); +} + +STATIC void +xfs_ioend_wake( + xfs_inode_t *ip) +{ + if (atomic_dec_and_test(&ip->i_iocount)) + wake_up(to_ioend_wq(ip)); +} + +void +xfs_count_page_state( + struct page *page, + int *delalloc, + int *unwritten) +{ + struct buffer_head *bh, *head; + + *delalloc = *unwritten = 0; + + bh = head = page_buffers(page); + do { + if (buffer_unwritten(bh)) + (*unwritten) = 1; + else if (buffer_delay(bh)) + (*delalloc) = 1; + } while ((bh = bh->b_this_page) != head); +} + +STATIC struct block_device * +xfs_find_bdev_for_inode( + struct inode *inode) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + + if (XFS_IS_REALTIME_INODE(ip)) + return mp->m_rtdev_targp->bt_bdev; + else + return mp->m_ddev_targp->bt_bdev; +} + +/* + * We're now finished for good with this ioend structure. + * Update the page state via the associated buffer_heads, + * release holds on the inode and bio, and finally free + * up memory. Do not use the ioend after this. + */ +STATIC void +xfs_destroy_ioend( + xfs_ioend_t *ioend) +{ + struct buffer_head *bh, *next; + struct xfs_inode *ip = XFS_I(ioend->io_inode); + + for (bh = ioend->io_buffer_head; bh; bh = next) { + next = bh->b_private; + bh->b_end_io(bh, !ioend->io_error); + } + + /* + * Volume managers supporting multiple paths can send back ENODEV + * when the final path disappears. In this case continuing to fill + * the page cache with dirty data which cannot be written out is + * evil, so prevent that. + */ + if (unlikely(ioend->io_error == -ENODEV)) { + xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, + __FILE__, __LINE__); + } + + xfs_ioend_wake(ip); + mempool_free(ioend, xfs_ioend_pool); +} + +/* + * If the end of the current ioend is beyond the current EOF, + * return the new EOF value, otherwise zero. + */ +STATIC xfs_fsize_t +xfs_ioend_new_eof( + xfs_ioend_t *ioend) +{ + xfs_inode_t *ip = XFS_I(ioend->io_inode); + xfs_fsize_t isize; + xfs_fsize_t bsize; + + bsize = ioend->io_offset + ioend->io_size; + isize = MAX(ip->i_size, ip->i_new_size); + isize = MIN(isize, bsize); + return isize > ip->i_d.di_size ? isize : 0; +} + +/* + * Update on-disk file size now that data has been written to disk. The + * current in-memory file size is i_size. If a write is beyond eof i_new_size + * will be the intended file size until i_size is updated. If this write does + * not extend all the way to the valid file size then restrict this update to + * the end of the write. + * + * This function does not block as blocking on the inode lock in IO completion + * can lead to IO completion order dependency deadlocks.. If it can't get the + * inode ilock it will return EAGAIN. Callers must handle this. + */ +STATIC int +xfs_setfilesize( + xfs_ioend_t *ioend) +{ + xfs_inode_t *ip = XFS_I(ioend->io_inode); + xfs_fsize_t isize; + + if (unlikely(ioend->io_error)) + return 0; + + if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) + return EAGAIN; + + isize = xfs_ioend_new_eof(ioend); + if (isize) { + trace_xfs_setfilesize(ip, ioend->io_offset, ioend->io_size); + ip->i_d.di_size = isize; + xfs_mark_inode_dirty(ip); + } + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return 0; +} + +/* + * Schedule IO completion handling on the final put of an ioend. + */ +STATIC void +xfs_finish_ioend( + struct xfs_ioend *ioend) +{ + if (atomic_dec_and_test(&ioend->io_remaining)) { + if (ioend->io_type == IO_UNWRITTEN) + queue_work(xfsconvertd_workqueue, &ioend->io_work); + else + queue_work(xfsdatad_workqueue, &ioend->io_work); + } +} + +/* + * IO write completion. + */ +STATIC void +xfs_end_io( + struct work_struct *work) +{ + xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work); + struct xfs_inode *ip = XFS_I(ioend->io_inode); + int error = 0; + + /* + * For unwritten extents we need to issue transactions to convert a + * range to normal written extens after the data I/O has finished. + */ + if (ioend->io_type == IO_UNWRITTEN && + likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) { + + error = xfs_iomap_write_unwritten(ip, ioend->io_offset, + ioend->io_size); + if (error) + ioend->io_error = error; + } + + /* + * We might have to update the on-disk file size after extending + * writes. + */ + error = xfs_setfilesize(ioend); + ASSERT(!error || error == EAGAIN); + + /* + * If we didn't complete processing of the ioend, requeue it to the + * tail of the workqueue for another attempt later. Otherwise destroy + * it. + */ + if (error == EAGAIN) { + atomic_inc(&ioend->io_remaining); + xfs_finish_ioend(ioend); + /* ensure we don't spin on blocked ioends */ + delay(1); + } else { + if (ioend->io_iocb) + aio_complete(ioend->io_iocb, ioend->io_result, 0); + xfs_destroy_ioend(ioend); + } +} + +/* + * Call IO completion handling in caller context on the final put of an ioend. + */ +STATIC void +xfs_finish_ioend_sync( + struct xfs_ioend *ioend) +{ + if (atomic_dec_and_test(&ioend->io_remaining)) + xfs_end_io(&ioend->io_work); +} + +/* + * Allocate and initialise an IO completion structure. + * We need to track unwritten extent write completion here initially. + * We'll need to extend this for updating the ondisk inode size later + * (vs. incore size). + */ +STATIC xfs_ioend_t * +xfs_alloc_ioend( + struct inode *inode, + unsigned int type) +{ + xfs_ioend_t *ioend; + + ioend = mempool_alloc(xfs_ioend_pool, GFP_NOFS); + + /* + * Set the count to 1 initially, which will prevent an I/O + * completion callback from happening before we have started + * all the I/O from calling the completion routine too early. + */ + atomic_set(&ioend->io_remaining, 1); + ioend->io_error = 0; + ioend->io_list = NULL; + ioend->io_type = type; + ioend->io_inode = inode; + ioend->io_buffer_head = NULL; + ioend->io_buffer_tail = NULL; + atomic_inc(&XFS_I(ioend->io_inode)->i_iocount); + ioend->io_offset = 0; + ioend->io_size = 0; + ioend->io_iocb = NULL; + ioend->io_result = 0; + + INIT_WORK(&ioend->io_work, xfs_end_io); + return ioend; +} + +STATIC int +xfs_map_blocks( + struct inode *inode, + loff_t offset, + struct xfs_bmbt_irec *imap, + int type, + int nonblocking) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + ssize_t count = 1 << inode->i_blkbits; + xfs_fileoff_t offset_fsb, end_fsb; + int error = 0; + int bmapi_flags = XFS_BMAPI_ENTIRE; + int nimaps = 1; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -XFS_ERROR(EIO); + + if (type == IO_UNWRITTEN) + bmapi_flags |= XFS_BMAPI_IGSTATE; + + if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) { + if (nonblocking) + return -XFS_ERROR(EAGAIN); + xfs_ilock(ip, XFS_ILOCK_SHARED); + } + + ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE || + (ip->i_df.if_flags & XFS_IFEXTENTS)); + ASSERT(offset <= mp->m_maxioffset); + + if (offset + count > mp->m_maxioffset) + count = mp->m_maxioffset - offset; + end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count); + offset_fsb = XFS_B_TO_FSBT(mp, offset); + error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb, + bmapi_flags, NULL, 0, imap, &nimaps, NULL); + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + if (error) + return -XFS_ERROR(error); + + if (type == IO_DELALLOC && + (!nimaps || isnullstartblock(imap->br_startblock))) { + error = xfs_iomap_write_allocate(ip, offset, count, imap); + if (!error) + trace_xfs_map_blocks_alloc(ip, offset, count, type, imap); + return -XFS_ERROR(error); + } + +#ifdef DEBUG + if (type == IO_UNWRITTEN) { + ASSERT(nimaps); + ASSERT(imap->br_startblock != HOLESTARTBLOCK); + ASSERT(imap->br_startblock != DELAYSTARTBLOCK); + } +#endif + if (nimaps) + trace_xfs_map_blocks_found(ip, offset, count, type, imap); + return 0; +} + +STATIC int +xfs_imap_valid( + struct inode *inode, + struct xfs_bmbt_irec *imap, + xfs_off_t offset) +{ + offset >>= inode->i_blkbits; + + return offset >= imap->br_startoff && + offset < imap->br_startoff + imap->br_blockcount; +} + +/* + * BIO completion handler for buffered IO. + */ +STATIC void +xfs_end_bio( + struct bio *bio, + int error) +{ + xfs_ioend_t *ioend = bio->bi_private; + + ASSERT(atomic_read(&bio->bi_cnt) >= 1); + ioend->io_error = test_bit(BIO_UPTODATE, &bio->bi_flags) ? 0 : error; + + /* Toss bio and pass work off to an xfsdatad thread */ + bio->bi_private = NULL; + bio->bi_end_io = NULL; + bio_put(bio); + + xfs_finish_ioend(ioend); +} + +STATIC void +xfs_submit_ioend_bio( + struct writeback_control *wbc, + xfs_ioend_t *ioend, + struct bio *bio) +{ + atomic_inc(&ioend->io_remaining); + bio->bi_private = ioend; + bio->bi_end_io = xfs_end_bio; + + /* + * If the I/O is beyond EOF we mark the inode dirty immediately + * but don't update the inode size until I/O completion. + */ + if (xfs_ioend_new_eof(ioend)) + xfs_mark_inode_dirty(XFS_I(ioend->io_inode)); + + submit_bio(wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE, bio); +} + +STATIC struct bio * +xfs_alloc_ioend_bio( + struct buffer_head *bh) +{ + int nvecs = bio_get_nr_vecs(bh->b_bdev); + struct bio *bio = bio_alloc(GFP_NOIO, nvecs); + + ASSERT(bio->bi_private == NULL); + bio->bi_sector = bh->b_blocknr * (bh->b_size >> 9); + bio->bi_bdev = bh->b_bdev; + return bio; +} + +STATIC void +xfs_start_buffer_writeback( + struct buffer_head *bh) +{ + ASSERT(buffer_mapped(bh)); + ASSERT(buffer_locked(bh)); + ASSERT(!buffer_delay(bh)); + ASSERT(!buffer_unwritten(bh)); + + mark_buffer_async_write(bh); + set_buffer_uptodate(bh); + clear_buffer_dirty(bh); +} + +STATIC void +xfs_start_page_writeback( + struct page *page, + int clear_dirty, + int buffers) +{ + ASSERT(PageLocked(page)); + ASSERT(!PageWriteback(page)); + if (clear_dirty) + clear_page_dirty_for_io(page); + set_page_writeback(page); + unlock_page(page); + /* If no buffers on the page are to be written, finish it here */ + if (!buffers) + end_page_writeback(page); +} + +static inline int bio_add_buffer(struct bio *bio, struct buffer_head *bh) +{ + return bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh)); +} + +/* + * Submit all of the bios for all of the ioends we have saved up, covering the + * initial writepage page and also any probed pages. + * + * Because we may have multiple ioends spanning a page, we need to start + * writeback on all the buffers before we submit them for I/O. If we mark the + * buffers as we got, then we can end up with a page that only has buffers + * marked async write and I/O complete on can occur before we mark the other + * buffers async write. + * + * The end result of this is that we trip a bug in end_page_writeback() because + * we call it twice for the one page as the code in end_buffer_async_write() + * assumes that all buffers on the page are started at the same time. + * + * The fix is two passes across the ioend list - one to start writeback on the + * buffer_heads, and then submit them for I/O on the second pass. + */ +STATIC void +xfs_submit_ioend( + struct writeback_control *wbc, + xfs_ioend_t *ioend) +{ + xfs_ioend_t *head = ioend; + xfs_ioend_t *next; + struct buffer_head *bh; + struct bio *bio; + sector_t lastblock = 0; + + /* Pass 1 - start writeback */ + do { + next = ioend->io_list; + for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) + xfs_start_buffer_writeback(bh); + } while ((ioend = next) != NULL); + + /* Pass 2 - submit I/O */ + ioend = head; + do { + next = ioend->io_list; + bio = NULL; + + for (bh = ioend->io_buffer_head; bh; bh = bh->b_private) { + + if (!bio) { + retry: + bio = xfs_alloc_ioend_bio(bh); + } else if (bh->b_blocknr != lastblock + 1) { + xfs_submit_ioend_bio(wbc, ioend, bio); + goto retry; + } + + if (bio_add_buffer(bio, bh) != bh->b_size) { + xfs_submit_ioend_bio(wbc, ioend, bio); + goto retry; + } + + lastblock = bh->b_blocknr; + } + if (bio) + xfs_submit_ioend_bio(wbc, ioend, bio); + xfs_finish_ioend(ioend); + } while ((ioend = next) != NULL); +} + +/* + * Cancel submission of all buffer_heads so far in this endio. + * Toss the endio too. Only ever called for the initial page + * in a writepage request, so only ever one page. + */ +STATIC void +xfs_cancel_ioend( + xfs_ioend_t *ioend) +{ + xfs_ioend_t *next; + struct buffer_head *bh, *next_bh; + + do { + next = ioend->io_list; + bh = ioend->io_buffer_head; + do { + next_bh = bh->b_private; + clear_buffer_async_write(bh); + unlock_buffer(bh); + } while ((bh = next_bh) != NULL); + + xfs_ioend_wake(XFS_I(ioend->io_inode)); + mempool_free(ioend, xfs_ioend_pool); + } while ((ioend = next) != NULL); +} + +/* + * Test to see if we've been building up a completion structure for + * earlier buffers -- if so, we try to append to this ioend if we + * can, otherwise we finish off any current ioend and start another. + * Return true if we've finished the given ioend. + */ +STATIC void +xfs_add_to_ioend( + struct inode *inode, + struct buffer_head *bh, + xfs_off_t offset, + unsigned int type, + xfs_ioend_t **result, + int need_ioend) +{ + xfs_ioend_t *ioend = *result; + + if (!ioend || need_ioend || type != ioend->io_type) { + xfs_ioend_t *previous = *result; + + ioend = xfs_alloc_ioend(inode, type); + ioend->io_offset = offset; + ioend->io_buffer_head = bh; + ioend->io_buffer_tail = bh; + if (previous) + previous->io_list = ioend; + *result = ioend; + } else { + ioend->io_buffer_tail->b_private = bh; + ioend->io_buffer_tail = bh; + } + + bh->b_private = NULL; + ioend->io_size += bh->b_size; +} + +STATIC void +xfs_map_buffer( + struct inode *inode, + struct buffer_head *bh, + struct xfs_bmbt_irec *imap, + xfs_off_t offset) +{ + sector_t bn; + struct xfs_mount *m = XFS_I(inode)->i_mount; + xfs_off_t iomap_offset = XFS_FSB_TO_B(m, imap->br_startoff); + xfs_daddr_t iomap_bn = xfs_fsb_to_db(XFS_I(inode), imap->br_startblock); + + ASSERT(imap->br_startblock != HOLESTARTBLOCK); + ASSERT(imap->br_startblock != DELAYSTARTBLOCK); + + bn = (iomap_bn >> (inode->i_blkbits - BBSHIFT)) + + ((offset - iomap_offset) >> inode->i_blkbits); + + ASSERT(bn || XFS_IS_REALTIME_INODE(XFS_I(inode))); + + bh->b_blocknr = bn; + set_buffer_mapped(bh); +} + +STATIC void +xfs_map_at_offset( + struct inode *inode, + struct buffer_head *bh, + struct xfs_bmbt_irec *imap, + xfs_off_t offset) +{ + ASSERT(imap->br_startblock != HOLESTARTBLOCK); + ASSERT(imap->br_startblock != DELAYSTARTBLOCK); + + xfs_map_buffer(inode, bh, imap, offset); + set_buffer_mapped(bh); + clear_buffer_delay(bh); + clear_buffer_unwritten(bh); +} + +/* + * Test if a given page is suitable for writing as part of an unwritten + * or delayed allocate extent. + */ +STATIC int +xfs_is_delayed_page( + struct page *page, + unsigned int type) +{ + if (PageWriteback(page)) + return 0; + + if (page->mapping && page_has_buffers(page)) { + struct buffer_head *bh, *head; + int acceptable = 0; + + bh = head = page_buffers(page); + do { + if (buffer_unwritten(bh)) + acceptable = (type == IO_UNWRITTEN); + else if (buffer_delay(bh)) + acceptable = (type == IO_DELALLOC); + else if (buffer_dirty(bh) && buffer_mapped(bh)) + acceptable = (type == IO_OVERWRITE); + else + break; + } while ((bh = bh->b_this_page) != head); + + if (acceptable) + return 1; + } + + return 0; +} + +/* + * Allocate & map buffers for page given the extent map. Write it out. + * except for the original page of a writepage, this is called on + * delalloc/unwritten pages only, for the original page it is possible + * that the page has no mapping at all. + */ +STATIC int +xfs_convert_page( + struct inode *inode, + struct page *page, + loff_t tindex, + struct xfs_bmbt_irec *imap, + xfs_ioend_t **ioendp, + struct writeback_control *wbc) +{ + struct buffer_head *bh, *head; + xfs_off_t end_offset; + unsigned long p_offset; + unsigned int type; + int len, page_dirty; + int count = 0, done = 0, uptodate = 1; + xfs_off_t offset = page_offset(page); + + if (page->index != tindex) + goto fail; + if (!trylock_page(page)) + goto fail; + if (PageWriteback(page)) + goto fail_unlock_page; + if (page->mapping != inode->i_mapping) + goto fail_unlock_page; + if (!xfs_is_delayed_page(page, (*ioendp)->io_type)) + goto fail_unlock_page; + + /* + * page_dirty is initially a count of buffers on the page before + * EOF and is decremented as we move each into a cleanable state. + * + * Derivation: + * + * End offset is the highest offset that this page should represent. + * If we are on the last page, (end_offset & (PAGE_CACHE_SIZE - 1)) + * will evaluate non-zero and be less than PAGE_CACHE_SIZE and + * hence give us the correct page_dirty count. On any other page, + * it will be zero and in that case we need page_dirty to be the + * count of buffers on the page. + */ + end_offset = min_t(unsigned long long, + (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, + i_size_read(inode)); + + len = 1 << inode->i_blkbits; + p_offset = min_t(unsigned long, end_offset & (PAGE_CACHE_SIZE - 1), + PAGE_CACHE_SIZE); + p_offset = p_offset ? roundup(p_offset, len) : PAGE_CACHE_SIZE; + page_dirty = p_offset / len; + + bh = head = page_buffers(page); + do { + if (offset >= end_offset) + break; + if (!buffer_uptodate(bh)) + uptodate = 0; + if (!(PageUptodate(page) || buffer_uptodate(bh))) { + done = 1; + continue; + } + + if (buffer_unwritten(bh) || buffer_delay(bh) || + buffer_mapped(bh)) { + if (buffer_unwritten(bh)) + type = IO_UNWRITTEN; + else if (buffer_delay(bh)) + type = IO_DELALLOC; + else + type = IO_OVERWRITE; + + if (!xfs_imap_valid(inode, imap, offset)) { + done = 1; + continue; + } + + lock_buffer(bh); + if (type != IO_OVERWRITE) + xfs_map_at_offset(inode, bh, imap, offset); + xfs_add_to_ioend(inode, bh, offset, type, + ioendp, done); + + page_dirty--; + count++; + } else { + done = 1; + } + } while (offset += len, (bh = bh->b_this_page) != head); + + if (uptodate && bh == head) + SetPageUptodate(page); + + if (count) { + if (--wbc->nr_to_write <= 0 && + wbc->sync_mode == WB_SYNC_NONE) + done = 1; + } + xfs_start_page_writeback(page, !page_dirty, count); + + return done; + fail_unlock_page: + unlock_page(page); + fail: + return 1; +} + +/* + * Convert & write out a cluster of pages in the same extent as defined + * by mp and following the start page. + */ +STATIC void +xfs_cluster_write( + struct inode *inode, + pgoff_t tindex, + struct xfs_bmbt_irec *imap, + xfs_ioend_t **ioendp, + struct writeback_control *wbc, + pgoff_t tlast) +{ + struct pagevec pvec; + int done = 0, i; + + pagevec_init(&pvec, 0); + while (!done && tindex <= tlast) { + unsigned len = min_t(pgoff_t, PAGEVEC_SIZE, tlast - tindex + 1); + + if (!pagevec_lookup(&pvec, inode->i_mapping, tindex, len)) + break; + + for (i = 0; i < pagevec_count(&pvec); i++) { + done = xfs_convert_page(inode, pvec.pages[i], tindex++, + imap, ioendp, wbc); + if (done) + break; + } + + pagevec_release(&pvec); + cond_resched(); + } +} + +STATIC void +xfs_vm_invalidatepage( + struct page *page, + unsigned long offset) +{ + trace_xfs_invalidatepage(page->mapping->host, page, offset); + block_invalidatepage(page, offset); +} + +/* + * If the page has delalloc buffers on it, we need to punch them out before we + * invalidate the page. If we don't, we leave a stale delalloc mapping on the + * inode that can trip a BUG() in xfs_get_blocks() later on if a direct IO read + * is done on that same region - the delalloc extent is returned when none is + * supposed to be there. + * + * We prevent this by truncating away the delalloc regions on the page before + * invalidating it. Because they are delalloc, we can do this without needing a + * transaction. Indeed - if we get ENOSPC errors, we have to be able to do this + * truncation without a transaction as there is no space left for block + * reservation (typically why we see a ENOSPC in writeback). + * + * This is not a performance critical path, so for now just do the punching a + * buffer head at a time. + */ +STATIC void +xfs_aops_discard_page( + struct page *page) +{ + struct inode *inode = page->mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct buffer_head *bh, *head; + loff_t offset = page_offset(page); + + if (!xfs_is_delayed_page(page, IO_DELALLOC)) + goto out_invalidate; + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + goto out_invalidate; + + xfs_alert(ip->i_mount, + "page discard on page %p, inode 0x%llx, offset %llu.", + page, ip->i_ino, offset); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + bh = head = page_buffers(page); + do { + int error; + xfs_fileoff_t start_fsb; + + if (!buffer_delay(bh)) + goto next_buffer; + + start_fsb = XFS_B_TO_FSBT(ip->i_mount, offset); + error = xfs_bmap_punch_delalloc_range(ip, start_fsb, 1); + if (error) { + /* something screwed, just bail */ + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_alert(ip->i_mount, + "page discard unable to remove delalloc mapping."); + } + break; + } +next_buffer: + offset += 1 << inode->i_blkbits; + + } while ((bh = bh->b_this_page) != head); + + xfs_iunlock(ip, XFS_ILOCK_EXCL); +out_invalidate: + xfs_vm_invalidatepage(page, 0); + return; +} + +/* + * Write out a dirty page. + * + * For delalloc space on the page we need to allocate space and flush it. + * For unwritten space on the page we need to start the conversion to + * regular allocated space. + * For any other dirty buffer heads on the page we should flush them. + */ +STATIC int +xfs_vm_writepage( + struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct buffer_head *bh, *head; + struct xfs_bmbt_irec imap; + xfs_ioend_t *ioend = NULL, *iohead = NULL; + loff_t offset; + unsigned int type; + __uint64_t end_offset; + pgoff_t end_index, last_index; + ssize_t len; + int err, imap_valid = 0, uptodate = 1; + int count = 0; + int nonblocking = 0; + + trace_xfs_writepage(inode, page, 0); + + ASSERT(page_has_buffers(page)); + + /* + * Refuse to write the page out if we are called from reclaim context. + * + * This avoids stack overflows when called from deeply used stacks in + * random callers for direct reclaim or memcg reclaim. We explicitly + * allow reclaim from kswapd as the stack usage there is relatively low. + * + * This should really be done by the core VM, but until that happens + * filesystems like XFS, btrfs and ext4 have to take care of this + * by themselves. + */ + if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC) + goto redirty; + + /* + * Given that we do not allow direct reclaim to call us, we should + * never be called while in a filesystem transaction. + */ + if (WARN_ON(current->flags & PF_FSTRANS)) + goto redirty; + + /* Is this page beyond the end of the file? */ + offset = i_size_read(inode); + end_index = offset >> PAGE_CACHE_SHIFT; + last_index = (offset - 1) >> PAGE_CACHE_SHIFT; + if (page->index >= end_index) { + if ((page->index >= end_index + 1) || + !(i_size_read(inode) & (PAGE_CACHE_SIZE - 1))) { + unlock_page(page); + return 0; + } + } + + end_offset = min_t(unsigned long long, + (xfs_off_t)(page->index + 1) << PAGE_CACHE_SHIFT, + offset); + len = 1 << inode->i_blkbits; + + bh = head = page_buffers(page); + offset = page_offset(page); + type = IO_OVERWRITE; + + if (wbc->sync_mode == WB_SYNC_NONE) + nonblocking = 1; + + do { + int new_ioend = 0; + + if (offset >= end_offset) + break; + if (!buffer_uptodate(bh)) + uptodate = 0; + + /* + * set_page_dirty dirties all buffers in a page, independent + * of their state. The dirty state however is entirely + * meaningless for holes (!mapped && uptodate), so skip + * buffers covering holes here. + */ + if (!buffer_mapped(bh) && buffer_uptodate(bh)) { + imap_valid = 0; + continue; + } + + if (buffer_unwritten(bh)) { + if (type != IO_UNWRITTEN) { + type = IO_UNWRITTEN; + imap_valid = 0; + } + } else if (buffer_delay(bh)) { + if (type != IO_DELALLOC) { + type = IO_DELALLOC; + imap_valid = 0; + } + } else if (buffer_uptodate(bh)) { + if (type != IO_OVERWRITE) { + type = IO_OVERWRITE; + imap_valid = 0; + } + } else { + if (PageUptodate(page)) { + ASSERT(buffer_mapped(bh)); + imap_valid = 0; + } + continue; + } + + if (imap_valid) + imap_valid = xfs_imap_valid(inode, &imap, offset); + if (!imap_valid) { + /* + * If we didn't have a valid mapping then we need to + * put the new mapping into a separate ioend structure. + * This ensures non-contiguous extents always have + * separate ioends, which is particularly important + * for unwritten extent conversion at I/O completion + * time. + */ + new_ioend = 1; + err = xfs_map_blocks(inode, offset, &imap, type, + nonblocking); + if (err) + goto error; + imap_valid = xfs_imap_valid(inode, &imap, offset); + } + if (imap_valid) { + lock_buffer(bh); + if (type != IO_OVERWRITE) + xfs_map_at_offset(inode, bh, &imap, offset); + xfs_add_to_ioend(inode, bh, offset, type, &ioend, + new_ioend); + count++; + } + + if (!iohead) + iohead = ioend; + + } while (offset += len, ((bh = bh->b_this_page) != head)); + + if (uptodate && bh == head) + SetPageUptodate(page); + + xfs_start_page_writeback(page, 1, count); + + if (ioend && imap_valid) { + xfs_off_t end_index; + + end_index = imap.br_startoff + imap.br_blockcount; + + /* to bytes */ + end_index <<= inode->i_blkbits; + + /* to pages */ + end_index = (end_index - 1) >> PAGE_CACHE_SHIFT; + + /* check against file size */ + if (end_index > last_index) + end_index = last_index; + + xfs_cluster_write(inode, page->index + 1, &imap, &ioend, + wbc, end_index); + } + + if (iohead) + xfs_submit_ioend(wbc, iohead); + + return 0; + +error: + if (iohead) + xfs_cancel_ioend(iohead); + + if (err == -EAGAIN) + goto redirty; + + xfs_aops_discard_page(page); + ClearPageUptodate(page); + unlock_page(page); + return err; + +redirty: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; +} + +STATIC int +xfs_vm_writepages( + struct address_space *mapping, + struct writeback_control *wbc) +{ + xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); + return generic_writepages(mapping, wbc); +} + +/* + * Called to move a page into cleanable state - and from there + * to be released. The page should already be clean. We always + * have buffer heads in this call. + * + * Returns 1 if the page is ok to release, 0 otherwise. + */ +STATIC int +xfs_vm_releasepage( + struct page *page, + gfp_t gfp_mask) +{ + int delalloc, unwritten; + + trace_xfs_releasepage(page->mapping->host, page, 0); + + xfs_count_page_state(page, &delalloc, &unwritten); + + if (WARN_ON(delalloc)) + return 0; + if (WARN_ON(unwritten)) + return 0; + + return try_to_free_buffers(page); +} + +STATIC int +__xfs_get_blocks( + struct inode *inode, + sector_t iblock, + struct buffer_head *bh_result, + int create, + int direct) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + xfs_fileoff_t offset_fsb, end_fsb; + int error = 0; + int lockmode = 0; + struct xfs_bmbt_irec imap; + int nimaps = 1; + xfs_off_t offset; + ssize_t size; + int new = 0; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -XFS_ERROR(EIO); + + offset = (xfs_off_t)iblock << inode->i_blkbits; + ASSERT(bh_result->b_size >= (1 << inode->i_blkbits)); + size = bh_result->b_size; + + if (!create && direct && offset >= i_size_read(inode)) + return 0; + + if (create) { + lockmode = XFS_ILOCK_EXCL; + xfs_ilock(ip, lockmode); + } else { + lockmode = xfs_ilock_map_shared(ip); + } + + ASSERT(offset <= mp->m_maxioffset); + if (offset + size > mp->m_maxioffset) + size = mp->m_maxioffset - offset; + end_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + size); + offset_fsb = XFS_B_TO_FSBT(mp, offset); + + error = xfs_bmapi(NULL, ip, offset_fsb, end_fsb - offset_fsb, + XFS_BMAPI_ENTIRE, NULL, 0, &imap, &nimaps, NULL); + if (error) + goto out_unlock; + + if (create && + (!nimaps || + (imap.br_startblock == HOLESTARTBLOCK || + imap.br_startblock == DELAYSTARTBLOCK))) { + if (direct) { + error = xfs_iomap_write_direct(ip, offset, size, + &imap, nimaps); + } else { + error = xfs_iomap_write_delay(ip, offset, size, &imap); + } + if (error) + goto out_unlock; + + trace_xfs_get_blocks_alloc(ip, offset, size, 0, &imap); + } else if (nimaps) { + trace_xfs_get_blocks_found(ip, offset, size, 0, &imap); + } else { + trace_xfs_get_blocks_notfound(ip, offset, size); + goto out_unlock; + } + xfs_iunlock(ip, lockmode); + + if (imap.br_startblock != HOLESTARTBLOCK && + imap.br_startblock != DELAYSTARTBLOCK) { + /* + * For unwritten extents do not report a disk address on + * the read case (treat as if we're reading into a hole). + */ + if (create || !ISUNWRITTEN(&imap)) + xfs_map_buffer(inode, bh_result, &imap, offset); + if (create && ISUNWRITTEN(&imap)) { + if (direct) + bh_result->b_private = inode; + set_buffer_unwritten(bh_result); + } + } + + /* + * If this is a realtime file, data may be on a different device. + * to that pointed to from the buffer_head b_bdev currently. + */ + bh_result->b_bdev = xfs_find_bdev_for_inode(inode); + + /* + * If we previously allocated a block out beyond eof and we are now + * coming back to use it then we will need to flag it as new even if it + * has a disk address. + * + * With sub-block writes into unwritten extents we also need to mark + * the buffer as new so that the unwritten parts of the buffer gets + * correctly zeroed. + */ + if (create && + ((!buffer_mapped(bh_result) && !buffer_uptodate(bh_result)) || + (offset >= i_size_read(inode)) || + (new || ISUNWRITTEN(&imap)))) + set_buffer_new(bh_result); + + if (imap.br_startblock == DELAYSTARTBLOCK) { + BUG_ON(direct); + if (create) { + set_buffer_uptodate(bh_result); + set_buffer_mapped(bh_result); + set_buffer_delay(bh_result); + } + } + + /* + * If this is O_DIRECT or the mpage code calling tell them how large + * the mapping is, so that we can avoid repeated get_blocks calls. + */ + if (direct || size > (1 << inode->i_blkbits)) { + xfs_off_t mapping_size; + + mapping_size = imap.br_startoff + imap.br_blockcount - iblock; + mapping_size <<= inode->i_blkbits; + + ASSERT(mapping_size > 0); + if (mapping_size > size) + mapping_size = size; + if (mapping_size > LONG_MAX) + mapping_size = LONG_MAX; + + bh_result->b_size = mapping_size; + } + + return 0; + +out_unlock: + xfs_iunlock(ip, lockmode); + return -error; +} + +int +xfs_get_blocks( + struct inode *inode, + sector_t iblock, + struct buffer_head *bh_result, + int create) +{ + return __xfs_get_blocks(inode, iblock, bh_result, create, 0); +} + +STATIC int +xfs_get_blocks_direct( + struct inode *inode, + sector_t iblock, + struct buffer_head *bh_result, + int create) +{ + return __xfs_get_blocks(inode, iblock, bh_result, create, 1); +} + +/* + * Complete a direct I/O write request. + * + * If the private argument is non-NULL __xfs_get_blocks signals us that we + * need to issue a transaction to convert the range from unwritten to written + * extents. In case this is regular synchronous I/O we just call xfs_end_io + * to do this and we are done. But in case this was a successful AIO + * request this handler is called from interrupt context, from which we + * can't start transactions. In that case offload the I/O completion to + * the workqueues we also use for buffered I/O completion. + */ +STATIC void +xfs_end_io_direct_write( + struct kiocb *iocb, + loff_t offset, + ssize_t size, + void *private, + int ret, + bool is_async) +{ + struct xfs_ioend *ioend = iocb->private; + + /* + * blockdev_direct_IO can return an error even after the I/O + * completion handler was called. Thus we need to protect + * against double-freeing. + */ + iocb->private = NULL; + + ioend->io_offset = offset; + ioend->io_size = size; + if (private && size > 0) + ioend->io_type = IO_UNWRITTEN; + + if (is_async) { + /* + * If we are converting an unwritten extent we need to delay + * the AIO completion until after the unwrittent extent + * conversion has completed, otherwise do it ASAP. + */ + if (ioend->io_type == IO_UNWRITTEN) { + ioend->io_iocb = iocb; + ioend->io_result = ret; + } else { + aio_complete(iocb, ret, 0); + } + xfs_finish_ioend(ioend); + } else { + xfs_finish_ioend_sync(ioend); + } + + /* XXX: probably should move into the real I/O completion handler */ + inode_dio_done(ioend->io_inode); +} + +STATIC ssize_t +xfs_vm_direct_IO( + int rw, + struct kiocb *iocb, + const struct iovec *iov, + loff_t offset, + unsigned long nr_segs) +{ + struct inode *inode = iocb->ki_filp->f_mapping->host; + struct block_device *bdev = xfs_find_bdev_for_inode(inode); + ssize_t ret; + + if (rw & WRITE) { + iocb->private = xfs_alloc_ioend(inode, IO_DIRECT); + + ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov, + offset, nr_segs, + xfs_get_blocks_direct, + xfs_end_io_direct_write, NULL, 0); + if (ret != -EIOCBQUEUED && iocb->private) + xfs_destroy_ioend(iocb->private); + } else { + ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iov, + offset, nr_segs, + xfs_get_blocks_direct, + NULL, NULL, 0); + } + + return ret; +} + +STATIC void +xfs_vm_write_failed( + struct address_space *mapping, + loff_t to) +{ + struct inode *inode = mapping->host; + + if (to > inode->i_size) { + /* + * punch out the delalloc blocks we have already allocated. We + * don't call xfs_setattr() to do this as we may be in the + * middle of a multi-iovec write and so the vfs inode->i_size + * will not match the xfs ip->i_size and so it will zero too + * much. Hence we jus truncate the page cache to zero what is + * necessary and punch the delalloc blocks directly. + */ + struct xfs_inode *ip = XFS_I(inode); + xfs_fileoff_t start_fsb; + xfs_fileoff_t end_fsb; + int error; + + truncate_pagecache(inode, to, inode->i_size); + + /* + * Check if there are any blocks that are outside of i_size + * that need to be trimmed back. + */ + start_fsb = XFS_B_TO_FSB(ip->i_mount, inode->i_size) + 1; + end_fsb = XFS_B_TO_FSB(ip->i_mount, to); + if (end_fsb <= start_fsb) + return; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_bmap_punch_delalloc_range(ip, start_fsb, + end_fsb - start_fsb); + if (error) { + /* something screwed, just bail */ + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_alert(ip->i_mount, + "xfs_vm_write_failed: unable to clean up ino %lld", + ip->i_ino); + } + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } +} + +STATIC int +xfs_vm_write_begin( + struct file *file, + struct address_space *mapping, + loff_t pos, + unsigned len, + unsigned flags, + struct page **pagep, + void **fsdata) +{ + int ret; + + ret = block_write_begin(mapping, pos, len, flags | AOP_FLAG_NOFS, + pagep, xfs_get_blocks); + if (unlikely(ret)) + xfs_vm_write_failed(mapping, pos + len); + return ret; +} + +STATIC int +xfs_vm_write_end( + struct file *file, + struct address_space *mapping, + loff_t pos, + unsigned len, + unsigned copied, + struct page *page, + void *fsdata) +{ + int ret; + + ret = generic_write_end(file, mapping, pos, len, copied, page, fsdata); + if (unlikely(ret < len)) + xfs_vm_write_failed(mapping, pos + len); + return ret; +} + +STATIC sector_t +xfs_vm_bmap( + struct address_space *mapping, + sector_t block) +{ + struct inode *inode = (struct inode *)mapping->host; + struct xfs_inode *ip = XFS_I(inode); + + trace_xfs_vm_bmap(XFS_I(inode)); + xfs_ilock(ip, XFS_IOLOCK_SHARED); + xfs_flush_pages(ip, (xfs_off_t)0, -1, 0, FI_REMAPF); + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + return generic_block_bmap(mapping, block, xfs_get_blocks); +} + +STATIC int +xfs_vm_readpage( + struct file *unused, + struct page *page) +{ + return mpage_readpage(page, xfs_get_blocks); +} + +STATIC int +xfs_vm_readpages( + struct file *unused, + struct address_space *mapping, + struct list_head *pages, + unsigned nr_pages) +{ + return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks); +} + +const struct address_space_operations xfs_address_space_operations = { + .readpage = xfs_vm_readpage, + .readpages = xfs_vm_readpages, + .writepage = xfs_vm_writepage, + .writepages = xfs_vm_writepages, + .releasepage = xfs_vm_releasepage, + .invalidatepage = xfs_vm_invalidatepage, + .write_begin = xfs_vm_write_begin, + .write_end = xfs_vm_write_end, + .bmap = xfs_vm_bmap, + .direct_IO = xfs_vm_direct_IO, + .migratepage = buffer_migrate_page, + .is_partially_uptodate = block_is_partially_uptodate, + .error_remove_page = generic_error_remove_page, +}; diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h new file mode 100644 index 000000000000..71f721e1a71f --- /dev/null +++ b/fs/xfs/xfs_aops.h @@ -0,0 +1,68 @@ +/* + * Copyright (c) 2005-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_AOPS_H__ +#define __XFS_AOPS_H__ + +extern struct workqueue_struct *xfsdatad_workqueue; +extern struct workqueue_struct *xfsconvertd_workqueue; +extern mempool_t *xfs_ioend_pool; + +/* + * Types of I/O for bmap clustering and I/O completion tracking. + */ +enum { + IO_DIRECT = 0, /* special case for direct I/O ioends */ + IO_DELALLOC, /* mapping covers delalloc region */ + IO_UNWRITTEN, /* mapping covers allocated but uninitialized data */ + IO_OVERWRITE, /* mapping covers already allocated extent */ +}; + +#define XFS_IO_TYPES \ + { 0, "" }, \ + { IO_DELALLOC, "delalloc" }, \ + { IO_UNWRITTEN, "unwritten" }, \ + { IO_OVERWRITE, "overwrite" } + +/* + * xfs_ioend struct manages large extent writes for XFS. + * It can manage several multi-page bio's at once. + */ +typedef struct xfs_ioend { + struct xfs_ioend *io_list; /* next ioend in chain */ + unsigned int io_type; /* delalloc / unwritten */ + int io_error; /* I/O error code */ + atomic_t io_remaining; /* hold count */ + struct inode *io_inode; /* file being written to */ + struct buffer_head *io_buffer_head;/* buffer linked list head */ + struct buffer_head *io_buffer_tail;/* buffer linked list tail */ + size_t io_size; /* size of the extent */ + xfs_off_t io_offset; /* offset in the file */ + struct work_struct io_work; /* xfsdatad work queue */ + struct kiocb *io_iocb; + int io_result; +} xfs_ioend_t; + +extern const struct address_space_operations xfs_address_space_operations; +extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int); + +extern void xfs_ioend_init(void); +extern void xfs_ioend_wait(struct xfs_inode *); + +extern void xfs_count_page_state(struct page *, int *, int *); + +#endif /* __XFS_AOPS_H__ */ diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c new file mode 100644 index 000000000000..c57836dc778f --- /dev/null +++ b/fs/xfs/xfs_buf.c @@ -0,0 +1,1876 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "xfs_sb.h" +#include "xfs_inum.h" +#include "xfs_log.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_trace.h" + +static kmem_zone_t *xfs_buf_zone; +STATIC int xfsbufd(void *); +STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int); + +static struct workqueue_struct *xfslogd_workqueue; +struct workqueue_struct *xfsdatad_workqueue; +struct workqueue_struct *xfsconvertd_workqueue; + +#ifdef XFS_BUF_LOCK_TRACKING +# define XB_SET_OWNER(bp) ((bp)->b_last_holder = current->pid) +# define XB_CLEAR_OWNER(bp) ((bp)->b_last_holder = -1) +# define XB_GET_OWNER(bp) ((bp)->b_last_holder) +#else +# define XB_SET_OWNER(bp) do { } while (0) +# define XB_CLEAR_OWNER(bp) do { } while (0) +# define XB_GET_OWNER(bp) do { } while (0) +#endif + +#define xb_to_gfp(flags) \ + ((((flags) & XBF_READ_AHEAD) ? __GFP_NORETRY : \ + ((flags) & XBF_DONT_BLOCK) ? GFP_NOFS : GFP_KERNEL) | __GFP_NOWARN) + +#define xb_to_km(flags) \ + (((flags) & XBF_DONT_BLOCK) ? KM_NOFS : KM_SLEEP) + +#define xfs_buf_allocate(flags) \ + kmem_zone_alloc(xfs_buf_zone, xb_to_km(flags)) +#define xfs_buf_deallocate(bp) \ + kmem_zone_free(xfs_buf_zone, (bp)); + +static inline int +xfs_buf_is_vmapped( + struct xfs_buf *bp) +{ + /* + * Return true if the buffer is vmapped. + * + * The XBF_MAPPED flag is set if the buffer should be mapped, but the + * code is clever enough to know it doesn't have to map a single page, + * so the check has to be both for XBF_MAPPED and bp->b_page_count > 1. + */ + return (bp->b_flags & XBF_MAPPED) && bp->b_page_count > 1; +} + +static inline int +xfs_buf_vmap_len( + struct xfs_buf *bp) +{ + return (bp->b_page_count * PAGE_SIZE) - bp->b_offset; +} + +/* + * xfs_buf_lru_add - add a buffer to the LRU. + * + * The LRU takes a new reference to the buffer so that it will only be freed + * once the shrinker takes the buffer off the LRU. + */ +STATIC void +xfs_buf_lru_add( + struct xfs_buf *bp) +{ + struct xfs_buftarg *btp = bp->b_target; + + spin_lock(&btp->bt_lru_lock); + if (list_empty(&bp->b_lru)) { + atomic_inc(&bp->b_hold); + list_add_tail(&bp->b_lru, &btp->bt_lru); + btp->bt_lru_nr++; + } + spin_unlock(&btp->bt_lru_lock); +} + +/* + * xfs_buf_lru_del - remove a buffer from the LRU + * + * The unlocked check is safe here because it only occurs when there are not + * b_lru_ref counts left on the inode under the pag->pag_buf_lock. it is there + * to optimise the shrinker removing the buffer from the LRU and calling + * xfs_buf_free(). i.e. it removes an unnecessary round trip on the + * bt_lru_lock. + */ +STATIC void +xfs_buf_lru_del( + struct xfs_buf *bp) +{ + struct xfs_buftarg *btp = bp->b_target; + + if (list_empty(&bp->b_lru)) + return; + + spin_lock(&btp->bt_lru_lock); + if (!list_empty(&bp->b_lru)) { + list_del_init(&bp->b_lru); + btp->bt_lru_nr--; + } + spin_unlock(&btp->bt_lru_lock); +} + +/* + * When we mark a buffer stale, we remove the buffer from the LRU and clear the + * b_lru_ref count so that the buffer is freed immediately when the buffer + * reference count falls to zero. If the buffer is already on the LRU, we need + * to remove the reference that LRU holds on the buffer. + * + * This prevents build-up of stale buffers on the LRU. + */ +void +xfs_buf_stale( + struct xfs_buf *bp) +{ + bp->b_flags |= XBF_STALE; + atomic_set(&(bp)->b_lru_ref, 0); + if (!list_empty(&bp->b_lru)) { + struct xfs_buftarg *btp = bp->b_target; + + spin_lock(&btp->bt_lru_lock); + if (!list_empty(&bp->b_lru)) { + list_del_init(&bp->b_lru); + btp->bt_lru_nr--; + atomic_dec(&bp->b_hold); + } + spin_unlock(&btp->bt_lru_lock); + } + ASSERT(atomic_read(&bp->b_hold) >= 1); +} + +STATIC void +_xfs_buf_initialize( + xfs_buf_t *bp, + xfs_buftarg_t *target, + xfs_off_t range_base, + size_t range_length, + xfs_buf_flags_t flags) +{ + /* + * We don't want certain flags to appear in b_flags. + */ + flags &= ~(XBF_LOCK|XBF_MAPPED|XBF_DONT_BLOCK|XBF_READ_AHEAD); + + memset(bp, 0, sizeof(xfs_buf_t)); + atomic_set(&bp->b_hold, 1); + atomic_set(&bp->b_lru_ref, 1); + init_completion(&bp->b_iowait); + INIT_LIST_HEAD(&bp->b_lru); + INIT_LIST_HEAD(&bp->b_list); + RB_CLEAR_NODE(&bp->b_rbnode); + sema_init(&bp->b_sema, 0); /* held, no waiters */ + XB_SET_OWNER(bp); + bp->b_target = target; + bp->b_file_offset = range_base; + /* + * Set buffer_length and count_desired to the same value initially. + * I/O routines should use count_desired, which will be the same in + * most cases but may be reset (e.g. XFS recovery). + */ + bp->b_buffer_length = bp->b_count_desired = range_length; + bp->b_flags = flags; + bp->b_bn = XFS_BUF_DADDR_NULL; + atomic_set(&bp->b_pin_count, 0); + init_waitqueue_head(&bp->b_waiters); + + XFS_STATS_INC(xb_create); + + trace_xfs_buf_init(bp, _RET_IP_); +} + +/* + * Allocate a page array capable of holding a specified number + * of pages, and point the page buf at it. + */ +STATIC int +_xfs_buf_get_pages( + xfs_buf_t *bp, + int page_count, + xfs_buf_flags_t flags) +{ + /* Make sure that we have a page list */ + if (bp->b_pages == NULL) { + bp->b_offset = xfs_buf_poff(bp->b_file_offset); + bp->b_page_count = page_count; + if (page_count <= XB_PAGES) { + bp->b_pages = bp->b_page_array; + } else { + bp->b_pages = kmem_alloc(sizeof(struct page *) * + page_count, xb_to_km(flags)); + if (bp->b_pages == NULL) + return -ENOMEM; + } + memset(bp->b_pages, 0, sizeof(struct page *) * page_count); + } + return 0; +} + +/* + * Frees b_pages if it was allocated. + */ +STATIC void +_xfs_buf_free_pages( + xfs_buf_t *bp) +{ + if (bp->b_pages != bp->b_page_array) { + kmem_free(bp->b_pages); + bp->b_pages = NULL; + } +} + +/* + * Releases the specified buffer. + * + * The modification state of any associated pages is left unchanged. + * The buffer most not be on any hash - use xfs_buf_rele instead for + * hashed and refcounted buffers + */ +void +xfs_buf_free( + xfs_buf_t *bp) +{ + trace_xfs_buf_free(bp, _RET_IP_); + + ASSERT(list_empty(&bp->b_lru)); + + if (bp->b_flags & _XBF_PAGES) { + uint i; + + if (xfs_buf_is_vmapped(bp)) + vm_unmap_ram(bp->b_addr - bp->b_offset, + bp->b_page_count); + + for (i = 0; i < bp->b_page_count; i++) { + struct page *page = bp->b_pages[i]; + + __free_page(page); + } + } else if (bp->b_flags & _XBF_KMEM) + kmem_free(bp->b_addr); + _xfs_buf_free_pages(bp); + xfs_buf_deallocate(bp); +} + +/* + * Allocates all the pages for buffer in question and builds it's page list. + */ +STATIC int +xfs_buf_allocate_memory( + xfs_buf_t *bp, + uint flags) +{ + size_t size = bp->b_count_desired; + size_t nbytes, offset; + gfp_t gfp_mask = xb_to_gfp(flags); + unsigned short page_count, i; + xfs_off_t end; + int error; + + /* + * for buffers that are contained within a single page, just allocate + * the memory from the heap - there's no need for the complexity of + * page arrays to keep allocation down to order 0. + */ + if (bp->b_buffer_length < PAGE_SIZE) { + bp->b_addr = kmem_alloc(bp->b_buffer_length, xb_to_km(flags)); + if (!bp->b_addr) { + /* low memory - use alloc_page loop instead */ + goto use_alloc_page; + } + + if (((unsigned long)(bp->b_addr + bp->b_buffer_length - 1) & + PAGE_MASK) != + ((unsigned long)bp->b_addr & PAGE_MASK)) { + /* b_addr spans two pages - use alloc_page instead */ + kmem_free(bp->b_addr); + bp->b_addr = NULL; + goto use_alloc_page; + } + bp->b_offset = offset_in_page(bp->b_addr); + bp->b_pages = bp->b_page_array; + bp->b_pages[0] = virt_to_page(bp->b_addr); + bp->b_page_count = 1; + bp->b_flags |= XBF_MAPPED | _XBF_KMEM; + return 0; + } + +use_alloc_page: + end = bp->b_file_offset + bp->b_buffer_length; + page_count = xfs_buf_btoc(end) - xfs_buf_btoct(bp->b_file_offset); + error = _xfs_buf_get_pages(bp, page_count, flags); + if (unlikely(error)) + return error; + + offset = bp->b_offset; + bp->b_flags |= _XBF_PAGES; + + for (i = 0; i < bp->b_page_count; i++) { + struct page *page; + uint retries = 0; +retry: + page = alloc_page(gfp_mask); + if (unlikely(page == NULL)) { + if (flags & XBF_READ_AHEAD) { + bp->b_page_count = i; + error = ENOMEM; + goto out_free_pages; + } + + /* + * This could deadlock. + * + * But until all the XFS lowlevel code is revamped to + * handle buffer allocation failures we can't do much. + */ + if (!(++retries % 100)) + xfs_err(NULL, + "possible memory allocation deadlock in %s (mode:0x%x)", + __func__, gfp_mask); + + XFS_STATS_INC(xb_page_retries); + congestion_wait(BLK_RW_ASYNC, HZ/50); + goto retry; + } + + XFS_STATS_INC(xb_page_found); + + nbytes = min_t(size_t, size, PAGE_SIZE - offset); + size -= nbytes; + bp->b_pages[i] = page; + offset = 0; + } + return 0; + +out_free_pages: + for (i = 0; i < bp->b_page_count; i++) + __free_page(bp->b_pages[i]); + return error; +} + +/* + * Map buffer into kernel address-space if necessary. + */ +STATIC int +_xfs_buf_map_pages( + xfs_buf_t *bp, + uint flags) +{ + ASSERT(bp->b_flags & _XBF_PAGES); + if (bp->b_page_count == 1) { + /* A single page buffer is always mappable */ + bp->b_addr = page_address(bp->b_pages[0]) + bp->b_offset; + bp->b_flags |= XBF_MAPPED; + } else if (flags & XBF_MAPPED) { + int retried = 0; + + do { + bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count, + -1, PAGE_KERNEL); + if (bp->b_addr) + break; + vm_unmap_aliases(); + } while (retried++ <= 1); + + if (!bp->b_addr) + return -ENOMEM; + bp->b_addr += bp->b_offset; + bp->b_flags |= XBF_MAPPED; + } + + return 0; +} + +/* + * Finding and Reading Buffers + */ + +/* + * Look up, and creates if absent, a lockable buffer for + * a given range of an inode. The buffer is returned + * locked. If other overlapping buffers exist, they are + * released before the new buffer is created and locked, + * which may imply that this call will block until those buffers + * are unlocked. No I/O is implied by this call. + */ +xfs_buf_t * +_xfs_buf_find( + xfs_buftarg_t *btp, /* block device target */ + xfs_off_t ioff, /* starting offset of range */ + size_t isize, /* length of range */ + xfs_buf_flags_t flags, + xfs_buf_t *new_bp) +{ + xfs_off_t range_base; + size_t range_length; + struct xfs_perag *pag; + struct rb_node **rbp; + struct rb_node *parent; + xfs_buf_t *bp; + + range_base = (ioff << BBSHIFT); + range_length = (isize << BBSHIFT); + + /* Check for IOs smaller than the sector size / not sector aligned */ + ASSERT(!(range_length < (1 << btp->bt_sshift))); + ASSERT(!(range_base & (xfs_off_t)btp->bt_smask)); + + /* get tree root */ + pag = xfs_perag_get(btp->bt_mount, + xfs_daddr_to_agno(btp->bt_mount, ioff)); + + /* walk tree */ + spin_lock(&pag->pag_buf_lock); + rbp = &pag->pag_buf_tree.rb_node; + parent = NULL; + bp = NULL; + while (*rbp) { + parent = *rbp; + bp = rb_entry(parent, struct xfs_buf, b_rbnode); + + if (range_base < bp->b_file_offset) + rbp = &(*rbp)->rb_left; + else if (range_base > bp->b_file_offset) + rbp = &(*rbp)->rb_right; + else { + /* + * found a block offset match. If the range doesn't + * match, the only way this is allowed is if the buffer + * in the cache is stale and the transaction that made + * it stale has not yet committed. i.e. we are + * reallocating a busy extent. Skip this buffer and + * continue searching to the right for an exact match. + */ + if (bp->b_buffer_length != range_length) { + ASSERT(bp->b_flags & XBF_STALE); + rbp = &(*rbp)->rb_right; + continue; + } + atomic_inc(&bp->b_hold); + goto found; + } + } + + /* No match found */ + if (new_bp) { + _xfs_buf_initialize(new_bp, btp, range_base, + range_length, flags); + rb_link_node(&new_bp->b_rbnode, parent, rbp); + rb_insert_color(&new_bp->b_rbnode, &pag->pag_buf_tree); + /* the buffer keeps the perag reference until it is freed */ + new_bp->b_pag = pag; + spin_unlock(&pag->pag_buf_lock); + } else { + XFS_STATS_INC(xb_miss_locked); + spin_unlock(&pag->pag_buf_lock); + xfs_perag_put(pag); + } + return new_bp; + +found: + spin_unlock(&pag->pag_buf_lock); + xfs_perag_put(pag); + + if (!xfs_buf_trylock(bp)) { + if (flags & XBF_TRYLOCK) { + xfs_buf_rele(bp); + XFS_STATS_INC(xb_busy_locked); + return NULL; + } + xfs_buf_lock(bp); + XFS_STATS_INC(xb_get_locked_waited); + } + + /* + * if the buffer is stale, clear all the external state associated with + * it. We need to keep flags such as how we allocated the buffer memory + * intact here. + */ + if (bp->b_flags & XBF_STALE) { + ASSERT((bp->b_flags & _XBF_DELWRI_Q) == 0); + bp->b_flags &= XBF_MAPPED | _XBF_KMEM | _XBF_PAGES; + } + + trace_xfs_buf_find(bp, flags, _RET_IP_); + XFS_STATS_INC(xb_get_locked); + return bp; +} + +/* + * Assembles a buffer covering the specified range. + * Storage in memory for all portions of the buffer will be allocated, + * although backing storage may not be. + */ +xfs_buf_t * +xfs_buf_get( + xfs_buftarg_t *target,/* target for buffer */ + xfs_off_t ioff, /* starting offset of range */ + size_t isize, /* length of range */ + xfs_buf_flags_t flags) +{ + xfs_buf_t *bp, *new_bp; + int error = 0; + + new_bp = xfs_buf_allocate(flags); + if (unlikely(!new_bp)) + return NULL; + + bp = _xfs_buf_find(target, ioff, isize, flags, new_bp); + if (bp == new_bp) { + error = xfs_buf_allocate_memory(bp, flags); + if (error) + goto no_buffer; + } else { + xfs_buf_deallocate(new_bp); + if (unlikely(bp == NULL)) + return NULL; + } + + if (!(bp->b_flags & XBF_MAPPED)) { + error = _xfs_buf_map_pages(bp, flags); + if (unlikely(error)) { + xfs_warn(target->bt_mount, + "%s: failed to map pages\n", __func__); + goto no_buffer; + } + } + + XFS_STATS_INC(xb_get); + + /* + * Always fill in the block number now, the mapped cases can do + * their own overlay of this later. + */ + bp->b_bn = ioff; + bp->b_count_desired = bp->b_buffer_length; + + trace_xfs_buf_get(bp, flags, _RET_IP_); + return bp; + + no_buffer: + if (flags & (XBF_LOCK | XBF_TRYLOCK)) + xfs_buf_unlock(bp); + xfs_buf_rele(bp); + return NULL; +} + +STATIC int +_xfs_buf_read( + xfs_buf_t *bp, + xfs_buf_flags_t flags) +{ + int status; + + ASSERT(!(flags & (XBF_DELWRI|XBF_WRITE))); + ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL); + + bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_DELWRI | XBF_READ_AHEAD); + bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | XBF_READ_AHEAD); + + status = xfs_buf_iorequest(bp); + if (status || bp->b_error || (flags & XBF_ASYNC)) + return status; + return xfs_buf_iowait(bp); +} + +xfs_buf_t * +xfs_buf_read( + xfs_buftarg_t *target, + xfs_off_t ioff, + size_t isize, + xfs_buf_flags_t flags) +{ + xfs_buf_t *bp; + + flags |= XBF_READ; + + bp = xfs_buf_get(target, ioff, isize, flags); + if (bp) { + trace_xfs_buf_read(bp, flags, _RET_IP_); + + if (!XFS_BUF_ISDONE(bp)) { + XFS_STATS_INC(xb_get_read); + _xfs_buf_read(bp, flags); + } else if (flags & XBF_ASYNC) { + /* + * Read ahead call which is already satisfied, + * drop the buffer + */ + goto no_buffer; + } else { + /* We do not want read in the flags */ + bp->b_flags &= ~XBF_READ; + } + } + + return bp; + + no_buffer: + if (flags & (XBF_LOCK | XBF_TRYLOCK)) + xfs_buf_unlock(bp); + xfs_buf_rele(bp); + return NULL; +} + +/* + * If we are not low on memory then do the readahead in a deadlock + * safe manner. + */ +void +xfs_buf_readahead( + xfs_buftarg_t *target, + xfs_off_t ioff, + size_t isize) +{ + if (bdi_read_congested(target->bt_bdi)) + return; + + xfs_buf_read(target, ioff, isize, + XBF_TRYLOCK|XBF_ASYNC|XBF_READ_AHEAD|XBF_DONT_BLOCK); +} + +/* + * Read an uncached buffer from disk. Allocates and returns a locked + * buffer containing the disk contents or nothing. + */ +struct xfs_buf * +xfs_buf_read_uncached( + struct xfs_mount *mp, + struct xfs_buftarg *target, + xfs_daddr_t daddr, + size_t length, + int flags) +{ + xfs_buf_t *bp; + int error; + + bp = xfs_buf_get_uncached(target, length, flags); + if (!bp) + return NULL; + + /* set up the buffer for a read IO */ + XFS_BUF_SET_ADDR(bp, daddr); + XFS_BUF_READ(bp); + + xfsbdstrat(mp, bp); + error = xfs_buf_iowait(bp); + if (error || bp->b_error) { + xfs_buf_relse(bp); + return NULL; + } + return bp; +} + +xfs_buf_t * +xfs_buf_get_empty( + size_t len, + xfs_buftarg_t *target) +{ + xfs_buf_t *bp; + + bp = xfs_buf_allocate(0); + if (bp) + _xfs_buf_initialize(bp, target, 0, len, 0); + return bp; +} + +/* + * Return a buffer allocated as an empty buffer and associated to external + * memory via xfs_buf_associate_memory() back to it's empty state. + */ +void +xfs_buf_set_empty( + struct xfs_buf *bp, + size_t len) +{ + if (bp->b_pages) + _xfs_buf_free_pages(bp); + + bp->b_pages = NULL; + bp->b_page_count = 0; + bp->b_addr = NULL; + bp->b_file_offset = 0; + bp->b_buffer_length = bp->b_count_desired = len; + bp->b_bn = XFS_BUF_DADDR_NULL; + bp->b_flags &= ~XBF_MAPPED; +} + +static inline struct page * +mem_to_page( + void *addr) +{ + if ((!is_vmalloc_addr(addr))) { + return virt_to_page(addr); + } else { + return vmalloc_to_page(addr); + } +} + +int +xfs_buf_associate_memory( + xfs_buf_t *bp, + void *mem, + size_t len) +{ + int rval; + int i = 0; + unsigned long pageaddr; + unsigned long offset; + size_t buflen; + int page_count; + + pageaddr = (unsigned long)mem & PAGE_MASK; + offset = (unsigned long)mem - pageaddr; + buflen = PAGE_ALIGN(len + offset); + page_count = buflen >> PAGE_SHIFT; + + /* Free any previous set of page pointers */ + if (bp->b_pages) + _xfs_buf_free_pages(bp); + + bp->b_pages = NULL; + bp->b_addr = mem; + + rval = _xfs_buf_get_pages(bp, page_count, XBF_DONT_BLOCK); + if (rval) + return rval; + + bp->b_offset = offset; + + for (i = 0; i < bp->b_page_count; i++) { + bp->b_pages[i] = mem_to_page((void *)pageaddr); + pageaddr += PAGE_SIZE; + } + + bp->b_count_desired = len; + bp->b_buffer_length = buflen; + bp->b_flags |= XBF_MAPPED; + + return 0; +} + +xfs_buf_t * +xfs_buf_get_uncached( + struct xfs_buftarg *target, + size_t len, + int flags) +{ + unsigned long page_count = PAGE_ALIGN(len) >> PAGE_SHIFT; + int error, i; + xfs_buf_t *bp; + + bp = xfs_buf_allocate(0); + if (unlikely(bp == NULL)) + goto fail; + _xfs_buf_initialize(bp, target, 0, len, 0); + + error = _xfs_buf_get_pages(bp, page_count, 0); + if (error) + goto fail_free_buf; + + for (i = 0; i < page_count; i++) { + bp->b_pages[i] = alloc_page(xb_to_gfp(flags)); + if (!bp->b_pages[i]) + goto fail_free_mem; + } + bp->b_flags |= _XBF_PAGES; + + error = _xfs_buf_map_pages(bp, XBF_MAPPED); + if (unlikely(error)) { + xfs_warn(target->bt_mount, + "%s: failed to map pages\n", __func__); + goto fail_free_mem; + } + + trace_xfs_buf_get_uncached(bp, _RET_IP_); + return bp; + + fail_free_mem: + while (--i >= 0) + __free_page(bp->b_pages[i]); + _xfs_buf_free_pages(bp); + fail_free_buf: + xfs_buf_deallocate(bp); + fail: + return NULL; +} + +/* + * Increment reference count on buffer, to hold the buffer concurrently + * with another thread which may release (free) the buffer asynchronously. + * Must hold the buffer already to call this function. + */ +void +xfs_buf_hold( + xfs_buf_t *bp) +{ + trace_xfs_buf_hold(bp, _RET_IP_); + atomic_inc(&bp->b_hold); +} + +/* + * Releases a hold on the specified buffer. If the + * the hold count is 1, calls xfs_buf_free. + */ +void +xfs_buf_rele( + xfs_buf_t *bp) +{ + struct xfs_perag *pag = bp->b_pag; + + trace_xfs_buf_rele(bp, _RET_IP_); + + if (!pag) { + ASSERT(list_empty(&bp->b_lru)); + ASSERT(RB_EMPTY_NODE(&bp->b_rbnode)); + if (atomic_dec_and_test(&bp->b_hold)) + xfs_buf_free(bp); + return; + } + + ASSERT(!RB_EMPTY_NODE(&bp->b_rbnode)); + + ASSERT(atomic_read(&bp->b_hold) > 0); + if (atomic_dec_and_lock(&bp->b_hold, &pag->pag_buf_lock)) { + if (!(bp->b_flags & XBF_STALE) && + atomic_read(&bp->b_lru_ref)) { + xfs_buf_lru_add(bp); + spin_unlock(&pag->pag_buf_lock); + } else { + xfs_buf_lru_del(bp); + ASSERT(!(bp->b_flags & (XBF_DELWRI|_XBF_DELWRI_Q))); + rb_erase(&bp->b_rbnode, &pag->pag_buf_tree); + spin_unlock(&pag->pag_buf_lock); + xfs_perag_put(pag); + xfs_buf_free(bp); + } + } +} + + +/* + * Lock a buffer object, if it is not already locked. + * + * If we come across a stale, pinned, locked buffer, we know that we are + * being asked to lock a buffer that has been reallocated. Because it is + * pinned, we know that the log has not been pushed to disk and hence it + * will still be locked. Rather than continuing to have trylock attempts + * fail until someone else pushes the log, push it ourselves before + * returning. This means that the xfsaild will not get stuck trying + * to push on stale inode buffers. + */ +int +xfs_buf_trylock( + struct xfs_buf *bp) +{ + int locked; + + locked = down_trylock(&bp->b_sema) == 0; + if (locked) + XB_SET_OWNER(bp); + else if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) + xfs_log_force(bp->b_target->bt_mount, 0); + + trace_xfs_buf_trylock(bp, _RET_IP_); + return locked; +} + +/* + * Lock a buffer object. + * + * If we come across a stale, pinned, locked buffer, we know that we + * are being asked to lock a buffer that has been reallocated. Because + * it is pinned, we know that the log has not been pushed to disk and + * hence it will still be locked. Rather than sleeping until someone + * else pushes the log, push it ourselves before trying to get the lock. + */ +void +xfs_buf_lock( + struct xfs_buf *bp) +{ + trace_xfs_buf_lock(bp, _RET_IP_); + + if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) + xfs_log_force(bp->b_target->bt_mount, 0); + down(&bp->b_sema); + XB_SET_OWNER(bp); + + trace_xfs_buf_lock_done(bp, _RET_IP_); +} + +/* + * Releases the lock on the buffer object. + * If the buffer is marked delwri but is not queued, do so before we + * unlock the buffer as we need to set flags correctly. We also need to + * take a reference for the delwri queue because the unlocker is going to + * drop their's and they don't know we just queued it. + */ +void +xfs_buf_unlock( + struct xfs_buf *bp) +{ + if ((bp->b_flags & (XBF_DELWRI|_XBF_DELWRI_Q)) == XBF_DELWRI) { + atomic_inc(&bp->b_hold); + bp->b_flags |= XBF_ASYNC; + xfs_buf_delwri_queue(bp, 0); + } + + XB_CLEAR_OWNER(bp); + up(&bp->b_sema); + + trace_xfs_buf_unlock(bp, _RET_IP_); +} + +STATIC void +xfs_buf_wait_unpin( + xfs_buf_t *bp) +{ + DECLARE_WAITQUEUE (wait, current); + + if (atomic_read(&bp->b_pin_count) == 0) + return; + + add_wait_queue(&bp->b_waiters, &wait); + for (;;) { + set_current_state(TASK_UNINTERRUPTIBLE); + if (atomic_read(&bp->b_pin_count) == 0) + break; + io_schedule(); + } + remove_wait_queue(&bp->b_waiters, &wait); + set_current_state(TASK_RUNNING); +} + +/* + * Buffer Utility Routines + */ + +STATIC void +xfs_buf_iodone_work( + struct work_struct *work) +{ + xfs_buf_t *bp = + container_of(work, xfs_buf_t, b_iodone_work); + + if (bp->b_iodone) + (*(bp->b_iodone))(bp); + else if (bp->b_flags & XBF_ASYNC) + xfs_buf_relse(bp); +} + +void +xfs_buf_ioend( + xfs_buf_t *bp, + int schedule) +{ + trace_xfs_buf_iodone(bp, _RET_IP_); + + bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_READ_AHEAD); + if (bp->b_error == 0) + bp->b_flags |= XBF_DONE; + + if ((bp->b_iodone) || (bp->b_flags & XBF_ASYNC)) { + if (schedule) { + INIT_WORK(&bp->b_iodone_work, xfs_buf_iodone_work); + queue_work(xfslogd_workqueue, &bp->b_iodone_work); + } else { + xfs_buf_iodone_work(&bp->b_iodone_work); + } + } else { + complete(&bp->b_iowait); + } +} + +void +xfs_buf_ioerror( + xfs_buf_t *bp, + int error) +{ + ASSERT(error >= 0 && error <= 0xffff); + bp->b_error = (unsigned short)error; + trace_xfs_buf_ioerror(bp, error, _RET_IP_); +} + +int +xfs_bwrite( + struct xfs_mount *mp, + struct xfs_buf *bp) +{ + int error; + + bp->b_flags |= XBF_WRITE; + bp->b_flags &= ~(XBF_ASYNC | XBF_READ); + + xfs_buf_delwri_dequeue(bp); + xfs_bdstrat_cb(bp); + + error = xfs_buf_iowait(bp); + if (error) + xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR); + xfs_buf_relse(bp); + return error; +} + +void +xfs_bdwrite( + void *mp, + struct xfs_buf *bp) +{ + trace_xfs_buf_bdwrite(bp, _RET_IP_); + + bp->b_flags &= ~XBF_READ; + bp->b_flags |= (XBF_DELWRI | XBF_ASYNC); + + xfs_buf_delwri_queue(bp, 1); +} + +/* + * Called when we want to stop a buffer from getting written or read. + * We attach the EIO error, muck with its flags, and call xfs_buf_ioend + * so that the proper iodone callbacks get called. + */ +STATIC int +xfs_bioerror( + xfs_buf_t *bp) +{ +#ifdef XFSERRORDEBUG + ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone); +#endif + + /* + * No need to wait until the buffer is unpinned, we aren't flushing it. + */ + xfs_buf_ioerror(bp, EIO); + + /* + * We're calling xfs_buf_ioend, so delete XBF_DONE flag. + */ + XFS_BUF_UNREAD(bp); + XFS_BUF_UNDELAYWRITE(bp); + XFS_BUF_UNDONE(bp); + XFS_BUF_STALE(bp); + + xfs_buf_ioend(bp, 0); + + return EIO; +} + +/* + * Same as xfs_bioerror, except that we are releasing the buffer + * here ourselves, and avoiding the xfs_buf_ioend call. + * This is meant for userdata errors; metadata bufs come with + * iodone functions attached, so that we can track down errors. + */ +STATIC int +xfs_bioerror_relse( + struct xfs_buf *bp) +{ + int64_t fl = bp->b_flags; + /* + * No need to wait until the buffer is unpinned. + * We aren't flushing it. + * + * chunkhold expects B_DONE to be set, whether + * we actually finish the I/O or not. We don't want to + * change that interface. + */ + XFS_BUF_UNREAD(bp); + XFS_BUF_UNDELAYWRITE(bp); + XFS_BUF_DONE(bp); + XFS_BUF_STALE(bp); + bp->b_iodone = NULL; + if (!(fl & XBF_ASYNC)) { + /* + * Mark b_error and B_ERROR _both_. + * Lot's of chunkcache code assumes that. + * There's no reason to mark error for + * ASYNC buffers. + */ + xfs_buf_ioerror(bp, EIO); + XFS_BUF_FINISH_IOWAIT(bp); + } else { + xfs_buf_relse(bp); + } + + return EIO; +} + + +/* + * All xfs metadata buffers except log state machine buffers + * get this attached as their b_bdstrat callback function. + * This is so that we can catch a buffer + * after prematurely unpinning it to forcibly shutdown the filesystem. + */ +int +xfs_bdstrat_cb( + struct xfs_buf *bp) +{ + if (XFS_FORCED_SHUTDOWN(bp->b_target->bt_mount)) { + trace_xfs_bdstrat_shut(bp, _RET_IP_); + /* + * Metadata write that didn't get logged but + * written delayed anyway. These aren't associated + * with a transaction, and can be ignored. + */ + if (!bp->b_iodone && !XFS_BUF_ISREAD(bp)) + return xfs_bioerror_relse(bp); + else + return xfs_bioerror(bp); + } + + xfs_buf_iorequest(bp); + return 0; +} + +/* + * Wrapper around bdstrat so that we can stop data from going to disk in case + * we are shutting down the filesystem. Typically user data goes thru this + * path; one of the exceptions is the superblock. + */ +void +xfsbdstrat( + struct xfs_mount *mp, + struct xfs_buf *bp) +{ + if (XFS_FORCED_SHUTDOWN(mp)) { + trace_xfs_bdstrat_shut(bp, _RET_IP_); + xfs_bioerror_relse(bp); + return; + } + + xfs_buf_iorequest(bp); +} + +STATIC void +_xfs_buf_ioend( + xfs_buf_t *bp, + int schedule) +{ + if (atomic_dec_and_test(&bp->b_io_remaining) == 1) + xfs_buf_ioend(bp, schedule); +} + +STATIC void +xfs_buf_bio_end_io( + struct bio *bio, + int error) +{ + xfs_buf_t *bp = (xfs_buf_t *)bio->bi_private; + + xfs_buf_ioerror(bp, -error); + + if (!error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ)) + invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp)); + + _xfs_buf_ioend(bp, 1); + bio_put(bio); +} + +STATIC void +_xfs_buf_ioapply( + xfs_buf_t *bp) +{ + int rw, map_i, total_nr_pages, nr_pages; + struct bio *bio; + int offset = bp->b_offset; + int size = bp->b_count_desired; + sector_t sector = bp->b_bn; + + total_nr_pages = bp->b_page_count; + map_i = 0; + + if (bp->b_flags & XBF_WRITE) { + if (bp->b_flags & XBF_SYNCIO) + rw = WRITE_SYNC; + else + rw = WRITE; + if (bp->b_flags & XBF_FUA) + rw |= REQ_FUA; + if (bp->b_flags & XBF_FLUSH) + rw |= REQ_FLUSH; + } else if (bp->b_flags & XBF_READ_AHEAD) { + rw = READA; + } else { + rw = READ; + } + + /* we only use the buffer cache for meta-data */ + rw |= REQ_META; + +next_chunk: + atomic_inc(&bp->b_io_remaining); + nr_pages = BIO_MAX_SECTORS >> (PAGE_SHIFT - BBSHIFT); + if (nr_pages > total_nr_pages) + nr_pages = total_nr_pages; + + bio = bio_alloc(GFP_NOIO, nr_pages); + bio->bi_bdev = bp->b_target->bt_bdev; + bio->bi_sector = sector; + bio->bi_end_io = xfs_buf_bio_end_io; + bio->bi_private = bp; + + + for (; size && nr_pages; nr_pages--, map_i++) { + int rbytes, nbytes = PAGE_SIZE - offset; + + if (nbytes > size) + nbytes = size; + + rbytes = bio_add_page(bio, bp->b_pages[map_i], nbytes, offset); + if (rbytes < nbytes) + break; + + offset = 0; + sector += nbytes >> BBSHIFT; + size -= nbytes; + total_nr_pages--; + } + + if (likely(bio->bi_size)) { + if (xfs_buf_is_vmapped(bp)) { + flush_kernel_vmap_range(bp->b_addr, + xfs_buf_vmap_len(bp)); + } + submit_bio(rw, bio); + if (size) + goto next_chunk; + } else { + xfs_buf_ioerror(bp, EIO); + bio_put(bio); + } +} + +int +xfs_buf_iorequest( + xfs_buf_t *bp) +{ + trace_xfs_buf_iorequest(bp, _RET_IP_); + + if (bp->b_flags & XBF_DELWRI) { + xfs_buf_delwri_queue(bp, 1); + return 0; + } + + if (bp->b_flags & XBF_WRITE) { + xfs_buf_wait_unpin(bp); + } + + xfs_buf_hold(bp); + + /* Set the count to 1 initially, this will stop an I/O + * completion callout which happens before we have started + * all the I/O from calling xfs_buf_ioend too early. + */ + atomic_set(&bp->b_io_remaining, 1); + _xfs_buf_ioapply(bp); + _xfs_buf_ioend(bp, 0); + + xfs_buf_rele(bp); + return 0; +} + +/* + * Waits for I/O to complete on the buffer supplied. + * It returns immediately if no I/O is pending. + * It returns the I/O error code, if any, or 0 if there was no error. + */ +int +xfs_buf_iowait( + xfs_buf_t *bp) +{ + trace_xfs_buf_iowait(bp, _RET_IP_); + + wait_for_completion(&bp->b_iowait); + + trace_xfs_buf_iowait_done(bp, _RET_IP_); + return bp->b_error; +} + +xfs_caddr_t +xfs_buf_offset( + xfs_buf_t *bp, + size_t offset) +{ + struct page *page; + + if (bp->b_flags & XBF_MAPPED) + return bp->b_addr + offset; + + offset += bp->b_offset; + page = bp->b_pages[offset >> PAGE_SHIFT]; + return (xfs_caddr_t)page_address(page) + (offset & (PAGE_SIZE-1)); +} + +/* + * Move data into or out of a buffer. + */ +void +xfs_buf_iomove( + xfs_buf_t *bp, /* buffer to process */ + size_t boff, /* starting buffer offset */ + size_t bsize, /* length to copy */ + void *data, /* data address */ + xfs_buf_rw_t mode) /* read/write/zero flag */ +{ + size_t bend, cpoff, csize; + struct page *page; + + bend = boff + bsize; + while (boff < bend) { + page = bp->b_pages[xfs_buf_btoct(boff + bp->b_offset)]; + cpoff = xfs_buf_poff(boff + bp->b_offset); + csize = min_t(size_t, + PAGE_SIZE-cpoff, bp->b_count_desired-boff); + + ASSERT(((csize + cpoff) <= PAGE_SIZE)); + + switch (mode) { + case XBRW_ZERO: + memset(page_address(page) + cpoff, 0, csize); + break; + case XBRW_READ: + memcpy(data, page_address(page) + cpoff, csize); + break; + case XBRW_WRITE: + memcpy(page_address(page) + cpoff, data, csize); + } + + boff += csize; + data += csize; + } +} + +/* + * Handling of buffer targets (buftargs). + */ + +/* + * Wait for any bufs with callbacks that have been submitted but have not yet + * returned. These buffers will have an elevated hold count, so wait on those + * while freeing all the buffers only held by the LRU. + */ +void +xfs_wait_buftarg( + struct xfs_buftarg *btp) +{ + struct xfs_buf *bp; + +restart: + spin_lock(&btp->bt_lru_lock); + while (!list_empty(&btp->bt_lru)) { + bp = list_first_entry(&btp->bt_lru, struct xfs_buf, b_lru); + if (atomic_read(&bp->b_hold) > 1) { + spin_unlock(&btp->bt_lru_lock); + delay(100); + goto restart; + } + /* + * clear the LRU reference count so the bufer doesn't get + * ignored in xfs_buf_rele(). + */ + atomic_set(&bp->b_lru_ref, 0); + spin_unlock(&btp->bt_lru_lock); + xfs_buf_rele(bp); + spin_lock(&btp->bt_lru_lock); + } + spin_unlock(&btp->bt_lru_lock); +} + +int +xfs_buftarg_shrink( + struct shrinker *shrink, + struct shrink_control *sc) +{ + struct xfs_buftarg *btp = container_of(shrink, + struct xfs_buftarg, bt_shrinker); + struct xfs_buf *bp; + int nr_to_scan = sc->nr_to_scan; + LIST_HEAD(dispose); + + if (!nr_to_scan) + return btp->bt_lru_nr; + + spin_lock(&btp->bt_lru_lock); + while (!list_empty(&btp->bt_lru)) { + if (nr_to_scan-- <= 0) + break; + + bp = list_first_entry(&btp->bt_lru, struct xfs_buf, b_lru); + + /* + * Decrement the b_lru_ref count unless the value is already + * zero. If the value is already zero, we need to reclaim the + * buffer, otherwise it gets another trip through the LRU. + */ + if (!atomic_add_unless(&bp->b_lru_ref, -1, 0)) { + list_move_tail(&bp->b_lru, &btp->bt_lru); + continue; + } + + /* + * remove the buffer from the LRU now to avoid needing another + * lock round trip inside xfs_buf_rele(). + */ + list_move(&bp->b_lru, &dispose); + btp->bt_lru_nr--; + } + spin_unlock(&btp->bt_lru_lock); + + while (!list_empty(&dispose)) { + bp = list_first_entry(&dispose, struct xfs_buf, b_lru); + list_del_init(&bp->b_lru); + xfs_buf_rele(bp); + } + + return btp->bt_lru_nr; +} + +void +xfs_free_buftarg( + struct xfs_mount *mp, + struct xfs_buftarg *btp) +{ + unregister_shrinker(&btp->bt_shrinker); + + xfs_flush_buftarg(btp, 1); + if (mp->m_flags & XFS_MOUNT_BARRIER) + xfs_blkdev_issue_flush(btp); + + kthread_stop(btp->bt_task); + kmem_free(btp); +} + +STATIC int +xfs_setsize_buftarg_flags( + xfs_buftarg_t *btp, + unsigned int blocksize, + unsigned int sectorsize, + int verbose) +{ + btp->bt_bsize = blocksize; + btp->bt_sshift = ffs(sectorsize) - 1; + btp->bt_smask = sectorsize - 1; + + if (set_blocksize(btp->bt_bdev, sectorsize)) { + xfs_warn(btp->bt_mount, + "Cannot set_blocksize to %u on device %s\n", + sectorsize, xfs_buf_target_name(btp)); + return EINVAL; + } + + return 0; +} + +/* + * When allocating the initial buffer target we have not yet + * read in the superblock, so don't know what sized sectors + * are being used is at this early stage. Play safe. + */ +STATIC int +xfs_setsize_buftarg_early( + xfs_buftarg_t *btp, + struct block_device *bdev) +{ + return xfs_setsize_buftarg_flags(btp, + PAGE_SIZE, bdev_logical_block_size(bdev), 0); +} + +int +xfs_setsize_buftarg( + xfs_buftarg_t *btp, + unsigned int blocksize, + unsigned int sectorsize) +{ + return xfs_setsize_buftarg_flags(btp, blocksize, sectorsize, 1); +} + +STATIC int +xfs_alloc_delwrite_queue( + xfs_buftarg_t *btp, + const char *fsname) +{ + INIT_LIST_HEAD(&btp->bt_delwrite_queue); + spin_lock_init(&btp->bt_delwrite_lock); + btp->bt_flags = 0; + btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname); + if (IS_ERR(btp->bt_task)) + return PTR_ERR(btp->bt_task); + return 0; +} + +xfs_buftarg_t * +xfs_alloc_buftarg( + struct xfs_mount *mp, + struct block_device *bdev, + int external, + const char *fsname) +{ + xfs_buftarg_t *btp; + + btp = kmem_zalloc(sizeof(*btp), KM_SLEEP); + + btp->bt_mount = mp; + btp->bt_dev = bdev->bd_dev; + btp->bt_bdev = bdev; + btp->bt_bdi = blk_get_backing_dev_info(bdev); + if (!btp->bt_bdi) + goto error; + + INIT_LIST_HEAD(&btp->bt_lru); + spin_lock_init(&btp->bt_lru_lock); + if (xfs_setsize_buftarg_early(btp, bdev)) + goto error; + if (xfs_alloc_delwrite_queue(btp, fsname)) + goto error; + btp->bt_shrinker.shrink = xfs_buftarg_shrink; + btp->bt_shrinker.seeks = DEFAULT_SEEKS; + register_shrinker(&btp->bt_shrinker); + return btp; + +error: + kmem_free(btp); + return NULL; +} + + +/* + * Delayed write buffer handling + */ +STATIC void +xfs_buf_delwri_queue( + xfs_buf_t *bp, + int unlock) +{ + struct list_head *dwq = &bp->b_target->bt_delwrite_queue; + spinlock_t *dwlk = &bp->b_target->bt_delwrite_lock; + + trace_xfs_buf_delwri_queue(bp, _RET_IP_); + + ASSERT((bp->b_flags&(XBF_DELWRI|XBF_ASYNC)) == (XBF_DELWRI|XBF_ASYNC)); + + spin_lock(dwlk); + /* If already in the queue, dequeue and place at tail */ + if (!list_empty(&bp->b_list)) { + ASSERT(bp->b_flags & _XBF_DELWRI_Q); + if (unlock) + atomic_dec(&bp->b_hold); + list_del(&bp->b_list); + } + + if (list_empty(dwq)) { + /* start xfsbufd as it is about to have something to do */ + wake_up_process(bp->b_target->bt_task); + } + + bp->b_flags |= _XBF_DELWRI_Q; + list_add_tail(&bp->b_list, dwq); + bp->b_queuetime = jiffies; + spin_unlock(dwlk); + + if (unlock) + xfs_buf_unlock(bp); +} + +void +xfs_buf_delwri_dequeue( + xfs_buf_t *bp) +{ + spinlock_t *dwlk = &bp->b_target->bt_delwrite_lock; + int dequeued = 0; + + spin_lock(dwlk); + if ((bp->b_flags & XBF_DELWRI) && !list_empty(&bp->b_list)) { + ASSERT(bp->b_flags & _XBF_DELWRI_Q); + list_del_init(&bp->b_list); + dequeued = 1; + } + bp->b_flags &= ~(XBF_DELWRI|_XBF_DELWRI_Q); + spin_unlock(dwlk); + + if (dequeued) + xfs_buf_rele(bp); + + trace_xfs_buf_delwri_dequeue(bp, _RET_IP_); +} + +/* + * If a delwri buffer needs to be pushed before it has aged out, then promote + * it to the head of the delwri queue so that it will be flushed on the next + * xfsbufd run. We do this by resetting the queuetime of the buffer to be older + * than the age currently needed to flush the buffer. Hence the next time the + * xfsbufd sees it is guaranteed to be considered old enough to flush. + */ +void +xfs_buf_delwri_promote( + struct xfs_buf *bp) +{ + struct xfs_buftarg *btp = bp->b_target; + long age = xfs_buf_age_centisecs * msecs_to_jiffies(10) + 1; + + ASSERT(bp->b_flags & XBF_DELWRI); + ASSERT(bp->b_flags & _XBF_DELWRI_Q); + + /* + * Check the buffer age before locking the delayed write queue as we + * don't need to promote buffers that are already past the flush age. + */ + if (bp->b_queuetime < jiffies - age) + return; + bp->b_queuetime = jiffies - age; + spin_lock(&btp->bt_delwrite_lock); + list_move(&bp->b_list, &btp->bt_delwrite_queue); + spin_unlock(&btp->bt_delwrite_lock); +} + +STATIC void +xfs_buf_runall_queues( + struct workqueue_struct *queue) +{ + flush_workqueue(queue); +} + +/* + * Move as many buffers as specified to the supplied list + * idicating if we skipped any buffers to prevent deadlocks. + */ +STATIC int +xfs_buf_delwri_split( + xfs_buftarg_t *target, + struct list_head *list, + unsigned long age) +{ + xfs_buf_t *bp, *n; + struct list_head *dwq = &target->bt_delwrite_queue; + spinlock_t *dwlk = &target->bt_delwrite_lock; + int skipped = 0; + int force; + + force = test_and_clear_bit(XBT_FORCE_FLUSH, &target->bt_flags); + INIT_LIST_HEAD(list); + spin_lock(dwlk); + list_for_each_entry_safe(bp, n, dwq, b_list) { + ASSERT(bp->b_flags & XBF_DELWRI); + + if (!xfs_buf_ispinned(bp) && xfs_buf_trylock(bp)) { + if (!force && + time_before(jiffies, bp->b_queuetime + age)) { + xfs_buf_unlock(bp); + break; + } + + bp->b_flags &= ~(XBF_DELWRI | _XBF_DELWRI_Q); + bp->b_flags |= XBF_WRITE; + list_move_tail(&bp->b_list, list); + trace_xfs_buf_delwri_split(bp, _RET_IP_); + } else + skipped++; + } + spin_unlock(dwlk); + + return skipped; + +} + +/* + * Compare function is more complex than it needs to be because + * the return value is only 32 bits and we are doing comparisons + * on 64 bit values + */ +static int +xfs_buf_cmp( + void *priv, + struct list_head *a, + struct list_head *b) +{ + struct xfs_buf *ap = container_of(a, struct xfs_buf, b_list); + struct xfs_buf *bp = container_of(b, struct xfs_buf, b_list); + xfs_daddr_t diff; + + diff = ap->b_bn - bp->b_bn; + if (diff < 0) + return -1; + if (diff > 0) + return 1; + return 0; +} + +STATIC int +xfsbufd( + void *data) +{ + xfs_buftarg_t *target = (xfs_buftarg_t *)data; + + current->flags |= PF_MEMALLOC; + + set_freezable(); + + do { + long age = xfs_buf_age_centisecs * msecs_to_jiffies(10); + long tout = xfs_buf_timer_centisecs * msecs_to_jiffies(10); + struct list_head tmp; + struct blk_plug plug; + + if (unlikely(freezing(current))) { + set_bit(XBT_FORCE_SLEEP, &target->bt_flags); + refrigerator(); + } else { + clear_bit(XBT_FORCE_SLEEP, &target->bt_flags); + } + + /* sleep for a long time if there is nothing to do. */ + if (list_empty(&target->bt_delwrite_queue)) + tout = MAX_SCHEDULE_TIMEOUT; + schedule_timeout_interruptible(tout); + + xfs_buf_delwri_split(target, &tmp, age); + list_sort(NULL, &tmp, xfs_buf_cmp); + + blk_start_plug(&plug); + while (!list_empty(&tmp)) { + struct xfs_buf *bp; + bp = list_first_entry(&tmp, struct xfs_buf, b_list); + list_del_init(&bp->b_list); + xfs_bdstrat_cb(bp); + } + blk_finish_plug(&plug); + } while (!kthread_should_stop()); + + return 0; +} + +/* + * Go through all incore buffers, and release buffers if they belong to + * the given device. This is used in filesystem error handling to + * preserve the consistency of its metadata. + */ +int +xfs_flush_buftarg( + xfs_buftarg_t *target, + int wait) +{ + xfs_buf_t *bp; + int pincount = 0; + LIST_HEAD(tmp_list); + LIST_HEAD(wait_list); + struct blk_plug plug; + + xfs_buf_runall_queues(xfsconvertd_workqueue); + xfs_buf_runall_queues(xfsdatad_workqueue); + xfs_buf_runall_queues(xfslogd_workqueue); + + set_bit(XBT_FORCE_FLUSH, &target->bt_flags); + pincount = xfs_buf_delwri_split(target, &tmp_list, 0); + + /* + * Dropped the delayed write list lock, now walk the temporary list. + * All I/O is issued async and then if we need to wait for completion + * we do that after issuing all the IO. + */ + list_sort(NULL, &tmp_list, xfs_buf_cmp); + + blk_start_plug(&plug); + while (!list_empty(&tmp_list)) { + bp = list_first_entry(&tmp_list, struct xfs_buf, b_list); + ASSERT(target == bp->b_target); + list_del_init(&bp->b_list); + if (wait) { + bp->b_flags &= ~XBF_ASYNC; + list_add(&bp->b_list, &wait_list); + } + xfs_bdstrat_cb(bp); + } + blk_finish_plug(&plug); + + if (wait) { + /* Wait for IO to complete. */ + while (!list_empty(&wait_list)) { + bp = list_first_entry(&wait_list, struct xfs_buf, b_list); + + list_del_init(&bp->b_list); + xfs_buf_iowait(bp); + xfs_buf_relse(bp); + } + } + + return pincount; +} + +int __init +xfs_buf_init(void) +{ + xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf", + KM_ZONE_HWALIGN, NULL); + if (!xfs_buf_zone) + goto out; + + xfslogd_workqueue = alloc_workqueue("xfslogd", + WQ_MEM_RECLAIM | WQ_HIGHPRI, 1); + if (!xfslogd_workqueue) + goto out_free_buf_zone; + + xfsdatad_workqueue = alloc_workqueue("xfsdatad", WQ_MEM_RECLAIM, 1); + if (!xfsdatad_workqueue) + goto out_destroy_xfslogd_workqueue; + + xfsconvertd_workqueue = alloc_workqueue("xfsconvertd", + WQ_MEM_RECLAIM, 1); + if (!xfsconvertd_workqueue) + goto out_destroy_xfsdatad_workqueue; + + return 0; + + out_destroy_xfsdatad_workqueue: + destroy_workqueue(xfsdatad_workqueue); + out_destroy_xfslogd_workqueue: + destroy_workqueue(xfslogd_workqueue); + out_free_buf_zone: + kmem_zone_destroy(xfs_buf_zone); + out: + return -ENOMEM; +} + +void +xfs_buf_terminate(void) +{ + destroy_workqueue(xfsconvertd_workqueue); + destroy_workqueue(xfsdatad_workqueue); + destroy_workqueue(xfslogd_workqueue); + kmem_zone_destroy(xfs_buf_zone); +} + +#ifdef CONFIG_KDB_MODULES +struct list_head * +xfs_get_buftarg_list(void) +{ + return &xfs_buftarg_list; +} +#endif diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h new file mode 100644 index 000000000000..620972b8094d --- /dev/null +++ b/fs/xfs/xfs_buf.h @@ -0,0 +1,326 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_BUF_H__ +#define __XFS_BUF_H__ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Base types + */ + +#define XFS_BUF_DADDR_NULL ((xfs_daddr_t) (-1LL)) + +#define xfs_buf_ctob(pp) ((pp) * PAGE_CACHE_SIZE) +#define xfs_buf_btoc(dd) (((dd) + PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) +#define xfs_buf_btoct(dd) ((dd) >> PAGE_CACHE_SHIFT) +#define xfs_buf_poff(aa) ((aa) & ~PAGE_CACHE_MASK) + +typedef enum { + XBRW_READ = 1, /* transfer into target memory */ + XBRW_WRITE = 2, /* transfer from target memory */ + XBRW_ZERO = 3, /* Zero target memory */ +} xfs_buf_rw_t; + +#define XBF_READ (1 << 0) /* buffer intended for reading from device */ +#define XBF_WRITE (1 << 1) /* buffer intended for writing to device */ +#define XBF_READ_AHEAD (1 << 2) /* asynchronous read-ahead */ +#define XBF_MAPPED (1 << 3) /* buffer mapped (b_addr valid) */ +#define XBF_ASYNC (1 << 4) /* initiator will not wait for completion */ +#define XBF_DONE (1 << 5) /* all pages in the buffer uptodate */ +#define XBF_DELWRI (1 << 6) /* buffer has dirty pages */ +#define XBF_STALE (1 << 7) /* buffer has been staled, do not find it */ + +/* I/O hints for the BIO layer */ +#define XBF_SYNCIO (1 << 10)/* treat this buffer as synchronous I/O */ +#define XBF_FUA (1 << 11)/* force cache write through mode */ +#define XBF_FLUSH (1 << 12)/* flush the disk cache before a write */ + +/* flags used only as arguments to access routines */ +#define XBF_LOCK (1 << 15)/* lock requested */ +#define XBF_TRYLOCK (1 << 16)/* lock requested, but do not wait */ +#define XBF_DONT_BLOCK (1 << 17)/* do not block in current thread */ + +/* flags used only internally */ +#define _XBF_PAGES (1 << 20)/* backed by refcounted pages */ +#define _XBF_KMEM (1 << 21)/* backed by heap memory */ +#define _XBF_DELWRI_Q (1 << 22)/* buffer on delwri queue */ + +typedef unsigned int xfs_buf_flags_t; + +#define XFS_BUF_FLAGS \ + { XBF_READ, "READ" }, \ + { XBF_WRITE, "WRITE" }, \ + { XBF_READ_AHEAD, "READ_AHEAD" }, \ + { XBF_MAPPED, "MAPPED" }, \ + { XBF_ASYNC, "ASYNC" }, \ + { XBF_DONE, "DONE" }, \ + { XBF_DELWRI, "DELWRI" }, \ + { XBF_STALE, "STALE" }, \ + { XBF_SYNCIO, "SYNCIO" }, \ + { XBF_FUA, "FUA" }, \ + { XBF_FLUSH, "FLUSH" }, \ + { XBF_LOCK, "LOCK" }, /* should never be set */\ + { XBF_TRYLOCK, "TRYLOCK" }, /* ditto */\ + { XBF_DONT_BLOCK, "DONT_BLOCK" }, /* ditto */\ + { _XBF_PAGES, "PAGES" }, \ + { _XBF_KMEM, "KMEM" }, \ + { _XBF_DELWRI_Q, "DELWRI_Q" } + +typedef enum { + XBT_FORCE_SLEEP = 0, + XBT_FORCE_FLUSH = 1, +} xfs_buftarg_flags_t; + +typedef struct xfs_buftarg { + dev_t bt_dev; + struct block_device *bt_bdev; + struct backing_dev_info *bt_bdi; + struct xfs_mount *bt_mount; + unsigned int bt_bsize; + unsigned int bt_sshift; + size_t bt_smask; + + /* per device delwri queue */ + struct task_struct *bt_task; + struct list_head bt_delwrite_queue; + spinlock_t bt_delwrite_lock; + unsigned long bt_flags; + + /* LRU control structures */ + struct shrinker bt_shrinker; + struct list_head bt_lru; + spinlock_t bt_lru_lock; + unsigned int bt_lru_nr; +} xfs_buftarg_t; + +struct xfs_buf; +typedef void (*xfs_buf_iodone_t)(struct xfs_buf *); + +#define XB_PAGES 2 + +typedef struct xfs_buf { + /* + * first cacheline holds all the fields needed for an uncontended cache + * hit to be fully processed. The semaphore straddles the cacheline + * boundary, but the counter and lock sits on the first cacheline, + * which is the only bit that is touched if we hit the semaphore + * fast-path on locking. + */ + struct rb_node b_rbnode; /* rbtree node */ + xfs_off_t b_file_offset; /* offset in file */ + size_t b_buffer_length;/* size of buffer in bytes */ + atomic_t b_hold; /* reference count */ + atomic_t b_lru_ref; /* lru reclaim ref count */ + xfs_buf_flags_t b_flags; /* status flags */ + struct semaphore b_sema; /* semaphore for lockables */ + + struct list_head b_lru; /* lru list */ + wait_queue_head_t b_waiters; /* unpin waiters */ + struct list_head b_list; + struct xfs_perag *b_pag; /* contains rbtree root */ + xfs_buftarg_t *b_target; /* buffer target (device) */ + xfs_daddr_t b_bn; /* block number for I/O */ + size_t b_count_desired;/* desired transfer size */ + void *b_addr; /* virtual address of buffer */ + struct work_struct b_iodone_work; + xfs_buf_iodone_t b_iodone; /* I/O completion function */ + struct completion b_iowait; /* queue for I/O waiters */ + void *b_fspriv; + struct xfs_trans *b_transp; + struct page **b_pages; /* array of page pointers */ + struct page *b_page_array[XB_PAGES]; /* inline pages */ + unsigned long b_queuetime; /* time buffer was queued */ + atomic_t b_pin_count; /* pin count */ + atomic_t b_io_remaining; /* #outstanding I/O requests */ + unsigned int b_page_count; /* size of page array */ + unsigned int b_offset; /* page offset in first page */ + unsigned short b_error; /* error code on I/O */ +#ifdef XFS_BUF_LOCK_TRACKING + int b_last_holder; +#endif +} xfs_buf_t; + + +/* Finding and Reading Buffers */ +extern xfs_buf_t *_xfs_buf_find(xfs_buftarg_t *, xfs_off_t, size_t, + xfs_buf_flags_t, xfs_buf_t *); +#define xfs_incore(buftarg,blkno,len,lockit) \ + _xfs_buf_find(buftarg, blkno ,len, lockit, NULL) + +extern xfs_buf_t *xfs_buf_get(xfs_buftarg_t *, xfs_off_t, size_t, + xfs_buf_flags_t); +extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t, + xfs_buf_flags_t); + +extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *); +extern void xfs_buf_set_empty(struct xfs_buf *bp, size_t len); +extern xfs_buf_t *xfs_buf_get_uncached(struct xfs_buftarg *, size_t, int); +extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t); +extern void xfs_buf_hold(xfs_buf_t *); +extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t); +struct xfs_buf *xfs_buf_read_uncached(struct xfs_mount *mp, + struct xfs_buftarg *target, + xfs_daddr_t daddr, size_t length, int flags); + +/* Releasing Buffers */ +extern void xfs_buf_free(xfs_buf_t *); +extern void xfs_buf_rele(xfs_buf_t *); + +/* Locking and Unlocking Buffers */ +extern int xfs_buf_trylock(xfs_buf_t *); +extern void xfs_buf_lock(xfs_buf_t *); +extern void xfs_buf_unlock(xfs_buf_t *); +#define xfs_buf_islocked(bp) \ + ((bp)->b_sema.count <= 0) + +/* Buffer Read and Write Routines */ +extern int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp); +extern void xfs_bdwrite(void *mp, xfs_buf_t *bp); + +extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *); +extern int xfs_bdstrat_cb(struct xfs_buf *); + +extern void xfs_buf_ioend(xfs_buf_t *, int); +extern void xfs_buf_ioerror(xfs_buf_t *, int); +extern int xfs_buf_iorequest(xfs_buf_t *); +extern int xfs_buf_iowait(xfs_buf_t *); +extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *, + xfs_buf_rw_t); +#define xfs_buf_zero(bp, off, len) \ + xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO) + +static inline int xfs_buf_geterror(xfs_buf_t *bp) +{ + return bp ? bp->b_error : ENOMEM; +} + +/* Buffer Utility Routines */ +extern xfs_caddr_t xfs_buf_offset(xfs_buf_t *, size_t); + +/* Delayed Write Buffer Routines */ +extern void xfs_buf_delwri_dequeue(xfs_buf_t *); +extern void xfs_buf_delwri_promote(xfs_buf_t *); + +/* Buffer Daemon Setup Routines */ +extern int xfs_buf_init(void); +extern void xfs_buf_terminate(void); + +static inline const char * +xfs_buf_target_name(struct xfs_buftarg *target) +{ + static char __b[BDEVNAME_SIZE]; + + return bdevname(target->bt_bdev, __b); +} + + +#define XFS_BUF_ZEROFLAGS(bp) \ + ((bp)->b_flags &= ~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI| \ + XBF_SYNCIO|XBF_FUA|XBF_FLUSH)) + +void xfs_buf_stale(struct xfs_buf *bp); +#define XFS_BUF_STALE(bp) xfs_buf_stale(bp); +#define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XBF_STALE) +#define XFS_BUF_ISSTALE(bp) ((bp)->b_flags & XBF_STALE) +#define XFS_BUF_SUPER_STALE(bp) do { \ + XFS_BUF_STALE(bp); \ + xfs_buf_delwri_dequeue(bp); \ + XFS_BUF_DONE(bp); \ + } while (0) + +#define XFS_BUF_DELAYWRITE(bp) ((bp)->b_flags |= XBF_DELWRI) +#define XFS_BUF_UNDELAYWRITE(bp) xfs_buf_delwri_dequeue(bp) +#define XFS_BUF_ISDELAYWRITE(bp) ((bp)->b_flags & XBF_DELWRI) + +#define XFS_BUF_DONE(bp) ((bp)->b_flags |= XBF_DONE) +#define XFS_BUF_UNDONE(bp) ((bp)->b_flags &= ~XBF_DONE) +#define XFS_BUF_ISDONE(bp) ((bp)->b_flags & XBF_DONE) + +#define XFS_BUF_ASYNC(bp) ((bp)->b_flags |= XBF_ASYNC) +#define XFS_BUF_UNASYNC(bp) ((bp)->b_flags &= ~XBF_ASYNC) +#define XFS_BUF_ISASYNC(bp) ((bp)->b_flags & XBF_ASYNC) + +#define XFS_BUF_READ(bp) ((bp)->b_flags |= XBF_READ) +#define XFS_BUF_UNREAD(bp) ((bp)->b_flags &= ~XBF_READ) +#define XFS_BUF_ISREAD(bp) ((bp)->b_flags & XBF_READ) + +#define XFS_BUF_WRITE(bp) ((bp)->b_flags |= XBF_WRITE) +#define XFS_BUF_UNWRITE(bp) ((bp)->b_flags &= ~XBF_WRITE) +#define XFS_BUF_ISWRITE(bp) ((bp)->b_flags & XBF_WRITE) + +#define XFS_BUF_ADDR(bp) ((bp)->b_bn) +#define XFS_BUF_SET_ADDR(bp, bno) ((bp)->b_bn = (xfs_daddr_t)(bno)) +#define XFS_BUF_OFFSET(bp) ((bp)->b_file_offset) +#define XFS_BUF_SET_OFFSET(bp, off) ((bp)->b_file_offset = (off)) +#define XFS_BUF_COUNT(bp) ((bp)->b_count_desired) +#define XFS_BUF_SET_COUNT(bp, cnt) ((bp)->b_count_desired = (cnt)) +#define XFS_BUF_SIZE(bp) ((bp)->b_buffer_length) +#define XFS_BUF_SET_SIZE(bp, cnt) ((bp)->b_buffer_length = (cnt)) + +static inline void +xfs_buf_set_ref( + struct xfs_buf *bp, + int lru_ref) +{ + atomic_set(&bp->b_lru_ref, lru_ref); +} +#define XFS_BUF_SET_VTYPE_REF(bp, type, ref) xfs_buf_set_ref(bp, ref) +#define XFS_BUF_SET_VTYPE(bp, type) do { } while (0) + +static inline int xfs_buf_ispinned(struct xfs_buf *bp) +{ + return atomic_read(&bp->b_pin_count); +} + +#define XFS_BUF_FINISH_IOWAIT(bp) complete(&bp->b_iowait); + +static inline void xfs_buf_relse(xfs_buf_t *bp) +{ + xfs_buf_unlock(bp); + xfs_buf_rele(bp); +} + +/* + * Handling of buftargs. + */ +extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *, + struct block_device *, int, const char *); +extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *); +extern void xfs_wait_buftarg(xfs_buftarg_t *); +extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int); +extern int xfs_flush_buftarg(xfs_buftarg_t *, int); + +#ifdef CONFIG_KDB_MODULES +extern struct list_head *xfs_get_buftarg_list(void); +#endif + +#define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev) +#define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev) + +#define xfs_binval(buftarg) xfs_flush_buftarg(buftarg, 1) +#define XFS_bflush(buftarg) xfs_flush_buftarg(buftarg, 1) + +#endif /* __XFS_BUF_H__ */ diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c new file mode 100644 index 000000000000..244e797dae32 --- /dev/null +++ b/fs/xfs/xfs_discard.c @@ -0,0 +1,222 @@ +/* + * Copyright (C) 2010 Red Hat, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_sb.h" +#include "xfs_inum.h" +#include "xfs_log.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_quota.h" +#include "xfs_trans.h" +#include "xfs_alloc_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_btree.h" +#include "xfs_inode.h" +#include "xfs_alloc.h" +#include "xfs_error.h" +#include "xfs_discard.h" +#include "xfs_trace.h" + +STATIC int +xfs_trim_extents( + struct xfs_mount *mp, + xfs_agnumber_t agno, + xfs_fsblock_t start, + xfs_fsblock_t len, + xfs_fsblock_t minlen, + __uint64_t *blocks_trimmed) +{ + struct block_device *bdev = mp->m_ddev_targp->bt_bdev; + struct xfs_btree_cur *cur; + struct xfs_buf *agbp; + struct xfs_perag *pag; + int error; + int i; + + pag = xfs_perag_get(mp, agno); + + error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp); + if (error || !agbp) + goto out_put_perag; + + cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT); + + /* + * Force out the log. This means any transactions that might have freed + * space before we took the AGF buffer lock are now on disk, and the + * volatile disk cache is flushed. + */ + xfs_log_force(mp, XFS_LOG_SYNC); + + /* + * Look up the longest btree in the AGF and start with it. + */ + error = xfs_alloc_lookup_le(cur, 0, + XFS_BUF_TO_AGF(agbp)->agf_longest, &i); + if (error) + goto out_del_cursor; + + /* + * Loop until we are done with all extents that are large + * enough to be worth discarding. + */ + while (i) { + xfs_agblock_t fbno; + xfs_extlen_t flen; + + error = xfs_alloc_get_rec(cur, &fbno, &flen, &i); + if (error) + goto out_del_cursor; + XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor); + ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest); + + /* + * Too small? Give up. + */ + if (flen < minlen) { + trace_xfs_discard_toosmall(mp, agno, fbno, flen); + goto out_del_cursor; + } + + /* + * If the extent is entirely outside of the range we are + * supposed to discard skip it. Do not bother to trim + * down partially overlapping ranges for now. + */ + if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start || + XFS_AGB_TO_FSB(mp, agno, fbno) >= start + len) { + trace_xfs_discard_exclude(mp, agno, fbno, flen); + goto next_extent; + } + + /* + * If any blocks in the range are still busy, skip the + * discard and try again the next time. + */ + if (xfs_alloc_busy_search(mp, agno, fbno, flen)) { + trace_xfs_discard_busy(mp, agno, fbno, flen); + goto next_extent; + } + + trace_xfs_discard_extent(mp, agno, fbno, flen); + error = -blkdev_issue_discard(bdev, + XFS_AGB_TO_DADDR(mp, agno, fbno), + XFS_FSB_TO_BB(mp, flen), + GFP_NOFS, 0); + if (error) + goto out_del_cursor; + *blocks_trimmed += flen; + +next_extent: + error = xfs_btree_decrement(cur, 0, &i); + if (error) + goto out_del_cursor; + } + +out_del_cursor: + xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR); + xfs_buf_relse(agbp); +out_put_perag: + xfs_perag_put(pag); + return error; +} + +int +xfs_ioc_trim( + struct xfs_mount *mp, + struct fstrim_range __user *urange) +{ + struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue; + unsigned int granularity = q->limits.discard_granularity; + struct fstrim_range range; + xfs_fsblock_t start, len, minlen; + xfs_agnumber_t start_agno, end_agno, agno; + __uint64_t blocks_trimmed = 0; + int error, last_error = 0; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + if (!blk_queue_discard(q)) + return -XFS_ERROR(EOPNOTSUPP); + if (copy_from_user(&range, urange, sizeof(range))) + return -XFS_ERROR(EFAULT); + + /* + * Truncating down the len isn't actually quite correct, but using + * XFS_B_TO_FSB would mean we trivially get overflows for values + * of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default + * used by the fstrim application. In the end it really doesn't + * matter as trimming blocks is an advisory interface. + */ + start = XFS_B_TO_FSBT(mp, range.start); + len = XFS_B_TO_FSBT(mp, range.len); + minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen)); + + start_agno = XFS_FSB_TO_AGNO(mp, start); + if (start_agno >= mp->m_sb.sb_agcount) + return -XFS_ERROR(EINVAL); + + end_agno = XFS_FSB_TO_AGNO(mp, start + len); + if (end_agno >= mp->m_sb.sb_agcount) + end_agno = mp->m_sb.sb_agcount - 1; + + for (agno = start_agno; agno <= end_agno; agno++) { + error = -xfs_trim_extents(mp, agno, start, len, minlen, + &blocks_trimmed); + if (error) + last_error = error; + } + + if (last_error) + return last_error; + + range.len = XFS_FSB_TO_B(mp, blocks_trimmed); + if (copy_to_user(urange, &range, sizeof(range))) + return -XFS_ERROR(EFAULT); + return 0; +} + +int +xfs_discard_extents( + struct xfs_mount *mp, + struct list_head *list) +{ + struct xfs_busy_extent *busyp; + int error = 0; + + list_for_each_entry(busyp, list, list) { + trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, + busyp->length); + + error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, + XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), + XFS_FSB_TO_BB(mp, busyp->length), + GFP_NOFS, 0); + if (error && error != EOPNOTSUPP) { + xfs_info(mp, + "discard failed for extent [0x%llu,%u], error %d", + (unsigned long long)busyp->bno, + busyp->length, + error); + return error; + } + } + + return 0; +} diff --git a/fs/xfs/xfs_discard.h b/fs/xfs/xfs_discard.h new file mode 100644 index 000000000000..344879aea646 --- /dev/null +++ b/fs/xfs/xfs_discard.h @@ -0,0 +1,10 @@ +#ifndef XFS_DISCARD_H +#define XFS_DISCARD_H 1 + +struct fstrim_range; +struct list_head; + +extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *); +extern int xfs_discard_extents(struct xfs_mount *, struct list_head *); + +#endif /* XFS_DISCARD_H */ diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c new file mode 100644 index 000000000000..db62959bed13 --- /dev/null +++ b/fs/xfs/xfs_dquot.c @@ -0,0 +1,1454 @@ +/* + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_itable.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_trans_space.h" +#include "xfs_trans_priv.h" +#include "xfs_qm.h" +#include "xfs_trace.h" + + +/* + LOCK ORDER + + inode lock (ilock) + dquot hash-chain lock (hashlock) + xqm dquot freelist lock (freelistlock + mount's dquot list lock (mplistlock) + user dquot lock - lock ordering among dquots is based on the uid or gid + group dquot lock - similar to udquots. Between the two dquots, the udquot + has to be locked first. + pin lock - the dquot lock must be held to take this lock. + flush lock - ditto. +*/ + +#ifdef DEBUG +xfs_buftarg_t *xfs_dqerror_target; +int xfs_do_dqerror; +int xfs_dqreq_num; +int xfs_dqerror_mod = 33; +#endif + +static struct lock_class_key xfs_dquot_other_class; + +/* + * Allocate and initialize a dquot. We don't always allocate fresh memory; + * we try to reclaim a free dquot if the number of incore dquots are above + * a threshold. + * The only field inside the core that gets initialized at this point + * is the d_id field. The idea is to fill in the entire q_core + * when we read in the on disk dquot. + */ +STATIC xfs_dquot_t * +xfs_qm_dqinit( + xfs_mount_t *mp, + xfs_dqid_t id, + uint type) +{ + xfs_dquot_t *dqp; + boolean_t brandnewdquot; + + brandnewdquot = xfs_qm_dqalloc_incore(&dqp); + dqp->dq_flags = type; + dqp->q_core.d_id = cpu_to_be32(id); + dqp->q_mount = mp; + + /* + * No need to re-initialize these if this is a reclaimed dquot. + */ + if (brandnewdquot) { + INIT_LIST_HEAD(&dqp->q_freelist); + mutex_init(&dqp->q_qlock); + init_waitqueue_head(&dqp->q_pinwait); + + /* + * Because we want to use a counting completion, complete + * the flush completion once to allow a single access to + * the flush completion without blocking. + */ + init_completion(&dqp->q_flush); + complete(&dqp->q_flush); + + trace_xfs_dqinit(dqp); + } else { + /* + * Only the q_core portion was zeroed in dqreclaim_one(). + * So, we need to reset others. + */ + dqp->q_nrefs = 0; + dqp->q_blkno = 0; + INIT_LIST_HEAD(&dqp->q_mplist); + INIT_LIST_HEAD(&dqp->q_hashlist); + dqp->q_bufoffset = 0; + dqp->q_fileoffset = 0; + dqp->q_transp = NULL; + dqp->q_gdquot = NULL; + dqp->q_res_bcount = 0; + dqp->q_res_icount = 0; + dqp->q_res_rtbcount = 0; + atomic_set(&dqp->q_pincount, 0); + dqp->q_hash = NULL; + ASSERT(list_empty(&dqp->q_freelist)); + + trace_xfs_dqreuse(dqp); + } + + /* + * In either case we need to make sure group quotas have a different + * lock class than user quotas, to make sure lockdep knows we can + * locks of one of each at the same time. + */ + if (!(type & XFS_DQ_USER)) + lockdep_set_class(&dqp->q_qlock, &xfs_dquot_other_class); + + /* + * log item gets initialized later + */ + return (dqp); +} + +/* + * This is called to free all the memory associated with a dquot + */ +void +xfs_qm_dqdestroy( + xfs_dquot_t *dqp) +{ + ASSERT(list_empty(&dqp->q_freelist)); + + mutex_destroy(&dqp->q_qlock); + kmem_zone_free(xfs_Gqm->qm_dqzone, dqp); + + atomic_dec(&xfs_Gqm->qm_totaldquots); +} + +/* + * This is what a 'fresh' dquot inside a dquot chunk looks like on disk. + */ +STATIC void +xfs_qm_dqinit_core( + xfs_dqid_t id, + uint type, + xfs_dqblk_t *d) +{ + /* + * Caller has zero'd the entire dquot 'chunk' already. + */ + d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + d->dd_diskdq.d_version = XFS_DQUOT_VERSION; + d->dd_diskdq.d_id = cpu_to_be32(id); + d->dd_diskdq.d_flags = type; +} + +/* + * If default limits are in force, push them into the dquot now. + * We overwrite the dquot limits only if they are zero and this + * is not the root dquot. + */ +void +xfs_qm_adjust_dqlimits( + xfs_mount_t *mp, + xfs_disk_dquot_t *d) +{ + xfs_quotainfo_t *q = mp->m_quotainfo; + + ASSERT(d->d_id); + + if (q->qi_bsoftlimit && !d->d_blk_softlimit) + d->d_blk_softlimit = cpu_to_be64(q->qi_bsoftlimit); + if (q->qi_bhardlimit && !d->d_blk_hardlimit) + d->d_blk_hardlimit = cpu_to_be64(q->qi_bhardlimit); + if (q->qi_isoftlimit && !d->d_ino_softlimit) + d->d_ino_softlimit = cpu_to_be64(q->qi_isoftlimit); + if (q->qi_ihardlimit && !d->d_ino_hardlimit) + d->d_ino_hardlimit = cpu_to_be64(q->qi_ihardlimit); + if (q->qi_rtbsoftlimit && !d->d_rtb_softlimit) + d->d_rtb_softlimit = cpu_to_be64(q->qi_rtbsoftlimit); + if (q->qi_rtbhardlimit && !d->d_rtb_hardlimit) + d->d_rtb_hardlimit = cpu_to_be64(q->qi_rtbhardlimit); +} + +/* + * Check the limits and timers of a dquot and start or reset timers + * if necessary. + * This gets called even when quota enforcement is OFF, which makes our + * life a little less complicated. (We just don't reject any quota + * reservations in that case, when enforcement is off). + * We also return 0 as the values of the timers in Q_GETQUOTA calls, when + * enforcement's off. + * In contrast, warnings are a little different in that they don't + * 'automatically' get started when limits get exceeded. They do + * get reset to zero, however, when we find the count to be under + * the soft limit (they are only ever set non-zero via userspace). + */ +void +xfs_qm_adjust_dqtimers( + xfs_mount_t *mp, + xfs_disk_dquot_t *d) +{ + ASSERT(d->d_id); + +#ifdef DEBUG + if (d->d_blk_hardlimit) + ASSERT(be64_to_cpu(d->d_blk_softlimit) <= + be64_to_cpu(d->d_blk_hardlimit)); + if (d->d_ino_hardlimit) + ASSERT(be64_to_cpu(d->d_ino_softlimit) <= + be64_to_cpu(d->d_ino_hardlimit)); + if (d->d_rtb_hardlimit) + ASSERT(be64_to_cpu(d->d_rtb_softlimit) <= + be64_to_cpu(d->d_rtb_hardlimit)); +#endif + + if (!d->d_btimer) { + if ((d->d_blk_softlimit && + (be64_to_cpu(d->d_bcount) >= + be64_to_cpu(d->d_blk_softlimit))) || + (d->d_blk_hardlimit && + (be64_to_cpu(d->d_bcount) >= + be64_to_cpu(d->d_blk_hardlimit)))) { + d->d_btimer = cpu_to_be32(get_seconds() + + mp->m_quotainfo->qi_btimelimit); + } else { + d->d_bwarns = 0; + } + } else { + if ((!d->d_blk_softlimit || + (be64_to_cpu(d->d_bcount) < + be64_to_cpu(d->d_blk_softlimit))) && + (!d->d_blk_hardlimit || + (be64_to_cpu(d->d_bcount) < + be64_to_cpu(d->d_blk_hardlimit)))) { + d->d_btimer = 0; + } + } + + if (!d->d_itimer) { + if ((d->d_ino_softlimit && + (be64_to_cpu(d->d_icount) >= + be64_to_cpu(d->d_ino_softlimit))) || + (d->d_ino_hardlimit && + (be64_to_cpu(d->d_icount) >= + be64_to_cpu(d->d_ino_hardlimit)))) { + d->d_itimer = cpu_to_be32(get_seconds() + + mp->m_quotainfo->qi_itimelimit); + } else { + d->d_iwarns = 0; + } + } else { + if ((!d->d_ino_softlimit || + (be64_to_cpu(d->d_icount) < + be64_to_cpu(d->d_ino_softlimit))) && + (!d->d_ino_hardlimit || + (be64_to_cpu(d->d_icount) < + be64_to_cpu(d->d_ino_hardlimit)))) { + d->d_itimer = 0; + } + } + + if (!d->d_rtbtimer) { + if ((d->d_rtb_softlimit && + (be64_to_cpu(d->d_rtbcount) >= + be64_to_cpu(d->d_rtb_softlimit))) || + (d->d_rtb_hardlimit && + (be64_to_cpu(d->d_rtbcount) >= + be64_to_cpu(d->d_rtb_hardlimit)))) { + d->d_rtbtimer = cpu_to_be32(get_seconds() + + mp->m_quotainfo->qi_rtbtimelimit); + } else { + d->d_rtbwarns = 0; + } + } else { + if ((!d->d_rtb_softlimit || + (be64_to_cpu(d->d_rtbcount) < + be64_to_cpu(d->d_rtb_softlimit))) && + (!d->d_rtb_hardlimit || + (be64_to_cpu(d->d_rtbcount) < + be64_to_cpu(d->d_rtb_hardlimit)))) { + d->d_rtbtimer = 0; + } + } +} + +/* + * initialize a buffer full of dquots and log the whole thing + */ +STATIC void +xfs_qm_init_dquot_blk( + xfs_trans_t *tp, + xfs_mount_t *mp, + xfs_dqid_t id, + uint type, + xfs_buf_t *bp) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + xfs_dqblk_t *d; + int curid, i; + + ASSERT(tp); + ASSERT(xfs_buf_islocked(bp)); + + d = bp->b_addr; + + /* + * ID of the first dquot in the block - id's are zero based. + */ + curid = id - (id % q->qi_dqperchunk); + ASSERT(curid >= 0); + memset(d, 0, BBTOB(q->qi_dqchunklen)); + for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) + xfs_qm_dqinit_core(curid, type, d); + xfs_trans_dquot_buf(tp, bp, + (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF : + ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF : + XFS_BLF_GDQUOT_BUF))); + xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1); +} + + + +/* + * Allocate a block and fill it with dquots. + * This is called when the bmapi finds a hole. + */ +STATIC int +xfs_qm_dqalloc( + xfs_trans_t **tpp, + xfs_mount_t *mp, + xfs_dquot_t *dqp, + xfs_inode_t *quotip, + xfs_fileoff_t offset_fsb, + xfs_buf_t **O_bpp) +{ + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + xfs_bmbt_irec_t map; + int nmaps, error, committed; + xfs_buf_t *bp; + xfs_trans_t *tp = *tpp; + + ASSERT(tp != NULL); + + trace_xfs_dqalloc(dqp); + + /* + * Initialize the bmap freelist prior to calling bmapi code. + */ + xfs_bmap_init(&flist, &firstblock); + xfs_ilock(quotip, XFS_ILOCK_EXCL); + /* + * Return if this type of quotas is turned off while we didn't + * have an inode lock + */ + if (XFS_IS_THIS_QUOTA_OFF(dqp)) { + xfs_iunlock(quotip, XFS_ILOCK_EXCL); + return (ESRCH); + } + + xfs_trans_ijoin_ref(tp, quotip, XFS_ILOCK_EXCL); + nmaps = 1; + if ((error = xfs_bmapi(tp, quotip, + offset_fsb, XFS_DQUOT_CLUSTER_SIZE_FSB, + XFS_BMAPI_METADATA | XFS_BMAPI_WRITE, + &firstblock, + XFS_QM_DQALLOC_SPACE_RES(mp), + &map, &nmaps, &flist))) { + goto error0; + } + ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB); + ASSERT(nmaps == 1); + ASSERT((map.br_startblock != DELAYSTARTBLOCK) && + (map.br_startblock != HOLESTARTBLOCK)); + + /* + * Keep track of the blkno to save a lookup later + */ + dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); + + /* now we can just get the buffer (there's nothing to read yet) */ + bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, + dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, + 0); + if (!bp || (error = xfs_buf_geterror(bp))) + goto error1; + /* + * Make a chunk of dquots out of this buffer and log + * the entire thing. + */ + xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id), + dqp->dq_flags & XFS_DQ_ALLTYPES, bp); + + /* + * xfs_bmap_finish() may commit the current transaction and + * start a second transaction if the freelist is not empty. + * + * Since we still want to modify this buffer, we need to + * ensure that the buffer is not released on commit of + * the first transaction and ensure the buffer is added to the + * second transaction. + * + * If there is only one transaction then don't stop the buffer + * from being released when it commits later on. + */ + + xfs_trans_bhold(tp, bp); + + if ((error = xfs_bmap_finish(tpp, &flist, &committed))) { + goto error1; + } + + if (committed) { + tp = *tpp; + xfs_trans_bjoin(tp, bp); + } else { + xfs_trans_bhold_release(tp, bp); + } + + *O_bpp = bp; + return 0; + + error1: + xfs_bmap_cancel(&flist); + error0: + xfs_iunlock(quotip, XFS_ILOCK_EXCL); + + return (error); +} + +/* + * Maps a dquot to the buffer containing its on-disk version. + * This returns a ptr to the buffer containing the on-disk dquot + * in the bpp param, and a ptr to the on-disk dquot within that buffer + */ +STATIC int +xfs_qm_dqtobp( + xfs_trans_t **tpp, + xfs_dquot_t *dqp, + xfs_disk_dquot_t **O_ddpp, + xfs_buf_t **O_bpp, + uint flags) +{ + xfs_bmbt_irec_t map; + int nmaps = 1, error; + xfs_buf_t *bp; + xfs_inode_t *quotip = XFS_DQ_TO_QIP(dqp); + xfs_mount_t *mp = dqp->q_mount; + xfs_disk_dquot_t *ddq; + xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id); + xfs_trans_t *tp = (tpp ? *tpp : NULL); + + dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk; + + xfs_ilock(quotip, XFS_ILOCK_SHARED); + if (XFS_IS_THIS_QUOTA_OFF(dqp)) { + /* + * Return if this type of quotas is turned off while we + * didn't have the quota inode lock. + */ + xfs_iunlock(quotip, XFS_ILOCK_SHARED); + return ESRCH; + } + + /* + * Find the block map; no allocations yet + */ + error = xfs_bmapi(NULL, quotip, dqp->q_fileoffset, + XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, + NULL, 0, &map, &nmaps, NULL); + + xfs_iunlock(quotip, XFS_ILOCK_SHARED); + if (error) + return error; + + ASSERT(nmaps == 1); + ASSERT(map.br_blockcount == 1); + + /* + * Offset of dquot in the (fixed sized) dquot chunk. + */ + dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) * + sizeof(xfs_dqblk_t); + + ASSERT(map.br_startblock != DELAYSTARTBLOCK); + if (map.br_startblock == HOLESTARTBLOCK) { + /* + * We don't allocate unless we're asked to + */ + if (!(flags & XFS_QMOPT_DQALLOC)) + return ENOENT; + + ASSERT(tp); + error = xfs_qm_dqalloc(tpp, mp, dqp, quotip, + dqp->q_fileoffset, &bp); + if (error) + return error; + tp = *tpp; + } else { + trace_xfs_dqtobp_read(dqp); + + /* + * store the blkno etc so that we don't have to do the + * mapping all the time + */ + dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock); + + error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, + dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, + 0, &bp); + if (error || !bp) + return XFS_ERROR(error); + } + + ASSERT(xfs_buf_islocked(bp)); + + /* + * calculate the location of the dquot inside the buffer. + */ + ddq = bp->b_addr + dqp->q_bufoffset; + + /* + * A simple sanity check in case we got a corrupted dquot... + */ + error = xfs_qm_dqcheck(mp, ddq, id, dqp->dq_flags & XFS_DQ_ALLTYPES, + flags & (XFS_QMOPT_DQREPAIR|XFS_QMOPT_DOWARN), + "dqtobp"); + if (error) { + if (!(flags & XFS_QMOPT_DQREPAIR)) { + xfs_trans_brelse(tp, bp); + return XFS_ERROR(EIO); + } + } + + *O_bpp = bp; + *O_ddpp = ddq; + + return (0); +} + + +/* + * Read in the ondisk dquot using dqtobp() then copy it to an incore version, + * and release the buffer immediately. + * + */ +/* ARGSUSED */ +STATIC int +xfs_qm_dqread( + xfs_trans_t **tpp, + xfs_dqid_t id, + xfs_dquot_t *dqp, /* dquot to get filled in */ + uint flags) +{ + xfs_disk_dquot_t *ddqp; + xfs_buf_t *bp; + int error; + xfs_trans_t *tp; + + ASSERT(tpp); + + trace_xfs_dqread(dqp); + + /* + * get a pointer to the on-disk dquot and the buffer containing it + * dqp already knows its own type (GROUP/USER). + */ + if ((error = xfs_qm_dqtobp(tpp, dqp, &ddqp, &bp, flags))) { + return (error); + } + tp = *tpp; + + /* copy everything from disk dquot to the incore dquot */ + memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t)); + ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); + xfs_qm_dquot_logitem_init(dqp); + + /* + * Reservation counters are defined as reservation plus current usage + * to avoid having to add every time. + */ + dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount); + dqp->q_res_icount = be64_to_cpu(ddqp->d_icount); + dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount); + + /* Mark the buf so that this will stay incore a little longer */ + XFS_BUF_SET_VTYPE_REF(bp, B_FS_DQUOT, XFS_DQUOT_REF); + + /* + * We got the buffer with a xfs_trans_read_buf() (in dqtobp()) + * So we need to release with xfs_trans_brelse(). + * The strategy here is identical to that of inodes; we lock + * the dquot in xfs_qm_dqget() before making it accessible to + * others. This is because dquots, like inodes, need a good level of + * concurrency, and we don't want to take locks on the entire buffers + * for dquot accesses. + * Note also that the dquot buffer may even be dirty at this point, if + * this particular dquot was repaired. We still aren't afraid to + * brelse it because we have the changes incore. + */ + ASSERT(xfs_buf_islocked(bp)); + xfs_trans_brelse(tp, bp); + + return (error); +} + + +/* + * allocate an incore dquot from the kernel heap, + * and fill its core with quota information kept on disk. + * If XFS_QMOPT_DQALLOC is set, it'll allocate a dquot on disk + * if it wasn't already allocated. + */ +STATIC int +xfs_qm_idtodq( + xfs_mount_t *mp, + xfs_dqid_t id, /* gid or uid, depending on type */ + uint type, /* UDQUOT or GDQUOT */ + uint flags, /* DQALLOC, DQREPAIR */ + xfs_dquot_t **O_dqpp)/* OUT : incore dquot, not locked */ +{ + xfs_dquot_t *dqp; + int error; + xfs_trans_t *tp; + int cancelflags=0; + + dqp = xfs_qm_dqinit(mp, id, type); + tp = NULL; + if (flags & XFS_QMOPT_DQALLOC) { + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC); + error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp), + XFS_WRITE_LOG_RES(mp) + + BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 + + 128, + 0, + XFS_TRANS_PERM_LOG_RES, + XFS_WRITE_LOG_COUNT); + if (error) { + cancelflags = 0; + goto error0; + } + cancelflags = XFS_TRANS_RELEASE_LOG_RES; + } + + /* + * Read it from disk; xfs_dqread() takes care of + * all the necessary initialization of dquot's fields (locks, etc) + */ + if ((error = xfs_qm_dqread(&tp, id, dqp, flags))) { + /* + * This can happen if quotas got turned off (ESRCH), + * or if the dquot didn't exist on disk and we ask to + * allocate (ENOENT). + */ + trace_xfs_dqread_fail(dqp); + cancelflags |= XFS_TRANS_ABORT; + goto error0; + } + if (tp) { + if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) + goto error1; + } + + *O_dqpp = dqp; + return (0); + + error0: + ASSERT(error); + if (tp) + xfs_trans_cancel(tp, cancelflags); + error1: + xfs_qm_dqdestroy(dqp); + *O_dqpp = NULL; + return (error); +} + +/* + * Lookup a dquot in the incore dquot hashtable. We keep two separate + * hashtables for user and group dquots; and, these are global tables + * inside the XQM, not per-filesystem tables. + * The hash chain must be locked by caller, and it is left locked + * on return. Returning dquot is locked. + */ +STATIC int +xfs_qm_dqlookup( + xfs_mount_t *mp, + xfs_dqid_t id, + xfs_dqhash_t *qh, + xfs_dquot_t **O_dqpp) +{ + xfs_dquot_t *dqp; + uint flist_locked; + + ASSERT(mutex_is_locked(&qh->qh_lock)); + + flist_locked = B_FALSE; + + /* + * Traverse the hashchain looking for a match + */ + list_for_each_entry(dqp, &qh->qh_list, q_hashlist) { + /* + * We already have the hashlock. We don't need the + * dqlock to look at the id field of the dquot, since the + * id can't be modified without the hashlock anyway. + */ + if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) { + trace_xfs_dqlookup_found(dqp); + + /* + * All in core dquots must be on the dqlist of mp + */ + ASSERT(!list_empty(&dqp->q_mplist)); + + xfs_dqlock(dqp); + if (dqp->q_nrefs == 0) { + ASSERT(!list_empty(&dqp->q_freelist)); + if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { + trace_xfs_dqlookup_want(dqp); + + /* + * We may have raced with dqreclaim_one() + * (and lost). So, flag that we don't + * want the dquot to be reclaimed. + */ + dqp->dq_flags |= XFS_DQ_WANT; + xfs_dqunlock(dqp); + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + xfs_dqlock(dqp); + dqp->dq_flags &= ~(XFS_DQ_WANT); + } + flist_locked = B_TRUE; + } + + /* + * id couldn't have changed; we had the hashlock all + * along + */ + ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); + + if (flist_locked) { + if (dqp->q_nrefs != 0) { + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + flist_locked = B_FALSE; + } else { + /* take it off the freelist */ + trace_xfs_dqlookup_freelist(dqp); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + } + } + + XFS_DQHOLD(dqp); + + if (flist_locked) + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + /* + * move the dquot to the front of the hashchain + */ + ASSERT(mutex_is_locked(&qh->qh_lock)); + list_move(&dqp->q_hashlist, &qh->qh_list); + trace_xfs_dqlookup_done(dqp); + *O_dqpp = dqp; + return 0; + } + } + + *O_dqpp = NULL; + ASSERT(mutex_is_locked(&qh->qh_lock)); + return (1); +} + +/* + * Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a + * a locked dquot, doing an allocation (if requested) as needed. + * When both an inode and an id are given, the inode's id takes precedence. + * That is, if the id changes while we don't hold the ilock inside this + * function, the new dquot is returned, not necessarily the one requested + * in the id argument. + */ +int +xfs_qm_dqget( + xfs_mount_t *mp, + xfs_inode_t *ip, /* locked inode (optional) */ + xfs_dqid_t id, /* uid/projid/gid depending on type */ + uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */ + uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */ + xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */ +{ + xfs_dquot_t *dqp; + xfs_dqhash_t *h; + uint version; + int error; + + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) || + (! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) || + (! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) { + return (ESRCH); + } + h = XFS_DQ_HASH(mp, id, type); + +#ifdef DEBUG + if (xfs_do_dqerror) { + if ((xfs_dqerror_target == mp->m_ddev_targp) && + (xfs_dqreq_num++ % xfs_dqerror_mod) == 0) { + xfs_debug(mp, "Returning error in dqget"); + return (EIO); + } + } +#endif + + again: + +#ifdef DEBUG + ASSERT(type == XFS_DQ_USER || + type == XFS_DQ_PROJ || + type == XFS_DQ_GROUP); + if (ip) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + if (type == XFS_DQ_USER) + ASSERT(ip->i_udquot == NULL); + else + ASSERT(ip->i_gdquot == NULL); + } +#endif + mutex_lock(&h->qh_lock); + + /* + * Look in the cache (hashtable). + * The chain is kept locked during lookup. + */ + if (xfs_qm_dqlookup(mp, id, h, O_dqpp) == 0) { + XQM_STATS_INC(xqmstats.xs_qm_dqcachehits); + /* + * The dquot was found, moved to the front of the chain, + * taken off the freelist if it was on it, and locked + * at this point. Just unlock the hashchain and return. + */ + ASSERT(*O_dqpp); + ASSERT(XFS_DQ_IS_LOCKED(*O_dqpp)); + mutex_unlock(&h->qh_lock); + trace_xfs_dqget_hit(*O_dqpp); + return (0); /* success */ + } + XQM_STATS_INC(xqmstats.xs_qm_dqcachemisses); + + /* + * Dquot cache miss. We don't want to keep the inode lock across + * a (potential) disk read. Also we don't want to deal with the lock + * ordering between quotainode and this inode. OTOH, dropping the inode + * lock here means dealing with a chown that can happen before + * we re-acquire the lock. + */ + if (ip) + xfs_iunlock(ip, XFS_ILOCK_EXCL); + /* + * Save the hashchain version stamp, and unlock the chain, so that + * we don't keep the lock across a disk read + */ + version = h->qh_version; + mutex_unlock(&h->qh_lock); + + /* + * Allocate the dquot on the kernel heap, and read the ondisk + * portion off the disk. Also, do all the necessary initialization + * This can return ENOENT if dquot didn't exist on disk and we didn't + * ask it to allocate; ESRCH if quotas got turned off suddenly. + */ + if ((error = xfs_qm_idtodq(mp, id, type, + flags & (XFS_QMOPT_DQALLOC|XFS_QMOPT_DQREPAIR| + XFS_QMOPT_DOWARN), + &dqp))) { + if (ip) + xfs_ilock(ip, XFS_ILOCK_EXCL); + return (error); + } + + /* + * See if this is mount code calling to look at the overall quota limits + * which are stored in the id == 0 user or group's dquot. + * Since we may not have done a quotacheck by this point, just return + * the dquot without attaching it to any hashtables, lists, etc, or even + * taking a reference. + * The caller must dqdestroy this once done. + */ + if (flags & XFS_QMOPT_DQSUSER) { + ASSERT(id == 0); + ASSERT(! ip); + goto dqret; + } + + /* + * Dquot lock comes after hashlock in the lock ordering + */ + if (ip) { + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * A dquot could be attached to this inode by now, since + * we had dropped the ilock. + */ + if (type == XFS_DQ_USER) { + if (!XFS_IS_UQUOTA_ON(mp)) { + /* inode stays locked on return */ + xfs_qm_dqdestroy(dqp); + return XFS_ERROR(ESRCH); + } + if (ip->i_udquot) { + xfs_qm_dqdestroy(dqp); + dqp = ip->i_udquot; + xfs_dqlock(dqp); + goto dqret; + } + } else { + if (!XFS_IS_OQUOTA_ON(mp)) { + /* inode stays locked on return */ + xfs_qm_dqdestroy(dqp); + return XFS_ERROR(ESRCH); + } + if (ip->i_gdquot) { + xfs_qm_dqdestroy(dqp); + dqp = ip->i_gdquot; + xfs_dqlock(dqp); + goto dqret; + } + } + } + + /* + * Hashlock comes after ilock in lock order + */ + mutex_lock(&h->qh_lock); + if (version != h->qh_version) { + xfs_dquot_t *tmpdqp; + /* + * Now, see if somebody else put the dquot in the + * hashtable before us. This can happen because we didn't + * keep the hashchain lock. We don't have to worry about + * lock order between the two dquots here since dqp isn't + * on any findable lists yet. + */ + if (xfs_qm_dqlookup(mp, id, h, &tmpdqp) == 0) { + /* + * Duplicate found. Just throw away the new dquot + * and start over. + */ + xfs_qm_dqput(tmpdqp); + mutex_unlock(&h->qh_lock); + xfs_qm_dqdestroy(dqp); + XQM_STATS_INC(xqmstats.xs_qm_dquot_dups); + goto again; + } + } + + /* + * Put the dquot at the beginning of the hash-chain and mp's list + * LOCK ORDER: hashlock, freelistlock, mplistlock, udqlock, gdqlock .. + */ + ASSERT(mutex_is_locked(&h->qh_lock)); + dqp->q_hash = h; + list_add(&dqp->q_hashlist, &h->qh_list); + h->qh_version++; + + /* + * Attach this dquot to this filesystem's list of all dquots, + * kept inside the mount structure in m_quotainfo field + */ + mutex_lock(&mp->m_quotainfo->qi_dqlist_lock); + + /* + * We return a locked dquot to the caller, with a reference taken + */ + xfs_dqlock(dqp); + dqp->q_nrefs = 1; + + list_add(&dqp->q_mplist, &mp->m_quotainfo->qi_dqlist); + mp->m_quotainfo->qi_dquots++; + mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); + mutex_unlock(&h->qh_lock); + dqret: + ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL)); + trace_xfs_dqget_miss(dqp); + *O_dqpp = dqp; + return (0); +} + + +/* + * Release a reference to the dquot (decrement ref-count) + * and unlock it. If there is a group quota attached to this + * dquot, carefully release that too without tripping over + * deadlocks'n'stuff. + */ +void +xfs_qm_dqput( + xfs_dquot_t *dqp) +{ + xfs_dquot_t *gdqp; + + ASSERT(dqp->q_nrefs > 0); + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + + trace_xfs_dqput(dqp); + + if (dqp->q_nrefs != 1) { + dqp->q_nrefs--; + xfs_dqunlock(dqp); + return; + } + + /* + * drop the dqlock and acquire the freelist and dqlock + * in the right order; but try to get it out-of-order first + */ + if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) { + trace_xfs_dqput_wait(dqp); + xfs_dqunlock(dqp); + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + xfs_dqlock(dqp); + } + + while (1) { + gdqp = NULL; + + /* We can't depend on nrefs being == 1 here */ + if (--dqp->q_nrefs == 0) { + trace_xfs_dqput_free(dqp); + + list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist); + xfs_Gqm->qm_dqfrlist_cnt++; + + /* + * If we just added a udquot to the freelist, then + * we want to release the gdquot reference that + * it (probably) has. Otherwise it'll keep the + * gdquot from getting reclaimed. + */ + if ((gdqp = dqp->q_gdquot)) { + /* + * Avoid a recursive dqput call + */ + xfs_dqlock(gdqp); + dqp->q_gdquot = NULL; + } + } + xfs_dqunlock(dqp); + + /* + * If we had a group quota inside the user quota as a hint, + * release it now. + */ + if (! gdqp) + break; + dqp = gdqp; + } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); +} + +/* + * Release a dquot. Flush it if dirty, then dqput() it. + * dquot must not be locked. + */ +void +xfs_qm_dqrele( + xfs_dquot_t *dqp) +{ + if (!dqp) + return; + + trace_xfs_dqrele(dqp); + + xfs_dqlock(dqp); + /* + * We don't care to flush it if the dquot is dirty here. + * That will create stutters that we want to avoid. + * Instead we do a delayed write when we try to reclaim + * a dirty dquot. Also xfs_sync will take part of the burden... + */ + xfs_qm_dqput(dqp); +} + +/* + * This is the dquot flushing I/O completion routine. It is called + * from interrupt level when the buffer containing the dquot is + * flushed to disk. It is responsible for removing the dquot logitem + * from the AIL if it has not been re-logged, and unlocking the dquot's + * flush lock. This behavior is very similar to that of inodes.. + */ +STATIC void +xfs_qm_dqflush_done( + struct xfs_buf *bp, + struct xfs_log_item *lip) +{ + xfs_dq_logitem_t *qip = (struct xfs_dq_logitem *)lip; + xfs_dquot_t *dqp = qip->qli_dquot; + struct xfs_ail *ailp = lip->li_ailp; + + /* + * We only want to pull the item from the AIL if its + * location in the log has not changed since we started the flush. + * Thus, we only bother if the dquot's lsn has + * not changed. First we check the lsn outside the lock + * since it's cheaper, and then we recheck while + * holding the lock before removing the dquot from the AIL. + */ + if ((lip->li_flags & XFS_LI_IN_AIL) && + lip->li_lsn == qip->qli_flush_lsn) { + + /* xfs_trans_ail_delete() drops the AIL lock. */ + spin_lock(&ailp->xa_lock); + if (lip->li_lsn == qip->qli_flush_lsn) + xfs_trans_ail_delete(ailp, lip); + else + spin_unlock(&ailp->xa_lock); + } + + /* + * Release the dq's flush lock since we're done with it. + */ + xfs_dqfunlock(dqp); +} + +/* + * Write a modified dquot to disk. + * The dquot must be locked and the flush lock too taken by caller. + * The flush lock will not be unlocked until the dquot reaches the disk, + * but the dquot is free to be unlocked and modified by the caller + * in the interim. Dquot is still locked on return. This behavior is + * identical to that of inodes. + */ +int +xfs_qm_dqflush( + xfs_dquot_t *dqp, + uint flags) +{ + struct xfs_mount *mp = dqp->q_mount; + struct xfs_buf *bp; + struct xfs_disk_dquot *ddqp; + int error; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + ASSERT(!completion_done(&dqp->q_flush)); + + trace_xfs_dqflush(dqp); + + /* + * If not dirty, or it's pinned and we are not supposed to block, nada. + */ + if (!XFS_DQ_IS_DIRTY(dqp) || + (!(flags & SYNC_WAIT) && atomic_read(&dqp->q_pincount) > 0)) { + xfs_dqfunlock(dqp); + return 0; + } + xfs_qm_dqunpin_wait(dqp); + + /* + * This may have been unpinned because the filesystem is shutting + * down forcibly. If that's the case we must not write this dquot + * to disk, because the log record didn't make it to disk! + */ + if (XFS_FORCED_SHUTDOWN(mp)) { + dqp->dq_flags &= ~XFS_DQ_DIRTY; + xfs_dqfunlock(dqp); + return XFS_ERROR(EIO); + } + + /* + * Get the buffer containing the on-disk dquot + */ + error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, 0, &bp); + if (error) { + ASSERT(error != ENOENT); + xfs_dqfunlock(dqp); + return error; + } + + /* + * Calculate the location of the dquot inside the buffer. + */ + ddqp = bp->b_addr + dqp->q_bufoffset; + + /* + * A simple sanity check in case we got a corrupted dquot.. + */ + error = xfs_qm_dqcheck(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0, + XFS_QMOPT_DOWARN, "dqflush (incore copy)"); + if (error) { + xfs_buf_relse(bp); + xfs_dqfunlock(dqp); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); + return XFS_ERROR(EIO); + } + + /* This is the only portion of data that needs to persist */ + memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t)); + + /* + * Clear the dirty field and remember the flush lsn for later use. + */ + dqp->dq_flags &= ~XFS_DQ_DIRTY; + + xfs_trans_ail_copy_lsn(mp->m_ail, &dqp->q_logitem.qli_flush_lsn, + &dqp->q_logitem.qli_item.li_lsn); + + /* + * Attach an iodone routine so that we can remove this dquot from the + * AIL and release the flush lock once the dquot is synced to disk. + */ + xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done, + &dqp->q_logitem.qli_item); + + /* + * If the buffer is pinned then push on the log so we won't + * get stuck waiting in the write for too long. + */ + if (xfs_buf_ispinned(bp)) { + trace_xfs_dqflush_force(dqp); + xfs_log_force(mp, 0); + } + + if (flags & SYNC_WAIT) + error = xfs_bwrite(mp, bp); + else + xfs_bdwrite(mp, bp); + + trace_xfs_dqflush_done(dqp); + + /* + * dqp is still locked, but caller is free to unlock it now. + */ + return error; + +} + +int +xfs_qm_dqlock_nowait( + xfs_dquot_t *dqp) +{ + return mutex_trylock(&dqp->q_qlock); +} + +void +xfs_dqlock( + xfs_dquot_t *dqp) +{ + mutex_lock(&dqp->q_qlock); +} + +void +xfs_dqunlock( + xfs_dquot_t *dqp) +{ + mutex_unlock(&(dqp->q_qlock)); + if (dqp->q_logitem.qli_dquot == dqp) { + /* Once was dqp->q_mount, but might just have been cleared */ + xfs_trans_unlocked_item(dqp->q_logitem.qli_item.li_ailp, + (xfs_log_item_t*)&(dqp->q_logitem)); + } +} + + +void +xfs_dqunlock_nonotify( + xfs_dquot_t *dqp) +{ + mutex_unlock(&(dqp->q_qlock)); +} + +/* + * Lock two xfs_dquot structures. + * + * To avoid deadlocks we always lock the quota structure with + * the lowerd id first. + */ +void +xfs_dqlock2( + xfs_dquot_t *d1, + xfs_dquot_t *d2) +{ + if (d1 && d2) { + ASSERT(d1 != d2); + if (be32_to_cpu(d1->q_core.d_id) > + be32_to_cpu(d2->q_core.d_id)) { + mutex_lock(&d2->q_qlock); + mutex_lock_nested(&d1->q_qlock, XFS_QLOCK_NESTED); + } else { + mutex_lock(&d1->q_qlock); + mutex_lock_nested(&d2->q_qlock, XFS_QLOCK_NESTED); + } + } else if (d1) { + mutex_lock(&d1->q_qlock); + } else if (d2) { + mutex_lock(&d2->q_qlock); + } +} + + +/* + * Take a dquot out of the mount's dqlist as well as the hashlist. + * This is called via unmount as well as quotaoff, and the purge + * will always succeed unless there are soft (temp) references + * outstanding. + * + * This returns 0 if it was purged, 1 if it wasn't. It's not an error code + * that we're returning! XXXsup - not cool. + */ +/* ARGSUSED */ +int +xfs_qm_dqpurge( + xfs_dquot_t *dqp) +{ + xfs_dqhash_t *qh = dqp->q_hash; + xfs_mount_t *mp = dqp->q_mount; + + ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock)); + ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock)); + + xfs_dqlock(dqp); + /* + * We really can't afford to purge a dquot that is + * referenced, because these are hard refs. + * It shouldn't happen in general because we went thru _all_ inodes in + * dqrele_all_inodes before calling this and didn't let the mountlock go. + * However it is possible that we have dquots with temporary + * references that are not attached to an inode. e.g. see xfs_setattr(). + */ + if (dqp->q_nrefs != 0) { + xfs_dqunlock(dqp); + mutex_unlock(&dqp->q_hash->qh_lock); + return (1); + } + + ASSERT(!list_empty(&dqp->q_freelist)); + + /* + * If we're turning off quotas, we have to make sure that, for + * example, we don't delete quota disk blocks while dquots are + * in the process of getting written to those disk blocks. + * This dquot might well be on AIL, and we can't leave it there + * if we're turning off quotas. Basically, we need this flush + * lock, and are willing to block on it. + */ + if (!xfs_dqflock_nowait(dqp)) { + /* + * Block on the flush lock after nudging dquot buffer, + * if it is incore. + */ + xfs_qm_dqflock_pushbuf_wait(dqp); + } + + /* + * XXXIf we're turning this type of quotas off, we don't care + * about the dirty metadata sitting in this dquot. OTOH, if + * we're unmounting, we do care, so we flush it and wait. + */ + if (XFS_DQ_IS_DIRTY(dqp)) { + int error; + + /* dqflush unlocks dqflock */ + /* + * Given that dqpurge is a very rare occurrence, it is OK + * that we're holding the hashlist and mplist locks + * across the disk write. But, ... XXXsup + * + * We don't care about getting disk errors here. We need + * to purge this dquot anyway, so we go ahead regardless. + */ + error = xfs_qm_dqflush(dqp, SYNC_WAIT); + if (error) + xfs_warn(mp, "%s: dquot %p flush failed", + __func__, dqp); + xfs_dqflock(dqp); + } + ASSERT(atomic_read(&dqp->q_pincount) == 0); + ASSERT(XFS_FORCED_SHUTDOWN(mp) || + !(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL)); + + list_del_init(&dqp->q_hashlist); + qh->qh_version++; + list_del_init(&dqp->q_mplist); + mp->m_quotainfo->qi_dqreclaims++; + mp->m_quotainfo->qi_dquots--; + /* + * XXX Move this to the front of the freelist, if we can get the + * freelist lock. + */ + ASSERT(!list_empty(&dqp->q_freelist)); + + dqp->q_mount = NULL; + dqp->q_hash = NULL; + dqp->dq_flags = XFS_DQ_INACTIVE; + memset(&dqp->q_core, 0, sizeof(dqp->q_core)); + xfs_dqfunlock(dqp); + xfs_dqunlock(dqp); + mutex_unlock(&qh->qh_lock); + return (0); +} + + +/* + * Give the buffer a little push if it is incore and + * wait on the flush lock. + */ +void +xfs_qm_dqflock_pushbuf_wait( + xfs_dquot_t *dqp) +{ + xfs_mount_t *mp = dqp->q_mount; + xfs_buf_t *bp; + + /* + * Check to see if the dquot has been flushed delayed + * write. If so, grab its buffer and send it + * out immediately. We'll be able to acquire + * the flush lock when the I/O completes. + */ + bp = xfs_incore(mp->m_ddev_targp, dqp->q_blkno, + mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK); + if (!bp) + goto out_lock; + + if (XFS_BUF_ISDELAYWRITE(bp)) { + if (xfs_buf_ispinned(bp)) + xfs_log_force(mp, 0); + xfs_buf_delwri_promote(bp); + wake_up_process(bp->b_target->bt_task); + } + xfs_buf_relse(bp); +out_lock: + xfs_dqflock(dqp); +} diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h new file mode 100644 index 000000000000..34b7e945dbfa --- /dev/null +++ b/fs/xfs/xfs_dquot.h @@ -0,0 +1,137 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_DQUOT_H__ +#define __XFS_DQUOT_H__ + +/* + * Dquots are structures that hold quota information about a user or a group, + * much like inodes are for files. In fact, dquots share many characteristics + * with inodes. However, dquots can also be a centralized resource, relative + * to a collection of inodes. In this respect, dquots share some characteristics + * of the superblock. + * XFS dquots exploit both those in its algorithms. They make every attempt + * to not be a bottleneck when quotas are on and have minimal impact, if any, + * when quotas are off. + */ + +/* + * The hash chain headers (hash buckets) + */ +typedef struct xfs_dqhash { + struct list_head qh_list; + struct mutex qh_lock; + uint qh_version; /* ever increasing version */ + uint qh_nelems; /* number of dquots on the list */ +} xfs_dqhash_t; + +struct xfs_mount; +struct xfs_trans; + +/* + * The incore dquot structure + */ +typedef struct xfs_dquot { + uint dq_flags; /* various flags (XFS_DQ_*) */ + struct list_head q_freelist; /* global free list of dquots */ + struct list_head q_mplist; /* mount's list of dquots */ + struct list_head q_hashlist; /* gloabl hash list of dquots */ + xfs_dqhash_t *q_hash; /* the hashchain header */ + struct xfs_mount*q_mount; /* filesystem this relates to */ + struct xfs_trans*q_transp; /* trans this belongs to currently */ + uint q_nrefs; /* # active refs from inodes */ + xfs_daddr_t q_blkno; /* blkno of dquot buffer */ + int q_bufoffset; /* off of dq in buffer (# dquots) */ + xfs_fileoff_t q_fileoffset; /* offset in quotas file */ + + struct xfs_dquot*q_gdquot; /* group dquot, hint only */ + xfs_disk_dquot_t q_core; /* actual usage & quotas */ + xfs_dq_logitem_t q_logitem; /* dquot log item */ + xfs_qcnt_t q_res_bcount; /* total regular nblks used+reserved */ + xfs_qcnt_t q_res_icount; /* total inos allocd+reserved */ + xfs_qcnt_t q_res_rtbcount;/* total realtime blks used+reserved */ + struct mutex q_qlock; /* quota lock */ + struct completion q_flush; /* flush completion queue */ + atomic_t q_pincount; /* dquot pin count */ + wait_queue_head_t q_pinwait; /* dquot pinning wait queue */ +} xfs_dquot_t; + +/* + * Lock hierarchy for q_qlock: + * XFS_QLOCK_NORMAL is the implicit default, + * XFS_QLOCK_NESTED is the dquot with the higher id in xfs_dqlock2 + */ +enum { + XFS_QLOCK_NORMAL = 0, + XFS_QLOCK_NESTED, +}; + +#define XFS_DQHOLD(dqp) ((dqp)->q_nrefs++) + +/* + * Manage the q_flush completion queue embedded in the dquot. This completion + * queue synchronizes processes attempting to flush the in-core dquot back to + * disk. + */ +static inline void xfs_dqflock(xfs_dquot_t *dqp) +{ + wait_for_completion(&dqp->q_flush); +} + +static inline int xfs_dqflock_nowait(xfs_dquot_t *dqp) +{ + return try_wait_for_completion(&dqp->q_flush); +} + +static inline void xfs_dqfunlock(xfs_dquot_t *dqp) +{ + complete(&dqp->q_flush); +} + +#define XFS_DQ_IS_LOCKED(dqp) (mutex_is_locked(&((dqp)->q_qlock))) +#define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY) +#define XFS_QM_ISUDQ(dqp) ((dqp)->dq_flags & XFS_DQ_USER) +#define XFS_QM_ISPDQ(dqp) ((dqp)->dq_flags & XFS_DQ_PROJ) +#define XFS_QM_ISGDQ(dqp) ((dqp)->dq_flags & XFS_DQ_GROUP) +#define XFS_DQ_TO_QINF(dqp) ((dqp)->q_mount->m_quotainfo) +#define XFS_DQ_TO_QIP(dqp) (XFS_QM_ISUDQ(dqp) ? \ + XFS_DQ_TO_QINF(dqp)->qi_uquotaip : \ + XFS_DQ_TO_QINF(dqp)->qi_gquotaip) + +#define XFS_IS_THIS_QUOTA_OFF(d) (! (XFS_QM_ISUDQ(d) ? \ + (XFS_IS_UQUOTA_ON((d)->q_mount)) : \ + (XFS_IS_OQUOTA_ON((d)->q_mount)))) + +extern void xfs_qm_dqdestroy(xfs_dquot_t *); +extern int xfs_qm_dqflush(xfs_dquot_t *, uint); +extern int xfs_qm_dqpurge(xfs_dquot_t *); +extern void xfs_qm_dqunpin_wait(xfs_dquot_t *); +extern int xfs_qm_dqlock_nowait(xfs_dquot_t *); +extern void xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp); +extern void xfs_qm_adjust_dqtimers(xfs_mount_t *, + xfs_disk_dquot_t *); +extern void xfs_qm_adjust_dqlimits(xfs_mount_t *, + xfs_disk_dquot_t *); +extern int xfs_qm_dqget(xfs_mount_t *, xfs_inode_t *, + xfs_dqid_t, uint, uint, xfs_dquot_t **); +extern void xfs_qm_dqput(xfs_dquot_t *); +extern void xfs_dqlock(xfs_dquot_t *); +extern void xfs_dqlock2(xfs_dquot_t *, xfs_dquot_t *); +extern void xfs_dqunlock(xfs_dquot_t *); +extern void xfs_dqunlock_nonotify(xfs_dquot_t *); + +#endif /* __XFS_DQUOT_H__ */ diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c new file mode 100644 index 000000000000..9e0e2fa3f2c8 --- /dev/null +++ b/fs/xfs/xfs_dquot_item.c @@ -0,0 +1,529 @@ +/* + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_itable.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_trans_priv.h" +#include "xfs_qm.h" + +static inline struct xfs_dq_logitem *DQUOT_ITEM(struct xfs_log_item *lip) +{ + return container_of(lip, struct xfs_dq_logitem, qli_item); +} + +/* + * returns the number of iovecs needed to log the given dquot item. + */ +STATIC uint +xfs_qm_dquot_logitem_size( + struct xfs_log_item *lip) +{ + /* + * we need only two iovecs, one for the format, one for the real thing + */ + return 2; +} + +/* + * fills in the vector of log iovecs for the given dquot log item. + */ +STATIC void +xfs_qm_dquot_logitem_format( + struct xfs_log_item *lip, + struct xfs_log_iovec *logvec) +{ + struct xfs_dq_logitem *qlip = DQUOT_ITEM(lip); + + logvec->i_addr = &qlip->qli_format; + logvec->i_len = sizeof(xfs_dq_logformat_t); + logvec->i_type = XLOG_REG_TYPE_QFORMAT; + logvec++; + logvec->i_addr = &qlip->qli_dquot->q_core; + logvec->i_len = sizeof(xfs_disk_dquot_t); + logvec->i_type = XLOG_REG_TYPE_DQUOT; + + ASSERT(2 == lip->li_desc->lid_size); + qlip->qli_format.qlf_size = 2; + +} + +/* + * Increment the pin count of the given dquot. + */ +STATIC void +xfs_qm_dquot_logitem_pin( + struct xfs_log_item *lip) +{ + struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + atomic_inc(&dqp->q_pincount); +} + +/* + * Decrement the pin count of the given dquot, and wake up + * anyone in xfs_dqwait_unpin() if the count goes to 0. The + * dquot must have been previously pinned with a call to + * xfs_qm_dquot_logitem_pin(). + */ +STATIC void +xfs_qm_dquot_logitem_unpin( + struct xfs_log_item *lip, + int remove) +{ + struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; + + ASSERT(atomic_read(&dqp->q_pincount) > 0); + if (atomic_dec_and_test(&dqp->q_pincount)) + wake_up(&dqp->q_pinwait); +} + +/* + * Given the logitem, this writes the corresponding dquot entry to disk + * asynchronously. This is called with the dquot entry securely locked; + * we simply get xfs_qm_dqflush() to do the work, and unlock the dquot + * at the end. + */ +STATIC void +xfs_qm_dquot_logitem_push( + struct xfs_log_item *lip) +{ + struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; + int error; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + ASSERT(!completion_done(&dqp->q_flush)); + + /* + * Since we were able to lock the dquot's flush lock and + * we found it on the AIL, the dquot must be dirty. This + * is because the dquot is removed from the AIL while still + * holding the flush lock in xfs_dqflush_done(). Thus, if + * we found it in the AIL and were able to obtain the flush + * lock without sleeping, then there must not have been + * anyone in the process of flushing the dquot. + */ + error = xfs_qm_dqflush(dqp, 0); + if (error) + xfs_warn(dqp->q_mount, "%s: push error %d on dqp %p", + __func__, error, dqp); + xfs_dqunlock(dqp); +} + +STATIC xfs_lsn_t +xfs_qm_dquot_logitem_committed( + struct xfs_log_item *lip, + xfs_lsn_t lsn) +{ + /* + * We always re-log the entire dquot when it becomes dirty, + * so, the latest copy _is_ the only one that matters. + */ + return lsn; +} + +/* + * This is called to wait for the given dquot to be unpinned. + * Most of these pin/unpin routines are plagiarized from inode code. + */ +void +xfs_qm_dqunpin_wait( + struct xfs_dquot *dqp) +{ + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + if (atomic_read(&dqp->q_pincount) == 0) + return; + + /* + * Give the log a push so we don't wait here too long. + */ + xfs_log_force(dqp->q_mount, 0); + wait_event(dqp->q_pinwait, (atomic_read(&dqp->q_pincount) == 0)); +} + +/* + * This is called when IOP_TRYLOCK returns XFS_ITEM_PUSHBUF to indicate that + * the dquot is locked by us, but the flush lock isn't. So, here we are + * going to see if the relevant dquot buffer is incore, waiting on DELWRI. + * If so, we want to push it out to help us take this item off the AIL as soon + * as possible. + * + * We must not be holding the AIL lock at this point. Calling incore() to + * search the buffer cache can be a time consuming thing, and AIL lock is a + * spinlock. + */ +STATIC void +xfs_qm_dquot_logitem_pushbuf( + struct xfs_log_item *lip) +{ + struct xfs_dq_logitem *qlip = DQUOT_ITEM(lip); + struct xfs_dquot *dqp = qlip->qli_dquot; + struct xfs_buf *bp; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + + /* + * If flushlock isn't locked anymore, chances are that the + * inode flush completed and the inode was taken off the AIL. + * So, just get out. + */ + if (completion_done(&dqp->q_flush) || + !(lip->li_flags & XFS_LI_IN_AIL)) { + xfs_dqunlock(dqp); + return; + } + + bp = xfs_incore(dqp->q_mount->m_ddev_targp, qlip->qli_format.qlf_blkno, + dqp->q_mount->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK); + xfs_dqunlock(dqp); + if (!bp) + return; + if (XFS_BUF_ISDELAYWRITE(bp)) + xfs_buf_delwri_promote(bp); + xfs_buf_relse(bp); +} + +/* + * This is called to attempt to lock the dquot associated with this + * dquot log item. Don't sleep on the dquot lock or the flush lock. + * If the flush lock is already held, indicating that the dquot has + * been or is in the process of being flushed, then see if we can + * find the dquot's buffer in the buffer cache without sleeping. If + * we can and it is marked delayed write, then we want to send it out. + * We delay doing so until the push routine, though, to avoid sleeping + * in any device strategy routines. + */ +STATIC uint +xfs_qm_dquot_logitem_trylock( + struct xfs_log_item *lip) +{ + struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; + + if (atomic_read(&dqp->q_pincount) > 0) + return XFS_ITEM_PINNED; + + if (!xfs_qm_dqlock_nowait(dqp)) + return XFS_ITEM_LOCKED; + + if (!xfs_dqflock_nowait(dqp)) { + /* + * dquot has already been flushed to the backing buffer, + * leave it locked, pushbuf routine will unlock it. + */ + return XFS_ITEM_PUSHBUF; + } + + ASSERT(lip->li_flags & XFS_LI_IN_AIL); + return XFS_ITEM_SUCCESS; +} + +/* + * Unlock the dquot associated with the log item. + * Clear the fields of the dquot and dquot log item that + * are specific to the current transaction. If the + * hold flags is set, do not unlock the dquot. + */ +STATIC void +xfs_qm_dquot_logitem_unlock( + struct xfs_log_item *lip) +{ + struct xfs_dquot *dqp = DQUOT_ITEM(lip)->qli_dquot; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + + /* + * Clear the transaction pointer in the dquot + */ + dqp->q_transp = NULL; + + /* + * dquots are never 'held' from getting unlocked at the end of + * a transaction. Their locking and unlocking is hidden inside the + * transaction layer, within trans_commit. Hence, no LI_HOLD flag + * for the logitem. + */ + xfs_dqunlock(dqp); +} + +/* + * this needs to stamp an lsn into the dquot, I think. + * rpc's that look at user dquot's would then have to + * push on the dependency recorded in the dquot + */ +STATIC void +xfs_qm_dquot_logitem_committing( + struct xfs_log_item *lip, + xfs_lsn_t lsn) +{ +} + +/* + * This is the ops vector for dquots + */ +static struct xfs_item_ops xfs_dquot_item_ops = { + .iop_size = xfs_qm_dquot_logitem_size, + .iop_format = xfs_qm_dquot_logitem_format, + .iop_pin = xfs_qm_dquot_logitem_pin, + .iop_unpin = xfs_qm_dquot_logitem_unpin, + .iop_trylock = xfs_qm_dquot_logitem_trylock, + .iop_unlock = xfs_qm_dquot_logitem_unlock, + .iop_committed = xfs_qm_dquot_logitem_committed, + .iop_push = xfs_qm_dquot_logitem_push, + .iop_pushbuf = xfs_qm_dquot_logitem_pushbuf, + .iop_committing = xfs_qm_dquot_logitem_committing +}; + +/* + * Initialize the dquot log item for a newly allocated dquot. + * The dquot isn't locked at this point, but it isn't on any of the lists + * either, so we don't care. + */ +void +xfs_qm_dquot_logitem_init( + struct xfs_dquot *dqp) +{ + struct xfs_dq_logitem *lp = &dqp->q_logitem; + + xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT, + &xfs_dquot_item_ops); + lp->qli_dquot = dqp; + lp->qli_format.qlf_type = XFS_LI_DQUOT; + lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id); + lp->qli_format.qlf_blkno = dqp->q_blkno; + lp->qli_format.qlf_len = 1; + /* + * This is just the offset of this dquot within its buffer + * (which is currently 1 FSB and probably won't change). + * Hence 32 bits for this offset should be just fine. + * Alternatively, we can store (bufoffset / sizeof(xfs_dqblk_t)) + * here, and recompute it at recovery time. + */ + lp->qli_format.qlf_boffset = (__uint32_t)dqp->q_bufoffset; +} + +/*------------------ QUOTAOFF LOG ITEMS -------------------*/ + +static inline struct xfs_qoff_logitem *QOFF_ITEM(struct xfs_log_item *lip) +{ + return container_of(lip, struct xfs_qoff_logitem, qql_item); +} + + +/* + * This returns the number of iovecs needed to log the given quotaoff item. + * We only need 1 iovec for an quotaoff item. It just logs the + * quotaoff_log_format structure. + */ +STATIC uint +xfs_qm_qoff_logitem_size( + struct xfs_log_item *lip) +{ + return 1; +} + +/* + * This is called to fill in the vector of log iovecs for the + * given quotaoff log item. We use only 1 iovec, and we point that + * at the quotaoff_log_format structure embedded in the quotaoff item. + * It is at this point that we assert that all of the extent + * slots in the quotaoff item have been filled. + */ +STATIC void +xfs_qm_qoff_logitem_format( + struct xfs_log_item *lip, + struct xfs_log_iovec *log_vector) +{ + struct xfs_qoff_logitem *qflip = QOFF_ITEM(lip); + + ASSERT(qflip->qql_format.qf_type == XFS_LI_QUOTAOFF); + + log_vector->i_addr = &qflip->qql_format; + log_vector->i_len = sizeof(xfs_qoff_logitem_t); + log_vector->i_type = XLOG_REG_TYPE_QUOTAOFF; + qflip->qql_format.qf_size = 1; +} + +/* + * Pinning has no meaning for an quotaoff item, so just return. + */ +STATIC void +xfs_qm_qoff_logitem_pin( + struct xfs_log_item *lip) +{ +} + +/* + * Since pinning has no meaning for an quotaoff item, unpinning does + * not either. + */ +STATIC void +xfs_qm_qoff_logitem_unpin( + struct xfs_log_item *lip, + int remove) +{ +} + +/* + * Quotaoff items have no locking, so just return success. + */ +STATIC uint +xfs_qm_qoff_logitem_trylock( + struct xfs_log_item *lip) +{ + return XFS_ITEM_LOCKED; +} + +/* + * Quotaoff items have no locking or pushing, so return failure + * so that the caller doesn't bother with us. + */ +STATIC void +xfs_qm_qoff_logitem_unlock( + struct xfs_log_item *lip) +{ +} + +/* + * The quotaoff-start-item is logged only once and cannot be moved in the log, + * so simply return the lsn at which it's been logged. + */ +STATIC xfs_lsn_t +xfs_qm_qoff_logitem_committed( + struct xfs_log_item *lip, + xfs_lsn_t lsn) +{ + return lsn; +} + +/* + * There isn't much you can do to push on an quotaoff item. It is simply + * stuck waiting for the log to be flushed to disk. + */ +STATIC void +xfs_qm_qoff_logitem_push( + struct xfs_log_item *lip) +{ +} + + +STATIC xfs_lsn_t +xfs_qm_qoffend_logitem_committed( + struct xfs_log_item *lip, + xfs_lsn_t lsn) +{ + struct xfs_qoff_logitem *qfe = QOFF_ITEM(lip); + struct xfs_qoff_logitem *qfs = qfe->qql_start_lip; + struct xfs_ail *ailp = qfs->qql_item.li_ailp; + + /* + * Delete the qoff-start logitem from the AIL. + * xfs_trans_ail_delete() drops the AIL lock. + */ + spin_lock(&ailp->xa_lock); + xfs_trans_ail_delete(ailp, (xfs_log_item_t *)qfs); + + kmem_free(qfs); + kmem_free(qfe); + return (xfs_lsn_t)-1; +} + +/* + * XXX rcc - don't know quite what to do with this. I think we can + * just ignore it. The only time that isn't the case is if we allow + * the client to somehow see that quotas have been turned off in which + * we can't allow that to get back until the quotaoff hits the disk. + * So how would that happen? Also, do we need different routines for + * quotaoff start and quotaoff end? I suspect the answer is yes but + * to be sure, I need to look at the recovery code and see how quota off + * recovery is handled (do we roll forward or back or do something else). + * If we roll forwards or backwards, then we need two separate routines, + * one that does nothing and one that stamps in the lsn that matters + * (truly makes the quotaoff irrevocable). If we do something else, + * then maybe we don't need two. + */ +STATIC void +xfs_qm_qoff_logitem_committing( + struct xfs_log_item *lip, + xfs_lsn_t commit_lsn) +{ +} + +static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = { + .iop_size = xfs_qm_qoff_logitem_size, + .iop_format = xfs_qm_qoff_logitem_format, + .iop_pin = xfs_qm_qoff_logitem_pin, + .iop_unpin = xfs_qm_qoff_logitem_unpin, + .iop_trylock = xfs_qm_qoff_logitem_trylock, + .iop_unlock = xfs_qm_qoff_logitem_unlock, + .iop_committed = xfs_qm_qoffend_logitem_committed, + .iop_push = xfs_qm_qoff_logitem_push, + .iop_committing = xfs_qm_qoff_logitem_committing +}; + +/* + * This is the ops vector shared by all quotaoff-start log items. + */ +static struct xfs_item_ops xfs_qm_qoff_logitem_ops = { + .iop_size = xfs_qm_qoff_logitem_size, + .iop_format = xfs_qm_qoff_logitem_format, + .iop_pin = xfs_qm_qoff_logitem_pin, + .iop_unpin = xfs_qm_qoff_logitem_unpin, + .iop_trylock = xfs_qm_qoff_logitem_trylock, + .iop_unlock = xfs_qm_qoff_logitem_unlock, + .iop_committed = xfs_qm_qoff_logitem_committed, + .iop_push = xfs_qm_qoff_logitem_push, + .iop_committing = xfs_qm_qoff_logitem_committing +}; + +/* + * Allocate and initialize an quotaoff item of the correct quota type(s). + */ +struct xfs_qoff_logitem * +xfs_qm_qoff_logitem_init( + struct xfs_mount *mp, + struct xfs_qoff_logitem *start, + uint flags) +{ + struct xfs_qoff_logitem *qf; + + qf = kmem_zalloc(sizeof(struct xfs_qoff_logitem), KM_SLEEP); + + xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ? + &xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops); + qf->qql_item.li_mountp = mp; + qf->qql_format.qf_type = XFS_LI_QUOTAOFF; + qf->qql_format.qf_flags = flags; + qf->qql_start_lip = start; + return qf; +} diff --git a/fs/xfs/xfs_dquot_item.h b/fs/xfs/xfs_dquot_item.h new file mode 100644 index 000000000000..5acae2ada70b --- /dev/null +++ b/fs/xfs/xfs_dquot_item.h @@ -0,0 +1,48 @@ +/* + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_DQUOT_ITEM_H__ +#define __XFS_DQUOT_ITEM_H__ + +struct xfs_dquot; +struct xfs_trans; +struct xfs_mount; +struct xfs_qoff_logitem; + +typedef struct xfs_dq_logitem { + xfs_log_item_t qli_item; /* common portion */ + struct xfs_dquot *qli_dquot; /* dquot ptr */ + xfs_lsn_t qli_flush_lsn; /* lsn at last flush */ + xfs_dq_logformat_t qli_format; /* logged structure */ +} xfs_dq_logitem_t; + +typedef struct xfs_qoff_logitem { + xfs_log_item_t qql_item; /* common portion */ + struct xfs_qoff_logitem *qql_start_lip; /* qoff-start logitem, if any */ + xfs_qoff_logformat_t qql_format; /* logged structure */ +} xfs_qoff_logitem_t; + + +extern void xfs_qm_dquot_logitem_init(struct xfs_dquot *); +extern xfs_qoff_logitem_t *xfs_qm_qoff_logitem_init(struct xfs_mount *, + struct xfs_qoff_logitem *, uint); +extern xfs_qoff_logitem_t *xfs_trans_get_qoff_item(struct xfs_trans *, + struct xfs_qoff_logitem *, uint); +extern void xfs_trans_log_quotaoff_item(struct xfs_trans *, + struct xfs_qoff_logitem *); + +#endif /* __XFS_DQUOT_ITEM_H__ */ diff --git a/fs/xfs/xfs_export.c b/fs/xfs/xfs_export.c new file mode 100644 index 000000000000..75e5d322e48f --- /dev/null +++ b/fs/xfs/xfs_export.c @@ -0,0 +1,250 @@ +/* + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_types.h" +#include "xfs_inum.h" +#include "xfs_log.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_dir2.h" +#include "xfs_mount.h" +#include "xfs_export.h" +#include "xfs_vnodeops.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_trace.h" + +/* + * Note that we only accept fileids which are long enough rather than allow + * the parent generation number to default to zero. XFS considers zero a + * valid generation number not an invalid/wildcard value. + */ +static int xfs_fileid_length(int fileid_type) +{ + switch (fileid_type) { + case FILEID_INO32_GEN: + return 2; + case FILEID_INO32_GEN_PARENT: + return 4; + case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: + return 3; + case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: + return 6; + } + return 255; /* invalid */ +} + +STATIC int +xfs_fs_encode_fh( + struct dentry *dentry, + __u32 *fh, + int *max_len, + int connectable) +{ + struct fid *fid = (struct fid *)fh; + struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fh; + struct inode *inode = dentry->d_inode; + int fileid_type; + int len; + + /* Directories don't need their parent encoded, they have ".." */ + if (S_ISDIR(inode->i_mode) || !connectable) + fileid_type = FILEID_INO32_GEN; + else + fileid_type = FILEID_INO32_GEN_PARENT; + + /* + * If the the filesystem may contain 64bit inode numbers, we need + * to use larger file handles that can represent them. + * + * While we only allocate inodes that do not fit into 32 bits any + * large enough filesystem may contain them, thus the slightly + * confusing looking conditional below. + */ + if (!(XFS_M(inode->i_sb)->m_flags & XFS_MOUNT_SMALL_INUMS) || + (XFS_M(inode->i_sb)->m_flags & XFS_MOUNT_32BITINODES)) + fileid_type |= XFS_FILEID_TYPE_64FLAG; + + /* + * Only encode if there is enough space given. In practice + * this means we can't export a filesystem with 64bit inodes + * over NFSv2 with the subtree_check export option; the other + * seven combinations work. The real answer is "don't use v2". + */ + len = xfs_fileid_length(fileid_type); + if (*max_len < len) { + *max_len = len; + return 255; + } + *max_len = len; + + switch (fileid_type) { + case FILEID_INO32_GEN_PARENT: + spin_lock(&dentry->d_lock); + fid->i32.parent_ino = dentry->d_parent->d_inode->i_ino; + fid->i32.parent_gen = dentry->d_parent->d_inode->i_generation; + spin_unlock(&dentry->d_lock); + /*FALLTHRU*/ + case FILEID_INO32_GEN: + fid->i32.ino = inode->i_ino; + fid->i32.gen = inode->i_generation; + break; + case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: + spin_lock(&dentry->d_lock); + fid64->parent_ino = dentry->d_parent->d_inode->i_ino; + fid64->parent_gen = dentry->d_parent->d_inode->i_generation; + spin_unlock(&dentry->d_lock); + /*FALLTHRU*/ + case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: + fid64->ino = inode->i_ino; + fid64->gen = inode->i_generation; + break; + } + + return fileid_type; +} + +STATIC struct inode * +xfs_nfs_get_inode( + struct super_block *sb, + u64 ino, + u32 generation) + { + xfs_mount_t *mp = XFS_M(sb); + xfs_inode_t *ip; + int error; + + /* + * NFS can sometimes send requests for ino 0. Fail them gracefully. + */ + if (ino == 0) + return ERR_PTR(-ESTALE); + + /* + * The XFS_IGET_UNTRUSTED means that an invalid inode number is just + * fine and not an indication of a corrupted filesystem as clients can + * send invalid file handles and we have to handle it gracefully.. + */ + error = xfs_iget(mp, NULL, ino, XFS_IGET_UNTRUSTED, 0, &ip); + if (error) { + /* + * EINVAL means the inode cluster doesn't exist anymore. + * This implies the filehandle is stale, so we should + * translate it here. + * We don't use ESTALE directly down the chain to not + * confuse applications using bulkstat that expect EINVAL. + */ + if (error == EINVAL || error == ENOENT) + error = ESTALE; + return ERR_PTR(-error); + } + + if (ip->i_d.di_gen != generation) { + IRELE(ip); + return ERR_PTR(-ESTALE); + } + + return VFS_I(ip); +} + +STATIC struct dentry * +xfs_fs_fh_to_dentry(struct super_block *sb, struct fid *fid, + int fh_len, int fileid_type) +{ + struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fid; + struct inode *inode = NULL; + + if (fh_len < xfs_fileid_length(fileid_type)) + return NULL; + + switch (fileid_type) { + case FILEID_INO32_GEN_PARENT: + case FILEID_INO32_GEN: + inode = xfs_nfs_get_inode(sb, fid->i32.ino, fid->i32.gen); + break; + case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: + case FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG: + inode = xfs_nfs_get_inode(sb, fid64->ino, fid64->gen); + break; + } + + return d_obtain_alias(inode); +} + +STATIC struct dentry * +xfs_fs_fh_to_parent(struct super_block *sb, struct fid *fid, + int fh_len, int fileid_type) +{ + struct xfs_fid64 *fid64 = (struct xfs_fid64 *)fid; + struct inode *inode = NULL; + + switch (fileid_type) { + case FILEID_INO32_GEN_PARENT: + inode = xfs_nfs_get_inode(sb, fid->i32.parent_ino, + fid->i32.parent_gen); + break; + case FILEID_INO32_GEN_PARENT | XFS_FILEID_TYPE_64FLAG: + inode = xfs_nfs_get_inode(sb, fid64->parent_ino, + fid64->parent_gen); + break; + } + + return d_obtain_alias(inode); +} + +STATIC struct dentry * +xfs_fs_get_parent( + struct dentry *child) +{ + int error; + struct xfs_inode *cip; + + error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL); + if (unlikely(error)) + return ERR_PTR(-error); + + return d_obtain_alias(VFS_I(cip)); +} + +STATIC int +xfs_fs_nfs_commit_metadata( + struct inode *inode) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + int error = 0; + + xfs_ilock(ip, XFS_ILOCK_SHARED); + if (xfs_ipincount(ip)) { + error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn, + XFS_LOG_SYNC, NULL); + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + return error; +} + +const struct export_operations xfs_export_operations = { + .encode_fh = xfs_fs_encode_fh, + .fh_to_dentry = xfs_fs_fh_to_dentry, + .fh_to_parent = xfs_fs_fh_to_parent, + .get_parent = xfs_fs_get_parent, + .commit_metadata = xfs_fs_nfs_commit_metadata, +}; diff --git a/fs/xfs/xfs_export.h b/fs/xfs/xfs_export.h new file mode 100644 index 000000000000..3272b6ae7a35 --- /dev/null +++ b/fs/xfs/xfs_export.h @@ -0,0 +1,72 @@ +/* + * Copyright (c) 2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_EXPORT_H__ +#define __XFS_EXPORT_H__ + +/* + * Common defines for code related to exporting XFS filesystems over NFS. + * + * The NFS fileid goes out on the wire as an array of + * 32bit unsigned ints in host order. There are 5 possible + * formats. + * + * (1) fileid_type=0x00 + * (no fileid data; handled by the generic code) + * + * (2) fileid_type=0x01 + * inode-num + * generation + * + * (3) fileid_type=0x02 + * inode-num + * generation + * parent-inode-num + * parent-generation + * + * (4) fileid_type=0x81 + * inode-num-lo32 + * inode-num-hi32 + * generation + * + * (5) fileid_type=0x82 + * inode-num-lo32 + * inode-num-hi32 + * generation + * parent-inode-num-lo32 + * parent-inode-num-hi32 + * parent-generation + * + * Note, the NFS filehandle also includes an fsid portion which + * may have an inode number in it. That number is hardcoded to + * 32bits and there is no way for XFS to intercept it. In + * practice this means when exporting an XFS filesystem with 64bit + * inodes you should either export the mountpoint (rather than + * a subdirectory) or use the "fsid" export option. + */ + +struct xfs_fid64 { + u64 ino; + u32 gen; + u64 parent_ino; + u32 parent_gen; +} __attribute__((packed)); + +/* This flag goes on the wire. Don't play with it. */ +#define XFS_FILEID_TYPE_64FLAG 0x80 /* NFS fileid has 64bit inodes */ + +#endif /* __XFS_EXPORT_H__ */ diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c new file mode 100644 index 000000000000..7f7b42469ea7 --- /dev/null +++ b/fs/xfs/xfs_file.c @@ -0,0 +1,1096 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_trans.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_bmap.h" +#include "xfs_error.h" +#include "xfs_vnodeops.h" +#include "xfs_da_btree.h" +#include "xfs_ioctl.h" +#include "xfs_trace.h" + +#include +#include + +static const struct vm_operations_struct xfs_file_vm_ops; + +/* + * Locking primitives for read and write IO paths to ensure we consistently use + * and order the inode->i_mutex, ip->i_lock and ip->i_iolock. + */ +static inline void +xfs_rw_ilock( + struct xfs_inode *ip, + int type) +{ + if (type & XFS_IOLOCK_EXCL) + mutex_lock(&VFS_I(ip)->i_mutex); + xfs_ilock(ip, type); +} + +static inline void +xfs_rw_iunlock( + struct xfs_inode *ip, + int type) +{ + xfs_iunlock(ip, type); + if (type & XFS_IOLOCK_EXCL) + mutex_unlock(&VFS_I(ip)->i_mutex); +} + +static inline void +xfs_rw_ilock_demote( + struct xfs_inode *ip, + int type) +{ + xfs_ilock_demote(ip, type); + if (type & XFS_IOLOCK_EXCL) + mutex_unlock(&VFS_I(ip)->i_mutex); +} + +/* + * xfs_iozero + * + * xfs_iozero clears the specified range of buffer supplied, + * and marks all the affected blocks as valid and modified. If + * an affected block is not allocated, it will be allocated. If + * an affected block is not completely overwritten, and is not + * valid before the operation, it will be read from disk before + * being partially zeroed. + */ +STATIC int +xfs_iozero( + struct xfs_inode *ip, /* inode */ + loff_t pos, /* offset in file */ + size_t count) /* size of data to zero */ +{ + struct page *page; + struct address_space *mapping; + int status; + + mapping = VFS_I(ip)->i_mapping; + do { + unsigned offset, bytes; + void *fsdata; + + offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ + bytes = PAGE_CACHE_SIZE - offset; + if (bytes > count) + bytes = count; + + status = pagecache_write_begin(NULL, mapping, pos, bytes, + AOP_FLAG_UNINTERRUPTIBLE, + &page, &fsdata); + if (status) + break; + + zero_user(page, offset, bytes); + + status = pagecache_write_end(NULL, mapping, pos, bytes, bytes, + page, fsdata); + WARN_ON(status <= 0); /* can't return less than zero! */ + pos += bytes; + count -= bytes; + status = 0; + } while (count); + + return (-status); +} + +STATIC int +xfs_file_fsync( + struct file *file, + loff_t start, + loff_t end, + int datasync) +{ + struct inode *inode = file->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error = 0; + int log_flushed = 0; + + trace_xfs_file_fsync(ip); + + error = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (error) + return error; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -XFS_ERROR(EIO); + + xfs_iflags_clear(ip, XFS_ITRUNCATED); + + xfs_ilock(ip, XFS_IOLOCK_SHARED); + xfs_ioend_wait(ip); + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + + if (mp->m_flags & XFS_MOUNT_BARRIER) { + /* + * If we have an RT and/or log subvolume we need to make sure + * to flush the write cache the device used for file data + * first. This is to ensure newly written file data make + * it to disk before logging the new inode size in case of + * an extending write. + */ + if (XFS_IS_REALTIME_INODE(ip)) + xfs_blkdev_issue_flush(mp->m_rtdev_targp); + else if (mp->m_logdev_targp != mp->m_ddev_targp) + xfs_blkdev_issue_flush(mp->m_ddev_targp); + } + + /* + * We always need to make sure that the required inode state is safe on + * disk. The inode might be clean but we still might need to force the + * log because of committed transactions that haven't hit the disk yet. + * Likewise, there could be unflushed non-transactional changes to the + * inode core that have to go to disk and this requires us to issue + * a synchronous transaction to capture these changes correctly. + * + * This code relies on the assumption that if the i_update_core field + * of the inode is clear and the inode is unpinned then it is clean + * and no action is required. + */ + xfs_ilock(ip, XFS_ILOCK_SHARED); + + /* + * First check if the VFS inode is marked dirty. All the dirtying + * of non-transactional updates no goes through mark_inode_dirty*, + * which allows us to distinguish beteeen pure timestamp updates + * and i_size updates which need to be caught for fdatasync. + * After that also theck for the dirty state in the XFS inode, which + * might gets cleared when the inode gets written out via the AIL + * or xfs_iflush_cluster. + */ + if (((inode->i_state & I_DIRTY_DATASYNC) || + ((inode->i_state & I_DIRTY_SYNC) && !datasync)) && + ip->i_update_core) { + /* + * Kick off a transaction to log the inode core to get the + * updates. The sync transaction will also force the log. + */ + xfs_iunlock(ip, XFS_ILOCK_SHARED); + tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); + error = xfs_trans_reserve(tp, 0, + XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + if (error) { + xfs_trans_cancel(tp, 0); + return -error; + } + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * Note - it's possible that we might have pushed ourselves out + * of the way during trans_reserve which would flush the inode. + * But there's no guarantee that the inode buffer has actually + * gone out yet (it's delwri). Plus the buffer could be pinned + * anyway if it's part of an inode in another recent + * transaction. So we play it safe and fire off the + * transaction anyway. + */ + xfs_trans_ijoin(tp, ip); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + xfs_trans_set_sync(tp); + error = _xfs_trans_commit(tp, 0, &log_flushed); + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + } else { + /* + * Timestamps/size haven't changed since last inode flush or + * inode transaction commit. That means either nothing got + * written or a transaction committed which caught the updates. + * If the latter happened and the transaction hasn't hit the + * disk yet, the inode will be still be pinned. If it is, + * force the log. + */ + if (xfs_ipincount(ip)) { + error = _xfs_log_force_lsn(mp, + ip->i_itemp->ili_last_lsn, + XFS_LOG_SYNC, &log_flushed); + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + } + + /* + * If we only have a single device, and the log force about was + * a no-op we might have to flush the data device cache here. + * This can only happen for fdatasync/O_DSYNC if we were overwriting + * an already allocated file and thus do not have any metadata to + * commit. + */ + if ((mp->m_flags & XFS_MOUNT_BARRIER) && + mp->m_logdev_targp == mp->m_ddev_targp && + !XFS_IS_REALTIME_INODE(ip) && + !log_flushed) + xfs_blkdev_issue_flush(mp->m_ddev_targp); + + return -error; +} + +STATIC ssize_t +xfs_file_aio_read( + struct kiocb *iocb, + const struct iovec *iovp, + unsigned long nr_segs, + loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + size_t size = 0; + ssize_t ret = 0; + int ioflags = 0; + xfs_fsize_t n; + unsigned long seg; + + XFS_STATS_INC(xs_read_calls); + + BUG_ON(iocb->ki_pos != pos); + + if (unlikely(file->f_flags & O_DIRECT)) + ioflags |= IO_ISDIRECT; + if (file->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + /* START copy & waste from filemap.c */ + for (seg = 0; seg < nr_segs; seg++) { + const struct iovec *iv = &iovp[seg]; + + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + size += iv->iov_len; + if (unlikely((ssize_t)(size|iv->iov_len) < 0)) + return XFS_ERROR(-EINVAL); + } + /* END copy & waste from filemap.c */ + + if (unlikely(ioflags & IO_ISDIRECT)) { + xfs_buftarg_t *target = + XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp; + if ((iocb->ki_pos & target->bt_smask) || + (size & target->bt_smask)) { + if (iocb->ki_pos == ip->i_size) + return 0; + return -XFS_ERROR(EINVAL); + } + } + + n = XFS_MAXIOFFSET(mp) - iocb->ki_pos; + if (n <= 0 || size == 0) + return 0; + + if (n < size) + size = n; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -EIO; + + if (unlikely(ioflags & IO_ISDIRECT)) { + xfs_rw_ilock(ip, XFS_IOLOCK_EXCL); + + if (inode->i_mapping->nrpages) { + ret = -xfs_flushinval_pages(ip, + (iocb->ki_pos & PAGE_CACHE_MASK), + -1, FI_REMAPF_LOCKED); + if (ret) { + xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL); + return ret; + } + } + xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL); + } else + xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); + + trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags); + + ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos); + if (ret > 0) + XFS_STATS_ADD(xs_read_bytes, ret); + + xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED); + return ret; +} + +STATIC ssize_t +xfs_file_splice_read( + struct file *infilp, + loff_t *ppos, + struct pipe_inode_info *pipe, + size_t count, + unsigned int flags) +{ + struct xfs_inode *ip = XFS_I(infilp->f_mapping->host); + int ioflags = 0; + ssize_t ret; + + XFS_STATS_INC(xs_read_calls); + + if (infilp->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -EIO; + + xfs_rw_ilock(ip, XFS_IOLOCK_SHARED); + + trace_xfs_file_splice_read(ip, count, *ppos, ioflags); + + ret = generic_file_splice_read(infilp, ppos, pipe, count, flags); + if (ret > 0) + XFS_STATS_ADD(xs_read_bytes, ret); + + xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED); + return ret; +} + +STATIC void +xfs_aio_write_isize_update( + struct inode *inode, + loff_t *ppos, + ssize_t bytes_written) +{ + struct xfs_inode *ip = XFS_I(inode); + xfs_fsize_t isize = i_size_read(inode); + + if (bytes_written > 0) + XFS_STATS_ADD(xs_write_bytes, bytes_written); + + if (unlikely(bytes_written < 0 && bytes_written != -EFAULT && + *ppos > isize)) + *ppos = isize; + + if (*ppos > ip->i_size) { + xfs_rw_ilock(ip, XFS_ILOCK_EXCL); + if (*ppos > ip->i_size) + ip->i_size = *ppos; + xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); + } +} + +/* + * If this was a direct or synchronous I/O that failed (such as ENOSPC) then + * part of the I/O may have been written to disk before the error occurred. In + * this case the on-disk file size may have been adjusted beyond the in-memory + * file size and now needs to be truncated back. + */ +STATIC void +xfs_aio_write_newsize_update( + struct xfs_inode *ip) +{ + if (ip->i_new_size) { + xfs_rw_ilock(ip, XFS_ILOCK_EXCL); + ip->i_new_size = 0; + if (ip->i_d.di_size > ip->i_size) + ip->i_d.di_size = ip->i_size; + xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); + } +} + +/* + * xfs_file_splice_write() does not use xfs_rw_ilock() because + * generic_file_splice_write() takes the i_mutex itself. This, in theory, + * couuld cause lock inversions between the aio_write path and the splice path + * if someone is doing concurrent splice(2) based writes and write(2) based + * writes to the same inode. The only real way to fix this is to re-implement + * the generic code here with correct locking orders. + */ +STATIC ssize_t +xfs_file_splice_write( + struct pipe_inode_info *pipe, + struct file *outfilp, + loff_t *ppos, + size_t count, + unsigned int flags) +{ + struct inode *inode = outfilp->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + xfs_fsize_t new_size; + int ioflags = 0; + ssize_t ret; + + XFS_STATS_INC(xs_write_calls); + + if (outfilp->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -EIO; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + new_size = *ppos + count; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (new_size > ip->i_size) + ip->i_new_size = new_size; + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + trace_xfs_file_splice_write(ip, count, *ppos, ioflags); + + ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags); + + xfs_aio_write_isize_update(inode, ppos, ret); + xfs_aio_write_newsize_update(ip); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return ret; +} + +/* + * This routine is called to handle zeroing any space in the last + * block of the file that is beyond the EOF. We do this since the + * size is being increased without writing anything to that block + * and we don't want anyone to read the garbage on the disk. + */ +STATIC int /* error (positive) */ +xfs_zero_last_block( + xfs_inode_t *ip, + xfs_fsize_t offset, + xfs_fsize_t isize) +{ + xfs_fileoff_t last_fsb; + xfs_mount_t *mp = ip->i_mount; + int nimaps; + int zero_offset; + int zero_len; + int error = 0; + xfs_bmbt_irec_t imap; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + zero_offset = XFS_B_FSB_OFFSET(mp, isize); + if (zero_offset == 0) { + /* + * There are no extra bytes in the last block on disk to + * zero, so return. + */ + return 0; + } + + last_fsb = XFS_B_TO_FSBT(mp, isize); + nimaps = 1; + error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap, + &nimaps, NULL); + if (error) { + return error; + } + ASSERT(nimaps > 0); + /* + * If the block underlying isize is just a hole, then there + * is nothing to zero. + */ + if (imap.br_startblock == HOLESTARTBLOCK) { + return 0; + } + /* + * Zero the part of the last block beyond the EOF, and write it + * out sync. We need to drop the ilock while we do this so we + * don't deadlock when the buffer cache calls back to us. + */ + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + zero_len = mp->m_sb.sb_blocksize - zero_offset; + if (isize + zero_len > offset) + zero_len = offset - isize; + error = xfs_iozero(ip, isize, zero_len); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + ASSERT(error >= 0); + return error; +} + +/* + * Zero any on disk space between the current EOF and the new, + * larger EOF. This handles the normal case of zeroing the remainder + * of the last block in the file and the unusual case of zeroing blocks + * out beyond the size of the file. This second case only happens + * with fixed size extents and when the system crashes before the inode + * size was updated but after blocks were allocated. If fill is set, + * then any holes in the range are filled and zeroed. If not, the holes + * are left alone as holes. + */ + +int /* error (positive) */ +xfs_zero_eof( + xfs_inode_t *ip, + xfs_off_t offset, /* starting I/O offset */ + xfs_fsize_t isize) /* current inode size */ +{ + xfs_mount_t *mp = ip->i_mount; + xfs_fileoff_t start_zero_fsb; + xfs_fileoff_t end_zero_fsb; + xfs_fileoff_t zero_count_fsb; + xfs_fileoff_t last_fsb; + xfs_fileoff_t zero_off; + xfs_fsize_t zero_len; + int nimaps; + int error = 0; + xfs_bmbt_irec_t imap; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + ASSERT(offset > isize); + + /* + * First handle zeroing the block on which isize resides. + * We only zero a part of that block so it is handled specially. + */ + error = xfs_zero_last_block(ip, offset, isize); + if (error) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + return error; + } + + /* + * Calculate the range between the new size and the old + * where blocks needing to be zeroed may exist. To get the + * block where the last byte in the file currently resides, + * we need to subtract one from the size and truncate back + * to a block boundary. We subtract 1 in case the size is + * exactly on a block boundary. + */ + last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1; + start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize); + end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1); + ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb); + if (last_fsb == end_zero_fsb) { + /* + * The size was only incremented on its last block. + * We took care of that above, so just return. + */ + return 0; + } + + ASSERT(start_zero_fsb <= end_zero_fsb); + while (start_zero_fsb <= end_zero_fsb) { + nimaps = 1; + zero_count_fsb = end_zero_fsb - start_zero_fsb + 1; + error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb, + 0, NULL, 0, &imap, &nimaps, NULL); + if (error) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL)); + return error; + } + ASSERT(nimaps > 0); + + if (imap.br_state == XFS_EXT_UNWRITTEN || + imap.br_startblock == HOLESTARTBLOCK) { + /* + * This loop handles initializing pages that were + * partially initialized by the code below this + * loop. It basically zeroes the part of the page + * that sits on a hole and sets the page as P_HOLE + * and calls remapf if it is a mapped file. + */ + start_zero_fsb = imap.br_startoff + imap.br_blockcount; + ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); + continue; + } + + /* + * There are blocks we need to zero. + * Drop the inode lock while we're doing the I/O. + * We'll still have the iolock to protect us. + */ + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + zero_off = XFS_FSB_TO_B(mp, start_zero_fsb); + zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount); + + if ((zero_off + zero_len) > offset) + zero_len = offset - zero_off; + + error = xfs_iozero(ip, zero_off, zero_len); + if (error) { + goto out_lock; + } + + start_zero_fsb = imap.br_startoff + imap.br_blockcount; + ASSERT(start_zero_fsb <= (end_zero_fsb + 1)); + + xfs_ilock(ip, XFS_ILOCK_EXCL); + } + + return 0; + +out_lock: + xfs_ilock(ip, XFS_ILOCK_EXCL); + ASSERT(error >= 0); + return error; +} + +/* + * Common pre-write limit and setup checks. + * + * Returns with iolock held according to @iolock. + */ +STATIC ssize_t +xfs_file_aio_write_checks( + struct file *file, + loff_t *pos, + size_t *count, + int *iolock) +{ + struct inode *inode = file->f_mapping->host; + struct xfs_inode *ip = XFS_I(inode); + xfs_fsize_t new_size; + int error = 0; + + error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode)); + if (error) { + xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock); + *iolock = 0; + return error; + } + + new_size = *pos + *count; + if (new_size > ip->i_size) + ip->i_new_size = new_size; + + if (likely(!(file->f_mode & FMODE_NOCMTIME))) + file_update_time(file); + + /* + * If the offset is beyond the size of the file, we need to zero any + * blocks that fall between the existing EOF and the start of this + * write. + */ + if (*pos > ip->i_size) + error = -xfs_zero_eof(ip, *pos, ip->i_size); + + xfs_rw_iunlock(ip, XFS_ILOCK_EXCL); + if (error) + return error; + + /* + * If we're writing the file then make sure to clear the setuid and + * setgid bits if the process is not being run by root. This keeps + * people from modifying setuid and setgid binaries. + */ + return file_remove_suid(file); + +} + +/* + * xfs_file_dio_aio_write - handle direct IO writes + * + * Lock the inode appropriately to prepare for and issue a direct IO write. + * By separating it from the buffered write path we remove all the tricky to + * follow locking changes and looping. + * + * If there are cached pages or we're extending the file, we need IOLOCK_EXCL + * until we're sure the bytes at the new EOF have been zeroed and/or the cached + * pages are flushed out. + * + * In most cases the direct IO writes will be done holding IOLOCK_SHARED + * allowing them to be done in parallel with reads and other direct IO writes. + * However, if the IO is not aligned to filesystem blocks, the direct IO layer + * needs to do sub-block zeroing and that requires serialisation against other + * direct IOs to the same block. In this case we need to serialise the + * submission of the unaligned IOs so that we don't get racing block zeroing in + * the dio layer. To avoid the problem with aio, we also need to wait for + * outstanding IOs to complete so that unwritten extent conversion is completed + * before we try to map the overlapping block. This is currently implemented by + * hitting it with a big hammer (i.e. xfs_ioend_wait()). + * + * Returns with locks held indicated by @iolock and errors indicated by + * negative return values. + */ +STATIC ssize_t +xfs_file_dio_aio_write( + struct kiocb *iocb, + const struct iovec *iovp, + unsigned long nr_segs, + loff_t pos, + size_t ocount, + int *iolock) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + ssize_t ret = 0; + size_t count = ocount; + int unaligned_io = 0; + struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp; + + *iolock = 0; + if ((pos & target->bt_smask) || (count & target->bt_smask)) + return -XFS_ERROR(EINVAL); + + if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask)) + unaligned_io = 1; + + if (unaligned_io || mapping->nrpages || pos > ip->i_size) + *iolock = XFS_IOLOCK_EXCL; + else + *iolock = XFS_IOLOCK_SHARED; + xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock); + + ret = xfs_file_aio_write_checks(file, &pos, &count, iolock); + if (ret) + return ret; + + if (mapping->nrpages) { + WARN_ON(*iolock != XFS_IOLOCK_EXCL); + ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1, + FI_REMAPF_LOCKED); + if (ret) + return ret; + } + + /* + * If we are doing unaligned IO, wait for all other IO to drain, + * otherwise demote the lock if we had to flush cached pages + */ + if (unaligned_io) + xfs_ioend_wait(ip); + else if (*iolock == XFS_IOLOCK_EXCL) { + xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL); + *iolock = XFS_IOLOCK_SHARED; + } + + trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0); + ret = generic_file_direct_write(iocb, iovp, + &nr_segs, pos, &iocb->ki_pos, count, ocount); + + /* No fallback to buffered IO on errors for XFS. */ + ASSERT(ret < 0 || ret == count); + return ret; +} + +STATIC ssize_t +xfs_file_buffered_aio_write( + struct kiocb *iocb, + const struct iovec *iovp, + unsigned long nr_segs, + loff_t pos, + size_t ocount, + int *iolock) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct xfs_inode *ip = XFS_I(inode); + ssize_t ret; + int enospc = 0; + size_t count = ocount; + + *iolock = XFS_IOLOCK_EXCL; + xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock); + + ret = xfs_file_aio_write_checks(file, &pos, &count, iolock); + if (ret) + return ret; + + /* We can write back this queue in page reclaim */ + current->backing_dev_info = mapping->backing_dev_info; + +write_retry: + trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0); + ret = generic_file_buffered_write(iocb, iovp, nr_segs, + pos, &iocb->ki_pos, count, ret); + /* + * if we just got an ENOSPC, flush the inode now we aren't holding any + * page locks and retry *once* + */ + if (ret == -ENOSPC && !enospc) { + ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE); + if (ret) + return ret; + enospc = 1; + goto write_retry; + } + current->backing_dev_info = NULL; + return ret; +} + +STATIC ssize_t +xfs_file_aio_write( + struct kiocb *iocb, + const struct iovec *iovp, + unsigned long nr_segs, + loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct xfs_inode *ip = XFS_I(inode); + ssize_t ret; + int iolock; + size_t ocount = 0; + + XFS_STATS_INC(xs_write_calls); + + BUG_ON(iocb->ki_pos != pos); + + ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ); + if (ret) + return ret; + + if (ocount == 0) + return 0; + + xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE); + + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return -EIO; + + if (unlikely(file->f_flags & O_DIRECT)) + ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos, + ocount, &iolock); + else + ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos, + ocount, &iolock); + + xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret); + + if (ret <= 0) + goto out_unlock; + + /* Handle various SYNC-type writes */ + if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { + loff_t end = pos + ret - 1; + int error; + + xfs_rw_iunlock(ip, iolock); + error = xfs_file_fsync(file, pos, end, + (file->f_flags & __O_SYNC) ? 0 : 1); + xfs_rw_ilock(ip, iolock); + if (error) + ret = error; + } + +out_unlock: + xfs_aio_write_newsize_update(ip); + xfs_rw_iunlock(ip, iolock); + return ret; +} + +STATIC long +xfs_file_fallocate( + struct file *file, + int mode, + loff_t offset, + loff_t len) +{ + struct inode *inode = file->f_path.dentry->d_inode; + long error; + loff_t new_size = 0; + xfs_flock64_t bf; + xfs_inode_t *ip = XFS_I(inode); + int cmd = XFS_IOC_RESVSP; + int attr_flags = XFS_ATTR_NOLOCK; + + if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE)) + return -EOPNOTSUPP; + + bf.l_whence = 0; + bf.l_start = offset; + bf.l_len = len; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + if (mode & FALLOC_FL_PUNCH_HOLE) + cmd = XFS_IOC_UNRESVSP; + + /* check the new inode size is valid before allocating */ + if (!(mode & FALLOC_FL_KEEP_SIZE) && + offset + len > i_size_read(inode)) { + new_size = offset + len; + error = inode_newsize_ok(inode, new_size); + if (error) + goto out_unlock; + } + + if (file->f_flags & O_DSYNC) + attr_flags |= XFS_ATTR_SYNC; + + error = -xfs_change_file_space(ip, cmd, &bf, 0, attr_flags); + if (error) + goto out_unlock; + + /* Change file size if needed */ + if (new_size) { + struct iattr iattr; + + iattr.ia_valid = ATTR_SIZE; + iattr.ia_size = new_size; + error = -xfs_setattr_size(ip, &iattr, XFS_ATTR_NOLOCK); + } + +out_unlock: + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + return error; +} + + +STATIC int +xfs_file_open( + struct inode *inode, + struct file *file) +{ + if (!(file->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS) + return -EFBIG; + if (XFS_FORCED_SHUTDOWN(XFS_M(inode->i_sb))) + return -EIO; + return 0; +} + +STATIC int +xfs_dir_open( + struct inode *inode, + struct file *file) +{ + struct xfs_inode *ip = XFS_I(inode); + int mode; + int error; + + error = xfs_file_open(inode, file); + if (error) + return error; + + /* + * If there are any blocks, read-ahead block 0 as we're almost + * certain to have the next operation be a read there. + */ + mode = xfs_ilock_map_shared(ip); + if (ip->i_d.di_nextents > 0) + xfs_da_reada_buf(NULL, ip, 0, XFS_DATA_FORK); + xfs_iunlock(ip, mode); + return 0; +} + +STATIC int +xfs_file_release( + struct inode *inode, + struct file *filp) +{ + return -xfs_release(XFS_I(inode)); +} + +STATIC int +xfs_file_readdir( + struct file *filp, + void *dirent, + filldir_t filldir) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + xfs_inode_t *ip = XFS_I(inode); + int error; + size_t bufsize; + + /* + * The Linux API doesn't pass down the total size of the buffer + * we read into down to the filesystem. With the filldir concept + * it's not needed for correct information, but the XFS dir2 leaf + * code wants an estimate of the buffer size to calculate it's + * readahead window and size the buffers used for mapping to + * physical blocks. + * + * Try to give it an estimate that's good enough, maybe at some + * point we can change the ->readdir prototype to include the + * buffer size. For now we use the current glibc buffer size. + */ + bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size); + + error = xfs_readdir(ip, dirent, bufsize, + (xfs_off_t *)&filp->f_pos, filldir); + if (error) + return -error; + return 0; +} + +STATIC int +xfs_file_mmap( + struct file *filp, + struct vm_area_struct *vma) +{ + vma->vm_ops = &xfs_file_vm_ops; + vma->vm_flags |= VM_CAN_NONLINEAR; + + file_accessed(filp); + return 0; +} + +/* + * mmap()d file has taken write protection fault and is being made + * writable. We can set the page state up correctly for a writable + * page, which means we can do correct delalloc accounting (ENOSPC + * checking!) and unwritten extent mapping. + */ +STATIC int +xfs_vm_page_mkwrite( + struct vm_area_struct *vma, + struct vm_fault *vmf) +{ + return block_page_mkwrite(vma, vmf, xfs_get_blocks); +} + +const struct file_operations xfs_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = xfs_file_aio_read, + .aio_write = xfs_file_aio_write, + .splice_read = xfs_file_splice_read, + .splice_write = xfs_file_splice_write, + .unlocked_ioctl = xfs_file_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = xfs_file_compat_ioctl, +#endif + .mmap = xfs_file_mmap, + .open = xfs_file_open, + .release = xfs_file_release, + .fsync = xfs_file_fsync, + .fallocate = xfs_file_fallocate, +}; + +const struct file_operations xfs_dir_file_operations = { + .open = xfs_dir_open, + .read = generic_read_dir, + .readdir = xfs_file_readdir, + .llseek = generic_file_llseek, + .unlocked_ioctl = xfs_file_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = xfs_file_compat_ioctl, +#endif + .fsync = xfs_file_fsync, +}; + +static const struct vm_operations_struct xfs_file_vm_ops = { + .fault = filemap_fault, + .page_mkwrite = xfs_vm_page_mkwrite, +}; diff --git a/fs/xfs/xfs_fs_subr.c b/fs/xfs/xfs_fs_subr.c new file mode 100644 index 000000000000..ed88ed16811c --- /dev/null +++ b/fs/xfs/xfs_fs_subr.c @@ -0,0 +1,96 @@ +/* + * Copyright (c) 2000-2002,2005-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_vnodeops.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_trace.h" + +/* + * note: all filemap functions return negative error codes. These + * need to be inverted before returning to the xfs core functions. + */ +void +xfs_tosspages( + xfs_inode_t *ip, + xfs_off_t first, + xfs_off_t last, + int fiopt) +{ + /* can't toss partial tail pages, so mask them out */ + last &= ~(PAGE_SIZE - 1); + truncate_inode_pages_range(VFS_I(ip)->i_mapping, first, last - 1); +} + +int +xfs_flushinval_pages( + xfs_inode_t *ip, + xfs_off_t first, + xfs_off_t last, + int fiopt) +{ + struct address_space *mapping = VFS_I(ip)->i_mapping; + int ret = 0; + + trace_xfs_pagecache_inval(ip, first, last); + + xfs_iflags_clear(ip, XFS_ITRUNCATED); + ret = filemap_write_and_wait_range(mapping, first, + last == -1 ? LLONG_MAX : last); + if (!ret) + truncate_inode_pages_range(mapping, first, last); + return -ret; +} + +int +xfs_flush_pages( + xfs_inode_t *ip, + xfs_off_t first, + xfs_off_t last, + uint64_t flags, + int fiopt) +{ + struct address_space *mapping = VFS_I(ip)->i_mapping; + int ret = 0; + int ret2; + + xfs_iflags_clear(ip, XFS_ITRUNCATED); + ret = -filemap_fdatawrite_range(mapping, first, + last == -1 ? LLONG_MAX : last); + if (flags & XBF_ASYNC) + return ret; + ret2 = xfs_wait_on_pages(ip, first, last); + if (!ret) + ret = ret2; + return ret; +} + +int +xfs_wait_on_pages( + xfs_inode_t *ip, + xfs_off_t first, + xfs_off_t last) +{ + struct address_space *mapping = VFS_I(ip)->i_mapping; + + if (mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK)) { + return -filemap_fdatawait_range(mapping, first, + last == -1 ? ip->i_size - 1 : last); + } + return 0; +} diff --git a/fs/xfs/xfs_globals.c b/fs/xfs/xfs_globals.c new file mode 100644 index 000000000000..76e81cff70b9 --- /dev/null +++ b/fs/xfs/xfs_globals.c @@ -0,0 +1,43 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_sysctl.h" + +/* + * Tunable XFS parameters. xfs_params is required even when CONFIG_SYSCTL=n, + * other XFS code uses these values. Times are measured in centisecs (i.e. + * 100ths of a second). + */ +xfs_param_t xfs_params = { + /* MIN DFLT MAX */ + .sgid_inherit = { 0, 0, 1 }, + .symlink_mode = { 0, 0, 1 }, + .panic_mask = { 0, 0, 255 }, + .error_level = { 0, 3, 11 }, + .syncd_timer = { 1*100, 30*100, 7200*100}, + .stats_clear = { 0, 0, 1 }, + .inherit_sync = { 0, 1, 1 }, + .inherit_nodump = { 0, 1, 1 }, + .inherit_noatim = { 0, 1, 1 }, + .xfs_buf_timer = { 100/2, 1*100, 30*100 }, + .xfs_buf_age = { 1*100, 15*100, 7200*100}, + .inherit_nosym = { 0, 0, 1 }, + .rotorstep = { 1, 1, 255 }, + .inherit_nodfrg = { 0, 1, 1 }, + .fstrm_timer = { 1, 30*100, 3600*100}, +}; diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c new file mode 100644 index 000000000000..f7ce7debe14c --- /dev/null +++ b/fs/xfs/xfs_ioctl.c @@ -0,0 +1,1556 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_ioctl.h" +#include "xfs_rtalloc.h" +#include "xfs_itable.h" +#include "xfs_error.h" +#include "xfs_attr.h" +#include "xfs_bmap.h" +#include "xfs_buf_item.h" +#include "xfs_utils.h" +#include "xfs_dfrag.h" +#include "xfs_fsops.h" +#include "xfs_vnodeops.h" +#include "xfs_discard.h" +#include "xfs_quota.h" +#include "xfs_inode_item.h" +#include "xfs_export.h" +#include "xfs_trace.h" + +#include +#include +#include +#include +#include +#include +#include + +/* + * xfs_find_handle maps from userspace xfs_fsop_handlereq structure to + * a file or fs handle. + * + * XFS_IOC_PATH_TO_FSHANDLE + * returns fs handle for a mount point or path within that mount point + * XFS_IOC_FD_TO_HANDLE + * returns full handle for a FD opened in user space + * XFS_IOC_PATH_TO_HANDLE + * returns full handle for a path + */ +int +xfs_find_handle( + unsigned int cmd, + xfs_fsop_handlereq_t *hreq) +{ + int hsize; + xfs_handle_t handle; + struct inode *inode; + struct file *file = NULL; + struct path path; + int error; + struct xfs_inode *ip; + + if (cmd == XFS_IOC_FD_TO_HANDLE) { + file = fget(hreq->fd); + if (!file) + return -EBADF; + inode = file->f_path.dentry->d_inode; + } else { + error = user_lpath((const char __user *)hreq->path, &path); + if (error) + return error; + inode = path.dentry->d_inode; + } + ip = XFS_I(inode); + + /* + * We can only generate handles for inodes residing on a XFS filesystem, + * and only for regular files, directories or symbolic links. + */ + error = -EINVAL; + if (inode->i_sb->s_magic != XFS_SB_MAGIC) + goto out_put; + + error = -EBADF; + if (!S_ISREG(inode->i_mode) && + !S_ISDIR(inode->i_mode) && + !S_ISLNK(inode->i_mode)) + goto out_put; + + + memcpy(&handle.ha_fsid, ip->i_mount->m_fixedfsid, sizeof(xfs_fsid_t)); + + if (cmd == XFS_IOC_PATH_TO_FSHANDLE) { + /* + * This handle only contains an fsid, zero the rest. + */ + memset(&handle.ha_fid, 0, sizeof(handle.ha_fid)); + hsize = sizeof(xfs_fsid_t); + } else { + int lock_mode; + + lock_mode = xfs_ilock_map_shared(ip); + handle.ha_fid.fid_len = sizeof(xfs_fid_t) - + sizeof(handle.ha_fid.fid_len); + handle.ha_fid.fid_pad = 0; + handle.ha_fid.fid_gen = ip->i_d.di_gen; + handle.ha_fid.fid_ino = ip->i_ino; + xfs_iunlock_map_shared(ip, lock_mode); + + hsize = XFS_HSIZE(handle); + } + + error = -EFAULT; + if (copy_to_user(hreq->ohandle, &handle, hsize) || + copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) + goto out_put; + + error = 0; + + out_put: + if (cmd == XFS_IOC_FD_TO_HANDLE) + fput(file); + else + path_put(&path); + return error; +} + +/* + * No need to do permission checks on the various pathname components + * as the handle operations are privileged. + */ +STATIC int +xfs_handle_acceptable( + void *context, + struct dentry *dentry) +{ + return 1; +} + +/* + * Convert userspace handle data into a dentry. + */ +struct dentry * +xfs_handle_to_dentry( + struct file *parfilp, + void __user *uhandle, + u32 hlen) +{ + xfs_handle_t handle; + struct xfs_fid64 fid; + + /* + * Only allow handle opens under a directory. + */ + if (!S_ISDIR(parfilp->f_path.dentry->d_inode->i_mode)) + return ERR_PTR(-ENOTDIR); + + if (hlen != sizeof(xfs_handle_t)) + return ERR_PTR(-EINVAL); + if (copy_from_user(&handle, uhandle, hlen)) + return ERR_PTR(-EFAULT); + if (handle.ha_fid.fid_len != + sizeof(handle.ha_fid) - sizeof(handle.ha_fid.fid_len)) + return ERR_PTR(-EINVAL); + + memset(&fid, 0, sizeof(struct fid)); + fid.ino = handle.ha_fid.fid_ino; + fid.gen = handle.ha_fid.fid_gen; + + return exportfs_decode_fh(parfilp->f_path.mnt, (struct fid *)&fid, 3, + FILEID_INO32_GEN | XFS_FILEID_TYPE_64FLAG, + xfs_handle_acceptable, NULL); +} + +STATIC struct dentry * +xfs_handlereq_to_dentry( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + return xfs_handle_to_dentry(parfilp, hreq->ihandle, hreq->ihandlen); +} + +int +xfs_open_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + const struct cred *cred = current_cred(); + int error; + int fd; + int permflag; + struct file *filp; + struct inode *inode; + struct dentry *dentry; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + + dentry = xfs_handlereq_to_dentry(parfilp, hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + inode = dentry->d_inode; + + /* Restrict xfs_open_by_handle to directories & regular files. */ + if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) { + error = -XFS_ERROR(EPERM); + goto out_dput; + } + +#if BITS_PER_LONG != 32 + hreq->oflags |= O_LARGEFILE; +#endif + + /* Put open permission in namei format. */ + permflag = hreq->oflags; + if ((permflag+1) & O_ACCMODE) + permflag++; + if (permflag & O_TRUNC) + permflag |= 2; + + if ((!(permflag & O_APPEND) || (permflag & O_TRUNC)) && + (permflag & FMODE_WRITE) && IS_APPEND(inode)) { + error = -XFS_ERROR(EPERM); + goto out_dput; + } + + if ((permflag & FMODE_WRITE) && IS_IMMUTABLE(inode)) { + error = -XFS_ERROR(EACCES); + goto out_dput; + } + + /* Can't write directories. */ + if (S_ISDIR(inode->i_mode) && (permflag & FMODE_WRITE)) { + error = -XFS_ERROR(EISDIR); + goto out_dput; + } + + fd = get_unused_fd(); + if (fd < 0) { + error = fd; + goto out_dput; + } + + filp = dentry_open(dentry, mntget(parfilp->f_path.mnt), + hreq->oflags, cred); + if (IS_ERR(filp)) { + put_unused_fd(fd); + return PTR_ERR(filp); + } + + if (S_ISREG(inode->i_mode)) { + filp->f_flags |= O_NOATIME; + filp->f_mode |= FMODE_NOCMTIME; + } + + fd_install(fd, filp); + return fd; + + out_dput: + dput(dentry); + return error; +} + +/* + * This is a copy from fs/namei.c:vfs_readlink(), except for removing it's + * unused first argument. + */ +STATIC int +do_readlink( + char __user *buffer, + int buflen, + const char *link) +{ + int len; + + len = PTR_ERR(link); + if (IS_ERR(link)) + goto out; + + len = strlen(link); + if (len > (unsigned) buflen) + len = buflen; + if (copy_to_user(buffer, link, len)) + len = -EFAULT; + out: + return len; +} + + +int +xfs_readlink_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq) +{ + struct dentry *dentry; + __u32 olen; + void *link; + int error; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + + dentry = xfs_handlereq_to_dentry(parfilp, hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + /* Restrict this handle operation to symlinks only. */ + if (!S_ISLNK(dentry->d_inode->i_mode)) { + error = -XFS_ERROR(EINVAL); + goto out_dput; + } + + if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) { + error = -XFS_ERROR(EFAULT); + goto out_dput; + } + + link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); + if (!link) { + error = -XFS_ERROR(ENOMEM); + goto out_dput; + } + + error = -xfs_readlink(XFS_I(dentry->d_inode), link); + if (error) + goto out_kfree; + error = do_readlink(hreq->ohandle, olen, link); + if (error) + goto out_kfree; + + out_kfree: + kfree(link); + out_dput: + dput(dentry); + return error; +} + +STATIC int +xfs_fssetdm_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + struct fsdmidata fsd; + xfs_fsop_setdm_handlereq_t dmhreq; + struct dentry *dentry; + + if (!capable(CAP_MKNOD)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&dmhreq, arg, sizeof(xfs_fsop_setdm_handlereq_t))) + return -XFS_ERROR(EFAULT); + + dentry = xfs_handlereq_to_dentry(parfilp, &dmhreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) { + error = -XFS_ERROR(EPERM); + goto out; + } + + if (copy_from_user(&fsd, dmhreq.data, sizeof(fsd))) { + error = -XFS_ERROR(EFAULT); + goto out; + } + + error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask, + fsd.fsd_dmstate); + + out: + dput(dentry); + return error; +} + +STATIC int +xfs_attrlist_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error = -ENOMEM; + attrlist_cursor_kern_t *cursor; + xfs_fsop_attrlist_handlereq_t al_hreq; + struct dentry *dentry; + char *kbuf; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&al_hreq, arg, sizeof(xfs_fsop_attrlist_handlereq_t))) + return -XFS_ERROR(EFAULT); + if (al_hreq.buflen > XATTR_LIST_MAX) + return -XFS_ERROR(EINVAL); + + /* + * Reject flags, only allow namespaces. + */ + if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) + return -XFS_ERROR(EINVAL); + + dentry = xfs_handlereq_to_dentry(parfilp, &al_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + kbuf = kzalloc(al_hreq.buflen, GFP_KERNEL); + if (!kbuf) + goto out_dput; + + cursor = (attrlist_cursor_kern_t *)&al_hreq.pos; + error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen, + al_hreq.flags, cursor); + if (error) + goto out_kfree; + + if (copy_to_user(al_hreq.buffer, kbuf, al_hreq.buflen)) + error = -EFAULT; + + out_kfree: + kfree(kbuf); + out_dput: + dput(dentry); + return error; +} + +int +xfs_attrmulti_attr_get( + struct inode *inode, + unsigned char *name, + unsigned char __user *ubuf, + __uint32_t *len, + __uint32_t flags) +{ + unsigned char *kbuf; + int error = EFAULT; + + if (*len > XATTR_SIZE_MAX) + return EINVAL; + kbuf = kmalloc(*len, GFP_KERNEL); + if (!kbuf) + return ENOMEM; + + error = xfs_attr_get(XFS_I(inode), name, kbuf, (int *)len, flags); + if (error) + goto out_kfree; + + if (copy_to_user(ubuf, kbuf, *len)) + error = EFAULT; + + out_kfree: + kfree(kbuf); + return error; +} + +int +xfs_attrmulti_attr_set( + struct inode *inode, + unsigned char *name, + const unsigned char __user *ubuf, + __uint32_t len, + __uint32_t flags) +{ + unsigned char *kbuf; + int error = EFAULT; + + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return EPERM; + if (len > XATTR_SIZE_MAX) + return EINVAL; + + kbuf = memdup_user(ubuf, len); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); + + error = xfs_attr_set(XFS_I(inode), name, kbuf, len, flags); + + return error; +} + +int +xfs_attrmulti_attr_remove( + struct inode *inode, + unsigned char *name, + __uint32_t flags) +{ + if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) + return EPERM; + return xfs_attr_remove(XFS_I(inode), name, flags); +} + +STATIC int +xfs_attrmulti_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + xfs_attr_multiop_t *ops; + xfs_fsop_attrmulti_handlereq_t am_hreq; + struct dentry *dentry; + unsigned int i, size; + unsigned char *attr_name; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t))) + return -XFS_ERROR(EFAULT); + + /* overflow check */ + if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t)) + return -E2BIG; + + dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + error = E2BIG; + size = am_hreq.opcount * sizeof(xfs_attr_multiop_t); + if (!size || size > 16 * PAGE_SIZE) + goto out_dput; + + ops = memdup_user(am_hreq.ops, size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); + goto out_dput; + } + + attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); + if (!attr_name) + goto out_kfree_ops; + + error = 0; + for (i = 0; i < am_hreq.opcount; i++) { + ops[i].am_error = strncpy_from_user((char *)attr_name, + ops[i].am_attrname, MAXNAMELEN); + if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) + error = -ERANGE; + if (ops[i].am_error < 0) + break; + + switch (ops[i].am_opcode) { + case ATTR_OP_GET: + ops[i].am_error = xfs_attrmulti_attr_get( + dentry->d_inode, attr_name, + ops[i].am_attrvalue, &ops[i].am_length, + ops[i].am_flags); + break; + case ATTR_OP_SET: + ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + if (ops[i].am_error) + break; + ops[i].am_error = xfs_attrmulti_attr_set( + dentry->d_inode, attr_name, + ops[i].am_attrvalue, ops[i].am_length, + ops[i].am_flags); + mnt_drop_write(parfilp->f_path.mnt); + break; + case ATTR_OP_REMOVE: + ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + if (ops[i].am_error) + break; + ops[i].am_error = xfs_attrmulti_attr_remove( + dentry->d_inode, attr_name, + ops[i].am_flags); + mnt_drop_write(parfilp->f_path.mnt); + break; + default: + ops[i].am_error = EINVAL; + } + } + + if (copy_to_user(am_hreq.ops, ops, size)) + error = XFS_ERROR(EFAULT); + + kfree(attr_name); + out_kfree_ops: + kfree(ops); + out_dput: + dput(dentry); + return -error; +} + +int +xfs_ioc_space( + struct xfs_inode *ip, + struct inode *inode, + struct file *filp, + int ioflags, + unsigned int cmd, + xfs_flock64_t *bf) +{ + int attr_flags = 0; + int error; + + /* + * Only allow the sys admin to reserve space unless + * unwritten extents are enabled. + */ + if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) && + !capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + + if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) + return -XFS_ERROR(EPERM); + + if (!(filp->f_mode & FMODE_WRITE)) + return -XFS_ERROR(EBADF); + + if (!S_ISREG(inode->i_mode)) + return -XFS_ERROR(EINVAL); + + if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) + attr_flags |= XFS_ATTR_NONBLOCK; + + if (filp->f_flags & O_DSYNC) + attr_flags |= XFS_ATTR_SYNC; + + if (ioflags & IO_INVIS) + attr_flags |= XFS_ATTR_DMI; + + error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags); + return -error; +} + +STATIC int +xfs_ioc_bulkstat( + xfs_mount_t *mp, + unsigned int cmd, + void __user *arg) +{ + xfs_fsop_bulkreq_t bulkreq; + int count; /* # of records returned */ + xfs_ino_t inlast; /* last inode number */ + int done; + int error; + + /* done = 1 if there are more stats to get and if bulkstat */ + /* should be called again (unused here, but used in dmapi) */ + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (XFS_FORCED_SHUTDOWN(mp)) + return -XFS_ERROR(EIO); + + if (copy_from_user(&bulkreq, arg, sizeof(xfs_fsop_bulkreq_t))) + return -XFS_ERROR(EFAULT); + + if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64))) + return -XFS_ERROR(EFAULT); + + if ((count = bulkreq.icount) <= 0) + return -XFS_ERROR(EINVAL); + + if (bulkreq.ubuffer == NULL) + return -XFS_ERROR(EINVAL); + + if (cmd == XFS_IOC_FSINUMBERS) + error = xfs_inumbers(mp, &inlast, &count, + bulkreq.ubuffer, xfs_inumbers_fmt); + else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE) + error = xfs_bulkstat_single(mp, &inlast, + bulkreq.ubuffer, &done); + else /* XFS_IOC_FSBULKSTAT */ + error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one, + sizeof(xfs_bstat_t), bulkreq.ubuffer, + &done); + + if (error) + return -error; + + if (bulkreq.ocount != NULL) { + if (copy_to_user(bulkreq.lastip, &inlast, + sizeof(xfs_ino_t))) + return -XFS_ERROR(EFAULT); + + if (copy_to_user(bulkreq.ocount, &count, sizeof(count))) + return -XFS_ERROR(EFAULT); + } + + return 0; +} + +STATIC int +xfs_ioc_fsgeometry_v1( + xfs_mount_t *mp, + void __user *arg) +{ + xfs_fsop_geom_t fsgeo; + int error; + + error = xfs_fs_geometry(mp, &fsgeo, 3); + if (error) + return -error; + + /* + * Caller should have passed an argument of type + * xfs_fsop_geom_v1_t. This is a proper subset of the + * xfs_fsop_geom_t that xfs_fs_geometry() fills in. + */ + if (copy_to_user(arg, &fsgeo, sizeof(xfs_fsop_geom_v1_t))) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_ioc_fsgeometry( + xfs_mount_t *mp, + void __user *arg) +{ + xfs_fsop_geom_t fsgeo; + int error; + + error = xfs_fs_geometry(mp, &fsgeo, 4); + if (error) + return -error; + + if (copy_to_user(arg, &fsgeo, sizeof(fsgeo))) + return -XFS_ERROR(EFAULT); + return 0; +} + +/* + * Linux extended inode flags interface. + */ + +STATIC unsigned int +xfs_merge_ioc_xflags( + unsigned int flags, + unsigned int start) +{ + unsigned int xflags = start; + + if (flags & FS_IMMUTABLE_FL) + xflags |= XFS_XFLAG_IMMUTABLE; + else + xflags &= ~XFS_XFLAG_IMMUTABLE; + if (flags & FS_APPEND_FL) + xflags |= XFS_XFLAG_APPEND; + else + xflags &= ~XFS_XFLAG_APPEND; + if (flags & FS_SYNC_FL) + xflags |= XFS_XFLAG_SYNC; + else + xflags &= ~XFS_XFLAG_SYNC; + if (flags & FS_NOATIME_FL) + xflags |= XFS_XFLAG_NOATIME; + else + xflags &= ~XFS_XFLAG_NOATIME; + if (flags & FS_NODUMP_FL) + xflags |= XFS_XFLAG_NODUMP; + else + xflags &= ~XFS_XFLAG_NODUMP; + + return xflags; +} + +STATIC unsigned int +xfs_di2lxflags( + __uint16_t di_flags) +{ + unsigned int flags = 0; + + if (di_flags & XFS_DIFLAG_IMMUTABLE) + flags |= FS_IMMUTABLE_FL; + if (di_flags & XFS_DIFLAG_APPEND) + flags |= FS_APPEND_FL; + if (di_flags & XFS_DIFLAG_SYNC) + flags |= FS_SYNC_FL; + if (di_flags & XFS_DIFLAG_NOATIME) + flags |= FS_NOATIME_FL; + if (di_flags & XFS_DIFLAG_NODUMP) + flags |= FS_NODUMP_FL; + return flags; +} + +STATIC int +xfs_ioc_fsgetxattr( + xfs_inode_t *ip, + int attr, + void __user *arg) +{ + struct fsxattr fa; + + memset(&fa, 0, sizeof(struct fsxattr)); + + xfs_ilock(ip, XFS_ILOCK_SHARED); + fa.fsx_xflags = xfs_ip2xflags(ip); + fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog; + fa.fsx_projid = xfs_get_projid(ip); + + if (attr) { + if (ip->i_afp) { + if (ip->i_afp->if_flags & XFS_IFEXTENTS) + fa.fsx_nextents = ip->i_afp->if_bytes / + sizeof(xfs_bmbt_rec_t); + else + fa.fsx_nextents = ip->i_d.di_anextents; + } else + fa.fsx_nextents = 0; + } else { + if (ip->i_df.if_flags & XFS_IFEXTENTS) + fa.fsx_nextents = ip->i_df.if_bytes / + sizeof(xfs_bmbt_rec_t); + else + fa.fsx_nextents = ip->i_d.di_nextents; + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + + if (copy_to_user(arg, &fa, sizeof(fa))) + return -EFAULT; + return 0; +} + +STATIC void +xfs_set_diflags( + struct xfs_inode *ip, + unsigned int xflags) +{ + unsigned int di_flags; + + /* can't set PREALLOC this way, just preserve it */ + di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC); + if (xflags & XFS_XFLAG_IMMUTABLE) + di_flags |= XFS_DIFLAG_IMMUTABLE; + if (xflags & XFS_XFLAG_APPEND) + di_flags |= XFS_DIFLAG_APPEND; + if (xflags & XFS_XFLAG_SYNC) + di_flags |= XFS_DIFLAG_SYNC; + if (xflags & XFS_XFLAG_NOATIME) + di_flags |= XFS_DIFLAG_NOATIME; + if (xflags & XFS_XFLAG_NODUMP) + di_flags |= XFS_DIFLAG_NODUMP; + if (xflags & XFS_XFLAG_PROJINHERIT) + di_flags |= XFS_DIFLAG_PROJINHERIT; + if (xflags & XFS_XFLAG_NODEFRAG) + di_flags |= XFS_DIFLAG_NODEFRAG; + if (xflags & XFS_XFLAG_FILESTREAM) + di_flags |= XFS_DIFLAG_FILESTREAM; + if (S_ISDIR(ip->i_d.di_mode)) { + if (xflags & XFS_XFLAG_RTINHERIT) + di_flags |= XFS_DIFLAG_RTINHERIT; + if (xflags & XFS_XFLAG_NOSYMLINKS) + di_flags |= XFS_DIFLAG_NOSYMLINKS; + if (xflags & XFS_XFLAG_EXTSZINHERIT) + di_flags |= XFS_DIFLAG_EXTSZINHERIT; + } else if (S_ISREG(ip->i_d.di_mode)) { + if (xflags & XFS_XFLAG_REALTIME) + di_flags |= XFS_DIFLAG_REALTIME; + if (xflags & XFS_XFLAG_EXTSIZE) + di_flags |= XFS_DIFLAG_EXTSIZE; + } + + ip->i_d.di_flags = di_flags; +} + +STATIC void +xfs_diflags_to_linux( + struct xfs_inode *ip) +{ + struct inode *inode = VFS_I(ip); + unsigned int xflags = xfs_ip2xflags(ip); + + if (xflags & XFS_XFLAG_IMMUTABLE) + inode->i_flags |= S_IMMUTABLE; + else + inode->i_flags &= ~S_IMMUTABLE; + if (xflags & XFS_XFLAG_APPEND) + inode->i_flags |= S_APPEND; + else + inode->i_flags &= ~S_APPEND; + if (xflags & XFS_XFLAG_SYNC) + inode->i_flags |= S_SYNC; + else + inode->i_flags &= ~S_SYNC; + if (xflags & XFS_XFLAG_NOATIME) + inode->i_flags |= S_NOATIME; + else + inode->i_flags &= ~S_NOATIME; +} + +#define FSX_PROJID 1 +#define FSX_EXTSIZE 2 +#define FSX_XFLAGS 4 +#define FSX_NONBLOCK 8 + +STATIC int +xfs_ioctl_setattr( + xfs_inode_t *ip, + struct fsxattr *fa, + int mask) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + unsigned int lock_flags = 0; + struct xfs_dquot *udqp = NULL; + struct xfs_dquot *gdqp = NULL; + struct xfs_dquot *olddquot = NULL; + int code; + + trace_xfs_ioctl_setattr(ip); + + if (mp->m_flags & XFS_MOUNT_RDONLY) + return XFS_ERROR(EROFS); + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + /* + * Disallow 32bit project ids when projid32bit feature is not enabled. + */ + if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1) && + !xfs_sb_version_hasprojid32bit(&ip->i_mount->m_sb)) + return XFS_ERROR(EINVAL); + + /* + * If disk quotas is on, we make sure that the dquots do exist on disk, + * before we start any other transactions. Trying to do this later + * is messy. We don't care to take a readlock to look at the ids + * in inode here, because we can't hold it across the trans_reserve. + * If the IDs do change before we take the ilock, we're covered + * because the i_*dquot fields will get updated anyway. + */ + if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) { + code = xfs_qm_vop_dqalloc(ip, ip->i_d.di_uid, + ip->i_d.di_gid, fa->fsx_projid, + XFS_QMOPT_PQUOTA, &udqp, &gdqp); + if (code) + return code; + } + + /* + * For the other attributes, we acquire the inode lock and + * first do an error checking pass. + */ + tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); + code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + if (code) + goto error_return; + + lock_flags = XFS_ILOCK_EXCL; + xfs_ilock(ip, lock_flags); + + /* + * CAP_FOWNER overrides the following restrictions: + * + * The user ID of the calling process must be equal + * to the file owner ID, except in cases where the + * CAP_FSETID capability is applicable. + */ + if (current_fsuid() != ip->i_d.di_uid && !capable(CAP_FOWNER)) { + code = XFS_ERROR(EPERM); + goto error_return; + } + + /* + * Do a quota reservation only if projid is actually going to change. + */ + if (mask & FSX_PROJID) { + if (XFS_IS_QUOTA_RUNNING(mp) && + XFS_IS_PQUOTA_ON(mp) && + xfs_get_projid(ip) != fa->fsx_projid) { + ASSERT(tp); + code = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, + capable(CAP_FOWNER) ? + XFS_QMOPT_FORCE_RES : 0); + if (code) /* out of quota */ + goto error_return; + } + } + + if (mask & FSX_EXTSIZE) { + /* + * Can't change extent size if any extents are allocated. + */ + if (ip->i_d.di_nextents && + ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != + fa->fsx_extsize)) { + code = XFS_ERROR(EINVAL); /* EFBIG? */ + goto error_return; + } + + /* + * Extent size must be a multiple of the appropriate block + * size, if set at all. It must also be smaller than the + * maximum extent size supported by the filesystem. + * + * Also, for non-realtime files, limit the extent size hint to + * half the size of the AGs in the filesystem so alignment + * doesn't result in extents larger than an AG. + */ + if (fa->fsx_extsize != 0) { + xfs_extlen_t size; + xfs_fsblock_t extsize_fsb; + + extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize); + if (extsize_fsb > MAXEXTLEN) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + + if (XFS_IS_REALTIME_INODE(ip) || + ((mask & FSX_XFLAGS) && + (fa->fsx_xflags & XFS_XFLAG_REALTIME))) { + size = mp->m_sb.sb_rextsize << + mp->m_sb.sb_blocklog; + } else { + size = mp->m_sb.sb_blocksize; + if (extsize_fsb > mp->m_sb.sb_agblocks / 2) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + } + + if (fa->fsx_extsize % size) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + } + } + + + if (mask & FSX_XFLAGS) { + /* + * Can't change realtime flag if any extents are allocated. + */ + if ((ip->i_d.di_nextents || ip->i_delayed_blks) && + (XFS_IS_REALTIME_INODE(ip)) != + (fa->fsx_xflags & XFS_XFLAG_REALTIME)) { + code = XFS_ERROR(EINVAL); /* EFBIG? */ + goto error_return; + } + + /* + * If realtime flag is set then must have realtime data. + */ + if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) { + if ((mp->m_sb.sb_rblocks == 0) || + (mp->m_sb.sb_rextsize == 0) || + (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) { + code = XFS_ERROR(EINVAL); + goto error_return; + } + } + + /* + * Can't modify an immutable/append-only file unless + * we have appropriate permission. + */ + if ((ip->i_d.di_flags & + (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) || + (fa->fsx_xflags & + (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) && + !capable(CAP_LINUX_IMMUTABLE)) { + code = XFS_ERROR(EPERM); + goto error_return; + } + } + + xfs_trans_ijoin(tp, ip); + + /* + * Change file ownership. Must be the owner or privileged. + */ + if (mask & FSX_PROJID) { + /* + * CAP_FSETID overrides the following restrictions: + * + * The set-user-ID and set-group-ID bits of a file will be + * cleared upon successful return from chown() + */ + if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && + !capable(CAP_FSETID)) + ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); + + /* + * Change the ownerships and register quota modifications + * in the transaction. + */ + if (xfs_get_projid(ip) != fa->fsx_projid) { + if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp)) { + olddquot = xfs_qm_vop_chown(tp, ip, + &ip->i_gdquot, gdqp); + } + xfs_set_projid(ip, fa->fsx_projid); + + /* + * We may have to rev the inode as well as + * the superblock version number since projids didn't + * exist before DINODE_VERSION_2 and SB_VERSION_NLINK. + */ + if (ip->i_d.di_version == 1) + xfs_bump_ino_vers2(tp, ip); + } + + } + + if (mask & FSX_EXTSIZE) + ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog; + if (mask & FSX_XFLAGS) { + xfs_set_diflags(ip, fa->fsx_xflags); + xfs_diflags_to_linux(ip); + } + + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + XFS_STATS_INC(xs_ig_attrchg); + + /* + * If this is a synchronous mount, make sure that the + * transaction goes to disk before returning to the user. + * This is slightly sub-optimal in that truncates require + * two sync transactions instead of one for wsync filesystems. + * One for the truncate and one for the timestamps since we + * don't want to change the timestamps unless we're sure the + * truncate worked. Truncates are less than 1% of the laddis + * mix so this probably isn't worth the trouble to optimize. + */ + if (mp->m_flags & XFS_MOUNT_WSYNC) + xfs_trans_set_sync(tp); + code = xfs_trans_commit(tp, 0); + xfs_iunlock(ip, lock_flags); + + /* + * Release any dquot(s) the inode had kept before chown. + */ + xfs_qm_dqrele(olddquot); + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + + return code; + + error_return: + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + xfs_trans_cancel(tp, 0); + if (lock_flags) + xfs_iunlock(ip, lock_flags); + return code; +} + +STATIC int +xfs_ioc_fssetxattr( + xfs_inode_t *ip, + struct file *filp, + void __user *arg) +{ + struct fsxattr fa; + unsigned int mask; + + if (copy_from_user(&fa, arg, sizeof(fa))) + return -EFAULT; + + mask = FSX_XFLAGS | FSX_EXTSIZE | FSX_PROJID; + if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) + mask |= FSX_NONBLOCK; + + return -xfs_ioctl_setattr(ip, &fa, mask); +} + +STATIC int +xfs_ioc_getxflags( + xfs_inode_t *ip, + void __user *arg) +{ + unsigned int flags; + + flags = xfs_di2lxflags(ip->i_d.di_flags); + if (copy_to_user(arg, &flags, sizeof(flags))) + return -EFAULT; + return 0; +} + +STATIC int +xfs_ioc_setxflags( + xfs_inode_t *ip, + struct file *filp, + void __user *arg) +{ + struct fsxattr fa; + unsigned int flags; + unsigned int mask; + + if (copy_from_user(&flags, arg, sizeof(flags))) + return -EFAULT; + + if (flags & ~(FS_IMMUTABLE_FL | FS_APPEND_FL | \ + FS_NOATIME_FL | FS_NODUMP_FL | \ + FS_SYNC_FL)) + return -EOPNOTSUPP; + + mask = FSX_XFLAGS; + if (filp->f_flags & (O_NDELAY|O_NONBLOCK)) + mask |= FSX_NONBLOCK; + fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip)); + + return -xfs_ioctl_setattr(ip, &fa, mask); +} + +STATIC int +xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full) +{ + struct getbmap __user *base = *ap; + + /* copy only getbmap portion (not getbmapx) */ + if (copy_to_user(base, bmv, sizeof(struct getbmap))) + return XFS_ERROR(EFAULT); + + *ap += sizeof(struct getbmap); + return 0; +} + +STATIC int +xfs_ioc_getbmap( + struct xfs_inode *ip, + int ioflags, + unsigned int cmd, + void __user *arg) +{ + struct getbmapx bmx; + int error; + + if (copy_from_user(&bmx, arg, sizeof(struct getbmapx))) + return -XFS_ERROR(EFAULT); + + if (bmx.bmv_count < 2) + return -XFS_ERROR(EINVAL); + + bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0); + if (ioflags & IO_INVIS) + bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ; + + error = xfs_getbmap(ip, &bmx, xfs_getbmap_format, + (struct getbmap *)arg+1); + if (error) + return -error; + + /* copy back header - only size of getbmap */ + if (copy_to_user(arg, &bmx, sizeof(struct getbmap))) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full) +{ + struct getbmapx __user *base = *ap; + + if (copy_to_user(base, bmv, sizeof(struct getbmapx))) + return XFS_ERROR(EFAULT); + + *ap += sizeof(struct getbmapx); + return 0; +} + +STATIC int +xfs_ioc_getbmapx( + struct xfs_inode *ip, + void __user *arg) +{ + struct getbmapx bmx; + int error; + + if (copy_from_user(&bmx, arg, sizeof(bmx))) + return -XFS_ERROR(EFAULT); + + if (bmx.bmv_count < 2) + return -XFS_ERROR(EINVAL); + + if (bmx.bmv_iflags & (~BMV_IF_VALID)) + return -XFS_ERROR(EINVAL); + + error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format, + (struct getbmapx *)arg+1); + if (error) + return -error; + + /* copy back header */ + if (copy_to_user(arg, &bmx, sizeof(struct getbmapx))) + return -XFS_ERROR(EFAULT); + + return 0; +} + +/* + * Note: some of the ioctl's return positive numbers as a + * byte count indicating success, such as readlink_by_handle. + * So we don't "sign flip" like most other routines. This means + * true errors need to be returned as a negative value. + */ +long +xfs_file_ioctl( + struct file *filp, + unsigned int cmd, + unsigned long p) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + void __user *arg = (void __user *)p; + int ioflags = 0; + int error; + + if (filp->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + trace_xfs_file_ioctl(ip); + + switch (cmd) { + case FITRIM: + return xfs_ioc_trim(mp, arg); + case XFS_IOC_ALLOCSP: + case XFS_IOC_FREESP: + case XFS_IOC_RESVSP: + case XFS_IOC_UNRESVSP: + case XFS_IOC_ALLOCSP64: + case XFS_IOC_FREESP64: + case XFS_IOC_RESVSP64: + case XFS_IOC_UNRESVSP64: + case XFS_IOC_ZERO_RANGE: { + xfs_flock64_t bf; + + if (copy_from_user(&bf, arg, sizeof(bf))) + return -XFS_ERROR(EFAULT); + return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf); + } + case XFS_IOC_DIOINFO: { + struct dioattr da; + xfs_buftarg_t *target = + XFS_IS_REALTIME_INODE(ip) ? + mp->m_rtdev_targp : mp->m_ddev_targp; + + da.d_mem = da.d_miniosz = 1 << target->bt_sshift; + da.d_maxiosz = INT_MAX & ~(da.d_miniosz - 1); + + if (copy_to_user(arg, &da, sizeof(da))) + return -XFS_ERROR(EFAULT); + return 0; + } + + case XFS_IOC_FSBULKSTAT_SINGLE: + case XFS_IOC_FSBULKSTAT: + case XFS_IOC_FSINUMBERS: + return xfs_ioc_bulkstat(mp, cmd, arg); + + case XFS_IOC_FSGEOMETRY_V1: + return xfs_ioc_fsgeometry_v1(mp, arg); + + case XFS_IOC_FSGEOMETRY: + return xfs_ioc_fsgeometry(mp, arg); + + case XFS_IOC_GETVERSION: + return put_user(inode->i_generation, (int __user *)arg); + + case XFS_IOC_FSGETXATTR: + return xfs_ioc_fsgetxattr(ip, 0, arg); + case XFS_IOC_FSGETXATTRA: + return xfs_ioc_fsgetxattr(ip, 1, arg); + case XFS_IOC_FSSETXATTR: + return xfs_ioc_fssetxattr(ip, filp, arg); + case XFS_IOC_GETXFLAGS: + return xfs_ioc_getxflags(ip, arg); + case XFS_IOC_SETXFLAGS: + return xfs_ioc_setxflags(ip, filp, arg); + + case XFS_IOC_FSSETDM: { + struct fsdmidata dmi; + + if (copy_from_user(&dmi, arg, sizeof(dmi))) + return -XFS_ERROR(EFAULT); + + error = xfs_set_dmattrs(ip, dmi.fsd_dmevmask, + dmi.fsd_dmstate); + return -error; + } + + case XFS_IOC_GETBMAP: + case XFS_IOC_GETBMAPA: + return xfs_ioc_getbmap(ip, ioflags, cmd, arg); + + case XFS_IOC_GETBMAPX: + return xfs_ioc_getbmapx(ip, arg); + + case XFS_IOC_FD_TO_HANDLE: + case XFS_IOC_PATH_TO_HANDLE: + case XFS_IOC_PATH_TO_FSHANDLE: { + xfs_fsop_handlereq_t hreq; + + if (copy_from_user(&hreq, arg, sizeof(hreq))) + return -XFS_ERROR(EFAULT); + return xfs_find_handle(cmd, &hreq); + } + case XFS_IOC_OPEN_BY_HANDLE: { + xfs_fsop_handlereq_t hreq; + + if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) + return -XFS_ERROR(EFAULT); + return xfs_open_by_handle(filp, &hreq); + } + case XFS_IOC_FSSETDM_BY_HANDLE: + return xfs_fssetdm_by_handle(filp, arg); + + case XFS_IOC_READLINK_BY_HANDLE: { + xfs_fsop_handlereq_t hreq; + + if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t))) + return -XFS_ERROR(EFAULT); + return xfs_readlink_by_handle(filp, &hreq); + } + case XFS_IOC_ATTRLIST_BY_HANDLE: + return xfs_attrlist_by_handle(filp, arg); + + case XFS_IOC_ATTRMULTI_BY_HANDLE: + return xfs_attrmulti_by_handle(filp, arg); + + case XFS_IOC_SWAPEXT: { + struct xfs_swapext sxp; + + if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t))) + return -XFS_ERROR(EFAULT); + error = xfs_swapext(&sxp); + return -error; + } + + case XFS_IOC_FSCOUNTS: { + xfs_fsop_counts_t out; + + error = xfs_fs_counts(mp, &out); + if (error) + return -error; + + if (copy_to_user(arg, &out, sizeof(out))) + return -XFS_ERROR(EFAULT); + return 0; + } + + case XFS_IOC_SET_RESBLKS: { + xfs_fsop_resblks_t inout; + __uint64_t in; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (mp->m_flags & XFS_MOUNT_RDONLY) + return -XFS_ERROR(EROFS); + + if (copy_from_user(&inout, arg, sizeof(inout))) + return -XFS_ERROR(EFAULT); + + /* input parameter is passed in resblks field of structure */ + in = inout.resblks; + error = xfs_reserve_blocks(mp, &in, &inout); + if (error) + return -error; + + if (copy_to_user(arg, &inout, sizeof(inout))) + return -XFS_ERROR(EFAULT); + return 0; + } + + case XFS_IOC_GET_RESBLKS: { + xfs_fsop_resblks_t out; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + error = xfs_reserve_blocks(mp, NULL, &out); + if (error) + return -error; + + if (copy_to_user(arg, &out, sizeof(out))) + return -XFS_ERROR(EFAULT); + + return 0; + } + + case XFS_IOC_FSGROWFSDATA: { + xfs_growfs_data_t in; + + if (copy_from_user(&in, arg, sizeof(in))) + return -XFS_ERROR(EFAULT); + + error = xfs_growfs_data(mp, &in); + return -error; + } + + case XFS_IOC_FSGROWFSLOG: { + xfs_growfs_log_t in; + + if (copy_from_user(&in, arg, sizeof(in))) + return -XFS_ERROR(EFAULT); + + error = xfs_growfs_log(mp, &in); + return -error; + } + + case XFS_IOC_FSGROWFSRT: { + xfs_growfs_rt_t in; + + if (copy_from_user(&in, arg, sizeof(in))) + return -XFS_ERROR(EFAULT); + + error = xfs_growfs_rt(mp, &in); + return -error; + } + + case XFS_IOC_GOINGDOWN: { + __uint32_t in; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (get_user(in, (__uint32_t __user *)arg)) + return -XFS_ERROR(EFAULT); + + error = xfs_fs_goingdown(mp, in); + return -error; + } + + case XFS_IOC_ERROR_INJECTION: { + xfs_error_injection_t in; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(&in, arg, sizeof(in))) + return -XFS_ERROR(EFAULT); + + error = xfs_errortag_add(in.errtag, mp); + return -error; + } + + case XFS_IOC_ERROR_CLEARALL: + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + error = xfs_errortag_clearall(mp, 1); + return -error; + + default: + return -ENOTTY; + } +} diff --git a/fs/xfs/xfs_ioctl.h b/fs/xfs/xfs_ioctl.h new file mode 100644 index 000000000000..d56173b34a2a --- /dev/null +++ b/fs/xfs/xfs_ioctl.h @@ -0,0 +1,85 @@ +/* + * Copyright (c) 2008 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_IOCTL_H__ +#define __XFS_IOCTL_H__ + +extern int +xfs_ioc_space( + struct xfs_inode *ip, + struct inode *inode, + struct file *filp, + int ioflags, + unsigned int cmd, + xfs_flock64_t *bf); + +extern int +xfs_find_handle( + unsigned int cmd, + xfs_fsop_handlereq_t *hreq); + +extern int +xfs_open_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq); + +extern int +xfs_readlink_by_handle( + struct file *parfilp, + xfs_fsop_handlereq_t *hreq); + +extern int +xfs_attrmulti_attr_get( + struct inode *inode, + unsigned char *name, + unsigned char __user *ubuf, + __uint32_t *len, + __uint32_t flags); + +extern int +xfs_attrmulti_attr_set( + struct inode *inode, + unsigned char *name, + const unsigned char __user *ubuf, + __uint32_t len, + __uint32_t flags); + +extern int +xfs_attrmulti_attr_remove( + struct inode *inode, + unsigned char *name, + __uint32_t flags); + +extern struct dentry * +xfs_handle_to_dentry( + struct file *parfilp, + void __user *uhandle, + u32 hlen); + +extern long +xfs_file_ioctl( + struct file *filp, + unsigned int cmd, + unsigned long p); + +extern long +xfs_file_compat_ioctl( + struct file *file, + unsigned int cmd, + unsigned long arg); + +#endif diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c new file mode 100644 index 000000000000..54e623bfbb85 --- /dev/null +++ b/fs/xfs/xfs_ioctl32.c @@ -0,0 +1,672 @@ +/* + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include +#include +#include +#include +#include +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_vnode.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_itable.h" +#include "xfs_error.h" +#include "xfs_dfrag.h" +#include "xfs_vnodeops.h" +#include "xfs_fsops.h" +#include "xfs_alloc.h" +#include "xfs_rtalloc.h" +#include "xfs_attr.h" +#include "xfs_ioctl.h" +#include "xfs_ioctl32.h" +#include "xfs_trace.h" + +#define _NATIVE_IOC(cmd, type) \ + _IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type)) + +#ifdef BROKEN_X86_ALIGNMENT +STATIC int +xfs_compat_flock64_copyin( + xfs_flock64_t *bf, + compat_xfs_flock64_t __user *arg32) +{ + if (get_user(bf->l_type, &arg32->l_type) || + get_user(bf->l_whence, &arg32->l_whence) || + get_user(bf->l_start, &arg32->l_start) || + get_user(bf->l_len, &arg32->l_len) || + get_user(bf->l_sysid, &arg32->l_sysid) || + get_user(bf->l_pid, &arg32->l_pid) || + copy_from_user(bf->l_pad, &arg32->l_pad, 4*sizeof(u32))) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_compat_ioc_fsgeometry_v1( + struct xfs_mount *mp, + compat_xfs_fsop_geom_v1_t __user *arg32) +{ + xfs_fsop_geom_t fsgeo; + int error; + + error = xfs_fs_geometry(mp, &fsgeo, 3); + if (error) + return -error; + /* The 32-bit variant simply has some padding at the end */ + if (copy_to_user(arg32, &fsgeo, sizeof(struct compat_xfs_fsop_geom_v1))) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_compat_growfs_data_copyin( + struct xfs_growfs_data *in, + compat_xfs_growfs_data_t __user *arg32) +{ + if (get_user(in->newblocks, &arg32->newblocks) || + get_user(in->imaxpct, &arg32->imaxpct)) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_compat_growfs_rt_copyin( + struct xfs_growfs_rt *in, + compat_xfs_growfs_rt_t __user *arg32) +{ + if (get_user(in->newblocks, &arg32->newblocks) || + get_user(in->extsize, &arg32->extsize)) + return -XFS_ERROR(EFAULT); + return 0; +} + +STATIC int +xfs_inumbers_fmt_compat( + void __user *ubuffer, + const xfs_inogrp_t *buffer, + long count, + long *written) +{ + compat_xfs_inogrp_t __user *p32 = ubuffer; + long i; + + for (i = 0; i < count; i++) { + if (put_user(buffer[i].xi_startino, &p32[i].xi_startino) || + put_user(buffer[i].xi_alloccount, &p32[i].xi_alloccount) || + put_user(buffer[i].xi_allocmask, &p32[i].xi_allocmask)) + return -XFS_ERROR(EFAULT); + } + *written = count * sizeof(*p32); + return 0; +} + +#else +#define xfs_inumbers_fmt_compat xfs_inumbers_fmt +#endif /* BROKEN_X86_ALIGNMENT */ + +STATIC int +xfs_ioctl32_bstime_copyin( + xfs_bstime_t *bstime, + compat_xfs_bstime_t __user *bstime32) +{ + compat_time_t sec32; /* tv_sec differs on 64 vs. 32 */ + + if (get_user(sec32, &bstime32->tv_sec) || + get_user(bstime->tv_nsec, &bstime32->tv_nsec)) + return -XFS_ERROR(EFAULT); + bstime->tv_sec = sec32; + return 0; +} + +/* xfs_bstat_t has differing alignment on intel, & bstime_t sizes everywhere */ +STATIC int +xfs_ioctl32_bstat_copyin( + xfs_bstat_t *bstat, + compat_xfs_bstat_t __user *bstat32) +{ + if (get_user(bstat->bs_ino, &bstat32->bs_ino) || + get_user(bstat->bs_mode, &bstat32->bs_mode) || + get_user(bstat->bs_nlink, &bstat32->bs_nlink) || + get_user(bstat->bs_uid, &bstat32->bs_uid) || + get_user(bstat->bs_gid, &bstat32->bs_gid) || + get_user(bstat->bs_rdev, &bstat32->bs_rdev) || + get_user(bstat->bs_blksize, &bstat32->bs_blksize) || + get_user(bstat->bs_size, &bstat32->bs_size) || + xfs_ioctl32_bstime_copyin(&bstat->bs_atime, &bstat32->bs_atime) || + xfs_ioctl32_bstime_copyin(&bstat->bs_mtime, &bstat32->bs_mtime) || + xfs_ioctl32_bstime_copyin(&bstat->bs_ctime, &bstat32->bs_ctime) || + get_user(bstat->bs_blocks, &bstat32->bs_size) || + get_user(bstat->bs_xflags, &bstat32->bs_size) || + get_user(bstat->bs_extsize, &bstat32->bs_extsize) || + get_user(bstat->bs_extents, &bstat32->bs_extents) || + get_user(bstat->bs_gen, &bstat32->bs_gen) || + get_user(bstat->bs_projid_lo, &bstat32->bs_projid_lo) || + get_user(bstat->bs_projid_hi, &bstat32->bs_projid_hi) || + get_user(bstat->bs_dmevmask, &bstat32->bs_dmevmask) || + get_user(bstat->bs_dmstate, &bstat32->bs_dmstate) || + get_user(bstat->bs_aextents, &bstat32->bs_aextents)) + return -XFS_ERROR(EFAULT); + return 0; +} + +/* XFS_IOC_FSBULKSTAT and friends */ + +STATIC int +xfs_bstime_store_compat( + compat_xfs_bstime_t __user *p32, + const xfs_bstime_t *p) +{ + __s32 sec32; + + sec32 = p->tv_sec; + if (put_user(sec32, &p32->tv_sec) || + put_user(p->tv_nsec, &p32->tv_nsec)) + return -XFS_ERROR(EFAULT); + return 0; +} + +/* Return 0 on success or positive error (to xfs_bulkstat()) */ +STATIC int +xfs_bulkstat_one_fmt_compat( + void __user *ubuffer, + int ubsize, + int *ubused, + const xfs_bstat_t *buffer) +{ + compat_xfs_bstat_t __user *p32 = ubuffer; + + if (ubsize < sizeof(*p32)) + return XFS_ERROR(ENOMEM); + + if (put_user(buffer->bs_ino, &p32->bs_ino) || + put_user(buffer->bs_mode, &p32->bs_mode) || + put_user(buffer->bs_nlink, &p32->bs_nlink) || + put_user(buffer->bs_uid, &p32->bs_uid) || + put_user(buffer->bs_gid, &p32->bs_gid) || + put_user(buffer->bs_rdev, &p32->bs_rdev) || + put_user(buffer->bs_blksize, &p32->bs_blksize) || + put_user(buffer->bs_size, &p32->bs_size) || + xfs_bstime_store_compat(&p32->bs_atime, &buffer->bs_atime) || + xfs_bstime_store_compat(&p32->bs_mtime, &buffer->bs_mtime) || + xfs_bstime_store_compat(&p32->bs_ctime, &buffer->bs_ctime) || + put_user(buffer->bs_blocks, &p32->bs_blocks) || + put_user(buffer->bs_xflags, &p32->bs_xflags) || + put_user(buffer->bs_extsize, &p32->bs_extsize) || + put_user(buffer->bs_extents, &p32->bs_extents) || + put_user(buffer->bs_gen, &p32->bs_gen) || + put_user(buffer->bs_projid, &p32->bs_projid) || + put_user(buffer->bs_projid_hi, &p32->bs_projid_hi) || + put_user(buffer->bs_dmevmask, &p32->bs_dmevmask) || + put_user(buffer->bs_dmstate, &p32->bs_dmstate) || + put_user(buffer->bs_aextents, &p32->bs_aextents)) + return XFS_ERROR(EFAULT); + if (ubused) + *ubused = sizeof(*p32); + return 0; +} + +STATIC int +xfs_bulkstat_one_compat( + xfs_mount_t *mp, /* mount point for filesystem */ + xfs_ino_t ino, /* inode number to get data for */ + void __user *buffer, /* buffer to place output in */ + int ubsize, /* size of buffer */ + int *ubused, /* bytes used by me */ + int *stat) /* BULKSTAT_RV_... */ +{ + return xfs_bulkstat_one_int(mp, ino, buffer, ubsize, + xfs_bulkstat_one_fmt_compat, + ubused, stat); +} + +/* copied from xfs_ioctl.c */ +STATIC int +xfs_compat_ioc_bulkstat( + xfs_mount_t *mp, + unsigned int cmd, + compat_xfs_fsop_bulkreq_t __user *p32) +{ + u32 addr; + xfs_fsop_bulkreq_t bulkreq; + int count; /* # of records returned */ + xfs_ino_t inlast; /* last inode number */ + int done; + int error; + + /* done = 1 if there are more stats to get and if bulkstat */ + /* should be called again (unused here, but used in dmapi) */ + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + + if (XFS_FORCED_SHUTDOWN(mp)) + return -XFS_ERROR(EIO); + + if (get_user(addr, &p32->lastip)) + return -XFS_ERROR(EFAULT); + bulkreq.lastip = compat_ptr(addr); + if (get_user(bulkreq.icount, &p32->icount) || + get_user(addr, &p32->ubuffer)) + return -XFS_ERROR(EFAULT); + bulkreq.ubuffer = compat_ptr(addr); + if (get_user(addr, &p32->ocount)) + return -XFS_ERROR(EFAULT); + bulkreq.ocount = compat_ptr(addr); + + if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64))) + return -XFS_ERROR(EFAULT); + + if ((count = bulkreq.icount) <= 0) + return -XFS_ERROR(EINVAL); + + if (bulkreq.ubuffer == NULL) + return -XFS_ERROR(EINVAL); + + if (cmd == XFS_IOC_FSINUMBERS_32) { + error = xfs_inumbers(mp, &inlast, &count, + bulkreq.ubuffer, xfs_inumbers_fmt_compat); + } else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE_32) { + int res; + + error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer, + sizeof(compat_xfs_bstat_t), 0, &res); + } else if (cmd == XFS_IOC_FSBULKSTAT_32) { + error = xfs_bulkstat(mp, &inlast, &count, + xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t), + bulkreq.ubuffer, &done); + } else + error = XFS_ERROR(EINVAL); + if (error) + return -error; + + if (bulkreq.ocount != NULL) { + if (copy_to_user(bulkreq.lastip, &inlast, + sizeof(xfs_ino_t))) + return -XFS_ERROR(EFAULT); + + if (copy_to_user(bulkreq.ocount, &count, sizeof(count))) + return -XFS_ERROR(EFAULT); + } + + return 0; +} + +STATIC int +xfs_compat_handlereq_copyin( + xfs_fsop_handlereq_t *hreq, + compat_xfs_fsop_handlereq_t __user *arg32) +{ + compat_xfs_fsop_handlereq_t hreq32; + + if (copy_from_user(&hreq32, arg32, sizeof(compat_xfs_fsop_handlereq_t))) + return -XFS_ERROR(EFAULT); + + hreq->fd = hreq32.fd; + hreq->path = compat_ptr(hreq32.path); + hreq->oflags = hreq32.oflags; + hreq->ihandle = compat_ptr(hreq32.ihandle); + hreq->ihandlen = hreq32.ihandlen; + hreq->ohandle = compat_ptr(hreq32.ohandle); + hreq->ohandlen = compat_ptr(hreq32.ohandlen); + + return 0; +} + +STATIC struct dentry * +xfs_compat_handlereq_to_dentry( + struct file *parfilp, + compat_xfs_fsop_handlereq_t *hreq) +{ + return xfs_handle_to_dentry(parfilp, + compat_ptr(hreq->ihandle), hreq->ihandlen); +} + +STATIC int +xfs_compat_attrlist_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + attrlist_cursor_kern_t *cursor; + compat_xfs_fsop_attrlist_handlereq_t al_hreq; + struct dentry *dentry; + char *kbuf; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&al_hreq, arg, + sizeof(compat_xfs_fsop_attrlist_handlereq_t))) + return -XFS_ERROR(EFAULT); + if (al_hreq.buflen > XATTR_LIST_MAX) + return -XFS_ERROR(EINVAL); + + /* + * Reject flags, only allow namespaces. + */ + if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE)) + return -XFS_ERROR(EINVAL); + + dentry = xfs_compat_handlereq_to_dentry(parfilp, &al_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + error = -ENOMEM; + kbuf = kmalloc(al_hreq.buflen, GFP_KERNEL); + if (!kbuf) + goto out_dput; + + cursor = (attrlist_cursor_kern_t *)&al_hreq.pos; + error = -xfs_attr_list(XFS_I(dentry->d_inode), kbuf, al_hreq.buflen, + al_hreq.flags, cursor); + if (error) + goto out_kfree; + + if (copy_to_user(compat_ptr(al_hreq.buffer), kbuf, al_hreq.buflen)) + error = -EFAULT; + + out_kfree: + kfree(kbuf); + out_dput: + dput(dentry); + return error; +} + +STATIC int +xfs_compat_attrmulti_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + compat_xfs_attr_multiop_t *ops; + compat_xfs_fsop_attrmulti_handlereq_t am_hreq; + struct dentry *dentry; + unsigned int i, size; + unsigned char *attr_name; + + if (!capable(CAP_SYS_ADMIN)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&am_hreq, arg, + sizeof(compat_xfs_fsop_attrmulti_handlereq_t))) + return -XFS_ERROR(EFAULT); + + /* overflow check */ + if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t)) + return -E2BIG; + + dentry = xfs_compat_handlereq_to_dentry(parfilp, &am_hreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + error = E2BIG; + size = am_hreq.opcount * sizeof(compat_xfs_attr_multiop_t); + if (!size || size > 16 * PAGE_SIZE) + goto out_dput; + + ops = memdup_user(compat_ptr(am_hreq.ops), size); + if (IS_ERR(ops)) { + error = PTR_ERR(ops); + goto out_dput; + } + + attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL); + if (!attr_name) + goto out_kfree_ops; + + error = 0; + for (i = 0; i < am_hreq.opcount; i++) { + ops[i].am_error = strncpy_from_user((char *)attr_name, + compat_ptr(ops[i].am_attrname), + MAXNAMELEN); + if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN) + error = -ERANGE; + if (ops[i].am_error < 0) + break; + + switch (ops[i].am_opcode) { + case ATTR_OP_GET: + ops[i].am_error = xfs_attrmulti_attr_get( + dentry->d_inode, attr_name, + compat_ptr(ops[i].am_attrvalue), + &ops[i].am_length, ops[i].am_flags); + break; + case ATTR_OP_SET: + ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + if (ops[i].am_error) + break; + ops[i].am_error = xfs_attrmulti_attr_set( + dentry->d_inode, attr_name, + compat_ptr(ops[i].am_attrvalue), + ops[i].am_length, ops[i].am_flags); + mnt_drop_write(parfilp->f_path.mnt); + break; + case ATTR_OP_REMOVE: + ops[i].am_error = mnt_want_write(parfilp->f_path.mnt); + if (ops[i].am_error) + break; + ops[i].am_error = xfs_attrmulti_attr_remove( + dentry->d_inode, attr_name, + ops[i].am_flags); + mnt_drop_write(parfilp->f_path.mnt); + break; + default: + ops[i].am_error = EINVAL; + } + } + + if (copy_to_user(compat_ptr(am_hreq.ops), ops, size)) + error = XFS_ERROR(EFAULT); + + kfree(attr_name); + out_kfree_ops: + kfree(ops); + out_dput: + dput(dentry); + return -error; +} + +STATIC int +xfs_compat_fssetdm_by_handle( + struct file *parfilp, + void __user *arg) +{ + int error; + struct fsdmidata fsd; + compat_xfs_fsop_setdm_handlereq_t dmhreq; + struct dentry *dentry; + + if (!capable(CAP_MKNOD)) + return -XFS_ERROR(EPERM); + if (copy_from_user(&dmhreq, arg, + sizeof(compat_xfs_fsop_setdm_handlereq_t))) + return -XFS_ERROR(EFAULT); + + dentry = xfs_compat_handlereq_to_dentry(parfilp, &dmhreq.hreq); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + + if (IS_IMMUTABLE(dentry->d_inode) || IS_APPEND(dentry->d_inode)) { + error = -XFS_ERROR(EPERM); + goto out; + } + + if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) { + error = -XFS_ERROR(EFAULT); + goto out; + } + + error = -xfs_set_dmattrs(XFS_I(dentry->d_inode), fsd.fsd_dmevmask, + fsd.fsd_dmstate); + +out: + dput(dentry); + return error; +} + +long +xfs_file_compat_ioctl( + struct file *filp, + unsigned cmd, + unsigned long p) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + void __user *arg = (void __user *)p; + int ioflags = 0; + int error; + + if (filp->f_mode & FMODE_NOCMTIME) + ioflags |= IO_INVIS; + + trace_xfs_file_compat_ioctl(ip); + + switch (cmd) { + /* No size or alignment issues on any arch */ + case XFS_IOC_DIOINFO: + case XFS_IOC_FSGEOMETRY: + case XFS_IOC_FSGETXATTR: + case XFS_IOC_FSSETXATTR: + case XFS_IOC_FSGETXATTRA: + case XFS_IOC_FSSETDM: + case XFS_IOC_GETBMAP: + case XFS_IOC_GETBMAPA: + case XFS_IOC_GETBMAPX: + case XFS_IOC_FSCOUNTS: + case XFS_IOC_SET_RESBLKS: + case XFS_IOC_GET_RESBLKS: + case XFS_IOC_FSGROWFSLOG: + case XFS_IOC_GOINGDOWN: + case XFS_IOC_ERROR_INJECTION: + case XFS_IOC_ERROR_CLEARALL: + return xfs_file_ioctl(filp, cmd, p); +#ifndef BROKEN_X86_ALIGNMENT + /* These are handled fine if no alignment issues */ + case XFS_IOC_ALLOCSP: + case XFS_IOC_FREESP: + case XFS_IOC_RESVSP: + case XFS_IOC_UNRESVSP: + case XFS_IOC_ALLOCSP64: + case XFS_IOC_FREESP64: + case XFS_IOC_RESVSP64: + case XFS_IOC_UNRESVSP64: + case XFS_IOC_FSGEOMETRY_V1: + case XFS_IOC_FSGROWFSDATA: + case XFS_IOC_FSGROWFSRT: + case XFS_IOC_ZERO_RANGE: + return xfs_file_ioctl(filp, cmd, p); +#else + case XFS_IOC_ALLOCSP_32: + case XFS_IOC_FREESP_32: + case XFS_IOC_ALLOCSP64_32: + case XFS_IOC_FREESP64_32: + case XFS_IOC_RESVSP_32: + case XFS_IOC_UNRESVSP_32: + case XFS_IOC_RESVSP64_32: + case XFS_IOC_UNRESVSP64_32: + case XFS_IOC_ZERO_RANGE_32: { + struct xfs_flock64 bf; + + if (xfs_compat_flock64_copyin(&bf, arg)) + return -XFS_ERROR(EFAULT); + cmd = _NATIVE_IOC(cmd, struct xfs_flock64); + return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf); + } + case XFS_IOC_FSGEOMETRY_V1_32: + return xfs_compat_ioc_fsgeometry_v1(mp, arg); + case XFS_IOC_FSGROWFSDATA_32: { + struct xfs_growfs_data in; + + if (xfs_compat_growfs_data_copyin(&in, arg)) + return -XFS_ERROR(EFAULT); + error = xfs_growfs_data(mp, &in); + return -error; + } + case XFS_IOC_FSGROWFSRT_32: { + struct xfs_growfs_rt in; + + if (xfs_compat_growfs_rt_copyin(&in, arg)) + return -XFS_ERROR(EFAULT); + error = xfs_growfs_rt(mp, &in); + return -error; + } +#endif + /* long changes size, but xfs only copiese out 32 bits */ + case XFS_IOC_GETXFLAGS_32: + case XFS_IOC_SETXFLAGS_32: + case XFS_IOC_GETVERSION_32: + cmd = _NATIVE_IOC(cmd, long); + return xfs_file_ioctl(filp, cmd, p); + case XFS_IOC_SWAPEXT_32: { + struct xfs_swapext sxp; + struct compat_xfs_swapext __user *sxu = arg; + + /* Bulk copy in up to the sx_stat field, then copy bstat */ + if (copy_from_user(&sxp, sxu, + offsetof(struct xfs_swapext, sx_stat)) || + xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat)) + return -XFS_ERROR(EFAULT); + error = xfs_swapext(&sxp); + return -error; + } + case XFS_IOC_FSBULKSTAT_32: + case XFS_IOC_FSBULKSTAT_SINGLE_32: + case XFS_IOC_FSINUMBERS_32: + return xfs_compat_ioc_bulkstat(mp, cmd, arg); + case XFS_IOC_FD_TO_HANDLE_32: + case XFS_IOC_PATH_TO_HANDLE_32: + case XFS_IOC_PATH_TO_FSHANDLE_32: { + struct xfs_fsop_handlereq hreq; + + if (xfs_compat_handlereq_copyin(&hreq, arg)) + return -XFS_ERROR(EFAULT); + cmd = _NATIVE_IOC(cmd, struct xfs_fsop_handlereq); + return xfs_find_handle(cmd, &hreq); + } + case XFS_IOC_OPEN_BY_HANDLE_32: { + struct xfs_fsop_handlereq hreq; + + if (xfs_compat_handlereq_copyin(&hreq, arg)) + return -XFS_ERROR(EFAULT); + return xfs_open_by_handle(filp, &hreq); + } + case XFS_IOC_READLINK_BY_HANDLE_32: { + struct xfs_fsop_handlereq hreq; + + if (xfs_compat_handlereq_copyin(&hreq, arg)) + return -XFS_ERROR(EFAULT); + return xfs_readlink_by_handle(filp, &hreq); + } + case XFS_IOC_ATTRLIST_BY_HANDLE_32: + return xfs_compat_attrlist_by_handle(filp, arg); + case XFS_IOC_ATTRMULTI_BY_HANDLE_32: + return xfs_compat_attrmulti_by_handle(filp, arg); + case XFS_IOC_FSSETDM_BY_HANDLE_32: + return xfs_compat_fssetdm_by_handle(filp, arg); + default: + return -XFS_ERROR(ENOIOCTLCMD); + } +} diff --git a/fs/xfs/xfs_ioctl32.h b/fs/xfs/xfs_ioctl32.h new file mode 100644 index 000000000000..80f4060e8970 --- /dev/null +++ b/fs/xfs/xfs_ioctl32.h @@ -0,0 +1,237 @@ +/* + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_IOCTL32_H__ +#define __XFS_IOCTL32_H__ + +#include + +/* + * on 32-bit arches, ioctl argument structures may have different sizes + * and/or alignment. We define compat structures which match the + * 32-bit sizes/alignments here, and their associated ioctl numbers. + * + * xfs_ioctl32.c contains routines to copy these structures in and out. + */ + +/* stock kernel-level ioctls we support */ +#define XFS_IOC_GETXFLAGS_32 FS_IOC32_GETFLAGS +#define XFS_IOC_SETXFLAGS_32 FS_IOC32_SETFLAGS +#define XFS_IOC_GETVERSION_32 FS_IOC32_GETVERSION + +/* + * On intel, even if sizes match, alignment and/or padding may differ. + */ +#if defined(CONFIG_IA64) || defined(CONFIG_X86_64) +#define BROKEN_X86_ALIGNMENT +#define __compat_packed __attribute__((packed)) +#else +#define __compat_packed +#endif + +typedef struct compat_xfs_bstime { + compat_time_t tv_sec; /* seconds */ + __s32 tv_nsec; /* and nanoseconds */ +} compat_xfs_bstime_t; + +typedef struct compat_xfs_bstat { + __u64 bs_ino; /* inode number */ + __u16 bs_mode; /* type and mode */ + __u16 bs_nlink; /* number of links */ + __u32 bs_uid; /* user id */ + __u32 bs_gid; /* group id */ + __u32 bs_rdev; /* device value */ + __s32 bs_blksize; /* block size */ + __s64 bs_size; /* file size */ + compat_xfs_bstime_t bs_atime; /* access time */ + compat_xfs_bstime_t bs_mtime; /* modify time */ + compat_xfs_bstime_t bs_ctime; /* inode change time */ + int64_t bs_blocks; /* number of blocks */ + __u32 bs_xflags; /* extended flags */ + __s32 bs_extsize; /* extent size */ + __s32 bs_extents; /* number of extents */ + __u32 bs_gen; /* generation count */ + __u16 bs_projid_lo; /* lower part of project id */ +#define bs_projid bs_projid_lo /* (previously just bs_projid) */ + __u16 bs_projid_hi; /* high part of project id */ + unsigned char bs_pad[12]; /* pad space, unused */ + __u32 bs_dmevmask; /* DMIG event mask */ + __u16 bs_dmstate; /* DMIG state info */ + __u16 bs_aextents; /* attribute number of extents */ +} __compat_packed compat_xfs_bstat_t; + +typedef struct compat_xfs_fsop_bulkreq { + compat_uptr_t lastip; /* last inode # pointer */ + __s32 icount; /* count of entries in buffer */ + compat_uptr_t ubuffer; /* user buffer for inode desc. */ + compat_uptr_t ocount; /* output count pointer */ +} compat_xfs_fsop_bulkreq_t; + +#define XFS_IOC_FSBULKSTAT_32 \ + _IOWR('X', 101, struct compat_xfs_fsop_bulkreq) +#define XFS_IOC_FSBULKSTAT_SINGLE_32 \ + _IOWR('X', 102, struct compat_xfs_fsop_bulkreq) +#define XFS_IOC_FSINUMBERS_32 \ + _IOWR('X', 103, struct compat_xfs_fsop_bulkreq) + +typedef struct compat_xfs_fsop_handlereq { + __u32 fd; /* fd for FD_TO_HANDLE */ + compat_uptr_t path; /* user pathname */ + __u32 oflags; /* open flags */ + compat_uptr_t ihandle; /* user supplied handle */ + __u32 ihandlen; /* user supplied length */ + compat_uptr_t ohandle; /* user buffer for handle */ + compat_uptr_t ohandlen; /* user buffer length */ +} compat_xfs_fsop_handlereq_t; + +#define XFS_IOC_PATH_TO_FSHANDLE_32 \ + _IOWR('X', 104, struct compat_xfs_fsop_handlereq) +#define XFS_IOC_PATH_TO_HANDLE_32 \ + _IOWR('X', 105, struct compat_xfs_fsop_handlereq) +#define XFS_IOC_FD_TO_HANDLE_32 \ + _IOWR('X', 106, struct compat_xfs_fsop_handlereq) +#define XFS_IOC_OPEN_BY_HANDLE_32 \ + _IOWR('X', 107, struct compat_xfs_fsop_handlereq) +#define XFS_IOC_READLINK_BY_HANDLE_32 \ + _IOWR('X', 108, struct compat_xfs_fsop_handlereq) + +/* The bstat field in the swapext struct needs translation */ +typedef struct compat_xfs_swapext { + __int64_t sx_version; /* version */ + __int64_t sx_fdtarget; /* fd of target file */ + __int64_t sx_fdtmp; /* fd of tmp file */ + xfs_off_t sx_offset; /* offset into file */ + xfs_off_t sx_length; /* leng from offset */ + char sx_pad[16]; /* pad space, unused */ + compat_xfs_bstat_t sx_stat; /* stat of target b4 copy */ +} __compat_packed compat_xfs_swapext_t; + +#define XFS_IOC_SWAPEXT_32 _IOWR('X', 109, struct compat_xfs_swapext) + +typedef struct compat_xfs_fsop_attrlist_handlereq { + struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */ + struct xfs_attrlist_cursor pos; /* opaque cookie, list offset */ + __u32 flags; /* which namespace to use */ + __u32 buflen; /* length of buffer supplied */ + compat_uptr_t buffer; /* returned names */ +} __compat_packed compat_xfs_fsop_attrlist_handlereq_t; + +/* Note: actually this is read/write */ +#define XFS_IOC_ATTRLIST_BY_HANDLE_32 \ + _IOW('X', 122, struct compat_xfs_fsop_attrlist_handlereq) + +/* am_opcodes defined in xfs_fs.h */ +typedef struct compat_xfs_attr_multiop { + __u32 am_opcode; + __s32 am_error; + compat_uptr_t am_attrname; + compat_uptr_t am_attrvalue; + __u32 am_length; + __u32 am_flags; +} compat_xfs_attr_multiop_t; + +typedef struct compat_xfs_fsop_attrmulti_handlereq { + struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */ + __u32 opcount;/* count of following multiop */ + /* ptr to compat_xfs_attr_multiop */ + compat_uptr_t ops; /* attr_multi data */ +} compat_xfs_fsop_attrmulti_handlereq_t; + +#define XFS_IOC_ATTRMULTI_BY_HANDLE_32 \ + _IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq) + +typedef struct compat_xfs_fsop_setdm_handlereq { + struct compat_xfs_fsop_handlereq hreq; /* handle information */ + /* ptr to struct fsdmidata */ + compat_uptr_t data; /* DMAPI data */ +} compat_xfs_fsop_setdm_handlereq_t; + +#define XFS_IOC_FSSETDM_BY_HANDLE_32 \ + _IOW('X', 121, struct compat_xfs_fsop_setdm_handlereq) + +#ifdef BROKEN_X86_ALIGNMENT +/* on ia32 l_start is on a 32-bit boundary */ +typedef struct compat_xfs_flock64 { + __s16 l_type; + __s16 l_whence; + __s64 l_start __attribute__((packed)); + /* len == 0 means until end of file */ + __s64 l_len __attribute__((packed)); + __s32 l_sysid; + __u32 l_pid; + __s32 l_pad[4]; /* reserve area */ +} compat_xfs_flock64_t; + +#define XFS_IOC_ALLOCSP_32 _IOW('X', 10, struct compat_xfs_flock64) +#define XFS_IOC_FREESP_32 _IOW('X', 11, struct compat_xfs_flock64) +#define XFS_IOC_ALLOCSP64_32 _IOW('X', 36, struct compat_xfs_flock64) +#define XFS_IOC_FREESP64_32 _IOW('X', 37, struct compat_xfs_flock64) +#define XFS_IOC_RESVSP_32 _IOW('X', 40, struct compat_xfs_flock64) +#define XFS_IOC_UNRESVSP_32 _IOW('X', 41, struct compat_xfs_flock64) +#define XFS_IOC_RESVSP64_32 _IOW('X', 42, struct compat_xfs_flock64) +#define XFS_IOC_UNRESVSP64_32 _IOW('X', 43, struct compat_xfs_flock64) +#define XFS_IOC_ZERO_RANGE_32 _IOW('X', 57, struct compat_xfs_flock64) + +typedef struct compat_xfs_fsop_geom_v1 { + __u32 blocksize; /* filesystem (data) block size */ + __u32 rtextsize; /* realtime extent size */ + __u32 agblocks; /* fsblocks in an AG */ + __u32 agcount; /* number of allocation groups */ + __u32 logblocks; /* fsblocks in the log */ + __u32 sectsize; /* (data) sector size, bytes */ + __u32 inodesize; /* inode size in bytes */ + __u32 imaxpct; /* max allowed inode space(%) */ + __u64 datablocks; /* fsblocks in data subvolume */ + __u64 rtblocks; /* fsblocks in realtime subvol */ + __u64 rtextents; /* rt extents in realtime subvol*/ + __u64 logstart; /* starting fsblock of the log */ + unsigned char uuid[16]; /* unique id of the filesystem */ + __u32 sunit; /* stripe unit, fsblocks */ + __u32 swidth; /* stripe width, fsblocks */ + __s32 version; /* structure version */ + __u32 flags; /* superblock version flags */ + __u32 logsectsize; /* log sector size, bytes */ + __u32 rtsectsize; /* realtime sector size, bytes */ + __u32 dirblocksize; /* directory block size, bytes */ +} __attribute__((packed)) compat_xfs_fsop_geom_v1_t; + +#define XFS_IOC_FSGEOMETRY_V1_32 \ + _IOR('X', 100, struct compat_xfs_fsop_geom_v1) + +typedef struct compat_xfs_inogrp { + __u64 xi_startino; /* starting inode number */ + __s32 xi_alloccount; /* # bits set in allocmask */ + __u64 xi_allocmask; /* mask of allocated inodes */ +} __attribute__((packed)) compat_xfs_inogrp_t; + +/* These growfs input structures have padding on the end, so must translate */ +typedef struct compat_xfs_growfs_data { + __u64 newblocks; /* new data subvol size, fsblocks */ + __u32 imaxpct; /* new inode space percentage limit */ +} __attribute__((packed)) compat_xfs_growfs_data_t; + +typedef struct compat_xfs_growfs_rt { + __u64 newblocks; /* new realtime size, fsblocks */ + __u32 extsize; /* new realtime extent size, fsblocks */ +} __attribute__((packed)) compat_xfs_growfs_rt_t; + +#define XFS_IOC_FSGROWFSDATA_32 _IOW('X', 110, struct compat_xfs_growfs_data) +#define XFS_IOC_FSGROWFSRT_32 _IOW('X', 112, struct compat_xfs_growfs_rt) + +#endif /* BROKEN_X86_ALIGNMENT */ + +#endif /* __XFS_IOCTL32_H__ */ diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c new file mode 100644 index 000000000000..b9c172b3fbbe --- /dev/null +++ b/fs/xfs/xfs_iops.c @@ -0,0 +1,1210 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_acl.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_itable.h" +#include "xfs_rw.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_utils.h" +#include "xfs_vnodeops.h" +#include "xfs_inode_item.h" +#include "xfs_trace.h" + +#include +#include +#include +#include +#include +#include +#include + +/* + * Bring the timestamps in the XFS inode uptodate. + * + * Used before writing the inode to disk. + */ +void +xfs_synchronize_times( + xfs_inode_t *ip) +{ + struct inode *inode = VFS_I(ip); + + ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec; + ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec; + ip->i_d.di_ctime.t_sec = (__int32_t)inode->i_ctime.tv_sec; + ip->i_d.di_ctime.t_nsec = (__int32_t)inode->i_ctime.tv_nsec; + ip->i_d.di_mtime.t_sec = (__int32_t)inode->i_mtime.tv_sec; + ip->i_d.di_mtime.t_nsec = (__int32_t)inode->i_mtime.tv_nsec; +} + +/* + * If the linux inode is valid, mark it dirty. + * Used when committing a dirty inode into a transaction so that + * the inode will get written back by the linux code + */ +void +xfs_mark_inode_dirty_sync( + xfs_inode_t *ip) +{ + struct inode *inode = VFS_I(ip); + + if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) + mark_inode_dirty_sync(inode); +} + +void +xfs_mark_inode_dirty( + xfs_inode_t *ip) +{ + struct inode *inode = VFS_I(ip); + + if (!(inode->i_state & (I_WILL_FREE|I_FREEING))) + mark_inode_dirty(inode); +} + +/* + * Hook in SELinux. This is not quite correct yet, what we really need + * here (as we do for default ACLs) is a mechanism by which creation of + * these attrs can be journalled at inode creation time (along with the + * inode, of course, such that log replay can't cause these to be lost). + */ +STATIC int +xfs_init_security( + struct inode *inode, + struct inode *dir, + const struct qstr *qstr) +{ + struct xfs_inode *ip = XFS_I(inode); + size_t length; + void *value; + unsigned char *name; + int error; + + error = security_inode_init_security(inode, dir, qstr, (char **)&name, + &value, &length); + if (error) { + if (error == -EOPNOTSUPP) + return 0; + return -error; + } + + error = xfs_attr_set(ip, name, value, length, ATTR_SECURE); + + kfree(name); + kfree(value); + return error; +} + +static void +xfs_dentry_to_name( + struct xfs_name *namep, + struct dentry *dentry) +{ + namep->name = dentry->d_name.name; + namep->len = dentry->d_name.len; +} + +STATIC void +xfs_cleanup_inode( + struct inode *dir, + struct inode *inode, + struct dentry *dentry) +{ + struct xfs_name teardown; + + /* Oh, the horror. + * If we can't add the ACL or we fail in + * xfs_init_security we must back out. + * ENOSPC can hit here, among other things. + */ + xfs_dentry_to_name(&teardown, dentry); + + xfs_remove(XFS_I(dir), &teardown, XFS_I(inode)); + iput(inode); +} + +STATIC int +xfs_vn_mknod( + struct inode *dir, + struct dentry *dentry, + int mode, + dev_t rdev) +{ + struct inode *inode; + struct xfs_inode *ip = NULL; + struct posix_acl *default_acl = NULL; + struct xfs_name name; + int error; + + /* + * Irix uses Missed'em'V split, but doesn't want to see + * the upper 5 bits of (14bit) major. + */ + if (S_ISCHR(mode) || S_ISBLK(mode)) { + if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff)) + return -EINVAL; + rdev = sysv_encode_dev(rdev); + } else { + rdev = 0; + } + + if (IS_POSIXACL(dir)) { + default_acl = xfs_get_acl(dir, ACL_TYPE_DEFAULT); + if (IS_ERR(default_acl)) + return PTR_ERR(default_acl); + + if (!default_acl) + mode &= ~current_umask(); + } + + xfs_dentry_to_name(&name, dentry); + error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); + if (unlikely(error)) + goto out_free_acl; + + inode = VFS_I(ip); + + error = xfs_init_security(inode, dir, &dentry->d_name); + if (unlikely(error)) + goto out_cleanup_inode; + + if (default_acl) { + error = -xfs_inherit_acl(inode, default_acl); + default_acl = NULL; + if (unlikely(error)) + goto out_cleanup_inode; + } + + + d_instantiate(dentry, inode); + return -error; + + out_cleanup_inode: + xfs_cleanup_inode(dir, inode, dentry); + out_free_acl: + posix_acl_release(default_acl); + return -error; +} + +STATIC int +xfs_vn_create( + struct inode *dir, + struct dentry *dentry, + int mode, + struct nameidata *nd) +{ + return xfs_vn_mknod(dir, dentry, mode, 0); +} + +STATIC int +xfs_vn_mkdir( + struct inode *dir, + struct dentry *dentry, + int mode) +{ + return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0); +} + +STATIC struct dentry * +xfs_vn_lookup( + struct inode *dir, + struct dentry *dentry, + struct nameidata *nd) +{ + struct xfs_inode *cip; + struct xfs_name name; + int error; + + if (dentry->d_name.len >= MAXNAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + xfs_dentry_to_name(&name, dentry); + error = xfs_lookup(XFS_I(dir), &name, &cip, NULL); + if (unlikely(error)) { + if (unlikely(error != ENOENT)) + return ERR_PTR(-error); + d_add(dentry, NULL); + return NULL; + } + + return d_splice_alias(VFS_I(cip), dentry); +} + +STATIC struct dentry * +xfs_vn_ci_lookup( + struct inode *dir, + struct dentry *dentry, + struct nameidata *nd) +{ + struct xfs_inode *ip; + struct xfs_name xname; + struct xfs_name ci_name; + struct qstr dname; + int error; + + if (dentry->d_name.len >= MAXNAMELEN) + return ERR_PTR(-ENAMETOOLONG); + + xfs_dentry_to_name(&xname, dentry); + error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name); + if (unlikely(error)) { + if (unlikely(error != ENOENT)) + return ERR_PTR(-error); + /* + * call d_add(dentry, NULL) here when d_drop_negative_children + * is called in xfs_vn_mknod (ie. allow negative dentries + * with CI filesystems). + */ + return NULL; + } + + /* if exact match, just splice and exit */ + if (!ci_name.name) + return d_splice_alias(VFS_I(ip), dentry); + + /* else case-insensitive match... */ + dname.name = ci_name.name; + dname.len = ci_name.len; + dentry = d_add_ci(dentry, VFS_I(ip), &dname); + kmem_free(ci_name.name); + return dentry; +} + +STATIC int +xfs_vn_link( + struct dentry *old_dentry, + struct inode *dir, + struct dentry *dentry) +{ + struct inode *inode = old_dentry->d_inode; + struct xfs_name name; + int error; + + xfs_dentry_to_name(&name, dentry); + + error = xfs_link(XFS_I(dir), XFS_I(inode), &name); + if (unlikely(error)) + return -error; + + ihold(inode); + d_instantiate(dentry, inode); + return 0; +} + +STATIC int +xfs_vn_unlink( + struct inode *dir, + struct dentry *dentry) +{ + struct xfs_name name; + int error; + + xfs_dentry_to_name(&name, dentry); + + error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode)); + if (error) + return error; + + /* + * With unlink, the VFS makes the dentry "negative": no inode, + * but still hashed. This is incompatible with case-insensitive + * mode, so invalidate (unhash) the dentry in CI-mode. + */ + if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb)) + d_invalidate(dentry); + return 0; +} + +STATIC int +xfs_vn_symlink( + struct inode *dir, + struct dentry *dentry, + const char *symname) +{ + struct inode *inode; + struct xfs_inode *cip = NULL; + struct xfs_name name; + int error; + mode_t mode; + + mode = S_IFLNK | + (irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO); + xfs_dentry_to_name(&name, dentry); + + error = xfs_symlink(XFS_I(dir), &name, symname, mode, &cip); + if (unlikely(error)) + goto out; + + inode = VFS_I(cip); + + error = xfs_init_security(inode, dir, &dentry->d_name); + if (unlikely(error)) + goto out_cleanup_inode; + + d_instantiate(dentry, inode); + return 0; + + out_cleanup_inode: + xfs_cleanup_inode(dir, inode, dentry); + out: + return -error; +} + +STATIC int +xfs_vn_rename( + struct inode *odir, + struct dentry *odentry, + struct inode *ndir, + struct dentry *ndentry) +{ + struct inode *new_inode = ndentry->d_inode; + struct xfs_name oname; + struct xfs_name nname; + + xfs_dentry_to_name(&oname, odentry); + xfs_dentry_to_name(&nname, ndentry); + + return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode), + XFS_I(ndir), &nname, new_inode ? + XFS_I(new_inode) : NULL); +} + +/* + * careful here - this function can get called recursively, so + * we need to be very careful about how much stack we use. + * uio is kmalloced for this reason... + */ +STATIC void * +xfs_vn_follow_link( + struct dentry *dentry, + struct nameidata *nd) +{ + char *link; + int error = -ENOMEM; + + link = kmalloc(MAXPATHLEN+1, GFP_KERNEL); + if (!link) + goto out_err; + + error = -xfs_readlink(XFS_I(dentry->d_inode), link); + if (unlikely(error)) + goto out_kfree; + + nd_set_link(nd, link); + return NULL; + + out_kfree: + kfree(link); + out_err: + nd_set_link(nd, ERR_PTR(error)); + return NULL; +} + +STATIC void +xfs_vn_put_link( + struct dentry *dentry, + struct nameidata *nd, + void *p) +{ + char *s = nd_get_link(nd); + + if (!IS_ERR(s)) + kfree(s); +} + +STATIC int +xfs_vn_getattr( + struct vfsmount *mnt, + struct dentry *dentry, + struct kstat *stat) +{ + struct inode *inode = dentry->d_inode; + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + + trace_xfs_getattr(ip); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + stat->size = XFS_ISIZE(ip); + stat->dev = inode->i_sb->s_dev; + stat->mode = ip->i_d.di_mode; + stat->nlink = ip->i_d.di_nlink; + stat->uid = ip->i_d.di_uid; + stat->gid = ip->i_d.di_gid; + stat->ino = ip->i_ino; + stat->atime = inode->i_atime; + stat->mtime = inode->i_mtime; + stat->ctime = inode->i_ctime; + stat->blocks = + XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks); + + + switch (inode->i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + stat->blksize = BLKDEV_IOSIZE; + stat->rdev = MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, + sysv_minor(ip->i_df.if_u2.if_rdev)); + break; + default: + if (XFS_IS_REALTIME_INODE(ip)) { + /* + * If the file blocks are being allocated from a + * realtime volume, then return the inode's realtime + * extent size or the realtime volume's extent size. + */ + stat->blksize = + xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog; + } else + stat->blksize = xfs_preferred_iosize(mp); + stat->rdev = 0; + break; + } + + return 0; +} + +int +xfs_setattr_nonsize( + struct xfs_inode *ip, + struct iattr *iattr, + int flags) +{ + xfs_mount_t *mp = ip->i_mount; + struct inode *inode = VFS_I(ip); + int mask = iattr->ia_valid; + xfs_trans_t *tp; + int error; + uid_t uid = 0, iuid = 0; + gid_t gid = 0, igid = 0; + struct xfs_dquot *udqp = NULL, *gdqp = NULL; + struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL; + + trace_xfs_setattr(ip); + + if (mp->m_flags & XFS_MOUNT_RDONLY) + return XFS_ERROR(EROFS); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + error = -inode_change_ok(inode, iattr); + if (error) + return XFS_ERROR(error); + + ASSERT((mask & ATTR_SIZE) == 0); + + /* + * If disk quotas is on, we make sure that the dquots do exist on disk, + * before we start any other transactions. Trying to do this later + * is messy. We don't care to take a readlock to look at the ids + * in inode here, because we can't hold it across the trans_reserve. + * If the IDs do change before we take the ilock, we're covered + * because the i_*dquot fields will get updated anyway. + */ + if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) { + uint qflags = 0; + + if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) { + uid = iattr->ia_uid; + qflags |= XFS_QMOPT_UQUOTA; + } else { + uid = ip->i_d.di_uid; + } + if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) { + gid = iattr->ia_gid; + qflags |= XFS_QMOPT_GQUOTA; + } else { + gid = ip->i_d.di_gid; + } + + /* + * We take a reference when we initialize udqp and gdqp, + * so it is important that we never blindly double trip on + * the same variable. See xfs_create() for an example. + */ + ASSERT(udqp == NULL); + ASSERT(gdqp == NULL); + error = xfs_qm_vop_dqalloc(ip, uid, gid, xfs_get_projid(ip), + qflags, &udqp, &gdqp); + if (error) + return error; + } + + tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE); + error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0); + if (error) + goto out_dqrele; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * Change file ownership. Must be the owner or privileged. + */ + if (mask & (ATTR_UID|ATTR_GID)) { + /* + * These IDs could have changed since we last looked at them. + * But, we're assured that if the ownership did change + * while we didn't have the inode locked, inode's dquot(s) + * would have changed also. + */ + iuid = ip->i_d.di_uid; + igid = ip->i_d.di_gid; + gid = (mask & ATTR_GID) ? iattr->ia_gid : igid; + uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid; + + /* + * Do a quota reservation only if uid/gid is actually + * going to change. + */ + if (XFS_IS_QUOTA_RUNNING(mp) && + ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) || + (XFS_IS_GQUOTA_ON(mp) && igid != gid))) { + ASSERT(tp); + error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, + capable(CAP_FOWNER) ? + XFS_QMOPT_FORCE_RES : 0); + if (error) /* out of quota */ + goto out_trans_cancel; + } + } + + xfs_trans_ijoin(tp, ip); + + /* + * Change file ownership. Must be the owner or privileged. + */ + if (mask & (ATTR_UID|ATTR_GID)) { + /* + * CAP_FSETID overrides the following restrictions: + * + * The set-user-ID and set-group-ID bits of a file will be + * cleared upon successful return from chown() + */ + if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) && + !capable(CAP_FSETID)) + ip->i_d.di_mode &= ~(S_ISUID|S_ISGID); + + /* + * Change the ownerships and register quota modifications + * in the transaction. + */ + if (iuid != uid) { + if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) { + ASSERT(mask & ATTR_UID); + ASSERT(udqp); + olddquot1 = xfs_qm_vop_chown(tp, ip, + &ip->i_udquot, udqp); + } + ip->i_d.di_uid = uid; + inode->i_uid = uid; + } + if (igid != gid) { + if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) { + ASSERT(!XFS_IS_PQUOTA_ON(mp)); + ASSERT(mask & ATTR_GID); + ASSERT(gdqp); + olddquot2 = xfs_qm_vop_chown(tp, ip, + &ip->i_gdquot, gdqp); + } + ip->i_d.di_gid = gid; + inode->i_gid = gid; + } + } + + /* + * Change file access modes. + */ + if (mask & ATTR_MODE) { + umode_t mode = iattr->ia_mode; + + if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) + mode &= ~S_ISGID; + + ip->i_d.di_mode &= S_IFMT; + ip->i_d.di_mode |= mode & ~S_IFMT; + + inode->i_mode &= S_IFMT; + inode->i_mode |= mode & ~S_IFMT; + } + + /* + * Change file access or modified times. + */ + if (mask & ATTR_ATIME) { + inode->i_atime = iattr->ia_atime; + ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec; + ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec; + ip->i_update_core = 1; + } + if (mask & ATTR_CTIME) { + inode->i_ctime = iattr->ia_ctime; + ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; + ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec; + ip->i_update_core = 1; + } + if (mask & ATTR_MTIME) { + inode->i_mtime = iattr->ia_mtime; + ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec; + ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec; + ip->i_update_core = 1; + } + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + XFS_STATS_INC(xs_ig_attrchg); + + if (mp->m_flags & XFS_MOUNT_WSYNC) + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp, 0); + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + /* + * Release any dquot(s) the inode had kept before chown. + */ + xfs_qm_dqrele(olddquot1); + xfs_qm_dqrele(olddquot2); + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + + if (error) + return XFS_ERROR(error); + + /* + * XXX(hch): Updating the ACL entries is not atomic vs the i_mode + * update. We could avoid this with linked transactions + * and passing down the transaction pointer all the way + * to attr_set. No previous user of the generic + * Posix ACL code seems to care about this issue either. + */ + if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) { + error = -xfs_acl_chmod(inode); + if (error) + return XFS_ERROR(error); + } + + return 0; + +out_trans_cancel: + xfs_trans_cancel(tp, 0); + xfs_iunlock(ip, XFS_ILOCK_EXCL); +out_dqrele: + xfs_qm_dqrele(udqp); + xfs_qm_dqrele(gdqp); + return error; +} + +/* + * Truncate file. Must have write permission and not be a directory. + */ +int +xfs_setattr_size( + struct xfs_inode *ip, + struct iattr *iattr, + int flags) +{ + struct xfs_mount *mp = ip->i_mount; + struct inode *inode = VFS_I(ip); + int mask = iattr->ia_valid; + struct xfs_trans *tp; + int error; + uint lock_flags; + uint commit_flags = 0; + + trace_xfs_setattr(ip); + + if (mp->m_flags & XFS_MOUNT_RDONLY) + return XFS_ERROR(EROFS); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + error = -inode_change_ok(inode, iattr); + if (error) + return XFS_ERROR(error); + + ASSERT(S_ISREG(ip->i_d.di_mode)); + ASSERT((mask & (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET| + ATTR_MTIME_SET|ATTR_KILL_SUID|ATTR_KILL_SGID| + ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0); + + lock_flags = XFS_ILOCK_EXCL; + if (!(flags & XFS_ATTR_NOLOCK)) + lock_flags |= XFS_IOLOCK_EXCL; + xfs_ilock(ip, lock_flags); + + /* + * Short circuit the truncate case for zero length files. + */ + if (iattr->ia_size == 0 && + ip->i_size == 0 && ip->i_d.di_nextents == 0) { + if (!(mask & (ATTR_CTIME|ATTR_MTIME))) + goto out_unlock; + + /* + * Use the regular setattr path to update the timestamps. + */ + xfs_iunlock(ip, lock_flags); + iattr->ia_valid &= ~ATTR_SIZE; + return xfs_setattr_nonsize(ip, iattr, 0); + } + + /* + * Make sure that the dquots are attached to the inode. + */ + error = xfs_qm_dqattach_locked(ip, 0); + if (error) + goto out_unlock; + + /* + * Now we can make the changes. Before we join the inode to the + * transaction, take care of the part of the truncation that must be + * done without the inode lock. This needs to be done before joining + * the inode to the transaction, because the inode cannot be unlocked + * once it is a part of the transaction. + */ + if (iattr->ia_size > ip->i_size) { + /* + * Do the first part of growing a file: zero any data in the + * last block that is beyond the old EOF. We need to do this + * before the inode is joined to the transaction to modify + * i_size. + */ + error = xfs_zero_eof(ip, iattr->ia_size, ip->i_size); + if (error) + goto out_unlock; + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + lock_flags &= ~XFS_ILOCK_EXCL; + + /* + * We are going to log the inode size change in this transaction so + * any previous writes that are beyond the on disk EOF and the new + * EOF that have not been written out need to be written here. If we + * do not write the data out, we expose ourselves to the null files + * problem. + * + * Only flush from the on disk size to the smaller of the in memory + * file size or the new size as that's the range we really care about + * here and prevents waiting for other data not within the range we + * care about here. + */ + if (ip->i_size != ip->i_d.di_size && iattr->ia_size > ip->i_d.di_size) { + error = xfs_flush_pages(ip, ip->i_d.di_size, iattr->ia_size, + XBF_ASYNC, FI_NONE); + if (error) + goto out_unlock; + } + + /* + * Wait for all I/O to complete. + */ + xfs_ioend_wait(ip); + + error = -block_truncate_page(inode->i_mapping, iattr->ia_size, + xfs_get_blocks); + if (error) + goto out_unlock; + + tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE); + error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, + XFS_TRANS_PERM_LOG_RES, + XFS_ITRUNCATE_LOG_COUNT); + if (error) + goto out_trans_cancel; + + truncate_setsize(inode, iattr->ia_size); + + commit_flags = XFS_TRANS_RELEASE_LOG_RES; + lock_flags |= XFS_ILOCK_EXCL; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + + xfs_trans_ijoin(tp, ip); + + /* + * Only change the c/mtime if we are changing the size or we are + * explicitly asked to change it. This handles the semantic difference + * between truncate() and ftruncate() as implemented in the VFS. + * + * The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a + * special case where we need to update the times despite not having + * these flags set. For all other operations the VFS set these flags + * explicitly if it wants a timestamp update. + */ + if (iattr->ia_size != ip->i_size && + (!(mask & (ATTR_CTIME | ATTR_MTIME)))) { + iattr->ia_ctime = iattr->ia_mtime = + current_fs_time(inode->i_sb); + mask |= ATTR_CTIME | ATTR_MTIME; + } + + if (iattr->ia_size > ip->i_size) { + ip->i_d.di_size = iattr->ia_size; + ip->i_size = iattr->ia_size; + } else if (iattr->ia_size <= ip->i_size || + (iattr->ia_size == 0 && ip->i_d.di_nextents)) { + error = xfs_itruncate_data(&tp, ip, iattr->ia_size); + if (error) + goto out_trans_abort; + + /* + * Truncated "down", so we're removing references to old data + * here - if we delay flushing for a long time, we expose + * ourselves unduly to the notorious NULL files problem. So, + * we mark this inode and flush it when the file is closed, + * and do not wait the usual (long) time for writeout. + */ + xfs_iflags_set(ip, XFS_ITRUNCATED); + } + + if (mask & ATTR_CTIME) { + inode->i_ctime = iattr->ia_ctime; + ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec; + ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec; + ip->i_update_core = 1; + } + if (mask & ATTR_MTIME) { + inode->i_mtime = iattr->ia_mtime; + ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec; + ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec; + ip->i_update_core = 1; + } + + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + + XFS_STATS_INC(xs_ig_attrchg); + + if (mp->m_flags & XFS_MOUNT_WSYNC) + xfs_trans_set_sync(tp); + + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); +out_unlock: + if (lock_flags) + xfs_iunlock(ip, lock_flags); + return error; + +out_trans_abort: + commit_flags |= XFS_TRANS_ABORT; +out_trans_cancel: + xfs_trans_cancel(tp, commit_flags); + goto out_unlock; +} + +STATIC int +xfs_vn_setattr( + struct dentry *dentry, + struct iattr *iattr) +{ + if (iattr->ia_valid & ATTR_SIZE) + return -xfs_setattr_size(XFS_I(dentry->d_inode), iattr, 0); + return -xfs_setattr_nonsize(XFS_I(dentry->d_inode), iattr, 0); +} + +#define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR) + +/* + * Call fiemap helper to fill in user data. + * Returns positive errors to xfs_getbmap. + */ +STATIC int +xfs_fiemap_format( + void **arg, + struct getbmapx *bmv, + int *full) +{ + int error; + struct fiemap_extent_info *fieinfo = *arg; + u32 fiemap_flags = 0; + u64 logical, physical, length; + + /* Do nothing for a hole */ + if (bmv->bmv_block == -1LL) + return 0; + + logical = BBTOB(bmv->bmv_offset); + physical = BBTOB(bmv->bmv_block); + length = BBTOB(bmv->bmv_length); + + if (bmv->bmv_oflags & BMV_OF_PREALLOC) + fiemap_flags |= FIEMAP_EXTENT_UNWRITTEN; + else if (bmv->bmv_oflags & BMV_OF_DELALLOC) { + fiemap_flags |= FIEMAP_EXTENT_DELALLOC; + physical = 0; /* no block yet */ + } + if (bmv->bmv_oflags & BMV_OF_LAST) + fiemap_flags |= FIEMAP_EXTENT_LAST; + + error = fiemap_fill_next_extent(fieinfo, logical, physical, + length, fiemap_flags); + if (error > 0) { + error = 0; + *full = 1; /* user array now full */ + } + + return -error; +} + +STATIC int +xfs_vn_fiemap( + struct inode *inode, + struct fiemap_extent_info *fieinfo, + u64 start, + u64 length) +{ + xfs_inode_t *ip = XFS_I(inode); + struct getbmapx bm; + int error; + + error = fiemap_check_flags(fieinfo, XFS_FIEMAP_FLAGS); + if (error) + return error; + + /* Set up bmap header for xfs internal routine */ + bm.bmv_offset = BTOBB(start); + /* Special case for whole file */ + if (length == FIEMAP_MAX_OFFSET) + bm.bmv_length = -1LL; + else + bm.bmv_length = BTOBB(length); + + /* We add one because in getbmap world count includes the header */ + bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM : + fieinfo->fi_extents_max + 1; + bm.bmv_count = min_t(__s32, bm.bmv_count, + (PAGE_SIZE * 16 / sizeof(struct getbmapx))); + bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES; + if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) + bm.bmv_iflags |= BMV_IF_ATTRFORK; + if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC)) + bm.bmv_iflags |= BMV_IF_DELALLOC; + + error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo); + if (error) + return -error; + + return 0; +} + +static const struct inode_operations xfs_inode_operations = { + .get_acl = xfs_get_acl, + .getattr = xfs_vn_getattr, + .setattr = xfs_vn_setattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, + .listxattr = xfs_vn_listxattr, + .fiemap = xfs_vn_fiemap, +}; + +static const struct inode_operations xfs_dir_inode_operations = { + .create = xfs_vn_create, + .lookup = xfs_vn_lookup, + .link = xfs_vn_link, + .unlink = xfs_vn_unlink, + .symlink = xfs_vn_symlink, + .mkdir = xfs_vn_mkdir, + /* + * Yes, XFS uses the same method for rmdir and unlink. + * + * There are some subtile differences deeper in the code, + * but we use S_ISDIR to check for those. + */ + .rmdir = xfs_vn_unlink, + .mknod = xfs_vn_mknod, + .rename = xfs_vn_rename, + .get_acl = xfs_get_acl, + .getattr = xfs_vn_getattr, + .setattr = xfs_vn_setattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, + .listxattr = xfs_vn_listxattr, +}; + +static const struct inode_operations xfs_dir_ci_inode_operations = { + .create = xfs_vn_create, + .lookup = xfs_vn_ci_lookup, + .link = xfs_vn_link, + .unlink = xfs_vn_unlink, + .symlink = xfs_vn_symlink, + .mkdir = xfs_vn_mkdir, + /* + * Yes, XFS uses the same method for rmdir and unlink. + * + * There are some subtile differences deeper in the code, + * but we use S_ISDIR to check for those. + */ + .rmdir = xfs_vn_unlink, + .mknod = xfs_vn_mknod, + .rename = xfs_vn_rename, + .get_acl = xfs_get_acl, + .getattr = xfs_vn_getattr, + .setattr = xfs_vn_setattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, + .listxattr = xfs_vn_listxattr, +}; + +static const struct inode_operations xfs_symlink_inode_operations = { + .readlink = generic_readlink, + .follow_link = xfs_vn_follow_link, + .put_link = xfs_vn_put_link, + .get_acl = xfs_get_acl, + .getattr = xfs_vn_getattr, + .setattr = xfs_vn_setattr, + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .removexattr = generic_removexattr, + .listxattr = xfs_vn_listxattr, +}; + +STATIC void +xfs_diflags_to_iflags( + struct inode *inode, + struct xfs_inode *ip) +{ + if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE) + inode->i_flags |= S_IMMUTABLE; + else + inode->i_flags &= ~S_IMMUTABLE; + if (ip->i_d.di_flags & XFS_DIFLAG_APPEND) + inode->i_flags |= S_APPEND; + else + inode->i_flags &= ~S_APPEND; + if (ip->i_d.di_flags & XFS_DIFLAG_SYNC) + inode->i_flags |= S_SYNC; + else + inode->i_flags &= ~S_SYNC; + if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME) + inode->i_flags |= S_NOATIME; + else + inode->i_flags &= ~S_NOATIME; +} + +/* + * Initialize the Linux inode, set up the operation vectors and + * unlock the inode. + * + * When reading existing inodes from disk this is called directly + * from xfs_iget, when creating a new inode it is called from + * xfs_ialloc after setting up the inode. + * + * We are always called with an uninitialised linux inode here. + * We need to initialise the necessary fields and take a reference + * on it. + */ +void +xfs_setup_inode( + struct xfs_inode *ip) +{ + struct inode *inode = &ip->i_vnode; + + inode->i_ino = ip->i_ino; + inode->i_state = I_NEW; + + inode_sb_list_add(inode); + /* make the inode look hashed for the writeback code */ + hlist_add_fake(&inode->i_hash); + + inode->i_mode = ip->i_d.di_mode; + inode->i_nlink = ip->i_d.di_nlink; + inode->i_uid = ip->i_d.di_uid; + inode->i_gid = ip->i_d.di_gid; + + switch (inode->i_mode & S_IFMT) { + case S_IFBLK: + case S_IFCHR: + inode->i_rdev = + MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff, + sysv_minor(ip->i_df.if_u2.if_rdev)); + break; + default: + inode->i_rdev = 0; + break; + } + + inode->i_generation = ip->i_d.di_gen; + i_size_write(inode, ip->i_d.di_size); + inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec; + inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec; + inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec; + inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec; + inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec; + inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec; + xfs_diflags_to_iflags(inode, ip); + + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + inode->i_op = &xfs_inode_operations; + inode->i_fop = &xfs_file_operations; + inode->i_mapping->a_ops = &xfs_address_space_operations; + break; + case S_IFDIR: + if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb)) + inode->i_op = &xfs_dir_ci_inode_operations; + else + inode->i_op = &xfs_dir_inode_operations; + inode->i_fop = &xfs_dir_file_operations; + break; + case S_IFLNK: + inode->i_op = &xfs_symlink_inode_operations; + if (!(ip->i_df.if_flags & XFS_IFINLINE)) + inode->i_mapping->a_ops = &xfs_address_space_operations; + break; + default: + inode->i_op = &xfs_inode_operations; + init_special_inode(inode, inode->i_mode, inode->i_rdev); + break; + } + + /* + * If there is no attribute fork no ACL can exist on this inode, + * and it can't have any file capabilities attached to it either. + */ + if (!XFS_IFORK_Q(ip)) { + inode_has_no_xattr(inode); + cache_no_acl(inode); + } + + xfs_iflags_clear(ip, XFS_INEW); + barrier(); + + unlock_new_inode(inode); +} diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h new file mode 100644 index 000000000000..ef41c92ce66e --- /dev/null +++ b/fs/xfs/xfs_iops.h @@ -0,0 +1,30 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_IOPS_H__ +#define __XFS_IOPS_H__ + +struct xfs_inode; + +extern const struct file_operations xfs_file_operations; +extern const struct file_operations xfs_dir_file_operations; + +extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); + +extern void xfs_setup_inode(struct xfs_inode *); + +#endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h new file mode 100644 index 000000000000..1e8a45e74c3e --- /dev/null +++ b/fs/xfs/xfs_linux.h @@ -0,0 +1,309 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_LINUX__ +#define __XFS_LINUX__ + +#include + +/* + * XFS_BIG_BLKNOS needs block layer disk addresses to be 64 bits. + * XFS_BIG_INUMS requires XFS_BIG_BLKNOS to be set. + */ +#if defined(CONFIG_LBDAF) || (BITS_PER_LONG == 64) +# define XFS_BIG_BLKNOS 1 +# define XFS_BIG_INUMS 1 +#else +# define XFS_BIG_BLKNOS 0 +# define XFS_BIG_INUMS 0 +#endif + +#include "xfs_types.h" + +#include "kmem.h" +#include "mrlock.h" +#include "time.h" +#include "uuid.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include + +#include "xfs_vnode.h" +#include "xfs_stats.h" +#include "xfs_sysctl.h" +#include "xfs_iops.h" +#include "xfs_aops.h" +#include "xfs_super.h" +#include "xfs_buf.h" +#include "xfs_message.h" + +#ifdef __BIG_ENDIAN +#define XFS_NATIVE_HOST 1 +#else +#undef XFS_NATIVE_HOST +#endif + +/* + * Feature macros (disable/enable) + */ +#ifdef CONFIG_SMP +#define HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */ +#else +#undef HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */ +#endif + +#define irix_sgid_inherit xfs_params.sgid_inherit.val +#define irix_symlink_mode xfs_params.symlink_mode.val +#define xfs_panic_mask xfs_params.panic_mask.val +#define xfs_error_level xfs_params.error_level.val +#define xfs_syncd_centisecs xfs_params.syncd_timer.val +#define xfs_stats_clear xfs_params.stats_clear.val +#define xfs_inherit_sync xfs_params.inherit_sync.val +#define xfs_inherit_nodump xfs_params.inherit_nodump.val +#define xfs_inherit_noatime xfs_params.inherit_noatim.val +#define xfs_buf_timer_centisecs xfs_params.xfs_buf_timer.val +#define xfs_buf_age_centisecs xfs_params.xfs_buf_age.val +#define xfs_inherit_nosymlinks xfs_params.inherit_nosym.val +#define xfs_rotorstep xfs_params.rotorstep.val +#define xfs_inherit_nodefrag xfs_params.inherit_nodfrg.val +#define xfs_fstrm_centisecs xfs_params.fstrm_timer.val + +#define current_cpu() (raw_smp_processor_id()) +#define current_pid() (current->pid) +#define current_test_flags(f) (current->flags & (f)) +#define current_set_flags_nested(sp, f) \ + (*(sp) = current->flags, current->flags |= (f)) +#define current_clear_flags_nested(sp, f) \ + (*(sp) = current->flags, current->flags &= ~(f)) +#define current_restore_flags_nested(sp, f) \ + (current->flags = ((current->flags & ~(f)) | (*(sp) & (f)))) + +#define spinlock_destroy(lock) + +#define NBBY 8 /* number of bits per byte */ + +/* + * Size of block device i/o is parameterized here. + * Currently the system supports page-sized i/o. + */ +#define BLKDEV_IOSHIFT PAGE_CACHE_SHIFT +#define BLKDEV_IOSIZE (1<> 32; + __low = c; + if (__high) { + __upper = __high % (b); + __high = __high / (b); + } + asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper)); + asm("":"=A" (c):"a" (__low),"d" (__high)); + *(__u64 *)a = c; + return __mod; + } + } + + /* NOTREACHED */ + return 0; +} + +/* Side effect free 64 bit mod operation */ +static inline __u32 xfs_do_mod(void *a, __u32 b, int n) +{ + switch (n) { + case 4: + return *(__u32 *)a % b; + case 8: + { + unsigned long __upper, __low, __high, __mod; + __u64 c = *(__u64 *)a; + __upper = __high = c >> 32; + __low = c; + if (__high) { + __upper = __high % (b); + __high = __high / (b); + } + asm("divl %2":"=a" (__low), "=d" (__mod):"rm" (b), "0" (__low), "1" (__upper)); + asm("":"=A" (c):"a" (__low),"d" (__high)); + return __mod; + } + } + + /* NOTREACHED */ + return 0; +} +#else +static inline __u32 xfs_do_div(void *a, __u32 b, int n) +{ + __u32 mod; + + switch (n) { + case 4: + mod = *(__u32 *)a % b; + *(__u32 *)a = *(__u32 *)a / b; + return mod; + case 8: + mod = do_div(*(__u64 *)a, b); + return mod; + } + + /* NOTREACHED */ + return 0; +} + +/* Side effect free 64 bit mod operation */ +static inline __u32 xfs_do_mod(void *a, __u32 b, int n) +{ + switch (n) { + case 4: + return *(__u32 *)a % b; + case 8: + { + __u64 c = *(__u64 *)a; + return do_div(c, b); + } + } + + /* NOTREACHED */ + return 0; +} +#endif + +#undef do_div +#define do_div(a, b) xfs_do_div(&(a), (b), sizeof(a)) +#define do_mod(a, b) xfs_do_mod(&(a), (b), sizeof(a)) + +static inline __uint64_t roundup_64(__uint64_t x, __uint32_t y) +{ + x += y - 1; + do_div(x, y); + return(x * y); +} + +static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y) +{ + x += y - 1; + do_div(x, y); + return x; +} + +/* ARM old ABI has some weird alignment/padding */ +#if defined(__arm__) && !defined(__ARM_EABI__) +#define __arch_pack __attribute__((packed)) +#else +#define __arch_pack +#endif + +#define ASSERT_ALWAYS(expr) \ + (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) + +#ifndef DEBUG +#define ASSERT(expr) ((void)0) + +#ifndef STATIC +# define STATIC static noinline +#endif + +#else /* DEBUG */ + +#define ASSERT(expr) \ + (unlikely(expr) ? (void)0 : assfail(#expr, __FILE__, __LINE__)) + +#ifndef STATIC +# define STATIC noinline +#endif + +#endif /* DEBUG */ + +#endif /* __XFS_LINUX__ */ diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c new file mode 100644 index 000000000000..bd672def95ac --- /dev/null +++ b/fs/xfs/xfs_message.c @@ -0,0 +1,108 @@ +/* + * Copyright (c) 2011 Red Hat, Inc. All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" + +/* + * XFS logging functions + */ +static void +__xfs_printk( + const char *level, + const struct xfs_mount *mp, + struct va_format *vaf) +{ + if (mp && mp->m_fsname) { + printk("%sXFS (%s): %pV\n", level, mp->m_fsname, vaf); + return; + } + printk("%sXFS: %pV\n", level, vaf); +} + +#define define_xfs_printk_level(func, kern_level) \ +void func(const struct xfs_mount *mp, const char *fmt, ...) \ +{ \ + struct va_format vaf; \ + va_list args; \ + \ + va_start(args, fmt); \ + \ + vaf.fmt = fmt; \ + vaf.va = &args; \ + \ + __xfs_printk(kern_level, mp, &vaf); \ + va_end(args); \ +} \ + +define_xfs_printk_level(xfs_emerg, KERN_EMERG); +define_xfs_printk_level(xfs_alert, KERN_ALERT); +define_xfs_printk_level(xfs_crit, KERN_CRIT); +define_xfs_printk_level(xfs_err, KERN_ERR); +define_xfs_printk_level(xfs_warn, KERN_WARNING); +define_xfs_printk_level(xfs_notice, KERN_NOTICE); +define_xfs_printk_level(xfs_info, KERN_INFO); +#ifdef DEBUG +define_xfs_printk_level(xfs_debug, KERN_DEBUG); +#endif + +void +xfs_alert_tag( + const struct xfs_mount *mp, + int panic_tag, + const char *fmt, ...) +{ + struct va_format vaf; + va_list args; + int do_panic = 0; + + if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) { + xfs_alert(mp, "Transforming an alert into a BUG."); + do_panic = 1; + } + + va_start(args, fmt); + + vaf.fmt = fmt; + vaf.va = &args; + + __xfs_printk(KERN_ALERT, mp, &vaf); + va_end(args); + + BUG_ON(do_panic); +} + +void +assfail(char *expr, char *file, int line) +{ + xfs_emerg(NULL, "Assertion failed: %s, file: %s, line: %d", + expr, file, line); + BUG(); +} + +void +xfs_hex_dump(void *p, int length) +{ + print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_ADDRESS, 16, 1, p, length, 1); +} diff --git a/fs/xfs/xfs_message.h b/fs/xfs/xfs_message.h new file mode 100644 index 000000000000..7fb7ea007672 --- /dev/null +++ b/fs/xfs/xfs_message.h @@ -0,0 +1,39 @@ +#ifndef __XFS_MESSAGE_H +#define __XFS_MESSAGE_H 1 + +struct xfs_mount; + +extern void xfs_emerg(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_alert(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_alert_tag(const struct xfs_mount *mp, int tag, + const char *fmt, ...) + __attribute__ ((format (printf, 3, 4))); +extern void xfs_crit(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_err(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_warn(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_notice(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +extern void xfs_info(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); + +#ifdef DEBUG +extern void xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) + __attribute__ ((format (printf, 2, 3))); +#else +static inline void +__attribute__ ((format (printf, 2, 3))) +xfs_debug(const struct xfs_mount *mp, const char *fmt, ...) +{ +} +#endif + +extern void assfail(char *expr, char *f, int l); + +extern void xfs_hex_dump(void *p, int length); + +#endif /* __XFS_MESSAGE_H */ diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c new file mode 100644 index 000000000000..9a0aa76facdf --- /dev/null +++ b/fs/xfs/xfs_qm.c @@ -0,0 +1,2416 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_bmap.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_trans_space.h" +#include "xfs_utils.h" +#include "xfs_qm.h" +#include "xfs_trace.h" + +/* + * The global quota manager. There is only one of these for the entire + * system, _not_ one per file system. XQM keeps track of the overall + * quota functionality, including maintaining the freelist and hash + * tables of dquots. + */ +struct mutex xfs_Gqm_lock; +struct xfs_qm *xfs_Gqm; +uint ndquot; + +kmem_zone_t *qm_dqzone; +kmem_zone_t *qm_dqtrxzone; + +STATIC void xfs_qm_list_init(xfs_dqlist_t *, char *, int); +STATIC void xfs_qm_list_destroy(xfs_dqlist_t *); + +STATIC int xfs_qm_init_quotainos(xfs_mount_t *); +STATIC int xfs_qm_init_quotainfo(xfs_mount_t *); +STATIC int xfs_qm_shake(struct shrinker *, struct shrink_control *); + +static struct shrinker xfs_qm_shaker = { + .shrink = xfs_qm_shake, + .seeks = DEFAULT_SEEKS, +}; + +/* + * Initialize the XQM structure. + * Note that there is not one quota manager per file system. + */ +STATIC struct xfs_qm * +xfs_Gqm_init(void) +{ + xfs_dqhash_t *udqhash, *gdqhash; + xfs_qm_t *xqm; + size_t hsize; + uint i; + + /* + * Initialize the dquot hash tables. + */ + udqhash = kmem_zalloc_greedy(&hsize, + XFS_QM_HASHSIZE_LOW * sizeof(xfs_dqhash_t), + XFS_QM_HASHSIZE_HIGH * sizeof(xfs_dqhash_t)); + if (!udqhash) + goto out; + + gdqhash = kmem_zalloc_large(hsize); + if (!gdqhash) + goto out_free_udqhash; + + hsize /= sizeof(xfs_dqhash_t); + ndquot = hsize << 8; + + xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP); + xqm->qm_dqhashmask = hsize - 1; + xqm->qm_usr_dqhtable = udqhash; + xqm->qm_grp_dqhtable = gdqhash; + ASSERT(xqm->qm_usr_dqhtable != NULL); + ASSERT(xqm->qm_grp_dqhtable != NULL); + + for (i = 0; i < hsize; i++) { + xfs_qm_list_init(&(xqm->qm_usr_dqhtable[i]), "uxdqh", i); + xfs_qm_list_init(&(xqm->qm_grp_dqhtable[i]), "gxdqh", i); + } + + /* + * Freelist of all dquots of all file systems + */ + INIT_LIST_HEAD(&xqm->qm_dqfrlist); + xqm->qm_dqfrlist_cnt = 0; + mutex_init(&xqm->qm_dqfrlist_lock); + + /* + * dquot zone. we register our own low-memory callback. + */ + if (!qm_dqzone) { + xqm->qm_dqzone = kmem_zone_init(sizeof(xfs_dquot_t), + "xfs_dquots"); + qm_dqzone = xqm->qm_dqzone; + } else + xqm->qm_dqzone = qm_dqzone; + + register_shrinker(&xfs_qm_shaker); + + /* + * The t_dqinfo portion of transactions. + */ + if (!qm_dqtrxzone) { + xqm->qm_dqtrxzone = kmem_zone_init(sizeof(xfs_dquot_acct_t), + "xfs_dqtrx"); + qm_dqtrxzone = xqm->qm_dqtrxzone; + } else + xqm->qm_dqtrxzone = qm_dqtrxzone; + + atomic_set(&xqm->qm_totaldquots, 0); + xqm->qm_dqfree_ratio = XFS_QM_DQFREE_RATIO; + xqm->qm_nrefs = 0; + return xqm; + + out_free_udqhash: + kmem_free_large(udqhash); + out: + return NULL; +} + +/* + * Destroy the global quota manager when its reference count goes to zero. + */ +STATIC void +xfs_qm_destroy( + struct xfs_qm *xqm) +{ + struct xfs_dquot *dqp, *n; + int hsize, i; + + ASSERT(xqm != NULL); + ASSERT(xqm->qm_nrefs == 0); + unregister_shrinker(&xfs_qm_shaker); + hsize = xqm->qm_dqhashmask + 1; + for (i = 0; i < hsize; i++) { + xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i])); + xfs_qm_list_destroy(&(xqm->qm_grp_dqhtable[i])); + } + kmem_free_large(xqm->qm_usr_dqhtable); + kmem_free_large(xqm->qm_grp_dqhtable); + xqm->qm_usr_dqhtable = NULL; + xqm->qm_grp_dqhtable = NULL; + xqm->qm_dqhashmask = 0; + + /* frlist cleanup */ + mutex_lock(&xqm->qm_dqfrlist_lock); + list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) { + xfs_dqlock(dqp); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + xfs_dqunlock(dqp); + xfs_qm_dqdestroy(dqp); + } + mutex_unlock(&xqm->qm_dqfrlist_lock); + mutex_destroy(&xqm->qm_dqfrlist_lock); + kmem_free(xqm); +} + +/* + * Called at mount time to let XQM know that another file system is + * starting quotas. This isn't crucial information as the individual mount + * structures are pretty independent, but it helps the XQM keep a + * global view of what's going on. + */ +/* ARGSUSED */ +STATIC int +xfs_qm_hold_quotafs_ref( + struct xfs_mount *mp) +{ + /* + * Need to lock the xfs_Gqm structure for things like this. For example, + * the structure could disappear between the entry to this routine and + * a HOLD operation if not locked. + */ + mutex_lock(&xfs_Gqm_lock); + + if (!xfs_Gqm) { + xfs_Gqm = xfs_Gqm_init(); + if (!xfs_Gqm) { + mutex_unlock(&xfs_Gqm_lock); + return ENOMEM; + } + } + + /* + * We can keep a list of all filesystems with quotas mounted for + * debugging and statistical purposes, but ... + * Just take a reference and get out. + */ + xfs_Gqm->qm_nrefs++; + mutex_unlock(&xfs_Gqm_lock); + + return 0; +} + + +/* + * Release the reference that a filesystem took at mount time, + * so that we know when we need to destroy the entire quota manager. + */ +/* ARGSUSED */ +STATIC void +xfs_qm_rele_quotafs_ref( + struct xfs_mount *mp) +{ + xfs_dquot_t *dqp, *n; + + ASSERT(xfs_Gqm); + ASSERT(xfs_Gqm->qm_nrefs > 0); + + /* + * Go thru the freelist and destroy all inactive dquots. + */ + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + + list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) { + xfs_dqlock(dqp); + if (dqp->dq_flags & XFS_DQ_INACTIVE) { + ASSERT(dqp->q_mount == NULL); + ASSERT(! XFS_DQ_IS_DIRTY(dqp)); + ASSERT(list_empty(&dqp->q_hashlist)); + ASSERT(list_empty(&dqp->q_mplist)); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + xfs_dqunlock(dqp); + xfs_qm_dqdestroy(dqp); + } else { + xfs_dqunlock(dqp); + } + } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + + /* + * Destroy the entire XQM. If somebody mounts with quotaon, this'll + * be restarted. + */ + mutex_lock(&xfs_Gqm_lock); + if (--xfs_Gqm->qm_nrefs == 0) { + xfs_qm_destroy(xfs_Gqm); + xfs_Gqm = NULL; + } + mutex_unlock(&xfs_Gqm_lock); +} + +/* + * Just destroy the quotainfo structure. + */ +void +xfs_qm_unmount( + struct xfs_mount *mp) +{ + if (mp->m_quotainfo) { + xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL); + xfs_qm_destroy_quotainfo(mp); + } +} + + +/* + * This is called from xfs_mountfs to start quotas and initialize all + * necessary data structures like quotainfo. This is also responsible for + * running a quotacheck as necessary. We are guaranteed that the superblock + * is consistently read in at this point. + * + * If we fail here, the mount will continue with quota turned off. We don't + * need to inidicate success or failure at all. + */ +void +xfs_qm_mount_quotas( + xfs_mount_t *mp) +{ + int error = 0; + uint sbf; + + /* + * If quotas on realtime volumes is not supported, we disable + * quotas immediately. + */ + if (mp->m_sb.sb_rextents) { + xfs_notice(mp, "Cannot turn on quotas for realtime filesystem"); + mp->m_qflags = 0; + goto write_changes; + } + + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + /* + * Allocate the quotainfo structure inside the mount struct, and + * create quotainode(s), and change/rev superblock if necessary. + */ + error = xfs_qm_init_quotainfo(mp); + if (error) { + /* + * We must turn off quotas. + */ + ASSERT(mp->m_quotainfo == NULL); + mp->m_qflags = 0; + goto write_changes; + } + /* + * If any of the quotas are not consistent, do a quotacheck. + */ + if (XFS_QM_NEED_QUOTACHECK(mp)) { + error = xfs_qm_quotacheck(mp); + if (error) { + /* Quotacheck failed and disabled quotas. */ + return; + } + } + /* + * If one type of quotas is off, then it will lose its + * quotachecked status, since we won't be doing accounting for + * that type anymore. + */ + if (!XFS_IS_UQUOTA_ON(mp)) + mp->m_qflags &= ~XFS_UQUOTA_CHKD; + if (!(XFS_IS_GQUOTA_ON(mp) || XFS_IS_PQUOTA_ON(mp))) + mp->m_qflags &= ~XFS_OQUOTA_CHKD; + + write_changes: + /* + * We actually don't have to acquire the m_sb_lock at all. + * This can only be called from mount, and that's single threaded. XXX + */ + spin_lock(&mp->m_sb_lock); + sbf = mp->m_sb.sb_qflags; + mp->m_sb.sb_qflags = mp->m_qflags & XFS_MOUNT_QUOTA_ALL; + spin_unlock(&mp->m_sb_lock); + + if (sbf != (mp->m_qflags & XFS_MOUNT_QUOTA_ALL)) { + if (xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS)) { + /* + * We could only have been turning quotas off. + * We aren't in very good shape actually because + * the incore structures are convinced that quotas are + * off, but the on disk superblock doesn't know that ! + */ + ASSERT(!(XFS_IS_QUOTA_RUNNING(mp))); + xfs_alert(mp, "%s: Superblock update failed!", + __func__); + } + } + + if (error) { + xfs_warn(mp, "Failed to initialize disk quotas."); + return; + } +} + +/* + * Called from the vfsops layer. + */ +void +xfs_qm_unmount_quotas( + xfs_mount_t *mp) +{ + /* + * Release the dquots that root inode, et al might be holding, + * before we flush quotas and blow away the quotainfo structure. + */ + ASSERT(mp->m_rootip); + xfs_qm_dqdetach(mp->m_rootip); + if (mp->m_rbmip) + xfs_qm_dqdetach(mp->m_rbmip); + if (mp->m_rsumip) + xfs_qm_dqdetach(mp->m_rsumip); + + /* + * Release the quota inodes. + */ + if (mp->m_quotainfo) { + if (mp->m_quotainfo->qi_uquotaip) { + IRELE(mp->m_quotainfo->qi_uquotaip); + mp->m_quotainfo->qi_uquotaip = NULL; + } + if (mp->m_quotainfo->qi_gquotaip) { + IRELE(mp->m_quotainfo->qi_gquotaip); + mp->m_quotainfo->qi_gquotaip = NULL; + } + } +} + +/* + * Flush all dquots of the given file system to disk. The dquots are + * _not_ purged from memory here, just their data written to disk. + */ +STATIC int +xfs_qm_dqflush_all( + struct xfs_mount *mp, + int sync_mode) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + int recl; + struct xfs_dquot *dqp; + int error; + + if (!q) + return 0; +again: + mutex_lock(&q->qi_dqlist_lock); + list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { + xfs_dqlock(dqp); + if (! XFS_DQ_IS_DIRTY(dqp)) { + xfs_dqunlock(dqp); + continue; + } + + /* XXX a sentinel would be better */ + recl = q->qi_dqreclaims; + if (!xfs_dqflock_nowait(dqp)) { + /* + * If we can't grab the flush lock then check + * to see if the dquot has been flushed delayed + * write. If so, grab its buffer and send it + * out immediately. We'll be able to acquire + * the flush lock when the I/O completes. + */ + xfs_qm_dqflock_pushbuf_wait(dqp); + } + /* + * Let go of the mplist lock. We don't want to hold it + * across a disk write. + */ + mutex_unlock(&q->qi_dqlist_lock); + error = xfs_qm_dqflush(dqp, sync_mode); + xfs_dqunlock(dqp); + if (error) + return error; + + mutex_lock(&q->qi_dqlist_lock); + if (recl != q->qi_dqreclaims) { + mutex_unlock(&q->qi_dqlist_lock); + /* XXX restart limit */ + goto again; + } + } + + mutex_unlock(&q->qi_dqlist_lock); + /* return ! busy */ + return 0; +} +/* + * Release the group dquot pointers the user dquots may be + * carrying around as a hint. mplist is locked on entry and exit. + */ +STATIC void +xfs_qm_detach_gdquots( + struct xfs_mount *mp) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + struct xfs_dquot *dqp, *gdqp; + int nrecl; + + again: + ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); + list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { + xfs_dqlock(dqp); + if ((gdqp = dqp->q_gdquot)) { + xfs_dqlock(gdqp); + dqp->q_gdquot = NULL; + } + xfs_dqunlock(dqp); + + if (gdqp) { + /* + * Can't hold the mplist lock across a dqput. + * XXXmust convert to marker based iterations here. + */ + nrecl = q->qi_dqreclaims; + mutex_unlock(&q->qi_dqlist_lock); + xfs_qm_dqput(gdqp); + + mutex_lock(&q->qi_dqlist_lock); + if (nrecl != q->qi_dqreclaims) + goto again; + } + } +} + +/* + * Go through all the incore dquots of this file system and take them + * off the mplist and hashlist, if the dquot type matches the dqtype + * parameter. This is used when turning off quota accounting for + * users and/or groups, as well as when the filesystem is unmounting. + */ +STATIC int +xfs_qm_dqpurge_int( + struct xfs_mount *mp, + uint flags) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + struct xfs_dquot *dqp, *n; + uint dqtype; + int nrecl; + int nmisses; + + if (!q) + return 0; + + dqtype = (flags & XFS_QMOPT_UQUOTA) ? XFS_DQ_USER : 0; + dqtype |= (flags & XFS_QMOPT_PQUOTA) ? XFS_DQ_PROJ : 0; + dqtype |= (flags & XFS_QMOPT_GQUOTA) ? XFS_DQ_GROUP : 0; + + mutex_lock(&q->qi_dqlist_lock); + + /* + * In the first pass through all incore dquots of this filesystem, + * we release the group dquot pointers the user dquots may be + * carrying around as a hint. We need to do this irrespective of + * what's being turned off. + */ + xfs_qm_detach_gdquots(mp); + + again: + nmisses = 0; + ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); + /* + * Try to get rid of all of the unwanted dquots. The idea is to + * get them off mplist and hashlist, but leave them on freelist. + */ + list_for_each_entry_safe(dqp, n, &q->qi_dqlist, q_mplist) { + /* + * It's OK to look at the type without taking dqlock here. + * We're holding the mplist lock here, and that's needed for + * a dqreclaim. + */ + if ((dqp->dq_flags & dqtype) == 0) + continue; + + if (!mutex_trylock(&dqp->q_hash->qh_lock)) { + nrecl = q->qi_dqreclaims; + mutex_unlock(&q->qi_dqlist_lock); + mutex_lock(&dqp->q_hash->qh_lock); + mutex_lock(&q->qi_dqlist_lock); + + /* + * XXXTheoretically, we can get into a very long + * ping pong game here. + * No one can be adding dquots to the mplist at + * this point, but somebody might be taking things off. + */ + if (nrecl != q->qi_dqreclaims) { + mutex_unlock(&dqp->q_hash->qh_lock); + goto again; + } + } + + /* + * Take the dquot off the mplist and hashlist. It may remain on + * freelist in INACTIVE state. + */ + nmisses += xfs_qm_dqpurge(dqp); + } + mutex_unlock(&q->qi_dqlist_lock); + return nmisses; +} + +int +xfs_qm_dqpurge_all( + xfs_mount_t *mp, + uint flags) +{ + int ndquots; + + /* + * Purge the dquot cache. + * None of the dquots should really be busy at this point. + */ + if (mp->m_quotainfo) { + while ((ndquots = xfs_qm_dqpurge_int(mp, flags))) { + delay(ndquots * 10); + } + } + return 0; +} + +STATIC int +xfs_qm_dqattach_one( + xfs_inode_t *ip, + xfs_dqid_t id, + uint type, + uint doalloc, + xfs_dquot_t *udqhint, /* hint */ + xfs_dquot_t **IO_idqpp) +{ + xfs_dquot_t *dqp; + int error; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + error = 0; + + /* + * See if we already have it in the inode itself. IO_idqpp is + * &i_udquot or &i_gdquot. This made the code look weird, but + * made the logic a lot simpler. + */ + dqp = *IO_idqpp; + if (dqp) { + trace_xfs_dqattach_found(dqp); + return 0; + } + + /* + * udqhint is the i_udquot field in inode, and is non-NULL only + * when the type arg is group/project. Its purpose is to save a + * lookup by dqid (xfs_qm_dqget) by caching a group dquot inside + * the user dquot. + */ + if (udqhint) { + ASSERT(type == XFS_DQ_GROUP || type == XFS_DQ_PROJ); + xfs_dqlock(udqhint); + + /* + * No need to take dqlock to look at the id. + * + * The ID can't change until it gets reclaimed, and it won't + * be reclaimed as long as we have a ref from inode and we + * hold the ilock. + */ + dqp = udqhint->q_gdquot; + if (dqp && be32_to_cpu(dqp->q_core.d_id) == id) { + xfs_dqlock(dqp); + XFS_DQHOLD(dqp); + ASSERT(*IO_idqpp == NULL); + *IO_idqpp = dqp; + + xfs_dqunlock(dqp); + xfs_dqunlock(udqhint); + return 0; + } + + /* + * We can't hold a dquot lock when we call the dqget code. + * We'll deadlock in no time, because of (not conforming to) + * lock ordering - the inodelock comes before any dquot lock, + * and we may drop and reacquire the ilock in xfs_qm_dqget(). + */ + xfs_dqunlock(udqhint); + } + + /* + * Find the dquot from somewhere. This bumps the + * reference count of dquot and returns it locked. + * This can return ENOENT if dquot didn't exist on + * disk and we didn't ask it to allocate; + * ESRCH if quotas got turned off suddenly. + */ + error = xfs_qm_dqget(ip->i_mount, ip, id, type, XFS_QMOPT_DOWARN, &dqp); + if (error) + return error; + + trace_xfs_dqattach_get(dqp); + + /* + * dqget may have dropped and re-acquired the ilock, but it guarantees + * that the dquot returned is the one that should go in the inode. + */ + *IO_idqpp = dqp; + xfs_dqunlock(dqp); + return 0; +} + + +/* + * Given a udquot and gdquot, attach a ptr to the group dquot in the + * udquot as a hint for future lookups. The idea sounds simple, but the + * execution isn't, because the udquot might have a group dquot attached + * already and getting rid of that gets us into lock ordering constraints. + * The process is complicated more by the fact that the dquots may or may not + * be locked on entry. + */ +STATIC void +xfs_qm_dqattach_grouphint( + xfs_dquot_t *udq, + xfs_dquot_t *gdq) +{ + xfs_dquot_t *tmp; + + xfs_dqlock(udq); + + if ((tmp = udq->q_gdquot)) { + if (tmp == gdq) { + xfs_dqunlock(udq); + return; + } + + udq->q_gdquot = NULL; + /* + * We can't keep any dqlocks when calling dqrele, + * because the freelist lock comes before dqlocks. + */ + xfs_dqunlock(udq); + /* + * we took a hard reference once upon a time in dqget, + * so give it back when the udquot no longer points at it + * dqput() does the unlocking of the dquot. + */ + xfs_qm_dqrele(tmp); + + xfs_dqlock(udq); + xfs_dqlock(gdq); + + } else { + ASSERT(XFS_DQ_IS_LOCKED(udq)); + xfs_dqlock(gdq); + } + + ASSERT(XFS_DQ_IS_LOCKED(udq)); + ASSERT(XFS_DQ_IS_LOCKED(gdq)); + /* + * Somebody could have attached a gdquot here, + * when we dropped the uqlock. If so, just do nothing. + */ + if (udq->q_gdquot == NULL) { + XFS_DQHOLD(gdq); + udq->q_gdquot = gdq; + } + + xfs_dqunlock(gdq); + xfs_dqunlock(udq); +} + + +/* + * Given a locked inode, attach dquot(s) to it, taking U/G/P-QUOTAON + * into account. + * If XFS_QMOPT_DQALLOC, the dquot(s) will be allocated if needed. + * Inode may get unlocked and relocked in here, and the caller must deal with + * the consequences. + */ +int +xfs_qm_dqattach_locked( + xfs_inode_t *ip, + uint flags) +{ + xfs_mount_t *mp = ip->i_mount; + uint nquotas = 0; + int error = 0; + + if (!XFS_IS_QUOTA_RUNNING(mp) || + !XFS_IS_QUOTA_ON(mp) || + !XFS_NOT_DQATTACHED(mp, ip) || + ip->i_ino == mp->m_sb.sb_uquotino || + ip->i_ino == mp->m_sb.sb_gquotino) + return 0; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + if (XFS_IS_UQUOTA_ON(mp)) { + error = xfs_qm_dqattach_one(ip, ip->i_d.di_uid, XFS_DQ_USER, + flags & XFS_QMOPT_DQALLOC, + NULL, &ip->i_udquot); + if (error) + goto done; + nquotas++; + } + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + if (XFS_IS_OQUOTA_ON(mp)) { + error = XFS_IS_GQUOTA_ON(mp) ? + xfs_qm_dqattach_one(ip, ip->i_d.di_gid, XFS_DQ_GROUP, + flags & XFS_QMOPT_DQALLOC, + ip->i_udquot, &ip->i_gdquot) : + xfs_qm_dqattach_one(ip, xfs_get_projid(ip), XFS_DQ_PROJ, + flags & XFS_QMOPT_DQALLOC, + ip->i_udquot, &ip->i_gdquot); + /* + * Don't worry about the udquot that we may have + * attached above. It'll get detached, if not already. + */ + if (error) + goto done; + nquotas++; + } + + /* + * Attach this group quota to the user quota as a hint. + * This WON'T, in general, result in a thrash. + */ + if (nquotas == 2) { + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(ip->i_udquot); + ASSERT(ip->i_gdquot); + + /* + * We may or may not have the i_udquot locked at this point, + * but this check is OK since we don't depend on the i_gdquot to + * be accurate 100% all the time. It is just a hint, and this + * will succeed in general. + */ + if (ip->i_udquot->q_gdquot == ip->i_gdquot) + goto done; + /* + * Attach i_gdquot to the gdquot hint inside the i_udquot. + */ + xfs_qm_dqattach_grouphint(ip->i_udquot, ip->i_gdquot); + } + + done: +#ifdef DEBUG + if (!error) { + if (XFS_IS_UQUOTA_ON(mp)) + ASSERT(ip->i_udquot); + if (XFS_IS_OQUOTA_ON(mp)) + ASSERT(ip->i_gdquot); + } + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); +#endif + return error; +} + +int +xfs_qm_dqattach( + struct xfs_inode *ip, + uint flags) +{ + int error; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_qm_dqattach_locked(ip, flags); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + return error; +} + +/* + * Release dquots (and their references) if any. + * The inode should be locked EXCL except when this's called by + * xfs_ireclaim. + */ +void +xfs_qm_dqdetach( + xfs_inode_t *ip) +{ + if (!(ip->i_udquot || ip->i_gdquot)) + return; + + trace_xfs_dquot_dqdetach(ip); + + ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_uquotino); + ASSERT(ip->i_ino != ip->i_mount->m_sb.sb_gquotino); + if (ip->i_udquot) { + xfs_qm_dqrele(ip->i_udquot); + ip->i_udquot = NULL; + } + if (ip->i_gdquot) { + xfs_qm_dqrele(ip->i_gdquot); + ip->i_gdquot = NULL; + } +} + +int +xfs_qm_sync( + struct xfs_mount *mp, + int flags) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + int recl, restarts; + struct xfs_dquot *dqp; + int error; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + + restarts = 0; + + again: + mutex_lock(&q->qi_dqlist_lock); + /* + * dqpurge_all() also takes the mplist lock and iterate thru all dquots + * in quotaoff. However, if the QUOTA_ACTIVE bits are not cleared + * when we have the mplist lock, we know that dquots will be consistent + * as long as we have it locked. + */ + if (!XFS_IS_QUOTA_ON(mp)) { + mutex_unlock(&q->qi_dqlist_lock); + return 0; + } + ASSERT(mutex_is_locked(&q->qi_dqlist_lock)); + list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) { + /* + * If this is vfs_sync calling, then skip the dquots that + * don't 'seem' to be dirty. ie. don't acquire dqlock. + * This is very similar to what xfs_sync does with inodes. + */ + if (flags & SYNC_TRYLOCK) { + if (!XFS_DQ_IS_DIRTY(dqp)) + continue; + if (!xfs_qm_dqlock_nowait(dqp)) + continue; + } else { + xfs_dqlock(dqp); + } + + /* + * Now, find out for sure if this dquot is dirty or not. + */ + if (! XFS_DQ_IS_DIRTY(dqp)) { + xfs_dqunlock(dqp); + continue; + } + + /* XXX a sentinel would be better */ + recl = q->qi_dqreclaims; + if (!xfs_dqflock_nowait(dqp)) { + if (flags & SYNC_TRYLOCK) { + xfs_dqunlock(dqp); + continue; + } + /* + * If we can't grab the flush lock then if the caller + * really wanted us to give this our best shot, so + * see if we can give a push to the buffer before we wait + * on the flush lock. At this point, we know that + * even though the dquot is being flushed, + * it has (new) dirty data. + */ + xfs_qm_dqflock_pushbuf_wait(dqp); + } + /* + * Let go of the mplist lock. We don't want to hold it + * across a disk write + */ + mutex_unlock(&q->qi_dqlist_lock); + error = xfs_qm_dqflush(dqp, flags); + xfs_dqunlock(dqp); + if (error && XFS_FORCED_SHUTDOWN(mp)) + return 0; /* Need to prevent umount failure */ + else if (error) + return error; + + mutex_lock(&q->qi_dqlist_lock); + if (recl != q->qi_dqreclaims) { + if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS) + break; + + mutex_unlock(&q->qi_dqlist_lock); + goto again; + } + } + + mutex_unlock(&q->qi_dqlist_lock); + return 0; +} + +/* + * The hash chains and the mplist use the same xfs_dqhash structure as + * their list head, but we can take the mplist qh_lock and one of the + * hash qh_locks at the same time without any problem as they aren't + * related. + */ +static struct lock_class_key xfs_quota_mplist_class; + +/* + * This initializes all the quota information that's kept in the + * mount structure + */ +STATIC int +xfs_qm_init_quotainfo( + xfs_mount_t *mp) +{ + xfs_quotainfo_t *qinf; + int error; + xfs_dquot_t *dqp; + + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + /* + * Tell XQM that we exist as soon as possible. + */ + if ((error = xfs_qm_hold_quotafs_ref(mp))) { + return error; + } + + qinf = mp->m_quotainfo = kmem_zalloc(sizeof(xfs_quotainfo_t), KM_SLEEP); + + /* + * See if quotainodes are setup, and if not, allocate them, + * and change the superblock accordingly. + */ + if ((error = xfs_qm_init_quotainos(mp))) { + kmem_free(qinf); + mp->m_quotainfo = NULL; + return error; + } + + INIT_LIST_HEAD(&qinf->qi_dqlist); + mutex_init(&qinf->qi_dqlist_lock); + lockdep_set_class(&qinf->qi_dqlist_lock, &xfs_quota_mplist_class); + + qinf->qi_dqreclaims = 0; + + /* mutex used to serialize quotaoffs */ + mutex_init(&qinf->qi_quotaofflock); + + /* Precalc some constants */ + qinf->qi_dqchunklen = XFS_FSB_TO_BB(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); + ASSERT(qinf->qi_dqchunklen); + qinf->qi_dqperchunk = BBTOB(qinf->qi_dqchunklen); + do_div(qinf->qi_dqperchunk, sizeof(xfs_dqblk_t)); + + mp->m_qflags |= (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_CHKD); + + /* + * We try to get the limits from the superuser's limits fields. + * This is quite hacky, but it is standard quota practice. + * We look at the USR dquot with id == 0 first, but if user quotas + * are not enabled we goto the GRP dquot with id == 0. + * We don't really care to keep separate default limits for user + * and group quotas, at least not at this point. + */ + error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)0, + XFS_IS_UQUOTA_RUNNING(mp) ? XFS_DQ_USER : + (XFS_IS_GQUOTA_RUNNING(mp) ? XFS_DQ_GROUP : + XFS_DQ_PROJ), + XFS_QMOPT_DQSUSER|XFS_QMOPT_DOWARN, + &dqp); + if (! error) { + xfs_disk_dquot_t *ddqp = &dqp->q_core; + + /* + * The warnings and timers set the grace period given to + * a user or group before he or she can not perform any + * more writing. If it is zero, a default is used. + */ + qinf->qi_btimelimit = ddqp->d_btimer ? + be32_to_cpu(ddqp->d_btimer) : XFS_QM_BTIMELIMIT; + qinf->qi_itimelimit = ddqp->d_itimer ? + be32_to_cpu(ddqp->d_itimer) : XFS_QM_ITIMELIMIT; + qinf->qi_rtbtimelimit = ddqp->d_rtbtimer ? + be32_to_cpu(ddqp->d_rtbtimer) : XFS_QM_RTBTIMELIMIT; + qinf->qi_bwarnlimit = ddqp->d_bwarns ? + be16_to_cpu(ddqp->d_bwarns) : XFS_QM_BWARNLIMIT; + qinf->qi_iwarnlimit = ddqp->d_iwarns ? + be16_to_cpu(ddqp->d_iwarns) : XFS_QM_IWARNLIMIT; + qinf->qi_rtbwarnlimit = ddqp->d_rtbwarns ? + be16_to_cpu(ddqp->d_rtbwarns) : XFS_QM_RTBWARNLIMIT; + qinf->qi_bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); + qinf->qi_bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit); + qinf->qi_ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); + qinf->qi_isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); + qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); + qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); + + /* + * We sent the XFS_QMOPT_DQSUSER flag to dqget because + * we don't want this dquot cached. We haven't done a + * quotacheck yet, and quotacheck doesn't like incore dquots. + */ + xfs_qm_dqdestroy(dqp); + } else { + qinf->qi_btimelimit = XFS_QM_BTIMELIMIT; + qinf->qi_itimelimit = XFS_QM_ITIMELIMIT; + qinf->qi_rtbtimelimit = XFS_QM_RTBTIMELIMIT; + qinf->qi_bwarnlimit = XFS_QM_BWARNLIMIT; + qinf->qi_iwarnlimit = XFS_QM_IWARNLIMIT; + qinf->qi_rtbwarnlimit = XFS_QM_RTBWARNLIMIT; + } + + return 0; +} + + +/* + * Gets called when unmounting a filesystem or when all quotas get + * turned off. + * This purges the quota inodes, destroys locks and frees itself. + */ +void +xfs_qm_destroy_quotainfo( + xfs_mount_t *mp) +{ + xfs_quotainfo_t *qi; + + qi = mp->m_quotainfo; + ASSERT(qi != NULL); + ASSERT(xfs_Gqm != NULL); + + /* + * Release the reference that XQM kept, so that we know + * when the XQM structure should be freed. We cannot assume + * that xfs_Gqm is non-null after this point. + */ + xfs_qm_rele_quotafs_ref(mp); + + ASSERT(list_empty(&qi->qi_dqlist)); + mutex_destroy(&qi->qi_dqlist_lock); + + if (qi->qi_uquotaip) { + IRELE(qi->qi_uquotaip); + qi->qi_uquotaip = NULL; /* paranoia */ + } + if (qi->qi_gquotaip) { + IRELE(qi->qi_gquotaip); + qi->qi_gquotaip = NULL; + } + mutex_destroy(&qi->qi_quotaofflock); + kmem_free(qi); + mp->m_quotainfo = NULL; +} + + + +/* ------------------- PRIVATE STATIC FUNCTIONS ----------------------- */ + +/* ARGSUSED */ +STATIC void +xfs_qm_list_init( + xfs_dqlist_t *list, + char *str, + int n) +{ + mutex_init(&list->qh_lock); + INIT_LIST_HEAD(&list->qh_list); + list->qh_version = 0; + list->qh_nelems = 0; +} + +STATIC void +xfs_qm_list_destroy( + xfs_dqlist_t *list) +{ + mutex_destroy(&(list->qh_lock)); +} + +/* + * Create an inode and return with a reference already taken, but unlocked + * This is how we create quota inodes + */ +STATIC int +xfs_qm_qino_alloc( + xfs_mount_t *mp, + xfs_inode_t **ip, + __int64_t sbfields, + uint flags) +{ + xfs_trans_t *tp; + int error; + int committed; + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE); + if ((error = xfs_trans_reserve(tp, + XFS_QM_QINOCREATE_SPACE_RES(mp), + XFS_CREATE_LOG_RES(mp), 0, + XFS_TRANS_PERM_LOG_RES, + XFS_CREATE_LOG_COUNT))) { + xfs_trans_cancel(tp, 0); + return error; + } + + error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, 1, ip, &committed); + if (error) { + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | + XFS_TRANS_ABORT); + return error; + } + + /* + * Make the changes in the superblock, and log those too. + * sbfields arg may contain fields other than *QUOTINO; + * VERSIONNUM for example. + */ + spin_lock(&mp->m_sb_lock); + if (flags & XFS_QMOPT_SBVERSION) { + ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); + ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | + XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == + (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | + XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); + + xfs_sb_version_addquota(&mp->m_sb); + mp->m_sb.sb_uquotino = NULLFSINO; + mp->m_sb.sb_gquotino = NULLFSINO; + + /* qflags will get updated _after_ quotacheck */ + mp->m_sb.sb_qflags = 0; + } + if (flags & XFS_QMOPT_UQUOTA) + mp->m_sb.sb_uquotino = (*ip)->i_ino; + else + mp->m_sb.sb_gquotino = (*ip)->i_ino; + spin_unlock(&mp->m_sb_lock); + xfs_mod_sb(tp, sbfields); + + if ((error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES))) { + xfs_alert(mp, "%s failed (error %d)!", __func__, error); + return error; + } + return 0; +} + + +STATIC void +xfs_qm_reset_dqcounts( + xfs_mount_t *mp, + xfs_buf_t *bp, + xfs_dqid_t id, + uint type) +{ + xfs_disk_dquot_t *ddq; + int j; + + trace_xfs_reset_dqcounts(bp, _RET_IP_); + + /* + * Reset all counters and timers. They'll be + * started afresh by xfs_qm_quotacheck. + */ +#ifdef DEBUG + j = XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB); + do_div(j, sizeof(xfs_dqblk_t)); + ASSERT(mp->m_quotainfo->qi_dqperchunk == j); +#endif + ddq = bp->b_addr; + for (j = 0; j < mp->m_quotainfo->qi_dqperchunk; j++) { + /* + * Do a sanity check, and if needed, repair the dqblk. Don't + * output any warnings because it's perfectly possible to + * find uninitialised dquot blks. See comment in xfs_qm_dqcheck. + */ + (void) xfs_qm_dqcheck(mp, ddq, id+j, type, XFS_QMOPT_DQREPAIR, + "xfs_quotacheck"); + ddq->d_bcount = 0; + ddq->d_icount = 0; + ddq->d_rtbcount = 0; + ddq->d_btimer = 0; + ddq->d_itimer = 0; + ddq->d_rtbtimer = 0; + ddq->d_bwarns = 0; + ddq->d_iwarns = 0; + ddq->d_rtbwarns = 0; + ddq = (xfs_disk_dquot_t *) ((xfs_dqblk_t *)ddq + 1); + } +} + +STATIC int +xfs_qm_dqiter_bufs( + xfs_mount_t *mp, + xfs_dqid_t firstid, + xfs_fsblock_t bno, + xfs_filblks_t blkcnt, + uint flags) +{ + xfs_buf_t *bp; + int error; + int type; + + ASSERT(blkcnt > 0); + type = flags & XFS_QMOPT_UQUOTA ? XFS_DQ_USER : + (flags & XFS_QMOPT_PQUOTA ? XFS_DQ_PROJ : XFS_DQ_GROUP); + error = 0; + + /* + * Blkcnt arg can be a very big number, and might even be + * larger than the log itself. So, we have to break it up into + * manageable-sized transactions. + * Note that we don't start a permanent transaction here; we might + * not be able to get a log reservation for the whole thing up front, + * and we don't really care to either, because we just discard + * everything if we were to crash in the middle of this loop. + */ + while (blkcnt--) { + error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, bno), + mp->m_quotainfo->qi_dqchunklen, 0, &bp); + if (error) + break; + + xfs_qm_reset_dqcounts(mp, bp, firstid, type); + xfs_bdwrite(mp, bp); + /* + * goto the next block. + */ + bno++; + firstid += mp->m_quotainfo->qi_dqperchunk; + } + return error; +} + +/* + * Iterate over all allocated USR/GRP/PRJ dquots in the system, calling a + * caller supplied function for every chunk of dquots that we find. + */ +STATIC int +xfs_qm_dqiterate( + xfs_mount_t *mp, + xfs_inode_t *qip, + uint flags) +{ + xfs_bmbt_irec_t *map; + int i, nmaps; /* number of map entries */ + int error; /* return value */ + xfs_fileoff_t lblkno; + xfs_filblks_t maxlblkcnt; + xfs_dqid_t firstid; + xfs_fsblock_t rablkno; + xfs_filblks_t rablkcnt; + + error = 0; + /* + * This looks racy, but we can't keep an inode lock across a + * trans_reserve. But, this gets called during quotacheck, and that + * happens only at mount time which is single threaded. + */ + if (qip->i_d.di_nblocks == 0) + return 0; + + map = kmem_alloc(XFS_DQITER_MAP_SIZE * sizeof(*map), KM_SLEEP); + + lblkno = 0; + maxlblkcnt = XFS_B_TO_FSB(mp, (xfs_ufsize_t)XFS_MAXIOFFSET(mp)); + do { + nmaps = XFS_DQITER_MAP_SIZE; + /* + * We aren't changing the inode itself. Just changing + * some of its data. No new blocks are added here, and + * the inode is never added to the transaction. + */ + xfs_ilock(qip, XFS_ILOCK_SHARED); + error = xfs_bmapi(NULL, qip, lblkno, + maxlblkcnt - lblkno, + XFS_BMAPI_METADATA, + NULL, + 0, map, &nmaps, NULL); + xfs_iunlock(qip, XFS_ILOCK_SHARED); + if (error) + break; + + ASSERT(nmaps <= XFS_DQITER_MAP_SIZE); + for (i = 0; i < nmaps; i++) { + ASSERT(map[i].br_startblock != DELAYSTARTBLOCK); + ASSERT(map[i].br_blockcount); + + + lblkno += map[i].br_blockcount; + + if (map[i].br_startblock == HOLESTARTBLOCK) + continue; + + firstid = (xfs_dqid_t) map[i].br_startoff * + mp->m_quotainfo->qi_dqperchunk; + /* + * Do a read-ahead on the next extent. + */ + if ((i+1 < nmaps) && + (map[i+1].br_startblock != HOLESTARTBLOCK)) { + rablkcnt = map[i+1].br_blockcount; + rablkno = map[i+1].br_startblock; + while (rablkcnt--) { + xfs_buf_readahead(mp->m_ddev_targp, + XFS_FSB_TO_DADDR(mp, rablkno), + mp->m_quotainfo->qi_dqchunklen); + rablkno++; + } + } + /* + * Iterate thru all the blks in the extent and + * reset the counters of all the dquots inside them. + */ + if ((error = xfs_qm_dqiter_bufs(mp, + firstid, + map[i].br_startblock, + map[i].br_blockcount, + flags))) { + break; + } + } + + if (error) + break; + } while (nmaps > 0); + + kmem_free(map); + + return error; +} + +/* + * Called by dqusage_adjust in doing a quotacheck. + * + * Given the inode, and a dquot id this updates both the incore dqout as well + * as the buffer copy. This is so that once the quotacheck is done, we can + * just log all the buffers, as opposed to logging numerous updates to + * individual dquots. + */ +STATIC int +xfs_qm_quotacheck_dqadjust( + struct xfs_inode *ip, + xfs_dqid_t id, + uint type, + xfs_qcnt_t nblks, + xfs_qcnt_t rtblks) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_dquot *dqp; + int error; + + error = xfs_qm_dqget(mp, ip, id, type, + XFS_QMOPT_DQALLOC | XFS_QMOPT_DOWARN, &dqp); + if (error) { + /* + * Shouldn't be able to turn off quotas here. + */ + ASSERT(error != ESRCH); + ASSERT(error != ENOENT); + return error; + } + + trace_xfs_dqadjust(dqp); + + /* + * Adjust the inode count and the block count to reflect this inode's + * resource usage. + */ + be64_add_cpu(&dqp->q_core.d_icount, 1); + dqp->q_res_icount++; + if (nblks) { + be64_add_cpu(&dqp->q_core.d_bcount, nblks); + dqp->q_res_bcount += nblks; + } + if (rtblks) { + be64_add_cpu(&dqp->q_core.d_rtbcount, rtblks); + dqp->q_res_rtbcount += rtblks; + } + + /* + * Set default limits, adjust timers (since we changed usages) + * + * There are no timers for the default values set in the root dquot. + */ + if (dqp->q_core.d_id) { + xfs_qm_adjust_dqlimits(mp, &dqp->q_core); + xfs_qm_adjust_dqtimers(mp, &dqp->q_core); + } + + dqp->dq_flags |= XFS_DQ_DIRTY; + xfs_qm_dqput(dqp); + return 0; +} + +STATIC int +xfs_qm_get_rtblks( + xfs_inode_t *ip, + xfs_qcnt_t *O_rtblks) +{ + xfs_filblks_t rtblks; /* total rt blks */ + xfs_extnum_t idx; /* extent record index */ + xfs_ifork_t *ifp; /* inode fork pointer */ + xfs_extnum_t nextents; /* number of extent entries */ + int error; + + ASSERT(XFS_IS_REALTIME_INODE(ip)); + ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK); + if (!(ifp->if_flags & XFS_IFEXTENTS)) { + if ((error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK))) + return error; + } + rtblks = 0; + nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); + for (idx = 0; idx < nextents; idx++) + rtblks += xfs_bmbt_get_blockcount(xfs_iext_get_ext(ifp, idx)); + *O_rtblks = (xfs_qcnt_t)rtblks; + return 0; +} + +/* + * callback routine supplied to bulkstat(). Given an inumber, find its + * dquots and update them to account for resources taken by that inode. + */ +/* ARGSUSED */ +STATIC int +xfs_qm_dqusage_adjust( + xfs_mount_t *mp, /* mount point for filesystem */ + xfs_ino_t ino, /* inode number to get data for */ + void __user *buffer, /* not used */ + int ubsize, /* not used */ + int *ubused, /* not used */ + int *res) /* result code value */ +{ + xfs_inode_t *ip; + xfs_qcnt_t nblks, rtblks = 0; + int error; + + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + /* + * rootino must have its resources accounted for, not so with the quota + * inodes. + */ + if (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino) { + *res = BULKSTAT_RV_NOTHING; + return XFS_ERROR(EINVAL); + } + + /* + * We don't _need_ to take the ilock EXCL. However, the xfs_qm_dqget + * interface expects the inode to be exclusively locked because that's + * the case in all other instances. It's OK that we do this because + * quotacheck is done only at mount time. + */ + error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip); + if (error) { + *res = BULKSTAT_RV_NOTHING; + return error; + } + + ASSERT(ip->i_delayed_blks == 0); + + if (XFS_IS_REALTIME_INODE(ip)) { + /* + * Walk thru the extent list and count the realtime blocks. + */ + error = xfs_qm_get_rtblks(ip, &rtblks); + if (error) + goto error0; + } + + nblks = (xfs_qcnt_t)ip->i_d.di_nblocks - rtblks; + + /* + * Add the (disk blocks and inode) resources occupied by this + * inode to its dquots. We do this adjustment in the incore dquot, + * and also copy the changes to its buffer. + * We don't care about putting these changes in a transaction + * envelope because if we crash in the middle of a 'quotacheck' + * we have to start from the beginning anyway. + * Once we're done, we'll log all the dquot bufs. + * + * The *QUOTA_ON checks below may look pretty racy, but quotachecks + * and quotaoffs don't race. (Quotachecks happen at mount time only). + */ + if (XFS_IS_UQUOTA_ON(mp)) { + error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_uid, + XFS_DQ_USER, nblks, rtblks); + if (error) + goto error0; + } + + if (XFS_IS_GQUOTA_ON(mp)) { + error = xfs_qm_quotacheck_dqadjust(ip, ip->i_d.di_gid, + XFS_DQ_GROUP, nblks, rtblks); + if (error) + goto error0; + } + + if (XFS_IS_PQUOTA_ON(mp)) { + error = xfs_qm_quotacheck_dqadjust(ip, xfs_get_projid(ip), + XFS_DQ_PROJ, nblks, rtblks); + if (error) + goto error0; + } + + xfs_iunlock(ip, XFS_ILOCK_EXCL); + IRELE(ip); + *res = BULKSTAT_RV_DIDONE; + return 0; + +error0: + xfs_iunlock(ip, XFS_ILOCK_EXCL); + IRELE(ip); + *res = BULKSTAT_RV_GIVEUP; + return error; +} + +/* + * Walk thru all the filesystem inodes and construct a consistent view + * of the disk quota world. If the quotacheck fails, disable quotas. + */ +int +xfs_qm_quotacheck( + xfs_mount_t *mp) +{ + int done, count, error; + xfs_ino_t lastino; + size_t structsz; + xfs_inode_t *uip, *gip; + uint flags; + + count = INT_MAX; + structsz = 1; + lastino = 0; + flags = 0; + + ASSERT(mp->m_quotainfo->qi_uquotaip || mp->m_quotainfo->qi_gquotaip); + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + /* + * There should be no cached dquots. The (simplistic) quotacheck + * algorithm doesn't like that. + */ + ASSERT(list_empty(&mp->m_quotainfo->qi_dqlist)); + + xfs_notice(mp, "Quotacheck needed: Please wait."); + + /* + * First we go thru all the dquots on disk, USR and GRP/PRJ, and reset + * their counters to zero. We need a clean slate. + * We don't log our changes till later. + */ + uip = mp->m_quotainfo->qi_uquotaip; + if (uip) { + error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA); + if (error) + goto error_return; + flags |= XFS_UQUOTA_CHKD; + } + + gip = mp->m_quotainfo->qi_gquotaip; + if (gip) { + error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ? + XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA); + if (error) + goto error_return; + flags |= XFS_OQUOTA_CHKD; + } + + do { + /* + * Iterate thru all the inodes in the file system, + * adjusting the corresponding dquot counters in core. + */ + error = xfs_bulkstat(mp, &lastino, &count, + xfs_qm_dqusage_adjust, + structsz, NULL, &done); + if (error) + break; + + } while (!done); + + /* + * We've made all the changes that we need to make incore. + * Flush them down to disk buffers if everything was updated + * successfully. + */ + if (!error) + error = xfs_qm_dqflush_all(mp, 0); + + /* + * We can get this error if we couldn't do a dquot allocation inside + * xfs_qm_dqusage_adjust (via bulkstat). We don't care about the + * dirty dquots that might be cached, we just want to get rid of them + * and turn quotaoff. The dquots won't be attached to any of the inodes + * at this point (because we intentionally didn't in dqget_noattach). + */ + if (error) { + xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL); + goto error_return; + } + + /* + * We didn't log anything, because if we crashed, we'll have to + * start the quotacheck from scratch anyway. However, we must make + * sure that our dquot changes are secure before we put the + * quotacheck'd stamp on the superblock. So, here we do a synchronous + * flush. + */ + XFS_bflush(mp->m_ddev_targp); + + /* + * If one type of quotas is off, then it will lose its + * quotachecked status, since we won't be doing accounting for + * that type anymore. + */ + mp->m_qflags &= ~(XFS_OQUOTA_CHKD | XFS_UQUOTA_CHKD); + mp->m_qflags |= flags; + + error_return: + if (error) { + xfs_warn(mp, + "Quotacheck: Unsuccessful (Error %d): Disabling quotas.", + error); + /* + * We must turn off quotas. + */ + ASSERT(mp->m_quotainfo != NULL); + ASSERT(xfs_Gqm != NULL); + xfs_qm_destroy_quotainfo(mp); + if (xfs_mount_reset_sbqflags(mp)) { + xfs_warn(mp, + "Quotacheck: Failed to reset quota flags."); + } + } else + xfs_notice(mp, "Quotacheck: Done."); + return (error); +} + +/* + * This is called after the superblock has been read in and we're ready to + * iget the quota inodes. + */ +STATIC int +xfs_qm_init_quotainos( + xfs_mount_t *mp) +{ + xfs_inode_t *uip, *gip; + int error; + __int64_t sbflags; + uint flags; + + ASSERT(mp->m_quotainfo); + uip = gip = NULL; + sbflags = 0; + flags = 0; + + /* + * Get the uquota and gquota inodes + */ + if (xfs_sb_version_hasquota(&mp->m_sb)) { + if (XFS_IS_UQUOTA_ON(mp) && + mp->m_sb.sb_uquotino != NULLFSINO) { + ASSERT(mp->m_sb.sb_uquotino > 0); + if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, + 0, 0, &uip))) + return XFS_ERROR(error); + } + if (XFS_IS_OQUOTA_ON(mp) && + mp->m_sb.sb_gquotino != NULLFSINO) { + ASSERT(mp->m_sb.sb_gquotino > 0); + if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, + 0, 0, &gip))) { + if (uip) + IRELE(uip); + return XFS_ERROR(error); + } + } + } else { + flags |= XFS_QMOPT_SBVERSION; + sbflags |= (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | + XFS_SB_GQUOTINO | XFS_SB_QFLAGS); + } + + /* + * Create the two inodes, if they don't exist already. The changes + * made above will get added to a transaction and logged in one of + * the qino_alloc calls below. If the device is readonly, + * temporarily switch to read-write to do this. + */ + if (XFS_IS_UQUOTA_ON(mp) && uip == NULL) { + if ((error = xfs_qm_qino_alloc(mp, &uip, + sbflags | XFS_SB_UQUOTINO, + flags | XFS_QMOPT_UQUOTA))) + return XFS_ERROR(error); + + flags &= ~XFS_QMOPT_SBVERSION; + } + if (XFS_IS_OQUOTA_ON(mp) && gip == NULL) { + flags |= (XFS_IS_GQUOTA_ON(mp) ? + XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA); + error = xfs_qm_qino_alloc(mp, &gip, + sbflags | XFS_SB_GQUOTINO, flags); + if (error) { + if (uip) + IRELE(uip); + + return XFS_ERROR(error); + } + } + + mp->m_quotainfo->qi_uquotaip = uip; + mp->m_quotainfo->qi_gquotaip = gip; + + return 0; +} + + + +/* + * Just pop the least recently used dquot off the freelist and + * recycle it. The returned dquot is locked. + */ +STATIC xfs_dquot_t * +xfs_qm_dqreclaim_one(void) +{ + xfs_dquot_t *dqpout; + xfs_dquot_t *dqp; + int restarts; + int startagain; + + restarts = 0; + dqpout = NULL; + + /* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */ +again: + startagain = 0; + mutex_lock(&xfs_Gqm->qm_dqfrlist_lock); + + list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) { + struct xfs_mount *mp = dqp->q_mount; + xfs_dqlock(dqp); + + /* + * We are racing with dqlookup here. Naturally we don't + * want to reclaim a dquot that lookup wants. We release the + * freelist lock and start over, so that lookup will grab + * both the dquot and the freelistlock. + */ + if (dqp->dq_flags & XFS_DQ_WANT) { + ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE)); + + trace_xfs_dqreclaim_want(dqp); + XQM_STATS_INC(xqmstats.xs_qm_dqwants); + restarts++; + startagain = 1; + goto dqunlock; + } + + /* + * If the dquot is inactive, we are assured that it is + * not on the mplist or the hashlist, and that makes our + * life easier. + */ + if (dqp->dq_flags & XFS_DQ_INACTIVE) { + ASSERT(mp == NULL); + ASSERT(! XFS_DQ_IS_DIRTY(dqp)); + ASSERT(list_empty(&dqp->q_hashlist)); + ASSERT(list_empty(&dqp->q_mplist)); + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + dqpout = dqp; + XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims); + goto dqunlock; + } + + ASSERT(dqp->q_hash); + ASSERT(!list_empty(&dqp->q_mplist)); + + /* + * Try to grab the flush lock. If this dquot is in the process + * of getting flushed to disk, we don't want to reclaim it. + */ + if (!xfs_dqflock_nowait(dqp)) + goto dqunlock; + + /* + * We have the flush lock so we know that this is not in the + * process of being flushed. So, if this is dirty, flush it + * DELWRI so that we don't get a freelist infested with + * dirty dquots. + */ + if (XFS_DQ_IS_DIRTY(dqp)) { + int error; + + trace_xfs_dqreclaim_dirty(dqp); + + /* + * We flush it delayed write, so don't bother + * releasing the freelist lock. + */ + error = xfs_qm_dqflush(dqp, 0); + if (error) { + xfs_warn(mp, "%s: dquot %p flush failed", + __func__, dqp); + } + goto dqunlock; + } + + /* + * We're trying to get the hashlock out of order. This races + * with dqlookup; so, we giveup and goto the next dquot if + * we couldn't get the hashlock. This way, we won't starve + * a dqlookup process that holds the hashlock that is + * waiting for the freelist lock. + */ + if (!mutex_trylock(&dqp->q_hash->qh_lock)) { + restarts++; + goto dqfunlock; + } + + /* + * This races with dquot allocation code as well as dqflush_all + * and reclaim code. So, if we failed to grab the mplist lock, + * giveup everything and start over. + */ + if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) { + restarts++; + startagain = 1; + goto qhunlock; + } + + ASSERT(dqp->q_nrefs == 0); + list_del_init(&dqp->q_mplist); + mp->m_quotainfo->qi_dquots--; + mp->m_quotainfo->qi_dqreclaims++; + list_del_init(&dqp->q_hashlist); + dqp->q_hash->qh_version++; + list_del_init(&dqp->q_freelist); + xfs_Gqm->qm_dqfrlist_cnt--; + dqpout = dqp; + mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock); +qhunlock: + mutex_unlock(&dqp->q_hash->qh_lock); +dqfunlock: + xfs_dqfunlock(dqp); +dqunlock: + xfs_dqunlock(dqp); + if (dqpout) + break; + if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS) + break; + if (startagain) { + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + goto again; + } + } + mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock); + return dqpout; +} + +/* + * Traverse the freelist of dquots and attempt to reclaim a maximum of + * 'howmany' dquots. This operation races with dqlookup(), and attempts to + * favor the lookup function ... + */ +STATIC int +xfs_qm_shake_freelist( + int howmany) +{ + int nreclaimed = 0; + xfs_dquot_t *dqp; + + if (howmany <= 0) + return 0; + + while (nreclaimed < howmany) { + dqp = xfs_qm_dqreclaim_one(); + if (!dqp) + return nreclaimed; + xfs_qm_dqdestroy(dqp); + nreclaimed++; + } + return nreclaimed; +} + +/* + * The kmem_shake interface is invoked when memory is running low. + */ +/* ARGSUSED */ +STATIC int +xfs_qm_shake( + struct shrinker *shrink, + struct shrink_control *sc) +{ + int ndqused, nfree, n; + gfp_t gfp_mask = sc->gfp_mask; + + if (!kmem_shake_allow(gfp_mask)) + return 0; + if (!xfs_Gqm) + return 0; + + nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */ + /* incore dquots in all f/s's */ + ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree; + + ASSERT(ndqused >= 0); + + if (nfree <= ndqused && nfree < ndquot) + return 0; + + ndqused *= xfs_Gqm->qm_dqfree_ratio; /* target # of free dquots */ + n = nfree - ndqused - ndquot; /* # over target */ + + return xfs_qm_shake_freelist(MAX(nfree, n)); +} + + +/*------------------------------------------------------------------*/ + +/* + * Return a new incore dquot. Depending on the number of + * dquots in the system, we either allocate a new one on the kernel heap, + * or reclaim a free one. + * Return value is B_TRUE if we allocated a new dquot, B_FALSE if we managed + * to reclaim an existing one from the freelist. + */ +boolean_t +xfs_qm_dqalloc_incore( + xfs_dquot_t **O_dqpp) +{ + xfs_dquot_t *dqp; + + /* + * Check against high water mark to see if we want to pop + * a nincompoop dquot off the freelist. + */ + if (atomic_read(&xfs_Gqm->qm_totaldquots) >= ndquot) { + /* + * Try to recycle a dquot from the freelist. + */ + if ((dqp = xfs_qm_dqreclaim_one())) { + XQM_STATS_INC(xqmstats.xs_qm_dqreclaims); + /* + * Just zero the core here. The rest will get + * reinitialized by caller. XXX we shouldn't even + * do this zero ... + */ + memset(&dqp->q_core, 0, sizeof(dqp->q_core)); + *O_dqpp = dqp; + return B_FALSE; + } + XQM_STATS_INC(xqmstats.xs_qm_dqreclaim_misses); + } + + /* + * Allocate a brand new dquot on the kernel heap and return it + * to the caller to initialize. + */ + ASSERT(xfs_Gqm->qm_dqzone != NULL); + *O_dqpp = kmem_zone_zalloc(xfs_Gqm->qm_dqzone, KM_SLEEP); + atomic_inc(&xfs_Gqm->qm_totaldquots); + + return B_TRUE; +} + + +/* + * Start a transaction and write the incore superblock changes to + * disk. flags parameter indicates which fields have changed. + */ +int +xfs_qm_write_sb_changes( + xfs_mount_t *mp, + __int64_t flags) +{ + xfs_trans_t *tp; + int error; + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SBCHANGE); + if ((error = xfs_trans_reserve(tp, 0, + mp->m_sb.sb_sectsize + 128, 0, + 0, + XFS_DEFAULT_LOG_COUNT))) { + xfs_trans_cancel(tp, 0); + return error; + } + + xfs_mod_sb(tp, flags); + error = xfs_trans_commit(tp, 0); + + return error; +} + + +/* --------------- utility functions for vnodeops ---------------- */ + + +/* + * Given an inode, a uid, gid and prid make sure that we have + * allocated relevant dquot(s) on disk, and that we won't exceed inode + * quotas by creating this file. + * This also attaches dquot(s) to the given inode after locking it, + * and returns the dquots corresponding to the uid and/or gid. + * + * in : inode (unlocked) + * out : udquot, gdquot with references taken and unlocked + */ +int +xfs_qm_vop_dqalloc( + struct xfs_inode *ip, + uid_t uid, + gid_t gid, + prid_t prid, + uint flags, + struct xfs_dquot **O_udqpp, + struct xfs_dquot **O_gdqpp) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_dquot *uq, *gq; + int error; + uint lockflags; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + + lockflags = XFS_ILOCK_EXCL; + xfs_ilock(ip, lockflags); + + if ((flags & XFS_QMOPT_INHERIT) && XFS_INHERIT_GID(ip)) + gid = ip->i_d.di_gid; + + /* + * Attach the dquot(s) to this inode, doing a dquot allocation + * if necessary. The dquot(s) will not be locked. + */ + if (XFS_NOT_DQATTACHED(mp, ip)) { + error = xfs_qm_dqattach_locked(ip, XFS_QMOPT_DQALLOC); + if (error) { + xfs_iunlock(ip, lockflags); + return error; + } + } + + uq = gq = NULL; + if ((flags & XFS_QMOPT_UQUOTA) && XFS_IS_UQUOTA_ON(mp)) { + if (ip->i_d.di_uid != uid) { + /* + * What we need is the dquot that has this uid, and + * if we send the inode to dqget, the uid of the inode + * takes priority over what's sent in the uid argument. + * We must unlock inode here before calling dqget if + * we're not sending the inode, because otherwise + * we'll deadlock by doing trans_reserve while + * holding ilock. + */ + xfs_iunlock(ip, lockflags); + if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t) uid, + XFS_DQ_USER, + XFS_QMOPT_DQALLOC | + XFS_QMOPT_DOWARN, + &uq))) { + ASSERT(error != ENOENT); + return error; + } + /* + * Get the ilock in the right order. + */ + xfs_dqunlock(uq); + lockflags = XFS_ILOCK_SHARED; + xfs_ilock(ip, lockflags); + } else { + /* + * Take an extra reference, because we'll return + * this to caller + */ + ASSERT(ip->i_udquot); + uq = ip->i_udquot; + xfs_dqlock(uq); + XFS_DQHOLD(uq); + xfs_dqunlock(uq); + } + } + if ((flags & XFS_QMOPT_GQUOTA) && XFS_IS_GQUOTA_ON(mp)) { + if (ip->i_d.di_gid != gid) { + xfs_iunlock(ip, lockflags); + if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)gid, + XFS_DQ_GROUP, + XFS_QMOPT_DQALLOC | + XFS_QMOPT_DOWARN, + &gq))) { + if (uq) + xfs_qm_dqrele(uq); + ASSERT(error != ENOENT); + return error; + } + xfs_dqunlock(gq); + lockflags = XFS_ILOCK_SHARED; + xfs_ilock(ip, lockflags); + } else { + ASSERT(ip->i_gdquot); + gq = ip->i_gdquot; + xfs_dqlock(gq); + XFS_DQHOLD(gq); + xfs_dqunlock(gq); + } + } else if ((flags & XFS_QMOPT_PQUOTA) && XFS_IS_PQUOTA_ON(mp)) { + if (xfs_get_projid(ip) != prid) { + xfs_iunlock(ip, lockflags); + if ((error = xfs_qm_dqget(mp, NULL, (xfs_dqid_t)prid, + XFS_DQ_PROJ, + XFS_QMOPT_DQALLOC | + XFS_QMOPT_DOWARN, + &gq))) { + if (uq) + xfs_qm_dqrele(uq); + ASSERT(error != ENOENT); + return (error); + } + xfs_dqunlock(gq); + lockflags = XFS_ILOCK_SHARED; + xfs_ilock(ip, lockflags); + } else { + ASSERT(ip->i_gdquot); + gq = ip->i_gdquot; + xfs_dqlock(gq); + XFS_DQHOLD(gq); + xfs_dqunlock(gq); + } + } + if (uq) + trace_xfs_dquot_dqalloc(ip); + + xfs_iunlock(ip, lockflags); + if (O_udqpp) + *O_udqpp = uq; + else if (uq) + xfs_qm_dqrele(uq); + if (O_gdqpp) + *O_gdqpp = gq; + else if (gq) + xfs_qm_dqrele(gq); + return 0; +} + +/* + * Actually transfer ownership, and do dquot modifications. + * These were already reserved. + */ +xfs_dquot_t * +xfs_qm_vop_chown( + xfs_trans_t *tp, + xfs_inode_t *ip, + xfs_dquot_t **IO_olddq, + xfs_dquot_t *newdq) +{ + xfs_dquot_t *prevdq; + uint bfield = XFS_IS_REALTIME_INODE(ip) ? + XFS_TRANS_DQ_RTBCOUNT : XFS_TRANS_DQ_BCOUNT; + + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(XFS_IS_QUOTA_RUNNING(ip->i_mount)); + + /* old dquot */ + prevdq = *IO_olddq; + ASSERT(prevdq); + ASSERT(prevdq != newdq); + + xfs_trans_mod_dquot(tp, prevdq, bfield, -(ip->i_d.di_nblocks)); + xfs_trans_mod_dquot(tp, prevdq, XFS_TRANS_DQ_ICOUNT, -1); + + /* the sparkling new dquot */ + xfs_trans_mod_dquot(tp, newdq, bfield, ip->i_d.di_nblocks); + xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_ICOUNT, 1); + + /* + * Take an extra reference, because the inode + * is going to keep this dquot pointer even + * after the trans_commit. + */ + xfs_dqlock(newdq); + XFS_DQHOLD(newdq); + xfs_dqunlock(newdq); + *IO_olddq = newdq; + + return prevdq; +} + +/* + * Quota reservations for setattr(AT_UID|AT_GID|AT_PROJID). + */ +int +xfs_qm_vop_chown_reserve( + xfs_trans_t *tp, + xfs_inode_t *ip, + xfs_dquot_t *udqp, + xfs_dquot_t *gdqp, + uint flags) +{ + xfs_mount_t *mp = ip->i_mount; + uint delblks, blkflags, prjflags = 0; + xfs_dquot_t *unresudq, *unresgdq, *delblksudq, *delblksgdq; + int error; + + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + delblks = ip->i_delayed_blks; + delblksudq = delblksgdq = unresudq = unresgdq = NULL; + blkflags = XFS_IS_REALTIME_INODE(ip) ? + XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS; + + if (XFS_IS_UQUOTA_ON(mp) && udqp && + ip->i_d.di_uid != (uid_t)be32_to_cpu(udqp->q_core.d_id)) { + delblksudq = udqp; + /* + * If there are delayed allocation blocks, then we have to + * unreserve those from the old dquot, and add them to the + * new dquot. + */ + if (delblks) { + ASSERT(ip->i_udquot); + unresudq = ip->i_udquot; + } + } + if (XFS_IS_OQUOTA_ON(ip->i_mount) && gdqp) { + if (XFS_IS_PQUOTA_ON(ip->i_mount) && + xfs_get_projid(ip) != be32_to_cpu(gdqp->q_core.d_id)) + prjflags = XFS_QMOPT_ENOSPC; + + if (prjflags || + (XFS_IS_GQUOTA_ON(ip->i_mount) && + ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id))) { + delblksgdq = gdqp; + if (delblks) { + ASSERT(ip->i_gdquot); + unresgdq = ip->i_gdquot; + } + } + } + + if ((error = xfs_trans_reserve_quota_bydquots(tp, ip->i_mount, + delblksudq, delblksgdq, ip->i_d.di_nblocks, 1, + flags | blkflags | prjflags))) + return (error); + + /* + * Do the delayed blks reservations/unreservations now. Since, these + * are done without the help of a transaction, if a reservation fails + * its previous reservations won't be automatically undone by trans + * code. So, we have to do it manually here. + */ + if (delblks) { + /* + * Do the reservations first. Unreservation can't fail. + */ + ASSERT(delblksudq || delblksgdq); + ASSERT(unresudq || unresgdq); + if ((error = xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, + delblksudq, delblksgdq, (xfs_qcnt_t)delblks, 0, + flags | blkflags | prjflags))) + return (error); + xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, + unresudq, unresgdq, -((xfs_qcnt_t)delblks), 0, + blkflags); + } + + return (0); +} + +int +xfs_qm_vop_rename_dqattach( + struct xfs_inode **i_tab) +{ + struct xfs_mount *mp = i_tab[0]->i_mount; + int i; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + + for (i = 0; (i < 4 && i_tab[i]); i++) { + struct xfs_inode *ip = i_tab[i]; + int error; + + /* + * Watch out for duplicate entries in the table. + */ + if (i == 0 || ip != i_tab[i-1]) { + if (XFS_NOT_DQATTACHED(mp, ip)) { + error = xfs_qm_dqattach(ip, 0); + if (error) + return error; + } + } + } + return 0; +} + +void +xfs_qm_vop_create_dqattach( + struct xfs_trans *tp, + struct xfs_inode *ip, + struct xfs_dquot *udqp, + struct xfs_dquot *gdqp) +{ + struct xfs_mount *mp = tp->t_mountp; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT(XFS_IS_QUOTA_RUNNING(mp)); + + if (udqp) { + xfs_dqlock(udqp); + XFS_DQHOLD(udqp); + xfs_dqunlock(udqp); + ASSERT(ip->i_udquot == NULL); + ip->i_udquot = udqp; + ASSERT(XFS_IS_UQUOTA_ON(mp)); + ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id)); + xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1); + } + if (gdqp) { + xfs_dqlock(gdqp); + XFS_DQHOLD(gdqp); + xfs_dqunlock(gdqp); + ASSERT(ip->i_gdquot == NULL); + ip->i_gdquot = gdqp; + ASSERT(XFS_IS_OQUOTA_ON(mp)); + ASSERT((XFS_IS_GQUOTA_ON(mp) ? + ip->i_d.di_gid : xfs_get_projid(ip)) == + be32_to_cpu(gdqp->q_core.d_id)); + xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1); + } +} + diff --git a/fs/xfs/xfs_qm.h b/fs/xfs/xfs_qm.h new file mode 100644 index 000000000000..43b9abe1052c --- /dev/null +++ b/fs/xfs/xfs_qm.h @@ -0,0 +1,166 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_QM_H__ +#define __XFS_QM_H__ + +#include "xfs_dquot_item.h" +#include "xfs_dquot.h" +#include "xfs_quota_priv.h" +#include "xfs_qm_stats.h" + +struct xfs_qm; +struct xfs_inode; + +extern uint ndquot; +extern struct mutex xfs_Gqm_lock; +extern struct xfs_qm *xfs_Gqm; +extern kmem_zone_t *qm_dqzone; +extern kmem_zone_t *qm_dqtrxzone; + +/* + * Used in xfs_qm_sync called by xfs_sync to count the max times that it can + * iterate over the mountpt's dquot list in one call. + */ +#define XFS_QM_SYNC_MAX_RESTARTS 7 + +/* + * Ditto, for xfs_qm_dqreclaim_one. + */ +#define XFS_QM_RECLAIM_MAX_RESTARTS 4 + +/* + * Ideal ratio of free to in use dquots. Quota manager makes an attempt + * to keep this balance. + */ +#define XFS_QM_DQFREE_RATIO 2 + +/* + * Dquot hashtable constants/threshold values. + */ +#define XFS_QM_HASHSIZE_LOW (PAGE_SIZE / sizeof(xfs_dqhash_t)) +#define XFS_QM_HASHSIZE_HIGH ((PAGE_SIZE * 4) / sizeof(xfs_dqhash_t)) + +/* + * This defines the unit of allocation of dquots. + * Currently, it is just one file system block, and a 4K blk contains 30 + * (136 * 30 = 4080) dquots. It's probably not worth trying to make + * this more dynamic. + * XXXsup However, if this number is changed, we have to make sure that we don't + * implicitly assume that we do allocations in chunks of a single filesystem + * block in the dquot/xqm code. + */ +#define XFS_DQUOT_CLUSTER_SIZE_FSB (xfs_filblks_t)1 + +typedef xfs_dqhash_t xfs_dqlist_t; + +/* + * Quota Manager (global) structure. Lives only in core. + */ +typedef struct xfs_qm { + xfs_dqlist_t *qm_usr_dqhtable;/* udquot hash table */ + xfs_dqlist_t *qm_grp_dqhtable;/* gdquot hash table */ + uint qm_dqhashmask; /* # buckets in dq hashtab - 1 */ + struct list_head qm_dqfrlist; /* freelist of dquots */ + struct mutex qm_dqfrlist_lock; + int qm_dqfrlist_cnt; + atomic_t qm_totaldquots; /* total incore dquots */ + uint qm_nrefs; /* file systems with quota on */ + int qm_dqfree_ratio;/* ratio of free to inuse dquots */ + kmem_zone_t *qm_dqzone; /* dquot mem-alloc zone */ + kmem_zone_t *qm_dqtrxzone; /* t_dqinfo of transactions */ +} xfs_qm_t; + +/* + * Various quota information for individual filesystems. + * The mount structure keeps a pointer to this. + */ +typedef struct xfs_quotainfo { + xfs_inode_t *qi_uquotaip; /* user quota inode */ + xfs_inode_t *qi_gquotaip; /* group quota inode */ + struct list_head qi_dqlist; /* all dquots in filesys */ + struct mutex qi_dqlist_lock; + int qi_dquots; + int qi_dqreclaims; /* a change here indicates + a removal in the dqlist */ + time_t qi_btimelimit; /* limit for blks timer */ + time_t qi_itimelimit; /* limit for inodes timer */ + time_t qi_rtbtimelimit;/* limit for rt blks timer */ + xfs_qwarncnt_t qi_bwarnlimit; /* limit for blks warnings */ + xfs_qwarncnt_t qi_iwarnlimit; /* limit for inodes warnings */ + xfs_qwarncnt_t qi_rtbwarnlimit;/* limit for rt blks warnings */ + struct mutex qi_quotaofflock;/* to serialize quotaoff */ + xfs_filblks_t qi_dqchunklen; /* # BBs in a chunk of dqs */ + uint qi_dqperchunk; /* # ondisk dqs in above chunk */ + xfs_qcnt_t qi_bhardlimit; /* default data blk hard limit */ + xfs_qcnt_t qi_bsoftlimit; /* default data blk soft limit */ + xfs_qcnt_t qi_ihardlimit; /* default inode count hard limit */ + xfs_qcnt_t qi_isoftlimit; /* default inode count soft limit */ + xfs_qcnt_t qi_rtbhardlimit;/* default realtime blk hard limit */ + xfs_qcnt_t qi_rtbsoftlimit;/* default realtime blk soft limit */ +} xfs_quotainfo_t; + + +extern void xfs_trans_mod_dquot(xfs_trans_t *, xfs_dquot_t *, uint, long); +extern int xfs_trans_reserve_quota_bydquots(xfs_trans_t *, xfs_mount_t *, + xfs_dquot_t *, xfs_dquot_t *, long, long, uint); +extern void xfs_trans_dqjoin(xfs_trans_t *, xfs_dquot_t *); +extern void xfs_trans_log_dquot(xfs_trans_t *, xfs_dquot_t *); + +/* + * We keep the usr and grp dquots separately so that locking will be easier + * to do at commit time. All transactions that we know of at this point + * affect no more than two dquots of one type. Hence, the TRANS_MAXDQS value. + */ +#define XFS_QM_TRANS_MAXDQS 2 +typedef struct xfs_dquot_acct { + xfs_dqtrx_t dqa_usrdquots[XFS_QM_TRANS_MAXDQS]; + xfs_dqtrx_t dqa_grpdquots[XFS_QM_TRANS_MAXDQS]; +} xfs_dquot_acct_t; + +/* + * Users are allowed to have a usage exceeding their softlimit for + * a period this long. + */ +#define XFS_QM_BTIMELIMIT (7 * 24*60*60) /* 1 week */ +#define XFS_QM_RTBTIMELIMIT (7 * 24*60*60) /* 1 week */ +#define XFS_QM_ITIMELIMIT (7 * 24*60*60) /* 1 week */ + +#define XFS_QM_BWARNLIMIT 5 +#define XFS_QM_IWARNLIMIT 5 +#define XFS_QM_RTBWARNLIMIT 5 + +extern void xfs_qm_destroy_quotainfo(xfs_mount_t *); +extern int xfs_qm_quotacheck(xfs_mount_t *); +extern int xfs_qm_write_sb_changes(xfs_mount_t *, __int64_t); + +/* dquot stuff */ +extern boolean_t xfs_qm_dqalloc_incore(xfs_dquot_t **); +extern int xfs_qm_dqpurge_all(xfs_mount_t *, uint); +extern void xfs_qm_dqrele_all_inodes(xfs_mount_t *, uint); + +/* quota ops */ +extern int xfs_qm_scall_trunc_qfiles(xfs_mount_t *, uint); +extern int xfs_qm_scall_getquota(xfs_mount_t *, xfs_dqid_t, uint, + fs_disk_quota_t *); +extern int xfs_qm_scall_setqlim(xfs_mount_t *, xfs_dqid_t, uint, + fs_disk_quota_t *); +extern int xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *); +extern int xfs_qm_scall_quotaon(xfs_mount_t *, uint); +extern int xfs_qm_scall_quotaoff(xfs_mount_t *, uint); + +#endif /* __XFS_QM_H__ */ diff --git a/fs/xfs/xfs_qm_bhv.c b/fs/xfs/xfs_qm_bhv.c new file mode 100644 index 000000000000..a0a829addca9 --- /dev/null +++ b/fs/xfs/xfs_qm_bhv.c @@ -0,0 +1,176 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_itable.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_qm.h" + + +STATIC void +xfs_fill_statvfs_from_dquot( + struct kstatfs *statp, + xfs_disk_dquot_t *dp) +{ + __uint64_t limit; + + limit = dp->d_blk_softlimit ? + be64_to_cpu(dp->d_blk_softlimit) : + be64_to_cpu(dp->d_blk_hardlimit); + if (limit && statp->f_blocks > limit) { + statp->f_blocks = limit; + statp->f_bfree = statp->f_bavail = + (statp->f_blocks > be64_to_cpu(dp->d_bcount)) ? + (statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0; + } + + limit = dp->d_ino_softlimit ? + be64_to_cpu(dp->d_ino_softlimit) : + be64_to_cpu(dp->d_ino_hardlimit); + if (limit && statp->f_files > limit) { + statp->f_files = limit; + statp->f_ffree = + (statp->f_files > be64_to_cpu(dp->d_icount)) ? + (statp->f_ffree - be64_to_cpu(dp->d_icount)) : 0; + } +} + + +/* + * Directory tree accounting is implemented using project quotas, where + * the project identifier is inherited from parent directories. + * A statvfs (df, etc.) of a directory that is using project quota should + * return a statvfs of the project, not the entire filesystem. + * This makes such trees appear as if they are filesystems in themselves. + */ +void +xfs_qm_statvfs( + xfs_inode_t *ip, + struct kstatfs *statp) +{ + xfs_mount_t *mp = ip->i_mount; + xfs_dquot_t *dqp; + + if (!xfs_qm_dqget(mp, NULL, xfs_get_projid(ip), XFS_DQ_PROJ, 0, &dqp)) { + xfs_fill_statvfs_from_dquot(statp, &dqp->q_core); + xfs_qm_dqput(dqp); + } +} + +int +xfs_qm_newmount( + xfs_mount_t *mp, + uint *needquotamount, + uint *quotaflags) +{ + uint quotaondisk; + uint uquotaondisk = 0, gquotaondisk = 0, pquotaondisk = 0; + + quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) && + (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT); + + if (quotaondisk) { + uquotaondisk = mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT; + pquotaondisk = mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT; + gquotaondisk = mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT; + } + + /* + * If the device itself is read-only, we can't allow + * the user to change the state of quota on the mount - + * this would generate a transaction on the ro device, + * which would lead to an I/O error and shutdown + */ + + if (((uquotaondisk && !XFS_IS_UQUOTA_ON(mp)) || + (!uquotaondisk && XFS_IS_UQUOTA_ON(mp)) || + (pquotaondisk && !XFS_IS_PQUOTA_ON(mp)) || + (!pquotaondisk && XFS_IS_PQUOTA_ON(mp)) || + (gquotaondisk && !XFS_IS_GQUOTA_ON(mp)) || + (!gquotaondisk && XFS_IS_OQUOTA_ON(mp))) && + xfs_dev_is_read_only(mp, "changing quota state")) { + xfs_warn(mp, "please mount with%s%s%s%s.", + (!quotaondisk ? "out quota" : ""), + (uquotaondisk ? " usrquota" : ""), + (pquotaondisk ? " prjquota" : ""), + (gquotaondisk ? " grpquota" : "")); + return XFS_ERROR(EPERM); + } + + if (XFS_IS_QUOTA_ON(mp) || quotaondisk) { + /* + * Call mount_quotas at this point only if we won't have to do + * a quotacheck. + */ + if (quotaondisk && !XFS_QM_NEED_QUOTACHECK(mp)) { + /* + * If an error occurred, qm_mount_quotas code + * has already disabled quotas. So, just finish + * mounting, and get on with the boring life + * without disk quotas. + */ + xfs_qm_mount_quotas(mp); + } else { + /* + * Clear the quota flags, but remember them. This + * is so that the quota code doesn't get invoked + * before we're ready. This can happen when an + * inode goes inactive and wants to free blocks, + * or via xfs_log_mount_finish. + */ + *needquotamount = B_TRUE; + *quotaflags = mp->m_qflags; + mp->m_qflags = 0; + } + } + + return 0; +} + +void __init +xfs_qm_init(void) +{ + printk(KERN_INFO "SGI XFS Quota Management subsystem\n"); + mutex_init(&xfs_Gqm_lock); + xfs_qm_init_procfs(); +} + +void __exit +xfs_qm_exit(void) +{ + xfs_qm_cleanup_procfs(); + if (qm_dqzone) + kmem_zone_destroy(qm_dqzone); + if (qm_dqtrxzone) + kmem_zone_destroy(qm_dqtrxzone); +} diff --git a/fs/xfs/xfs_qm_stats.c b/fs/xfs/xfs_qm_stats.c new file mode 100644 index 000000000000..8671a0b32644 --- /dev/null +++ b/fs/xfs/xfs_qm_stats.c @@ -0,0 +1,105 @@ +/* + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_itable.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_qm.h" + +struct xqmstats xqmstats; + +static int xqm_proc_show(struct seq_file *m, void *v) +{ + /* maximum; incore; ratio free to inuse; freelist */ + seq_printf(m, "%d\t%d\t%d\t%u\n", + ndquot, + xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0, + xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0, + xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0); + return 0; +} + +static int xqm_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, xqm_proc_show, NULL); +} + +static const struct file_operations xqm_proc_fops = { + .owner = THIS_MODULE, + .open = xqm_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int xqmstat_proc_show(struct seq_file *m, void *v) +{ + /* quota performance statistics */ + seq_printf(m, "qm %u %u %u %u %u %u %u %u\n", + xqmstats.xs_qm_dqreclaims, + xqmstats.xs_qm_dqreclaim_misses, + xqmstats.xs_qm_dquot_dups, + xqmstats.xs_qm_dqcachemisses, + xqmstats.xs_qm_dqcachehits, + xqmstats.xs_qm_dqwants, + xqmstats.xs_qm_dqshake_reclaims, + xqmstats.xs_qm_dqinact_reclaims); + return 0; +} + +static int xqmstat_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, xqmstat_proc_show, NULL); +} + +static const struct file_operations xqmstat_proc_fops = { + .owner = THIS_MODULE, + .open = xqmstat_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +void +xfs_qm_init_procfs(void) +{ + proc_create("fs/xfs/xqmstat", 0, NULL, &xqmstat_proc_fops); + proc_create("fs/xfs/xqm", 0, NULL, &xqm_proc_fops); +} + +void +xfs_qm_cleanup_procfs(void) +{ + remove_proc_entry("fs/xfs/xqm", NULL); + remove_proc_entry("fs/xfs/xqmstat", NULL); +} diff --git a/fs/xfs/xfs_qm_stats.h b/fs/xfs/xfs_qm_stats.h new file mode 100644 index 000000000000..5b964fc0dc09 --- /dev/null +++ b/fs/xfs/xfs_qm_stats.h @@ -0,0 +1,53 @@ +/* + * Copyright (c) 2002 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_QM_STATS_H__ +#define __XFS_QM_STATS_H__ + +#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF) + +/* + * XQM global statistics + */ +struct xqmstats { + __uint32_t xs_qm_dqreclaims; + __uint32_t xs_qm_dqreclaim_misses; + __uint32_t xs_qm_dquot_dups; + __uint32_t xs_qm_dqcachemisses; + __uint32_t xs_qm_dqcachehits; + __uint32_t xs_qm_dqwants; + __uint32_t xs_qm_dqshake_reclaims; + __uint32_t xs_qm_dqinact_reclaims; +}; + +extern struct xqmstats xqmstats; + +# define XQM_STATS_INC(count) ( (count)++ ) + +extern void xfs_qm_init_procfs(void); +extern void xfs_qm_cleanup_procfs(void); + +#else + +# define XQM_STATS_INC(count) do { } while (0) + +static inline void xfs_qm_init_procfs(void) { }; +static inline void xfs_qm_cleanup_procfs(void) { }; + +#endif + +#endif /* __XFS_QM_STATS_H__ */ diff --git a/fs/xfs/xfs_qm_syscalls.c b/fs/xfs/xfs_qm_syscalls.c new file mode 100644 index 000000000000..609246f42e6c --- /dev/null +++ b/fs/xfs/xfs_qm_syscalls.c @@ -0,0 +1,906 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include + +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_itable.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_utils.h" +#include "xfs_qm.h" +#include "xfs_trace.h" + +STATIC int xfs_qm_log_quotaoff(xfs_mount_t *, xfs_qoff_logitem_t **, uint); +STATIC int xfs_qm_log_quotaoff_end(xfs_mount_t *, xfs_qoff_logitem_t *, + uint); +STATIC uint xfs_qm_export_flags(uint); +STATIC uint xfs_qm_export_qtype_flags(uint); +STATIC void xfs_qm_export_dquot(xfs_mount_t *, xfs_disk_dquot_t *, + fs_disk_quota_t *); + + +/* + * Turn off quota accounting and/or enforcement for all udquots and/or + * gdquots. Called only at unmount time. + * + * This assumes that there are no dquots of this file system cached + * incore, and modifies the ondisk dquot directly. Therefore, for example, + * it is an error to call this twice, without purging the cache. + */ +int +xfs_qm_scall_quotaoff( + xfs_mount_t *mp, + uint flags) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + uint dqtype; + int error; + uint inactivate_flags; + xfs_qoff_logitem_t *qoffstart; + int nculprits; + + /* + * No file system can have quotas enabled on disk but not in core. + * Note that quota utilities (like quotaoff) _expect_ + * errno == EEXIST here. + */ + if ((mp->m_qflags & flags) == 0) + return XFS_ERROR(EEXIST); + error = 0; + + flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD); + + /* + * We don't want to deal with two quotaoffs messing up each other, + * so we're going to serialize it. quotaoff isn't exactly a performance + * critical thing. + * If quotaoff, then we must be dealing with the root filesystem. + */ + ASSERT(q); + mutex_lock(&q->qi_quotaofflock); + + /* + * If we're just turning off quota enforcement, change mp and go. + */ + if ((flags & XFS_ALL_QUOTA_ACCT) == 0) { + mp->m_qflags &= ~(flags); + + spin_lock(&mp->m_sb_lock); + mp->m_sb.sb_qflags = mp->m_qflags; + spin_unlock(&mp->m_sb_lock); + mutex_unlock(&q->qi_quotaofflock); + + /* XXX what to do if error ? Revert back to old vals incore ? */ + error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS); + return (error); + } + + dqtype = 0; + inactivate_flags = 0; + /* + * If accounting is off, we must turn enforcement off, clear the + * quota 'CHKD' certificate to make it known that we have to + * do a quotacheck the next time this quota is turned on. + */ + if (flags & XFS_UQUOTA_ACCT) { + dqtype |= XFS_QMOPT_UQUOTA; + flags |= (XFS_UQUOTA_CHKD | XFS_UQUOTA_ENFD); + inactivate_flags |= XFS_UQUOTA_ACTIVE; + } + if (flags & XFS_GQUOTA_ACCT) { + dqtype |= XFS_QMOPT_GQUOTA; + flags |= (XFS_OQUOTA_CHKD | XFS_OQUOTA_ENFD); + inactivate_flags |= XFS_GQUOTA_ACTIVE; + } else if (flags & XFS_PQUOTA_ACCT) { + dqtype |= XFS_QMOPT_PQUOTA; + flags |= (XFS_OQUOTA_CHKD | XFS_OQUOTA_ENFD); + inactivate_flags |= XFS_PQUOTA_ACTIVE; + } + + /* + * Nothing to do? Don't complain. This happens when we're just + * turning off quota enforcement. + */ + if ((mp->m_qflags & flags) == 0) + goto out_unlock; + + /* + * Write the LI_QUOTAOFF log record, and do SB changes atomically, + * and synchronously. If we fail to write, we should abort the + * operation as it cannot be recovered safely if we crash. + */ + error = xfs_qm_log_quotaoff(mp, &qoffstart, flags); + if (error) + goto out_unlock; + + /* + * Next we clear the XFS_MOUNT_*DQ_ACTIVE bit(s) in the mount struct + * to take care of the race between dqget and quotaoff. We don't take + * any special locks to reset these bits. All processes need to check + * these bits *after* taking inode lock(s) to see if the particular + * quota type is in the process of being turned off. If *ACTIVE, it is + * guaranteed that all dquot structures and all quotainode ptrs will all + * stay valid as long as that inode is kept locked. + * + * There is no turning back after this. + */ + mp->m_qflags &= ~inactivate_flags; + + /* + * Give back all the dquot reference(s) held by inodes. + * Here we go thru every single incore inode in this file system, and + * do a dqrele on the i_udquot/i_gdquot that it may have. + * Essentially, as long as somebody has an inode locked, this guarantees + * that quotas will not be turned off. This is handy because in a + * transaction once we lock the inode(s) and check for quotaon, we can + * depend on the quota inodes (and other things) being valid as long as + * we keep the lock(s). + */ + xfs_qm_dqrele_all_inodes(mp, flags); + + /* + * Next we make the changes in the quota flag in the mount struct. + * This isn't protected by a particular lock directly, because we + * don't want to take a mrlock every time we depend on quotas being on. + */ + mp->m_qflags &= ~(flags); + + /* + * Go through all the dquots of this file system and purge them, + * according to what was turned off. We may not be able to get rid + * of all dquots, because dquots can have temporary references that + * are not attached to inodes. eg. xfs_setattr, xfs_create. + * So, if we couldn't purge all the dquots from the filesystem, + * we can't get rid of the incore data structures. + */ + while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype))) + delay(10 * nculprits); + + /* + * Transactions that had started before ACTIVE state bit was cleared + * could have logged many dquots, so they'd have higher LSNs than + * the first QUOTAOFF log record does. If we happen to crash when + * the tail of the log has gone past the QUOTAOFF record, but + * before the last dquot modification, those dquots __will__ + * recover, and that's not good. + * + * So, we have QUOTAOFF start and end logitems; the start + * logitem won't get overwritten until the end logitem appears... + */ + error = xfs_qm_log_quotaoff_end(mp, qoffstart, flags); + if (error) { + /* We're screwed now. Shutdown is the only option. */ + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); + goto out_unlock; + } + + /* + * If quotas is completely disabled, close shop. + */ + if (((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET1) || + ((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET2)) { + mutex_unlock(&q->qi_quotaofflock); + xfs_qm_destroy_quotainfo(mp); + return (0); + } + + /* + * Release our quotainode references if we don't need them anymore. + */ + if ((dqtype & XFS_QMOPT_UQUOTA) && q->qi_uquotaip) { + IRELE(q->qi_uquotaip); + q->qi_uquotaip = NULL; + } + if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && q->qi_gquotaip) { + IRELE(q->qi_gquotaip); + q->qi_gquotaip = NULL; + } + +out_unlock: + mutex_unlock(&q->qi_quotaofflock); + return error; +} + +STATIC int +xfs_qm_scall_trunc_qfile( + struct xfs_mount *mp, + xfs_ino_t ino) +{ + struct xfs_inode *ip; + struct xfs_trans *tp; + int error; + + if (ino == NULLFSINO) + return 0; + + error = xfs_iget(mp, NULL, ino, 0, 0, &ip); + if (error) + return error; + + xfs_ilock(ip, XFS_IOLOCK_EXCL); + + tp = xfs_trans_alloc(mp, XFS_TRANS_TRUNCATE_FILE); + error = xfs_trans_reserve(tp, 0, XFS_ITRUNCATE_LOG_RES(mp), 0, + XFS_TRANS_PERM_LOG_RES, + XFS_ITRUNCATE_LOG_COUNT); + if (error) { + xfs_trans_cancel(tp, 0); + xfs_iunlock(ip, XFS_IOLOCK_EXCL); + goto out_put; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip); + + error = xfs_itruncate_data(&tp, ip, 0); + if (error) { + xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES | + XFS_TRANS_ABORT); + goto out_unlock; + } + + xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES); + +out_unlock: + xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); +out_put: + IRELE(ip); + return error; +} + +int +xfs_qm_scall_trunc_qfiles( + xfs_mount_t *mp, + uint flags) +{ + int error = 0, error2 = 0; + + if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { + xfs_debug(mp, "%s: flags=%x m_qflags=%x\n", + __func__, flags, mp->m_qflags); + return XFS_ERROR(EINVAL); + } + + if (flags & XFS_DQ_USER) + error = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_uquotino); + if (flags & (XFS_DQ_GROUP|XFS_DQ_PROJ)) + error2 = xfs_qm_scall_trunc_qfile(mp, mp->m_sb.sb_gquotino); + + return error ? error : error2; +} + +/* + * Switch on (a given) quota enforcement for a filesystem. This takes + * effect immediately. + * (Switching on quota accounting must be done at mount time.) + */ +int +xfs_qm_scall_quotaon( + xfs_mount_t *mp, + uint flags) +{ + int error; + uint qf; + __int64_t sbflags; + + flags &= (XFS_ALL_QUOTA_ACCT | XFS_ALL_QUOTA_ENFD); + /* + * Switching on quota accounting must be done at mount time. + */ + flags &= ~(XFS_ALL_QUOTA_ACCT); + + sbflags = 0; + + if (flags == 0) { + xfs_debug(mp, "%s: zero flags, m_qflags=%x\n", + __func__, mp->m_qflags); + return XFS_ERROR(EINVAL); + } + + /* No fs can turn on quotas with a delayed effect */ + ASSERT((flags & XFS_ALL_QUOTA_ACCT) == 0); + + /* + * Can't enforce without accounting. We check the superblock + * qflags here instead of m_qflags because rootfs can have + * quota acct on ondisk without m_qflags' knowing. + */ + if (((flags & XFS_UQUOTA_ACCT) == 0 && + (mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT) == 0 && + (flags & XFS_UQUOTA_ENFD)) + || + ((flags & XFS_PQUOTA_ACCT) == 0 && + (mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT) == 0 && + (flags & XFS_GQUOTA_ACCT) == 0 && + (mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) == 0 && + (flags & XFS_OQUOTA_ENFD))) { + xfs_debug(mp, + "%s: Can't enforce without acct, flags=%x sbflags=%x\n", + __func__, flags, mp->m_sb.sb_qflags); + return XFS_ERROR(EINVAL); + } + /* + * If everything's up to-date incore, then don't waste time. + */ + if ((mp->m_qflags & flags) == flags) + return XFS_ERROR(EEXIST); + + /* + * Change sb_qflags on disk but not incore mp->qflags + * if this is the root filesystem. + */ + spin_lock(&mp->m_sb_lock); + qf = mp->m_sb.sb_qflags; + mp->m_sb.sb_qflags = qf | flags; + spin_unlock(&mp->m_sb_lock); + + /* + * There's nothing to change if it's the same. + */ + if ((qf & flags) == flags && sbflags == 0) + return XFS_ERROR(EEXIST); + sbflags |= XFS_SB_QFLAGS; + + if ((error = xfs_qm_write_sb_changes(mp, sbflags))) + return (error); + /* + * If we aren't trying to switch on quota enforcement, we are done. + */ + if (((mp->m_sb.sb_qflags & XFS_UQUOTA_ACCT) != + (mp->m_qflags & XFS_UQUOTA_ACCT)) || + ((mp->m_sb.sb_qflags & XFS_PQUOTA_ACCT) != + (mp->m_qflags & XFS_PQUOTA_ACCT)) || + ((mp->m_sb.sb_qflags & XFS_GQUOTA_ACCT) != + (mp->m_qflags & XFS_GQUOTA_ACCT)) || + (flags & XFS_ALL_QUOTA_ENFD) == 0) + return (0); + + if (! XFS_IS_QUOTA_RUNNING(mp)) + return XFS_ERROR(ESRCH); + + /* + * Switch on quota enforcement in core. + */ + mutex_lock(&mp->m_quotainfo->qi_quotaofflock); + mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD); + mutex_unlock(&mp->m_quotainfo->qi_quotaofflock); + + return (0); +} + + +/* + * Return quota status information, such as uquota-off, enforcements, etc. + */ +int +xfs_qm_scall_getqstat( + struct xfs_mount *mp, + struct fs_quota_stat *out) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + struct xfs_inode *uip, *gip; + boolean_t tempuqip, tempgqip; + + uip = gip = NULL; + tempuqip = tempgqip = B_FALSE; + memset(out, 0, sizeof(fs_quota_stat_t)); + + out->qs_version = FS_QSTAT_VERSION; + if (!xfs_sb_version_hasquota(&mp->m_sb)) { + out->qs_uquota.qfs_ino = NULLFSINO; + out->qs_gquota.qfs_ino = NULLFSINO; + return (0); + } + out->qs_flags = (__uint16_t) xfs_qm_export_flags(mp->m_qflags & + (XFS_ALL_QUOTA_ACCT| + XFS_ALL_QUOTA_ENFD)); + out->qs_pad = 0; + out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino; + out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino; + + if (q) { + uip = q->qi_uquotaip; + gip = q->qi_gquotaip; + } + if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, + 0, 0, &uip) == 0) + tempuqip = B_TRUE; + } + if (!gip && mp->m_sb.sb_gquotino != NULLFSINO) { + if (xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, + 0, 0, &gip) == 0) + tempgqip = B_TRUE; + } + if (uip) { + out->qs_uquota.qfs_nblks = uip->i_d.di_nblocks; + out->qs_uquota.qfs_nextents = uip->i_d.di_nextents; + if (tempuqip) + IRELE(uip); + } + if (gip) { + out->qs_gquota.qfs_nblks = gip->i_d.di_nblocks; + out->qs_gquota.qfs_nextents = gip->i_d.di_nextents; + if (tempgqip) + IRELE(gip); + } + if (q) { + out->qs_incoredqs = q->qi_dquots; + out->qs_btimelimit = q->qi_btimelimit; + out->qs_itimelimit = q->qi_itimelimit; + out->qs_rtbtimelimit = q->qi_rtbtimelimit; + out->qs_bwarnlimit = q->qi_bwarnlimit; + out->qs_iwarnlimit = q->qi_iwarnlimit; + } + return 0; +} + +#define XFS_DQ_MASK \ + (FS_DQ_LIMIT_MASK | FS_DQ_TIMER_MASK | FS_DQ_WARNS_MASK) + +/* + * Adjust quota limits, and start/stop timers accordingly. + */ +int +xfs_qm_scall_setqlim( + xfs_mount_t *mp, + xfs_dqid_t id, + uint type, + fs_disk_quota_t *newlim) +{ + struct xfs_quotainfo *q = mp->m_quotainfo; + xfs_disk_dquot_t *ddq; + xfs_dquot_t *dqp; + xfs_trans_t *tp; + int error; + xfs_qcnt_t hard, soft; + + if (newlim->d_fieldmask & ~XFS_DQ_MASK) + return EINVAL; + if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0) + return 0; + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM); + if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128, + 0, 0, XFS_DEFAULT_LOG_COUNT))) { + xfs_trans_cancel(tp, 0); + return (error); + } + + /* + * We don't want to race with a quotaoff so take the quotaoff lock. + * (We don't hold an inode lock, so there's nothing else to stop + * a quotaoff from happening). (XXXThis doesn't currently happen + * because we take the vfslock before calling xfs_qm_sysent). + */ + mutex_lock(&q->qi_quotaofflock); + + /* + * Get the dquot (locked), and join it to the transaction. + * Allocate the dquot if this doesn't exist. + */ + if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) { + xfs_trans_cancel(tp, XFS_TRANS_ABORT); + ASSERT(error != ENOENT); + goto out_unlock; + } + xfs_trans_dqjoin(tp, dqp); + ddq = &dqp->q_core; + + /* + * Make sure that hardlimits are >= soft limits before changing. + */ + hard = (newlim->d_fieldmask & FS_DQ_BHARD) ? + (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_hardlimit) : + be64_to_cpu(ddq->d_blk_hardlimit); + soft = (newlim->d_fieldmask & FS_DQ_BSOFT) ? + (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_softlimit) : + be64_to_cpu(ddq->d_blk_softlimit); + if (hard == 0 || hard >= soft) { + ddq->d_blk_hardlimit = cpu_to_be64(hard); + ddq->d_blk_softlimit = cpu_to_be64(soft); + if (id == 0) { + q->qi_bhardlimit = hard; + q->qi_bsoftlimit = soft; + } + } else { + xfs_debug(mp, "blkhard %Ld < blksoft %Ld\n", hard, soft); + } + hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ? + (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) : + be64_to_cpu(ddq->d_rtb_hardlimit); + soft = (newlim->d_fieldmask & FS_DQ_RTBSOFT) ? + (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_softlimit) : + be64_to_cpu(ddq->d_rtb_softlimit); + if (hard == 0 || hard >= soft) { + ddq->d_rtb_hardlimit = cpu_to_be64(hard); + ddq->d_rtb_softlimit = cpu_to_be64(soft); + if (id == 0) { + q->qi_rtbhardlimit = hard; + q->qi_rtbsoftlimit = soft; + } + } else { + xfs_debug(mp, "rtbhard %Ld < rtbsoft %Ld\n", hard, soft); + } + + hard = (newlim->d_fieldmask & FS_DQ_IHARD) ? + (xfs_qcnt_t) newlim->d_ino_hardlimit : + be64_to_cpu(ddq->d_ino_hardlimit); + soft = (newlim->d_fieldmask & FS_DQ_ISOFT) ? + (xfs_qcnt_t) newlim->d_ino_softlimit : + be64_to_cpu(ddq->d_ino_softlimit); + if (hard == 0 || hard >= soft) { + ddq->d_ino_hardlimit = cpu_to_be64(hard); + ddq->d_ino_softlimit = cpu_to_be64(soft); + if (id == 0) { + q->qi_ihardlimit = hard; + q->qi_isoftlimit = soft; + } + } else { + xfs_debug(mp, "ihard %Ld < isoft %Ld\n", hard, soft); + } + + /* + * Update warnings counter(s) if requested + */ + if (newlim->d_fieldmask & FS_DQ_BWARNS) + ddq->d_bwarns = cpu_to_be16(newlim->d_bwarns); + if (newlim->d_fieldmask & FS_DQ_IWARNS) + ddq->d_iwarns = cpu_to_be16(newlim->d_iwarns); + if (newlim->d_fieldmask & FS_DQ_RTBWARNS) + ddq->d_rtbwarns = cpu_to_be16(newlim->d_rtbwarns); + + if (id == 0) { + /* + * Timelimits for the super user set the relative time + * the other users can be over quota for this file system. + * If it is zero a default is used. Ditto for the default + * soft and hard limit values (already done, above), and + * for warnings. + */ + if (newlim->d_fieldmask & FS_DQ_BTIMER) { + q->qi_btimelimit = newlim->d_btimer; + ddq->d_btimer = cpu_to_be32(newlim->d_btimer); + } + if (newlim->d_fieldmask & FS_DQ_ITIMER) { + q->qi_itimelimit = newlim->d_itimer; + ddq->d_itimer = cpu_to_be32(newlim->d_itimer); + } + if (newlim->d_fieldmask & FS_DQ_RTBTIMER) { + q->qi_rtbtimelimit = newlim->d_rtbtimer; + ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer); + } + if (newlim->d_fieldmask & FS_DQ_BWARNS) + q->qi_bwarnlimit = newlim->d_bwarns; + if (newlim->d_fieldmask & FS_DQ_IWARNS) + q->qi_iwarnlimit = newlim->d_iwarns; + if (newlim->d_fieldmask & FS_DQ_RTBWARNS) + q->qi_rtbwarnlimit = newlim->d_rtbwarns; + } else { + /* + * If the user is now over quota, start the timelimit. + * The user will not be 'warned'. + * Note that we keep the timers ticking, whether enforcement + * is on or off. We don't really want to bother with iterating + * over all ondisk dquots and turning the timers on/off. + */ + xfs_qm_adjust_dqtimers(mp, ddq); + } + dqp->dq_flags |= XFS_DQ_DIRTY; + xfs_trans_log_dquot(tp, dqp); + + error = xfs_trans_commit(tp, 0); + xfs_qm_dqrele(dqp); + + out_unlock: + mutex_unlock(&q->qi_quotaofflock); + return error; +} + +int +xfs_qm_scall_getquota( + xfs_mount_t *mp, + xfs_dqid_t id, + uint type, + fs_disk_quota_t *out) +{ + xfs_dquot_t *dqp; + int error; + + /* + * Try to get the dquot. We don't want it allocated on disk, so + * we aren't passing the XFS_QMOPT_DOALLOC flag. If it doesn't + * exist, we'll get ENOENT back. + */ + if ((error = xfs_qm_dqget(mp, NULL, id, type, 0, &dqp))) { + return (error); + } + + /* + * If everything's NULL, this dquot doesn't quite exist as far as + * our utility programs are concerned. + */ + if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) { + xfs_qm_dqput(dqp); + return XFS_ERROR(ENOENT); + } + /* + * Convert the disk dquot to the exportable format + */ + xfs_qm_export_dquot(mp, &dqp->q_core, out); + xfs_qm_dqput(dqp); + return (error ? XFS_ERROR(EFAULT) : 0); +} + + +STATIC int +xfs_qm_log_quotaoff_end( + xfs_mount_t *mp, + xfs_qoff_logitem_t *startqoff, + uint flags) +{ + xfs_trans_t *tp; + int error; + xfs_qoff_logitem_t *qoffi; + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF_END); + + if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_qoff_logitem_t) * 2, + 0, 0, XFS_DEFAULT_LOG_COUNT))) { + xfs_trans_cancel(tp, 0); + return (error); + } + + qoffi = xfs_trans_get_qoff_item(tp, startqoff, + flags & XFS_ALL_QUOTA_ACCT); + xfs_trans_log_quotaoff_item(tp, qoffi); + + /* + * We have to make sure that the transaction is secure on disk before we + * return and actually stop quota accounting. So, make it synchronous. + * We don't care about quotoff's performance. + */ + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp, 0); + return (error); +} + + +STATIC int +xfs_qm_log_quotaoff( + xfs_mount_t *mp, + xfs_qoff_logitem_t **qoffstartp, + uint flags) +{ + xfs_trans_t *tp; + int error; + xfs_qoff_logitem_t *qoffi=NULL; + uint oldsbqflag=0; + + tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QUOTAOFF); + if ((error = xfs_trans_reserve(tp, 0, + sizeof(xfs_qoff_logitem_t) * 2 + + mp->m_sb.sb_sectsize + 128, + 0, + 0, + XFS_DEFAULT_LOG_COUNT))) { + goto error0; + } + + qoffi = xfs_trans_get_qoff_item(tp, NULL, flags & XFS_ALL_QUOTA_ACCT); + xfs_trans_log_quotaoff_item(tp, qoffi); + + spin_lock(&mp->m_sb_lock); + oldsbqflag = mp->m_sb.sb_qflags; + mp->m_sb.sb_qflags = (mp->m_qflags & ~(flags)) & XFS_MOUNT_QUOTA_ALL; + spin_unlock(&mp->m_sb_lock); + + xfs_mod_sb(tp, XFS_SB_QFLAGS); + + /* + * We have to make sure that the transaction is secure on disk before we + * return and actually stop quota accounting. So, make it synchronous. + * We don't care about quotoff's performance. + */ + xfs_trans_set_sync(tp); + error = xfs_trans_commit(tp, 0); + +error0: + if (error) { + xfs_trans_cancel(tp, 0); + /* + * No one else is modifying sb_qflags, so this is OK. + * We still hold the quotaofflock. + */ + spin_lock(&mp->m_sb_lock); + mp->m_sb.sb_qflags = oldsbqflag; + spin_unlock(&mp->m_sb_lock); + } + *qoffstartp = qoffi; + return (error); +} + + +/* + * Translate an internal style on-disk-dquot to the exportable format. + * The main differences are that the counters/limits are all in Basic + * Blocks (BBs) instead of the internal FSBs, and all on-disk data has + * to be converted to the native endianness. + */ +STATIC void +xfs_qm_export_dquot( + xfs_mount_t *mp, + xfs_disk_dquot_t *src, + struct fs_disk_quota *dst) +{ + memset(dst, 0, sizeof(*dst)); + dst->d_version = FS_DQUOT_VERSION; /* different from src->d_version */ + dst->d_flags = xfs_qm_export_qtype_flags(src->d_flags); + dst->d_id = be32_to_cpu(src->d_id); + dst->d_blk_hardlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_hardlimit)); + dst->d_blk_softlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_softlimit)); + dst->d_ino_hardlimit = be64_to_cpu(src->d_ino_hardlimit); + dst->d_ino_softlimit = be64_to_cpu(src->d_ino_softlimit); + dst->d_bcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_bcount)); + dst->d_icount = be64_to_cpu(src->d_icount); + dst->d_btimer = be32_to_cpu(src->d_btimer); + dst->d_itimer = be32_to_cpu(src->d_itimer); + dst->d_iwarns = be16_to_cpu(src->d_iwarns); + dst->d_bwarns = be16_to_cpu(src->d_bwarns); + dst->d_rtb_hardlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_hardlimit)); + dst->d_rtb_softlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_softlimit)); + dst->d_rtbcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtbcount)); + dst->d_rtbtimer = be32_to_cpu(src->d_rtbtimer); + dst->d_rtbwarns = be16_to_cpu(src->d_rtbwarns); + + /* + * Internally, we don't reset all the timers when quota enforcement + * gets turned off. No need to confuse the user level code, + * so return zeroes in that case. + */ + if ((!XFS_IS_UQUOTA_ENFORCED(mp) && src->d_flags == XFS_DQ_USER) || + (!XFS_IS_OQUOTA_ENFORCED(mp) && + (src->d_flags & (XFS_DQ_PROJ | XFS_DQ_GROUP)))) { + dst->d_btimer = 0; + dst->d_itimer = 0; + dst->d_rtbtimer = 0; + } + +#ifdef DEBUG + if (((XFS_IS_UQUOTA_ENFORCED(mp) && dst->d_flags == FS_USER_QUOTA) || + (XFS_IS_OQUOTA_ENFORCED(mp) && + (dst->d_flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)))) && + dst->d_id != 0) { + if (((int) dst->d_bcount >= (int) dst->d_blk_softlimit) && + (dst->d_blk_softlimit > 0)) { + ASSERT(dst->d_btimer != 0); + } + if (((int) dst->d_icount >= (int) dst->d_ino_softlimit) && + (dst->d_ino_softlimit > 0)) { + ASSERT(dst->d_itimer != 0); + } + } +#endif +} + +STATIC uint +xfs_qm_export_qtype_flags( + uint flags) +{ + /* + * Can't be more than one, or none. + */ + ASSERT((flags & (FS_PROJ_QUOTA | FS_USER_QUOTA)) != + (FS_PROJ_QUOTA | FS_USER_QUOTA)); + ASSERT((flags & (FS_PROJ_QUOTA | FS_GROUP_QUOTA)) != + (FS_PROJ_QUOTA | FS_GROUP_QUOTA)); + ASSERT((flags & (FS_USER_QUOTA | FS_GROUP_QUOTA)) != + (FS_USER_QUOTA | FS_GROUP_QUOTA)); + ASSERT((flags & (FS_PROJ_QUOTA|FS_USER_QUOTA|FS_GROUP_QUOTA)) != 0); + + return (flags & XFS_DQ_USER) ? + FS_USER_QUOTA : (flags & XFS_DQ_PROJ) ? + FS_PROJ_QUOTA : FS_GROUP_QUOTA; +} + +STATIC uint +xfs_qm_export_flags( + uint flags) +{ + uint uflags; + + uflags = 0; + if (flags & XFS_UQUOTA_ACCT) + uflags |= FS_QUOTA_UDQ_ACCT; + if (flags & XFS_PQUOTA_ACCT) + uflags |= FS_QUOTA_PDQ_ACCT; + if (flags & XFS_GQUOTA_ACCT) + uflags |= FS_QUOTA_GDQ_ACCT; + if (flags & XFS_UQUOTA_ENFD) + uflags |= FS_QUOTA_UDQ_ENFD; + if (flags & (XFS_OQUOTA_ENFD)) { + uflags |= (flags & XFS_GQUOTA_ACCT) ? + FS_QUOTA_GDQ_ENFD : FS_QUOTA_PDQ_ENFD; + } + return (uflags); +} + + +STATIC int +xfs_dqrele_inode( + struct xfs_inode *ip, + struct xfs_perag *pag, + int flags) +{ + /* skip quota inodes */ + if (ip == ip->i_mount->m_quotainfo->qi_uquotaip || + ip == ip->i_mount->m_quotainfo->qi_gquotaip) { + ASSERT(ip->i_udquot == NULL); + ASSERT(ip->i_gdquot == NULL); + return 0; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + if ((flags & XFS_UQUOTA_ACCT) && ip->i_udquot) { + xfs_qm_dqrele(ip->i_udquot); + ip->i_udquot = NULL; + } + if (flags & (XFS_PQUOTA_ACCT|XFS_GQUOTA_ACCT) && ip->i_gdquot) { + xfs_qm_dqrele(ip->i_gdquot); + ip->i_gdquot = NULL; + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return 0; +} + + +/* + * Go thru all the inodes in the file system, releasing their dquots. + * + * Note that the mount structure gets modified to indicate that quotas are off + * AFTER this, in the case of quotaoff. + */ +void +xfs_qm_dqrele_all_inodes( + struct xfs_mount *mp, + uint flags) +{ + ASSERT(mp->m_quotainfo); + xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags); +} diff --git a/fs/xfs/xfs_quota_priv.h b/fs/xfs/xfs_quota_priv.h new file mode 100644 index 000000000000..94a3d927d716 --- /dev/null +++ b/fs/xfs/xfs_quota_priv.h @@ -0,0 +1,53 @@ +/* + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_QUOTA_PRIV_H__ +#define __XFS_QUOTA_PRIV_H__ + +/* + * Number of bmaps that we ask from bmapi when doing a quotacheck. + * We make this restriction to keep the memory usage to a minimum. + */ +#define XFS_DQITER_MAP_SIZE 10 + +/* + * Hash into a bucket in the dquot hash table, based on . + */ +#define XFS_DQ_HASHVAL(mp, id) (((__psunsigned_t)(mp) + \ + (__psunsigned_t)(id)) & \ + (xfs_Gqm->qm_dqhashmask - 1)) +#define XFS_DQ_HASH(mp, id, type) (type == XFS_DQ_USER ? \ + (xfs_Gqm->qm_usr_dqhtable + \ + XFS_DQ_HASHVAL(mp, id)) : \ + (xfs_Gqm->qm_grp_dqhtable + \ + XFS_DQ_HASHVAL(mp, id))) +#define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \ + !dqp->q_core.d_blk_hardlimit && \ + !dqp->q_core.d_blk_softlimit && \ + !dqp->q_core.d_rtb_hardlimit && \ + !dqp->q_core.d_rtb_softlimit && \ + !dqp->q_core.d_ino_hardlimit && \ + !dqp->q_core.d_ino_softlimit && \ + !dqp->q_core.d_bcount && \ + !dqp->q_core.d_rtbcount && \ + !dqp->q_core.d_icount) + +#define DQFLAGTO_TYPESTR(d) (((d)->dq_flags & XFS_DQ_USER) ? "USR" : \ + (((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \ + (((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???"))) + +#endif /* __XFS_QUOTA_PRIV_H__ */ diff --git a/fs/xfs/xfs_quotaops.c b/fs/xfs/xfs_quotaops.c new file mode 100644 index 000000000000..7e76f537abb7 --- /dev/null +++ b/fs/xfs/xfs_quotaops.c @@ -0,0 +1,139 @@ +/* + * Copyright (c) 2008, Christoph Hellwig + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_sb.h" +#include "xfs_inum.h" +#include "xfs_log.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_quota.h" +#include "xfs_trans.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_qm.h" +#include + + +STATIC int +xfs_quota_type(int type) +{ + switch (type) { + case USRQUOTA: + return XFS_DQ_USER; + case GRPQUOTA: + return XFS_DQ_GROUP; + default: + return XFS_DQ_PROJ; + } +} + +STATIC int +xfs_fs_get_xstate( + struct super_block *sb, + struct fs_quota_stat *fqs) +{ + struct xfs_mount *mp = XFS_M(sb); + + if (!XFS_IS_QUOTA_RUNNING(mp)) + return -ENOSYS; + return -xfs_qm_scall_getqstat(mp, fqs); +} + +STATIC int +xfs_fs_set_xstate( + struct super_block *sb, + unsigned int uflags, + int op) +{ + struct xfs_mount *mp = XFS_M(sb); + unsigned int flags = 0; + + if (sb->s_flags & MS_RDONLY) + return -EROFS; + if (op != Q_XQUOTARM && !XFS_IS_QUOTA_RUNNING(mp)) + return -ENOSYS; + + if (uflags & FS_QUOTA_UDQ_ACCT) + flags |= XFS_UQUOTA_ACCT; + if (uflags & FS_QUOTA_PDQ_ACCT) + flags |= XFS_PQUOTA_ACCT; + if (uflags & FS_QUOTA_GDQ_ACCT) + flags |= XFS_GQUOTA_ACCT; + if (uflags & FS_QUOTA_UDQ_ENFD) + flags |= XFS_UQUOTA_ENFD; + if (uflags & (FS_QUOTA_PDQ_ENFD|FS_QUOTA_GDQ_ENFD)) + flags |= XFS_OQUOTA_ENFD; + + switch (op) { + case Q_XQUOTAON: + return -xfs_qm_scall_quotaon(mp, flags); + case Q_XQUOTAOFF: + if (!XFS_IS_QUOTA_ON(mp)) + return -EINVAL; + return -xfs_qm_scall_quotaoff(mp, flags); + case Q_XQUOTARM: + if (XFS_IS_QUOTA_ON(mp)) + return -EINVAL; + return -xfs_qm_scall_trunc_qfiles(mp, flags); + } + + return -EINVAL; +} + +STATIC int +xfs_fs_get_dqblk( + struct super_block *sb, + int type, + qid_t id, + struct fs_disk_quota *fdq) +{ + struct xfs_mount *mp = XFS_M(sb); + + if (!XFS_IS_QUOTA_RUNNING(mp)) + return -ENOSYS; + if (!XFS_IS_QUOTA_ON(mp)) + return -ESRCH; + + return -xfs_qm_scall_getquota(mp, id, xfs_quota_type(type), fdq); +} + +STATIC int +xfs_fs_set_dqblk( + struct super_block *sb, + int type, + qid_t id, + struct fs_disk_quota *fdq) +{ + struct xfs_mount *mp = XFS_M(sb); + + if (sb->s_flags & MS_RDONLY) + return -EROFS; + if (!XFS_IS_QUOTA_RUNNING(mp)) + return -ENOSYS; + if (!XFS_IS_QUOTA_ON(mp)) + return -ESRCH; + + return -xfs_qm_scall_setqlim(mp, id, xfs_quota_type(type), fdq); +} + +const struct quotactl_ops xfs_quotactl_operations = { + .get_xstate = xfs_fs_get_xstate, + .set_xstate = xfs_fs_set_xstate, + .get_dqblk = xfs_fs_get_dqblk, + .set_dqblk = xfs_fs_set_dqblk, +}; diff --git a/fs/xfs/xfs_stats.c b/fs/xfs/xfs_stats.c new file mode 100644 index 000000000000..76fdc5861932 --- /dev/null +++ b/fs/xfs/xfs_stats.c @@ -0,0 +1,122 @@ +/* + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include + +DEFINE_PER_CPU(struct xfsstats, xfsstats); + +static int xfs_stat_proc_show(struct seq_file *m, void *v) +{ + int c, i, j, val; + __uint64_t xs_xstrat_bytes = 0; + __uint64_t xs_write_bytes = 0; + __uint64_t xs_read_bytes = 0; + + static const struct xstats_entry { + char *desc; + int endpoint; + } xstats[] = { + { "extent_alloc", XFSSTAT_END_EXTENT_ALLOC }, + { "abt", XFSSTAT_END_ALLOC_BTREE }, + { "blk_map", XFSSTAT_END_BLOCK_MAPPING }, + { "bmbt", XFSSTAT_END_BLOCK_MAP_BTREE }, + { "dir", XFSSTAT_END_DIRECTORY_OPS }, + { "trans", XFSSTAT_END_TRANSACTIONS }, + { "ig", XFSSTAT_END_INODE_OPS }, + { "log", XFSSTAT_END_LOG_OPS }, + { "push_ail", XFSSTAT_END_TAIL_PUSHING }, + { "xstrat", XFSSTAT_END_WRITE_CONVERT }, + { "rw", XFSSTAT_END_READ_WRITE_OPS }, + { "attr", XFSSTAT_END_ATTRIBUTE_OPS }, + { "icluster", XFSSTAT_END_INODE_CLUSTER }, + { "vnodes", XFSSTAT_END_VNODE_OPS }, + { "buf", XFSSTAT_END_BUF }, + { "abtb2", XFSSTAT_END_ABTB_V2 }, + { "abtc2", XFSSTAT_END_ABTC_V2 }, + { "bmbt2", XFSSTAT_END_BMBT_V2 }, + { "ibt2", XFSSTAT_END_IBT_V2 }, + }; + + /* Loop over all stats groups */ + for (i=j = 0; i < ARRAY_SIZE(xstats); i++) { + seq_printf(m, "%s", xstats[i].desc); + /* inner loop does each group */ + while (j < xstats[i].endpoint) { + val = 0; + /* sum over all cpus */ + for_each_possible_cpu(c) + val += *(((__u32*)&per_cpu(xfsstats, c) + j)); + seq_printf(m, " %u", val); + j++; + } + seq_putc(m, '\n'); + } + /* extra precision counters */ + for_each_possible_cpu(i) { + xs_xstrat_bytes += per_cpu(xfsstats, i).xs_xstrat_bytes; + xs_write_bytes += per_cpu(xfsstats, i).xs_write_bytes; + xs_read_bytes += per_cpu(xfsstats, i).xs_read_bytes; + } + + seq_printf(m, "xpc %Lu %Lu %Lu\n", + xs_xstrat_bytes, xs_write_bytes, xs_read_bytes); + seq_printf(m, "debug %u\n", +#if defined(DEBUG) + 1); +#else + 0); +#endif + return 0; +} + +static int xfs_stat_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, xfs_stat_proc_show, NULL); +} + +static const struct file_operations xfs_stat_proc_fops = { + .owner = THIS_MODULE, + .open = xfs_stat_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +int +xfs_init_procfs(void) +{ + if (!proc_mkdir("fs/xfs", NULL)) + goto out; + + if (!proc_create("fs/xfs/stat", 0, NULL, + &xfs_stat_proc_fops)) + goto out_remove_entry; + return 0; + + out_remove_entry: + remove_proc_entry("fs/xfs", NULL); + out: + return -ENOMEM; +} + +void +xfs_cleanup_procfs(void) +{ + remove_proc_entry("fs/xfs/stat", NULL); + remove_proc_entry("fs/xfs", NULL); +} diff --git a/fs/xfs/xfs_stats.h b/fs/xfs/xfs_stats.h new file mode 100644 index 000000000000..736854b1ca1a --- /dev/null +++ b/fs/xfs/xfs_stats.h @@ -0,0 +1,223 @@ +/* + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_STATS_H__ +#define __XFS_STATS_H__ + + +#if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF) + +#include + +/* + * XFS global statistics + */ +struct xfsstats { +# define XFSSTAT_END_EXTENT_ALLOC 4 + __uint32_t xs_allocx; + __uint32_t xs_allocb; + __uint32_t xs_freex; + __uint32_t xs_freeb; +# define XFSSTAT_END_ALLOC_BTREE (XFSSTAT_END_EXTENT_ALLOC+4) + __uint32_t xs_abt_lookup; + __uint32_t xs_abt_compare; + __uint32_t xs_abt_insrec; + __uint32_t xs_abt_delrec; +# define XFSSTAT_END_BLOCK_MAPPING (XFSSTAT_END_ALLOC_BTREE+7) + __uint32_t xs_blk_mapr; + __uint32_t xs_blk_mapw; + __uint32_t xs_blk_unmap; + __uint32_t xs_add_exlist; + __uint32_t xs_del_exlist; + __uint32_t xs_look_exlist; + __uint32_t xs_cmp_exlist; +# define XFSSTAT_END_BLOCK_MAP_BTREE (XFSSTAT_END_BLOCK_MAPPING+4) + __uint32_t xs_bmbt_lookup; + __uint32_t xs_bmbt_compare; + __uint32_t xs_bmbt_insrec; + __uint32_t xs_bmbt_delrec; +# define XFSSTAT_END_DIRECTORY_OPS (XFSSTAT_END_BLOCK_MAP_BTREE+4) + __uint32_t xs_dir_lookup; + __uint32_t xs_dir_create; + __uint32_t xs_dir_remove; + __uint32_t xs_dir_getdents; +# define XFSSTAT_END_TRANSACTIONS (XFSSTAT_END_DIRECTORY_OPS+3) + __uint32_t xs_trans_sync; + __uint32_t xs_trans_async; + __uint32_t xs_trans_empty; +# define XFSSTAT_END_INODE_OPS (XFSSTAT_END_TRANSACTIONS+7) + __uint32_t xs_ig_attempts; + __uint32_t xs_ig_found; + __uint32_t xs_ig_frecycle; + __uint32_t xs_ig_missed; + __uint32_t xs_ig_dup; + __uint32_t xs_ig_reclaims; + __uint32_t xs_ig_attrchg; +# define XFSSTAT_END_LOG_OPS (XFSSTAT_END_INODE_OPS+5) + __uint32_t xs_log_writes; + __uint32_t xs_log_blocks; + __uint32_t xs_log_noiclogs; + __uint32_t xs_log_force; + __uint32_t xs_log_force_sleep; +# define XFSSTAT_END_TAIL_PUSHING (XFSSTAT_END_LOG_OPS+10) + __uint32_t xs_try_logspace; + __uint32_t xs_sleep_logspace; + __uint32_t xs_push_ail; + __uint32_t xs_push_ail_success; + __uint32_t xs_push_ail_pushbuf; + __uint32_t xs_push_ail_pinned; + __uint32_t xs_push_ail_locked; + __uint32_t xs_push_ail_flushing; + __uint32_t xs_push_ail_restarts; + __uint32_t xs_push_ail_flush; +# define XFSSTAT_END_WRITE_CONVERT (XFSSTAT_END_TAIL_PUSHING+2) + __uint32_t xs_xstrat_quick; + __uint32_t xs_xstrat_split; +# define XFSSTAT_END_READ_WRITE_OPS (XFSSTAT_END_WRITE_CONVERT+2) + __uint32_t xs_write_calls; + __uint32_t xs_read_calls; +# define XFSSTAT_END_ATTRIBUTE_OPS (XFSSTAT_END_READ_WRITE_OPS+4) + __uint32_t xs_attr_get; + __uint32_t xs_attr_set; + __uint32_t xs_attr_remove; + __uint32_t xs_attr_list; +# define XFSSTAT_END_INODE_CLUSTER (XFSSTAT_END_ATTRIBUTE_OPS+3) + __uint32_t xs_iflush_count; + __uint32_t xs_icluster_flushcnt; + __uint32_t xs_icluster_flushinode; +# define XFSSTAT_END_VNODE_OPS (XFSSTAT_END_INODE_CLUSTER+8) + __uint32_t vn_active; /* # vnodes not on free lists */ + __uint32_t vn_alloc; /* # times vn_alloc called */ + __uint32_t vn_get; /* # times vn_get called */ + __uint32_t vn_hold; /* # times vn_hold called */ + __uint32_t vn_rele; /* # times vn_rele called */ + __uint32_t vn_reclaim; /* # times vn_reclaim called */ + __uint32_t vn_remove; /* # times vn_remove called */ + __uint32_t vn_free; /* # times vn_free called */ +#define XFSSTAT_END_BUF (XFSSTAT_END_VNODE_OPS+9) + __uint32_t xb_get; + __uint32_t xb_create; + __uint32_t xb_get_locked; + __uint32_t xb_get_locked_waited; + __uint32_t xb_busy_locked; + __uint32_t xb_miss_locked; + __uint32_t xb_page_retries; + __uint32_t xb_page_found; + __uint32_t xb_get_read; +/* Version 2 btree counters */ +#define XFSSTAT_END_ABTB_V2 (XFSSTAT_END_BUF+15) + __uint32_t xs_abtb_2_lookup; + __uint32_t xs_abtb_2_compare; + __uint32_t xs_abtb_2_insrec; + __uint32_t xs_abtb_2_delrec; + __uint32_t xs_abtb_2_newroot; + __uint32_t xs_abtb_2_killroot; + __uint32_t xs_abtb_2_increment; + __uint32_t xs_abtb_2_decrement; + __uint32_t xs_abtb_2_lshift; + __uint32_t xs_abtb_2_rshift; + __uint32_t xs_abtb_2_split; + __uint32_t xs_abtb_2_join; + __uint32_t xs_abtb_2_alloc; + __uint32_t xs_abtb_2_free; + __uint32_t xs_abtb_2_moves; +#define XFSSTAT_END_ABTC_V2 (XFSSTAT_END_ABTB_V2+15) + __uint32_t xs_abtc_2_lookup; + __uint32_t xs_abtc_2_compare; + __uint32_t xs_abtc_2_insrec; + __uint32_t xs_abtc_2_delrec; + __uint32_t xs_abtc_2_newroot; + __uint32_t xs_abtc_2_killroot; + __uint32_t xs_abtc_2_increment; + __uint32_t xs_abtc_2_decrement; + __uint32_t xs_abtc_2_lshift; + __uint32_t xs_abtc_2_rshift; + __uint32_t xs_abtc_2_split; + __uint32_t xs_abtc_2_join; + __uint32_t xs_abtc_2_alloc; + __uint32_t xs_abtc_2_free; + __uint32_t xs_abtc_2_moves; +#define XFSSTAT_END_BMBT_V2 (XFSSTAT_END_ABTC_V2+15) + __uint32_t xs_bmbt_2_lookup; + __uint32_t xs_bmbt_2_compare; + __uint32_t xs_bmbt_2_insrec; + __uint32_t xs_bmbt_2_delrec; + __uint32_t xs_bmbt_2_newroot; + __uint32_t xs_bmbt_2_killroot; + __uint32_t xs_bmbt_2_increment; + __uint32_t xs_bmbt_2_decrement; + __uint32_t xs_bmbt_2_lshift; + __uint32_t xs_bmbt_2_rshift; + __uint32_t xs_bmbt_2_split; + __uint32_t xs_bmbt_2_join; + __uint32_t xs_bmbt_2_alloc; + __uint32_t xs_bmbt_2_free; + __uint32_t xs_bmbt_2_moves; +#define XFSSTAT_END_IBT_V2 (XFSSTAT_END_BMBT_V2+15) + __uint32_t xs_ibt_2_lookup; + __uint32_t xs_ibt_2_compare; + __uint32_t xs_ibt_2_insrec; + __uint32_t xs_ibt_2_delrec; + __uint32_t xs_ibt_2_newroot; + __uint32_t xs_ibt_2_killroot; + __uint32_t xs_ibt_2_increment; + __uint32_t xs_ibt_2_decrement; + __uint32_t xs_ibt_2_lshift; + __uint32_t xs_ibt_2_rshift; + __uint32_t xs_ibt_2_split; + __uint32_t xs_ibt_2_join; + __uint32_t xs_ibt_2_alloc; + __uint32_t xs_ibt_2_free; + __uint32_t xs_ibt_2_moves; +/* Extra precision counters */ + __uint64_t xs_xstrat_bytes; + __uint64_t xs_write_bytes; + __uint64_t xs_read_bytes; +}; + +DECLARE_PER_CPU(struct xfsstats, xfsstats); + +/* + * We don't disable preempt, not too worried about poking the + * wrong CPU's stat for now (also aggregated before reporting). + */ +#define XFS_STATS_INC(v) (per_cpu(xfsstats, current_cpu()).v++) +#define XFS_STATS_DEC(v) (per_cpu(xfsstats, current_cpu()).v--) +#define XFS_STATS_ADD(v, inc) (per_cpu(xfsstats, current_cpu()).v += (inc)) + +extern int xfs_init_procfs(void); +extern void xfs_cleanup_procfs(void); + + +#else /* !CONFIG_PROC_FS */ + +# define XFS_STATS_INC(count) +# define XFS_STATS_DEC(count) +# define XFS_STATS_ADD(count, inc) + +static inline int xfs_init_procfs(void) +{ + return 0; +} + +static inline void xfs_cleanup_procfs(void) +{ +} + +#endif /* !CONFIG_PROC_FS */ + +#endif /* __XFS_STATS_H__ */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c new file mode 100644 index 000000000000..9a72dda58bd0 --- /dev/null +++ b/fs/xfs/xfs_super.c @@ -0,0 +1,1773 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "xfs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_dir2.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_itable.h" +#include "xfs_fsops.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_utils.h" +#include "xfs_vnodeops.h" +#include "xfs_log_priv.h" +#include "xfs_trans_priv.h" +#include "xfs_filestream.h" +#include "xfs_da_btree.h" +#include "xfs_extfree_item.h" +#include "xfs_mru_cache.h" +#include "xfs_inode_item.h" +#include "xfs_sync.h" +#include "xfs_trace.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static const struct super_operations xfs_super_operations; +static kmem_zone_t *xfs_ioend_zone; +mempool_t *xfs_ioend_pool; + +#define MNTOPT_LOGBUFS "logbufs" /* number of XFS log buffers */ +#define MNTOPT_LOGBSIZE "logbsize" /* size of XFS log buffers */ +#define MNTOPT_LOGDEV "logdev" /* log device */ +#define MNTOPT_RTDEV "rtdev" /* realtime I/O device */ +#define MNTOPT_BIOSIZE "biosize" /* log2 of preferred buffered io size */ +#define MNTOPT_WSYNC "wsync" /* safe-mode nfs compatible mount */ +#define MNTOPT_NOALIGN "noalign" /* turn off stripe alignment */ +#define MNTOPT_SWALLOC "swalloc" /* turn on stripe width allocation */ +#define MNTOPT_SUNIT "sunit" /* data volume stripe unit */ +#define MNTOPT_SWIDTH "swidth" /* data volume stripe width */ +#define MNTOPT_NOUUID "nouuid" /* ignore filesystem UUID */ +#define MNTOPT_MTPT "mtpt" /* filesystem mount point */ +#define MNTOPT_GRPID "grpid" /* group-ID from parent directory */ +#define MNTOPT_NOGRPID "nogrpid" /* group-ID from current process */ +#define MNTOPT_BSDGROUPS "bsdgroups" /* group-ID from parent directory */ +#define MNTOPT_SYSVGROUPS "sysvgroups" /* group-ID from current process */ +#define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ +#define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ +#define MNTOPT_BARRIER "barrier" /* use writer barriers for log write and + * unwritten extent conversion */ +#define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ +#define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ +#define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ +#define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ +#define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ +#define MNTOPT_NOLARGEIO "nolargeio" /* do not report large I/O sizes + * in stat(). */ +#define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */ +#define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */ +#define MNTOPT_FILESTREAM "filestreams" /* use filestreams allocator */ +#define MNTOPT_QUOTA "quota" /* disk quotas (user) */ +#define MNTOPT_NOQUOTA "noquota" /* no quotas */ +#define MNTOPT_USRQUOTA "usrquota" /* user quota enabled */ +#define MNTOPT_GRPQUOTA "grpquota" /* group quota enabled */ +#define MNTOPT_PRJQUOTA "prjquota" /* project quota enabled */ +#define MNTOPT_UQUOTA "uquota" /* user quota (IRIX variant) */ +#define MNTOPT_GQUOTA "gquota" /* group quota (IRIX variant) */ +#define MNTOPT_PQUOTA "pquota" /* project quota (IRIX variant) */ +#define MNTOPT_UQUOTANOENF "uqnoenforce"/* user quota limit enforcement */ +#define MNTOPT_GQUOTANOENF "gqnoenforce"/* group quota limit enforcement */ +#define MNTOPT_PQUOTANOENF "pqnoenforce"/* project quota limit enforcement */ +#define MNTOPT_QUOTANOENF "qnoenforce" /* same as uqnoenforce */ +#define MNTOPT_DELAYLOG "delaylog" /* Delayed logging enabled */ +#define MNTOPT_NODELAYLOG "nodelaylog" /* Delayed logging disabled */ +#define MNTOPT_DISCARD "discard" /* Discard unused blocks */ +#define MNTOPT_NODISCARD "nodiscard" /* Do not discard unused blocks */ + +/* + * Table driven mount option parser. + * + * Currently only used for remount, but it will be used for mount + * in the future, too. + */ +enum { + Opt_barrier, Opt_nobarrier, Opt_err +}; + +static const match_table_t tokens = { + {Opt_barrier, "barrier"}, + {Opt_nobarrier, "nobarrier"}, + {Opt_err, NULL} +}; + + +STATIC unsigned long +suffix_strtoul(char *s, char **endp, unsigned int base) +{ + int last, shift_left_factor = 0; + char *value = s; + + last = strlen(value) - 1; + if (value[last] == 'K' || value[last] == 'k') { + shift_left_factor = 10; + value[last] = '\0'; + } + if (value[last] == 'M' || value[last] == 'm') { + shift_left_factor = 20; + value[last] = '\0'; + } + if (value[last] == 'G' || value[last] == 'g') { + shift_left_factor = 30; + value[last] = '\0'; + } + + return simple_strtoul((const char *)s, endp, base) << shift_left_factor; +} + +/* + * This function fills in xfs_mount_t fields based on mount args. + * Note: the superblock has _not_ yet been read in. + * + * Note that this function leaks the various device name allocations on + * failure. The caller takes care of them. + */ +STATIC int +xfs_parseargs( + struct xfs_mount *mp, + char *options) +{ + struct super_block *sb = mp->m_super; + char *this_char, *value, *eov; + int dsunit = 0; + int dswidth = 0; + int iosize = 0; + __uint8_t iosizelog = 0; + + /* + * set up the mount name first so all the errors will refer to the + * correct device. + */ + mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL); + if (!mp->m_fsname) + return ENOMEM; + mp->m_fsname_len = strlen(mp->m_fsname) + 1; + + /* + * Copy binary VFS mount flags we are interested in. + */ + if (sb->s_flags & MS_RDONLY) + mp->m_flags |= XFS_MOUNT_RDONLY; + if (sb->s_flags & MS_DIRSYNC) + mp->m_flags |= XFS_MOUNT_DIRSYNC; + if (sb->s_flags & MS_SYNCHRONOUS) + mp->m_flags |= XFS_MOUNT_WSYNC; + + /* + * Set some default flags that could be cleared by the mount option + * parsing. + */ + mp->m_flags |= XFS_MOUNT_BARRIER; + mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; + mp->m_flags |= XFS_MOUNT_SMALL_INUMS; + mp->m_flags |= XFS_MOUNT_DELAYLOG; + + /* + * These can be overridden by the mount option parsing. + */ + mp->m_logbufs = -1; + mp->m_logbsize = -1; + + if (!options) + goto done; + + while ((this_char = strsep(&options, ",")) != NULL) { + if (!*this_char) + continue; + if ((value = strchr(this_char, '=')) != NULL) + *value++ = 0; + + if (!strcmp(this_char, MNTOPT_LOGBUFS)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + mp->m_logbufs = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + mp->m_logbsize = suffix_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_LOGDEV)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL); + if (!mp->m_logname) + return ENOMEM; + } else if (!strcmp(this_char, MNTOPT_MTPT)) { + xfs_warn(mp, "%s option not allowed on this system", + this_char); + return EINVAL; + } else if (!strcmp(this_char, MNTOPT_RTDEV)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL); + if (!mp->m_rtname) + return ENOMEM; + } else if (!strcmp(this_char, MNTOPT_BIOSIZE)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + iosize = simple_strtoul(value, &eov, 10); + iosizelog = ffs(iosize) - 1; + } else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + iosize = suffix_strtoul(value, &eov, 10); + iosizelog = ffs(iosize) - 1; + } else if (!strcmp(this_char, MNTOPT_GRPID) || + !strcmp(this_char, MNTOPT_BSDGROUPS)) { + mp->m_flags |= XFS_MOUNT_GRPID; + } else if (!strcmp(this_char, MNTOPT_NOGRPID) || + !strcmp(this_char, MNTOPT_SYSVGROUPS)) { + mp->m_flags &= ~XFS_MOUNT_GRPID; + } else if (!strcmp(this_char, MNTOPT_WSYNC)) { + mp->m_flags |= XFS_MOUNT_WSYNC; + } else if (!strcmp(this_char, MNTOPT_NORECOVERY)) { + mp->m_flags |= XFS_MOUNT_NORECOVERY; + } else if (!strcmp(this_char, MNTOPT_NOALIGN)) { + mp->m_flags |= XFS_MOUNT_NOALIGN; + } else if (!strcmp(this_char, MNTOPT_SWALLOC)) { + mp->m_flags |= XFS_MOUNT_SWALLOC; + } else if (!strcmp(this_char, MNTOPT_SUNIT)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + dsunit = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_SWIDTH)) { + if (!value || !*value) { + xfs_warn(mp, "%s option requires an argument", + this_char); + return EINVAL; + } + dswidth = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { + mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS; +#if !XFS_BIG_INUMS + xfs_warn(mp, "%s option not allowed on this system", + this_char); + return EINVAL; +#endif + } else if (!strcmp(this_char, MNTOPT_NOUUID)) { + mp->m_flags |= XFS_MOUNT_NOUUID; + } else if (!strcmp(this_char, MNTOPT_BARRIER)) { + mp->m_flags |= XFS_MOUNT_BARRIER; + } else if (!strcmp(this_char, MNTOPT_NOBARRIER)) { + mp->m_flags &= ~XFS_MOUNT_BARRIER; + } else if (!strcmp(this_char, MNTOPT_IKEEP)) { + mp->m_flags |= XFS_MOUNT_IKEEP; + } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) { + mp->m_flags &= ~XFS_MOUNT_IKEEP; + } else if (!strcmp(this_char, MNTOPT_LARGEIO)) { + mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) { + mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_ATTR2)) { + mp->m_flags |= XFS_MOUNT_ATTR2; + } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { + mp->m_flags &= ~XFS_MOUNT_ATTR2; + mp->m_flags |= XFS_MOUNT_NOATTR2; + } else if (!strcmp(this_char, MNTOPT_FILESTREAM)) { + mp->m_flags |= XFS_MOUNT_FILESTREAMS; + } else if (!strcmp(this_char, MNTOPT_NOQUOTA)) { + mp->m_qflags &= ~(XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE | + XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE | + XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE | + XFS_UQUOTA_ENFD | XFS_OQUOTA_ENFD); + } else if (!strcmp(this_char, MNTOPT_QUOTA) || + !strcmp(this_char, MNTOPT_UQUOTA) || + !strcmp(this_char, MNTOPT_USRQUOTA)) { + mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE | + XFS_UQUOTA_ENFD); + } else if (!strcmp(this_char, MNTOPT_QUOTANOENF) || + !strcmp(this_char, MNTOPT_UQUOTANOENF)) { + mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE); + mp->m_qflags &= ~XFS_UQUOTA_ENFD; + } else if (!strcmp(this_char, MNTOPT_PQUOTA) || + !strcmp(this_char, MNTOPT_PRJQUOTA)) { + mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE | + XFS_OQUOTA_ENFD); + } else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) { + mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE); + mp->m_qflags &= ~XFS_OQUOTA_ENFD; + } else if (!strcmp(this_char, MNTOPT_GQUOTA) || + !strcmp(this_char, MNTOPT_GRPQUOTA)) { + mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE | + XFS_OQUOTA_ENFD); + } else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) { + mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE); + mp->m_qflags &= ~XFS_OQUOTA_ENFD; + } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) { + mp->m_flags |= XFS_MOUNT_DELAYLOG; + } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { + mp->m_flags &= ~XFS_MOUNT_DELAYLOG; + } else if (!strcmp(this_char, MNTOPT_DISCARD)) { + mp->m_flags |= XFS_MOUNT_DISCARD; + } else if (!strcmp(this_char, MNTOPT_NODISCARD)) { + mp->m_flags &= ~XFS_MOUNT_DISCARD; + } else if (!strcmp(this_char, "ihashsize")) { + xfs_warn(mp, + "ihashsize no longer used, option is deprecated."); + } else if (!strcmp(this_char, "osyncisdsync")) { + xfs_warn(mp, + "osyncisdsync has no effect, option is deprecated."); + } else if (!strcmp(this_char, "osyncisosync")) { + xfs_warn(mp, + "osyncisosync has no effect, option is deprecated."); + } else if (!strcmp(this_char, "irixsgid")) { + xfs_warn(mp, + "irixsgid is now a sysctl(2) variable, option is deprecated."); + } else { + xfs_warn(mp, "unknown mount option [%s].", this_char); + return EINVAL; + } + } + + /* + * no recovery flag requires a read-only mount + */ + if ((mp->m_flags & XFS_MOUNT_NORECOVERY) && + !(mp->m_flags & XFS_MOUNT_RDONLY)) { + xfs_warn(mp, "no-recovery mounts must be read-only."); + return EINVAL; + } + + if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) { + xfs_warn(mp, + "sunit and swidth options incompatible with the noalign option"); + return EINVAL; + } + + if ((mp->m_flags & XFS_MOUNT_DISCARD) && + !(mp->m_flags & XFS_MOUNT_DELAYLOG)) { + xfs_warn(mp, + "the discard option is incompatible with the nodelaylog option"); + return EINVAL; + } + +#ifndef CONFIG_XFS_QUOTA + if (XFS_IS_QUOTA_RUNNING(mp)) { + xfs_warn(mp, "quota support not available in this kernel."); + return EINVAL; + } +#endif + + if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) && + (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE))) { + xfs_warn(mp, "cannot mount with both project and group quota"); + return EINVAL; + } + + if ((dsunit && !dswidth) || (!dsunit && dswidth)) { + xfs_warn(mp, "sunit and swidth must be specified together"); + return EINVAL; + } + + if (dsunit && (dswidth % dsunit != 0)) { + xfs_warn(mp, + "stripe width (%d) must be a multiple of the stripe unit (%d)", + dswidth, dsunit); + return EINVAL; + } + +done: + if (!(mp->m_flags & XFS_MOUNT_NOALIGN)) { + /* + * At this point the superblock has not been read + * in, therefore we do not know the block size. + * Before the mount call ends we will convert + * these to FSBs. + */ + if (dsunit) { + mp->m_dalign = dsunit; + mp->m_flags |= XFS_MOUNT_RETERR; + } + + if (dswidth) + mp->m_swidth = dswidth; + } + + if (mp->m_logbufs != -1 && + mp->m_logbufs != 0 && + (mp->m_logbufs < XLOG_MIN_ICLOGS || + mp->m_logbufs > XLOG_MAX_ICLOGS)) { + xfs_warn(mp, "invalid logbufs value: %d [not %d-%d]", + mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS); + return XFS_ERROR(EINVAL); + } + if (mp->m_logbsize != -1 && + mp->m_logbsize != 0 && + (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE || + mp->m_logbsize > XLOG_MAX_RECORD_BSIZE || + !is_power_of_2(mp->m_logbsize))) { + xfs_warn(mp, + "invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]", + mp->m_logbsize); + return XFS_ERROR(EINVAL); + } + + if (iosizelog) { + if (iosizelog > XFS_MAX_IO_LOG || + iosizelog < XFS_MIN_IO_LOG) { + xfs_warn(mp, "invalid log iosize: %d [not %d-%d]", + iosizelog, XFS_MIN_IO_LOG, + XFS_MAX_IO_LOG); + return XFS_ERROR(EINVAL); + } + + mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE; + mp->m_readio_log = iosizelog; + mp->m_writeio_log = iosizelog; + } + + return 0; +} + +struct proc_xfs_info { + int flag; + char *str; +}; + +STATIC int +xfs_showargs( + struct xfs_mount *mp, + struct seq_file *m) +{ + static struct proc_xfs_info xfs_info_set[] = { + /* the few simple ones we can get from the mount struct */ + { XFS_MOUNT_IKEEP, "," MNTOPT_IKEEP }, + { XFS_MOUNT_WSYNC, "," MNTOPT_WSYNC }, + { XFS_MOUNT_NOALIGN, "," MNTOPT_NOALIGN }, + { XFS_MOUNT_SWALLOC, "," MNTOPT_SWALLOC }, + { XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID }, + { XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY }, + { XFS_MOUNT_ATTR2, "," MNTOPT_ATTR2 }, + { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, + { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, + { XFS_MOUNT_DELAYLOG, "," MNTOPT_DELAYLOG }, + { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, + { 0, NULL } + }; + static struct proc_xfs_info xfs_info_unset[] = { + /* the few simple ones we can get from the mount struct */ + { XFS_MOUNT_COMPAT_IOSIZE, "," MNTOPT_LARGEIO }, + { XFS_MOUNT_BARRIER, "," MNTOPT_NOBARRIER }, + { XFS_MOUNT_SMALL_INUMS, "," MNTOPT_64BITINODE }, + { 0, NULL } + }; + struct proc_xfs_info *xfs_infop; + + for (xfs_infop = xfs_info_set; xfs_infop->flag; xfs_infop++) { + if (mp->m_flags & xfs_infop->flag) + seq_puts(m, xfs_infop->str); + } + for (xfs_infop = xfs_info_unset; xfs_infop->flag; xfs_infop++) { + if (!(mp->m_flags & xfs_infop->flag)) + seq_puts(m, xfs_infop->str); + } + + if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) + seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk", + (int)(1 << mp->m_writeio_log) >> 10); + + if (mp->m_logbufs > 0) + seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); + if (mp->m_logbsize > 0) + seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10); + + if (mp->m_logname) + seq_printf(m, "," MNTOPT_LOGDEV "=%s", mp->m_logname); + if (mp->m_rtname) + seq_printf(m, "," MNTOPT_RTDEV "=%s", mp->m_rtname); + + if (mp->m_dalign > 0) + seq_printf(m, "," MNTOPT_SUNIT "=%d", + (int)XFS_FSB_TO_BB(mp, mp->m_dalign)); + if (mp->m_swidth > 0) + seq_printf(m, "," MNTOPT_SWIDTH "=%d", + (int)XFS_FSB_TO_BB(mp, mp->m_swidth)); + + if (mp->m_qflags & (XFS_UQUOTA_ACCT|XFS_UQUOTA_ENFD)) + seq_puts(m, "," MNTOPT_USRQUOTA); + else if (mp->m_qflags & XFS_UQUOTA_ACCT) + seq_puts(m, "," MNTOPT_UQUOTANOENF); + + /* Either project or group quotas can be active, not both */ + + if (mp->m_qflags & XFS_PQUOTA_ACCT) { + if (mp->m_qflags & XFS_OQUOTA_ENFD) + seq_puts(m, "," MNTOPT_PRJQUOTA); + else + seq_puts(m, "," MNTOPT_PQUOTANOENF); + } else if (mp->m_qflags & XFS_GQUOTA_ACCT) { + if (mp->m_qflags & XFS_OQUOTA_ENFD) + seq_puts(m, "," MNTOPT_GRPQUOTA); + else + seq_puts(m, "," MNTOPT_GQUOTANOENF); + } + + if (!(mp->m_qflags & XFS_ALL_QUOTA_ACCT)) + seq_puts(m, "," MNTOPT_NOQUOTA); + + return 0; +} +__uint64_t +xfs_max_file_offset( + unsigned int blockshift) +{ + unsigned int pagefactor = 1; + unsigned int bitshift = BITS_PER_LONG - 1; + + /* Figure out maximum filesize, on Linux this can depend on + * the filesystem blocksize (on 32 bit platforms). + * __block_write_begin does this in an [unsigned] long... + * page->index << (PAGE_CACHE_SHIFT - bbits) + * So, for page sized blocks (4K on 32 bit platforms), + * this wraps at around 8Tb (hence MAX_LFS_FILESIZE which is + * (((u64)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1) + * but for smaller blocksizes it is less (bbits = log2 bsize). + * Note1: get_block_t takes a long (implicit cast from above) + * Note2: The Large Block Device (LBD and HAVE_SECTOR_T) patch + * can optionally convert the [unsigned] long from above into + * an [unsigned] long long. + */ + +#if BITS_PER_LONG == 32 +# if defined(CONFIG_LBDAF) + ASSERT(sizeof(sector_t) == 8); + pagefactor = PAGE_CACHE_SIZE; + bitshift = BITS_PER_LONG; +# else + pagefactor = PAGE_CACHE_SIZE >> (PAGE_CACHE_SHIFT - blockshift); +# endif +#endif + + return (((__uint64_t)pagefactor) << bitshift) - 1; +} + +STATIC int +xfs_blkdev_get( + xfs_mount_t *mp, + const char *name, + struct block_device **bdevp) +{ + int error = 0; + + *bdevp = blkdev_get_by_path(name, FMODE_READ|FMODE_WRITE|FMODE_EXCL, + mp); + if (IS_ERR(*bdevp)) { + error = PTR_ERR(*bdevp); + xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error); + } + + return -error; +} + +STATIC void +xfs_blkdev_put( + struct block_device *bdev) +{ + if (bdev) + blkdev_put(bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL); +} + +void +xfs_blkdev_issue_flush( + xfs_buftarg_t *buftarg) +{ + blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL); +} + +STATIC void +xfs_close_devices( + struct xfs_mount *mp) +{ + if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { + struct block_device *logdev = mp->m_logdev_targp->bt_bdev; + xfs_free_buftarg(mp, mp->m_logdev_targp); + xfs_blkdev_put(logdev); + } + if (mp->m_rtdev_targp) { + struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev; + xfs_free_buftarg(mp, mp->m_rtdev_targp); + xfs_blkdev_put(rtdev); + } + xfs_free_buftarg(mp, mp->m_ddev_targp); +} + +/* + * The file system configurations are: + * (1) device (partition) with data and internal log + * (2) logical volume with data and log subvolumes. + * (3) logical volume with data, log, and realtime subvolumes. + * + * We only have to handle opening the log and realtime volumes here if + * they are present. The data subvolume has already been opened by + * get_sb_bdev() and is stored in sb->s_bdev. + */ +STATIC int +xfs_open_devices( + struct xfs_mount *mp) +{ + struct block_device *ddev = mp->m_super->s_bdev; + struct block_device *logdev = NULL, *rtdev = NULL; + int error; + + /* + * Open real time and log devices - order is important. + */ + if (mp->m_logname) { + error = xfs_blkdev_get(mp, mp->m_logname, &logdev); + if (error) + goto out; + } + + if (mp->m_rtname) { + error = xfs_blkdev_get(mp, mp->m_rtname, &rtdev); + if (error) + goto out_close_logdev; + + if (rtdev == ddev || rtdev == logdev) { + xfs_warn(mp, + "Cannot mount filesystem with identical rtdev and ddev/logdev."); + error = EINVAL; + goto out_close_rtdev; + } + } + + /* + * Setup xfs_mount buffer target pointers + */ + error = ENOMEM; + mp->m_ddev_targp = xfs_alloc_buftarg(mp, ddev, 0, mp->m_fsname); + if (!mp->m_ddev_targp) + goto out_close_rtdev; + + if (rtdev) { + mp->m_rtdev_targp = xfs_alloc_buftarg(mp, rtdev, 1, + mp->m_fsname); + if (!mp->m_rtdev_targp) + goto out_free_ddev_targ; + } + + if (logdev && logdev != ddev) { + mp->m_logdev_targp = xfs_alloc_buftarg(mp, logdev, 1, + mp->m_fsname); + if (!mp->m_logdev_targp) + goto out_free_rtdev_targ; + } else { + mp->m_logdev_targp = mp->m_ddev_targp; + } + + return 0; + + out_free_rtdev_targ: + if (mp->m_rtdev_targp) + xfs_free_buftarg(mp, mp->m_rtdev_targp); + out_free_ddev_targ: + xfs_free_buftarg(mp, mp->m_ddev_targp); + out_close_rtdev: + if (rtdev) + xfs_blkdev_put(rtdev); + out_close_logdev: + if (logdev && logdev != ddev) + xfs_blkdev_put(logdev); + out: + return error; +} + +/* + * Setup xfs_mount buffer target pointers based on superblock + */ +STATIC int +xfs_setup_devices( + struct xfs_mount *mp) +{ + int error; + + error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize, + mp->m_sb.sb_sectsize); + if (error) + return error; + + if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) { + unsigned int log_sector_size = BBSIZE; + + if (xfs_sb_version_hassector(&mp->m_sb)) + log_sector_size = mp->m_sb.sb_logsectsize; + error = xfs_setsize_buftarg(mp->m_logdev_targp, + mp->m_sb.sb_blocksize, + log_sector_size); + if (error) + return error; + } + if (mp->m_rtdev_targp) { + error = xfs_setsize_buftarg(mp->m_rtdev_targp, + mp->m_sb.sb_blocksize, + mp->m_sb.sb_sectsize); + if (error) + return error; + } + + return 0; +} + +/* Catch misguided souls that try to use this interface on XFS */ +STATIC struct inode * +xfs_fs_alloc_inode( + struct super_block *sb) +{ + BUG(); + return NULL; +} + +/* + * Now that the generic code is guaranteed not to be accessing + * the linux inode, we can reclaim the inode. + */ +STATIC void +xfs_fs_destroy_inode( + struct inode *inode) +{ + struct xfs_inode *ip = XFS_I(inode); + + trace_xfs_destroy_inode(ip); + + XFS_STATS_INC(vn_reclaim); + + /* bad inode, get out here ASAP */ + if (is_bad_inode(inode)) + goto out_reclaim; + + xfs_ioend_wait(ip); + + ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0); + + /* + * We should never get here with one of the reclaim flags already set. + */ + ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIMABLE)); + ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM)); + + /* + * We always use background reclaim here because even if the + * inode is clean, it still may be under IO and hence we have + * to take the flush lock. The background reclaim path handles + * this more efficiently than we can here, so simply let background + * reclaim tear down all inodes. + */ +out_reclaim: + xfs_inode_set_reclaim_tag(ip); +} + +/* + * Slab object creation initialisation for the XFS inode. + * This covers only the idempotent fields in the XFS inode; + * all other fields need to be initialised on allocation + * from the slab. This avoids the need to repeatedly initialise + * fields in the xfs inode that left in the initialise state + * when freeing the inode. + */ +STATIC void +xfs_fs_inode_init_once( + void *inode) +{ + struct xfs_inode *ip = inode; + + memset(ip, 0, sizeof(struct xfs_inode)); + + /* vfs inode */ + inode_init_once(VFS_I(ip)); + + /* xfs inode */ + atomic_set(&ip->i_iocount, 0); + atomic_set(&ip->i_pincount, 0); + spin_lock_init(&ip->i_flags_lock); + init_waitqueue_head(&ip->i_ipin_wait); + /* + * Because we want to use a counting completion, complete + * the flush completion once to allow a single access to + * the flush completion without blocking. + */ + init_completion(&ip->i_flush); + complete(&ip->i_flush); + + mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER, + "xfsino", ip->i_ino); +} + +/* + * Dirty the XFS inode when mark_inode_dirty_sync() is called so that + * we catch unlogged VFS level updates to the inode. + * + * We need the barrier() to maintain correct ordering between unlogged + * updates and the transaction commit code that clears the i_update_core + * field. This requires all updates to be completed before marking the + * inode dirty. + */ +STATIC void +xfs_fs_dirty_inode( + struct inode *inode, + int flags) +{ + barrier(); + XFS_I(inode)->i_update_core = 1; +} + +STATIC int +xfs_log_inode( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_trans *tp; + int error; + + xfs_iunlock(ip, XFS_ILOCK_SHARED); + tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS); + error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0); + + if (error) { + xfs_trans_cancel(tp, 0); + /* we need to return with the lock hold shared */ + xfs_ilock(ip, XFS_ILOCK_SHARED); + return error; + } + + xfs_ilock(ip, XFS_ILOCK_EXCL); + + /* + * Note - it's possible that we might have pushed ourselves out of the + * way during trans_reserve which would flush the inode. But there's + * no guarantee that the inode buffer has actually gone out yet (it's + * delwri). Plus the buffer could be pinned anyway if it's part of + * an inode in another recent transaction. So we play it safe and + * fire off the transaction anyway. + */ + xfs_trans_ijoin(tp, ip); + xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); + error = xfs_trans_commit(tp, 0); + xfs_ilock_demote(ip, XFS_ILOCK_EXCL); + + return error; +} + +STATIC int +xfs_fs_write_inode( + struct inode *inode, + struct writeback_control *wbc) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + int error = EAGAIN; + + trace_xfs_write_inode(ip); + + if (XFS_FORCED_SHUTDOWN(mp)) + return XFS_ERROR(EIO); + + if (wbc->sync_mode == WB_SYNC_ALL) { + /* + * Make sure the inode has made it it into the log. Instead + * of forcing it all the way to stable storage using a + * synchronous transaction we let the log force inside the + * ->sync_fs call do that for thus, which reduces the number + * of synchronous log foces dramatically. + */ + xfs_ioend_wait(ip); + xfs_ilock(ip, XFS_ILOCK_SHARED); + if (ip->i_update_core) { + error = xfs_log_inode(ip); + if (error) + goto out_unlock; + } + } else { + /* + * We make this non-blocking if the inode is contended, return + * EAGAIN to indicate to the caller that they did not succeed. + * This prevents the flush path from blocking on inodes inside + * another operation right now, they get caught later by + * xfs_sync. + */ + if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED)) + goto out; + + if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip)) + goto out_unlock; + + /* + * Now we have the flush lock and the inode is not pinned, we + * can check if the inode is really clean as we know that + * there are no pending transaction completions, it is not + * waiting on the delayed write queue and there is no IO in + * progress. + */ + if (xfs_inode_clean(ip)) { + xfs_ifunlock(ip); + error = 0; + goto out_unlock; + } + error = xfs_iflush(ip, SYNC_TRYLOCK); + } + + out_unlock: + xfs_iunlock(ip, XFS_ILOCK_SHARED); + out: + /* + * if we failed to write out the inode then mark + * it dirty again so we'll try again later. + */ + if (error) + xfs_mark_inode_dirty_sync(ip); + return -error; +} + +STATIC void +xfs_fs_evict_inode( + struct inode *inode) +{ + xfs_inode_t *ip = XFS_I(inode); + + trace_xfs_evict_inode(ip); + + truncate_inode_pages(&inode->i_data, 0); + end_writeback(inode); + XFS_STATS_INC(vn_rele); + XFS_STATS_INC(vn_remove); + XFS_STATS_DEC(vn_active); + + /* + * The iolock is used by the file system to coordinate reads, + * writes, and block truncates. Up to this point the lock + * protected concurrent accesses by users of the inode. But + * from here forward we're doing some final processing of the + * inode because we're done with it, and although we reuse the + * iolock for protection it is really a distinct lock class + * (in the lockdep sense) from before. To keep lockdep happy + * (and basically indicate what we are doing), we explicitly + * re-init the iolock here. + */ + ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock)); + mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino); + lockdep_set_class_and_name(&ip->i_iolock.mr_lock, + &xfs_iolock_reclaimable, "xfs_iolock_reclaimable"); + + xfs_inactive(ip); +} + +STATIC void +xfs_free_fsname( + struct xfs_mount *mp) +{ + kfree(mp->m_fsname); + kfree(mp->m_rtname); + kfree(mp->m_logname); +} + +STATIC void +xfs_fs_put_super( + struct super_block *sb) +{ + struct xfs_mount *mp = XFS_M(sb); + + xfs_syncd_stop(mp); + + /* + * Blow away any referenced inode in the filestreams cache. + * This can and will cause log traffic as inodes go inactive + * here. + */ + xfs_filestream_unmount(mp); + + XFS_bflush(mp->m_ddev_targp); + + xfs_unmountfs(mp); + xfs_freesb(mp); + xfs_icsb_destroy_counters(mp); + xfs_close_devices(mp); + xfs_free_fsname(mp); + kfree(mp); +} + +STATIC int +xfs_fs_sync_fs( + struct super_block *sb, + int wait) +{ + struct xfs_mount *mp = XFS_M(sb); + int error; + + /* + * Not much we can do for the first async pass. Writing out the + * superblock would be counter-productive as we are going to redirty + * when writing out other data and metadata (and writing out a single + * block is quite fast anyway). + * + * Try to asynchronously kick off quota syncing at least. + */ + if (!wait) { + xfs_qm_sync(mp, SYNC_TRYLOCK); + return 0; + } + + error = xfs_quiesce_data(mp); + if (error) + return -error; + + if (laptop_mode) { + /* + * The disk must be active because we're syncing. + * We schedule xfssyncd now (now that the disk is + * active) instead of later (when it might not be). + */ + flush_delayed_work_sync(&mp->m_sync_work); + } + + return 0; +} + +STATIC int +xfs_fs_statfs( + struct dentry *dentry, + struct kstatfs *statp) +{ + struct xfs_mount *mp = XFS_M(dentry->d_sb); + xfs_sb_t *sbp = &mp->m_sb; + struct xfs_inode *ip = XFS_I(dentry->d_inode); + __uint64_t fakeinos, id; + xfs_extlen_t lsize; + __int64_t ffree; + + statp->f_type = XFS_SB_MAGIC; + statp->f_namelen = MAXNAMELEN - 1; + + id = huge_encode_dev(mp->m_ddev_targp->bt_dev); + statp->f_fsid.val[0] = (u32)id; + statp->f_fsid.val[1] = (u32)(id >> 32); + + xfs_icsb_sync_counters(mp, XFS_ICSB_LAZY_COUNT); + + spin_lock(&mp->m_sb_lock); + statp->f_bsize = sbp->sb_blocksize; + lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0; + statp->f_blocks = sbp->sb_dblocks - lsize; + statp->f_bfree = statp->f_bavail = + sbp->sb_fdblocks - XFS_ALLOC_SET_ASIDE(mp); + fakeinos = statp->f_bfree << sbp->sb_inopblog; + statp->f_files = + MIN(sbp->sb_icount + fakeinos, (__uint64_t)XFS_MAXINUMBER); + if (mp->m_maxicount) + statp->f_files = min_t(typeof(statp->f_files), + statp->f_files, + mp->m_maxicount); + + /* make sure statp->f_ffree does not underflow */ + ffree = statp->f_files - (sbp->sb_icount - sbp->sb_ifree); + statp->f_ffree = max_t(__int64_t, ffree, 0); + + spin_unlock(&mp->m_sb_lock); + + if ((ip->i_d.di_flags & XFS_DIFLAG_PROJINHERIT) || + ((mp->m_qflags & (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD))) == + (XFS_PQUOTA_ACCT|XFS_OQUOTA_ENFD)) + xfs_qm_statvfs(ip, statp); + return 0; +} + +STATIC void +xfs_save_resvblks(struct xfs_mount *mp) +{ + __uint64_t resblks = 0; + + mp->m_resblks_save = mp->m_resblks; + xfs_reserve_blocks(mp, &resblks, NULL); +} + +STATIC void +xfs_restore_resvblks(struct xfs_mount *mp) +{ + __uint64_t resblks; + + if (mp->m_resblks_save) { + resblks = mp->m_resblks_save; + mp->m_resblks_save = 0; + } else + resblks = xfs_default_resblks(mp); + + xfs_reserve_blocks(mp, &resblks, NULL); +} + +STATIC int +xfs_fs_remount( + struct super_block *sb, + int *flags, + char *options) +{ + struct xfs_mount *mp = XFS_M(sb); + substring_t args[MAX_OPT_ARGS]; + char *p; + int error; + + while ((p = strsep(&options, ",")) != NULL) { + int token; + + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_barrier: + mp->m_flags |= XFS_MOUNT_BARRIER; + break; + case Opt_nobarrier: + mp->m_flags &= ~XFS_MOUNT_BARRIER; + break; + default: + /* + * Logically we would return an error here to prevent + * users from believing they might have changed + * mount options using remount which can't be changed. + * + * But unfortunately mount(8) adds all options from + * mtab and fstab to the mount arguments in some cases + * so we can't blindly reject options, but have to + * check for each specified option if it actually + * differs from the currently set option and only + * reject it if that's the case. + * + * Until that is implemented we return success for + * every remount request, and silently ignore all + * options that we can't actually change. + */ +#if 0 + xfs_info(mp, + "mount option \"%s\" not supported for remount\n", p); + return -EINVAL; +#else + break; +#endif + } + } + + /* ro -> rw */ + if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) { + mp->m_flags &= ~XFS_MOUNT_RDONLY; + + /* + * If this is the first remount to writeable state we + * might have some superblock changes to update. + */ + if (mp->m_update_flags) { + error = xfs_mount_log_sb(mp, mp->m_update_flags); + if (error) { + xfs_warn(mp, "failed to write sb changes"); + return error; + } + mp->m_update_flags = 0; + } + + /* + * Fill out the reserve pool if it is empty. Use the stashed + * value if it is non-zero, otherwise go with the default. + */ + xfs_restore_resvblks(mp); + } + + /* rw -> ro */ + if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) { + /* + * After we have synced the data but before we sync the + * metadata, we need to free up the reserve block pool so that + * the used block count in the superblock on disk is correct at + * the end of the remount. Stash the current reserve pool size + * so that if we get remounted rw, we can return it to the same + * size. + */ + + xfs_quiesce_data(mp); + xfs_save_resvblks(mp); + xfs_quiesce_attr(mp); + mp->m_flags |= XFS_MOUNT_RDONLY; + } + + return 0; +} + +/* + * Second stage of a freeze. The data is already frozen so we only + * need to take care of the metadata. Once that's done write a dummy + * record to dirty the log in case of a crash while frozen. + */ +STATIC int +xfs_fs_freeze( + struct super_block *sb) +{ + struct xfs_mount *mp = XFS_M(sb); + + xfs_save_resvblks(mp); + xfs_quiesce_attr(mp); + return -xfs_fs_log_dummy(mp); +} + +STATIC int +xfs_fs_unfreeze( + struct super_block *sb) +{ + struct xfs_mount *mp = XFS_M(sb); + + xfs_restore_resvblks(mp); + return 0; +} + +STATIC int +xfs_fs_show_options( + struct seq_file *m, + struct vfsmount *mnt) +{ + return -xfs_showargs(XFS_M(mnt->mnt_sb), m); +} + +/* + * This function fills in xfs_mount_t fields based on mount args. + * Note: the superblock _has_ now been read in. + */ +STATIC int +xfs_finish_flags( + struct xfs_mount *mp) +{ + int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); + + /* Fail a mount where the logbuf is smaller than the log stripe */ + if (xfs_sb_version_haslogv2(&mp->m_sb)) { + if (mp->m_logbsize <= 0 && + mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) { + mp->m_logbsize = mp->m_sb.sb_logsunit; + } else if (mp->m_logbsize > 0 && + mp->m_logbsize < mp->m_sb.sb_logsunit) { + xfs_warn(mp, + "logbuf size must be greater than or equal to log stripe size"); + return XFS_ERROR(EINVAL); + } + } else { + /* Fail a mount if the logbuf is larger than 32K */ + if (mp->m_logbsize > XLOG_BIG_RECORD_BSIZE) { + xfs_warn(mp, + "logbuf size for version 1 logs must be 16K or 32K"); + return XFS_ERROR(EINVAL); + } + } + + /* + * mkfs'ed attr2 will turn on attr2 mount unless explicitly + * told by noattr2 to turn it off + */ + if (xfs_sb_version_hasattr2(&mp->m_sb) && + !(mp->m_flags & XFS_MOUNT_NOATTR2)) + mp->m_flags |= XFS_MOUNT_ATTR2; + + /* + * prohibit r/w mounts of read-only filesystems + */ + if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) { + xfs_warn(mp, + "cannot mount a read-only filesystem as read-write"); + return XFS_ERROR(EROFS); + } + + return 0; +} + +STATIC int +xfs_fs_fill_super( + struct super_block *sb, + void *data, + int silent) +{ + struct inode *root; + struct xfs_mount *mp = NULL; + int flags = 0, error = ENOMEM; + + mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL); + if (!mp) + goto out; + + spin_lock_init(&mp->m_sb_lock); + mutex_init(&mp->m_growlock); + atomic_set(&mp->m_active_trans, 0); + + mp->m_super = sb; + sb->s_fs_info = mp; + + error = xfs_parseargs(mp, (char *)data); + if (error) + goto out_free_fsname; + + sb_min_blocksize(sb, BBSIZE); + sb->s_xattr = xfs_xattr_handlers; + sb->s_export_op = &xfs_export_operations; +#ifdef CONFIG_XFS_QUOTA + sb->s_qcop = &xfs_quotactl_operations; +#endif + sb->s_op = &xfs_super_operations; + + if (silent) + flags |= XFS_MFSI_QUIET; + + error = xfs_open_devices(mp); + if (error) + goto out_free_fsname; + + error = xfs_icsb_init_counters(mp); + if (error) + goto out_close_devices; + + error = xfs_readsb(mp, flags); + if (error) + goto out_destroy_counters; + + error = xfs_finish_flags(mp); + if (error) + goto out_free_sb; + + error = xfs_setup_devices(mp); + if (error) + goto out_free_sb; + + error = xfs_filestream_mount(mp); + if (error) + goto out_free_sb; + + /* + * we must configure the block size in the superblock before we run the + * full mount process as the mount process can lookup and cache inodes. + * For the same reason we must also initialise the syncd and register + * the inode cache shrinker so that inodes can be reclaimed during + * operations like a quotacheck that iterate all inodes in the + * filesystem. + */ + sb->s_magic = XFS_SB_MAGIC; + sb->s_blocksize = mp->m_sb.sb_blocksize; + sb->s_blocksize_bits = ffs(sb->s_blocksize) - 1; + sb->s_maxbytes = xfs_max_file_offset(sb->s_blocksize_bits); + sb->s_time_gran = 1; + set_posix_acl_flag(sb); + + error = xfs_mountfs(mp); + if (error) + goto out_filestream_unmount; + + error = xfs_syncd_init(mp); + if (error) + goto out_unmount; + + root = igrab(VFS_I(mp->m_rootip)); + if (!root) { + error = ENOENT; + goto out_syncd_stop; + } + if (is_bad_inode(root)) { + error = EINVAL; + goto out_syncd_stop; + } + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + error = ENOMEM; + goto out_iput; + } + + return 0; + + out_filestream_unmount: + xfs_filestream_unmount(mp); + out_free_sb: + xfs_freesb(mp); + out_destroy_counters: + xfs_icsb_destroy_counters(mp); + out_close_devices: + xfs_close_devices(mp); + out_free_fsname: + xfs_free_fsname(mp); + kfree(mp); + out: + return -error; + + out_iput: + iput(root); + out_syncd_stop: + xfs_syncd_stop(mp); + out_unmount: + /* + * Blow away any referenced inode in the filestreams cache. + * This can and will cause log traffic as inodes go inactive + * here. + */ + xfs_filestream_unmount(mp); + + XFS_bflush(mp->m_ddev_targp); + + xfs_unmountfs(mp); + goto out_free_sb; +} + +STATIC struct dentry * +xfs_fs_mount( + struct file_system_type *fs_type, + int flags, + const char *dev_name, + void *data) +{ + return mount_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super); +} + +static int +xfs_fs_nr_cached_objects( + struct super_block *sb) +{ + return xfs_reclaim_inodes_count(XFS_M(sb)); +} + +static void +xfs_fs_free_cached_objects( + struct super_block *sb, + int nr_to_scan) +{ + xfs_reclaim_inodes_nr(XFS_M(sb), nr_to_scan); +} + +static const struct super_operations xfs_super_operations = { + .alloc_inode = xfs_fs_alloc_inode, + .destroy_inode = xfs_fs_destroy_inode, + .dirty_inode = xfs_fs_dirty_inode, + .write_inode = xfs_fs_write_inode, + .evict_inode = xfs_fs_evict_inode, + .put_super = xfs_fs_put_super, + .sync_fs = xfs_fs_sync_fs, + .freeze_fs = xfs_fs_freeze, + .unfreeze_fs = xfs_fs_unfreeze, + .statfs = xfs_fs_statfs, + .remount_fs = xfs_fs_remount, + .show_options = xfs_fs_show_options, + .nr_cached_objects = xfs_fs_nr_cached_objects, + .free_cached_objects = xfs_fs_free_cached_objects, +}; + +static struct file_system_type xfs_fs_type = { + .owner = THIS_MODULE, + .name = "xfs", + .mount = xfs_fs_mount, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +STATIC int __init +xfs_init_zones(void) +{ + + xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend"); + if (!xfs_ioend_zone) + goto out; + + xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE, + xfs_ioend_zone); + if (!xfs_ioend_pool) + goto out_destroy_ioend_zone; + + xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t), + "xfs_log_ticket"); + if (!xfs_log_ticket_zone) + goto out_destroy_ioend_pool; + + xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t), + "xfs_bmap_free_item"); + if (!xfs_bmap_free_item_zone) + goto out_destroy_log_ticket_zone; + + xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t), + "xfs_btree_cur"); + if (!xfs_btree_cur_zone) + goto out_destroy_bmap_free_item_zone; + + xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t), + "xfs_da_state"); + if (!xfs_da_state_zone) + goto out_destroy_btree_cur_zone; + + xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf"); + if (!xfs_dabuf_zone) + goto out_destroy_da_state_zone; + + xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork"); + if (!xfs_ifork_zone) + goto out_destroy_dabuf_zone; + + xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans"); + if (!xfs_trans_zone) + goto out_destroy_ifork_zone; + + xfs_log_item_desc_zone = + kmem_zone_init(sizeof(struct xfs_log_item_desc), + "xfs_log_item_desc"); + if (!xfs_log_item_desc_zone) + goto out_destroy_trans_zone; + + /* + * The size of the zone allocated buf log item is the maximum + * size possible under XFS. This wastes a little bit of memory, + * but it is much faster. + */ + xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) + + (((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) / + NBWORD) * sizeof(int))), "xfs_buf_item"); + if (!xfs_buf_item_zone) + goto out_destroy_log_item_desc_zone; + + xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) + + ((XFS_EFD_MAX_FAST_EXTENTS - 1) * + sizeof(xfs_extent_t))), "xfs_efd_item"); + if (!xfs_efd_zone) + goto out_destroy_buf_item_zone; + + xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) + + ((XFS_EFI_MAX_FAST_EXTENTS - 1) * + sizeof(xfs_extent_t))), "xfs_efi_item"); + if (!xfs_efi_zone) + goto out_destroy_efd_zone; + + xfs_inode_zone = + kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode", + KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD, + xfs_fs_inode_init_once); + if (!xfs_inode_zone) + goto out_destroy_efi_zone; + + xfs_ili_zone = + kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili", + KM_ZONE_SPREAD, NULL); + if (!xfs_ili_zone) + goto out_destroy_inode_zone; + + return 0; + + out_destroy_inode_zone: + kmem_zone_destroy(xfs_inode_zone); + out_destroy_efi_zone: + kmem_zone_destroy(xfs_efi_zone); + out_destroy_efd_zone: + kmem_zone_destroy(xfs_efd_zone); + out_destroy_buf_item_zone: + kmem_zone_destroy(xfs_buf_item_zone); + out_destroy_log_item_desc_zone: + kmem_zone_destroy(xfs_log_item_desc_zone); + out_destroy_trans_zone: + kmem_zone_destroy(xfs_trans_zone); + out_destroy_ifork_zone: + kmem_zone_destroy(xfs_ifork_zone); + out_destroy_dabuf_zone: + kmem_zone_destroy(xfs_dabuf_zone); + out_destroy_da_state_zone: + kmem_zone_destroy(xfs_da_state_zone); + out_destroy_btree_cur_zone: + kmem_zone_destroy(xfs_btree_cur_zone); + out_destroy_bmap_free_item_zone: + kmem_zone_destroy(xfs_bmap_free_item_zone); + out_destroy_log_ticket_zone: + kmem_zone_destroy(xfs_log_ticket_zone); + out_destroy_ioend_pool: + mempool_destroy(xfs_ioend_pool); + out_destroy_ioend_zone: + kmem_zone_destroy(xfs_ioend_zone); + out: + return -ENOMEM; +} + +STATIC void +xfs_destroy_zones(void) +{ + kmem_zone_destroy(xfs_ili_zone); + kmem_zone_destroy(xfs_inode_zone); + kmem_zone_destroy(xfs_efi_zone); + kmem_zone_destroy(xfs_efd_zone); + kmem_zone_destroy(xfs_buf_item_zone); + kmem_zone_destroy(xfs_log_item_desc_zone); + kmem_zone_destroy(xfs_trans_zone); + kmem_zone_destroy(xfs_ifork_zone); + kmem_zone_destroy(xfs_dabuf_zone); + kmem_zone_destroy(xfs_da_state_zone); + kmem_zone_destroy(xfs_btree_cur_zone); + kmem_zone_destroy(xfs_bmap_free_item_zone); + kmem_zone_destroy(xfs_log_ticket_zone); + mempool_destroy(xfs_ioend_pool); + kmem_zone_destroy(xfs_ioend_zone); + +} + +STATIC int __init +xfs_init_workqueues(void) +{ + /* + * max_active is set to 8 to give enough concurency to allow + * multiple work operations on each CPU to run. This allows multiple + * filesystems to be running sync work concurrently, and scales with + * the number of CPUs in the system. + */ + xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8); + if (!xfs_syncd_wq) + goto out; + + xfs_ail_wq = alloc_workqueue("xfsail", WQ_CPU_INTENSIVE, 8); + if (!xfs_ail_wq) + goto out_destroy_syncd; + + return 0; + +out_destroy_syncd: + destroy_workqueue(xfs_syncd_wq); +out: + return -ENOMEM; +} + +STATIC void +xfs_destroy_workqueues(void) +{ + destroy_workqueue(xfs_ail_wq); + destroy_workqueue(xfs_syncd_wq); +} + +STATIC int __init +init_xfs_fs(void) +{ + int error; + + printk(KERN_INFO XFS_VERSION_STRING " with " + XFS_BUILD_OPTIONS " enabled\n"); + + xfs_ioend_init(); + xfs_dir_startup(); + + error = xfs_init_zones(); + if (error) + goto out; + + error = xfs_init_workqueues(); + if (error) + goto out_destroy_zones; + + error = xfs_mru_cache_init(); + if (error) + goto out_destroy_wq; + + error = xfs_filestream_init(); + if (error) + goto out_mru_cache_uninit; + + error = xfs_buf_init(); + if (error) + goto out_filestream_uninit; + + error = xfs_init_procfs(); + if (error) + goto out_buf_terminate; + + error = xfs_sysctl_register(); + if (error) + goto out_cleanup_procfs; + + vfs_initquota(); + + error = register_filesystem(&xfs_fs_type); + if (error) + goto out_sysctl_unregister; + return 0; + + out_sysctl_unregister: + xfs_sysctl_unregister(); + out_cleanup_procfs: + xfs_cleanup_procfs(); + out_buf_terminate: + xfs_buf_terminate(); + out_filestream_uninit: + xfs_filestream_uninit(); + out_mru_cache_uninit: + xfs_mru_cache_uninit(); + out_destroy_wq: + xfs_destroy_workqueues(); + out_destroy_zones: + xfs_destroy_zones(); + out: + return error; +} + +STATIC void __exit +exit_xfs_fs(void) +{ + vfs_exitquota(); + unregister_filesystem(&xfs_fs_type); + xfs_sysctl_unregister(); + xfs_cleanup_procfs(); + xfs_buf_terminate(); + xfs_filestream_uninit(); + xfs_mru_cache_uninit(); + xfs_destroy_workqueues(); + xfs_destroy_zones(); +} + +module_init(init_xfs_fs); +module_exit(exit_xfs_fs); + +MODULE_AUTHOR("Silicon Graphics, Inc."); +MODULE_DESCRIPTION(XFS_VERSION_STRING " with " XFS_BUILD_OPTIONS " enabled"); +MODULE_LICENSE("GPL"); diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h new file mode 100644 index 000000000000..50a3266c999e --- /dev/null +++ b/fs/xfs/xfs_super.h @@ -0,0 +1,87 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SUPER_H__ +#define __XFS_SUPER_H__ + +#include + +#ifdef CONFIG_XFS_QUOTA +extern void xfs_qm_init(void); +extern void xfs_qm_exit(void); +# define vfs_initquota() xfs_qm_init() +# define vfs_exitquota() xfs_qm_exit() +#else +# define vfs_initquota() do { } while (0) +# define vfs_exitquota() do { } while (0) +#endif + +#ifdef CONFIG_XFS_POSIX_ACL +# define XFS_ACL_STRING "ACLs, " +# define set_posix_acl_flag(sb) ((sb)->s_flags |= MS_POSIXACL) +#else +# define XFS_ACL_STRING +# define set_posix_acl_flag(sb) do { } while (0) +#endif + +#define XFS_SECURITY_STRING "security attributes, " + +#ifdef CONFIG_XFS_RT +# define XFS_REALTIME_STRING "realtime, " +#else +# define XFS_REALTIME_STRING +#endif + +#if XFS_BIG_BLKNOS +# if XFS_BIG_INUMS +# define XFS_BIGFS_STRING "large block/inode numbers, " +# else +# define XFS_BIGFS_STRING "large block numbers, " +# endif +#else +# define XFS_BIGFS_STRING +#endif + +#ifdef DEBUG +# define XFS_DBG_STRING "debug" +#else +# define XFS_DBG_STRING "no debug" +#endif + +#define XFS_VERSION_STRING "SGI XFS" +#define XFS_BUILD_OPTIONS XFS_ACL_STRING \ + XFS_SECURITY_STRING \ + XFS_REALTIME_STRING \ + XFS_BIGFS_STRING \ + XFS_DBG_STRING /* DBG must be last */ + +struct xfs_inode; +struct xfs_mount; +struct xfs_buftarg; +struct block_device; + +extern __uint64_t xfs_max_file_offset(unsigned int); + +extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); + +extern const struct export_operations xfs_export_operations; +extern const struct xattr_handler *xfs_xattr_handlers[]; +extern const struct quotactl_ops xfs_quotactl_operations; + +#define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info)) + +#endif /* __XFS_SUPER_H__ */ diff --git a/fs/xfs/xfs_sync.c b/fs/xfs/xfs_sync.c new file mode 100644 index 000000000000..4604f90f86a3 --- /dev/null +++ b/fs/xfs/xfs_sync.c @@ -0,0 +1,1065 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_trans_priv.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_dinode.h" +#include "xfs_error.h" +#include "xfs_filestream.h" +#include "xfs_vnodeops.h" +#include "xfs_inode_item.h" +#include "xfs_quota.h" +#include "xfs_trace.h" +#include "xfs_fsops.h" + +#include +#include + +struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ + +/* + * The inode lookup is done in batches to keep the amount of lock traffic and + * radix tree lookups to a minimum. The batch size is a trade off between + * lookup reduction and stack usage. This is in the reclaim path, so we can't + * be too greedy. + */ +#define XFS_LOOKUP_BATCH 32 + +STATIC int +xfs_inode_ag_walk_grab( + struct xfs_inode *ip) +{ + struct inode *inode = VFS_I(ip); + + ASSERT(rcu_read_lock_held()); + + /* + * check for stale RCU freed inode + * + * If the inode has been reallocated, it doesn't matter if it's not in + * the AG we are walking - we are walking for writeback, so if it + * passes all the "valid inode" checks and is dirty, then we'll write + * it back anyway. If it has been reallocated and still being + * initialised, the XFS_INEW check below will catch it. + */ + spin_lock(&ip->i_flags_lock); + if (!ip->i_ino) + goto out_unlock_noent; + + /* avoid new or reclaimable inodes. Leave for reclaim code to flush */ + if (__xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM)) + goto out_unlock_noent; + spin_unlock(&ip->i_flags_lock); + + /* nothing to sync during shutdown */ + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) + return EFSCORRUPTED; + + /* If we can't grab the inode, it must on it's way to reclaim. */ + if (!igrab(inode)) + return ENOENT; + + if (is_bad_inode(inode)) { + IRELE(ip); + return ENOENT; + } + + /* inode is valid */ + return 0; + +out_unlock_noent: + spin_unlock(&ip->i_flags_lock); + return ENOENT; +} + +STATIC int +xfs_inode_ag_walk( + struct xfs_mount *mp, + struct xfs_perag *pag, + int (*execute)(struct xfs_inode *ip, + struct xfs_perag *pag, int flags), + int flags) +{ + uint32_t first_index; + int last_error = 0; + int skipped; + int done; + int nr_found; + +restart: + done = 0; + skipped = 0; + first_index = 0; + nr_found = 0; + do { + struct xfs_inode *batch[XFS_LOOKUP_BATCH]; + int error = 0; + int i; + + rcu_read_lock(); + nr_found = radix_tree_gang_lookup(&pag->pag_ici_root, + (void **)batch, first_index, + XFS_LOOKUP_BATCH); + if (!nr_found) { + rcu_read_unlock(); + break; + } + + /* + * Grab the inodes before we drop the lock. if we found + * nothing, nr == 0 and the loop will be skipped. + */ + for (i = 0; i < nr_found; i++) { + struct xfs_inode *ip = batch[i]; + + if (done || xfs_inode_ag_walk_grab(ip)) + batch[i] = NULL; + + /* + * Update the index for the next lookup. Catch + * overflows into the next AG range which can occur if + * we have inodes in the last block of the AG and we + * are currently pointing to the last inode. + * + * Because we may see inodes that are from the wrong AG + * due to RCU freeing and reallocation, only update the + * index if it lies in this AG. It was a race that lead + * us to see this inode, so another lookup from the + * same index will not find it again. + */ + if (XFS_INO_TO_AGNO(mp, ip->i_ino) != pag->pag_agno) + continue; + first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); + if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) + done = 1; + } + + /* unlock now we've grabbed the inodes. */ + rcu_read_unlock(); + + for (i = 0; i < nr_found; i++) { + if (!batch[i]) + continue; + error = execute(batch[i], pag, flags); + IRELE(batch[i]); + if (error == EAGAIN) { + skipped++; + continue; + } + if (error && last_error != EFSCORRUPTED) + last_error = error; + } + + /* bail out if the filesystem is corrupted. */ + if (error == EFSCORRUPTED) + break; + + cond_resched(); + + } while (nr_found && !done); + + if (skipped) { + delay(1); + goto restart; + } + return last_error; +} + +int +xfs_inode_ag_iterator( + struct xfs_mount *mp, + int (*execute)(struct xfs_inode *ip, + struct xfs_perag *pag, int flags), + int flags) +{ + struct xfs_perag *pag; + int error = 0; + int last_error = 0; + xfs_agnumber_t ag; + + ag = 0; + while ((pag = xfs_perag_get(mp, ag))) { + ag = pag->pag_agno + 1; + error = xfs_inode_ag_walk(mp, pag, execute, flags); + xfs_perag_put(pag); + if (error) { + last_error = error; + if (error == EFSCORRUPTED) + break; + } + } + return XFS_ERROR(last_error); +} + +STATIC int +xfs_sync_inode_data( + struct xfs_inode *ip, + struct xfs_perag *pag, + int flags) +{ + struct inode *inode = VFS_I(ip); + struct address_space *mapping = inode->i_mapping; + int error = 0; + + if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) + goto out_wait; + + if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) { + if (flags & SYNC_TRYLOCK) + goto out_wait; + xfs_ilock(ip, XFS_IOLOCK_SHARED); + } + + error = xfs_flush_pages(ip, 0, -1, (flags & SYNC_WAIT) ? + 0 : XBF_ASYNC, FI_NONE); + xfs_iunlock(ip, XFS_IOLOCK_SHARED); + + out_wait: + if (flags & SYNC_WAIT) + xfs_ioend_wait(ip); + return error; +} + +STATIC int +xfs_sync_inode_attr( + struct xfs_inode *ip, + struct xfs_perag *pag, + int flags) +{ + int error = 0; + + xfs_ilock(ip, XFS_ILOCK_SHARED); + if (xfs_inode_clean(ip)) + goto out_unlock; + if (!xfs_iflock_nowait(ip)) { + if (!(flags & SYNC_WAIT)) + goto out_unlock; + xfs_iflock(ip); + } + + if (xfs_inode_clean(ip)) { + xfs_ifunlock(ip); + goto out_unlock; + } + + error = xfs_iflush(ip, flags); + + /* + * We don't want to try again on non-blocking flushes that can't run + * again immediately. If an inode really must be written, then that's + * what the SYNC_WAIT flag is for. + */ + if (error == EAGAIN) { + ASSERT(!(flags & SYNC_WAIT)); + error = 0; + } + + out_unlock: + xfs_iunlock(ip, XFS_ILOCK_SHARED); + return error; +} + +/* + * Write out pagecache data for the whole filesystem. + */ +STATIC int +xfs_sync_data( + struct xfs_mount *mp, + int flags) +{ + int error; + + ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0); + + error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags); + if (error) + return XFS_ERROR(error); + + xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0); + return 0; +} + +/* + * Write out inode metadata (attributes) for the whole filesystem. + */ +STATIC int +xfs_sync_attr( + struct xfs_mount *mp, + int flags) +{ + ASSERT((flags & ~SYNC_WAIT) == 0); + + return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags); +} + +STATIC int +xfs_sync_fsdata( + struct xfs_mount *mp) +{ + struct xfs_buf *bp; + + /* + * If the buffer is pinned then push on the log so we won't get stuck + * waiting in the write for someone, maybe ourselves, to flush the log. + * + * Even though we just pushed the log above, we did not have the + * superblock buffer locked at that point so it can become pinned in + * between there and here. + */ + bp = xfs_getsb(mp, 0); + if (xfs_buf_ispinned(bp)) + xfs_log_force(mp, 0); + + return xfs_bwrite(mp, bp); +} + +/* + * When remounting a filesystem read-only or freezing the filesystem, we have + * two phases to execute. This first phase is syncing the data before we + * quiesce the filesystem, and the second is flushing all the inodes out after + * we've waited for all the transactions created by the first phase to + * complete. The second phase ensures that the inodes are written to their + * location on disk rather than just existing in transactions in the log. This + * means after a quiesce there is no log replay required to write the inodes to + * disk (this is the main difference between a sync and a quiesce). + */ +/* + * First stage of freeze - no writers will make progress now we are here, + * so we flush delwri and delalloc buffers here, then wait for all I/O to + * complete. Data is frozen at that point. Metadata is not frozen, + * transactions can still occur here so don't bother flushing the buftarg + * because it'll just get dirty again. + */ +int +xfs_quiesce_data( + struct xfs_mount *mp) +{ + int error, error2 = 0; + + xfs_qm_sync(mp, SYNC_TRYLOCK); + xfs_qm_sync(mp, SYNC_WAIT); + + /* force out the newly dirtied log buffers */ + xfs_log_force(mp, XFS_LOG_SYNC); + + /* write superblock and hoover up shutdown errors */ + error = xfs_sync_fsdata(mp); + + /* make sure all delwri buffers are written out */ + xfs_flush_buftarg(mp->m_ddev_targp, 1); + + /* mark the log as covered if needed */ + if (xfs_log_need_covered(mp)) + error2 = xfs_fs_log_dummy(mp); + + /* flush data-only devices */ + if (mp->m_rtdev_targp) + XFS_bflush(mp->m_rtdev_targp); + + return error ? error : error2; +} + +STATIC void +xfs_quiesce_fs( + struct xfs_mount *mp) +{ + int count = 0, pincount; + + xfs_reclaim_inodes(mp, 0); + xfs_flush_buftarg(mp->m_ddev_targp, 0); + + /* + * This loop must run at least twice. The first instance of the loop + * will flush most meta data but that will generate more meta data + * (typically directory updates). Which then must be flushed and + * logged before we can write the unmount record. We also so sync + * reclaim of inodes to catch any that the above delwri flush skipped. + */ + do { + xfs_reclaim_inodes(mp, SYNC_WAIT); + xfs_sync_attr(mp, SYNC_WAIT); + pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1); + if (!pincount) { + delay(50); + count++; + } + } while (count < 2); +} + +/* + * Second stage of a quiesce. The data is already synced, now we have to take + * care of the metadata. New transactions are already blocked, so we need to + * wait for any remaining transactions to drain out before proceeding. + */ +void +xfs_quiesce_attr( + struct xfs_mount *mp) +{ + int error = 0; + + /* wait for all modifications to complete */ + while (atomic_read(&mp->m_active_trans) > 0) + delay(100); + + /* flush inodes and push all remaining buffers out to disk */ + xfs_quiesce_fs(mp); + + /* + * Just warn here till VFS can correctly support + * read-only remount without racing. + */ + WARN_ON(atomic_read(&mp->m_active_trans) != 0); + + /* Push the superblock and write an unmount record */ + error = xfs_log_sbcount(mp); + if (error) + xfs_warn(mp, "xfs_attr_quiesce: failed to log sb changes. " + "Frozen image may not be consistent."); + xfs_log_unmount_write(mp); + xfs_unmountfs_writesb(mp); +} + +static void +xfs_syncd_queue_sync( + struct xfs_mount *mp) +{ + queue_delayed_work(xfs_syncd_wq, &mp->m_sync_work, + msecs_to_jiffies(xfs_syncd_centisecs * 10)); +} + +/* + * Every sync period we need to unpin all items, reclaim inodes and sync + * disk quotas. We might need to cover the log to indicate that the + * filesystem is idle and not frozen. + */ +STATIC void +xfs_sync_worker( + struct work_struct *work) +{ + struct xfs_mount *mp = container_of(to_delayed_work(work), + struct xfs_mount, m_sync_work); + int error; + + if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { + /* dgc: errors ignored here */ + if (mp->m_super->s_frozen == SB_UNFROZEN && + xfs_log_need_covered(mp)) + error = xfs_fs_log_dummy(mp); + else + xfs_log_force(mp, 0); + error = xfs_qm_sync(mp, SYNC_TRYLOCK); + + /* start pushing all the metadata that is currently dirty */ + xfs_ail_push_all(mp->m_ail); + } + + /* queue us up again */ + xfs_syncd_queue_sync(mp); +} + +/* + * Queue a new inode reclaim pass if there are reclaimable inodes and there + * isn't a reclaim pass already in progress. By default it runs every 5s based + * on the xfs syncd work default of 30s. Perhaps this should have it's own + * tunable, but that can be done if this method proves to be ineffective or too + * aggressive. + */ +static void +xfs_syncd_queue_reclaim( + struct xfs_mount *mp) +{ + + /* + * We can have inodes enter reclaim after we've shut down the syncd + * workqueue during unmount, so don't allow reclaim work to be queued + * during unmount. + */ + if (!(mp->m_super->s_flags & MS_ACTIVE)) + return; + + rcu_read_lock(); + if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_RECLAIM_TAG)) { + queue_delayed_work(xfs_syncd_wq, &mp->m_reclaim_work, + msecs_to_jiffies(xfs_syncd_centisecs / 6 * 10)); + } + rcu_read_unlock(); +} + +/* + * This is a fast pass over the inode cache to try to get reclaim moving on as + * many inodes as possible in a short period of time. It kicks itself every few + * seconds, as well as being kicked by the inode cache shrinker when memory + * goes low. It scans as quickly as possible avoiding locked inodes or those + * already being flushed, and once done schedules a future pass. + */ +STATIC void +xfs_reclaim_worker( + struct work_struct *work) +{ + struct xfs_mount *mp = container_of(to_delayed_work(work), + struct xfs_mount, m_reclaim_work); + + xfs_reclaim_inodes(mp, SYNC_TRYLOCK); + xfs_syncd_queue_reclaim(mp); +} + +/* + * Flush delayed allocate data, attempting to free up reserved space + * from existing allocations. At this point a new allocation attempt + * has failed with ENOSPC and we are in the process of scratching our + * heads, looking about for more room. + * + * Queue a new data flush if there isn't one already in progress and + * wait for completion of the flush. This means that we only ever have one + * inode flush in progress no matter how many ENOSPC events are occurring and + * so will prevent the system from bogging down due to every concurrent + * ENOSPC event scanning all the active inodes in the system for writeback. + */ +void +xfs_flush_inodes( + struct xfs_inode *ip) +{ + struct xfs_mount *mp = ip->i_mount; + + queue_work(xfs_syncd_wq, &mp->m_flush_work); + flush_work_sync(&mp->m_flush_work); +} + +STATIC void +xfs_flush_worker( + struct work_struct *work) +{ + struct xfs_mount *mp = container_of(work, + struct xfs_mount, m_flush_work); + + xfs_sync_data(mp, SYNC_TRYLOCK); + xfs_sync_data(mp, SYNC_TRYLOCK | SYNC_WAIT); +} + +int +xfs_syncd_init( + struct xfs_mount *mp) +{ + INIT_WORK(&mp->m_flush_work, xfs_flush_worker); + INIT_DELAYED_WORK(&mp->m_sync_work, xfs_sync_worker); + INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker); + + xfs_syncd_queue_sync(mp); + xfs_syncd_queue_reclaim(mp); + + return 0; +} + +void +xfs_syncd_stop( + struct xfs_mount *mp) +{ + cancel_delayed_work_sync(&mp->m_sync_work); + cancel_delayed_work_sync(&mp->m_reclaim_work); + cancel_work_sync(&mp->m_flush_work); +} + +void +__xfs_inode_set_reclaim_tag( + struct xfs_perag *pag, + struct xfs_inode *ip) +{ + radix_tree_tag_set(&pag->pag_ici_root, + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); + + if (!pag->pag_ici_reclaimable) { + /* propagate the reclaim tag up into the perag radix tree */ + spin_lock(&ip->i_mount->m_perag_lock); + radix_tree_tag_set(&ip->i_mount->m_perag_tree, + XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); + spin_unlock(&ip->i_mount->m_perag_lock); + + /* schedule periodic background inode reclaim */ + xfs_syncd_queue_reclaim(ip->i_mount); + + trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno, + -1, _RET_IP_); + } + pag->pag_ici_reclaimable++; +} + +/* + * We set the inode flag atomically with the radix tree tag. + * Once we get tag lookups on the radix tree, this inode flag + * can go away. + */ +void +xfs_inode_set_reclaim_tag( + xfs_inode_t *ip) +{ + struct xfs_mount *mp = ip->i_mount; + struct xfs_perag *pag; + + pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); + spin_lock(&pag->pag_ici_lock); + spin_lock(&ip->i_flags_lock); + __xfs_inode_set_reclaim_tag(pag, ip); + __xfs_iflags_set(ip, XFS_IRECLAIMABLE); + spin_unlock(&ip->i_flags_lock); + spin_unlock(&pag->pag_ici_lock); + xfs_perag_put(pag); +} + +STATIC void +__xfs_inode_clear_reclaim( + xfs_perag_t *pag, + xfs_inode_t *ip) +{ + pag->pag_ici_reclaimable--; + if (!pag->pag_ici_reclaimable) { + /* clear the reclaim tag from the perag radix tree */ + spin_lock(&ip->i_mount->m_perag_lock); + radix_tree_tag_clear(&ip->i_mount->m_perag_tree, + XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), + XFS_ICI_RECLAIM_TAG); + spin_unlock(&ip->i_mount->m_perag_lock); + trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno, + -1, _RET_IP_); + } +} + +void +__xfs_inode_clear_reclaim_tag( + xfs_mount_t *mp, + xfs_perag_t *pag, + xfs_inode_t *ip) +{ + radix_tree_tag_clear(&pag->pag_ici_root, + XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG); + __xfs_inode_clear_reclaim(pag, ip); +} + +/* + * Grab the inode for reclaim exclusively. + * Return 0 if we grabbed it, non-zero otherwise. + */ +STATIC int +xfs_reclaim_inode_grab( + struct xfs_inode *ip, + int flags) +{ + ASSERT(rcu_read_lock_held()); + + /* quick check for stale RCU freed inode */ + if (!ip->i_ino) + return 1; + + /* + * do some unlocked checks first to avoid unnecessary lock traffic. + * The first is a flush lock check, the second is a already in reclaim + * check. Only do these checks if we are not going to block on locks. + */ + if ((flags & SYNC_TRYLOCK) && + (!ip->i_flush.done || __xfs_iflags_test(ip, XFS_IRECLAIM))) { + return 1; + } + + /* + * The radix tree lock here protects a thread in xfs_iget from racing + * with us starting reclaim on the inode. Once we have the + * XFS_IRECLAIM flag set it will not touch us. + * + * Due to RCU lookup, we may find inodes that have been freed and only + * have XFS_IRECLAIM set. Indeed, we may see reallocated inodes that + * aren't candidates for reclaim at all, so we must check the + * XFS_IRECLAIMABLE is set first before proceeding to reclaim. + */ + spin_lock(&ip->i_flags_lock); + if (!__xfs_iflags_test(ip, XFS_IRECLAIMABLE) || + __xfs_iflags_test(ip, XFS_IRECLAIM)) { + /* not a reclaim candidate. */ + spin_unlock(&ip->i_flags_lock); + return 1; + } + __xfs_iflags_set(ip, XFS_IRECLAIM); + spin_unlock(&ip->i_flags_lock); + return 0; +} + +/* + * Inodes in different states need to be treated differently, and the return + * value of xfs_iflush is not sufficient to get this right. The following table + * lists the inode states and the reclaim actions necessary for non-blocking + * reclaim: + * + * + * inode state iflush ret required action + * --------------- ---------- --------------- + * bad - reclaim + * shutdown EIO unpin and reclaim + * clean, unpinned 0 reclaim + * stale, unpinned 0 reclaim + * clean, pinned(*) 0 requeue + * stale, pinned EAGAIN requeue + * dirty, delwri ok 0 requeue + * dirty, delwri blocked EAGAIN requeue + * dirty, sync flush 0 reclaim + * + * (*) dgc: I don't think the clean, pinned state is possible but it gets + * handled anyway given the order of checks implemented. + * + * As can be seen from the table, the return value of xfs_iflush() is not + * sufficient to correctly decide the reclaim action here. The checks in + * xfs_iflush() might look like duplicates, but they are not. + * + * Also, because we get the flush lock first, we know that any inode that has + * been flushed delwri has had the flush completed by the time we check that + * the inode is clean. The clean inode check needs to be done before flushing + * the inode delwri otherwise we would loop forever requeuing clean inodes as + * we cannot tell apart a successful delwri flush and a clean inode from the + * return value of xfs_iflush(). + * + * Note that because the inode is flushed delayed write by background + * writeback, the flush lock may already be held here and waiting on it can + * result in very long latencies. Hence for sync reclaims, where we wait on the + * flush lock, the caller should push out delayed write inodes first before + * trying to reclaim them to minimise the amount of time spent waiting. For + * background relaim, we just requeue the inode for the next pass. + * + * Hence the order of actions after gaining the locks should be: + * bad => reclaim + * shutdown => unpin and reclaim + * pinned, delwri => requeue + * pinned, sync => unpin + * stale => reclaim + * clean => reclaim + * dirty, delwri => flush and requeue + * dirty, sync => flush, wait and reclaim + */ +STATIC int +xfs_reclaim_inode( + struct xfs_inode *ip, + struct xfs_perag *pag, + int sync_mode) +{ + int error; + +restart: + error = 0; + xfs_ilock(ip, XFS_ILOCK_EXCL); + if (!xfs_iflock_nowait(ip)) { + if (!(sync_mode & SYNC_WAIT)) + goto out; + xfs_iflock(ip); + } + + if (is_bad_inode(VFS_I(ip))) + goto reclaim; + if (XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_iunpin_wait(ip); + goto reclaim; + } + if (xfs_ipincount(ip)) { + if (!(sync_mode & SYNC_WAIT)) { + xfs_ifunlock(ip); + goto out; + } + xfs_iunpin_wait(ip); + } + if (xfs_iflags_test(ip, XFS_ISTALE)) + goto reclaim; + if (xfs_inode_clean(ip)) + goto reclaim; + + /* + * Now we have an inode that needs flushing. + * + * We do a nonblocking flush here even if we are doing a SYNC_WAIT + * reclaim as we can deadlock with inode cluster removal. + * xfs_ifree_cluster() can lock the inode buffer before it locks the + * ip->i_lock, and we are doing the exact opposite here. As a result, + * doing a blocking xfs_itobp() to get the cluster buffer will result + * in an ABBA deadlock with xfs_ifree_cluster(). + * + * As xfs_ifree_cluser() must gather all inodes that are active in the + * cache to mark them stale, if we hit this case we don't actually want + * to do IO here - we want the inode marked stale so we can simply + * reclaim it. Hence if we get an EAGAIN error on a SYNC_WAIT flush, + * just unlock the inode, back off and try again. Hopefully the next + * pass through will see the stale flag set on the inode. + */ + error = xfs_iflush(ip, SYNC_TRYLOCK | sync_mode); + if (sync_mode & SYNC_WAIT) { + if (error == EAGAIN) { + xfs_iunlock(ip, XFS_ILOCK_EXCL); + /* backoff longer than in xfs_ifree_cluster */ + delay(2); + goto restart; + } + xfs_iflock(ip); + goto reclaim; + } + + /* + * When we have to flush an inode but don't have SYNC_WAIT set, we + * flush the inode out using a delwri buffer and wait for the next + * call into reclaim to find it in a clean state instead of waiting for + * it now. We also don't return errors here - if the error is transient + * then the next reclaim pass will flush the inode, and if the error + * is permanent then the next sync reclaim will reclaim the inode and + * pass on the error. + */ + if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_warn(ip->i_mount, + "inode 0x%llx background reclaim flush failed with %d", + (long long)ip->i_ino, error); + } +out: + xfs_iflags_clear(ip, XFS_IRECLAIM); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + /* + * We could return EAGAIN here to make reclaim rescan the inode tree in + * a short while. However, this just burns CPU time scanning the tree + * waiting for IO to complete and xfssyncd never goes back to the idle + * state. Instead, return 0 to let the next scheduled background reclaim + * attempt to reclaim the inode again. + */ + return 0; + +reclaim: + xfs_ifunlock(ip); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + + XFS_STATS_INC(xs_ig_reclaims); + /* + * Remove the inode from the per-AG radix tree. + * + * Because radix_tree_delete won't complain even if the item was never + * added to the tree assert that it's been there before to catch + * problems with the inode life time early on. + */ + spin_lock(&pag->pag_ici_lock); + if (!radix_tree_delete(&pag->pag_ici_root, + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino))) + ASSERT(0); + __xfs_inode_clear_reclaim(pag, ip); + spin_unlock(&pag->pag_ici_lock); + + /* + * Here we do an (almost) spurious inode lock in order to coordinate + * with inode cache radix tree lookups. This is because the lookup + * can reference the inodes in the cache without taking references. + * + * We make that OK here by ensuring that we wait until the inode is + * unlocked after the lookup before we go ahead and free it. We get + * both the ilock and the iolock because the code may need to drop the + * ilock one but will still hold the iolock. + */ + xfs_ilock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + xfs_qm_dqdetach(ip); + xfs_iunlock(ip, XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL); + + xfs_inode_free(ip); + return error; + +} + +/* + * Walk the AGs and reclaim the inodes in them. Even if the filesystem is + * corrupted, we still want to try to reclaim all the inodes. If we don't, + * then a shut down during filesystem unmount reclaim walk leak all the + * unreclaimed inodes. + */ +int +xfs_reclaim_inodes_ag( + struct xfs_mount *mp, + int flags, + int *nr_to_scan) +{ + struct xfs_perag *pag; + int error = 0; + int last_error = 0; + xfs_agnumber_t ag; + int trylock = flags & SYNC_TRYLOCK; + int skipped; + +restart: + ag = 0; + skipped = 0; + while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) { + unsigned long first_index = 0; + int done = 0; + int nr_found = 0; + + ag = pag->pag_agno + 1; + + if (trylock) { + if (!mutex_trylock(&pag->pag_ici_reclaim_lock)) { + skipped++; + xfs_perag_put(pag); + continue; + } + first_index = pag->pag_ici_reclaim_cursor; + } else + mutex_lock(&pag->pag_ici_reclaim_lock); + + do { + struct xfs_inode *batch[XFS_LOOKUP_BATCH]; + int i; + + rcu_read_lock(); + nr_found = radix_tree_gang_lookup_tag( + &pag->pag_ici_root, + (void **)batch, first_index, + XFS_LOOKUP_BATCH, + XFS_ICI_RECLAIM_TAG); + if (!nr_found) { + done = 1; + rcu_read_unlock(); + break; + } + + /* + * Grab the inodes before we drop the lock. if we found + * nothing, nr == 0 and the loop will be skipped. + */ + for (i = 0; i < nr_found; i++) { + struct xfs_inode *ip = batch[i]; + + if (done || xfs_reclaim_inode_grab(ip, flags)) + batch[i] = NULL; + + /* + * Update the index for the next lookup. Catch + * overflows into the next AG range which can + * occur if we have inodes in the last block of + * the AG and we are currently pointing to the + * last inode. + * + * Because we may see inodes that are from the + * wrong AG due to RCU freeing and + * reallocation, only update the index if it + * lies in this AG. It was a race that lead us + * to see this inode, so another lookup from + * the same index will not find it again. + */ + if (XFS_INO_TO_AGNO(mp, ip->i_ino) != + pag->pag_agno) + continue; + first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1); + if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) + done = 1; + } + + /* unlock now we've grabbed the inodes. */ + rcu_read_unlock(); + + for (i = 0; i < nr_found; i++) { + if (!batch[i]) + continue; + error = xfs_reclaim_inode(batch[i], pag, flags); + if (error && last_error != EFSCORRUPTED) + last_error = error; + } + + *nr_to_scan -= XFS_LOOKUP_BATCH; + + cond_resched(); + + } while (nr_found && !done && *nr_to_scan > 0); + + if (trylock && !done) + pag->pag_ici_reclaim_cursor = first_index; + else + pag->pag_ici_reclaim_cursor = 0; + mutex_unlock(&pag->pag_ici_reclaim_lock); + xfs_perag_put(pag); + } + + /* + * if we skipped any AG, and we still have scan count remaining, do + * another pass this time using blocking reclaim semantics (i.e + * waiting on the reclaim locks and ignoring the reclaim cursors). This + * ensure that when we get more reclaimers than AGs we block rather + * than spin trying to execute reclaim. + */ + if (skipped && (flags & SYNC_WAIT) && *nr_to_scan > 0) { + trylock = 0; + goto restart; + } + return XFS_ERROR(last_error); +} + +int +xfs_reclaim_inodes( + xfs_mount_t *mp, + int mode) +{ + int nr_to_scan = INT_MAX; + + return xfs_reclaim_inodes_ag(mp, mode, &nr_to_scan); +} + +/* + * Scan a certain number of inodes for reclaim. + * + * When called we make sure that there is a background (fast) inode reclaim in + * progress, while we will throttle the speed of reclaim via doing synchronous + * reclaim of inodes. That means if we come across dirty inodes, we wait for + * them to be cleaned, which we hope will not be very long due to the + * background walker having already kicked the IO off on those dirty inodes. + */ +void +xfs_reclaim_inodes_nr( + struct xfs_mount *mp, + int nr_to_scan) +{ + /* kick background reclaimer and push the AIL */ + xfs_syncd_queue_reclaim(mp); + xfs_ail_push_all(mp->m_ail); + + xfs_reclaim_inodes_ag(mp, SYNC_TRYLOCK | SYNC_WAIT, &nr_to_scan); +} + +/* + * Return the number of reclaimable inodes in the filesystem for + * the shrinker to determine how much to reclaim. + */ +int +xfs_reclaim_inodes_count( + struct xfs_mount *mp) +{ + struct xfs_perag *pag; + xfs_agnumber_t ag = 0; + int reclaimable = 0; + + while ((pag = xfs_perag_get_tag(mp, ag, XFS_ICI_RECLAIM_TAG))) { + ag = pag->pag_agno + 1; + reclaimable += pag->pag_ici_reclaimable; + xfs_perag_put(pag); + } + return reclaimable; +} + diff --git a/fs/xfs/xfs_sync.h b/fs/xfs/xfs_sync.h new file mode 100644 index 000000000000..941202e7ac6e --- /dev/null +++ b/fs/xfs/xfs_sync.h @@ -0,0 +1,51 @@ +/* + * Copyright (c) 2000-2006 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef XFS_SYNC_H +#define XFS_SYNC_H 1 + +struct xfs_mount; +struct xfs_perag; + +#define SYNC_WAIT 0x0001 /* wait for i/o to complete */ +#define SYNC_TRYLOCK 0x0002 /* only try to lock inodes */ + +extern struct workqueue_struct *xfs_syncd_wq; /* sync workqueue */ + +int xfs_syncd_init(struct xfs_mount *mp); +void xfs_syncd_stop(struct xfs_mount *mp); + +int xfs_quiesce_data(struct xfs_mount *mp); +void xfs_quiesce_attr(struct xfs_mount *mp); + +void xfs_flush_inodes(struct xfs_inode *ip); + +int xfs_reclaim_inodes(struct xfs_mount *mp, int mode); +int xfs_reclaim_inodes_count(struct xfs_mount *mp); +void xfs_reclaim_inodes_nr(struct xfs_mount *mp, int nr_to_scan); + +void xfs_inode_set_reclaim_tag(struct xfs_inode *ip); +void __xfs_inode_set_reclaim_tag(struct xfs_perag *pag, struct xfs_inode *ip); +void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag, + struct xfs_inode *ip); + +int xfs_sync_inode_grab(struct xfs_inode *ip); +int xfs_inode_ag_iterator(struct xfs_mount *mp, + int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags), + int flags); + +#endif diff --git a/fs/xfs/xfs_sysctl.c b/fs/xfs/xfs_sysctl.c new file mode 100644 index 000000000000..ee2d2adaa438 --- /dev/null +++ b/fs/xfs/xfs_sysctl.c @@ -0,0 +1,252 @@ +/* + * Copyright (c) 2001-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include +#include +#include "xfs_error.h" + +static struct ctl_table_header *xfs_table_header; + +#ifdef CONFIG_PROC_FS +STATIC int +xfs_stats_clear_proc_handler( + ctl_table *ctl, + int write, + void __user *buffer, + size_t *lenp, + loff_t *ppos) +{ + int c, ret, *valp = ctl->data; + __uint32_t vn_active; + + ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); + + if (!ret && write && *valp) { + xfs_notice(NULL, "Clearing xfsstats"); + for_each_possible_cpu(c) { + preempt_disable(); + /* save vn_active, it's a universal truth! */ + vn_active = per_cpu(xfsstats, c).vn_active; + memset(&per_cpu(xfsstats, c), 0, + sizeof(struct xfsstats)); + per_cpu(xfsstats, c).vn_active = vn_active; + preempt_enable(); + } + xfs_stats_clear = 0; + } + + return ret; +} + +STATIC int +xfs_panic_mask_proc_handler( + ctl_table *ctl, + int write, + void __user *buffer, + size_t *lenp, + loff_t *ppos) +{ + int ret, *valp = ctl->data; + + ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); + if (!ret && write) { + xfs_panic_mask = *valp; +#ifdef DEBUG + xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES); +#endif + } + return ret; +} +#endif /* CONFIG_PROC_FS */ + +static ctl_table xfs_table[] = { + { + .procname = "irix_sgid_inherit", + .data = &xfs_params.sgid_inherit.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.sgid_inherit.min, + .extra2 = &xfs_params.sgid_inherit.max + }, + { + .procname = "irix_symlink_mode", + .data = &xfs_params.symlink_mode.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.symlink_mode.min, + .extra2 = &xfs_params.symlink_mode.max + }, + { + .procname = "panic_mask", + .data = &xfs_params.panic_mask.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = xfs_panic_mask_proc_handler, + .extra1 = &xfs_params.panic_mask.min, + .extra2 = &xfs_params.panic_mask.max + }, + + { + .procname = "error_level", + .data = &xfs_params.error_level.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.error_level.min, + .extra2 = &xfs_params.error_level.max + }, + { + .procname = "xfssyncd_centisecs", + .data = &xfs_params.syncd_timer.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.syncd_timer.min, + .extra2 = &xfs_params.syncd_timer.max + }, + { + .procname = "inherit_sync", + .data = &xfs_params.inherit_sync.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.inherit_sync.min, + .extra2 = &xfs_params.inherit_sync.max + }, + { + .procname = "inherit_nodump", + .data = &xfs_params.inherit_nodump.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.inherit_nodump.min, + .extra2 = &xfs_params.inherit_nodump.max + }, + { + .procname = "inherit_noatime", + .data = &xfs_params.inherit_noatim.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.inherit_noatim.min, + .extra2 = &xfs_params.inherit_noatim.max + }, + { + .procname = "xfsbufd_centisecs", + .data = &xfs_params.xfs_buf_timer.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.xfs_buf_timer.min, + .extra2 = &xfs_params.xfs_buf_timer.max + }, + { + .procname = "age_buffer_centisecs", + .data = &xfs_params.xfs_buf_age.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.xfs_buf_age.min, + .extra2 = &xfs_params.xfs_buf_age.max + }, + { + .procname = "inherit_nosymlinks", + .data = &xfs_params.inherit_nosym.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.inherit_nosym.min, + .extra2 = &xfs_params.inherit_nosym.max + }, + { + .procname = "rotorstep", + .data = &xfs_params.rotorstep.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.rotorstep.min, + .extra2 = &xfs_params.rotorstep.max + }, + { + .procname = "inherit_nodefrag", + .data = &xfs_params.inherit_nodfrg.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.inherit_nodfrg.min, + .extra2 = &xfs_params.inherit_nodfrg.max + }, + { + .procname = "filestream_centisecs", + .data = &xfs_params.fstrm_timer.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &xfs_params.fstrm_timer.min, + .extra2 = &xfs_params.fstrm_timer.max, + }, + /* please keep this the last entry */ +#ifdef CONFIG_PROC_FS + { + .procname = "stats_clear", + .data = &xfs_params.stats_clear.val, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = xfs_stats_clear_proc_handler, + .extra1 = &xfs_params.stats_clear.min, + .extra2 = &xfs_params.stats_clear.max + }, +#endif /* CONFIG_PROC_FS */ + + {} +}; + +static ctl_table xfs_dir_table[] = { + { + .procname = "xfs", + .mode = 0555, + .child = xfs_table + }, + {} +}; + +static ctl_table xfs_root_table[] = { + { + .procname = "fs", + .mode = 0555, + .child = xfs_dir_table + }, + {} +}; + +int +xfs_sysctl_register(void) +{ + xfs_table_header = register_sysctl_table(xfs_root_table); + if (!xfs_table_header) + return -ENOMEM; + return 0; +} + +void +xfs_sysctl_unregister(void) +{ + unregister_sysctl_table(xfs_table_header); +} diff --git a/fs/xfs/xfs_sysctl.h b/fs/xfs/xfs_sysctl.h new file mode 100644 index 000000000000..b9937d450f8e --- /dev/null +++ b/fs/xfs/xfs_sysctl.h @@ -0,0 +1,102 @@ +/* + * Copyright (c) 2001-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_SYSCTL_H__ +#define __XFS_SYSCTL_H__ + +#include + +/* + * Tunable xfs parameters + */ + +typedef struct xfs_sysctl_val { + int min; + int val; + int max; +} xfs_sysctl_val_t; + +typedef struct xfs_param { + xfs_sysctl_val_t sgid_inherit; /* Inherit S_ISGID if process' GID is + * not a member of parent dir GID. */ + xfs_sysctl_val_t symlink_mode; /* Link creat mode affected by umask */ + xfs_sysctl_val_t panic_mask; /* bitmask to cause panic on errors. */ + xfs_sysctl_val_t error_level; /* Degree of reporting for problems */ + xfs_sysctl_val_t syncd_timer; /* Interval between xfssyncd wakeups */ + xfs_sysctl_val_t stats_clear; /* Reset all XFS statistics to zero. */ + xfs_sysctl_val_t inherit_sync; /* Inherit the "sync" inode flag. */ + xfs_sysctl_val_t inherit_nodump;/* Inherit the "nodump" inode flag. */ + xfs_sysctl_val_t inherit_noatim;/* Inherit the "noatime" inode flag. */ + xfs_sysctl_val_t xfs_buf_timer; /* Interval between xfsbufd wakeups. */ + xfs_sysctl_val_t xfs_buf_age; /* Metadata buffer age before flush. */ + xfs_sysctl_val_t inherit_nosym; /* Inherit the "nosymlinks" flag. */ + xfs_sysctl_val_t rotorstep; /* inode32 AG rotoring control knob */ + xfs_sysctl_val_t inherit_nodfrg;/* Inherit the "nodefrag" inode flag. */ + xfs_sysctl_val_t fstrm_timer; /* Filestream dir-AG assoc'n timeout. */ +} xfs_param_t; + +/* + * xfs_error_level: + * + * How much error reporting will be done when internal problems are + * encountered. These problems normally return an EFSCORRUPTED to their + * caller, with no other information reported. + * + * 0 No error reports + * 1 Report EFSCORRUPTED errors that will cause a filesystem shutdown + * 5 Report all EFSCORRUPTED errors (all of the above errors, plus any + * additional errors that are known to not cause shutdowns) + * + * xfs_panic_mask bit 0x8 turns the error reports into panics + */ + +enum { + /* XFS_REFCACHE_SIZE = 1 */ + /* XFS_REFCACHE_PURGE = 2 */ + /* XFS_RESTRICT_CHOWN = 3 */ + XFS_SGID_INHERIT = 4, + XFS_SYMLINK_MODE = 5, + XFS_PANIC_MASK = 6, + XFS_ERRLEVEL = 7, + XFS_SYNCD_TIMER = 8, + /* XFS_PROBE_DMAPI = 9 */ + /* XFS_PROBE_IOOPS = 10 */ + /* XFS_PROBE_QUOTA = 11 */ + XFS_STATS_CLEAR = 12, + XFS_INHERIT_SYNC = 13, + XFS_INHERIT_NODUMP = 14, + XFS_INHERIT_NOATIME = 15, + XFS_BUF_TIMER = 16, + XFS_BUF_AGE = 17, + /* XFS_IO_BYPASS = 18 */ + XFS_INHERIT_NOSYM = 19, + XFS_ROTORSTEP = 20, + XFS_INHERIT_NODFRG = 21, + XFS_FILESTREAM_TIMER = 22, +}; + +extern xfs_param_t xfs_params; + +#ifdef CONFIG_SYSCTL +extern int xfs_sysctl_register(void); +extern void xfs_sysctl_unregister(void); +#else +# define xfs_sysctl_register() (0) +# define xfs_sysctl_unregister() do { } while (0) +#endif /* CONFIG_SYSCTL */ + +#endif /* __XFS_SYSCTL_H__ */ diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c new file mode 100644 index 000000000000..9010ce885e6a --- /dev/null +++ b/fs/xfs/xfs_trace.c @@ -0,0 +1,56 @@ +/* + * Copyright (c) 2009, Christoph Hellwig + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_mount.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" +#include "xfs_alloc.h" +#include "xfs_bmap.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_log_priv.h" +#include "xfs_buf_item.h" +#include "xfs_quota.h" +#include "xfs_iomap.h" +#include "xfs_aops.h" +#include "xfs_dquot_item.h" +#include "xfs_dquot.h" +#include "xfs_log_recover.h" +#include "xfs_inode_item.h" + +/* + * We include this last to have the helpers above available for the trace + * event implementations. + */ +#define CREATE_TRACE_POINTS +#include "xfs_trace.h" diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h new file mode 100644 index 000000000000..690fc7a7bd72 --- /dev/null +++ b/fs/xfs/xfs_trace.h @@ -0,0 +1,1746 @@ +/* + * Copyright (c) 2009, Christoph Hellwig + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM xfs + +#if !defined(_TRACE_XFS_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_XFS_H + +#include + +struct xfs_agf; +struct xfs_alloc_arg; +struct xfs_attr_list_context; +struct xfs_buf_log_item; +struct xfs_da_args; +struct xfs_da_node_entry; +struct xfs_dquot; +struct xlog_ticket; +struct log; +struct xlog_recover; +struct xlog_recover_item; +struct xfs_buf_log_format; +struct xfs_inode_log_format; + +DECLARE_EVENT_CLASS(xfs_attr_list_class, + TP_PROTO(struct xfs_attr_list_context *ctx), + TP_ARGS(ctx), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(u32, hashval) + __field(u32, blkno) + __field(u32, offset) + __field(void *, alist) + __field(int, bufsize) + __field(int, count) + __field(int, firstu) + __field(int, dupcnt) + __field(int, flags) + ), + TP_fast_assign( + __entry->dev = VFS_I(ctx->dp)->i_sb->s_dev; + __entry->ino = ctx->dp->i_ino; + __entry->hashval = ctx->cursor->hashval; + __entry->blkno = ctx->cursor->blkno; + __entry->offset = ctx->cursor->offset; + __entry->alist = ctx->alist; + __entry->bufsize = ctx->bufsize; + __entry->count = ctx->count; + __entry->firstu = ctx->firstu; + __entry->flags = ctx->flags; + ), + TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u " + "alist 0x%p size %u count %u firstu %u flags %d %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->hashval, + __entry->blkno, + __entry->offset, + __entry->dupcnt, + __entry->alist, + __entry->bufsize, + __entry->count, + __entry->firstu, + __entry->flags, + __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS) + ) +) + +#define DEFINE_ATTR_LIST_EVENT(name) \ +DEFINE_EVENT(xfs_attr_list_class, name, \ + TP_PROTO(struct xfs_attr_list_context *ctx), \ + TP_ARGS(ctx)) +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_sf_all); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_leaf_end); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_full); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_add); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_wrong_blk); +DEFINE_ATTR_LIST_EVENT(xfs_attr_list_notfound); + +DECLARE_EVENT_CLASS(xfs_perag_class, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, + unsigned long caller_ip), + TP_ARGS(mp, agno, refcount, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(int, refcount) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->refcount = refcount; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d agno %u refcount %d caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->refcount, + (char *)__entry->caller_ip) +); + +#define DEFINE_PERAG_REF_EVENT(name) \ +DEFINE_EVENT(xfs_perag_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, \ + unsigned long caller_ip), \ + TP_ARGS(mp, agno, refcount, caller_ip)) +DEFINE_PERAG_REF_EVENT(xfs_perag_get); +DEFINE_PERAG_REF_EVENT(xfs_perag_get_tag); +DEFINE_PERAG_REF_EVENT(xfs_perag_put); +DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim); +DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim); + +TRACE_EVENT(xfs_attr_list_node_descend, + TP_PROTO(struct xfs_attr_list_context *ctx, + struct xfs_da_node_entry *btree), + TP_ARGS(ctx, btree), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(u32, hashval) + __field(u32, blkno) + __field(u32, offset) + __field(void *, alist) + __field(int, bufsize) + __field(int, count) + __field(int, firstu) + __field(int, dupcnt) + __field(int, flags) + __field(u32, bt_hashval) + __field(u32, bt_before) + ), + TP_fast_assign( + __entry->dev = VFS_I(ctx->dp)->i_sb->s_dev; + __entry->ino = ctx->dp->i_ino; + __entry->hashval = ctx->cursor->hashval; + __entry->blkno = ctx->cursor->blkno; + __entry->offset = ctx->cursor->offset; + __entry->alist = ctx->alist; + __entry->bufsize = ctx->bufsize; + __entry->count = ctx->count; + __entry->firstu = ctx->firstu; + __entry->flags = ctx->flags; + __entry->bt_hashval = be32_to_cpu(btree->hashval); + __entry->bt_before = be32_to_cpu(btree->before); + ), + TP_printk("dev %d:%d ino 0x%llx cursor h/b/o 0x%x/0x%x/%u dupcnt %u " + "alist 0x%p size %u count %u firstu %u flags %d %s " + "node hashval %u, node before %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->hashval, + __entry->blkno, + __entry->offset, + __entry->dupcnt, + __entry->alist, + __entry->bufsize, + __entry->count, + __entry->firstu, + __entry->flags, + __print_flags(__entry->flags, "|", XFS_ATTR_FLAGS), + __entry->bt_hashval, + __entry->bt_before) +); + +TRACE_EVENT(xfs_iext_insert, + TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, + struct xfs_bmbt_irec *r, int state, unsigned long caller_ip), + TP_ARGS(ip, idx, r, state, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_extnum_t, idx) + __field(xfs_fileoff_t, startoff) + __field(xfs_fsblock_t, startblock) + __field(xfs_filblks_t, blockcount) + __field(xfs_exntst_t, state) + __field(int, bmap_state) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->idx = idx; + __entry->startoff = r->br_startoff; + __entry->startblock = r->br_startblock; + __entry->blockcount = r->br_blockcount; + __entry->state = r->br_state; + __entry->bmap_state = state; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " + "offset %lld block %lld count %lld flag %d caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), + (long)__entry->idx, + __entry->startoff, + (__int64_t)__entry->startblock, + __entry->blockcount, + __entry->state, + (char *)__entry->caller_ip) +); + +DECLARE_EVENT_CLASS(xfs_bmap_class, + TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, + unsigned long caller_ip), + TP_ARGS(ip, idx, state, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_extnum_t, idx) + __field(xfs_fileoff_t, startoff) + __field(xfs_fsblock_t, startblock) + __field(xfs_filblks_t, blockcount) + __field(xfs_exntst_t, state) + __field(int, bmap_state) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + struct xfs_ifork *ifp = (state & BMAP_ATTRFORK) ? + ip->i_afp : &ip->i_df; + struct xfs_bmbt_irec r; + + xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &r); + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->idx = idx; + __entry->startoff = r.br_startoff; + __entry->startblock = r.br_startblock; + __entry->blockcount = r.br_blockcount; + __entry->state = r.br_state; + __entry->bmap_state = state; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d ino 0x%llx state %s idx %ld " + "offset %lld block %lld count %lld flag %d caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS), + (long)__entry->idx, + __entry->startoff, + (__int64_t)__entry->startblock, + __entry->blockcount, + __entry->state, + (char *)__entry->caller_ip) +) + +#define DEFINE_BMAP_EVENT(name) \ +DEFINE_EVENT(xfs_bmap_class, name, \ + TP_PROTO(struct xfs_inode *ip, xfs_extnum_t idx, int state, \ + unsigned long caller_ip), \ + TP_ARGS(ip, idx, state, caller_ip)) +DEFINE_BMAP_EVENT(xfs_iext_remove); +DEFINE_BMAP_EVENT(xfs_bmap_pre_update); +DEFINE_BMAP_EVENT(xfs_bmap_post_update); +DEFINE_BMAP_EVENT(xfs_extlist); + +DECLARE_EVENT_CLASS(xfs_buf_class, + TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip), + TP_ARGS(bp, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_daddr_t, bno) + __field(size_t, buffer_length) + __field(int, hold) + __field(int, pincount) + __field(unsigned, lockval) + __field(unsigned, flags) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = bp->b_target->bt_dev; + __entry->bno = bp->b_bn; + __entry->buffer_length = bp->b_buffer_length; + __entry->hold = atomic_read(&bp->b_hold); + __entry->pincount = atomic_read(&bp->b_pin_count); + __entry->lockval = bp->b_sema.count; + __entry->flags = bp->b_flags; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " + "lock %d flags %s caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->bno, + __entry->buffer_length, + __entry->hold, + __entry->pincount, + __entry->lockval, + __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), + (void *)__entry->caller_ip) +) + +#define DEFINE_BUF_EVENT(name) \ +DEFINE_EVENT(xfs_buf_class, name, \ + TP_PROTO(struct xfs_buf *bp, unsigned long caller_ip), \ + TP_ARGS(bp, caller_ip)) +DEFINE_BUF_EVENT(xfs_buf_init); +DEFINE_BUF_EVENT(xfs_buf_free); +DEFINE_BUF_EVENT(xfs_buf_hold); +DEFINE_BUF_EVENT(xfs_buf_rele); +DEFINE_BUF_EVENT(xfs_buf_iodone); +DEFINE_BUF_EVENT(xfs_buf_iorequest); +DEFINE_BUF_EVENT(xfs_buf_bawrite); +DEFINE_BUF_EVENT(xfs_buf_bdwrite); +DEFINE_BUF_EVENT(xfs_buf_lock); +DEFINE_BUF_EVENT(xfs_buf_lock_done); +DEFINE_BUF_EVENT(xfs_buf_trylock); +DEFINE_BUF_EVENT(xfs_buf_unlock); +DEFINE_BUF_EVENT(xfs_buf_iowait); +DEFINE_BUF_EVENT(xfs_buf_iowait_done); +DEFINE_BUF_EVENT(xfs_buf_delwri_queue); +DEFINE_BUF_EVENT(xfs_buf_delwri_dequeue); +DEFINE_BUF_EVENT(xfs_buf_delwri_split); +DEFINE_BUF_EVENT(xfs_buf_get_uncached); +DEFINE_BUF_EVENT(xfs_bdstrat_shut); +DEFINE_BUF_EVENT(xfs_buf_item_relse); +DEFINE_BUF_EVENT(xfs_buf_item_iodone); +DEFINE_BUF_EVENT(xfs_buf_item_iodone_async); +DEFINE_BUF_EVENT(xfs_buf_error_relse); +DEFINE_BUF_EVENT(xfs_trans_read_buf_io); +DEFINE_BUF_EVENT(xfs_trans_read_buf_shut); + +/* not really buffer traces, but the buf provides useful information */ +DEFINE_BUF_EVENT(xfs_btree_corrupt); +DEFINE_BUF_EVENT(xfs_da_btree_corrupt); +DEFINE_BUF_EVENT(xfs_reset_dqcounts); +DEFINE_BUF_EVENT(xfs_inode_item_push); + +/* pass flags explicitly */ +DECLARE_EVENT_CLASS(xfs_buf_flags_class, + TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip), + TP_ARGS(bp, flags, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_daddr_t, bno) + __field(size_t, buffer_length) + __field(int, hold) + __field(int, pincount) + __field(unsigned, lockval) + __field(unsigned, flags) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = bp->b_target->bt_dev; + __entry->bno = bp->b_bn; + __entry->buffer_length = bp->b_buffer_length; + __entry->flags = flags; + __entry->hold = atomic_read(&bp->b_hold); + __entry->pincount = atomic_read(&bp->b_pin_count); + __entry->lockval = bp->b_sema.count; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " + "lock %d flags %s caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->bno, + __entry->buffer_length, + __entry->hold, + __entry->pincount, + __entry->lockval, + __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), + (void *)__entry->caller_ip) +) + +#define DEFINE_BUF_FLAGS_EVENT(name) \ +DEFINE_EVENT(xfs_buf_flags_class, name, \ + TP_PROTO(struct xfs_buf *bp, unsigned flags, unsigned long caller_ip), \ + TP_ARGS(bp, flags, caller_ip)) +DEFINE_BUF_FLAGS_EVENT(xfs_buf_find); +DEFINE_BUF_FLAGS_EVENT(xfs_buf_get); +DEFINE_BUF_FLAGS_EVENT(xfs_buf_read); + +TRACE_EVENT(xfs_buf_ioerror, + TP_PROTO(struct xfs_buf *bp, int error, unsigned long caller_ip), + TP_ARGS(bp, error, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_daddr_t, bno) + __field(size_t, buffer_length) + __field(unsigned, flags) + __field(int, hold) + __field(int, pincount) + __field(unsigned, lockval) + __field(int, error) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = bp->b_target->bt_dev; + __entry->bno = bp->b_bn; + __entry->buffer_length = bp->b_buffer_length; + __entry->hold = atomic_read(&bp->b_hold); + __entry->pincount = atomic_read(&bp->b_pin_count); + __entry->lockval = bp->b_sema.count; + __entry->error = error; + __entry->flags = bp->b_flags; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " + "lock %d error %d flags %s caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->bno, + __entry->buffer_length, + __entry->hold, + __entry->pincount, + __entry->lockval, + __entry->error, + __print_flags(__entry->flags, "|", XFS_BUF_FLAGS), + (void *)__entry->caller_ip) +); + +DECLARE_EVENT_CLASS(xfs_buf_item_class, + TP_PROTO(struct xfs_buf_log_item *bip), + TP_ARGS(bip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_daddr_t, buf_bno) + __field(size_t, buf_len) + __field(int, buf_hold) + __field(int, buf_pincount) + __field(int, buf_lockval) + __field(unsigned, buf_flags) + __field(unsigned, bli_recur) + __field(int, bli_refcount) + __field(unsigned, bli_flags) + __field(void *, li_desc) + __field(unsigned, li_flags) + ), + TP_fast_assign( + __entry->dev = bip->bli_buf->b_target->bt_dev; + __entry->bli_flags = bip->bli_flags; + __entry->bli_recur = bip->bli_recur; + __entry->bli_refcount = atomic_read(&bip->bli_refcount); + __entry->buf_bno = bip->bli_buf->b_bn; + __entry->buf_len = bip->bli_buf->b_buffer_length; + __entry->buf_flags = bip->bli_buf->b_flags; + __entry->buf_hold = atomic_read(&bip->bli_buf->b_hold); + __entry->buf_pincount = atomic_read(&bip->bli_buf->b_pin_count); + __entry->buf_lockval = bip->bli_buf->b_sema.count; + __entry->li_desc = bip->bli_item.li_desc; + __entry->li_flags = bip->bli_item.li_flags; + ), + TP_printk("dev %d:%d bno 0x%llx len 0x%zx hold %d pincount %d " + "lock %d flags %s recur %d refcount %d bliflags %s " + "lidesc 0x%p liflags %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->buf_bno, + __entry->buf_len, + __entry->buf_hold, + __entry->buf_pincount, + __entry->buf_lockval, + __print_flags(__entry->buf_flags, "|", XFS_BUF_FLAGS), + __entry->bli_recur, + __entry->bli_refcount, + __print_flags(__entry->bli_flags, "|", XFS_BLI_FLAGS), + __entry->li_desc, + __print_flags(__entry->li_flags, "|", XFS_LI_FLAGS)) +) + +#define DEFINE_BUF_ITEM_EVENT(name) \ +DEFINE_EVENT(xfs_buf_item_class, name, \ + TP_PROTO(struct xfs_buf_log_item *bip), \ + TP_ARGS(bip)) +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_size_stale); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_format_stale); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pin); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unpin_stale); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_trylock); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock_stale); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_committed); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_push); +DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pushbuf); +DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf); +DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf_recur); +DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb); +DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb_recur); +DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf); +DEFINE_BUF_ITEM_EVENT(xfs_trans_read_buf_recur); +DEFINE_BUF_ITEM_EVENT(xfs_trans_log_buf); +DEFINE_BUF_ITEM_EVENT(xfs_trans_brelse); +DEFINE_BUF_ITEM_EVENT(xfs_trans_bjoin); +DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold); +DEFINE_BUF_ITEM_EVENT(xfs_trans_bhold_release); +DEFINE_BUF_ITEM_EVENT(xfs_trans_binval); + +DECLARE_EVENT_CLASS(xfs_lock_class, + TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, + unsigned long caller_ip), + TP_ARGS(ip, lock_flags, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, lock_flags) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->lock_flags = lock_flags; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d ino 0x%llx flags %s caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->lock_flags, "|", XFS_LOCK_FLAGS), + (void *)__entry->caller_ip) +) + +#define DEFINE_LOCK_EVENT(name) \ +DEFINE_EVENT(xfs_lock_class, name, \ + TP_PROTO(struct xfs_inode *ip, unsigned lock_flags, \ + unsigned long caller_ip), \ + TP_ARGS(ip, lock_flags, caller_ip)) +DEFINE_LOCK_EVENT(xfs_ilock); +DEFINE_LOCK_EVENT(xfs_ilock_nowait); +DEFINE_LOCK_EVENT(xfs_ilock_demote); +DEFINE_LOCK_EVENT(xfs_iunlock); + +DECLARE_EVENT_CLASS(xfs_inode_class, + TP_PROTO(struct xfs_inode *ip), + TP_ARGS(ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + ), + TP_printk("dev %d:%d ino 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino) +) + +#define DEFINE_INODE_EVENT(name) \ +DEFINE_EVENT(xfs_inode_class, name, \ + TP_PROTO(struct xfs_inode *ip), \ + TP_ARGS(ip)) +DEFINE_INODE_EVENT(xfs_iget_skip); +DEFINE_INODE_EVENT(xfs_iget_reclaim); +DEFINE_INODE_EVENT(xfs_iget_reclaim_fail); +DEFINE_INODE_EVENT(xfs_iget_hit); +DEFINE_INODE_EVENT(xfs_iget_miss); + +DEFINE_INODE_EVENT(xfs_getattr); +DEFINE_INODE_EVENT(xfs_setattr); +DEFINE_INODE_EVENT(xfs_readlink); +DEFINE_INODE_EVENT(xfs_alloc_file_space); +DEFINE_INODE_EVENT(xfs_free_file_space); +DEFINE_INODE_EVENT(xfs_readdir); +#ifdef CONFIG_XFS_POSIX_ACL +DEFINE_INODE_EVENT(xfs_get_acl); +#endif +DEFINE_INODE_EVENT(xfs_vm_bmap); +DEFINE_INODE_EVENT(xfs_file_ioctl); +DEFINE_INODE_EVENT(xfs_file_compat_ioctl); +DEFINE_INODE_EVENT(xfs_ioctl_setattr); +DEFINE_INODE_EVENT(xfs_file_fsync); +DEFINE_INODE_EVENT(xfs_destroy_inode); +DEFINE_INODE_EVENT(xfs_write_inode); +DEFINE_INODE_EVENT(xfs_evict_inode); + +DEFINE_INODE_EVENT(xfs_dquot_dqalloc); +DEFINE_INODE_EVENT(xfs_dquot_dqdetach); + +DECLARE_EVENT_CLASS(xfs_iref_class, + TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), + TP_ARGS(ip, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, count) + __field(int, pincount) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->count = atomic_read(&VFS_I(ip)->i_count); + __entry->pincount = atomic_read(&ip->i_pincount); + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d ino 0x%llx count %d pincount %d caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->count, + __entry->pincount, + (char *)__entry->caller_ip) +) + +#define DEFINE_IREF_EVENT(name) \ +DEFINE_EVENT(xfs_iref_class, name, \ + TP_PROTO(struct xfs_inode *ip, unsigned long caller_ip), \ + TP_ARGS(ip, caller_ip)) +DEFINE_IREF_EVENT(xfs_ihold); +DEFINE_IREF_EVENT(xfs_irele); +DEFINE_IREF_EVENT(xfs_inode_pin); +DEFINE_IREF_EVENT(xfs_inode_unpin); +DEFINE_IREF_EVENT(xfs_inode_unpin_nowait); + +DECLARE_EVENT_CLASS(xfs_namespace_class, + TP_PROTO(struct xfs_inode *dp, struct xfs_name *name), + TP_ARGS(dp, name), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, dp_ino) + __dynamic_array(char, name, name->len) + ), + TP_fast_assign( + __entry->dev = VFS_I(dp)->i_sb->s_dev; + __entry->dp_ino = dp->i_ino; + memcpy(__get_str(name), name->name, name->len); + ), + TP_printk("dev %d:%d dp ino 0x%llx name %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->dp_ino, + __get_str(name)) +) + +#define DEFINE_NAMESPACE_EVENT(name) \ +DEFINE_EVENT(xfs_namespace_class, name, \ + TP_PROTO(struct xfs_inode *dp, struct xfs_name *name), \ + TP_ARGS(dp, name)) +DEFINE_NAMESPACE_EVENT(xfs_remove); +DEFINE_NAMESPACE_EVENT(xfs_link); +DEFINE_NAMESPACE_EVENT(xfs_lookup); +DEFINE_NAMESPACE_EVENT(xfs_create); +DEFINE_NAMESPACE_EVENT(xfs_symlink); + +TRACE_EVENT(xfs_rename, + TP_PROTO(struct xfs_inode *src_dp, struct xfs_inode *target_dp, + struct xfs_name *src_name, struct xfs_name *target_name), + TP_ARGS(src_dp, target_dp, src_name, target_name), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, src_dp_ino) + __field(xfs_ino_t, target_dp_ino) + __dynamic_array(char, src_name, src_name->len) + __dynamic_array(char, target_name, target_name->len) + ), + TP_fast_assign( + __entry->dev = VFS_I(src_dp)->i_sb->s_dev; + __entry->src_dp_ino = src_dp->i_ino; + __entry->target_dp_ino = target_dp->i_ino; + memcpy(__get_str(src_name), src_name->name, src_name->len); + memcpy(__get_str(target_name), target_name->name, target_name->len); + ), + TP_printk("dev %d:%d src dp ino 0x%llx target dp ino 0x%llx" + " src name %s target name %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->src_dp_ino, + __entry->target_dp_ino, + __get_str(src_name), + __get_str(target_name)) +) + +DECLARE_EVENT_CLASS(xfs_dquot_class, + TP_PROTO(struct xfs_dquot *dqp), + TP_ARGS(dqp), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(u32, id) + __field(unsigned, flags) + __field(unsigned, nrefs) + __field(unsigned long long, res_bcount) + __field(unsigned long long, bcount) + __field(unsigned long long, icount) + __field(unsigned long long, blk_hardlimit) + __field(unsigned long long, blk_softlimit) + __field(unsigned long long, ino_hardlimit) + __field(unsigned long long, ino_softlimit) + ), \ + TP_fast_assign( + __entry->dev = dqp->q_mount->m_super->s_dev; + __entry->id = be32_to_cpu(dqp->q_core.d_id); + __entry->flags = dqp->dq_flags; + __entry->nrefs = dqp->q_nrefs; + __entry->res_bcount = dqp->q_res_bcount; + __entry->bcount = be64_to_cpu(dqp->q_core.d_bcount); + __entry->icount = be64_to_cpu(dqp->q_core.d_icount); + __entry->blk_hardlimit = + be64_to_cpu(dqp->q_core.d_blk_hardlimit); + __entry->blk_softlimit = + be64_to_cpu(dqp->q_core.d_blk_softlimit); + __entry->ino_hardlimit = + be64_to_cpu(dqp->q_core.d_ino_hardlimit); + __entry->ino_softlimit = + be64_to_cpu(dqp->q_core.d_ino_softlimit); + ), + TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx " + "bcnt 0x%llx bhardlimit 0x%llx bsoftlimit 0x%llx " + "icnt 0x%llx ihardlimit 0x%llx isoftlimit 0x%llx]", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->id, + __print_flags(__entry->flags, "|", XFS_DQ_FLAGS), + __entry->nrefs, + __entry->res_bcount, + __entry->bcount, + __entry->blk_hardlimit, + __entry->blk_softlimit, + __entry->icount, + __entry->ino_hardlimit, + __entry->ino_softlimit) +) + +#define DEFINE_DQUOT_EVENT(name) \ +DEFINE_EVENT(xfs_dquot_class, name, \ + TP_PROTO(struct xfs_dquot *dqp), \ + TP_ARGS(dqp)) +DEFINE_DQUOT_EVENT(xfs_dqadjust); +DEFINE_DQUOT_EVENT(xfs_dqreclaim_want); +DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty); +DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink); +DEFINE_DQUOT_EVENT(xfs_dqattach_found); +DEFINE_DQUOT_EVENT(xfs_dqattach_get); +DEFINE_DQUOT_EVENT(xfs_dqinit); +DEFINE_DQUOT_EVENT(xfs_dqreuse); +DEFINE_DQUOT_EVENT(xfs_dqalloc); +DEFINE_DQUOT_EVENT(xfs_dqtobp_read); +DEFINE_DQUOT_EVENT(xfs_dqread); +DEFINE_DQUOT_EVENT(xfs_dqread_fail); +DEFINE_DQUOT_EVENT(xfs_dqlookup_found); +DEFINE_DQUOT_EVENT(xfs_dqlookup_want); +DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist); +DEFINE_DQUOT_EVENT(xfs_dqlookup_done); +DEFINE_DQUOT_EVENT(xfs_dqget_hit); +DEFINE_DQUOT_EVENT(xfs_dqget_miss); +DEFINE_DQUOT_EVENT(xfs_dqput); +DEFINE_DQUOT_EVENT(xfs_dqput_wait); +DEFINE_DQUOT_EVENT(xfs_dqput_free); +DEFINE_DQUOT_EVENT(xfs_dqrele); +DEFINE_DQUOT_EVENT(xfs_dqflush); +DEFINE_DQUOT_EVENT(xfs_dqflush_force); +DEFINE_DQUOT_EVENT(xfs_dqflush_done); + +DECLARE_EVENT_CLASS(xfs_loggrant_class, + TP_PROTO(struct log *log, struct xlog_ticket *tic), + TP_ARGS(log, tic), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned, trans_type) + __field(char, ocnt) + __field(char, cnt) + __field(int, curr_res) + __field(int, unit_res) + __field(unsigned int, flags) + __field(int, reserveq) + __field(int, writeq) + __field(int, grant_reserve_cycle) + __field(int, grant_reserve_bytes) + __field(int, grant_write_cycle) + __field(int, grant_write_bytes) + __field(int, curr_cycle) + __field(int, curr_block) + __field(xfs_lsn_t, tail_lsn) + ), + TP_fast_assign( + __entry->dev = log->l_mp->m_super->s_dev; + __entry->trans_type = tic->t_trans_type; + __entry->ocnt = tic->t_ocnt; + __entry->cnt = tic->t_cnt; + __entry->curr_res = tic->t_curr_res; + __entry->unit_res = tic->t_unit_res; + __entry->flags = tic->t_flags; + __entry->reserveq = list_empty(&log->l_reserveq); + __entry->writeq = list_empty(&log->l_writeq); + xlog_crack_grant_head(&log->l_grant_reserve_head, + &__entry->grant_reserve_cycle, + &__entry->grant_reserve_bytes); + xlog_crack_grant_head(&log->l_grant_write_head, + &__entry->grant_write_cycle, + &__entry->grant_write_bytes); + __entry->curr_cycle = log->l_curr_cycle; + __entry->curr_block = log->l_curr_block; + __entry->tail_lsn = atomic64_read(&log->l_tail_lsn); + ), + TP_printk("dev %d:%d type %s t_ocnt %u t_cnt %u t_curr_res %u " + "t_unit_res %u t_flags %s reserveq %s " + "writeq %s grant_reserve_cycle %d " + "grant_reserve_bytes %d grant_write_cycle %d " + "grant_write_bytes %d curr_cycle %d curr_block %d " + "tail_cycle %d tail_block %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __print_symbolic(__entry->trans_type, XFS_TRANS_TYPES), + __entry->ocnt, + __entry->cnt, + __entry->curr_res, + __entry->unit_res, + __print_flags(__entry->flags, "|", XLOG_TIC_FLAGS), + __entry->reserveq ? "empty" : "active", + __entry->writeq ? "empty" : "active", + __entry->grant_reserve_cycle, + __entry->grant_reserve_bytes, + __entry->grant_write_cycle, + __entry->grant_write_bytes, + __entry->curr_cycle, + __entry->curr_block, + CYCLE_LSN(__entry->tail_lsn), + BLOCK_LSN(__entry->tail_lsn) + ) +) + +#define DEFINE_LOGGRANT_EVENT(name) \ +DEFINE_EVENT(xfs_loggrant_class, name, \ + TP_PROTO(struct log *log, struct xlog_ticket *tic), \ + TP_ARGS(log, tic)) +DEFINE_LOGGRANT_EVENT(xfs_log_done_nonperm); +DEFINE_LOGGRANT_EVENT(xfs_log_done_perm); +DEFINE_LOGGRANT_EVENT(xfs_log_reserve); +DEFINE_LOGGRANT_EVENT(xfs_log_umount_write); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_enter); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_exit); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_error); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep1); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake1); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_sleep2); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake2); +DEFINE_LOGGRANT_EVENT(xfs_log_grant_wake_up); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_enter); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_exit); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_error); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep1); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake1); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_sleep2); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake2); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_write_wake_up); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_enter); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_exit); +DEFINE_LOGGRANT_EVENT(xfs_log_regrant_reserve_sub); +DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_enter); +DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_exit); +DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_sub); + +DECLARE_EVENT_CLASS(xfs_file_class, + TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), + TP_ARGS(ip, count, offset, flags), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_fsize_t, size) + __field(xfs_fsize_t, new_size) + __field(loff_t, offset) + __field(size_t, count) + __field(int, flags) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->size = ip->i_d.di_size; + __entry->new_size = ip->i_new_size; + __entry->offset = offset; + __entry->count = count; + __entry->flags = flags; + ), + TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " + "offset 0x%llx count 0x%zx ioflags %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->new_size, + __entry->offset, + __entry->count, + __print_flags(__entry->flags, "|", XFS_IO_FLAGS)) +) + +#define DEFINE_RW_EVENT(name) \ +DEFINE_EVENT(xfs_file_class, name, \ + TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), \ + TP_ARGS(ip, count, offset, flags)) +DEFINE_RW_EVENT(xfs_file_read); +DEFINE_RW_EVENT(xfs_file_buffered_write); +DEFINE_RW_EVENT(xfs_file_direct_write); +DEFINE_RW_EVENT(xfs_file_splice_read); +DEFINE_RW_EVENT(xfs_file_splice_write); + +DECLARE_EVENT_CLASS(xfs_page_class, + TP_PROTO(struct inode *inode, struct page *page, unsigned long off), + TP_ARGS(inode, page, off), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(pgoff_t, pgoff) + __field(loff_t, size) + __field(unsigned long, offset) + __field(int, delalloc) + __field(int, unwritten) + ), + TP_fast_assign( + int delalloc = -1, unwritten = -1; + + if (page_has_buffers(page)) + xfs_count_page_state(page, &delalloc, &unwritten); + __entry->dev = inode->i_sb->s_dev; + __entry->ino = XFS_I(inode)->i_ino; + __entry->pgoff = page_offset(page); + __entry->size = i_size_read(inode); + __entry->offset = off; + __entry->delalloc = delalloc; + __entry->unwritten = unwritten; + ), + TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx " + "delalloc %d unwritten %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->pgoff, + __entry->size, + __entry->offset, + __entry->delalloc, + __entry->unwritten) +) + +#define DEFINE_PAGE_EVENT(name) \ +DEFINE_EVENT(xfs_page_class, name, \ + TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \ + TP_ARGS(inode, page, off)) +DEFINE_PAGE_EVENT(xfs_writepage); +DEFINE_PAGE_EVENT(xfs_releasepage); +DEFINE_PAGE_EVENT(xfs_invalidatepage); + +DECLARE_EVENT_CLASS(xfs_imap_class, + TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, + int type, struct xfs_bmbt_irec *irec), + TP_ARGS(ip, offset, count, type, irec), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(loff_t, size) + __field(loff_t, new_size) + __field(loff_t, offset) + __field(size_t, count) + __field(int, type) + __field(xfs_fileoff_t, startoff) + __field(xfs_fsblock_t, startblock) + __field(xfs_filblks_t, blockcount) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->size = ip->i_d.di_size; + __entry->new_size = ip->i_new_size; + __entry->offset = offset; + __entry->count = count; + __entry->type = type; + __entry->startoff = irec ? irec->br_startoff : 0; + __entry->startblock = irec ? irec->br_startblock : 0; + __entry->blockcount = irec ? irec->br_blockcount : 0; + ), + TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " + "offset 0x%llx count %zd type %s " + "startoff 0x%llx startblock %lld blockcount 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->new_size, + __entry->offset, + __entry->count, + __print_symbolic(__entry->type, XFS_IO_TYPES), + __entry->startoff, + (__int64_t)__entry->startblock, + __entry->blockcount) +) + +#define DEFINE_IOMAP_EVENT(name) \ +DEFINE_EVENT(xfs_imap_class, name, \ + TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, \ + int type, struct xfs_bmbt_irec *irec), \ + TP_ARGS(ip, offset, count, type, irec)) +DEFINE_IOMAP_EVENT(xfs_map_blocks_found); +DEFINE_IOMAP_EVENT(xfs_map_blocks_alloc); +DEFINE_IOMAP_EVENT(xfs_get_blocks_found); +DEFINE_IOMAP_EVENT(xfs_get_blocks_alloc); + +DECLARE_EVENT_CLASS(xfs_simple_io_class, + TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), + TP_ARGS(ip, offset, count), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(loff_t, isize) + __field(loff_t, disize) + __field(loff_t, new_size) + __field(loff_t, offset) + __field(size_t, count) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->isize = ip->i_size; + __entry->disize = ip->i_d.di_size; + __entry->new_size = ip->i_new_size; + __entry->offset = offset; + __entry->count = count; + ), + TP_printk("dev %d:%d ino 0x%llx isize 0x%llx disize 0x%llx new_size 0x%llx " + "offset 0x%llx count %zd", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->isize, + __entry->disize, + __entry->new_size, + __entry->offset, + __entry->count) +); + +#define DEFINE_SIMPLE_IO_EVENT(name) \ +DEFINE_EVENT(xfs_simple_io_class, name, \ + TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), \ + TP_ARGS(ip, offset, count)) +DEFINE_SIMPLE_IO_EVENT(xfs_delalloc_enospc); +DEFINE_SIMPLE_IO_EVENT(xfs_unwritten_convert); +DEFINE_SIMPLE_IO_EVENT(xfs_get_blocks_notfound); +DEFINE_SIMPLE_IO_EVENT(xfs_setfilesize); + +DECLARE_EVENT_CLASS(xfs_itrunc_class, + TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), + TP_ARGS(ip, new_size), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_fsize_t, size) + __field(xfs_fsize_t, new_size) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->size = ip->i_d.di_size; + __entry->new_size = new_size; + ), + TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->new_size) +) + +#define DEFINE_ITRUNC_EVENT(name) \ +DEFINE_EVENT(xfs_itrunc_class, name, \ + TP_PROTO(struct xfs_inode *ip, xfs_fsize_t new_size), \ + TP_ARGS(ip, new_size)) +DEFINE_ITRUNC_EVENT(xfs_itruncate_data_start); +DEFINE_ITRUNC_EVENT(xfs_itruncate_data_end); + +TRACE_EVENT(xfs_pagecache_inval, + TP_PROTO(struct xfs_inode *ip, xfs_off_t start, xfs_off_t finish), + TP_ARGS(ip, start, finish), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_fsize_t, size) + __field(xfs_off_t, start) + __field(xfs_off_t, finish) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->size = ip->i_d.di_size; + __entry->start = start; + __entry->finish = finish; + ), + TP_printk("dev %d:%d ino 0x%llx size 0x%llx start 0x%llx finish 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->start, + __entry->finish) +); + +TRACE_EVENT(xfs_bunmap, + TP_PROTO(struct xfs_inode *ip, xfs_fileoff_t bno, xfs_filblks_t len, + int flags, unsigned long caller_ip), + TP_ARGS(ip, bno, len, flags, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(xfs_fsize_t, size) + __field(xfs_fileoff_t, bno) + __field(xfs_filblks_t, len) + __field(unsigned long, caller_ip) + __field(int, flags) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->ino = ip->i_ino; + __entry->size = ip->i_d.di_size; + __entry->bno = bno; + __entry->len = len; + __entry->caller_ip = caller_ip; + __entry->flags = flags; + ), + TP_printk("dev %d:%d ino 0x%llx size 0x%llx bno 0x%llx len 0x%llx" + "flags %s caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->bno, + __entry->len, + __print_flags(__entry->flags, "|", XFS_BMAPI_FLAGS), + (void *)__entry->caller_ip) + +); + +DECLARE_EVENT_CLASS(xfs_busy_class, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t agbno, xfs_extlen_t len), + TP_ARGS(mp, agno, agbno, len), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len) +); +#define DEFINE_BUSY_EVENT(name) \ +DEFINE_EVENT(xfs_busy_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ + xfs_agblock_t agbno, xfs_extlen_t len), \ + TP_ARGS(mp, agno, agbno, len)) +DEFINE_BUSY_EVENT(xfs_alloc_busy); +DEFINE_BUSY_EVENT(xfs_alloc_busy_enomem); +DEFINE_BUSY_EVENT(xfs_alloc_busy_force); +DEFINE_BUSY_EVENT(xfs_alloc_busy_reuse); +DEFINE_BUSY_EVENT(xfs_alloc_busy_clear); + +TRACE_EVENT(xfs_alloc_busy_trim, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t agbno, xfs_extlen_t len, + xfs_agblock_t tbno, xfs_extlen_t tlen), + TP_ARGS(mp, agno, agbno, len, tbno, tlen), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + __field(xfs_agblock_t, tbno) + __field(xfs_extlen_t, tlen) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + __entry->tbno = tbno; + __entry->tlen = tlen; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u tbno %u tlen %u", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len, + __entry->tbno, + __entry->tlen) +); + +TRACE_EVENT(xfs_trans_commit_lsn, + TP_PROTO(struct xfs_trans *trans), + TP_ARGS(trans), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(struct xfs_trans *, tp) + __field(xfs_lsn_t, lsn) + ), + TP_fast_assign( + __entry->dev = trans->t_mountp->m_super->s_dev; + __entry->tp = trans; + __entry->lsn = trans->t_commit_lsn; + ), + TP_printk("dev %d:%d trans 0x%p commit_lsn 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->tp, + __entry->lsn) +); + +TRACE_EVENT(xfs_agf, + TP_PROTO(struct xfs_mount *mp, struct xfs_agf *agf, int flags, + unsigned long caller_ip), + TP_ARGS(mp, agf, flags, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(int, flags) + __field(__u32, length) + __field(__u32, bno_root) + __field(__u32, cnt_root) + __field(__u32, bno_level) + __field(__u32, cnt_level) + __field(__u32, flfirst) + __field(__u32, fllast) + __field(__u32, flcount) + __field(__u32, freeblks) + __field(__u32, longest) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = be32_to_cpu(agf->agf_seqno), + __entry->flags = flags; + __entry->length = be32_to_cpu(agf->agf_length), + __entry->bno_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]), + __entry->cnt_root = be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]), + __entry->bno_level = + be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]), + __entry->cnt_level = + be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]), + __entry->flfirst = be32_to_cpu(agf->agf_flfirst), + __entry->fllast = be32_to_cpu(agf->agf_fllast), + __entry->flcount = be32_to_cpu(agf->agf_flcount), + __entry->freeblks = be32_to_cpu(agf->agf_freeblks), + __entry->longest = be32_to_cpu(agf->agf_longest); + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d agno %u flags %s length %u roots b %u c %u " + "levels b %u c %u flfirst %u fllast %u flcount %u " + "freeblks %u longest %u caller %pf", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __print_flags(__entry->flags, "|", XFS_AGF_FLAGS), + __entry->length, + __entry->bno_root, + __entry->cnt_root, + __entry->bno_level, + __entry->cnt_level, + __entry->flfirst, + __entry->fllast, + __entry->flcount, + __entry->freeblks, + __entry->longest, + (void *)__entry->caller_ip) +); + +TRACE_EVENT(xfs_free_extent, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno, + xfs_extlen_t len, bool isfl, int haveleft, int haveright), + TP_ARGS(mp, agno, agbno, len, isfl, haveleft, haveright), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + __field(int, isfl) + __field(int, haveleft) + __field(int, haveright) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + __entry->isfl = isfl; + __entry->haveleft = haveleft; + __entry->haveright = haveright; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u isfl %d %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len, + __entry->isfl, + __entry->haveleft ? + (__entry->haveright ? "both" : "left") : + (__entry->haveright ? "right" : "none")) + +); + +DECLARE_EVENT_CLASS(xfs_alloc_class, + TP_PROTO(struct xfs_alloc_arg *args), + TP_ARGS(args), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, minlen) + __field(xfs_extlen_t, maxlen) + __field(xfs_extlen_t, mod) + __field(xfs_extlen_t, prod) + __field(xfs_extlen_t, minleft) + __field(xfs_extlen_t, total) + __field(xfs_extlen_t, alignment) + __field(xfs_extlen_t, minalignslop) + __field(xfs_extlen_t, len) + __field(short, type) + __field(short, otype) + __field(char, wasdel) + __field(char, wasfromfl) + __field(char, isfl) + __field(char, userdata) + __field(xfs_fsblock_t, firstblock) + ), + TP_fast_assign( + __entry->dev = args->mp->m_super->s_dev; + __entry->agno = args->agno; + __entry->agbno = args->agbno; + __entry->minlen = args->minlen; + __entry->maxlen = args->maxlen; + __entry->mod = args->mod; + __entry->prod = args->prod; + __entry->minleft = args->minleft; + __entry->total = args->total; + __entry->alignment = args->alignment; + __entry->minalignslop = args->minalignslop; + __entry->len = args->len; + __entry->type = args->type; + __entry->otype = args->otype; + __entry->wasdel = args->wasdel; + __entry->wasfromfl = args->wasfromfl; + __entry->isfl = args->isfl; + __entry->userdata = args->userdata; + __entry->firstblock = args->firstblock; + ), + TP_printk("dev %d:%d agno %u agbno %u minlen %u maxlen %u mod %u " + "prod %u minleft %u total %u alignment %u minalignslop %u " + "len %u type %s otype %s wasdel %d wasfromfl %d isfl %d " + "userdata %d firstblock 0x%llx", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->minlen, + __entry->maxlen, + __entry->mod, + __entry->prod, + __entry->minleft, + __entry->total, + __entry->alignment, + __entry->minalignslop, + __entry->len, + __print_symbolic(__entry->type, XFS_ALLOC_TYPES), + __print_symbolic(__entry->otype, XFS_ALLOC_TYPES), + __entry->wasdel, + __entry->wasfromfl, + __entry->isfl, + __entry->userdata, + (unsigned long long)__entry->firstblock) +) + +#define DEFINE_ALLOC_EVENT(name) \ +DEFINE_EVENT(xfs_alloc_class, name, \ + TP_PROTO(struct xfs_alloc_arg *args), \ + TP_ARGS(args)) +DEFINE_ALLOC_EVENT(xfs_alloc_exact_done); +DEFINE_ALLOC_EVENT(xfs_alloc_exact_notfound); +DEFINE_ALLOC_EVENT(xfs_alloc_exact_error); +DEFINE_ALLOC_EVENT(xfs_alloc_near_nominleft); +DEFINE_ALLOC_EVENT(xfs_alloc_near_first); +DEFINE_ALLOC_EVENT(xfs_alloc_near_greater); +DEFINE_ALLOC_EVENT(xfs_alloc_near_lesser); +DEFINE_ALLOC_EVENT(xfs_alloc_near_error); +DEFINE_ALLOC_EVENT(xfs_alloc_near_noentry); +DEFINE_ALLOC_EVENT(xfs_alloc_near_busy); +DEFINE_ALLOC_EVENT(xfs_alloc_size_neither); +DEFINE_ALLOC_EVENT(xfs_alloc_size_noentry); +DEFINE_ALLOC_EVENT(xfs_alloc_size_nominleft); +DEFINE_ALLOC_EVENT(xfs_alloc_size_done); +DEFINE_ALLOC_EVENT(xfs_alloc_size_error); +DEFINE_ALLOC_EVENT(xfs_alloc_size_busy); +DEFINE_ALLOC_EVENT(xfs_alloc_small_freelist); +DEFINE_ALLOC_EVENT(xfs_alloc_small_notenough); +DEFINE_ALLOC_EVENT(xfs_alloc_small_done); +DEFINE_ALLOC_EVENT(xfs_alloc_small_error); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_badargs); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_nofix); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_noagbp); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_loopfailed); +DEFINE_ALLOC_EVENT(xfs_alloc_vextent_allfailed); + +DECLARE_EVENT_CLASS(xfs_dir2_class, + TP_PROTO(struct xfs_da_args *args), + TP_ARGS(args), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __dynamic_array(char, name, args->namelen) + __field(int, namelen) + __field(xfs_dahash_t, hashval) + __field(xfs_ino_t, inumber) + __field(int, op_flags) + ), + TP_fast_assign( + __entry->dev = VFS_I(args->dp)->i_sb->s_dev; + __entry->ino = args->dp->i_ino; + if (args->namelen) + memcpy(__get_str(name), args->name, args->namelen); + __entry->namelen = args->namelen; + __entry->hashval = args->hashval; + __entry->inumber = args->inumber; + __entry->op_flags = args->op_flags; + ), + TP_printk("dev %d:%d ino 0x%llx name %.*s namelen %d hashval 0x%x " + "inumber 0x%llx op_flags %s", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->namelen, + __entry->namelen ? __get_str(name) : NULL, + __entry->namelen, + __entry->hashval, + __entry->inumber, + __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS)) +) + +#define DEFINE_DIR2_EVENT(name) \ +DEFINE_EVENT(xfs_dir2_class, name, \ + TP_PROTO(struct xfs_da_args *args), \ + TP_ARGS(args)) +DEFINE_DIR2_EVENT(xfs_dir2_sf_addname); +DEFINE_DIR2_EVENT(xfs_dir2_sf_create); +DEFINE_DIR2_EVENT(xfs_dir2_sf_lookup); +DEFINE_DIR2_EVENT(xfs_dir2_sf_replace); +DEFINE_DIR2_EVENT(xfs_dir2_sf_removename); +DEFINE_DIR2_EVENT(xfs_dir2_sf_toino4); +DEFINE_DIR2_EVENT(xfs_dir2_sf_toino8); +DEFINE_DIR2_EVENT(xfs_dir2_sf_to_block); +DEFINE_DIR2_EVENT(xfs_dir2_block_addname); +DEFINE_DIR2_EVENT(xfs_dir2_block_lookup); +DEFINE_DIR2_EVENT(xfs_dir2_block_replace); +DEFINE_DIR2_EVENT(xfs_dir2_block_removename); +DEFINE_DIR2_EVENT(xfs_dir2_block_to_sf); +DEFINE_DIR2_EVENT(xfs_dir2_block_to_leaf); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_addname); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_lookup); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_replace); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_removename); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_block); +DEFINE_DIR2_EVENT(xfs_dir2_leaf_to_node); +DEFINE_DIR2_EVENT(xfs_dir2_node_addname); +DEFINE_DIR2_EVENT(xfs_dir2_node_lookup); +DEFINE_DIR2_EVENT(xfs_dir2_node_replace); +DEFINE_DIR2_EVENT(xfs_dir2_node_removename); +DEFINE_DIR2_EVENT(xfs_dir2_node_to_leaf); + +DECLARE_EVENT_CLASS(xfs_dir2_space_class, + TP_PROTO(struct xfs_da_args *args, int idx), + TP_ARGS(args, idx), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, op_flags) + __field(int, idx) + ), + TP_fast_assign( + __entry->dev = VFS_I(args->dp)->i_sb->s_dev; + __entry->ino = args->dp->i_ino; + __entry->op_flags = args->op_flags; + __entry->idx = idx; + ), + TP_printk("dev %d:%d ino 0x%llx op_flags %s index %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS), + __entry->idx) +) + +#define DEFINE_DIR2_SPACE_EVENT(name) \ +DEFINE_EVENT(xfs_dir2_space_class, name, \ + TP_PROTO(struct xfs_da_args *args, int idx), \ + TP_ARGS(args, idx)) +DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_add); +DEFINE_DIR2_SPACE_EVENT(xfs_dir2_leafn_remove); +DEFINE_DIR2_SPACE_EVENT(xfs_dir2_grow_inode); +DEFINE_DIR2_SPACE_EVENT(xfs_dir2_shrink_inode); + +TRACE_EVENT(xfs_dir2_leafn_moveents, + TP_PROTO(struct xfs_da_args *args, int src_idx, int dst_idx, int count), + TP_ARGS(args, src_idx, dst_idx, count), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(int, op_flags) + __field(int, src_idx) + __field(int, dst_idx) + __field(int, count) + ), + TP_fast_assign( + __entry->dev = VFS_I(args->dp)->i_sb->s_dev; + __entry->ino = args->dp->i_ino; + __entry->op_flags = args->op_flags; + __entry->src_idx = src_idx; + __entry->dst_idx = dst_idx; + __entry->count = count; + ), + TP_printk("dev %d:%d ino 0x%llx op_flags %s " + "src_idx %d dst_idx %d count %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_flags(__entry->op_flags, "|", XFS_DA_OP_FLAGS), + __entry->src_idx, + __entry->dst_idx, + __entry->count) +); + +#define XFS_SWAPEXT_INODES \ + { 0, "target" }, \ + { 1, "temp" } + +#define XFS_INODE_FORMAT_STR \ + { 0, "invalid" }, \ + { 1, "local" }, \ + { 2, "extent" }, \ + { 3, "btree" } + +DECLARE_EVENT_CLASS(xfs_swap_extent_class, + TP_PROTO(struct xfs_inode *ip, int which), + TP_ARGS(ip, which), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(int, which) + __field(xfs_ino_t, ino) + __field(int, format) + __field(int, nex) + __field(int, max_nex) + __field(int, broot_size) + __field(int, fork_off) + ), + TP_fast_assign( + __entry->dev = VFS_I(ip)->i_sb->s_dev; + __entry->which = which; + __entry->ino = ip->i_ino; + __entry->format = ip->i_d.di_format; + __entry->nex = ip->i_d.di_nextents; + __entry->max_nex = ip->i_df.if_ext_max; + __entry->broot_size = ip->i_df.if_broot_bytes; + __entry->fork_off = XFS_IFORK_BOFF(ip); + ), + TP_printk("dev %d:%d ino 0x%llx (%s), %s format, num_extents %d, " + "Max in-fork extents %d, broot size %d, fork offset %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __print_symbolic(__entry->which, XFS_SWAPEXT_INODES), + __print_symbolic(__entry->format, XFS_INODE_FORMAT_STR), + __entry->nex, + __entry->max_nex, + __entry->broot_size, + __entry->fork_off) +) + +#define DEFINE_SWAPEXT_EVENT(name) \ +DEFINE_EVENT(xfs_swap_extent_class, name, \ + TP_PROTO(struct xfs_inode *ip, int which), \ + TP_ARGS(ip, which)) + +DEFINE_SWAPEXT_EVENT(xfs_swap_extent_before); +DEFINE_SWAPEXT_EVENT(xfs_swap_extent_after); + +DECLARE_EVENT_CLASS(xfs_log_recover_item_class, + TP_PROTO(struct log *log, struct xlog_recover *trans, + struct xlog_recover_item *item, int pass), + TP_ARGS(log, trans, item, pass), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(unsigned long, item) + __field(xlog_tid_t, tid) + __field(int, type) + __field(int, pass) + __field(int, count) + __field(int, total) + ), + TP_fast_assign( + __entry->dev = log->l_mp->m_super->s_dev; + __entry->item = (unsigned long)item; + __entry->tid = trans->r_log_tid; + __entry->type = ITEM_TYPE(item); + __entry->pass = pass; + __entry->count = item->ri_cnt; + __entry->total = item->ri_total; + ), + TP_printk("dev %d:%d trans 0x%x, pass %d, item 0x%p, item type %s " + "item region count/total %d/%d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->tid, + __entry->pass, + (void *)__entry->item, + __print_symbolic(__entry->type, XFS_LI_TYPE_DESC), + __entry->count, + __entry->total) +) + +#define DEFINE_LOG_RECOVER_ITEM(name) \ +DEFINE_EVENT(xfs_log_recover_item_class, name, \ + TP_PROTO(struct log *log, struct xlog_recover *trans, \ + struct xlog_recover_item *item, int pass), \ + TP_ARGS(log, trans, item, pass)) + +DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add); +DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add_cont); +DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_head); +DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_tail); +DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_recover); + +DECLARE_EVENT_CLASS(xfs_log_recover_buf_item_class, + TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), + TP_ARGS(log, buf_f), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(__int64_t, blkno) + __field(unsigned short, len) + __field(unsigned short, flags) + __field(unsigned short, size) + __field(unsigned int, map_size) + ), + TP_fast_assign( + __entry->dev = log->l_mp->m_super->s_dev; + __entry->blkno = buf_f->blf_blkno; + __entry->len = buf_f->blf_len; + __entry->flags = buf_f->blf_flags; + __entry->size = buf_f->blf_size; + __entry->map_size = buf_f->blf_map_size; + ), + TP_printk("dev %d:%d blkno 0x%llx, len %u, flags 0x%x, size %d, " + "map_size %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->blkno, + __entry->len, + __entry->flags, + __entry->size, + __entry->map_size) +) + +#define DEFINE_LOG_RECOVER_BUF_ITEM(name) \ +DEFINE_EVENT(xfs_log_recover_buf_item_class, name, \ + TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), \ + TP_ARGS(log, buf_f)) + +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_not_cancel); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_add); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_ref_inc); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_recover); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_inode_buf); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_reg_buf); +DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_dquot_buf); + +DECLARE_EVENT_CLASS(xfs_log_recover_ino_item_class, + TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), + TP_ARGS(log, in_f), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_ino_t, ino) + __field(unsigned short, size) + __field(int, fields) + __field(unsigned short, asize) + __field(unsigned short, dsize) + __field(__int64_t, blkno) + __field(int, len) + __field(int, boffset) + ), + TP_fast_assign( + __entry->dev = log->l_mp->m_super->s_dev; + __entry->ino = in_f->ilf_ino; + __entry->size = in_f->ilf_size; + __entry->fields = in_f->ilf_fields; + __entry->asize = in_f->ilf_asize; + __entry->dsize = in_f->ilf_dsize; + __entry->blkno = in_f->ilf_blkno; + __entry->len = in_f->ilf_len; + __entry->boffset = in_f->ilf_boffset; + ), + TP_printk("dev %d:%d ino 0x%llx, size %u, fields 0x%x, asize %d, " + "dsize %d, blkno 0x%llx, len %d, boffset %d", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->ino, + __entry->size, + __entry->fields, + __entry->asize, + __entry->dsize, + __entry->blkno, + __entry->len, + __entry->boffset) +) +#define DEFINE_LOG_RECOVER_INO_ITEM(name) \ +DEFINE_EVENT(xfs_log_recover_ino_item_class, name, \ + TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), \ + TP_ARGS(log, in_f)) + +DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover); +DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel); +DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip); + +DECLARE_EVENT_CLASS(xfs_discard_class, + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, + xfs_agblock_t agbno, xfs_extlen_t len), + TP_ARGS(mp, agno, agbno, len), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(xfs_agnumber_t, agno) + __field(xfs_agblock_t, agbno) + __field(xfs_extlen_t, len) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->agno = agno; + __entry->agbno = agbno; + __entry->len = len; + ), + TP_printk("dev %d:%d agno %u agbno %u len %u\n", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->agno, + __entry->agbno, + __entry->len) +) + +#define DEFINE_DISCARD_EVENT(name) \ +DEFINE_EVENT(xfs_discard_class, name, \ + TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \ + xfs_agblock_t agbno, xfs_extlen_t len), \ + TP_ARGS(mp, agno, agbno, len)) +DEFINE_DISCARD_EVENT(xfs_discard_extent); +DEFINE_DISCARD_EVENT(xfs_discard_toosmall); +DEFINE_DISCARD_EVENT(xfs_discard_exclude); +DEFINE_DISCARD_EVENT(xfs_discard_busy); + +#endif /* _TRACE_XFS_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE xfs_trace +#include diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c new file mode 100644 index 000000000000..4d00ee67792d --- /dev/null +++ b/fs/xfs/xfs_trans_dquot.c @@ -0,0 +1,890 @@ +/* + * Copyright (c) 2000-2002 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#include "xfs.h" +#include "xfs_fs.h" +#include "xfs_bit.h" +#include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" +#include "xfs_alloc.h" +#include "xfs_quota.h" +#include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_itable.h" +#include "xfs_bmap.h" +#include "xfs_rtalloc.h" +#include "xfs_error.h" +#include "xfs_attr.h" +#include "xfs_buf_item.h" +#include "xfs_trans_priv.h" +#include "xfs_qm.h" + +STATIC void xfs_trans_alloc_dqinfo(xfs_trans_t *); + +/* + * Add the locked dquot to the transaction. + * The dquot must be locked, and it cannot be associated with any + * transaction. + */ +void +xfs_trans_dqjoin( + xfs_trans_t *tp, + xfs_dquot_t *dqp) +{ + ASSERT(dqp->q_transp != tp); + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + ASSERT(dqp->q_logitem.qli_dquot == dqp); + + /* + * Get a log_item_desc to point at the new item. + */ + xfs_trans_add_item(tp, &dqp->q_logitem.qli_item); + + /* + * Initialize d_transp so we can later determine if this dquot is + * associated with this transaction. + */ + dqp->q_transp = tp; +} + + +/* + * This is called to mark the dquot as needing + * to be logged when the transaction is committed. The dquot must + * already be associated with the given transaction. + * Note that it marks the entire transaction as dirty. In the ordinary + * case, this gets called via xfs_trans_commit, after the transaction + * is already dirty. However, there's nothing stop this from getting + * called directly, as done by xfs_qm_scall_setqlim. Hence, the TRANS_DIRTY + * flag. + */ +void +xfs_trans_log_dquot( + xfs_trans_t *tp, + xfs_dquot_t *dqp) +{ + ASSERT(dqp->q_transp == tp); + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + + tp->t_flags |= XFS_TRANS_DIRTY; + dqp->q_logitem.qli_item.li_desc->lid_flags |= XFS_LID_DIRTY; +} + +/* + * Carry forward whatever is left of the quota blk reservation to + * the spanky new transaction + */ +void +xfs_trans_dup_dqinfo( + xfs_trans_t *otp, + xfs_trans_t *ntp) +{ + xfs_dqtrx_t *oq, *nq; + int i,j; + xfs_dqtrx_t *oqa, *nqa; + + if (!otp->t_dqinfo) + return; + + xfs_trans_alloc_dqinfo(ntp); + oqa = otp->t_dqinfo->dqa_usrdquots; + nqa = ntp->t_dqinfo->dqa_usrdquots; + + /* + * Because the quota blk reservation is carried forward, + * it is also necessary to carry forward the DQ_DIRTY flag. + */ + if(otp->t_flags & XFS_TRANS_DQ_DIRTY) + ntp->t_flags |= XFS_TRANS_DQ_DIRTY; + + for (j = 0; j < 2; j++) { + for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { + if (oqa[i].qt_dquot == NULL) + break; + oq = &oqa[i]; + nq = &nqa[i]; + + nq->qt_dquot = oq->qt_dquot; + nq->qt_bcount_delta = nq->qt_icount_delta = 0; + nq->qt_rtbcount_delta = 0; + + /* + * Transfer whatever is left of the reservations. + */ + nq->qt_blk_res = oq->qt_blk_res - oq->qt_blk_res_used; + oq->qt_blk_res = oq->qt_blk_res_used; + + nq->qt_rtblk_res = oq->qt_rtblk_res - + oq->qt_rtblk_res_used; + oq->qt_rtblk_res = oq->qt_rtblk_res_used; + + nq->qt_ino_res = oq->qt_ino_res - oq->qt_ino_res_used; + oq->qt_ino_res = oq->qt_ino_res_used; + + } + oqa = otp->t_dqinfo->dqa_grpdquots; + nqa = ntp->t_dqinfo->dqa_grpdquots; + } +} + +/* + * Wrap around mod_dquot to account for both user and group quotas. + */ +void +xfs_trans_mod_dquot_byino( + xfs_trans_t *tp, + xfs_inode_t *ip, + uint field, + long delta) +{ + xfs_mount_t *mp = tp->t_mountp; + + if (!XFS_IS_QUOTA_RUNNING(mp) || + !XFS_IS_QUOTA_ON(mp) || + ip->i_ino == mp->m_sb.sb_uquotino || + ip->i_ino == mp->m_sb.sb_gquotino) + return; + + if (tp->t_dqinfo == NULL) + xfs_trans_alloc_dqinfo(tp); + + if (XFS_IS_UQUOTA_ON(mp) && ip->i_udquot) + (void) xfs_trans_mod_dquot(tp, ip->i_udquot, field, delta); + if (XFS_IS_OQUOTA_ON(mp) && ip->i_gdquot) + (void) xfs_trans_mod_dquot(tp, ip->i_gdquot, field, delta); +} + +STATIC xfs_dqtrx_t * +xfs_trans_get_dqtrx( + xfs_trans_t *tp, + xfs_dquot_t *dqp) +{ + int i; + xfs_dqtrx_t *qa; + + qa = XFS_QM_ISUDQ(dqp) ? + tp->t_dqinfo->dqa_usrdquots : tp->t_dqinfo->dqa_grpdquots; + + for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { + if (qa[i].qt_dquot == NULL || + qa[i].qt_dquot == dqp) + return &qa[i]; + } + + return NULL; +} + +/* + * Make the changes in the transaction structure. + * The moral equivalent to xfs_trans_mod_sb(). + * We don't touch any fields in the dquot, so we don't care + * if it's locked or not (most of the time it won't be). + */ +void +xfs_trans_mod_dquot( + xfs_trans_t *tp, + xfs_dquot_t *dqp, + uint field, + long delta) +{ + xfs_dqtrx_t *qtrx; + + ASSERT(tp); + ASSERT(XFS_IS_QUOTA_RUNNING(tp->t_mountp)); + qtrx = NULL; + + if (tp->t_dqinfo == NULL) + xfs_trans_alloc_dqinfo(tp); + /* + * Find either the first free slot or the slot that belongs + * to this dquot. + */ + qtrx = xfs_trans_get_dqtrx(tp, dqp); + ASSERT(qtrx); + if (qtrx->qt_dquot == NULL) + qtrx->qt_dquot = dqp; + + switch (field) { + + /* + * regular disk blk reservation + */ + case XFS_TRANS_DQ_RES_BLKS: + qtrx->qt_blk_res += (ulong)delta; + break; + + /* + * inode reservation + */ + case XFS_TRANS_DQ_RES_INOS: + qtrx->qt_ino_res += (ulong)delta; + break; + + /* + * disk blocks used. + */ + case XFS_TRANS_DQ_BCOUNT: + if (qtrx->qt_blk_res && delta > 0) { + qtrx->qt_blk_res_used += (ulong)delta; + ASSERT(qtrx->qt_blk_res >= qtrx->qt_blk_res_used); + } + qtrx->qt_bcount_delta += delta; + break; + + case XFS_TRANS_DQ_DELBCOUNT: + qtrx->qt_delbcnt_delta += delta; + break; + + /* + * Inode Count + */ + case XFS_TRANS_DQ_ICOUNT: + if (qtrx->qt_ino_res && delta > 0) { + qtrx->qt_ino_res_used += (ulong)delta; + ASSERT(qtrx->qt_ino_res >= qtrx->qt_ino_res_used); + } + qtrx->qt_icount_delta += delta; + break; + + /* + * rtblk reservation + */ + case XFS_TRANS_DQ_RES_RTBLKS: + qtrx->qt_rtblk_res += (ulong)delta; + break; + + /* + * rtblk count + */ + case XFS_TRANS_DQ_RTBCOUNT: + if (qtrx->qt_rtblk_res && delta > 0) { + qtrx->qt_rtblk_res_used += (ulong)delta; + ASSERT(qtrx->qt_rtblk_res >= qtrx->qt_rtblk_res_used); + } + qtrx->qt_rtbcount_delta += delta; + break; + + case XFS_TRANS_DQ_DELRTBCOUNT: + qtrx->qt_delrtb_delta += delta; + break; + + default: + ASSERT(0); + } + tp->t_flags |= XFS_TRANS_DQ_DIRTY; +} + + +/* + * Given an array of dqtrx structures, lock all the dquots associated + * and join them to the transaction, provided they have been modified. + * We know that the highest number of dquots (of one type - usr OR grp), + * involved in a transaction is 2 and that both usr and grp combined - 3. + * So, we don't attempt to make this very generic. + */ +STATIC void +xfs_trans_dqlockedjoin( + xfs_trans_t *tp, + xfs_dqtrx_t *q) +{ + ASSERT(q[0].qt_dquot != NULL); + if (q[1].qt_dquot == NULL) { + xfs_dqlock(q[0].qt_dquot); + xfs_trans_dqjoin(tp, q[0].qt_dquot); + } else { + ASSERT(XFS_QM_TRANS_MAXDQS == 2); + xfs_dqlock2(q[0].qt_dquot, q[1].qt_dquot); + xfs_trans_dqjoin(tp, q[0].qt_dquot); + xfs_trans_dqjoin(tp, q[1].qt_dquot); + } +} + + +/* + * Called by xfs_trans_commit() and similar in spirit to + * xfs_trans_apply_sb_deltas(). + * Go thru all the dquots belonging to this transaction and modify the + * INCORE dquot to reflect the actual usages. + * Unreserve just the reservations done by this transaction. + * dquot is still left locked at exit. + */ +void +xfs_trans_apply_dquot_deltas( + xfs_trans_t *tp) +{ + int i, j; + xfs_dquot_t *dqp; + xfs_dqtrx_t *qtrx, *qa; + xfs_disk_dquot_t *d; + long totalbdelta; + long totalrtbdelta; + + if (!(tp->t_flags & XFS_TRANS_DQ_DIRTY)) + return; + + ASSERT(tp->t_dqinfo); + qa = tp->t_dqinfo->dqa_usrdquots; + for (j = 0; j < 2; j++) { + if (qa[0].qt_dquot == NULL) { + qa = tp->t_dqinfo->dqa_grpdquots; + continue; + } + + /* + * Lock all of the dquots and join them to the transaction. + */ + xfs_trans_dqlockedjoin(tp, qa); + + for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { + qtrx = &qa[i]; + /* + * The array of dquots is filled + * sequentially, not sparsely. + */ + if ((dqp = qtrx->qt_dquot) == NULL) + break; + + ASSERT(XFS_DQ_IS_LOCKED(dqp)); + ASSERT(dqp->q_transp == tp); + + /* + * adjust the actual number of blocks used + */ + d = &dqp->q_core; + + /* + * The issue here is - sometimes we don't make a blkquota + * reservation intentionally to be fair to users + * (when the amount is small). On the other hand, + * delayed allocs do make reservations, but that's + * outside of a transaction, so we have no + * idea how much was really reserved. + * So, here we've accumulated delayed allocation blks and + * non-delay blks. The assumption is that the + * delayed ones are always reserved (outside of a + * transaction), and the others may or may not have + * quota reservations. + */ + totalbdelta = qtrx->qt_bcount_delta + + qtrx->qt_delbcnt_delta; + totalrtbdelta = qtrx->qt_rtbcount_delta + + qtrx->qt_delrtb_delta; +#ifdef DEBUG + if (totalbdelta < 0) + ASSERT(be64_to_cpu(d->d_bcount) >= + -totalbdelta); + + if (totalrtbdelta < 0) + ASSERT(be64_to_cpu(d->d_rtbcount) >= + -totalrtbdelta); + + if (qtrx->qt_icount_delta < 0) + ASSERT(be64_to_cpu(d->d_icount) >= + -qtrx->qt_icount_delta); +#endif + if (totalbdelta) + be64_add_cpu(&d->d_bcount, (xfs_qcnt_t)totalbdelta); + + if (qtrx->qt_icount_delta) + be64_add_cpu(&d->d_icount, (xfs_qcnt_t)qtrx->qt_icount_delta); + + if (totalrtbdelta) + be64_add_cpu(&d->d_rtbcount, (xfs_qcnt_t)totalrtbdelta); + + /* + * Get any default limits in use. + * Start/reset the timer(s) if needed. + */ + if (d->d_id) { + xfs_qm_adjust_dqlimits(tp->t_mountp, d); + xfs_qm_adjust_dqtimers(tp->t_mountp, d); + } + + dqp->dq_flags |= XFS_DQ_DIRTY; + /* + * add this to the list of items to get logged + */ + xfs_trans_log_dquot(tp, dqp); + /* + * Take off what's left of the original reservation. + * In case of delayed allocations, there's no + * reservation that a transaction structure knows of. + */ + if (qtrx->qt_blk_res != 0) { + if (qtrx->qt_blk_res != qtrx->qt_blk_res_used) { + if (qtrx->qt_blk_res > + qtrx->qt_blk_res_used) + dqp->q_res_bcount -= (xfs_qcnt_t) + (qtrx->qt_blk_res - + qtrx->qt_blk_res_used); + else + dqp->q_res_bcount -= (xfs_qcnt_t) + (qtrx->qt_blk_res_used - + qtrx->qt_blk_res); + } + } else { + /* + * These blks were never reserved, either inside + * a transaction or outside one (in a delayed + * allocation). Also, this isn't always a + * negative number since we sometimes + * deliberately skip quota reservations. + */ + if (qtrx->qt_bcount_delta) { + dqp->q_res_bcount += + (xfs_qcnt_t)qtrx->qt_bcount_delta; + } + } + /* + * Adjust the RT reservation. + */ + if (qtrx->qt_rtblk_res != 0) { + if (qtrx->qt_rtblk_res != qtrx->qt_rtblk_res_used) { + if (qtrx->qt_rtblk_res > + qtrx->qt_rtblk_res_used) + dqp->q_res_rtbcount -= (xfs_qcnt_t) + (qtrx->qt_rtblk_res - + qtrx->qt_rtblk_res_used); + else + dqp->q_res_rtbcount -= (xfs_qcnt_t) + (qtrx->qt_rtblk_res_used - + qtrx->qt_rtblk_res); + } + } else { + if (qtrx->qt_rtbcount_delta) + dqp->q_res_rtbcount += + (xfs_qcnt_t)qtrx->qt_rtbcount_delta; + } + + /* + * Adjust the inode reservation. + */ + if (qtrx->qt_ino_res != 0) { + ASSERT(qtrx->qt_ino_res >= + qtrx->qt_ino_res_used); + if (qtrx->qt_ino_res > qtrx->qt_ino_res_used) + dqp->q_res_icount -= (xfs_qcnt_t) + (qtrx->qt_ino_res - + qtrx->qt_ino_res_used); + } else { + if (qtrx->qt_icount_delta) + dqp->q_res_icount += + (xfs_qcnt_t)qtrx->qt_icount_delta; + } + + ASSERT(dqp->q_res_bcount >= + be64_to_cpu(dqp->q_core.d_bcount)); + ASSERT(dqp->q_res_icount >= + be64_to_cpu(dqp->q_core.d_icount)); + ASSERT(dqp->q_res_rtbcount >= + be64_to_cpu(dqp->q_core.d_rtbcount)); + } + /* + * Do the group quotas next + */ + qa = tp->t_dqinfo->dqa_grpdquots; + } +} + +/* + * Release the reservations, and adjust the dquots accordingly. + * This is called only when the transaction is being aborted. If by + * any chance we have done dquot modifications incore (ie. deltas) already, + * we simply throw those away, since that's the expected behavior + * when a transaction is curtailed without a commit. + */ +void +xfs_trans_unreserve_and_mod_dquots( + xfs_trans_t *tp) +{ + int i, j; + xfs_dquot_t *dqp; + xfs_dqtrx_t *qtrx, *qa; + boolean_t locked; + + if (!tp->t_dqinfo || !(tp->t_flags & XFS_TRANS_DQ_DIRTY)) + return; + + qa = tp->t_dqinfo->dqa_usrdquots; + + for (j = 0; j < 2; j++) { + for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) { + qtrx = &qa[i]; + /* + * We assume that the array of dquots is filled + * sequentially, not sparsely. + */ + if ((dqp = qtrx->qt_dquot) == NULL) + break; + /* + * Unreserve the original reservation. We don't care + * about the number of blocks used field, or deltas. + * Also we don't bother to zero the fields. + */ + locked = B_FALSE; + if (qtrx->qt_blk_res) { + xfs_dqlock(dqp); + locked = B_TRUE; + dqp->q_res_bcount -= + (xfs_qcnt_t)qtrx->qt_blk_res; + } + if (qtrx->qt_ino_res) { + if (!locked) { + xfs_dqlock(dqp); + locked = B_TRUE; + } + dqp->q_res_icount -= + (xfs_qcnt_t)qtrx->qt_ino_res; + } + + if (qtrx->qt_rtblk_res) { + if (!locked) { + xfs_dqlock(dqp); + locked = B_TRUE; + } + dqp->q_res_rtbcount -= + (xfs_qcnt_t)qtrx->qt_rtblk_res; + } + if (locked) + xfs_dqunlock(dqp); + + } + qa = tp->t_dqinfo->dqa_grpdquots; + } +} + +STATIC void +xfs_quota_warn( + struct xfs_mount *mp, + struct xfs_dquot *dqp, + int type) +{ + /* no warnings for project quotas - we just return ENOSPC later */ + if (dqp->dq_flags & XFS_DQ_PROJ) + return; + quota_send_warning((dqp->dq_flags & XFS_DQ_USER) ? USRQUOTA : GRPQUOTA, + be32_to_cpu(dqp->q_core.d_id), mp->m_super->s_dev, + type); +} + +/* + * This reserves disk blocks and inodes against a dquot. + * Flags indicate if the dquot is to be locked here and also + * if the blk reservation is for RT or regular blocks. + * Sending in XFS_QMOPT_FORCE_RES flag skips the quota check. + */ +STATIC int +xfs_trans_dqresv( + xfs_trans_t *tp, + xfs_mount_t *mp, + xfs_dquot_t *dqp, + long nblks, + long ninos, + uint flags) +{ + xfs_qcnt_t hardlimit; + xfs_qcnt_t softlimit; + time_t timer; + xfs_qwarncnt_t warns; + xfs_qwarncnt_t warnlimit; + xfs_qcnt_t count; + xfs_qcnt_t *resbcountp; + xfs_quotainfo_t *q = mp->m_quotainfo; + + + xfs_dqlock(dqp); + + if (flags & XFS_TRANS_DQ_RES_BLKS) { + hardlimit = be64_to_cpu(dqp->q_core.d_blk_hardlimit); + if (!hardlimit) + hardlimit = q->qi_bhardlimit; + softlimit = be64_to_cpu(dqp->q_core.d_blk_softlimit); + if (!softlimit) + softlimit = q->qi_bsoftlimit; + timer = be32_to_cpu(dqp->q_core.d_btimer); + warns = be16_to_cpu(dqp->q_core.d_bwarns); + warnlimit = dqp->q_mount->m_quotainfo->qi_bwarnlimit; + resbcountp = &dqp->q_res_bcount; + } else { + ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS); + hardlimit = be64_to_cpu(dqp->q_core.d_rtb_hardlimit); + if (!hardlimit) + hardlimit = q->qi_rtbhardlimit; + softlimit = be64_to_cpu(dqp->q_core.d_rtb_softlimit); + if (!softlimit) + softlimit = q->qi_rtbsoftlimit; + timer = be32_to_cpu(dqp->q_core.d_rtbtimer); + warns = be16_to_cpu(dqp->q_core.d_rtbwarns); + warnlimit = dqp->q_mount->m_quotainfo->qi_rtbwarnlimit; + resbcountp = &dqp->q_res_rtbcount; + } + + if ((flags & XFS_QMOPT_FORCE_RES) == 0 && + dqp->q_core.d_id && + ((XFS_IS_UQUOTA_ENFORCED(dqp->q_mount) && XFS_QM_ISUDQ(dqp)) || + (XFS_IS_OQUOTA_ENFORCED(dqp->q_mount) && + (XFS_QM_ISPDQ(dqp) || XFS_QM_ISGDQ(dqp))))) { + if (nblks > 0) { + /* + * dquot is locked already. See if we'd go over the + * hardlimit or exceed the timelimit if we allocate + * nblks. + */ + if (hardlimit > 0ULL && + hardlimit <= nblks + *resbcountp) { + xfs_quota_warn(mp, dqp, QUOTA_NL_BHARDWARN); + goto error_return; + } + if (softlimit > 0ULL && + softlimit <= nblks + *resbcountp) { + if ((timer != 0 && get_seconds() > timer) || + (warns != 0 && warns >= warnlimit)) { + xfs_quota_warn(mp, dqp, + QUOTA_NL_BSOFTLONGWARN); + goto error_return; + } + + xfs_quota_warn(mp, dqp, QUOTA_NL_BSOFTWARN); + } + } + if (ninos > 0) { + count = be64_to_cpu(dqp->q_core.d_icount); + timer = be32_to_cpu(dqp->q_core.d_itimer); + warns = be16_to_cpu(dqp->q_core.d_iwarns); + warnlimit = dqp->q_mount->m_quotainfo->qi_iwarnlimit; + hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit); + if (!hardlimit) + hardlimit = q->qi_ihardlimit; + softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit); + if (!softlimit) + softlimit = q->qi_isoftlimit; + + if (hardlimit > 0ULL && count >= hardlimit) { + xfs_quota_warn(mp, dqp, QUOTA_NL_IHARDWARN); + goto error_return; + } + if (softlimit > 0ULL && count >= softlimit) { + if ((timer != 0 && get_seconds() > timer) || + (warns != 0 && warns >= warnlimit)) { + xfs_quota_warn(mp, dqp, + QUOTA_NL_ISOFTLONGWARN); + goto error_return; + } + xfs_quota_warn(mp, dqp, QUOTA_NL_ISOFTWARN); + } + } + } + + /* + * Change the reservation, but not the actual usage. + * Note that q_res_bcount = q_core.d_bcount + resv + */ + (*resbcountp) += (xfs_qcnt_t)nblks; + if (ninos != 0) + dqp->q_res_icount += (xfs_qcnt_t)ninos; + + /* + * note the reservation amt in the trans struct too, + * so that the transaction knows how much was reserved by + * it against this particular dquot. + * We don't do this when we are reserving for a delayed allocation, + * because we don't have the luxury of a transaction envelope then. + */ + if (tp) { + ASSERT(tp->t_dqinfo); + ASSERT(flags & XFS_QMOPT_RESBLK_MASK); + if (nblks != 0) + xfs_trans_mod_dquot(tp, dqp, + flags & XFS_QMOPT_RESBLK_MASK, + nblks); + if (ninos != 0) + xfs_trans_mod_dquot(tp, dqp, + XFS_TRANS_DQ_RES_INOS, + ninos); + } + ASSERT(dqp->q_res_bcount >= be64_to_cpu(dqp->q_core.d_bcount)); + ASSERT(dqp->q_res_rtbcount >= be64_to_cpu(dqp->q_core.d_rtbcount)); + ASSERT(dqp->q_res_icount >= be64_to_cpu(dqp->q_core.d_icount)); + + xfs_dqunlock(dqp); + return 0; + +error_return: + xfs_dqunlock(dqp); + if (flags & XFS_QMOPT_ENOSPC) + return ENOSPC; + return EDQUOT; +} + + +/* + * Given dquot(s), make disk block and/or inode reservations against them. + * The fact that this does the reservation against both the usr and + * grp/prj quotas is important, because this follows a both-or-nothing + * approach. + * + * flags = XFS_QMOPT_FORCE_RES evades limit enforcement. Used by chown. + * XFS_QMOPT_ENOSPC returns ENOSPC not EDQUOT. Used by pquota. + * XFS_TRANS_DQ_RES_BLKS reserves regular disk blocks + * XFS_TRANS_DQ_RES_RTBLKS reserves realtime disk blocks + * dquots are unlocked on return, if they were not locked by caller. + */ +int +xfs_trans_reserve_quota_bydquots( + xfs_trans_t *tp, + xfs_mount_t *mp, + xfs_dquot_t *udqp, + xfs_dquot_t *gdqp, + long nblks, + long ninos, + uint flags) +{ + int resvd = 0, error; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + + if (tp && tp->t_dqinfo == NULL) + xfs_trans_alloc_dqinfo(tp); + + ASSERT(flags & XFS_QMOPT_RESBLK_MASK); + + if (udqp) { + error = xfs_trans_dqresv(tp, mp, udqp, nblks, ninos, + (flags & ~XFS_QMOPT_ENOSPC)); + if (error) + return error; + resvd = 1; + } + + if (gdqp) { + error = xfs_trans_dqresv(tp, mp, gdqp, nblks, ninos, flags); + if (error) { + /* + * can't do it, so backout previous reservation + */ + if (resvd) { + flags |= XFS_QMOPT_FORCE_RES; + xfs_trans_dqresv(tp, mp, udqp, + -nblks, -ninos, flags); + } + return error; + } + } + + /* + * Didn't change anything critical, so, no need to log + */ + return 0; +} + + +/* + * Lock the dquot and change the reservation if we can. + * This doesn't change the actual usage, just the reservation. + * The inode sent in is locked. + */ +int +xfs_trans_reserve_quota_nblks( + struct xfs_trans *tp, + struct xfs_inode *ip, + long nblks, + long ninos, + uint flags) +{ + struct xfs_mount *mp = ip->i_mount; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + if (XFS_IS_PQUOTA_ON(mp)) + flags |= XFS_QMOPT_ENOSPC; + + ASSERT(ip->i_ino != mp->m_sb.sb_uquotino); + ASSERT(ip->i_ino != mp->m_sb.sb_gquotino); + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + ASSERT((flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) == + XFS_TRANS_DQ_RES_RTBLKS || + (flags & ~(XFS_QMOPT_FORCE_RES | XFS_QMOPT_ENOSPC)) == + XFS_TRANS_DQ_RES_BLKS); + + /* + * Reserve nblks against these dquots, with trans as the mediator. + */ + return xfs_trans_reserve_quota_bydquots(tp, mp, + ip->i_udquot, ip->i_gdquot, + nblks, ninos, flags); +} + +/* + * This routine is called to allocate a quotaoff log item. + */ +xfs_qoff_logitem_t * +xfs_trans_get_qoff_item( + xfs_trans_t *tp, + xfs_qoff_logitem_t *startqoff, + uint flags) +{ + xfs_qoff_logitem_t *q; + + ASSERT(tp != NULL); + + q = xfs_qm_qoff_logitem_init(tp->t_mountp, startqoff, flags); + ASSERT(q != NULL); + + /* + * Get a log_item_desc to point at the new item. + */ + xfs_trans_add_item(tp, &q->qql_item); + return q; +} + + +/* + * This is called to mark the quotaoff logitem as needing + * to be logged when the transaction is committed. The logitem must + * already be associated with the given transaction. + */ +void +xfs_trans_log_quotaoff_item( + xfs_trans_t *tp, + xfs_qoff_logitem_t *qlp) +{ + tp->t_flags |= XFS_TRANS_DIRTY; + qlp->qql_item.li_desc->lid_flags |= XFS_LID_DIRTY; +} + +STATIC void +xfs_trans_alloc_dqinfo( + xfs_trans_t *tp) +{ + tp->t_dqinfo = kmem_zone_zalloc(xfs_Gqm->qm_dqtrxzone, KM_SLEEP); +} + +void +xfs_trans_free_dqinfo( + xfs_trans_t *tp) +{ + if (!tp->t_dqinfo) + return; + kmem_zone_free(xfs_Gqm->qm_dqtrxzone, tp->t_dqinfo); + tp->t_dqinfo = NULL; +} diff --git a/fs/xfs/xfs_vnode.h b/fs/xfs/xfs_vnode.h new file mode 100644 index 000000000000..7c220b4227bc --- /dev/null +++ b/fs/xfs/xfs_vnode.h @@ -0,0 +1,64 @@ +/* + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ +#ifndef __XFS_VNODE_H__ +#define __XFS_VNODE_H__ + +#include "xfs_fs.h" + +struct file; +struct xfs_inode; +struct xfs_iomap; +struct attrlist_cursor_kern; + +/* + * Return values for xfs_inactive. A return value of + * VN_INACTIVE_NOCACHE implies that the file system behavior + * has disassociated its state and bhv_desc_t from the vnode. + */ +#define VN_INACTIVE_CACHE 0 +#define VN_INACTIVE_NOCACHE 1 + +/* + * Flags for read/write calls - same values as IRIX + */ +#define IO_ISDIRECT 0x00004 /* bypass page cache */ +#define IO_INVIS 0x00020 /* don't update inode timestamps */ + +#define XFS_IO_FLAGS \ + { IO_ISDIRECT, "DIRECT" }, \ + { IO_INVIS, "INVIS"} + +/* + * Flush/Invalidate options for vop_toss/flush/flushinval_pages. + */ +#define FI_NONE 0 /* none */ +#define FI_REMAPF 1 /* Do a remapf prior to the operation */ +#define FI_REMAPF_LOCKED 2 /* Do a remapf prior to the operation. + Prevent VM access to the pages until + the operation completes. */ + +/* + * Some useful predicates. + */ +#define VN_MAPPED(vp) mapping_mapped(vp->i_mapping) +#define VN_CACHED(vp) (vp->i_mapping->nrpages) +#define VN_DIRTY(vp) mapping_tagged(vp->i_mapping, \ + PAGECACHE_TAG_DIRTY) + + +#endif /* __XFS_VNODE_H__ */ diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c new file mode 100644 index 000000000000..87d3e03878c8 --- /dev/null +++ b/fs/xfs/xfs_xattr.c @@ -0,0 +1,241 @@ +/* + * Copyright (C) 2008 Christoph Hellwig. + * Portions Copyright (C) 2000-2008 Silicon Graphics, Inc. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include "xfs.h" +#include "xfs_da_btree.h" +#include "xfs_bmap_btree.h" +#include "xfs_inode.h" +#include "xfs_attr.h" +#include "xfs_attr_leaf.h" +#include "xfs_acl.h" +#include "xfs_vnodeops.h" + +#include +#include + + +static int +xfs_xattr_get(struct dentry *dentry, const char *name, + void *value, size_t size, int xflags) +{ + struct xfs_inode *ip = XFS_I(dentry->d_inode); + int error, asize = size; + + if (strcmp(name, "") == 0) + return -EINVAL; + + /* Convert Linux syscall to XFS internal ATTR flags */ + if (!size) { + xflags |= ATTR_KERNOVAL; + value = NULL; + } + + error = -xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags); + if (error) + return error; + return asize; +} + +static int +xfs_xattr_set(struct dentry *dentry, const char *name, const void *value, + size_t size, int flags, int xflags) +{ + struct xfs_inode *ip = XFS_I(dentry->d_inode); + + if (strcmp(name, "") == 0) + return -EINVAL; + + /* Convert Linux syscall to XFS internal ATTR flags */ + if (flags & XATTR_CREATE) + xflags |= ATTR_CREATE; + if (flags & XATTR_REPLACE) + xflags |= ATTR_REPLACE; + + if (!value) + return -xfs_attr_remove(ip, (unsigned char *)name, xflags); + return -xfs_attr_set(ip, (unsigned char *)name, + (void *)value, size, xflags); +} + +static const struct xattr_handler xfs_xattr_user_handler = { + .prefix = XATTR_USER_PREFIX, + .flags = 0, /* no flags implies user namespace */ + .get = xfs_xattr_get, + .set = xfs_xattr_set, +}; + +static const struct xattr_handler xfs_xattr_trusted_handler = { + .prefix = XATTR_TRUSTED_PREFIX, + .flags = ATTR_ROOT, + .get = xfs_xattr_get, + .set = xfs_xattr_set, +}; + +static const struct xattr_handler xfs_xattr_security_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .flags = ATTR_SECURE, + .get = xfs_xattr_get, + .set = xfs_xattr_set, +}; + +const struct xattr_handler *xfs_xattr_handlers[] = { + &xfs_xattr_user_handler, + &xfs_xattr_trusted_handler, + &xfs_xattr_security_handler, +#ifdef CONFIG_XFS_POSIX_ACL + &xfs_xattr_acl_access_handler, + &xfs_xattr_acl_default_handler, +#endif + NULL +}; + +static unsigned int xfs_xattr_prefix_len(int flags) +{ + if (flags & XFS_ATTR_SECURE) + return sizeof("security"); + else if (flags & XFS_ATTR_ROOT) + return sizeof("trusted"); + else + return sizeof("user"); +} + +static const char *xfs_xattr_prefix(int flags) +{ + if (flags & XFS_ATTR_SECURE) + return xfs_xattr_security_handler.prefix; + else if (flags & XFS_ATTR_ROOT) + return xfs_xattr_trusted_handler.prefix; + else + return xfs_xattr_user_handler.prefix; +} + +static int +xfs_xattr_put_listent( + struct xfs_attr_list_context *context, + int flags, + unsigned char *name, + int namelen, + int valuelen, + unsigned char *value) +{ + unsigned int prefix_len = xfs_xattr_prefix_len(flags); + char *offset; + int arraytop; + + ASSERT(context->count >= 0); + + /* + * Only show root namespace entries if we are actually allowed to + * see them. + */ + if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN)) + return 0; + + arraytop = context->count + prefix_len + namelen + 1; + if (arraytop > context->firstu) { + context->count = -1; /* insufficient space */ + return 1; + } + offset = (char *)context->alist + context->count; + strncpy(offset, xfs_xattr_prefix(flags), prefix_len); + offset += prefix_len; + strncpy(offset, (char *)name, namelen); /* real name */ + offset += namelen; + *offset = '\0'; + context->count += prefix_len + namelen + 1; + return 0; +} + +static int +xfs_xattr_put_listent_sizes( + struct xfs_attr_list_context *context, + int flags, + unsigned char *name, + int namelen, + int valuelen, + unsigned char *value) +{ + context->count += xfs_xattr_prefix_len(flags) + namelen + 1; + return 0; +} + +static int +list_one_attr(const char *name, const size_t len, void *data, + size_t size, ssize_t *result) +{ + char *p = data + *result; + + *result += len; + if (!size) + return 0; + if (*result > size) + return -ERANGE; + + strcpy(p, name); + return 0; +} + +ssize_t +xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size) +{ + struct xfs_attr_list_context context; + struct attrlist_cursor_kern cursor = { 0 }; + struct inode *inode = dentry->d_inode; + int error; + + /* + * First read the regular on-disk attributes. + */ + memset(&context, 0, sizeof(context)); + context.dp = XFS_I(inode); + context.cursor = &cursor; + context.resynch = 1; + context.alist = data; + context.bufsize = size; + context.firstu = context.bufsize; + + if (size) + context.put_listent = xfs_xattr_put_listent; + else + context.put_listent = xfs_xattr_put_listent_sizes; + + xfs_attr_list_int(&context); + if (context.count < 0) + return -ERANGE; + + /* + * Then add the two synthetic ACL attributes. + */ + if (posix_acl_access_exists(inode)) { + error = list_one_attr(POSIX_ACL_XATTR_ACCESS, + strlen(POSIX_ACL_XATTR_ACCESS) + 1, + data, size, &context.count); + if (error) + return error; + } + + if (posix_acl_default_exists(inode)) { + error = list_one_attr(POSIX_ACL_XATTR_DEFAULT, + strlen(POSIX_ACL_XATTR_DEFAULT) + 1, + data, size, &context.count); + if (error) + return error; + } + + return context.count; +} diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 3b8e028b9601..e8bffbe2ba4b 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -1,6 +1,6 @@ #include #include -#include "../fs/xfs/linux-2.6/xfs_sysctl.h" +#include "../fs/xfs/xfs_sysctl.h" #include #include #include diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c index 4e4932a7b360..362da653813d 100644 --- a/kernel/sysctl_check.c +++ b/kernel/sysctl_check.c @@ -1,6 +1,6 @@ #include #include -#include "../fs/xfs/linux-2.6/xfs_sysctl.h" +#include "../fs/xfs/xfs_sysctl.h" #include #include #include -- cgit v1.3-8-gc7d7 From 144060fee07e9c22e179d00819c83c86fbcbf82c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 1 Aug 2011 12:49:14 +0200 Subject: perf: Add PM notifiers to fix CPU hotplug races Francis reports that s2r gets him spurious NMIs, this is because the suspend code leaves the boot cpu up and running. Cure this by adding a suspend notifier. The problem is that hotplug and suspend are completely un-serialized and the PM notifiers run before the suspend cpu unplug of all but the boot cpu. This leaves a window where the user can initialize another hotplug operation (either remove or add a cpu) resulting in either one too many or one too few hotplug ops. Thus we cannot use the hotplug code for the suspend case. There's another reason to not use the hotplug code, which is that the hotplug code totally destroys the perf state, we can do better for suspend and simply remove all counters from the PMU so that we can re-instate them on resume. Reported-by: Francis Moreau Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-1cvevybkgmv4s6v5y37t4847@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 97 ++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 95 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index b8785e26ee1c..d4c85425e3a0 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include #include @@ -6809,7 +6810,7 @@ static void __cpuinit perf_event_init_cpu(int cpu) struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); mutex_lock(&swhash->hlist_mutex); - if (swhash->hlist_refcount > 0) { + if (swhash->hlist_refcount > 0 && !swhash->swevent_hlist) { struct swevent_hlist *hlist; hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu)); @@ -6898,7 +6899,14 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; - switch (action & ~CPU_TASKS_FROZEN) { + /* + * Ignore suspend/resume action, the perf_pm_notifier will + * take care of that. + */ + if (action & CPU_TASKS_FROZEN) + return NOTIFY_OK; + + switch (action) { case CPU_UP_PREPARE: case CPU_DOWN_FAILED: @@ -6917,6 +6925,90 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } +static void perf_pm_resume_cpu(void *unused) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct pmu *pmu; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + ctx = cpuctx->task_ctx; + + perf_ctx_lock(cpuctx, ctx); + perf_pmu_disable(cpuctx->ctx.pmu); + + cpu_ctx_sched_out(cpuctx, EVENT_ALL); + if (ctx) + ctx_sched_out(ctx, cpuctx, EVENT_ALL); + + perf_pmu_enable(cpuctx->ctx.pmu); + perf_ctx_unlock(cpuctx, ctx); + } + srcu_read_unlock(&pmus_srcu, idx); +} + +static void perf_pm_suspend_cpu(void *unused) +{ + struct perf_cpu_context *cpuctx; + struct perf_event_context *ctx; + struct pmu *pmu; + int idx; + + idx = srcu_read_lock(&pmus_srcu); + list_for_each_entry_rcu(pmu, &pmus, entry) { + cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); + ctx = cpuctx->task_ctx; + + perf_ctx_lock(cpuctx, ctx); + perf_pmu_disable(cpuctx->ctx.pmu); + + perf_event_sched_in(cpuctx, ctx, current); + + perf_pmu_enable(cpuctx->ctx.pmu); + perf_ctx_unlock(cpuctx, ctx); + } + srcu_read_unlock(&pmus_srcu, idx); +} + +static int perf_resume(void) +{ + get_online_cpus(); + smp_call_function(perf_pm_resume_cpu, NULL, 1); + put_online_cpus(); + + return NOTIFY_OK; +} + +static int perf_suspend(void) +{ + get_online_cpus(); + smp_call_function(perf_pm_suspend_cpu, NULL, 1); + put_online_cpus(); + + return NOTIFY_OK; +} + +static int perf_pm(struct notifier_block *self, unsigned long action, void *ptr) +{ + switch (action) { + case PM_POST_HIBERNATION: + case PM_POST_SUSPEND: + return perf_resume(); + case PM_HIBERNATION_PREPARE: + case PM_SUSPEND_PREPARE: + return perf_suspend(); + default: + return NOTIFY_DONE; + } +} + +static struct notifier_block perf_pm_notifier = { + .notifier_call = perf_pm, +}; + void __init perf_event_init(void) { int ret; @@ -6931,6 +7023,7 @@ void __init perf_event_init(void) perf_tp_register(); perf_cpu_notifier(perf_cpu_notify); register_reboot_notifier(&perf_reboot_notifier); + register_pm_notifier(&perf_pm_notifier); ret = init_hw_breakpoint(); WARN(ret, "hw_breakpoint initialization failed with: %d", ret); -- cgit v1.3-8-gc7d7 From 7e5b2a01d2ca2eae4ef913b59f84341f9a70e206 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 11 Aug 2011 12:31:20 +0100 Subject: perf: provide PMU when initing events Currently, an event's 'pmu' field is set after pmu::event_init() is called. This means that pmu::event_init() must figure out which struct pmu the event was initialised from. This makes it difficult to consolidate common event initialisation code for similar PMUs, and very difficult to implement drivers for PMUs which can have multiple instances (e.g. a USB controller PMU, a GPU PMU, etc). This patch sets the 'pmu' field before initialising the event, allowing event init code to identify the struct pmu instance easily. In the event of failure to initialise an event, the event is destroyed via kfree() without calling perf_event::destroy(), so this shouldn't result in bad behaviour even if the destroy field was set before failure to initialise was noted. Signed-off-by: Mark Rutland Reviewed-by: Will Deacon Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1313062280-19123-1-git-send-email-mark.rutland@arm.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index d4c85425e3a0..adc3ef37b7e8 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5716,6 +5716,7 @@ struct pmu *perf_init_event(struct perf_event *event) pmu = idr_find(&pmu_idr, event->attr.type); rcu_read_unlock(); if (pmu) { + event->pmu = pmu; ret = pmu->event_init(event); if (ret) pmu = ERR_PTR(ret); @@ -5723,6 +5724,7 @@ struct pmu *perf_init_event(struct perf_event *event) } list_for_each_entry_rcu(pmu, &pmus, entry) { + event->pmu = pmu; ret = pmu->event_init(event); if (!ret) goto unlock; @@ -5849,8 +5851,6 @@ done: return ERR_PTR(err); } - event->pmu = pmu; - if (!event->parent) { if (event->attach_state & PERF_ATTACH_TASK) jump_label_inc(&perf_sched_events); -- cgit v1.3-8-gc7d7 From 18e5a45db30e0e338cdd663eda05a8288cc14fa5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 3 Aug 2011 13:59:04 -0400 Subject: watchdog: Make the kthreads NUMA affine Watchdog kthreads can use kthread_create_on_node() to NUMA affine their stack and task_struct. Signed-off-by: Eric Dumazet Signed-off-by: Don Zickus Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1312394344-18815-1-git-send-email-dzickus@redhat.com Signed-off-by: Ingo Molnar --- kernel/watchdog.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 36491cd5b7d4..e952a1394d26 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -438,7 +438,7 @@ static int watchdog_enable(int cpu) /* create the watchdog thread */ if (!p) { - p = kthread_create(watchdog, (void *)(unsigned long)cpu, "watchdog/%d", cpu); + p = kthread_create_on_node(watchdog, NULL, cpu_to_node(cpu), "watchdog/%d", cpu); if (IS_ERR(p)) { printk(KERN_ERR "softlockup watchdog for %i failed\n", cpu); if (!err) { -- cgit v1.3-8-gc7d7 From e2b245f89ee3f5b03fb42d843a79a58cf4773181 Mon Sep 17 00:00:00 2001 From: "Jan H. Schönherr" Date: Mon, 1 Aug 2011 11:03:28 +0200 Subject: sched: Remove rq->avg_load_per_task MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Since commit a2d47777 ("sched: fix stale value in average load per task") the variable rq->avg_load_per_task is no longer required. Remove it. Signed-off-by: Jan H. Schönherr Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1312189408-17172-1-git-send-email-schnhrr@cs.tu-berlin.de Signed-off-by: Ingo Molnar --- kernel/sched.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..b1e8f0eca1f9 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -520,8 +520,6 @@ struct rq { int cpu; int online; - unsigned long avg_load_per_task; - u64 rt_avg; u64 age_stamp; u64 idle_stamp; @@ -1569,11 +1567,9 @@ static unsigned long cpu_avg_load_per_task(int cpu) unsigned long nr_running = ACCESS_ONCE(rq->nr_running); if (nr_running) - rq->avg_load_per_task = rq->load.weight / nr_running; - else - rq->avg_load_per_task = 0; + return rq->load.weight / nr_running; - return rq->avg_load_per_task; + return 0; } #ifdef CONFIG_PREEMPT -- cgit v1.3-8-gc7d7 From 2c2efaed9bc973e3aeab1385c618017b56c8f6d7 Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Fri, 29 Jul 2011 16:20:33 +0800 Subject: sched: Kill WAKEUP_PREEMPT Remove the WAKEUP_PREEMPT feature, disabling it doesn't make any sense and its outlived its use by a long long while. Signed-off-by: Yong Zhang Acked-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110729082033.GB12106@zhy Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 9 +-------- kernel/sched_features.h | 5 ----- 2 files changed, 1 insertion(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index bc8ee9993814..241fc86bc613 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1095,9 +1095,6 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) * narrow margin doesn't have to wait for a full slice. * This also mitigates buddy induced latencies under load. */ - if (!sched_feat(WAKEUP_PREEMPT)) - return; - if (delta_exec < sysctl_sched_min_granularity) return; @@ -1233,7 +1230,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) return; #endif - if (cfs_rq->nr_running > 1 || !sched_feat(WAKEUP_PREEMPT)) + if (cfs_rq->nr_running > 1) check_preempt_tick(cfs_rq, curr); } @@ -1899,10 +1896,6 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (unlikely(p->policy != SCHED_NORMAL)) return; - - if (!sched_feat(WAKEUP_PREEMPT)) - return; - find_matching_se(&se, &pse); update_curr(cfs_rq_of(se)); BUG_ON(!pse); diff --git a/kernel/sched_features.h b/kernel/sched_features.h index 2e74677cb040..efa0a7b75dde 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -11,11 +11,6 @@ SCHED_FEAT(GENTLE_FAIR_SLEEPERS, 1) */ SCHED_FEAT(START_DEBIT, 1) -/* - * Should wakeups try to preempt running tasks. - */ -SCHED_FEAT(WAKEUP_PREEMPT, 1) - /* * Based on load and program behaviour, see if it makes sense to place * a newly woken task on the same cpu as the task that woke it -- -- cgit v1.3-8-gc7d7 From c350a04efd1c89cd256b2abc8f07a21d0d53ff24 Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Wed, 27 Jul 2011 17:14:55 +0200 Subject: sched: fix broken SCHED_RESET_ON_FORK handling Setting child->prio = current->normal_prio _after_ SCHED_RESET_ON_FORK has been handled for an RT parent gives birth to a deranged mutant child with non-RT policy, but RT prio and sched_class. Move PI leakage protection up, always set priorities and weight, and if the child is leaving RT class, reset rt_priority to the proper value. Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1311779695.8691.2.camel@marge.simson.net Signed-off-by: Ingo Molnar --- kernel/sched.c | 25 ++++++++++++------------- 1 file changed, 12 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index b1e8f0eca1f9..cf427bb2b65e 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2843,20 +2843,24 @@ void sched_fork(struct task_struct *p) */ p->state = TASK_RUNNING; + /* + * Make sure we do not leak PI boosting priority to the child. + */ + p->prio = current->normal_prio; + /* * Revert to default priority/policy on fork if requested. */ if (unlikely(p->sched_reset_on_fork)) { - if (p->policy == SCHED_FIFO || p->policy == SCHED_RR) { + if (task_has_rt_policy(p)) { p->policy = SCHED_NORMAL; - p->normal_prio = p->static_prio; - } - - if (PRIO_TO_NICE(p->static_prio) < 0) { p->static_prio = NICE_TO_PRIO(0); - p->normal_prio = p->static_prio; - set_load_weight(p); - } + p->rt_priority = 0; + } else if (PRIO_TO_NICE(p->static_prio) < 0) + p->static_prio = NICE_TO_PRIO(0); + + p->prio = p->normal_prio = __normal_prio(p); + set_load_weight(p); /* * We don't need the reset flag anymore after the fork. It has @@ -2865,11 +2869,6 @@ void sched_fork(struct task_struct *p) p->sched_reset_on_fork = 0; } - /* - * Make sure we do not leak PI boosting priority to the child. - */ - p->prio = current->normal_prio; - if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; -- cgit v1.3-8-gc7d7 From 67d955383ab2ef8866c494c14156a4f3d29e441c Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 16 Jun 2011 21:55:18 -0400 Subject: sched: Remove noop in next_prio() When computing the next priority for a given run-queue, the check for RT priority of the task determined by the pick_next_highest_task_rt() function could be removed, since only RT tasks are returned by the function. Reviewed-by: Yong Zhang Signed-off-by: Hillf Danton Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/BANLkTimxmWiof9s5AvS3v_0X+sMiE=0x5g@mail.gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 97540f0c9e47..e2698c0fc697 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -704,7 +704,7 @@ static inline int next_prio(struct rq *rq) { struct task_struct *next = pick_next_highest_task_rt(rq, rq->cpu); - if (next && rt_prio(next->prio)) + if (next) return next->prio; else return MAX_RT_PRIO; -- cgit v1.3-8-gc7d7 From 0835471697255b415edcefd6b1e25b6c034439f2 Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 16 Jun 2011 21:55:19 -0400 Subject: sched: Remove noop in lowest_flag_domain() Checking for the validity of sd is removed, since it is already checked by the for_each_domain macro. Signed-off-by: Hillf Danton Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/BANLkTimT+Tut-3TshCDm-NiLLXrOznibNA@mail.gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 241fc86bc613..f4b732a3552b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -3660,7 +3660,7 @@ static inline struct sched_domain *lowest_flag_domain(int cpu, int flag) struct sched_domain *sd; for_each_domain(cpu, sd) - if (sd && (sd->flags & flag)) + if (sd->flags & flag) break; return sd; -- cgit v1.3-8-gc7d7 From 311e800e16f63d909136a64ed17ca353a160be59 Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 16 Jun 2011 21:55:20 -0400 Subject: sched, rt: Fix rq->rt.pushable_tasks bug in push_rt_task() Do not call dequeue_pushable_task() when failing to push an eligible task, as it remains pushable, merely not at this particular moment. Signed-off-by: Hillf Danton Signed-off-by: Mike Galbraith Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: Yong Zhang Link: http://lkml.kernel.org/r/1306895385.4791.26.camel@marge.simson.net Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index e2698c0fc697..8e189455ed12 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1394,6 +1394,7 @@ static int push_rt_task(struct rq *rq) { struct task_struct *next_task; struct rq *lowest_rq; + int ret = 0; if (!rq->rt.overloaded) return 0; @@ -1426,7 +1427,7 @@ retry: if (!lowest_rq) { struct task_struct *task; /* - * find lock_lowest_rq releases rq->lock + * find_lock_lowest_rq releases rq->lock * so it is possible that next_task has migrated. * * We need to make sure that the task is still on the same @@ -1436,12 +1437,11 @@ retry: task = pick_next_pushable_task(rq); if (task_cpu(next_task) == rq->cpu && task == next_task) { /* - * If we get here, the task hasn't moved at all, but - * it has failed to push. We will not try again, - * since the other cpus will pull from us when they - * are ready. + * The task hasn't migrated, and is still the next + * eligible task, but we failed to find a run-queue + * to push it to. Do not retry in this case, since + * other cpus will pull from us when ready. */ - dequeue_pushable_task(rq, next_task); goto out; } @@ -1460,6 +1460,7 @@ retry: deactivate_task(rq, next_task, 0); set_task_cpu(next_task, lowest_rq->cpu); activate_task(lowest_rq, next_task, 0); + ret = 1; resched_task(lowest_rq->curr); @@ -1468,7 +1469,7 @@ retry: out: put_task_struct(next_task); - return 1; + return ret; } static void push_rt_tasks(struct rq *rq) -- cgit v1.3-8-gc7d7 From 1812a643ccbfeb61a00a7f0d7219606e63d8815b Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 16 Jun 2011 21:55:21 -0400 Subject: sched: Remove resetting exec_start in put_prev_task_rt() There's no reason to clean the exec_start in put_prev_task_rt() as it is reset when the task gets back to the run queue. This saves us doing a store() in the fast path. Signed-off-by: Hillf Danton Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: Mike Galbraith Cc: Yong Zhang Link: http://lkml.kernel.org/r/BANLkTimqWD=q6YnSDi-v9y=LMWecgEzEWg@mail.gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 8e189455ed12..70107a39b1ba 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1178,7 +1178,6 @@ static struct task_struct *pick_next_task_rt(struct rq *rq) static void put_prev_task_rt(struct rq *rq, struct task_struct *p) { update_curr_rt(rq); - p->se.exec_start = 0; /* * The previous task needs to be made eligible for pushing -- cgit v1.3-8-gc7d7 From c37495fd0f64fc139b5a07d242bcb485174d1206 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 16 Jun 2011 21:55:22 -0400 Subject: sched: Balance RT tasks when forked as well When a new task is woken, the code to balance the RT task is currently skipped in the select_task_rq() call. But it will be pushed if the rq is currently overloaded with RT tasks anyway. The issue is that we already queued the task, and if it does get pushed, it will have to be dequeued and requeued on the new run queue. The advantage with pushing it first is that we avoid this requeuing as we are pushing it off before the task is ever queued. See commit 318e0893ce3f524 ("sched: pre-route RT tasks on wakeup") for more details. The return of select_task_rq() when it is not a wake up has also been changed to return task_cpu() instead of smp_processor_id(). This is more of a sanity because the current only other user of select_task_rq() besides wake ups, is an exec, where task_cpu() should also be the same as smp_processor_id(). But if it is used for other purposes, lets keep the task on the same CPU. Why would we mant to migrate it to the current CPU? Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: Hillf Danton Link: http://lkml.kernel.org/r/20110617015919.832743148@goodmis.org Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 70107a39b1ba..2153a87e6157 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1017,10 +1017,12 @@ select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) struct rq *rq; int cpu; - if (sd_flag != SD_BALANCE_WAKE) - return smp_processor_id(); - cpu = task_cpu(p); + + /* For anything but wake ups, just return the task_cpu */ + if (sd_flag != SD_BALANCE_WAKE && sd_flag != SD_BALANCE_FORK) + goto out; + rq = cpu_rq(cpu); rcu_read_lock(); @@ -1059,6 +1061,7 @@ select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) } rcu_read_unlock(); +out: return cpu; } -- cgit v1.3-8-gc7d7 From 5181f4a46afd99e5e85c639b189e43e0a42b53df Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 16 Jun 2011 21:55:23 -0400 Subject: sched: Use pushable_tasks to determine next highest prio Hillf Danton proposed a patch (see link) that cleaned up the sched_rt code that calculates the priority of the next highest priority task to be used in finding run queues to pull from. His patch removed the calculating of the next prio to just use the current prio when deteriming if we should examine a run queue to pull from. The problem with his patch was that it caused more false checks. Because we check a run queue for pushable tasks if the current priority of that run queue is higher in priority than the task about to run on our run queue. But after grabbing the locks and doing the real check, we find that there may not be a task that has a higher prio task to pull. Thus the locks were taken with nothing to do. I added some trace_printks() to record when and how many times the run queue locks were taken to check for pullable tasks, compared to how many times we pulled a task. With the current method, it was: 3806 locks taken vs 2812 pulled tasks With Hillf's patch: 6728 locks taken vs 2804 pulled tasks The number of times locks were taken to pull a task went up almost double with no more success rate. But his patch did get me thinking. When we look at the priority of the highest task to consider taking the locks to do a pull, a failure to pull can be one of the following: (in order of most likely) o RT task was pushed off already between the check and taking the lock o Waiting RT task can not be migrated o RT task's CPU affinity does not include the target run queue's CPU o RT task's priority changed between the check and taking the lock And with Hillf's patch, the thing that caused most of the failures, is the RT task to pull was not at the right priority to pull (not greater than the current RT task priority on the target run queue). Most of the above cases we can't help. But the current method does not check if the next highest prio RT task can be migrated or not, and if it can not, we still grab the locks to do the test (we don't find out about this fact until after we have the locks). I thought about this case, and realized that the pushable task plist that is maintained only holds RT tasks that can migrate. If we move the calculating of the next highest prio task from the inc/dec_rt_task() functions into the queuing of the pushable tasks, then we only measure the priorities of those tasks that we push, and we get this basically for free. Not only does this patch make the code a little more efficient, it cleans it up and makes it a little simpler. Thanks to Hillf Danton for inspiring me on this patch. Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: Hillf Danton Cc: Gregory Haskins Link: http://lkml.kernel.org/r/BANLkTimQ67180HxCx5vgMqumqw1EkFh3qg@mail.gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 61 ++++++++++++++++--------------------------------------- 1 file changed, 18 insertions(+), 43 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 2153a87e6157..a8c207ff3492 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -124,21 +124,33 @@ static void dec_rt_migration(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) update_rt_migration(rt_rq); } +static inline int has_pushable_tasks(struct rq *rq) +{ + return !plist_head_empty(&rq->rt.pushable_tasks); +} + static void enqueue_pushable_task(struct rq *rq, struct task_struct *p) { plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); plist_node_init(&p->pushable_tasks, p->prio); plist_add(&p->pushable_tasks, &rq->rt.pushable_tasks); + + /* Update the highest prio pushable task */ + if (p->prio < rq->rt.highest_prio.next) + rq->rt.highest_prio.next = p->prio; } static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) { plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); -} -static inline int has_pushable_tasks(struct rq *rq) -{ - return !plist_head_empty(&rq->rt.pushable_tasks); + /* Update the new highest prio pushable task */ + if (has_pushable_tasks(rq)) { + p = plist_first_entry(&rq->rt.pushable_tasks, + struct task_struct, pushable_tasks); + rq->rt.highest_prio.next = p->prio; + } else + rq->rt.highest_prio.next = MAX_RT_PRIO; } #else @@ -698,47 +710,13 @@ static void update_curr_rt(struct rq *rq) #if defined CONFIG_SMP -static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu); - -static inline int next_prio(struct rq *rq) -{ - struct task_struct *next = pick_next_highest_task_rt(rq, rq->cpu); - - if (next) - return next->prio; - else - return MAX_RT_PRIO; -} - static void inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); - if (prio < prev_prio) { - - /* - * If the new task is higher in priority than anything on the - * run-queue, we know that the previous high becomes our - * next-highest. - */ - rt_rq->highest_prio.next = prev_prio; - - if (rq->online) - cpupri_set(&rq->rd->cpupri, rq->cpu, prio); - - } else if (prio == rt_rq->highest_prio.curr) - /* - * If the next task is equal in priority to the highest on - * the run-queue, then we implicitly know that the next highest - * task cannot be any lower than current - */ - rt_rq->highest_prio.next = prio; - else if (prio < rt_rq->highest_prio.next) - /* - * Otherwise, we need to recompute next-highest - */ - rt_rq->highest_prio.next = next_prio(rq); + if (rq->online && prio < prev_prio) + cpupri_set(&rq->rd->cpupri, rq->cpu, prio); } static void @@ -746,9 +724,6 @@ dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); - if (rt_rq->rt_nr_running && (prio <= rt_rq->highest_prio.next)) - rt_rq->highest_prio.next = next_prio(rq); - if (rq->online && rt_rq->highest_prio.curr != prev_prio) cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio.curr); } -- cgit v1.3-8-gc7d7 From c92211d9b772792a9dea530c042efb4ab5562f50 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 2 Aug 2011 16:36:12 -0400 Subject: sched/cpupri: Remove the vec->lock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit sched/cpupri: Remove the vec->lock The cpupri vec->lock has been showing up as a top contention lately. This is because of the RT push/pull logic takes an agressive approach for migrating RT tasks. The cpupri logic is in place to improve the performance of the push/pull when dealing with large number CPU machines. The problem though is a vec->lock is required, where a vec is a global per RT priority structure. That is, if there are lots of RT tasks at the same priority, every time they are added or removed from the RT queue, this global vec->lock is taken. Now that more kernel threads are becoming RT (RCU boost and threaded interrupts) this is becoming much more of an issue. There are two variables that are being synced by the vec->lock. The cpupri bitmask, and the vec->counter. The cpupri bitmask is one bit per priority. If a RT priority vec has a process queued, then the vec->count is > 0 and the cpupri bitmask is set for that RT priority. If the cpupri bitmask gets out of sync with the vec->counter, we could end up pushing a low proirity RT task to a high priority queue. That RT task that could have run immediately could be queued on a run queue with a higher priority task indefinitely. The solution is not to use the cpupri bitmask and just look at the vec->count directly when doing a pull. The cpupri bitmask is just a fast way to scan the RT priorities when a pull is made. Instead of using the bitmask, and just examine all RT priorities, and look at the vec->counts, we could eliminate the vec->lock. The scan of RT tasks is to find a run queue that we can push an RT task to, and we do not push to a high priority queue, thus the scan only needs to go from 1 to RT task->prio, and not all 100 RT priorities. The push algorithm, which does the scan of RT priorities (and scan of the bitmask) only happens when we have an overloaded RT run queue (more than one RT task queued). The grabbing of the vec->lock happens every time any RT task is queued or dequeued on the run queue for that priority. The slowing down of the scan by not using a bitmask is negligible by the speed up of removing the vec->lock contention, and replacing it with an atomic counter and memory barrier. To prove this, I wrote a patch that times both the loop and the code that grabs the vec->locks. I passed the patches to various people (and companies) to test and show the results. I let everyone choose their own load to test, giving different loads on the system, for various different setups. Here's some of the results: (snipping to a few CPUs to not make this change log huge, but the results were consistent across the entire system). System 1 (24 CPUs) Before patch: CPU: Name Count Max Min Average Total ---- ---- ----- --- --- ------- ----- [...] cpu 20: loop 3057 1.766 0.061 0.642 1963.170 vec 6782949 90.469 0.089 0.414 2811760.503 cpu 21: loop 2617 1.723 0.062 0.641 1679.074 vec 6782810 90.499 0.089 0.291 1978499.900 cpu 22: loop 2212 1.863 0.063 0.699 1547.160 vec 6767244 85.685 0.089 0.435 2949676.898 cpu 23: loop 2320 2.013 0.062 0.594 1380.265 vec 6781694 87.923 0.088 0.431 2928538.224 After patch: cpu 20: loop 2078 1.579 0.061 0.533 1108.006 vec 6164555 5.704 0.060 0.143 885185.809 cpu 21: loop 2268 1.712 0.065 0.575 1305.248 vec 6153376 5.558 0.060 0.187 1154960.469 cpu 22: loop 1542 1.639 0.095 0.533 823.249 vec 6156510 5.720 0.060 0.190 1172727.232 cpu 23: loop 1650 1.733 0.068 0.545 900.781 vec 6170784 5.533 0.060 0.167 1034287.953 All times are in microseconds. The 'loop' is the amount of time spent doing the loop across the priorities (before patch uses bitmask). the 'vec' is the amount of time in the code that requires grabbing the vec->lock. The second patch just does not have the vec lock, but encompasses the same code. Amazingly the loop code even went down on average. The vec code went from .5 down to .18, that's more than half the time spent! Note, more than one test was run, but they all had the same results. System 2 (64 CPUs) Before patch: CPU: Name Count Max Min Average Total ---- ---- ----- --- --- ------- ----- cpu 60: loop 0 0 0 0 0 vec 5410840 277.954 0.084 0.782 4232895.727 cpu 61: loop 0 0 0 0 0 vec 4915648 188.399 0.084 0.570 2803220.301 cpu 62: loop 0 0 0 0 0 vec 5356076 276.417 0.085 0.786 4214544.548 cpu 63: loop 0 0 0 0 0 vec 4891837 170.531 0.085 0.799 3910948.833 After patch: cpu 60: loop 0 0 0 0 0 vec 5365118 5.080 0.021 0.063 340490.267 cpu 61: loop 0 0 0 0 0 vec 4898590 1.757 0.019 0.071 347903.615 cpu 62: loop 0 0 0 0 0 vec 5737130 3.067 0.021 0.119 687108.734 cpu 63: loop 0 0 0 0 0 vec 4903228 1.822 0.021 0.071 348506.477 The test run during the measurement did not have any (very few, from other CPUs) RT tasks pushing. But this shows that it helped out tremendously with the contention, as the contention happens because the vec->lock is taken only on queuing at an RT priority, and different CPUs that queue tasks at the same priority will have contention. I tested on my own 4 CPU machine with the following results: Before patch: CPU: Name Count Max Min Average Total ---- ---- ----- --- --- ------- ----- cpu 0: loop 2377 1.489 0.158 0.588 1398.395 vec 4484 770.146 2.301 4.396 19711.755 cpu 1: loop 2169 1.962 0.160 0.576 1250.110 vec 4425 152.769 2.297 4.030 17834.228 cpu 2: loop 2324 1.749 0.155 0.559 1299.799 vec 4368 779.632 2.325 4.665 20379.268 cpu 3: loop 2325 1.629 0.157 0.561 1306.113 vec 4650 408.782 2.394 4.348 20222.577 After patch: CPU: Name Count Max Min Average Total ---- ---- ----- --- --- ------- ----- cpu 0: loop 2121 1.616 0.113 0.636 1349.189 vec 4303 1.151 0.225 0.421 1811.966 cpu 1: loop 2130 1.638 0.178 0.644 1372.927 vec 4627 1.379 0.235 0.428 1983.648 cpu 2: loop 2056 1.464 0.165 0.637 1310.141 vec 4471 1.311 0.217 0.433 1937.927 cpu 3: loop 2154 1.481 0.162 0.601 1295.083 vec 4236 1.253 0.230 0.425 1803.008 This was running my migrate.c code that can be found at: http://lwn.net/Articles/425763/ The migrate code does stress the RT tasks a bit. This shows that the loop did increase a little after the patch, but not by much. The vec code dropped dramatically. From 4.3us down to .42us. That's a 10x improvement! Tested-by: Mike Galbraith Tested-by: Luis Claudio R. Gonçalves Tested-by: Matthew Hank Sabins Signed-off-by: Steven Rostedt Reviewed-by: Gregory Haskins Acked-by: Hillf Danton Signed-off-by: Peter Zijlstra Cc: Chris Mason Link: http://lkml.kernel.org/r/1312317372.18583.101.camel@gandalf.stny.rr.com Signed-off-by: Ingo Molnar --- kernel/sched_cpupri.c | 62 ++++++++++++++++++++++++++++++++------------------- kernel/sched_cpupri.h | 5 ++--- 2 files changed, 41 insertions(+), 26 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 2722dc1b4138..7761a2669fff 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -47,9 +47,6 @@ static int convert_prio(int prio) return cpupri; } -#define for_each_cpupri_active(array, idx) \ - for_each_set_bit(idx, array, CPUPRI_NR_PRIORITIES) - /** * cpupri_find - find the best (lowest-pri) CPU in the system * @cp: The cpupri context @@ -71,11 +68,33 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, int idx = 0; int task_pri = convert_prio(p->prio); - for_each_cpupri_active(cp->pri_active, idx) { + if (task_pri >= MAX_RT_PRIO) + return 0; + + for (idx = 0; idx < task_pri; idx++) { struct cpupri_vec *vec = &cp->pri_to_cpu[idx]; - if (idx >= task_pri) - break; + if (!atomic_read(&(vec)->count)) + continue; + /* + * When looking at the vector, we need to read the counter, + * do a memory barrier, then read the mask. + * + * Note: This is still all racey, but we can deal with it. + * Ideally, we only want to look at masks that are set. + * + * If a mask is not set, then the only thing wrong is that we + * did a little more work than necessary. + * + * If we read a zero count but the mask is set, because of the + * memory barriers, that can only happen when the highest prio + * task for a run queue has left the run queue, in which case, + * it will be followed by a pull. If the task we are processing + * fails to find a proper place to go, that pull request will + * pull this task if the run queue is running at a lower + * priority. + */ + smp_rmb(); if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; @@ -115,7 +134,6 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) { int *currpri = &cp->cpu_to_pri[cpu]; int oldpri = *currpri; - unsigned long flags; newpri = convert_prio(newpri); @@ -134,26 +152,25 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) if (likely(newpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[newpri]; - raw_spin_lock_irqsave(&vec->lock, flags); - cpumask_set_cpu(cpu, vec->mask); - vec->count++; - if (vec->count == 1) - set_bit(newpri, cp->pri_active); - - raw_spin_unlock_irqrestore(&vec->lock, flags); + /* + * When adding a new vector, we update the mask first, + * do a write memory barrier, and then update the count, to + * make sure the vector is visible when count is set. + */ + smp_wmb(); + atomic_inc(&(vec)->count); } if (likely(oldpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; - raw_spin_lock_irqsave(&vec->lock, flags); - - vec->count--; - if (!vec->count) - clear_bit(oldpri, cp->pri_active); + /* + * When removing from the vector, we decrement the counter first + * do a memory barrier and then clear the mask. + */ + atomic_dec(&(vec)->count); + smp_wmb(); cpumask_clear_cpu(cpu, vec->mask); - - raw_spin_unlock_irqrestore(&vec->lock, flags); } *currpri = newpri; @@ -175,8 +192,7 @@ int cpupri_init(struct cpupri *cp) for (i = 0; i < CPUPRI_NR_PRIORITIES; i++) { struct cpupri_vec *vec = &cp->pri_to_cpu[i]; - raw_spin_lock_init(&vec->lock); - vec->count = 0; + atomic_set(&vec->count, 0); if (!zalloc_cpumask_var(&vec->mask, GFP_KERNEL)) goto cleanup; } diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h index 9fc7d386fea4..6b4cd17dead6 100644 --- a/kernel/sched_cpupri.h +++ b/kernel/sched_cpupri.h @@ -12,9 +12,8 @@ /* values 2-101 are RT priorities 0-99 */ struct cpupri_vec { - raw_spinlock_t lock; - int count; - cpumask_var_t mask; + atomic_t count; + cpumask_var_t mask; }; struct cpupri { -- cgit v1.3-8-gc7d7 From d473750b4073f16f23f46f30dc1bd3de45c35754 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 5 Aug 2011 08:27:49 -0400 Subject: sched/cpupri: Fix memory barriers for vec updates to always be in order [ This patch actually compiles. Thanks to Mike Galbraith for pointing that out. I compiled and booted this patch with no issues. ] Re-examining the cpupri patch, I see there's a possible race because the update of the two priorities vec->counts are not protected by a memory barrier. When a RT runqueue is overloaded and wants to push an RT task to another runqueue, it scans the RT priority vectors in a loop from lowest priority to highest. When we queue or dequeue an RT task that changes a runqueue's highest priority task, we update the vectors to show that a runqueue is rated at a different priority. To do this, we first set the new priority mask, and increment the vec->count, and then set the old priority mask by decrementing the vec->count. If we are lowering the runqueue's RT priority rating, it will trigger a RT pull, and we do not care if we miss pushing to this runqueue or not. But if we raise the priority, but the priority is still lower than an RT task that is looking to be pushed, we must make sure that this runqueue is still seen by the push algorithm (the loop). Because the loop reads from lowest to highest, and the new priority is set before the old one is cleared, we will either see the new or old priority set and the vector will be checked. But! Since there's no memory barrier between the updates of the two, the old count may be decremented first before the new count is incremented. This means the loop may see the old count of zero and skip it, and also the new count of zero before it was updated. A possible runqueue that the RT task could move to could be missed. A conditional memory barrier is placed between the vec->count updates and is only called when both updates are done. The smp_wmb() has also been changed to smp_mb__before_atomic_inc/dec(), as they are not needed by archs that already synchronize atomic_inc/dec(). The smp_rmb() has been moved to be called at every iteration of the loop so that the race between seeing the two updates is visible by each iteration of the loop, as an arch is free to optimize the reading of memory of the counters in the loop. Signed-off-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: Nick Piggin Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1312547269.18583.194.camel@gandalf.stny.rr.com Signed-off-by: Ingo Molnar --- kernel/sched_cpupri.c | 28 +++++++++++++++++++++++++--- 1 file changed, 25 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 7761a2669fff..90faffdbdf98 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -73,9 +73,10 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, for (idx = 0; idx < task_pri; idx++) { struct cpupri_vec *vec = &cp->pri_to_cpu[idx]; + int skip = 0; if (!atomic_read(&(vec)->count)) - continue; + skip = 1; /* * When looking at the vector, we need to read the counter, * do a memory barrier, then read the mask. @@ -96,6 +97,10 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, */ smp_rmb(); + /* Need to do the rmb for every iteration */ + if (skip) + continue; + if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; @@ -134,6 +139,7 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) { int *currpri = &cp->cpu_to_pri[cpu]; int oldpri = *currpri; + int do_mb = 0; newpri = convert_prio(newpri); @@ -158,18 +164,34 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) * do a write memory barrier, and then update the count, to * make sure the vector is visible when count is set. */ - smp_wmb(); + smp_mb__before_atomic_inc(); atomic_inc(&(vec)->count); + do_mb = 1; } if (likely(oldpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[oldpri]; + /* + * Because the order of modification of the vec->count + * is important, we must make sure that the update + * of the new prio is seen before we decrement the + * old prio. This makes sure that the loop sees + * one or the other when we raise the priority of + * the run queue. We don't care about when we lower the + * priority, as that will trigger an rt pull anyway. + * + * We only need to do a memory barrier if we updated + * the new priority vec. + */ + if (do_mb) + smp_mb__after_atomic_inc(); + /* * When removing from the vector, we decrement the counter first * do a memory barrier and then clear the mask. */ atomic_dec(&(vec)->count); - smp_wmb(); + smp_mb__after_atomic_inc(); cpumask_clear_cpu(cpu, vec->mask); } -- cgit v1.3-8-gc7d7 From 5710f15b52664ae0bfa60a66d75464769d297b2b Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Sat, 6 Aug 2011 08:10:04 +0800 Subject: sched/cpupri: Remove cpupri->pri_active Since [sched/cpupri: Remove the vec->lock], member pri_active of struct cpupri is not needed any more, just remove it. Also clean stuff related to it. Signed-off-by: Yong Zhang Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110806001004.GA2207@zhy Signed-off-by: Ingo Molnar --- kernel/sched_cpupri.c | 3 +-- kernel/sched_cpupri.h | 2 -- 2 files changed, 1 insertion(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 90faffdbdf98..a86cf9d9eb11 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -152,8 +152,7 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) * If the cpu was currently mapped to a different value, we * need to map it to the new value then remove the old value. * Note, we must add the new value first, otherwise we risk the - * cpu being cleared from pri_active, and this cpu could be - * missed for a push or pull. + * cpu being missed by the priority loop in cpupri_find. */ if (likely(newpri != CPUPRI_INVALID)) { struct cpupri_vec *vec = &cp->pri_to_cpu[newpri]; diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h index 6b4cd17dead6..f6d756173491 100644 --- a/kernel/sched_cpupri.h +++ b/kernel/sched_cpupri.h @@ -4,7 +4,6 @@ #include #define CPUPRI_NR_PRIORITIES (MAX_RT_PRIO + 2) -#define CPUPRI_NR_PRI_WORDS BITS_TO_LONGS(CPUPRI_NR_PRIORITIES) #define CPUPRI_INVALID -1 #define CPUPRI_IDLE 0 @@ -18,7 +17,6 @@ struct cpupri_vec { struct cpupri { struct cpupri_vec pri_to_cpu[CPUPRI_NR_PRIORITIES]; - long pri_active[CPUPRI_NR_PRI_WORDS]; int cpu_to_pri[NR_CPUS]; }; -- cgit v1.3-8-gc7d7 From 953bfcd10e6f3697233e8e5128c611d275da39c1 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:27 -0700 Subject: sched: Implement hierarchical task accounting for SCHED_OTHER Introduce hierarchical task accounting for the group scheduling case in CFS, as well as promoting the responsibility for maintaining rq->nr_running to the scheduling classes. The primary motivation for this is that with scheduling classes supporting bandwidth throttling it is possible for entities participating in throttled sub-trees to not have root visible changes in rq->nr_running across activate and de-activate operations. This in turn leads to incorrect idle and weight-per-task load balance decisions. This also allows us to make a small fixlet to the fastpath in pick_next_task() under group scheduling. Note: this issue also exists with the existing sched_rt throttling mechanism. This patch does not address that. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184756.878333391@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 6 ++---- kernel/sched_fair.c | 6 ++++++ kernel/sched_rt.c | 5 ++++- kernel/sched_stoptask.c | 2 ++ 4 files changed, 14 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index cf427bb2b65e..cd1a531ca8ff 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -311,7 +311,7 @@ struct task_group root_task_group; /* CFS-related fields in a runqueue */ struct cfs_rq { struct load_weight load; - unsigned long nr_running; + unsigned long nr_running, h_nr_running; u64 exec_clock; u64 min_vruntime; @@ -1802,7 +1802,6 @@ static void activate_task(struct rq *rq, struct task_struct *p, int flags) rq->nr_uninterruptible--; enqueue_task(rq, p, flags); - inc_nr_running(rq); } /* @@ -1814,7 +1813,6 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int flags) rq->nr_uninterruptible++; dequeue_task(rq, p, flags); - dec_nr_running(rq); } #ifdef CONFIG_IRQ_TIME_ACCOUNTING @@ -4258,7 +4256,7 @@ pick_next_task(struct rq *rq) * Optimization: we know that if all tasks are in * the fair class we can call that function directly: */ - if (likely(rq->nr_running == rq->cfs.nr_running)) { + if (likely(rq->nr_running == rq->cfs.h_nr_running)) { p = fair_sched_class.pick_next_task(rq); if (likely(p)) return p; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index f4b732a3552b..f86b0cb5eb29 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1310,16 +1310,19 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) break; cfs_rq = cfs_rq_of(se); enqueue_entity(cfs_rq, se, flags); + cfs_rq->h_nr_running++; flags = ENQUEUE_WAKEUP; } for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); + cfs_rq->h_nr_running++; update_cfs_load(cfs_rq, 0); update_cfs_shares(cfs_rq); } + inc_nr_running(rq); hrtick_update(rq); } @@ -1339,6 +1342,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); dequeue_entity(cfs_rq, se, flags); + cfs_rq->h_nr_running--; /* Don't dequeue parent if it has other entities besides us */ if (cfs_rq->load.weight) { @@ -1358,11 +1362,13 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); + cfs_rq->h_nr_running--; update_cfs_load(cfs_rq, 0); update_cfs_shares(cfs_rq); } + dec_nr_running(rq); hrtick_update(rq); } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index a8c207ff3492..a9d3c6bc684a 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -936,6 +936,8 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags) if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); + + inc_nr_running(rq); } static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags) @@ -946,6 +948,8 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags) dequeue_rt_entity(rt_se); dequeue_pushable_task(rq, p); + + dec_nr_running(rq); } /* @@ -1841,4 +1845,3 @@ static void print_rt_stats(struct seq_file *m, int cpu) rcu_read_unlock(); } #endif /* CONFIG_SCHED_DEBUG */ - diff --git a/kernel/sched_stoptask.c b/kernel/sched_stoptask.c index 6f437632afab..8b44e7fa7fb3 100644 --- a/kernel/sched_stoptask.c +++ b/kernel/sched_stoptask.c @@ -34,11 +34,13 @@ static struct task_struct *pick_next_task_stop(struct rq *rq) static void enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags) { + inc_nr_running(rq); } static void dequeue_task_stop(struct rq *rq, struct task_struct *p, int flags) { + dec_nr_running(rq); } static void yield_task_stop(struct rq *rq) -- cgit v1.3-8-gc7d7 From ab84d31e15502fb626169ba2663381e34bf965b2 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:28 -0700 Subject: sched: Introduce primitives to account for CFS bandwidth tracking In this patch we introduce the notion of CFS bandwidth, partitioned into globally unassigned bandwidth, and locally claimed bandwidth. - The global bandwidth is per task_group, it represents a pool of unclaimed bandwidth that cfs_rqs can allocate from. - The local bandwidth is tracked per-cfs_rq, this represents allotments from the global pool bandwidth assigned to a specific cpu. Bandwidth is managed via cgroupfs, adding two new interfaces to the cpu subsystem: - cpu.cfs_period_us : the bandwidth period in usecs - cpu.cfs_quota_us : the cpu bandwidth (in usecs) that this tg will be allowed to consume over period above. Signed-off-by: Paul Turner Signed-off-by: Nikhil Rao Signed-off-by: Bharata B Rao Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184756.972636699@google.com Signed-off-by: Ingo Molnar --- init/Kconfig | 12 ++++ kernel/sched.c | 196 ++++++++++++++++++++++++++++++++++++++++++++++++++-- kernel/sched_fair.c | 16 +++++ 3 files changed, 220 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/init/Kconfig b/init/Kconfig index d62778390e55..d19b3a77ab44 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -715,6 +715,18 @@ config FAIR_GROUP_SCHED depends on CGROUP_SCHED default CGROUP_SCHED +config CFS_BANDWIDTH + bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED" + depends on EXPERIMENTAL + depends on FAIR_GROUP_SCHED + default n + help + This option allows users to define CPU bandwidth rates (limits) for + tasks running within the fair group scheduler. Groups with no limit + set are considered to be unconstrained and will run with no + restriction. + See tip/Documentation/scheduler/sched-bwc.txt for more information. + config RT_GROUP_SCHED bool "Group scheduling for SCHED_RR/FIFO" depends on EXPERIMENTAL diff --git a/kernel/sched.c b/kernel/sched.c index cd1a531ca8ff..f08cb23be96c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -247,6 +247,14 @@ struct cfs_rq; static LIST_HEAD(task_groups); +struct cfs_bandwidth { +#ifdef CONFIG_CFS_BANDWIDTH + raw_spinlock_t lock; + ktime_t period; + u64 quota; +#endif +}; + /* task group related information */ struct task_group { struct cgroup_subsys_state css; @@ -278,6 +286,8 @@ struct task_group { #ifdef CONFIG_SCHED_AUTOGROUP struct autogroup *autogroup; #endif + + struct cfs_bandwidth cfs_bandwidth; }; /* task_group_lock serializes the addition/removal of task groups */ @@ -377,9 +387,48 @@ struct cfs_rq { unsigned long load_contribution; #endif +#ifdef CONFIG_CFS_BANDWIDTH + int runtime_enabled; + s64 runtime_remaining; +#endif #endif }; +#ifdef CONFIG_FAIR_GROUP_SCHED +#ifdef CONFIG_CFS_BANDWIDTH +static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) +{ + return &tg->cfs_bandwidth; +} + +static inline u64 default_cfs_period(void); + +static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) +{ + raw_spin_lock_init(&cfs_b->lock); + cfs_b->quota = RUNTIME_INF; + cfs_b->period = ns_to_ktime(default_cfs_period()); +} + +static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) +{ + cfs_rq->runtime_enabled = 0; +} + +static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) +{} +#else +static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} +static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} +static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} + +static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) +{ + return NULL; +} +#endif /* CONFIG_CFS_BANDWIDTH */ +#endif /* CONFIG_FAIR_GROUP_SCHED */ + /* Real-Time classes' related field in a runqueue: */ struct rt_rq { struct rt_prio_array active; @@ -7971,6 +8020,7 @@ static void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, /* allow initial update_cfs_load() to truncate */ cfs_rq->load_stamp = 1; #endif + init_cfs_rq_runtime(cfs_rq); tg->cfs_rq[cpu] = cfs_rq; tg->se[cpu] = se; @@ -8110,6 +8160,7 @@ void __init sched_init(void) * We achieve this by letting root_task_group's tasks sit * directly in rq->cfs (i.e root_task_group->se[] = NULL). */ + init_cfs_bandwidth(&root_task_group.cfs_bandwidth); init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -8351,6 +8402,8 @@ static void free_fair_sched_group(struct task_group *tg) { int i; + destroy_cfs_bandwidth(tg_cfs_bandwidth(tg)); + for_each_possible_cpu(i) { if (tg->cfs_rq) kfree(tg->cfs_rq[i]); @@ -8378,6 +8431,8 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) tg->shares = NICE_0_LOAD; + init_cfs_bandwidth(tg_cfs_bandwidth(tg)); + for_each_possible_cpu(i) { cfs_rq = kzalloc_node(sizeof(struct cfs_rq), GFP_KERNEL, cpu_to_node(i)); @@ -8753,7 +8808,7 @@ static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) return walk_tg_tree(tg_schedulable, tg_nop, &data); } -static int tg_set_bandwidth(struct task_group *tg, +static int tg_set_rt_bandwidth(struct task_group *tg, u64 rt_period, u64 rt_runtime) { int i, err = 0; @@ -8792,7 +8847,7 @@ int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us) if (rt_runtime_us < 0) rt_runtime = RUNTIME_INF; - return tg_set_bandwidth(tg, rt_period, rt_runtime); + return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); } long sched_group_rt_runtime(struct task_group *tg) @@ -8817,7 +8872,7 @@ int sched_group_set_rt_period(struct task_group *tg, long rt_period_us) if (rt_period == 0) return -EINVAL; - return tg_set_bandwidth(tg, rt_period, rt_runtime); + return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); } long sched_group_rt_period(struct task_group *tg) @@ -9007,6 +9062,128 @@ static u64 cpu_shares_read_u64(struct cgroup *cgrp, struct cftype *cft) return (u64) scale_load_down(tg->shares); } + +#ifdef CONFIG_CFS_BANDWIDTH +const u64 max_cfs_quota_period = 1 * NSEC_PER_SEC; /* 1s */ +const u64 min_cfs_quota_period = 1 * NSEC_PER_MSEC; /* 1ms */ + +static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) +{ + int i; + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); + static DEFINE_MUTEX(mutex); + + if (tg == &root_task_group) + return -EINVAL; + + /* + * Ensure we have at some amount of bandwidth every period. This is + * to prevent reaching a state of large arrears when throttled via + * entity_tick() resulting in prolonged exit starvation. + */ + if (quota < min_cfs_quota_period || period < min_cfs_quota_period) + return -EINVAL; + + /* + * Likewise, bound things on the otherside by preventing insane quota + * periods. This also allows us to normalize in computing quota + * feasibility. + */ + if (period > max_cfs_quota_period) + return -EINVAL; + + mutex_lock(&mutex); + raw_spin_lock_irq(&cfs_b->lock); + cfs_b->period = ns_to_ktime(period); + cfs_b->quota = quota; + raw_spin_unlock_irq(&cfs_b->lock); + + for_each_possible_cpu(i) { + struct cfs_rq *cfs_rq = tg->cfs_rq[i]; + struct rq *rq = rq_of(cfs_rq); + + raw_spin_lock_irq(&rq->lock); + cfs_rq->runtime_enabled = quota != RUNTIME_INF; + cfs_rq->runtime_remaining = 0; + raw_spin_unlock_irq(&rq->lock); + } + mutex_unlock(&mutex); + + return 0; +} + +int tg_set_cfs_quota(struct task_group *tg, long cfs_quota_us) +{ + u64 quota, period; + + period = ktime_to_ns(tg_cfs_bandwidth(tg)->period); + if (cfs_quota_us < 0) + quota = RUNTIME_INF; + else + quota = (u64)cfs_quota_us * NSEC_PER_USEC; + + return tg_set_cfs_bandwidth(tg, period, quota); +} + +long tg_get_cfs_quota(struct task_group *tg) +{ + u64 quota_us; + + if (tg_cfs_bandwidth(tg)->quota == RUNTIME_INF) + return -1; + + quota_us = tg_cfs_bandwidth(tg)->quota; + do_div(quota_us, NSEC_PER_USEC); + + return quota_us; +} + +int tg_set_cfs_period(struct task_group *tg, long cfs_period_us) +{ + u64 quota, period; + + period = (u64)cfs_period_us * NSEC_PER_USEC; + quota = tg_cfs_bandwidth(tg)->quota; + + if (period <= 0) + return -EINVAL; + + return tg_set_cfs_bandwidth(tg, period, quota); +} + +long tg_get_cfs_period(struct task_group *tg) +{ + u64 cfs_period_us; + + cfs_period_us = ktime_to_ns(tg_cfs_bandwidth(tg)->period); + do_div(cfs_period_us, NSEC_PER_USEC); + + return cfs_period_us; +} + +static s64 cpu_cfs_quota_read_s64(struct cgroup *cgrp, struct cftype *cft) +{ + return tg_get_cfs_quota(cgroup_tg(cgrp)); +} + +static int cpu_cfs_quota_write_s64(struct cgroup *cgrp, struct cftype *cftype, + s64 cfs_quota_us) +{ + return tg_set_cfs_quota(cgroup_tg(cgrp), cfs_quota_us); +} + +static u64 cpu_cfs_period_read_u64(struct cgroup *cgrp, struct cftype *cft) +{ + return tg_get_cfs_period(cgroup_tg(cgrp)); +} + +static int cpu_cfs_period_write_u64(struct cgroup *cgrp, struct cftype *cftype, + u64 cfs_period_us) +{ + return tg_set_cfs_period(cgroup_tg(cgrp), cfs_period_us); +} + +#endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED @@ -9041,6 +9218,18 @@ static struct cftype cpu_files[] = { .write_u64 = cpu_shares_write_u64, }, #endif +#ifdef CONFIG_CFS_BANDWIDTH + { + .name = "cfs_quota_us", + .read_s64 = cpu_cfs_quota_read_s64, + .write_s64 = cpu_cfs_quota_write_s64, + }, + { + .name = "cfs_period_us", + .read_u64 = cpu_cfs_period_read_u64, + .write_u64 = cpu_cfs_period_write_u64, + }, +#endif #ifdef CONFIG_RT_GROUP_SCHED { .name = "rt_runtime_us", @@ -9350,4 +9539,3 @@ struct cgroup_subsys cpuacct_subsys = { .subsys_id = cpuacct_subsys_id, }; #endif /* CONFIG_CGROUP_CPUACCT */ - diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index f86b0cb5eb29..f24f4171019d 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1234,6 +1234,22 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) check_preempt_tick(cfs_rq, curr); } + +/************************************************** + * CFS bandwidth control machinery + */ + +#ifdef CONFIG_CFS_BANDWIDTH +/* + * default period for cfs group bandwidth. + * default: 0.1s, units: nanoseconds + */ +static inline u64 default_cfs_period(void) +{ + return 100000000ULL; +} +#endif + /************************************************** * CFS operations on tasks: */ -- cgit v1.3-8-gc7d7 From a790de99599a29ad3f18667530cf4b9f4b7e3234 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:29 -0700 Subject: sched: Validate CFS quota hierarchies Add constraints validation for CFS bandwidth hierarchies. Validate that: max(child bandwidth) <= parent_bandwidth In a quota limited hierarchy, an unconstrained entity (e.g. bandwidth==RUNTIME_INF) inherits the bandwidth of its parent. This constraint is chosen over sum(child_bandwidth) as notion of over-commit is valuable within SCHED_OTHER. Some basic code from the RT case is re-factored for reuse. Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.083774572@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 112 +++++++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 98 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index f08cb23be96c..ea6850d93b2a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -252,6 +252,7 @@ struct cfs_bandwidth { raw_spinlock_t lock; ktime_t period; u64 quota; + s64 hierarchal_quota; #endif }; @@ -1518,7 +1519,8 @@ static inline void dec_cpu_load(struct rq *rq, unsigned long load) update_load_sub(&rq->load, load); } -#if (defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)) || defined(CONFIG_RT_GROUP_SCHED) +#if defined(CONFIG_RT_GROUP_SCHED) || (defined(CONFIG_FAIR_GROUP_SCHED) && \ + (defined(CONFIG_SMP) || defined(CONFIG_CFS_BANDWIDTH))) typedef int (*tg_visitor)(struct task_group *, void *); /* @@ -8708,12 +8710,7 @@ unsigned long sched_group_shares(struct task_group *tg) } #endif -#ifdef CONFIG_RT_GROUP_SCHED -/* - * Ensure that the real time constraints are schedulable. - */ -static DEFINE_MUTEX(rt_constraints_mutex); - +#if defined(CONFIG_RT_GROUP_SCHED) || defined(CONFIG_CFS_BANDWIDTH) static unsigned long to_ratio(u64 period, u64 runtime) { if (runtime == RUNTIME_INF) @@ -8721,6 +8718,13 @@ static unsigned long to_ratio(u64 period, u64 runtime) return div64_u64(runtime << 20, period); } +#endif + +#ifdef CONFIG_RT_GROUP_SCHED +/* + * Ensure that the real time constraints are schedulable. + */ +static DEFINE_MUTEX(rt_constraints_mutex); /* Must be called with tasklist_lock held */ static inline int tg_has_rt_tasks(struct task_group *tg) @@ -8741,7 +8745,7 @@ struct rt_schedulable_data { u64 rt_runtime; }; -static int tg_schedulable(struct task_group *tg, void *data) +static int tg_rt_schedulable(struct task_group *tg, void *data) { struct rt_schedulable_data *d = data; struct task_group *child; @@ -8805,7 +8809,7 @@ static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) .rt_runtime = runtime, }; - return walk_tg_tree(tg_schedulable, tg_nop, &data); + return walk_tg_tree(tg_rt_schedulable, tg_nop, &data); } static int tg_set_rt_bandwidth(struct task_group *tg, @@ -9064,14 +9068,17 @@ static u64 cpu_shares_read_u64(struct cgroup *cgrp, struct cftype *cft) } #ifdef CONFIG_CFS_BANDWIDTH +static DEFINE_MUTEX(cfs_constraints_mutex); + const u64 max_cfs_quota_period = 1 * NSEC_PER_SEC; /* 1s */ const u64 min_cfs_quota_period = 1 * NSEC_PER_MSEC; /* 1ms */ +static int __cfs_schedulable(struct task_group *tg, u64 period, u64 runtime); + static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) { - int i; + int i, ret = 0; struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); - static DEFINE_MUTEX(mutex); if (tg == &root_task_group) return -EINVAL; @@ -9092,7 +9099,11 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) if (period > max_cfs_quota_period) return -EINVAL; - mutex_lock(&mutex); + mutex_lock(&cfs_constraints_mutex); + ret = __cfs_schedulable(tg, period, quota); + if (ret) + goto out_unlock; + raw_spin_lock_irq(&cfs_b->lock); cfs_b->period = ns_to_ktime(period); cfs_b->quota = quota; @@ -9107,9 +9118,10 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) cfs_rq->runtime_remaining = 0; raw_spin_unlock_irq(&rq->lock); } - mutex_unlock(&mutex); +out_unlock: + mutex_unlock(&cfs_constraints_mutex); - return 0; + return ret; } int tg_set_cfs_quota(struct task_group *tg, long cfs_quota_us) @@ -9183,6 +9195,78 @@ static int cpu_cfs_period_write_u64(struct cgroup *cgrp, struct cftype *cftype, return tg_set_cfs_period(cgroup_tg(cgrp), cfs_period_us); } +struct cfs_schedulable_data { + struct task_group *tg; + u64 period, quota; +}; + +/* + * normalize group quota/period to be quota/max_period + * note: units are usecs + */ +static u64 normalize_cfs_quota(struct task_group *tg, + struct cfs_schedulable_data *d) +{ + u64 quota, period; + + if (tg == d->tg) { + period = d->period; + quota = d->quota; + } else { + period = tg_get_cfs_period(tg); + quota = tg_get_cfs_quota(tg); + } + + /* note: these should typically be equivalent */ + if (quota == RUNTIME_INF || quota == -1) + return RUNTIME_INF; + + return to_ratio(period, quota); +} + +static int tg_cfs_schedulable_down(struct task_group *tg, void *data) +{ + struct cfs_schedulable_data *d = data; + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); + s64 quota = 0, parent_quota = -1; + + if (!tg->parent) { + quota = RUNTIME_INF; + } else { + struct cfs_bandwidth *parent_b = tg_cfs_bandwidth(tg->parent); + + quota = normalize_cfs_quota(tg, d); + parent_quota = parent_b->hierarchal_quota; + + /* + * ensure max(child_quota) <= parent_quota, inherit when no + * limit is set + */ + if (quota == RUNTIME_INF) + quota = parent_quota; + else if (parent_quota != RUNTIME_INF && quota > parent_quota) + return -EINVAL; + } + cfs_b->hierarchal_quota = quota; + + return 0; +} + +static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) +{ + struct cfs_schedulable_data data = { + .tg = tg, + .period = period, + .quota = quota, + }; + + if (quota != RUNTIME_INF) { + do_div(data.period, NSEC_PER_USEC); + do_div(data.quota, NSEC_PER_USEC); + } + + return walk_tg_tree(tg_cfs_schedulable_down, tg_nop, &data); +} #endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ -- cgit v1.3-8-gc7d7 From ec12cb7f31e28854efae7dd6f9544e0a66379040 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:30 -0700 Subject: sched: Accumulate per-cfs_rq cpu usage and charge against bandwidth Account bandwidth usage on the cfs_rq level versus the task_groups to which they belong. Whether we are tracking bandwidth on a given cfs_rq is maintained under cfs_rq->runtime_enabled. cfs_rq's which belong to a bandwidth constrained task_group have their runtime accounted via the update_curr() path, which withdraws bandwidth from the global pool as desired. Updates involving the global pool are currently protected under cfs_bandwidth->lock, local runtime is protected by rq->lock. This patch only assigns and tracks quota, no action is taken in the case that cfs_rq->runtime_used exceeds cfs_rq->runtime_assigned. Signed-off-by: Paul Turner Signed-off-by: Nikhil Rao Signed-off-by: Bharata B Rao Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.179386821@google.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 4 +++ kernel/sched.c | 4 ++- kernel/sched_fair.c | 79 +++++++++++++++++++++++++++++++++++++++++++++++++-- kernel/sysctl.c | 10 +++++++ 4 files changed, 94 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4ac2c0578e0f..bc6f5f2e24fa 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2040,6 +2040,10 @@ static inline void sched_autogroup_fork(struct signal_struct *sig) { } static inline void sched_autogroup_exit(struct signal_struct *sig) { } #endif +#ifdef CONFIG_CFS_BANDWIDTH +extern unsigned int sysctl_sched_cfs_bandwidth_slice; +#endif + #ifdef CONFIG_RT_MUTEXES extern int rt_mutex_getprio(struct task_struct *p); extern void rt_mutex_setprio(struct task_struct *p, int prio); diff --git a/kernel/sched.c b/kernel/sched.c index ea6850d93b2a..35561c63a490 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -251,7 +251,7 @@ struct cfs_bandwidth { #ifdef CONFIG_CFS_BANDWIDTH raw_spinlock_t lock; ktime_t period; - u64 quota; + u64 quota, runtime; s64 hierarchal_quota; #endif }; @@ -407,6 +407,7 @@ static inline u64 default_cfs_period(void); static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { raw_spin_lock_init(&cfs_b->lock); + cfs_b->runtime = 0; cfs_b->quota = RUNTIME_INF; cfs_b->period = ns_to_ktime(default_cfs_period()); } @@ -9107,6 +9108,7 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) raw_spin_lock_irq(&cfs_b->lock); cfs_b->period = ns_to_ktime(period); cfs_b->quota = quota; + cfs_b->runtime = quota; raw_spin_unlock_irq(&cfs_b->lock); for_each_possible_cpu(i) { diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index f24f4171019d..9502aa899f73 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -89,6 +89,20 @@ const_debug unsigned int sysctl_sched_migration_cost = 500000UL; */ unsigned int __read_mostly sysctl_sched_shares_window = 10000000UL; +#ifdef CONFIG_CFS_BANDWIDTH +/* + * Amount of runtime to allocate from global (tg) to local (per-cfs_rq) pool + * each time a cfs_rq requests quota. + * + * Note: in the case that the slice exceeds the runtime remaining (either due + * to consumption or the quota being specified to be smaller than the slice) + * we will always only issue the remaining available time. + * + * default: 5 msec, units: microseconds + */ +unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL; +#endif + static const struct sched_class fair_sched_class; /************************************************************** @@ -292,6 +306,8 @@ find_matching_se(struct sched_entity **se, struct sched_entity **pse) #endif /* CONFIG_FAIR_GROUP_SCHED */ +static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, + unsigned long delta_exec); /************************************************************** * Scheduling class tree data structure manipulation methods: @@ -583,6 +599,8 @@ static void update_curr(struct cfs_rq *cfs_rq) cpuacct_charge(curtask, delta_exec); account_group_exec_runtime(curtask, delta_exec); } + + account_cfs_rq_runtime(cfs_rq, delta_exec); } static inline void @@ -1248,6 +1266,58 @@ static inline u64 default_cfs_period(void) { return 100000000ULL; } + +static inline u64 sched_cfs_bandwidth_slice(void) +{ + return (u64)sysctl_sched_cfs_bandwidth_slice * NSEC_PER_USEC; +} + +static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) +{ + struct task_group *tg = cfs_rq->tg; + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); + u64 amount = 0, min_amount; + + /* note: this is a positive sum as runtime_remaining <= 0 */ + min_amount = sched_cfs_bandwidth_slice() - cfs_rq->runtime_remaining; + + raw_spin_lock(&cfs_b->lock); + if (cfs_b->quota == RUNTIME_INF) + amount = min_amount; + else if (cfs_b->runtime > 0) { + amount = min(cfs_b->runtime, min_amount); + cfs_b->runtime -= amount; + } + raw_spin_unlock(&cfs_b->lock); + + cfs_rq->runtime_remaining += amount; +} + +static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, + unsigned long delta_exec) +{ + if (!cfs_rq->runtime_enabled) + return; + + cfs_rq->runtime_remaining -= delta_exec; + if (cfs_rq->runtime_remaining > 0) + return; + + assign_cfs_rq_runtime(cfs_rq); +} + +static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, + unsigned long delta_exec) +{ + if (!cfs_rq->runtime_enabled) + return; + + __account_cfs_rq_runtime(cfs_rq, delta_exec); +} + +#else +static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, + unsigned long delta_exec) {} #endif /************************************************** @@ -4266,8 +4336,13 @@ static void set_curr_task_fair(struct rq *rq) { struct sched_entity *se = &rq->curr->se; - for_each_sched_entity(se) - set_next_entity(cfs_rq_of(se), se); + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + set_next_entity(cfs_rq, se); + /* ensure bandwidth has been allocated on our new cfs_rq */ + account_cfs_rq_runtime(cfs_rq, 0); + } } #ifdef CONFIG_FAIR_GROUP_SCHED diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 11d65b531e50..2d2ecdcc8cdb 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -379,6 +379,16 @@ static struct ctl_table kern_table[] = { .extra2 = &one, }, #endif +#ifdef CONFIG_CFS_BANDWIDTH + { + .procname = "sched_cfs_bandwidth_slice_us", + .data = &sysctl_sched_cfs_bandwidth_slice, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &one, + }, +#endif #ifdef CONFIG_PROVE_LOCKING { .procname = "prove_locking", -- cgit v1.3-8-gc7d7 From 58088ad0152ba4b7997388c93d0ca208ec1ece75 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:31 -0700 Subject: sched: Add a timer to handle CFS bandwidth refresh This patch adds a per-task_group timer which handles the refresh of the global CFS bandwidth pool. Since the RT pool is using a similar timer there's some small refactoring to share this support. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.277271273@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 107 +++++++++++++++++++++++++++++++++++++++++----------- kernel/sched_fair.c | 40 ++++++++++++++++++-- 2 files changed, 123 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 35561c63a490..34bf8e6db9af 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -196,10 +196,28 @@ static inline int rt_bandwidth_enabled(void) return sysctl_sched_rt_runtime >= 0; } -static void start_rt_bandwidth(struct rt_bandwidth *rt_b) +static void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period) { - ktime_t now; + unsigned long delta; + ktime_t soft, hard, now; + for (;;) { + if (hrtimer_active(period_timer)) + break; + + now = hrtimer_cb_get_time(period_timer); + hrtimer_forward(period_timer, now, period); + + soft = hrtimer_get_softexpires(period_timer); + hard = hrtimer_get_expires(period_timer); + delta = ktime_to_ns(ktime_sub(hard, soft)); + __hrtimer_start_range_ns(period_timer, soft, delta, + HRTIMER_MODE_ABS_PINNED, 0); + } +} + +static void start_rt_bandwidth(struct rt_bandwidth *rt_b) +{ if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) return; @@ -207,22 +225,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) return; raw_spin_lock(&rt_b->rt_runtime_lock); - for (;;) { - unsigned long delta; - ktime_t soft, hard; - - if (hrtimer_active(&rt_b->rt_period_timer)) - break; - - now = hrtimer_cb_get_time(&rt_b->rt_period_timer); - hrtimer_forward(&rt_b->rt_period_timer, now, rt_b->rt_period); - - soft = hrtimer_get_softexpires(&rt_b->rt_period_timer); - hard = hrtimer_get_expires(&rt_b->rt_period_timer); - delta = ktime_to_ns(ktime_sub(hard, soft)); - __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta, - HRTIMER_MODE_ABS_PINNED, 0); - } + start_bandwidth_timer(&rt_b->rt_period_timer, rt_b->rt_period); raw_spin_unlock(&rt_b->rt_runtime_lock); } @@ -253,6 +256,9 @@ struct cfs_bandwidth { ktime_t period; u64 quota, runtime; s64 hierarchal_quota; + + int idle, timer_active; + struct hrtimer period_timer; #endif }; @@ -403,6 +409,28 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) } static inline u64 default_cfs_period(void); +static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun); + +static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) +{ + struct cfs_bandwidth *cfs_b = + container_of(timer, struct cfs_bandwidth, period_timer); + ktime_t now; + int overrun; + int idle = 0; + + for (;;) { + now = hrtimer_cb_get_time(timer); + overrun = hrtimer_forward(timer, now, cfs_b->period); + + if (!overrun) + break; + + idle = do_sched_cfs_period_timer(cfs_b, overrun); + } + + return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; +} static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { @@ -410,6 +438,9 @@ static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) cfs_b->runtime = 0; cfs_b->quota = RUNTIME_INF; cfs_b->period = ns_to_ktime(default_cfs_period()); + + hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + cfs_b->period_timer.function = sched_cfs_period_timer; } static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) @@ -417,8 +448,34 @@ static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) cfs_rq->runtime_enabled = 0; } +/* requires cfs_b->lock, may release to reprogram timer */ +static void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) +{ + /* + * The timer may be active because we're trying to set a new bandwidth + * period or because we're racing with the tear-down path + * (timer_active==0 becomes visible before the hrtimer call-back + * terminates). In either case we ensure that it's re-programmed + */ + while (unlikely(hrtimer_active(&cfs_b->period_timer))) { + raw_spin_unlock(&cfs_b->lock); + /* ensure cfs_b->lock is available while we wait */ + hrtimer_cancel(&cfs_b->period_timer); + + raw_spin_lock(&cfs_b->lock); + /* if someone else restarted the timer then we're done */ + if (cfs_b->timer_active) + return; + } + + cfs_b->timer_active = 1; + start_bandwidth_timer(&cfs_b->period_timer, cfs_b->period); +} + static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) -{} +{ + hrtimer_cancel(&cfs_b->period_timer); +} #else static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} @@ -9078,7 +9135,7 @@ static int __cfs_schedulable(struct task_group *tg, u64 period, u64 runtime); static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) { - int i, ret = 0; + int i, ret = 0, runtime_enabled; struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); if (tg == &root_task_group) @@ -9105,10 +9162,18 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) if (ret) goto out_unlock; + runtime_enabled = quota != RUNTIME_INF; raw_spin_lock_irq(&cfs_b->lock); cfs_b->period = ns_to_ktime(period); cfs_b->quota = quota; cfs_b->runtime = quota; + + /* restart the period timer (if active) to handle new period expiry */ + if (runtime_enabled && cfs_b->timer_active) { + /* force a reprogram */ + cfs_b->timer_active = 0; + __start_cfs_bandwidth(cfs_b); + } raw_spin_unlock_irq(&cfs_b->lock); for_each_possible_cpu(i) { @@ -9116,7 +9181,7 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) struct rq *rq = rq_of(cfs_rq); raw_spin_lock_irq(&rq->lock); - cfs_rq->runtime_enabled = quota != RUNTIME_INF; + cfs_rq->runtime_enabled = runtime_enabled; cfs_rq->runtime_remaining = 0; raw_spin_unlock_irq(&rq->lock); } diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 9502aa899f73..af73a8a85eef 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1284,9 +1284,16 @@ static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) raw_spin_lock(&cfs_b->lock); if (cfs_b->quota == RUNTIME_INF) amount = min_amount; - else if (cfs_b->runtime > 0) { - amount = min(cfs_b->runtime, min_amount); - cfs_b->runtime -= amount; + else { + /* ensure bandwidth timer remains active under consumption */ + if (!cfs_b->timer_active) + __start_cfs_bandwidth(cfs_b); + + if (cfs_b->runtime > 0) { + amount = min(cfs_b->runtime, min_amount); + cfs_b->runtime -= amount; + cfs_b->idle = 0; + } } raw_spin_unlock(&cfs_b->lock); @@ -1315,6 +1322,33 @@ static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, __account_cfs_rq_runtime(cfs_rq, delta_exec); } +/* + * Responsible for refilling a task_group's bandwidth and unthrottling its + * cfs_rqs as appropriate. If there has been no activity within the last + * period the timer is deactivated until scheduling resumes; cfs_b->idle is + * used to track this state. + */ +static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) +{ + int idle = 1; + + raw_spin_lock(&cfs_b->lock); + /* no need to continue the timer with no bandwidth constraint */ + if (cfs_b->quota == RUNTIME_INF) + goto out_unlock; + + idle = cfs_b->idle; + cfs_b->runtime = cfs_b->quota; + + /* mark as potentially idle for the upcoming period */ + cfs_b->idle = 1; +out_unlock: + if (idle) + cfs_b->timer_active = 0; + raw_spin_unlock(&cfs_b->lock); + + return idle; +} #else static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) {} -- cgit v1.3-8-gc7d7 From a9cf55b2861057a213e610da2fec52125439a11d Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:32 -0700 Subject: sched: Expire invalid runtime Since quota is managed using a global state but consumed on a per-cpu basis we need to ensure that our per-cpu state is appropriately synchronized. Most importantly, runtime that is state (from a previous period) should not be locally consumable. We take advantage of existing sched_clock synchronization about the jiffy to efficiently detect whether we have (globally) crossed a quota boundary above. One catch is that the direction of spread on sched_clock is undefined, specifically, we don't know whether our local clock is behind or ahead of the one responsible for the current expiration time. Fortunately we can differentiate these by considering whether the global deadline has advanced. If it has not, then we assume our clock to be "fast" and advance our local expiration; otherwise, we know the deadline has truly passed and we expire our local runtime. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.379275352@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 4 ++- kernel/sched_fair.c | 90 +++++++++++++++++++++++++++++++++++++++++++++++------ 2 files changed, 84 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 34bf8e6db9af..a2d55144bd9c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -256,6 +256,7 @@ struct cfs_bandwidth { ktime_t period; u64 quota, runtime; s64 hierarchal_quota; + u64 runtime_expires; int idle, timer_active; struct hrtimer period_timer; @@ -396,6 +397,7 @@ struct cfs_rq { #endif #ifdef CONFIG_CFS_BANDWIDTH int runtime_enabled; + u64 runtime_expires; s64 runtime_remaining; #endif #endif @@ -9166,8 +9168,8 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) raw_spin_lock_irq(&cfs_b->lock); cfs_b->period = ns_to_ktime(period); cfs_b->quota = quota; - cfs_b->runtime = quota; + __refill_cfs_bandwidth_runtime(cfs_b); /* restart the period timer (if active) to handle new period expiry */ if (runtime_enabled && cfs_b->timer_active) { /* force a reprogram */ diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index af73a8a85eef..9d1adbd0b615 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1272,11 +1272,30 @@ static inline u64 sched_cfs_bandwidth_slice(void) return (u64)sysctl_sched_cfs_bandwidth_slice * NSEC_PER_USEC; } +/* + * Replenish runtime according to assigned quota and update expiration time. + * We use sched_clock_cpu directly instead of rq->clock to avoid adding + * additional synchronization around rq->lock. + * + * requires cfs_b->lock + */ +static void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b) +{ + u64 now; + + if (cfs_b->quota == RUNTIME_INF) + return; + + now = sched_clock_cpu(smp_processor_id()); + cfs_b->runtime = cfs_b->quota; + cfs_b->runtime_expires = now + ktime_to_ns(cfs_b->period); +} + static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) { struct task_group *tg = cfs_rq->tg; struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); - u64 amount = 0, min_amount; + u64 amount = 0, min_amount, expires; /* note: this is a positive sum as runtime_remaining <= 0 */ min_amount = sched_cfs_bandwidth_slice() - cfs_rq->runtime_remaining; @@ -1285,9 +1304,16 @@ static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) if (cfs_b->quota == RUNTIME_INF) amount = min_amount; else { - /* ensure bandwidth timer remains active under consumption */ - if (!cfs_b->timer_active) + /* + * If the bandwidth pool has become inactive, then at least one + * period must have elapsed since the last consumption. + * Refresh the global state and ensure bandwidth timer becomes + * active. + */ + if (!cfs_b->timer_active) { + __refill_cfs_bandwidth_runtime(cfs_b); __start_cfs_bandwidth(cfs_b); + } if (cfs_b->runtime > 0) { amount = min(cfs_b->runtime, min_amount); @@ -1295,19 +1321,61 @@ static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) cfs_b->idle = 0; } } + expires = cfs_b->runtime_expires; raw_spin_unlock(&cfs_b->lock); cfs_rq->runtime_remaining += amount; + /* + * we may have advanced our local expiration to account for allowed + * spread between our sched_clock and the one on which runtime was + * issued. + */ + if ((s64)(expires - cfs_rq->runtime_expires) > 0) + cfs_rq->runtime_expires = expires; } -static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, - unsigned long delta_exec) +/* + * Note: This depends on the synchronization provided by sched_clock and the + * fact that rq->clock snapshots this value. + */ +static void expire_cfs_rq_runtime(struct cfs_rq *cfs_rq) { - if (!cfs_rq->runtime_enabled) + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + struct rq *rq = rq_of(cfs_rq); + + /* if the deadline is ahead of our clock, nothing to do */ + if (likely((s64)(rq->clock - cfs_rq->runtime_expires) < 0)) + return; + + if (cfs_rq->runtime_remaining < 0) return; + /* + * If the local deadline has passed we have to consider the + * possibility that our sched_clock is 'fast' and the global deadline + * has not truly expired. + * + * Fortunately we can check determine whether this the case by checking + * whether the global deadline has advanced. + */ + + if ((s64)(cfs_rq->runtime_expires - cfs_b->runtime_expires) >= 0) { + /* extend local deadline, drift is bounded above by 2 ticks */ + cfs_rq->runtime_expires += TICK_NSEC; + } else { + /* global deadline is ahead, expiration has passed */ + cfs_rq->runtime_remaining = 0; + } +} + +static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, + unsigned long delta_exec) +{ + /* dock delta_exec before expiring quota (as it could span periods) */ cfs_rq->runtime_remaining -= delta_exec; - if (cfs_rq->runtime_remaining > 0) + expire_cfs_rq_runtime(cfs_rq); + + if (likely(cfs_rq->runtime_remaining > 0)) return; assign_cfs_rq_runtime(cfs_rq); @@ -1338,7 +1406,12 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) goto out_unlock; idle = cfs_b->idle; - cfs_b->runtime = cfs_b->quota; + /* if we're going inactive then everything else can be deferred */ + if (idle) + goto out_unlock; + + __refill_cfs_bandwidth_runtime(cfs_b); + /* mark as potentially idle for the upcoming period */ cfs_b->idle = 1; @@ -1557,7 +1630,6 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg) return wl; } - #else static inline unsigned long effective_load(struct task_group *tg, int cpu, -- cgit v1.3-8-gc7d7 From 85dac906bec3bb41bfaa7ccaa65c4706de5cfdf8 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:33 -0700 Subject: sched: Add support for throttling group entities Now that consumption is tracked (via update_curr()) we add support to throttle group entities (and their corresponding cfs_rqs) in the case where this is no run-time remaining. Throttled entities are dequeued to prevent scheduling, additionally we mark them as throttled (using cfs_rq->throttled) to prevent them from becoming re-enqueued until they are unthrottled. A list of a task_group's throttled entities are maintained on the cfs_bandwidth structure. Note: While the machinery for throttling is added in this patch the act of throttling an entity exceeding its bandwidth is deferred until later within the series. Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.480608533@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 7 +++++ kernel/sched_fair.c | 89 ++++++++++++++++++++++++++++++++++++++++++++++++++--- 2 files changed, 92 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index a2d55144bd9c..044260a9418d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -260,6 +260,8 @@ struct cfs_bandwidth { int idle, timer_active; struct hrtimer period_timer; + struct list_head throttled_cfs_rq; + #endif }; @@ -399,6 +401,9 @@ struct cfs_rq { int runtime_enabled; u64 runtime_expires; s64 runtime_remaining; + + int throttled; + struct list_head throttled_list; #endif #endif }; @@ -441,6 +446,7 @@ static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) cfs_b->quota = RUNTIME_INF; cfs_b->period = ns_to_ktime(default_cfs_period()); + INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); cfs_b->period_timer.function = sched_cfs_period_timer; } @@ -448,6 +454,7 @@ static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) { cfs_rq->runtime_enabled = 0; + INIT_LIST_HEAD(&cfs_rq->throttled_list); } /* requires cfs_b->lock, may release to reprogram timer */ diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 9d1adbd0b615..72c9d4ed5991 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1291,7 +1291,8 @@ static void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b) cfs_b->runtime_expires = now + ktime_to_ns(cfs_b->period); } -static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) +/* returns 0 on failure to allocate runtime */ +static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) { struct task_group *tg = cfs_rq->tg; struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); @@ -1332,6 +1333,8 @@ static void assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) */ if ((s64)(expires - cfs_rq->runtime_expires) > 0) cfs_rq->runtime_expires = expires; + + return cfs_rq->runtime_remaining > 0; } /* @@ -1378,7 +1381,12 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, if (likely(cfs_rq->runtime_remaining > 0)) return; - assign_cfs_rq_runtime(cfs_rq); + /* + * if we're unable to extend our runtime we resched so that the active + * hierarchy can be throttled + */ + if (!assign_cfs_rq_runtime(cfs_rq) && likely(cfs_rq->curr)) + resched_task(rq_of(cfs_rq)->curr); } static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, @@ -1390,6 +1398,47 @@ static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, __account_cfs_rq_runtime(cfs_rq, delta_exec); } +static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) +{ + return cfs_rq->throttled; +} + +static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) +{ + struct rq *rq = rq_of(cfs_rq); + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + struct sched_entity *se; + long task_delta, dequeue = 1; + + se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; + + /* account load preceding throttle */ + update_cfs_load(cfs_rq, 0); + + task_delta = cfs_rq->h_nr_running; + for_each_sched_entity(se) { + struct cfs_rq *qcfs_rq = cfs_rq_of(se); + /* throttled entity or throttle-on-deactivate */ + if (!se->on_rq) + break; + + if (dequeue) + dequeue_entity(qcfs_rq, se, DEQUEUE_SLEEP); + qcfs_rq->h_nr_running -= task_delta; + + if (qcfs_rq->load.weight) + dequeue = 0; + } + + if (!se) + rq->nr_running -= task_delta; + + cfs_rq->throttled = 1; + raw_spin_lock(&cfs_b->lock); + list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); + raw_spin_unlock(&cfs_b->lock); +} + /* * Responsible for refilling a task_group's bandwidth and unthrottling its * cfs_rqs as appropriate. If there has been no activity within the last @@ -1425,6 +1474,11 @@ out_unlock: #else static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) {} + +static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) +{ + return 0; +} #endif /************************************************** @@ -1503,7 +1557,17 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) break; cfs_rq = cfs_rq_of(se); enqueue_entity(cfs_rq, se, flags); + + /* + * end evaluation on encountering a throttled cfs_rq + * + * note: in the case of encountering a throttled cfs_rq we will + * post the final h_nr_running increment below. + */ + if (cfs_rq_throttled(cfs_rq)) + break; cfs_rq->h_nr_running++; + flags = ENQUEUE_WAKEUP; } @@ -1511,11 +1575,15 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) cfs_rq = cfs_rq_of(se); cfs_rq->h_nr_running++; + if (cfs_rq_throttled(cfs_rq)) + break; + update_cfs_load(cfs_rq, 0); update_cfs_shares(cfs_rq); } - inc_nr_running(rq); + if (!se) + inc_nr_running(rq); hrtick_update(rq); } @@ -1535,6 +1603,15 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); dequeue_entity(cfs_rq, se, flags); + + /* + * end evaluation on encountering a throttled cfs_rq + * + * note: in the case of encountering a throttled cfs_rq we will + * post the final h_nr_running decrement below. + */ + if (cfs_rq_throttled(cfs_rq)) + break; cfs_rq->h_nr_running--; /* Don't dequeue parent if it has other entities besides us */ @@ -1557,11 +1634,15 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) cfs_rq = cfs_rq_of(se); cfs_rq->h_nr_running--; + if (cfs_rq_throttled(cfs_rq)) + break; + update_cfs_load(cfs_rq, 0); update_cfs_shares(cfs_rq); } - dec_nr_running(rq); + if (!se) + dec_nr_running(rq); hrtick_update(rq); } -- cgit v1.3-8-gc7d7 From 671fd9dabe5239ad218c7eb48b2b9edee50250e6 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:34 -0700 Subject: sched: Add support for unthrottling group entities At the start of each period we refresh the global bandwidth pool. At this time we must also unthrottle any cfs_rq entities who are now within bandwidth once more (as quota permits). Unthrottled entities have their corresponding cfs_rq->throttled flag cleared and their entities re-enqueued. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.574628950@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 3 ++ kernel/sched_fair.c | 127 ++++++++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 126 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 044260a9418d..4bbabc2c4a77 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -9192,6 +9192,9 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) raw_spin_lock_irq(&rq->lock); cfs_rq->runtime_enabled = runtime_enabled; cfs_rq->runtime_remaining = 0; + + if (cfs_rq_throttled(cfs_rq)) + unthrottle_cfs_rq(cfs_rq); raw_spin_unlock_irq(&rq->lock); } out_unlock: diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 72c9d4ed5991..76411950ff3b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1439,6 +1439,84 @@ static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) raw_spin_unlock(&cfs_b->lock); } +static void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) +{ + struct rq *rq = rq_of(cfs_rq); + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + struct sched_entity *se; + int enqueue = 1; + long task_delta; + + se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; + + cfs_rq->throttled = 0; + raw_spin_lock(&cfs_b->lock); + list_del_rcu(&cfs_rq->throttled_list); + raw_spin_unlock(&cfs_b->lock); + + if (!cfs_rq->load.weight) + return; + + task_delta = cfs_rq->h_nr_running; + for_each_sched_entity(se) { + if (se->on_rq) + enqueue = 0; + + cfs_rq = cfs_rq_of(se); + if (enqueue) + enqueue_entity(cfs_rq, se, ENQUEUE_WAKEUP); + cfs_rq->h_nr_running += task_delta; + + if (cfs_rq_throttled(cfs_rq)) + break; + } + + if (!se) + rq->nr_running += task_delta; + + /* determine whether we need to wake up potentially idle cpu */ + if (rq->curr == rq->idle && rq->cfs.nr_running) + resched_task(rq->curr); +} + +static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, + u64 remaining, u64 expires) +{ + struct cfs_rq *cfs_rq; + u64 runtime = remaining; + + rcu_read_lock(); + list_for_each_entry_rcu(cfs_rq, &cfs_b->throttled_cfs_rq, + throttled_list) { + struct rq *rq = rq_of(cfs_rq); + + raw_spin_lock(&rq->lock); + if (!cfs_rq_throttled(cfs_rq)) + goto next; + + runtime = -cfs_rq->runtime_remaining + 1; + if (runtime > remaining) + runtime = remaining; + remaining -= runtime; + + cfs_rq->runtime_remaining += runtime; + cfs_rq->runtime_expires = expires; + + /* we check whether we're throttled above */ + if (cfs_rq->runtime_remaining > 0) + unthrottle_cfs_rq(cfs_rq); + +next: + raw_spin_unlock(&rq->lock); + + if (!remaining) + break; + } + rcu_read_unlock(); + + return remaining; +} + /* * Responsible for refilling a task_group's bandwidth and unthrottling its * cfs_rqs as appropriate. If there has been no activity within the last @@ -1447,23 +1525,64 @@ static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) */ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) { - int idle = 1; + u64 runtime, runtime_expires; + int idle = 1, throttled; raw_spin_lock(&cfs_b->lock); /* no need to continue the timer with no bandwidth constraint */ if (cfs_b->quota == RUNTIME_INF) goto out_unlock; - idle = cfs_b->idle; + throttled = !list_empty(&cfs_b->throttled_cfs_rq); + /* idle depends on !throttled (for the case of a large deficit) */ + idle = cfs_b->idle && !throttled; + /* if we're going inactive then everything else can be deferred */ if (idle) goto out_unlock; __refill_cfs_bandwidth_runtime(cfs_b); + if (!throttled) { + /* mark as potentially idle for the upcoming period */ + cfs_b->idle = 1; + goto out_unlock; + } + + /* + * There are throttled entities so we must first use the new bandwidth + * to unthrottle them before making it generally available. This + * ensures that all existing debts will be paid before a new cfs_rq is + * allowed to run. + */ + runtime = cfs_b->runtime; + runtime_expires = cfs_b->runtime_expires; + cfs_b->runtime = 0; + + /* + * This check is repeated as we are holding onto the new bandwidth + * while we unthrottle. This can potentially race with an unthrottled + * group trying to acquire new bandwidth from the global pool. + */ + while (throttled && runtime > 0) { + raw_spin_unlock(&cfs_b->lock); + /* we can't nest cfs_b->lock while distributing bandwidth */ + runtime = distribute_cfs_runtime(cfs_b, runtime, + runtime_expires); + raw_spin_lock(&cfs_b->lock); + + throttled = !list_empty(&cfs_b->throttled_cfs_rq); + } - /* mark as potentially idle for the upcoming period */ - cfs_b->idle = 1; + /* return (any) remaining runtime */ + cfs_b->runtime = runtime; + /* + * While we are ensured activity in the period following an + * unthrottle, this also covers the case in which the new bandwidth is + * insufficient to cover the existing bandwidth deficit. (Forcing the + * timer to remain active while there are any throttled entities.) + */ + cfs_b->idle = 0; out_unlock: if (idle) cfs_b->timer_active = 0; -- cgit v1.3-8-gc7d7 From 8277434ef1202ce30315f8edb3fc760aa6e74493 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:35 -0700 Subject: sched: Allow for positional tg_tree walks Extend walk_tg_tree to accept a positional argument static int walk_tg_tree_from(struct task_group *from, tg_visitor down, tg_visitor up, void *data) Existing semantics are preserved, caller must hold rcu_lock() or sufficient analogue. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.677889157@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 50 +++++++++++++++++++++++++++++++++++++------------- 1 file changed, 37 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 4bbabc2c4a77..8ec1e7ac2894 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1591,20 +1591,23 @@ static inline void dec_cpu_load(struct rq *rq, unsigned long load) typedef int (*tg_visitor)(struct task_group *, void *); /* - * Iterate the full tree, calling @down when first entering a node and @up when - * leaving it for the final time. + * Iterate task_group tree rooted at *from, calling @down when first entering a + * node and @up when leaving it for the final time. + * + * Caller must hold rcu_lock or sufficient equivalent. */ -static int walk_tg_tree(tg_visitor down, tg_visitor up, void *data) +static int walk_tg_tree_from(struct task_group *from, + tg_visitor down, tg_visitor up, void *data) { struct task_group *parent, *child; int ret; - rcu_read_lock(); - parent = &root_task_group; + parent = from; + down: ret = (*down)(parent, data); if (ret) - goto out_unlock; + goto out; list_for_each_entry_rcu(child, &parent->children, siblings) { parent = child; goto down; @@ -1613,19 +1616,29 @@ up: continue; } ret = (*up)(parent, data); - if (ret) - goto out_unlock; + if (ret || parent == from) + goto out; child = parent; parent = parent->parent; if (parent) goto up; -out_unlock: - rcu_read_unlock(); - +out: return ret; } +/* + * Iterate the full tree, calling @down when first entering a node and @up when + * leaving it for the final time. + * + * Caller must hold rcu_lock or sufficient equivalent. + */ + +static inline int walk_tg_tree(tg_visitor down, tg_visitor up, void *data) +{ + return walk_tg_tree_from(&root_task_group, down, up, data); +} + static int tg_nop(struct task_group *tg, void *data) { return 0; @@ -8870,13 +8883,19 @@ static int tg_rt_schedulable(struct task_group *tg, void *data) static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) { + int ret; + struct rt_schedulable_data data = { .tg = tg, .rt_period = period, .rt_runtime = runtime, }; - return walk_tg_tree(tg_rt_schedulable, tg_nop, &data); + rcu_read_lock(); + ret = walk_tg_tree(tg_rt_schedulable, tg_nop, &data); + rcu_read_unlock(); + + return ret; } static int tg_set_rt_bandwidth(struct task_group *tg, @@ -9333,6 +9352,7 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data) static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) { + int ret; struct cfs_schedulable_data data = { .tg = tg, .period = period, @@ -9344,7 +9364,11 @@ static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) do_div(data.quota, NSEC_PER_USEC); } - return walk_tg_tree(tg_cfs_schedulable_down, tg_nop, &data); + rcu_read_lock(); + ret = walk_tg_tree(tg_cfs_schedulable_down, tg_nop, &data); + rcu_read_unlock(); + + return ret; } #endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ -- cgit v1.3-8-gc7d7 From 64660c864f46202b932b911a69deb09805bdbaf8 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:36 -0700 Subject: sched: Prevent interactions with throttled entities From the perspective of load-balance and shares distribution, throttled entities should be invisible. However, both of these operations work on 'active' lists and are not inherently aware of what group hierarchies may be present. In some cases this may be side-stepped (e.g. we could sideload via tg_load_down in load balance) while in others (e.g. update_shares()) it is more difficult to compute without incurring some O(n^2) costs. Instead, track hierarchicaal throttled state at time of transition. This allows us to easily identify whether an entity belongs to a throttled hierarchy and avoid incorrect interactions with it. Also, when an entity leaves a throttled hierarchy we need to advance its time averaging for shares averaging so that the elapsed throttled time is not considered as part of the cfs_rq's operation. We also use this information to prevent buddy interactions in the wakeup and yield_to() paths. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.777916795@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- kernel/sched_fair.c | 99 +++++++++++++++++++++++++++++++++++++++++++++++++---- 2 files changed, 94 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 8ec1e7ac2894..5db05f6fb470 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -402,7 +402,7 @@ struct cfs_rq { u64 runtime_expires; s64 runtime_remaining; - int throttled; + int throttled, throttle_count; struct list_head throttled_list; #endif #endif diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 76411950ff3b..5a2089492a98 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -706,6 +706,8 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) } #ifdef CONFIG_FAIR_GROUP_SCHED +/* we need this in update_cfs_load and load-balance functions below */ +static inline int throttled_hierarchy(struct cfs_rq *cfs_rq); # ifdef CONFIG_SMP static void update_cfs_rq_load_contribution(struct cfs_rq *cfs_rq, int global_update) @@ -728,7 +730,7 @@ static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update) u64 now, delta; unsigned long load = cfs_rq->load.weight; - if (cfs_rq->tg == &root_task_group) + if (cfs_rq->tg == &root_task_group || throttled_hierarchy(cfs_rq)) return; now = rq_of(cfs_rq)->clock_task; @@ -837,7 +839,7 @@ static void update_cfs_shares(struct cfs_rq *cfs_rq) tg = cfs_rq->tg; se = tg->se[cpu_of(rq_of(cfs_rq))]; - if (!se) + if (!se || throttled_hierarchy(cfs_rq)) return; #ifndef CONFIG_SMP if (likely(se->load.weight == tg->shares)) @@ -1403,6 +1405,65 @@ static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) return cfs_rq->throttled; } +/* check whether cfs_rq, or any parent, is throttled */ +static inline int throttled_hierarchy(struct cfs_rq *cfs_rq) +{ + return cfs_rq->throttle_count; +} + +/* + * Ensure that neither of the group entities corresponding to src_cpu or + * dest_cpu are members of a throttled hierarchy when performing group + * load-balance operations. + */ +static inline int throttled_lb_pair(struct task_group *tg, + int src_cpu, int dest_cpu) +{ + struct cfs_rq *src_cfs_rq, *dest_cfs_rq; + + src_cfs_rq = tg->cfs_rq[src_cpu]; + dest_cfs_rq = tg->cfs_rq[dest_cpu]; + + return throttled_hierarchy(src_cfs_rq) || + throttled_hierarchy(dest_cfs_rq); +} + +/* updated child weight may affect parent so we have to do this bottom up */ +static int tg_unthrottle_up(struct task_group *tg, void *data) +{ + struct rq *rq = data; + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; + + cfs_rq->throttle_count--; +#ifdef CONFIG_SMP + if (!cfs_rq->throttle_count) { + u64 delta = rq->clock_task - cfs_rq->load_stamp; + + /* leaving throttled state, advance shares averaging windows */ + cfs_rq->load_stamp += delta; + cfs_rq->load_last += delta; + + /* update entity weight now that we are on_rq again */ + update_cfs_shares(cfs_rq); + } +#endif + + return 0; +} + +static int tg_throttle_down(struct task_group *tg, void *data) +{ + struct rq *rq = data; + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; + + /* group is entering throttled state, record last load */ + if (!cfs_rq->throttle_count) + update_cfs_load(cfs_rq, 0); + cfs_rq->throttle_count++; + + return 0; +} + static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); @@ -1413,7 +1474,9 @@ static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; /* account load preceding throttle */ - update_cfs_load(cfs_rq, 0); + rcu_read_lock(); + walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq); + rcu_read_unlock(); task_delta = cfs_rq->h_nr_running; for_each_sched_entity(se) { @@ -1454,6 +1517,10 @@ static void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) list_del_rcu(&cfs_rq->throttled_list); raw_spin_unlock(&cfs_b->lock); + update_rq_clock(rq); + /* update hierarchical throttle state */ + walk_tg_tree_from(cfs_rq->tg, tg_nop, tg_unthrottle_up, (void *)rq); + if (!cfs_rq->load.weight) return; @@ -1598,6 +1665,17 @@ static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) { return 0; } + +static inline int throttled_hierarchy(struct cfs_rq *cfs_rq) +{ + return 0; +} + +static inline int throttled_lb_pair(struct task_group *tg, + int src_cpu, int dest_cpu) +{ + return 0; +} #endif /************************************************** @@ -2493,6 +2571,9 @@ move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest, for_each_leaf_cfs_rq(busiest, cfs_rq) { list_for_each_entry_safe(p, n, &cfs_rq->tasks, se.group_node) { + if (throttled_lb_pair(task_group(p), + busiest->cpu, this_cpu)) + break; if (!can_migrate_task(p, busiest, this_cpu, sd, idle, &pinned)) @@ -2608,8 +2689,13 @@ static void update_shares(int cpu) * Iterates the task_group tree in a bottom up fashion, see * list_add_leaf_cfs_rq() for details. */ - for_each_leaf_cfs_rq(rq, cfs_rq) + for_each_leaf_cfs_rq(rq, cfs_rq) { + /* throttled entities do not contribute to load */ + if (throttled_hierarchy(cfs_rq)) + continue; + update_shares_cpu(cfs_rq->tg, cpu); + } rcu_read_unlock(); } @@ -2659,9 +2745,10 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, u64 rem_load, moved_load; /* - * empty group + * empty group or part of a throttled hierarchy */ - if (!busiest_cfs_rq->task_weight) + if (!busiest_cfs_rq->task_weight || + throttled_lb_pair(busiest_cfs_rq->tg, cpu_of(busiest), this_cpu)) continue; rem_load = (u64)rem_load_move * busiest_weight; -- cgit v1.3-8-gc7d7 From 5238cdd3873e67a98b28c1161d65d2a615c320a3 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:37 -0700 Subject: sched: Prevent buddy interactions with throttled entities Buddies allow us to select "on-rq" entities without actually selecting them from a cfs_rq's rb_tree. As a result we must ensure that throttled entities are not falsely nominated as buddies. The fact that entities are dequeued within throttle_entity is not sufficient for clearing buddy status as the nomination may occur after throttling. Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.886850167@google.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 5a2089492a98..1d4acbea9e60 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -2348,6 +2348,15 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (unlikely(se == pse)) return; + /* + * This is possible from callers such as pull_task(), in which we + * unconditionally check_prempt_curr() after an enqueue (which may have + * lead to a throttle). This both saves work and prevents false + * next-buddy nomination below. + */ + if (unlikely(throttled_hierarchy(cfs_rq_of(pse)))) + return; + if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK)) { set_next_buddy(pse); next_buddy_marked = 1; @@ -2356,6 +2365,12 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ /* * We can come here with TIF_NEED_RESCHED already set from new task * wake up path. + * + * Note: this also catches the edge-case of curr being in a throttled + * group (e.g. via set_curr_task), since update_curr() (in the + * enqueue of curr) will have resulted in resched being set. This + * prevents us from potentially nominating it as a false LAST_BUDDY + * below. */ if (test_tsk_need_resched(curr)) return; @@ -2474,7 +2489,8 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp { struct sched_entity *se = &p->se; - if (!se->on_rq) + /* throttled hierarchies are not runnable */ + if (!se->on_rq || throttled_hierarchy(cfs_rq_of(se))) return false; /* Tell the scheduler that we'd really like pse to run next. */ -- cgit v1.3-8-gc7d7 From 8cb120d3e41a0464a559d639d519cef563717a4e Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:38 -0700 Subject: sched: Migrate throttled tasks on HOTPLUG Throttled tasks are invisisble to cpu-offline since they are not eligible for selection by pick_next_task(). The regular 'escape' path for a thread that is blocked at offline is via ttwu->select_task_rq, however this will not handle a throttled group since there are no individual thread wakeups on an unthrottle. Resolve this by unthrottling offline cpus so that threads can be migrated. Signed-off-by: Paul Turner Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184757.989000590@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 5db05f6fb470..397317248ddd 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6335,6 +6335,30 @@ static void calc_global_load_remove(struct rq *rq) rq->calc_load_active = 0; } +#ifdef CONFIG_CFS_BANDWIDTH +static void unthrottle_offline_cfs_rqs(struct rq *rq) +{ + struct cfs_rq *cfs_rq; + + for_each_leaf_cfs_rq(rq, cfs_rq) { + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + + if (!cfs_rq->runtime_enabled) + continue; + + /* + * clock_task is not advancing so we just need to make sure + * there's some valid quota amount + */ + cfs_rq->runtime_remaining = cfs_b->quota; + if (cfs_rq_throttled(cfs_rq)) + unthrottle_cfs_rq(cfs_rq); + } +} +#else +static void unthrottle_offline_cfs_rqs(struct rq *rq) {} +#endif + /* * Migrate all tasks from the rq, sleeping tasks will be migrated by * try_to_wake_up()->select_task_rq(). @@ -6360,6 +6384,9 @@ static void migrate_tasks(unsigned int dead_cpu) */ rq->stop = NULL; + /* Ensure any throttled groups are reachable by pick_next_task */ + unthrottle_offline_cfs_rqs(rq); + for ( ; ; ) { /* * There's this thread running, bail when that's the only -- cgit v1.3-8-gc7d7 From d3d9dc3302368269acf94b7381663b93000fe2fe Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:39 -0700 Subject: sched: Throttle entities exceeding their allowed bandwidth With the machinery in place to throttle and unthrottle entities, as well as handle their participation (or lack there of) we can now enable throttling. There are 2 points that we must check whether it's time to set throttled state: put_prev_entity() and enqueue_entity(). - put_prev_entity() is the typical throttle path, we reach it by exceeding our allocated run-time within update_curr()->account_cfs_rq_runtime() and going through a reschedule. - enqueue_entity() covers the case of a wake-up into an already throttled group. In this case we know the group cannot be on_rq and can throttle immediately. Checks are added at time of put_prev_entity() and enqueue_entity() Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184758.091415417@google.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 50 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 1d4acbea9e60..f9f671a7d0af 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -970,6 +970,8 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int initial) se->vruntime = vruntime; } +static void check_enqueue_throttle(struct cfs_rq *cfs_rq); + static void enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { @@ -999,8 +1001,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) __enqueue_entity(cfs_rq, se); se->on_rq = 1; - if (cfs_rq->nr_running == 1) + if (cfs_rq->nr_running == 1) { list_add_leaf_cfs_rq(cfs_rq); + check_enqueue_throttle(cfs_rq); + } } static void __clear_buddies_last(struct sched_entity *se) @@ -1202,6 +1206,8 @@ static struct sched_entity *pick_next_entity(struct cfs_rq *cfs_rq) return se; } +static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq); + static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev) { /* @@ -1211,6 +1217,9 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev) if (prev->on_rq) update_curr(cfs_rq); + /* throttle cfs_rqs exceeding runtime */ + check_cfs_rq_runtime(cfs_rq); + check_spread(cfs_rq, prev); if (prev->on_rq) { update_stats_wait_start(cfs_rq, prev); @@ -1464,7 +1473,7 @@ static int tg_throttle_down(struct task_group *tg, void *data) return 0; } -static __used void throttle_cfs_rq(struct cfs_rq *cfs_rq) +static void throttle_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); @@ -1657,9 +1666,48 @@ out_unlock: return idle; } + +/* + * When a group wakes up we want to make sure that its quota is not already + * expired/exceeded, otherwise it may be allowed to steal additional ticks of + * runtime as update_curr() throttling can not not trigger until it's on-rq. + */ +static void check_enqueue_throttle(struct cfs_rq *cfs_rq) +{ + /* an active group must be handled by the update_curr()->put() path */ + if (!cfs_rq->runtime_enabled || cfs_rq->curr) + return; + + /* ensure the group is not already throttled */ + if (cfs_rq_throttled(cfs_rq)) + return; + + /* update runtime allocation */ + account_cfs_rq_runtime(cfs_rq, 0); + if (cfs_rq->runtime_remaining <= 0) + throttle_cfs_rq(cfs_rq); +} + +/* conditionally throttle active cfs_rq's from put_prev_entity() */ +static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq) +{ + if (likely(!cfs_rq->runtime_enabled || cfs_rq->runtime_remaining > 0)) + return; + + /* + * it's possible for a throttled entity to be forced into a running + * state (e.g. set_curr_task), in this case we're finished. + */ + if (cfs_rq_throttled(cfs_rq)) + return; + + throttle_cfs_rq(cfs_rq); +} #else static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) {} +static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} +static void check_enqueue_throttle(struct cfs_rq *cfs_rq) {} static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) { -- cgit v1.3-8-gc7d7 From e8da1b18b32064c43881bceef0f051c2110c9ab9 Mon Sep 17 00:00:00 2001 From: Nikhil Rao Date: Thu, 21 Jul 2011 09:43:40 -0700 Subject: sched: Add exports tracking cfs bandwidth control statistics This change introduces statistics exports for the cpu sub-system, these are added through the use of a stat file similar to that exported by other subsystems. The following exports are included: nr_periods: number of periods in which execution occurred nr_throttled: the number of periods above in which execution was throttle throttled_time: cumulative wall-time that any cpus have been throttled for this group Signed-off-by: Paul Turner Signed-off-by: Nikhil Rao Signed-off-by: Bharata B Rao Reviewed-by: Hidetoshi Seto Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184758.198901931@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 21 +++++++++++++++++++++ kernel/sched_fair.c | 7 +++++++ 2 files changed, 28 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 397317248ddd..35c91859f8a6 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -262,6 +262,9 @@ struct cfs_bandwidth { struct hrtimer period_timer; struct list_head throttled_cfs_rq; + /* statistics */ + int nr_periods, nr_throttled; + u64 throttled_time; #endif }; @@ -402,6 +405,7 @@ struct cfs_rq { u64 runtime_expires; s64 runtime_remaining; + u64 throttled_timestamp; int throttled, throttle_count; struct list_head throttled_list; #endif @@ -9397,6 +9401,19 @@ static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) return ret; } + +static int cpu_stats_show(struct cgroup *cgrp, struct cftype *cft, + struct cgroup_map_cb *cb) +{ + struct task_group *tg = cgroup_tg(cgrp); + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); + + cb->fill(cb, "nr_periods", cfs_b->nr_periods); + cb->fill(cb, "nr_throttled", cfs_b->nr_throttled); + cb->fill(cb, "throttled_time", cfs_b->throttled_time); + + return 0; +} #endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -9443,6 +9460,10 @@ static struct cftype cpu_files[] = { .read_u64 = cpu_cfs_period_read_u64, .write_u64 = cpu_cfs_period_write_u64, }, + { + .name = "stat", + .read_map = cpu_stats_show, + }, #endif #ifdef CONFIG_RT_GROUP_SCHED { diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index f9f671a7d0af..d201f28c1de7 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1506,6 +1506,7 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq) rq->nr_running -= task_delta; cfs_rq->throttled = 1; + cfs_rq->throttled_timestamp = rq->clock; raw_spin_lock(&cfs_b->lock); list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); raw_spin_unlock(&cfs_b->lock); @@ -1523,8 +1524,10 @@ static void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) cfs_rq->throttled = 0; raw_spin_lock(&cfs_b->lock); + cfs_b->throttled_time += rq->clock - cfs_rq->throttled_timestamp; list_del_rcu(&cfs_rq->throttled_list); raw_spin_unlock(&cfs_b->lock); + cfs_rq->throttled_timestamp = 0; update_rq_clock(rq); /* update hierarchical throttle state */ @@ -1612,6 +1615,7 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) throttled = !list_empty(&cfs_b->throttled_cfs_rq); /* idle depends on !throttled (for the case of a large deficit) */ idle = cfs_b->idle && !throttled; + cfs_b->nr_periods += overrun; /* if we're going inactive then everything else can be deferred */ if (idle) @@ -1625,6 +1629,9 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) goto out_unlock; } + /* account preceding periods in which throttling occurred */ + cfs_b->nr_throttled += overrun; + /* * There are throttled entities so we must first use the new bandwidth * to unthrottle them before making it generally available. This -- cgit v1.3-8-gc7d7 From d8b4986d3dbc4fabc2054d63f1d31d6ed2fb1ca8 Mon Sep 17 00:00:00 2001 From: Paul Turner Date: Thu, 21 Jul 2011 09:43:41 -0700 Subject: sched: Return unused runtime on group dequeue When a local cfs_rq blocks we return the majority of its remaining quota to the global bandwidth pool for use by other runqueues. We do this only when the quota is current and there is more than min_cfs_rq_quota [1ms by default] of runtime remaining on the rq. In the case where there are throttled runqueues and we have sufficient bandwidth to meter out a slice, a second timer is kicked off to handle this delivery, unthrottling where appropriate. Using a 'worst case' antagonist which executes on each cpu for 1ms before moving onto the next on a fairly large machine: no quota generations: 197.47 ms /cgroup/a/cpuacct.usage 199.46 ms /cgroup/a/cpuacct.usage 205.46 ms /cgroup/a/cpuacct.usage 198.46 ms /cgroup/a/cpuacct.usage 208.39 ms /cgroup/a/cpuacct.usage Since we are allowed to use "stale" quota our usage is effectively bounded by the rate of input into the global pool and performance is relatively stable. with quota generations [1s increments]: 119.58 ms /cgroup/a/cpuacct.usage 119.65 ms /cgroup/a/cpuacct.usage 119.64 ms /cgroup/a/cpuacct.usage 119.63 ms /cgroup/a/cpuacct.usage 119.60 ms /cgroup/a/cpuacct.usage The large deficit here is due to quota generations (/intentionally/) preventing us from now using previously stranded slack quota. The cost is that this quota becomes unavailable. with quota generations and quota return: 200.09 ms /cgroup/a/cpuacct.usage 200.09 ms /cgroup/a/cpuacct.usage 198.09 ms /cgroup/a/cpuacct.usage 200.09 ms /cgroup/a/cpuacct.usage 200.06 ms /cgroup/a/cpuacct.usage By returning unused quota we're able to both stably consume our desired quota and prevent unintentional overages due to the abuse of slack quota from previous quota periods (especially on a large machine). Signed-off-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110721184758.306848658@google.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 15 +++++++- kernel/sched_fair.c | 108 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 122 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 35c91859f8a6..6baade0d7649 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -259,7 +259,7 @@ struct cfs_bandwidth { u64 runtime_expires; int idle, timer_active; - struct hrtimer period_timer; + struct hrtimer period_timer, slack_timer; struct list_head throttled_cfs_rq; /* statistics */ @@ -421,6 +421,16 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) static inline u64 default_cfs_period(void); static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun); +static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b); + +static enum hrtimer_restart sched_cfs_slack_timer(struct hrtimer *timer) +{ + struct cfs_bandwidth *cfs_b = + container_of(timer, struct cfs_bandwidth, slack_timer); + do_sched_cfs_slack_timer(cfs_b); + + return HRTIMER_NORESTART; +} static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) { @@ -453,6 +463,8 @@ static void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); cfs_b->period_timer.function = sched_cfs_period_timer; + hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + cfs_b->slack_timer.function = sched_cfs_slack_timer; } static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) @@ -488,6 +500,7 @@ static void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { hrtimer_cancel(&cfs_b->period_timer); + hrtimer_cancel(&cfs_b->slack_timer); } #else static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index d201f28c1de7..1ca2cd44d64a 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1052,6 +1052,8 @@ static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) __clear_buddies_skip(se); } +static void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); + static void dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { @@ -1090,6 +1092,9 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (!(flags & DEQUEUE_SLEEP)) se->vruntime -= cfs_rq->min_vruntime; + /* return excess runtime on last dequeue */ + return_cfs_rq_runtime(cfs_rq); + update_min_vruntime(cfs_rq); update_cfs_shares(cfs_rq); } @@ -1674,6 +1679,108 @@ out_unlock: return idle; } +/* a cfs_rq won't donate quota below this amount */ +static const u64 min_cfs_rq_runtime = 1 * NSEC_PER_MSEC; +/* minimum remaining period time to redistribute slack quota */ +static const u64 min_bandwidth_expiration = 2 * NSEC_PER_MSEC; +/* how long we wait to gather additional slack before distributing */ +static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC; + +/* are we near the end of the current quota period? */ +static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) +{ + struct hrtimer *refresh_timer = &cfs_b->period_timer; + u64 remaining; + + /* if the call-back is running a quota refresh is already occurring */ + if (hrtimer_callback_running(refresh_timer)) + return 1; + + /* is a quota refresh about to occur? */ + remaining = ktime_to_ns(hrtimer_expires_remaining(refresh_timer)); + if (remaining < min_expire) + return 1; + + return 0; +} + +static void start_cfs_slack_bandwidth(struct cfs_bandwidth *cfs_b) +{ + u64 min_left = cfs_bandwidth_slack_period + min_bandwidth_expiration; + + /* if there's a quota refresh soon don't bother with slack */ + if (runtime_refresh_within(cfs_b, min_left)) + return; + + start_bandwidth_timer(&cfs_b->slack_timer, + ns_to_ktime(cfs_bandwidth_slack_period)); +} + +/* we know any runtime found here is valid as update_curr() precedes return */ +static void __return_cfs_rq_runtime(struct cfs_rq *cfs_rq) +{ + struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + s64 slack_runtime = cfs_rq->runtime_remaining - min_cfs_rq_runtime; + + if (slack_runtime <= 0) + return; + + raw_spin_lock(&cfs_b->lock); + if (cfs_b->quota != RUNTIME_INF && + cfs_rq->runtime_expires == cfs_b->runtime_expires) { + cfs_b->runtime += slack_runtime; + + /* we are under rq->lock, defer unthrottling using a timer */ + if (cfs_b->runtime > sched_cfs_bandwidth_slice() && + !list_empty(&cfs_b->throttled_cfs_rq)) + start_cfs_slack_bandwidth(cfs_b); + } + raw_spin_unlock(&cfs_b->lock); + + /* even if it's not valid for return we don't want to try again */ + cfs_rq->runtime_remaining -= slack_runtime; +} + +static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) +{ + if (!cfs_rq->runtime_enabled || !cfs_rq->nr_running) + return; + + __return_cfs_rq_runtime(cfs_rq); +} + +/* + * This is done with a timer (instead of inline with bandwidth return) since + * it's necessary to juggle rq->locks to unthrottle their respective cfs_rqs. + */ +static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) +{ + u64 runtime = 0, slice = sched_cfs_bandwidth_slice(); + u64 expires; + + /* confirm we're still not at a refresh boundary */ + if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) + return; + + raw_spin_lock(&cfs_b->lock); + if (cfs_b->quota != RUNTIME_INF && cfs_b->runtime > slice) { + runtime = cfs_b->runtime; + cfs_b->runtime = 0; + } + expires = cfs_b->runtime_expires; + raw_spin_unlock(&cfs_b->lock); + + if (!runtime) + return; + + runtime = distribute_cfs_runtime(cfs_b, runtime, expires); + + raw_spin_lock(&cfs_b->lock); + if (expires == cfs_b->runtime_expires) + cfs_b->runtime = runtime; + raw_spin_unlock(&cfs_b->lock); +} + /* * When a group wakes up we want to make sure that its quota is not already * expired/exceeded, otherwise it may be allowed to steal additional ticks of @@ -1715,6 +1822,7 @@ static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, unsigned long delta_exec) {} static void check_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} static void check_enqueue_throttle(struct cfs_rq *cfs_rq) {} +static void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) { -- cgit v1.3-8-gc7d7 From 17f2ae7f677f023997e02fd2ebabd90ea2a0390d Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 14 Aug 2011 13:34:31 +0200 Subject: PM / Domains: Fix build for CONFIG_PM_RUNTIME unset Function genpd_queue_power_off_work() is not defined for CONFIG_PM_RUNTIME, so pm_genpd_poweroff_unused() causes a build error to happen in that case. Fix the problem by making pm_genpd_poweroff_unused() depend on CONFIG_PM_RUNTIME too. Signed-off-by: Rafael J. Wysocki --- drivers/base/power/domain.c | 30 +++++++++++++++--------------- include/linux/pm_domain.h | 10 +++++++--- kernel/power/Kconfig | 4 ++++ 3 files changed, 26 insertions(+), 18 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index e18566a0fedd..1c374579407c 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -460,6 +460,21 @@ static int pm_genpd_runtime_resume(struct device *dev) return 0; } +/** + * pm_genpd_poweroff_unused - Power off all PM domains with no devices in use. + */ +void pm_genpd_poweroff_unused(void) +{ + struct generic_pm_domain *genpd; + + mutex_lock(&gpd_list_lock); + + list_for_each_entry(genpd, &gpd_list, gpd_list_node) + genpd_queue_power_off_work(genpd); + + mutex_unlock(&gpd_list_lock); +} + #else static inline void genpd_power_off_work_fn(struct work_struct *work) {} @@ -1255,18 +1270,3 @@ void pm_genpd_init(struct generic_pm_domain *genpd, list_add(&genpd->gpd_list_node, &gpd_list); mutex_unlock(&gpd_list_lock); } - -/** - * pm_genpd_poweroff_unused - Power off all PM domains with no devices in use. - */ -void pm_genpd_poweroff_unused(void) -{ - struct generic_pm_domain *genpd; - - mutex_lock(&gpd_list_lock); - - list_for_each_entry(genpd, &gpd_list, gpd_list_node) - genpd_queue_power_off_work(genpd); - - mutex_unlock(&gpd_list_lock); -} diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h index 21097cb086fe..f9ec1736a116 100644 --- a/include/linux/pm_domain.h +++ b/include/linux/pm_domain.h @@ -72,8 +72,6 @@ extern int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd, extern void pm_genpd_init(struct generic_pm_domain *genpd, struct dev_power_governor *gov, bool is_off); extern int pm_genpd_poweron(struct generic_pm_domain *genpd); -extern void pm_genpd_poweroff_unused(void); -extern void genpd_queue_power_off_work(struct generic_pm_domain *genpd); #else static inline int pm_genpd_add_device(struct generic_pm_domain *genpd, struct device *dev) @@ -101,8 +99,14 @@ static inline int pm_genpd_poweron(struct generic_pm_domain *genpd) { return -ENOSYS; } -static inline void pm_genpd_poweroff_unused(void) {} +#endif + +#ifdef CONFIG_PM_GENERIC_DOMAINS_RUNTIME +extern void genpd_queue_power_off_work(struct generic_pm_domain *genpd); +extern void pm_genpd_poweroff_unused(void); +#else static inline void genpd_queue_power_off_work(struct generic_pm_domain *gpd) {} +static inline void pm_genpd_poweroff_unused(void) {} #endif #endif /* _LINUX_PM_DOMAIN_H */ diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index b1914cb9095c..3744c594b19b 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -231,3 +231,7 @@ config PM_CLK config PM_GENERIC_DOMAINS bool depends on PM + +config PM_GENERIC_DOMAINS_RUNTIME + def_bool y + depends on PM_RUNTIME && PM_GENERIC_DOMAINS -- cgit v1.3-8-gc7d7 From d522a0d17963e9c2e556db2cbd60c96d40505b6c Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 18 Aug 2011 12:19:27 -0700 Subject: irqdesc: fix new kernel-doc warning Fix kernel-doc warning in irqdesc.c: Warning(kernel/irq/irqdesc.c:353): No description found for parameter 'owner' Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds --- kernel/irq/irqdesc.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index cb65d0360e31..039b889ea053 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -344,6 +344,7 @@ EXPORT_SYMBOL_GPL(irq_free_descs); * @from: Start the search from this irq number * @cnt: Number of consecutive irqs to allocate. * @node: Preferred node on which the irq descriptor should be allocated + * @owner: Owning module (can be NULL) * * Returns the first irq number or error code */ -- cgit v1.3-8-gc7d7 From 81570d9caaad46a056580c9af078c5c55e6c764f Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:45 +0200 Subject: tracing/filter: Use static allocation for filter predicates Don't dynamically allocate filter_pred struct, use static memory. This way we can get rid of the code managing the dynamic filter_pred struct object. The create_pred function integrates create_logical_pred function. This way the static predicate memory is returned only from one place. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-2-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 57 +++++++++++--------------------------- 1 file changed, 16 insertions(+), 41 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 256764ecccd6..cb295a117ee7 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -630,11 +630,7 @@ find_event_field(struct ftrace_event_call *call, char *name) static void filter_free_pred(struct filter_pred *pred) { - if (!pred) - return; - kfree(pred->field_name); - kfree(pred); } static void filter_clear_pred(struct filter_pred *pred) @@ -1302,39 +1298,30 @@ parse_operand: return 0; } -static struct filter_pred *create_pred(int op, char *operand1, char *operand2) +static struct filter_pred *create_pred(struct filter_parse_state *ps, + int op, char *operand1, char *operand2) { - struct filter_pred *pred; + static struct filter_pred pred; - pred = kzalloc(sizeof(*pred), GFP_KERNEL); - if (!pred) - return NULL; + memset(&pred, 0, sizeof(pred)); + pred.op = op; - pred->field_name = kstrdup(operand1, GFP_KERNEL); - if (!pred->field_name) { - kfree(pred); + if (op == OP_AND || op == OP_OR) + return &pred; + + if (!operand1 || !operand2) { + parse_error(ps, FILT_ERR_MISSING_FIELD, 0); return NULL; } - strcpy(pred->regex.pattern, operand2); - pred->regex.len = strlen(pred->regex.pattern); - - pred->op = op; - - return pred; -} - -static struct filter_pred *create_logical_pred(int op) -{ - struct filter_pred *pred; - - pred = kzalloc(sizeof(*pred), GFP_KERNEL); - if (!pred) + pred.field_name = kstrdup(operand1, GFP_KERNEL); + if (!pred.field_name) return NULL; - pred->op = op; + strcpy(pred.regex.pattern, operand2); + pred.regex.len = strlen(pred.regex.pattern); - return pred; + return &pred; } static int check_preds(struct filter_parse_state *ps) @@ -1643,19 +1630,7 @@ static int replace_preds(struct ftrace_event_call *call, goto fail; } - if (elt->op == OP_AND || elt->op == OP_OR) { - pred = create_logical_pred(elt->op); - goto add_pred; - } - - if (!operand1 || !operand2) { - parse_error(ps, FILT_ERR_MISSING_FIELD, 0); - err = -EINVAL; - goto fail; - } - - pred = create_pred(elt->op, operand1, operand2); -add_pred: + pred = create_pred(ps, elt->op, operand1, operand2); if (!pred) { err = -ENOMEM; goto fail; -- cgit v1.3-8-gc7d7 From 9d96cd1743547f07a8a6c51a3f7741cfca0a0bee Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:46 +0200 Subject: tracing/filter: Separate predicate init and filter addition Making the code cleaner by having one function to fully prepare the predicate (create_pred), and another to add the predicate to the filter (filter_add_pred). As a benefit, this way the dry_run flag stays only inside the replace_preds function and is not passed deeper. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-3-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 56 ++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 33 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index cb295a117ee7..61c8dec3b2b7 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -685,8 +685,7 @@ __pop_pred_stack(struct pred_stack *stack) static int filter_set_pred(struct event_filter *filter, int idx, struct pred_stack *stack, - struct filter_pred *src, - filter_pred_fn_t fn) + struct filter_pred *src) { struct filter_pred *dest = &filter->preds[idx]; struct filter_pred *left; @@ -698,7 +697,6 @@ static int filter_set_pred(struct event_filter *filter, if (!dest->field_name) return -ENOMEM; } - dest->fn = fn; dest->index = idx; if (dest->op == OP_OR || dest->op == OP_AND) { @@ -836,12 +834,10 @@ static void filter_free_subsystem_filters(struct event_subsystem *system) } } -static int filter_add_pred_fn(struct filter_parse_state *ps, - struct ftrace_event_call *call, - struct event_filter *filter, - struct filter_pred *pred, - struct pred_stack *stack, - filter_pred_fn_t fn) +static int filter_add_pred(struct filter_parse_state *ps, + struct event_filter *filter, + struct filter_pred *pred, + struct pred_stack *stack) { int idx, err; @@ -852,7 +848,7 @@ static int filter_add_pred_fn(struct filter_parse_state *ps, idx = filter->n_preds; filter_clear_pred(&filter->preds[idx]); - err = filter_set_pred(filter, idx, stack, pred, fn); + err = filter_set_pred(filter, idx, stack, pred); if (err) return err; @@ -933,25 +929,16 @@ static filter_pred_fn_t select_comparison_fn(int op, int field_size, return fn; } -static int filter_add_pred(struct filter_parse_state *ps, - struct ftrace_event_call *call, - struct event_filter *filter, - struct filter_pred *pred, - struct pred_stack *stack, - bool dry_run) +static int init_pred(struct filter_parse_state *ps, + struct ftrace_event_call *call, + struct filter_pred *pred) + { struct ftrace_event_field *field; - filter_pred_fn_t fn; + filter_pred_fn_t fn = filter_pred_none; unsigned long long val; int ret; - fn = pred->fn = filter_pred_none; - - if (pred->op == OP_AND) - goto add_pred_fn; - else if (pred->op == OP_OR) - goto add_pred_fn; - field = find_event_field(call, pred->field_name); if (!field) { parse_error(ps, FILT_ERR_FIELD_NOT_FOUND, 0); @@ -997,9 +984,7 @@ static int filter_add_pred(struct filter_parse_state *ps, if (pred->op == OP_NE) pred->not = 1; -add_pred_fn: - if (!dry_run) - return filter_add_pred_fn(ps, call, filter, pred, stack, fn); + pred->fn = fn; return 0; } @@ -1299,6 +1284,7 @@ parse_operand: } static struct filter_pred *create_pred(struct filter_parse_state *ps, + struct ftrace_event_call *call, int op, char *operand1, char *operand2) { static struct filter_pred pred; @@ -1321,7 +1307,7 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps, strcpy(pred.regex.pattern, operand2); pred.regex.len = strlen(pred.regex.pattern); - return &pred; + return init_pred(ps, call, &pred) ? NULL : &pred; } static int check_preds(struct filter_parse_state *ps) @@ -1630,16 +1616,20 @@ static int replace_preds(struct ftrace_event_call *call, goto fail; } - pred = create_pred(ps, elt->op, operand1, operand2); + pred = create_pred(ps, call, elt->op, operand1, operand2); if (!pred) { err = -ENOMEM; goto fail; } - err = filter_add_pred(ps, call, filter, pred, &stack, dry_run); - filter_free_pred(pred); - if (err) - goto fail; + if (!dry_run) { + err = filter_add_pred(ps, filter, pred, &stack); + if (err) { + filter_free_pred(pred); + goto fail; + } + } + filter_free_pred(pred); operand1 = operand2 = NULL; } -- cgit v1.3-8-gc7d7 From 61aaef55300088e12d7f853adeea65d1aa1562db Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:47 +0200 Subject: tracing/filter: Remove field_name from filter_pred struct The field_name was used just for finding event's fields. This way we don't need to care about field_name allocation/free. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-4-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace.h | 11 +------- kernel/trace/trace_events_filter.c | 53 +++++++++----------------------------- 2 files changed, 13 insertions(+), 51 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 616846bcfee5..2eb3cf6d37bc 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -761,16 +761,7 @@ struct filter_pred { filter_pred_fn_t fn; u64 val; struct regex regex; - /* - * Leaf nodes use field_name, ops is used by AND and OR - * nodes. The field_name is always freed when freeing a pred. - * We can overload field_name for ops and have it freed - * as well. - */ - union { - char *field_name; - unsigned short *ops; - }; + unsigned short *ops; int offset; int not; int op; diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 61c8dec3b2b7..97b93f31884b 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -628,18 +628,6 @@ find_event_field(struct ftrace_event_call *call, char *name) return __find_event_field(head, name); } -static void filter_free_pred(struct filter_pred *pred) -{ - kfree(pred->field_name); -} - -static void filter_clear_pred(struct filter_pred *pred) -{ - kfree(pred->field_name); - pred->field_name = NULL; - pred->regex.len = 0; -} - static int __alloc_pred_stack(struct pred_stack *stack, int n_preds) { stack->preds = kzalloc(sizeof(*stack->preds)*(n_preds + 1), GFP_KERNEL); @@ -692,11 +680,6 @@ static int filter_set_pred(struct event_filter *filter, struct filter_pred *right; *dest = *src; - if (src->field_name) { - dest->field_name = kstrdup(src->field_name, GFP_KERNEL); - if (!dest->field_name) - return -ENOMEM; - } dest->index = idx; if (dest->op == OP_OR || dest->op == OP_AND) { @@ -737,11 +720,7 @@ static int filter_set_pred(struct event_filter *filter, static void __free_preds(struct event_filter *filter) { - int i; - if (filter->preds) { - for (i = 0; i < filter->a_preds; i++) - kfree(filter->preds[i].field_name); kfree(filter->preds); filter->preds = NULL; } @@ -839,16 +818,14 @@ static int filter_add_pred(struct filter_parse_state *ps, struct filter_pred *pred, struct pred_stack *stack) { - int idx, err; + int err; if (WARN_ON(filter->n_preds == filter->a_preds)) { parse_error(ps, FILT_ERR_TOO_MANY_PREDS, 0); return -ENOSPC; } - idx = filter->n_preds; - filter_clear_pred(&filter->preds[idx]); - err = filter_set_pred(filter, idx, stack, pred); + err = filter_set_pred(filter, filter->n_preds, stack, pred); if (err) return err; @@ -930,21 +907,14 @@ static filter_pred_fn_t select_comparison_fn(int op, int field_size, } static int init_pred(struct filter_parse_state *ps, - struct ftrace_event_call *call, + struct ftrace_event_field *field, struct filter_pred *pred) { - struct ftrace_event_field *field; filter_pred_fn_t fn = filter_pred_none; unsigned long long val; int ret; - field = find_event_field(call, pred->field_name); - if (!field) { - parse_error(ps, FILT_ERR_FIELD_NOT_FOUND, 0); - return -EINVAL; - } - pred->offset = field->offset; if (!is_legal_op(field, pred->op)) { @@ -1287,6 +1257,7 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps, struct ftrace_event_call *call, int op, char *operand1, char *operand2) { + struct ftrace_event_field *field; static struct filter_pred pred; memset(&pred, 0, sizeof(pred)); @@ -1300,14 +1271,16 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps, return NULL; } - pred.field_name = kstrdup(operand1, GFP_KERNEL); - if (!pred.field_name) + field = find_event_field(call, operand1); + if (!field) { + parse_error(ps, FILT_ERR_FIELD_NOT_FOUND, 0); return NULL; + } strcpy(pred.regex.pattern, operand2); pred.regex.len = strlen(pred.regex.pattern); - return init_pred(ps, call, &pred) ? NULL : &pred; + return init_pred(ps, field, &pred) ? NULL : &pred; } static int check_preds(struct filter_parse_state *ps) @@ -1618,18 +1591,16 @@ static int replace_preds(struct ftrace_event_call *call, pred = create_pred(ps, call, elt->op, operand1, operand2); if (!pred) { - err = -ENOMEM; + err = -EINVAL; goto fail; } + if (!dry_run) { err = filter_add_pred(ps, filter, pred, &stack); - if (err) { - filter_free_pred(pred); + if (err) goto fail; - } } - filter_free_pred(pred); operand1 = operand2 = NULL; } -- cgit v1.3-8-gc7d7 From 3f78f935e7fec3067b7990f9e4d324e1de70f79c Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:48 +0200 Subject: tracing/filter: Simplify tracepoint event lookup We dont need to perform lookup through the ftrace_events list, instead we can use the 'tp_event' field. Each perf_event contains tracepoint event field 'tp_event', which got initialized during the tracepoint event initialization. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-5-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 97b93f31884b..0948905dd39c 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1894,17 +1894,14 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id, int err; struct event_filter *filter; struct filter_parse_state *ps; - struct ftrace_event_call *call = NULL; + struct ftrace_event_call *call; mutex_lock(&event_mutex); - list_for_each_entry(call, &ftrace_events, list) { - if (call->event.type == event_id) - break; - } + call = event->tp_event; err = -EINVAL; - if (&call->list == &ftrace_events) + if (!call) goto out_unlock; err = -EEXIST; -- cgit v1.3-8-gc7d7 From f03f5979945c573801c25ba3089ef17c4d7edc61 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:49 +0200 Subject: tracing/filter: Unify predicate tree walking, change check_pred_tree function to use it Adding walk_pred_tree function to be used for walking throught the filter predicates. For each predicate the callback function is called, allowing users to add their own functionality or customize their way through the filter predicates. Changing check_pred_tree function to use walk_pred_tree. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-6-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 137 +++++++++++++++++++++++-------------- 1 file changed, 86 insertions(+), 51 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 0948905dd39c..5b889d43d856 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -381,6 +381,63 @@ get_pred_parent(struct filter_pred *pred, struct filter_pred *preds, return pred; } +enum walk_return { + WALK_PRED_ABORT, + WALK_PRED_PARENT, + WALK_PRED_DEFAULT, +}; + +typedef int (*filter_pred_walkcb_t) (enum move_type move, + struct filter_pred *pred, + int *err, void *data); + +static int walk_pred_tree(struct filter_pred *preds, + struct filter_pred *root, + filter_pred_walkcb_t cb, void *data) +{ + struct filter_pred *pred = root; + enum move_type move = MOVE_DOWN; + int done = 0; + + if (!preds) + return -EINVAL; + + do { + int err = 0, ret; + + ret = cb(move, pred, &err, data); + if (ret == WALK_PRED_ABORT) + return err; + if (ret == WALK_PRED_PARENT) + goto get_parent; + + switch (move) { + case MOVE_DOWN: + if (pred->left != FILTER_PRED_INVALID) { + pred = &preds[pred->left]; + continue; + } + goto get_parent; + case MOVE_UP_FROM_LEFT: + pred = &preds[pred->right]; + move = MOVE_DOWN; + continue; + case MOVE_UP_FROM_RIGHT: + get_parent: + if (pred == root) + break; + pred = get_pred_parent(pred, preds, + pred->parent, + &move); + continue; + } + done = 1; + } while (!done); + + /* We are fine. */ + return 0; +} + /* * A series of AND or ORs where found together. Instead of * climbing up and down the tree branches, an array of the @@ -1321,6 +1378,23 @@ static int count_preds(struct filter_parse_state *ps) return n_preds; } +struct check_pred_data { + int count; + int max; +}; + +static int check_pred_tree_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) +{ + struct check_pred_data *d = data; + + if (WARN_ON(d->count++ > d->max)) { + *err = -EINVAL; + return WALK_PRED_ABORT; + } + return WALK_PRED_DEFAULT; +} + /* * The tree is walked at filtering of an event. If the tree is not correctly * built, it may cause an infinite loop. Check here that the tree does @@ -1329,58 +1403,19 @@ static int count_preds(struct filter_parse_state *ps) static int check_pred_tree(struct event_filter *filter, struct filter_pred *root) { - struct filter_pred *preds; - struct filter_pred *pred; - enum move_type move = MOVE_DOWN; - int count = 0; - int done = 0; - int max; - - /* - * The max that we can hit a node is three times. - * Once going down, once coming up from left, and - * once coming up from right. This is more than enough - * since leafs are only hit a single time. - */ - max = 3 * filter->n_preds; - - preds = filter->preds; - if (!preds) - return -EINVAL; - pred = root; - - do { - if (WARN_ON(count++ > max)) - return -EINVAL; - - switch (move) { - case MOVE_DOWN: - if (pred->left != FILTER_PRED_INVALID) { - pred = &preds[pred->left]; - continue; - } - /* A leaf at the root is just a leaf in the tree */ - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - case MOVE_UP_FROM_LEFT: - pred = &preds[pred->right]; - move = MOVE_DOWN; - continue; - case MOVE_UP_FROM_RIGHT: - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - done = 1; - } while (!done); + struct check_pred_data data = { + /* + * The max that we can hit a node is three times. + * Once going down, once coming up from left, and + * once coming up from right. This is more than enough + * since leafs are only hit a single time. + */ + .max = 3 * filter->n_preds, + .count = 0, + }; - /* We are fine. */ - return 0; + return walk_pred_tree(filter->preds, root, + check_pred_tree_cb, &data); } static int count_leafs(struct filter_pred *preds, struct filter_pred *root) -- cgit v1.3-8-gc7d7 From c00b060f36e1238816ebcf2c8cccd5e9fa068980 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:50 +0200 Subject: tracing/filter: Change count_leafs function to use walk_pred_tree Changing count_leafs function to use unified predicates tree processing. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-7-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 47 ++++++++++++-------------------------- 1 file changed, 14 insertions(+), 33 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 5b889d43d856..ebbb2611982e 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1418,43 +1418,24 @@ static int check_pred_tree(struct event_filter *filter, check_pred_tree_cb, &data); } -static int count_leafs(struct filter_pred *preds, struct filter_pred *root) +static int count_leafs_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) { - struct filter_pred *pred; - enum move_type move = MOVE_DOWN; - int count = 0; - int done = 0; + int *count = data; - pred = root; + if ((move == MOVE_DOWN) && + (pred->left == FILTER_PRED_INVALID)) + (*count)++; - do { - switch (move) { - case MOVE_DOWN: - if (pred->left != FILTER_PRED_INVALID) { - pred = &preds[pred->left]; - continue; - } - /* A leaf at the root is just a leaf in the tree */ - if (pred == root) - return 1; - count++; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - case MOVE_UP_FROM_LEFT: - pred = &preds[pred->right]; - move = MOVE_DOWN; - continue; - case MOVE_UP_FROM_RIGHT: - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - done = 1; - } while (!done); + return WALK_PRED_DEFAULT; +} + +static int count_leafs(struct filter_pred *preds, struct filter_pred *root) +{ + int count = 0, ret; + ret = walk_pred_tree(preds, root, count_leafs_cb, &count); + WARN_ON(ret); return count; } -- cgit v1.3-8-gc7d7 From 1b797fe5aaac11e60fce1592119d0517e95aba95 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:51 +0200 Subject: tracing/filter: Change fold_pred_tree function to use walk_pred_tree Changing fold_pred_tree function to use unified predicates tree processing. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-8-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 65 ++++++++++++-------------------------- 1 file changed, 20 insertions(+), 45 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index ebbb2611982e..d8aa100cb22e 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1496,6 +1496,24 @@ static int fold_pred(struct filter_pred *preds, struct filter_pred *root) return 0; } +static int fold_pred_tree_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) +{ + struct filter_pred *preds = data; + + if (move != MOVE_DOWN) + return WALK_PRED_DEFAULT; + if (!(pred->index & FILTER_PRED_FOLD)) + return WALK_PRED_DEFAULT; + + *err = fold_pred(preds, pred); + if (*err) + return WALK_PRED_ABORT; + + /* eveyrhing below is folded, continue with parent */ + return WALK_PRED_PARENT; +} + /* * To optimize the processing of the ops, if we have several "ors" or * "ands" together, we can put them in an array and process them all @@ -1504,51 +1522,8 @@ static int fold_pred(struct filter_pred *preds, struct filter_pred *root) static int fold_pred_tree(struct event_filter *filter, struct filter_pred *root) { - struct filter_pred *preds; - struct filter_pred *pred; - enum move_type move = MOVE_DOWN; - int done = 0; - int err; - - preds = filter->preds; - if (!preds) - return -EINVAL; - pred = root; - - do { - switch (move) { - case MOVE_DOWN: - if (pred->index & FILTER_PRED_FOLD) { - err = fold_pred(preds, pred); - if (err) - return err; - /* Folded nodes are like leafs */ - } else if (pred->left != FILTER_PRED_INVALID) { - pred = &preds[pred->left]; - continue; - } - - /* A leaf at the root is just a leaf in the tree */ - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - case MOVE_UP_FROM_LEFT: - pred = &preds[pred->right]; - move = MOVE_DOWN; - continue; - case MOVE_UP_FROM_RIGHT: - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - done = 1; - } while (!done); - - return 0; + return walk_pred_tree(filter->preds, root, fold_pred_tree_cb, + filter->preds); } static int replace_preds(struct ftrace_event_call *call, -- cgit v1.3-8-gc7d7 From 96bc293a97e7f1651977976be7dd42031f6d8ea3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:52 +0200 Subject: tracing/filter: Change fold_pred function to use walk_pred_tree Changing fold_pred_tree function to use unified predicates tree processing. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-9-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 68 ++++++++++++++++++-------------------- 1 file changed, 33 insertions(+), 35 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index d8aa100cb22e..f44e68b89f15 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1439,13 +1439,40 @@ static int count_leafs(struct filter_pred *preds, struct filter_pred *root) return count; } +struct fold_pred_data { + struct filter_pred *root; + int count; + int children; +}; + +static int fold_pred_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) +{ + struct fold_pred_data *d = data; + struct filter_pred *root = d->root; + + if (move != MOVE_DOWN) + return WALK_PRED_DEFAULT; + if (pred->left != FILTER_PRED_INVALID) + return WALK_PRED_DEFAULT; + + if (WARN_ON(d->count == d->children)) { + *err = -EINVAL; + return WALK_PRED_ABORT; + } + + pred->index &= ~FILTER_PRED_FOLD; + root->ops[d->count++] = pred->index; + return WALK_PRED_DEFAULT; +} + static int fold_pred(struct filter_pred *preds, struct filter_pred *root) { - struct filter_pred *pred; - enum move_type move = MOVE_DOWN; - int count = 0; + struct fold_pred_data data = { + .root = root, + .count = 0, + }; int children; - int done = 0; /* No need to keep the fold flag */ root->index &= ~FILTER_PRED_FOLD; @@ -1463,37 +1490,8 @@ static int fold_pred(struct filter_pred *preds, struct filter_pred *root) return -ENOMEM; root->val = children; - - pred = root; - do { - switch (move) { - case MOVE_DOWN: - if (pred->left != FILTER_PRED_INVALID) { - pred = &preds[pred->left]; - continue; - } - if (WARN_ON(count == children)) - return -EINVAL; - pred->index &= ~FILTER_PRED_FOLD; - root->ops[count++] = pred->index; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - case MOVE_UP_FROM_LEFT: - pred = &preds[pred->right]; - move = MOVE_DOWN; - continue; - case MOVE_UP_FROM_RIGHT: - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - done = 1; - } while (!done); - - return 0; + data.children = children; + return walk_pred_tree(preds, root, fold_pred_cb, &data); } static int fold_pred_tree_cb(enum move_type move, struct filter_pred *pred, -- cgit v1.3-8-gc7d7 From f30120fce1efaa426f340a354d5ace36dab59f0e Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:53 +0200 Subject: tracing/filter: Change filter_match_preds function to use walk_pred_tree Changing filter_match_preds function to use unified predicates tree processing. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-10-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 124 +++++++++++++++++-------------------- 1 file changed, 58 insertions(+), 66 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index f44e68b89f15..319c3cac7d95 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -467,99 +467,91 @@ static int process_ops(struct filter_pred *preds, for (i = 0; i < op->val; i++) { pred = &preds[op->ops[i]]; - match = pred->fn(pred, rec); + if (!WARN_ON_ONCE(!pred->fn)) + match = pred->fn(pred, rec); if (!!match == type) return match; } return match; } +struct filter_match_preds_data { + struct filter_pred *preds; + int match; + void *rec; +}; + +static int filter_match_preds_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) +{ + struct filter_match_preds_data *d = data; + + *err = 0; + switch (move) { + case MOVE_DOWN: + /* only AND and OR have children */ + if (pred->left != FILTER_PRED_INVALID) { + /* If ops is set, then it was folded. */ + if (!pred->ops) + return WALK_PRED_DEFAULT; + /* We can treat folded ops as a leaf node */ + d->match = process_ops(d->preds, pred, d->rec); + } else { + if (!WARN_ON_ONCE(!pred->fn)) + d->match = pred->fn(pred, d->rec); + } + + return WALK_PRED_PARENT; + case MOVE_UP_FROM_LEFT: + /* + * Check for short circuits. + * + * Optimization: !!match == (pred->op == OP_OR) + * is the same as: + * if ((match && pred->op == OP_OR) || + * (!match && pred->op == OP_AND)) + */ + if (!!d->match == (pred->op == OP_OR)) + return WALK_PRED_PARENT; + break; + case MOVE_UP_FROM_RIGHT: + break; + } + + return WALK_PRED_DEFAULT; +} + /* return 1 if event matches, 0 otherwise (discard) */ int filter_match_preds(struct event_filter *filter, void *rec) { - int match = -1; - enum move_type move = MOVE_DOWN; struct filter_pred *preds; - struct filter_pred *pred; struct filter_pred *root; - int n_preds; - int done = 0; + struct filter_match_preds_data data = { + /* match is currently meaningless */ + .match = -1, + .rec = rec, + }; + int n_preds, ret; /* no filter is considered a match */ if (!filter) return 1; n_preds = filter->n_preds; - if (!n_preds) return 1; /* * n_preds, root and filter->preds are protect with preemption disabled. */ - preds = rcu_dereference_sched(filter->preds); root = rcu_dereference_sched(filter->root); if (!root) return 1; - pred = root; - - /* match is currently meaningless */ - match = -1; - - do { - switch (move) { - case MOVE_DOWN: - /* only AND and OR have children */ - if (pred->left != FILTER_PRED_INVALID) { - /* If ops is set, then it was folded. */ - if (!pred->ops) { - /* keep going to down the left side */ - pred = &preds[pred->left]; - continue; - } - /* We can treat folded ops as a leaf node */ - match = process_ops(preds, pred, rec); - } else - match = pred->fn(pred, rec); - /* If this pred is the only pred */ - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - case MOVE_UP_FROM_LEFT: - /* - * Check for short circuits. - * - * Optimization: !!match == (pred->op == OP_OR) - * is the same as: - * if ((match && pred->op == OP_OR) || - * (!match && pred->op == OP_AND)) - */ - if (!!match == (pred->op == OP_OR)) { - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - /* now go down the right side of the tree. */ - pred = &preds[pred->right]; - move = MOVE_DOWN; - continue; - case MOVE_UP_FROM_RIGHT: - /* We finished this equation. */ - if (pred == root) - break; - pred = get_pred_parent(pred, preds, - pred->parent, &move); - continue; - } - done = 1; - } while (!done); - - return match; + data.preds = preds = rcu_dereference_sched(filter->preds); + ret = walk_pred_tree(preds, root, filter_match_preds_cb, &data); + WARN_ON(ret); + return data.match; } EXPORT_SYMBOL_GPL(filter_match_preds); -- cgit v1.3-8-gc7d7 From 1d0e78e380cd2802aa603a50e08220dfc681141c Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 11 Aug 2011 16:25:54 +0200 Subject: tracing/filter: Add startup tests for events filter Adding automated tests running as late_initcall. Tests are compiled in with CONFIG_FTRACE_STARTUP_TEST option. Adding test event "ftrace_test_filter" used to simulate filter processing during event occurance. String filters are compiled and tested against several test events with different values. Also testing that evaluation of explicit predicates is ommited due to the lazy filter evaluation. Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/r/1313072754-4620-11-git-send-email-jolsa@redhat.com Signed-off-by: Steven Rostedt --- kernel/trace/Makefile | 2 + kernel/trace/trace.h | 3 + kernel/trace/trace_events_filter.c | 209 ++++++++++++++++++++++++++++++++ kernel/trace/trace_events_filter_test.h | 50 ++++++++ 4 files changed, 264 insertions(+) create mode 100644 kernel/trace/trace_events_filter_test.h (limited to 'kernel') diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 761c510a06c5..b384ed512bac 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -15,6 +15,8 @@ ifdef CONFIG_TRACING_BRANCHES KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING endif +CFLAGS_trace_events_filter.o := -I$(src) + # # Make the trace clocks available generally: it's infrastructure # relied on by ptrace for example: diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 2eb3cf6d37bc..4c7540ad5dcb 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -762,6 +762,9 @@ struct filter_pred { u64 val; struct regex regex; unsigned short *ops; +#ifdef CONFIG_FTRACE_STARTUP_TEST + struct ftrace_event_field *field; +#endif int offset; int not; int op; diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 319c3cac7d95..6a642e278241 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1329,6 +1329,9 @@ static struct filter_pred *create_pred(struct filter_parse_state *ps, strcpy(pred.regex.pattern, operand2); pred.regex.len = strlen(pred.regex.pattern); +#ifdef CONFIG_FTRACE_STARTUP_TEST + pred.field = field; +#endif return init_pred(ps, field, &pred) ? NULL : &pred; } @@ -1926,3 +1929,209 @@ out_unlock: #endif /* CONFIG_PERF_EVENTS */ +#ifdef CONFIG_FTRACE_STARTUP_TEST + +#include +#include + +#define CREATE_TRACE_POINTS +#include "trace_events_filter_test.h" + +static int test_get_filter(char *filter_str, struct ftrace_event_call *call, + struct event_filter **pfilter) +{ + struct event_filter *filter; + struct filter_parse_state *ps; + int err = -ENOMEM; + + filter = __alloc_filter(); + if (!filter) + goto out; + + ps = kzalloc(sizeof(*ps), GFP_KERNEL); + if (!ps) + goto free_filter; + + parse_init(ps, filter_ops, filter_str); + err = filter_parse(ps); + if (err) + goto free_ps; + + err = replace_preds(call, filter, ps, filter_str, false); + if (!err) + *pfilter = filter; + + free_ps: + filter_opstack_clear(ps); + postfix_clear(ps); + kfree(ps); + + free_filter: + if (err) + __free_filter(filter); + + out: + return err; +} + +#define DATA_REC(m, va, vb, vc, vd, ve, vf, vg, vh, nvisit) \ +{ \ + .filter = FILTER, \ + .rec = { .a = va, .b = vb, .c = vc, .d = vd, \ + .e = ve, .f = vf, .g = vg, .h = vh }, \ + .match = m, \ + .not_visited = nvisit, \ +} +#define YES 1 +#define NO 0 + +static struct test_filter_data_t { + char *filter; + struct ftrace_raw_ftrace_test_filter rec; + int match; + char *not_visited; +} test_filter_data[] = { +#define FILTER "a == 1 && b == 1 && c == 1 && d == 1 && " \ + "e == 1 && f == 1 && g == 1 && h == 1" + DATA_REC(YES, 1, 1, 1, 1, 1, 1, 1, 1, ""), + DATA_REC(NO, 0, 1, 1, 1, 1, 1, 1, 1, "bcdefgh"), + DATA_REC(NO, 1, 1, 1, 1, 1, 1, 1, 0, ""), +#undef FILTER +#define FILTER "a == 1 || b == 1 || c == 1 || d == 1 || " \ + "e == 1 || f == 1 || g == 1 || h == 1" + DATA_REC(NO, 0, 0, 0, 0, 0, 0, 0, 0, ""), + DATA_REC(YES, 0, 0, 0, 0, 0, 0, 0, 1, ""), + DATA_REC(YES, 1, 0, 0, 0, 0, 0, 0, 0, "bcdefgh"), +#undef FILTER +#define FILTER "(a == 1 || b == 1) && (c == 1 || d == 1) && " \ + "(e == 1 || f == 1) && (g == 1 || h == 1)" + DATA_REC(NO, 0, 0, 1, 1, 1, 1, 1, 1, "dfh"), + DATA_REC(YES, 0, 1, 0, 1, 0, 1, 0, 1, ""), + DATA_REC(YES, 1, 0, 1, 0, 0, 1, 0, 1, "bd"), + DATA_REC(NO, 1, 0, 1, 0, 0, 1, 0, 0, "bd"), +#undef FILTER +#define FILTER "(a == 1 && b == 1) || (c == 1 && d == 1) || " \ + "(e == 1 && f == 1) || (g == 1 && h == 1)" + DATA_REC(YES, 1, 0, 1, 1, 1, 1, 1, 1, "efgh"), + DATA_REC(YES, 0, 0, 0, 0, 0, 0, 1, 1, ""), + DATA_REC(NO, 0, 0, 0, 0, 0, 0, 0, 1, ""), +#undef FILTER +#define FILTER "(a == 1 && b == 1) && (c == 1 && d == 1) && " \ + "(e == 1 && f == 1) || (g == 1 && h == 1)" + DATA_REC(YES, 1, 1, 1, 1, 1, 1, 0, 0, "gh"), + DATA_REC(NO, 0, 0, 0, 0, 0, 0, 0, 1, ""), + DATA_REC(YES, 1, 1, 1, 1, 1, 0, 1, 1, ""), +#undef FILTER +#define FILTER "((a == 1 || b == 1) || (c == 1 || d == 1) || " \ + "(e == 1 || f == 1)) && (g == 1 || h == 1)" + DATA_REC(YES, 1, 1, 1, 1, 1, 1, 0, 1, "bcdef"), + DATA_REC(NO, 0, 0, 0, 0, 0, 0, 0, 0, ""), + DATA_REC(YES, 1, 1, 1, 1, 1, 0, 1, 1, "h"), +#undef FILTER +#define FILTER "((((((((a == 1) && (b == 1)) || (c == 1)) && (d == 1)) || " \ + "(e == 1)) && (f == 1)) || (g == 1)) && (h == 1))" + DATA_REC(YES, 1, 1, 1, 1, 1, 1, 1, 1, "ceg"), + DATA_REC(NO, 0, 1, 0, 1, 0, 1, 0, 1, ""), + DATA_REC(NO, 1, 0, 1, 0, 1, 0, 1, 0, ""), +#undef FILTER +#define FILTER "((((((((a == 1) || (b == 1)) && (c == 1)) || (d == 1)) && " \ + "(e == 1)) || (f == 1)) && (g == 1)) || (h == 1))" + DATA_REC(YES, 1, 1, 1, 1, 1, 1, 1, 1, "bdfh"), + DATA_REC(YES, 0, 1, 0, 1, 0, 1, 0, 1, ""), + DATA_REC(YES, 1, 0, 1, 0, 1, 0, 1, 0, "bdfh"), +}; + +#undef DATA_REC +#undef FILTER +#undef YES +#undef NO + +#define DATA_CNT (sizeof(test_filter_data)/sizeof(struct test_filter_data_t)) + +static int test_pred_visited; + +static int test_pred_visited_fn(struct filter_pred *pred, void *event) +{ + struct ftrace_event_field *field = pred->field; + + test_pred_visited = 1; + printk(KERN_INFO "\npred visited %s\n", field->name); + return 1; +} + +static int test_walk_pred_cb(enum move_type move, struct filter_pred *pred, + int *err, void *data) +{ + char *fields = data; + + if ((move == MOVE_DOWN) && + (pred->left == FILTER_PRED_INVALID)) { + struct ftrace_event_field *field = pred->field; + + if (!field) { + WARN(1, "all leafs should have field defined"); + return WALK_PRED_DEFAULT; + } + if (!strchr(fields, *field->name)) + return WALK_PRED_DEFAULT; + + WARN_ON(!pred->fn); + pred->fn = test_pred_visited_fn; + } + return WALK_PRED_DEFAULT; +} + +static __init int ftrace_test_event_filter(void) +{ + int i; + + printk(KERN_INFO "Testing ftrace filter: "); + + for (i = 0; i < DATA_CNT; i++) { + struct event_filter *filter = NULL; + struct test_filter_data_t *d = &test_filter_data[i]; + int err; + + err = test_get_filter(d->filter, &event_ftrace_test_filter, + &filter); + if (err) { + printk(KERN_INFO + "Failed to get filter for '%s', err %d\n", + d->filter, err); + break; + } + + if (*d->not_visited) + walk_pred_tree(filter->preds, filter->root, + test_walk_pred_cb, + d->not_visited); + + test_pred_visited = 0; + err = filter_match_preds(filter, &d->rec); + + __free_filter(filter); + + if (test_pred_visited) { + printk(KERN_INFO + "Failed, unwanted pred visited for filter %s\n", + d->filter); + break; + } + + if (err != d->match) { + printk(KERN_INFO + "Failed to match filter '%s', expected %d\n", + d->filter, d->match); + break; + } + } + + if (i == DATA_CNT) + printk(KERN_CONT "OK\n"); + + return 0; +} + +late_initcall(ftrace_test_event_filter); + +#endif /* CONFIG_FTRACE_STARTUP_TEST */ diff --git a/kernel/trace/trace_events_filter_test.h b/kernel/trace/trace_events_filter_test.h new file mode 100644 index 000000000000..bfd4dba0d603 --- /dev/null +++ b/kernel/trace/trace_events_filter_test.h @@ -0,0 +1,50 @@ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM test + +#if !defined(_TRACE_TEST_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_TEST_H + +#include + +TRACE_EVENT(ftrace_test_filter, + + TP_PROTO(int a, int b, int c, int d, int e, int f, int g, int h), + + TP_ARGS(a, b, c, d, e, f, g, h), + + TP_STRUCT__entry( + __field(int, a) + __field(int, b) + __field(int, c) + __field(int, d) + __field(int, e) + __field(int, f) + __field(int, g) + __field(int, h) + ), + + TP_fast_assign( + __entry->a = a; + __entry->b = b; + __entry->c = c; + __entry->d = d; + __entry->e = e; + __entry->f = f; + __entry->g = g; + __entry->h = h; + ), + + TP_printk("a %d, b %d, c %d, d %d, e %d, f %d, g %d, h %d", + __entry->a, __entry->b, __entry->c, __entry->d, + __entry->e, __entry->f, __entry->g, __entry->h) +); + +#endif /* _TRACE_TEST_H || TRACE_HEADER_MULTI_READ */ + +#undef TRACE_INCLUDE_PATH +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace_events_filter_test + +/* This part must be outside protection */ +#include -- cgit v1.3-8-gc7d7 From 69dd3d8e29e294caaf63eb5e8a72d250279f9e5f Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 23 Aug 2011 10:36:51 -0700 Subject: Revert "irq: Always set IRQF_ONESHOT if no primary handler is specified" This reverts commit f3637a5f2e2eb391ff5757bc83fb5de8f9726464. It turns out that this breaks several drivers, one example being OMAP boards which use the on-board OMAP UARTs and the omap-serial driver that will not boot to userspace after the commit. Paul Walmsley reports that enabling CONFIG_DEBUG_SHIRQ reveals 'IRQ handler type mismatch' errors: IRQ handler type mismatch for IRQ 74 current handler: serial idle ... and the reason is that setting IRQF_ONESHOT will now result in those interrupt handlers having different IRQF flags, and thus being unsharable. So the commit log in the reverted commit: "Since it is required for those users and there is no difference for others it makes sense to add this flag unconditionally." is simply not true: there may not be any difference from a "actions at irq time", but there is a *big* difference wrt this flag testing irq management (see __setup_irq() in kernel/irq/manage.c). One solution may be to stop verifying IRQF_ONESHOT in __setup_irq(), but right now the safe course of action is to revert the change. Let's revisit this in a later merge window. Reported-by: Paul Walmsley Cc: Sebastian Andrzej Siewior Requested-by: Alan Cox Acked-by: Thomas Gleixner Signed-off-by: Linus Torvalds --- kernel/irq/manage.c | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 2e9425889fa8..9b956fa20308 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1331,7 +1331,6 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, if (!thread_fn) return -EINVAL; handler = irq_default_primary_handler; - irqflags |= IRQF_ONESHOT; } action = kzalloc(sizeof(struct irqaction), GFP_KERNEL); -- cgit v1.3-8-gc7d7 From e8db0be1245de16a6cc6365506abc392c3c212d4 Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:03 +0200 Subject: PM QoS: Move and rename the implementation files The PM QoS implementation files are better named kernel/power/qos.c and include/linux/pm_qos.h. The PM QoS support is compiled under the CONFIG_PM option. Signed-off-by: Jean Pihet Acked-by: markgross Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- arch/arm/mach-msm/clock.c | 2 +- drivers/acpi/processor_idle.c | 2 +- drivers/cpuidle/cpuidle.c | 2 +- drivers/cpuidle/governors/ladder.c | 2 +- drivers/cpuidle/governors/menu.c | 2 +- drivers/media/video/via-camera.c | 2 +- drivers/net/e1000e/netdev.c | 2 +- drivers/net/wireless/ipw2x00/ipw2100.c | 2 +- include/linux/netdevice.h | 2 +- include/linux/pm_qos.h | 61 +++++ include/linux/pm_qos_params.h | 38 --- include/sound/pcm.h | 2 +- kernel/Makefile | 2 +- kernel/pm_qos_params.c | 481 --------------------------------- kernel/power/Makefile | 2 +- kernel/power/qos.c | 481 +++++++++++++++++++++++++++++++++ net/mac80211/main.c | 2 +- net/mac80211/mlme.c | 2 +- net/mac80211/scan.c | 2 +- sound/core/pcm_native.c | 2 +- 20 files changed, 558 insertions(+), 535 deletions(-) create mode 100644 include/linux/pm_qos.h delete mode 100644 include/linux/pm_qos_params.h delete mode 100644 kernel/pm_qos_params.c create mode 100644 kernel/power/qos.c (limited to 'kernel') diff --git a/arch/arm/mach-msm/clock.c b/arch/arm/mach-msm/clock.c index 22a537669624..d9145dfc2a3b 100644 --- a/arch/arm/mach-msm/clock.c +++ b/arch/arm/mach-msm/clock.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index 431ab11c8c1b..2e69e09ff03e 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -37,7 +37,7 @@ #include #include #include /* need_resched() */ -#include +#include #include #include #include diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index d4c542372886..0df014110097 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/cpuidle/governors/ladder.c b/drivers/cpuidle/governors/ladder.c index 12c98900dcf8..f62fde21e962 100644 --- a/drivers/cpuidle/governors/ladder.c +++ b/drivers/cpuidle/governors/ladder.c @@ -14,7 +14,7 @@ #include #include -#include +#include #include #include diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index c47f3d09c1ee..3600f1955e48 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -12,7 +12,7 @@ #include #include -#include +#include #include #include #include diff --git a/drivers/media/video/via-camera.c b/drivers/media/video/via-camera.c index 85d3048c1d67..b3ca3893f3de 100644 --- a/drivers/media/video/via-camera.c +++ b/drivers/media/video/via-camera.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c index 2198e615f241..07031af38964 100644 --- a/drivers/net/e1000e/netdev.c +++ b/drivers/net/e1000e/netdev.c @@ -47,7 +47,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/net/wireless/ipw2x00/ipw2100.c b/drivers/net/wireless/ipw2x00/ipw2100.c index 3774dd034746..aaab76ce6020 100644 --- a/drivers/net/wireless/ipw2x00/ipw2100.c +++ b/drivers/net/wireless/ipw2x00/ipw2100.c @@ -161,7 +161,7 @@ that only one external action is invoked at a time. #include #include #include -#include +#include #include diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ddee79bb8f15..f72ac6b972b7 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -31,7 +31,7 @@ #include #ifdef __KERNEL__ -#include +#include #include #include #include diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h new file mode 100644 index 000000000000..7ba675413b08 --- /dev/null +++ b/include/linux/pm_qos.h @@ -0,0 +1,61 @@ +#ifndef _LINUX_PM_QOS_H +#define _LINUX_PM_QOS_H +/* interface for the pm_qos_power infrastructure of the linux kernel. + * + * Mark Gross + */ +#include +#include +#include + +#define PM_QOS_RESERVED 0 +#define PM_QOS_CPU_DMA_LATENCY 1 +#define PM_QOS_NETWORK_LATENCY 2 +#define PM_QOS_NETWORK_THROUGHPUT 3 + +#define PM_QOS_NUM_CLASSES 4 +#define PM_QOS_DEFAULT_VALUE -1 + +#define PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) +#define PM_QOS_NETWORK_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) +#define PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE 0 + +struct pm_qos_request_list { + struct plist_node list; + int pm_qos_class; +}; + +#ifdef CONFIG_PM +void pm_qos_add_request(struct pm_qos_request_list *l, + int pm_qos_class, s32 value); +void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, + s32 new_value); +void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req); + +int pm_qos_request(int pm_qos_class); +int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier); +int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier); +int pm_qos_request_active(struct pm_qos_request_list *req); +#else +static inline void pm_qos_add_request(struct pm_qos_request_list *l, + int pm_qos_class, s32 value) + { return; } +static inline void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, + s32 new_value) + { return; } +static inline void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) + { return; } + +static inline int pm_qos_request(int pm_qos_class) + { return 0; } +static inline int pm_qos_add_notifier(int pm_qos_class, + struct notifier_block *notifier) + { return 0; } +static inline int pm_qos_remove_notifier(int pm_qos_class, + struct notifier_block *notifier) + { return 0; } +static inline int pm_qos_request_active(struct pm_qos_request_list *req) + { return 0; } +#endif + +#endif diff --git a/include/linux/pm_qos_params.h b/include/linux/pm_qos_params.h deleted file mode 100644 index a7d87f911cab..000000000000 --- a/include/linux/pm_qos_params.h +++ /dev/null @@ -1,38 +0,0 @@ -#ifndef _LINUX_PM_QOS_PARAMS_H -#define _LINUX_PM_QOS_PARAMS_H -/* interface for the pm_qos_power infrastructure of the linux kernel. - * - * Mark Gross - */ -#include -#include -#include - -#define PM_QOS_RESERVED 0 -#define PM_QOS_CPU_DMA_LATENCY 1 -#define PM_QOS_NETWORK_LATENCY 2 -#define PM_QOS_NETWORK_THROUGHPUT 3 - -#define PM_QOS_NUM_CLASSES 4 -#define PM_QOS_DEFAULT_VALUE -1 - -#define PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) -#define PM_QOS_NETWORK_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) -#define PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE 0 - -struct pm_qos_request_list { - struct plist_node list; - int pm_qos_class; -}; - -void pm_qos_add_request(struct pm_qos_request_list *l, int pm_qos_class, s32 value); -void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, - s32 new_value); -void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req); - -int pm_qos_request(int pm_qos_class); -int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier); -int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier); -int pm_qos_request_active(struct pm_qos_request_list *req); - -#endif diff --git a/include/sound/pcm.h b/include/sound/pcm.h index 57e71fa33f7c..666ee91e8a2e 100644 --- a/include/sound/pcm.h +++ b/include/sound/pcm.h @@ -29,7 +29,7 @@ #include #include #include -#include +#include #define snd_pcm_substream_chip(substream) ((substream)->private_data) #define snd_pcm_chip(pcm) ((pcm)->private_data) diff --git a/kernel/Makefile b/kernel/Makefile index eca595e2fd52..2da48d3515eb 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -9,7 +9,7 @@ obj-y = sched.o fork.o exec_domain.o panic.o printk.o \ rcupdate.o extable.o params.o posix-timers.o \ kthread.o wait.o kfifo.o sys_ni.o posix-cpu-timers.o mutex.o \ hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \ - notifier.o ksysfs.o pm_qos_params.o sched_clock.o cred.o \ + notifier.o ksysfs.o sched_clock.o cred.o \ async.o range.o obj-y += groups.o diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c deleted file mode 100644 index 37f05d0f0793..000000000000 --- a/kernel/pm_qos_params.c +++ /dev/null @@ -1,481 +0,0 @@ -/* - * This module exposes the interface to kernel space for specifying - * QoS dependencies. It provides infrastructure for registration of: - * - * Dependents on a QoS value : register requests - * Watchers of QoS value : get notified when target QoS value changes - * - * This QoS design is best effort based. Dependents register their QoS needs. - * Watchers register to keep track of the current QoS needs of the system. - * - * There are 3 basic classes of QoS parameter: latency, timeout, throughput - * each have defined units: - * latency: usec - * timeout: usec <-- currently not used. - * throughput: kbs (kilo byte / sec) - * - * There are lists of pm_qos_objects each one wrapping requests, notifiers - * - * User mode requests on a QOS parameter register themselves to the - * subsystem by opening the device node /dev/... and writing there request to - * the node. As long as the process holds a file handle open to the node the - * client continues to be accounted for. Upon file release the usermode - * request is removed and a new qos target is computed. This way when the - * request that the application has is cleaned up when closes the file - * pointer or exits the pm_qos_object will get an opportunity to clean up. - * - * Mark Gross - */ - -/*#define DEBUG*/ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include - -/* - * locking rule: all changes to requests or notifiers lists - * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock - * held, taken with _irqsave. One lock to rule them all - */ -enum pm_qos_type { - PM_QOS_MAX, /* return the largest value */ - PM_QOS_MIN /* return the smallest value */ -}; - -/* - * Note: The lockless read path depends on the CPU accessing - * target_value atomically. Atomic access is only guaranteed on all CPU - * types linux supports for 32 bit quantites - */ -struct pm_qos_object { - struct plist_head requests; - struct blocking_notifier_head *notifiers; - struct miscdevice pm_qos_power_miscdev; - char *name; - s32 target_value; /* Do not change to 64 bit */ - s32 default_value; - enum pm_qos_type type; -}; - -static DEFINE_SPINLOCK(pm_qos_lock); - -static struct pm_qos_object null_pm_qos; -static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier); -static struct pm_qos_object cpu_dma_pm_qos = { - .requests = PLIST_HEAD_INIT(cpu_dma_pm_qos.requests), - .notifiers = &cpu_dma_lat_notifier, - .name = "cpu_dma_latency", - .target_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, - .default_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, - .type = PM_QOS_MIN, -}; - -static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); -static struct pm_qos_object network_lat_pm_qos = { - .requests = PLIST_HEAD_INIT(network_lat_pm_qos.requests), - .notifiers = &network_lat_notifier, - .name = "network_latency", - .target_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, - .default_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, - .type = PM_QOS_MIN -}; - - -static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier); -static struct pm_qos_object network_throughput_pm_qos = { - .requests = PLIST_HEAD_INIT(network_throughput_pm_qos.requests), - .notifiers = &network_throughput_notifier, - .name = "network_throughput", - .target_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, - .default_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, - .type = PM_QOS_MAX, -}; - - -static struct pm_qos_object *pm_qos_array[] = { - &null_pm_qos, - &cpu_dma_pm_qos, - &network_lat_pm_qos, - &network_throughput_pm_qos -}; - -static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, - size_t count, loff_t *f_pos); -static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, - size_t count, loff_t *f_pos); -static int pm_qos_power_open(struct inode *inode, struct file *filp); -static int pm_qos_power_release(struct inode *inode, struct file *filp); - -static const struct file_operations pm_qos_power_fops = { - .write = pm_qos_power_write, - .read = pm_qos_power_read, - .open = pm_qos_power_open, - .release = pm_qos_power_release, - .llseek = noop_llseek, -}; - -/* unlocked internal variant */ -static inline int pm_qos_get_value(struct pm_qos_object *o) -{ - if (plist_head_empty(&o->requests)) - return o->default_value; - - switch (o->type) { - case PM_QOS_MIN: - return plist_first(&o->requests)->prio; - - case PM_QOS_MAX: - return plist_last(&o->requests)->prio; - - default: - /* runtime check for not using enum */ - BUG(); - } -} - -static inline s32 pm_qos_read_value(struct pm_qos_object *o) -{ - return o->target_value; -} - -static inline void pm_qos_set_value(struct pm_qos_object *o, s32 value) -{ - o->target_value = value; -} - -static void update_target(struct pm_qos_object *o, struct plist_node *node, - int del, int value) -{ - unsigned long flags; - int prev_value, curr_value; - - spin_lock_irqsave(&pm_qos_lock, flags); - prev_value = pm_qos_get_value(o); - /* PM_QOS_DEFAULT_VALUE is a signal that the value is unchanged */ - if (value != PM_QOS_DEFAULT_VALUE) { - /* - * to change the list, we atomically remove, reinit - * with new value and add, then see if the extremal - * changed - */ - plist_del(node, &o->requests); - plist_node_init(node, value); - plist_add(node, &o->requests); - } else if (del) { - plist_del(node, &o->requests); - } else { - plist_add(node, &o->requests); - } - curr_value = pm_qos_get_value(o); - pm_qos_set_value(o, curr_value); - spin_unlock_irqrestore(&pm_qos_lock, flags); - - if (prev_value != curr_value) - blocking_notifier_call_chain(o->notifiers, - (unsigned long)curr_value, - NULL); -} - -static int register_pm_qos_misc(struct pm_qos_object *qos) -{ - qos->pm_qos_power_miscdev.minor = MISC_DYNAMIC_MINOR; - qos->pm_qos_power_miscdev.name = qos->name; - qos->pm_qos_power_miscdev.fops = &pm_qos_power_fops; - - return misc_register(&qos->pm_qos_power_miscdev); -} - -static int find_pm_qos_object_by_minor(int minor) -{ - int pm_qos_class; - - for (pm_qos_class = 0; - pm_qos_class < PM_QOS_NUM_CLASSES; pm_qos_class++) { - if (minor == - pm_qos_array[pm_qos_class]->pm_qos_power_miscdev.minor) - return pm_qos_class; - } - return -1; -} - -/** - * pm_qos_request - returns current system wide qos expectation - * @pm_qos_class: identification of which qos value is requested - * - * This function returns the current target value. - */ -int pm_qos_request(int pm_qos_class) -{ - return pm_qos_read_value(pm_qos_array[pm_qos_class]); -} -EXPORT_SYMBOL_GPL(pm_qos_request); - -int pm_qos_request_active(struct pm_qos_request_list *req) -{ - return req->pm_qos_class != 0; -} -EXPORT_SYMBOL_GPL(pm_qos_request_active); - -/** - * pm_qos_add_request - inserts new qos request into the list - * @dep: pointer to a preallocated handle - * @pm_qos_class: identifies which list of qos request to use - * @value: defines the qos request - * - * This function inserts a new entry in the pm_qos_class list of requested qos - * performance characteristics. It recomputes the aggregate QoS expectations - * for the pm_qos_class of parameters and initializes the pm_qos_request_list - * handle. Caller needs to save this handle for later use in updates and - * removal. - */ - -void pm_qos_add_request(struct pm_qos_request_list *dep, - int pm_qos_class, s32 value) -{ - struct pm_qos_object *o = pm_qos_array[pm_qos_class]; - int new_value; - - if (pm_qos_request_active(dep)) { - WARN(1, KERN_ERR "pm_qos_add_request() called for already added request\n"); - return; - } - if (value == PM_QOS_DEFAULT_VALUE) - new_value = o->default_value; - else - new_value = value; - plist_node_init(&dep->list, new_value); - dep->pm_qos_class = pm_qos_class; - update_target(o, &dep->list, 0, PM_QOS_DEFAULT_VALUE); -} -EXPORT_SYMBOL_GPL(pm_qos_add_request); - -/** - * pm_qos_update_request - modifies an existing qos request - * @pm_qos_req : handle to list element holding a pm_qos request to use - * @value: defines the qos request - * - * Updates an existing qos request for the pm_qos_class of parameters along - * with updating the target pm_qos_class value. - * - * Attempts are made to make this code callable on hot code paths. - */ -void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, - s32 new_value) -{ - s32 temp; - struct pm_qos_object *o; - - if (!pm_qos_req) /*guard against callers passing in null */ - return; - - if (!pm_qos_request_active(pm_qos_req)) { - WARN(1, KERN_ERR "pm_qos_update_request() called for unknown object\n"); - return; - } - - o = pm_qos_array[pm_qos_req->pm_qos_class]; - - if (new_value == PM_QOS_DEFAULT_VALUE) - temp = o->default_value; - else - temp = new_value; - - if (temp != pm_qos_req->list.prio) - update_target(o, &pm_qos_req->list, 0, temp); -} -EXPORT_SYMBOL_GPL(pm_qos_update_request); - -/** - * pm_qos_remove_request - modifies an existing qos request - * @pm_qos_req: handle to request list element - * - * Will remove pm qos request from the list of requests and - * recompute the current target value for the pm_qos_class. Call this - * on slow code paths. - */ -void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) -{ - struct pm_qos_object *o; - - if (pm_qos_req == NULL) - return; - /* silent return to keep pcm code cleaner */ - - if (!pm_qos_request_active(pm_qos_req)) { - WARN(1, KERN_ERR "pm_qos_remove_request() called for unknown object\n"); - return; - } - - o = pm_qos_array[pm_qos_req->pm_qos_class]; - update_target(o, &pm_qos_req->list, 1, PM_QOS_DEFAULT_VALUE); - memset(pm_qos_req, 0, sizeof(*pm_qos_req)); -} -EXPORT_SYMBOL_GPL(pm_qos_remove_request); - -/** - * pm_qos_add_notifier - sets notification entry for changes to target value - * @pm_qos_class: identifies which qos target changes should be notified. - * @notifier: notifier block managed by caller. - * - * will register the notifier into a notification chain that gets called - * upon changes to the pm_qos_class target value. - */ -int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier) -{ - int retval; - - retval = blocking_notifier_chain_register( - pm_qos_array[pm_qos_class]->notifiers, notifier); - - return retval; -} -EXPORT_SYMBOL_GPL(pm_qos_add_notifier); - -/** - * pm_qos_remove_notifier - deletes notification entry from chain. - * @pm_qos_class: identifies which qos target changes are notified. - * @notifier: notifier block to be removed. - * - * will remove the notifier from the notification chain that gets called - * upon changes to the pm_qos_class target value. - */ -int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) -{ - int retval; - - retval = blocking_notifier_chain_unregister( - pm_qos_array[pm_qos_class]->notifiers, notifier); - - return retval; -} -EXPORT_SYMBOL_GPL(pm_qos_remove_notifier); - -static int pm_qos_power_open(struct inode *inode, struct file *filp) -{ - long pm_qos_class; - - pm_qos_class = find_pm_qos_object_by_minor(iminor(inode)); - if (pm_qos_class >= 0) { - struct pm_qos_request_list *req = kzalloc(sizeof(*req), GFP_KERNEL); - if (!req) - return -ENOMEM; - - pm_qos_add_request(req, pm_qos_class, PM_QOS_DEFAULT_VALUE); - filp->private_data = req; - - if (filp->private_data) - return 0; - } - return -EPERM; -} - -static int pm_qos_power_release(struct inode *inode, struct file *filp) -{ - struct pm_qos_request_list *req; - - req = filp->private_data; - pm_qos_remove_request(req); - kfree(req); - - return 0; -} - - -static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, - size_t count, loff_t *f_pos) -{ - s32 value; - unsigned long flags; - struct pm_qos_object *o; - struct pm_qos_request_list *pm_qos_req = filp->private_data; - - if (!pm_qos_req) - return -EINVAL; - if (!pm_qos_request_active(pm_qos_req)) - return -EINVAL; - - o = pm_qos_array[pm_qos_req->pm_qos_class]; - spin_lock_irqsave(&pm_qos_lock, flags); - value = pm_qos_get_value(o); - spin_unlock_irqrestore(&pm_qos_lock, flags); - - return simple_read_from_buffer(buf, count, f_pos, &value, sizeof(s32)); -} - -static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, - size_t count, loff_t *f_pos) -{ - s32 value; - struct pm_qos_request_list *pm_qos_req; - - if (count == sizeof(s32)) { - if (copy_from_user(&value, buf, sizeof(s32))) - return -EFAULT; - } else if (count <= 11) { /* ASCII perhaps? */ - char ascii_value[11]; - unsigned long int ulval; - int ret; - - if (copy_from_user(ascii_value, buf, count)) - return -EFAULT; - - if (count > 10) { - if (ascii_value[10] == '\n') - ascii_value[10] = '\0'; - else - return -EINVAL; - } else { - ascii_value[count] = '\0'; - } - ret = strict_strtoul(ascii_value, 16, &ulval); - if (ret) { - pr_debug("%s, 0x%lx, 0x%x\n", ascii_value, ulval, ret); - return -EINVAL; - } - value = (s32)lower_32_bits(ulval); - } else { - return -EINVAL; - } - - pm_qos_req = filp->private_data; - pm_qos_update_request(pm_qos_req, value); - - return count; -} - - -static int __init pm_qos_power_init(void) -{ - int ret = 0; - - ret = register_pm_qos_misc(&cpu_dma_pm_qos); - if (ret < 0) { - printk(KERN_ERR "pm_qos_param: cpu_dma_latency setup failed\n"); - return ret; - } - ret = register_pm_qos_misc(&network_lat_pm_qos); - if (ret < 0) { - printk(KERN_ERR "pm_qos_param: network_latency setup failed\n"); - return ret; - } - ret = register_pm_qos_misc(&network_throughput_pm_qos); - if (ret < 0) - printk(KERN_ERR - "pm_qos_param: network_throughput setup failed\n"); - - return ret; -} - -late_initcall(pm_qos_power_init); diff --git a/kernel/power/Makefile b/kernel/power/Makefile index c5ebc6a90643..ad6bdd8b401a 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -1,7 +1,7 @@ ccflags-$(CONFIG_PM_DEBUG) := -DDEBUG -obj-$(CONFIG_PM) += main.o +obj-$(CONFIG_PM) += main.o qos.o obj-$(CONFIG_PM_SLEEP) += console.o obj-$(CONFIG_FREEZER) += process.o obj-$(CONFIG_SUSPEND) += suspend.o diff --git a/kernel/power/qos.c b/kernel/power/qos.c new file mode 100644 index 000000000000..61b47384329e --- /dev/null +++ b/kernel/power/qos.c @@ -0,0 +1,481 @@ +/* + * This module exposes the interface to kernel space for specifying + * QoS dependencies. It provides infrastructure for registration of: + * + * Dependents on a QoS value : register requests + * Watchers of QoS value : get notified when target QoS value changes + * + * This QoS design is best effort based. Dependents register their QoS needs. + * Watchers register to keep track of the current QoS needs of the system. + * + * There are 3 basic classes of QoS parameter: latency, timeout, throughput + * each have defined units: + * latency: usec + * timeout: usec <-- currently not used. + * throughput: kbs (kilo byte / sec) + * + * There are lists of pm_qos_objects each one wrapping requests, notifiers + * + * User mode requests on a QOS parameter register themselves to the + * subsystem by opening the device node /dev/... and writing there request to + * the node. As long as the process holds a file handle open to the node the + * client continues to be accounted for. Upon file release the usermode + * request is removed and a new qos target is computed. This way when the + * request that the application has is cleaned up when closes the file + * pointer or exits the pm_qos_object will get an opportunity to clean up. + * + * Mark Gross + */ + +/*#define DEBUG*/ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +/* + * locking rule: all changes to requests or notifiers lists + * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock + * held, taken with _irqsave. One lock to rule them all + */ +enum pm_qos_type { + PM_QOS_MAX, /* return the largest value */ + PM_QOS_MIN /* return the smallest value */ +}; + +/* + * Note: The lockless read path depends on the CPU accessing + * target_value atomically. Atomic access is only guaranteed on all CPU + * types linux supports for 32 bit quantites + */ +struct pm_qos_object { + struct plist_head requests; + struct blocking_notifier_head *notifiers; + struct miscdevice pm_qos_power_miscdev; + char *name; + s32 target_value; /* Do not change to 64 bit */ + s32 default_value; + enum pm_qos_type type; +}; + +static DEFINE_SPINLOCK(pm_qos_lock); + +static struct pm_qos_object null_pm_qos; +static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier); +static struct pm_qos_object cpu_dma_pm_qos = { + .requests = PLIST_HEAD_INIT(cpu_dma_pm_qos.requests), + .notifiers = &cpu_dma_lat_notifier, + .name = "cpu_dma_latency", + .target_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, + .default_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, + .type = PM_QOS_MIN, +}; + +static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); +static struct pm_qos_object network_lat_pm_qos = { + .requests = PLIST_HEAD_INIT(network_lat_pm_qos.requests), + .notifiers = &network_lat_notifier, + .name = "network_latency", + .target_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, + .default_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, + .type = PM_QOS_MIN +}; + + +static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier); +static struct pm_qos_object network_throughput_pm_qos = { + .requests = PLIST_HEAD_INIT(network_throughput_pm_qos.requests), + .notifiers = &network_throughput_notifier, + .name = "network_throughput", + .target_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, + .default_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, + .type = PM_QOS_MAX, +}; + + +static struct pm_qos_object *pm_qos_array[] = { + &null_pm_qos, + &cpu_dma_pm_qos, + &network_lat_pm_qos, + &network_throughput_pm_qos +}; + +static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, + size_t count, loff_t *f_pos); +static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, + size_t count, loff_t *f_pos); +static int pm_qos_power_open(struct inode *inode, struct file *filp); +static int pm_qos_power_release(struct inode *inode, struct file *filp); + +static const struct file_operations pm_qos_power_fops = { + .write = pm_qos_power_write, + .read = pm_qos_power_read, + .open = pm_qos_power_open, + .release = pm_qos_power_release, + .llseek = noop_llseek, +}; + +/* unlocked internal variant */ +static inline int pm_qos_get_value(struct pm_qos_object *o) +{ + if (plist_head_empty(&o->requests)) + return o->default_value; + + switch (o->type) { + case PM_QOS_MIN: + return plist_first(&o->requests)->prio; + + case PM_QOS_MAX: + return plist_last(&o->requests)->prio; + + default: + /* runtime check for not using enum */ + BUG(); + } +} + +static inline s32 pm_qos_read_value(struct pm_qos_object *o) +{ + return o->target_value; +} + +static inline void pm_qos_set_value(struct pm_qos_object *o, s32 value) +{ + o->target_value = value; +} + +static void update_target(struct pm_qos_object *o, struct plist_node *node, + int del, int value) +{ + unsigned long flags; + int prev_value, curr_value; + + spin_lock_irqsave(&pm_qos_lock, flags); + prev_value = pm_qos_get_value(o); + /* PM_QOS_DEFAULT_VALUE is a signal that the value is unchanged */ + if (value != PM_QOS_DEFAULT_VALUE) { + /* + * to change the list, we atomically remove, reinit + * with new value and add, then see if the extremal + * changed + */ + plist_del(node, &o->requests); + plist_node_init(node, value); + plist_add(node, &o->requests); + } else if (del) { + plist_del(node, &o->requests); + } else { + plist_add(node, &o->requests); + } + curr_value = pm_qos_get_value(o); + pm_qos_set_value(o, curr_value); + spin_unlock_irqrestore(&pm_qos_lock, flags); + + if (prev_value != curr_value) + blocking_notifier_call_chain(o->notifiers, + (unsigned long)curr_value, + NULL); +} + +static int register_pm_qos_misc(struct pm_qos_object *qos) +{ + qos->pm_qos_power_miscdev.minor = MISC_DYNAMIC_MINOR; + qos->pm_qos_power_miscdev.name = qos->name; + qos->pm_qos_power_miscdev.fops = &pm_qos_power_fops; + + return misc_register(&qos->pm_qos_power_miscdev); +} + +static int find_pm_qos_object_by_minor(int minor) +{ + int pm_qos_class; + + for (pm_qos_class = 0; + pm_qos_class < PM_QOS_NUM_CLASSES; pm_qos_class++) { + if (minor == + pm_qos_array[pm_qos_class]->pm_qos_power_miscdev.minor) + return pm_qos_class; + } + return -1; +} + +/** + * pm_qos_request - returns current system wide qos expectation + * @pm_qos_class: identification of which qos value is requested + * + * This function returns the current target value. + */ +int pm_qos_request(int pm_qos_class) +{ + return pm_qos_read_value(pm_qos_array[pm_qos_class]); +} +EXPORT_SYMBOL_GPL(pm_qos_request); + +int pm_qos_request_active(struct pm_qos_request_list *req) +{ + return req->pm_qos_class != 0; +} +EXPORT_SYMBOL_GPL(pm_qos_request_active); + +/** + * pm_qos_add_request - inserts new qos request into the list + * @dep: pointer to a preallocated handle + * @pm_qos_class: identifies which list of qos request to use + * @value: defines the qos request + * + * This function inserts a new entry in the pm_qos_class list of requested qos + * performance characteristics. It recomputes the aggregate QoS expectations + * for the pm_qos_class of parameters and initializes the pm_qos_request_list + * handle. Caller needs to save this handle for later use in updates and + * removal. + */ + +void pm_qos_add_request(struct pm_qos_request_list *dep, + int pm_qos_class, s32 value) +{ + struct pm_qos_object *o = pm_qos_array[pm_qos_class]; + int new_value; + + if (pm_qos_request_active(dep)) { + WARN(1, KERN_ERR "pm_qos_add_request() called for already added request\n"); + return; + } + if (value == PM_QOS_DEFAULT_VALUE) + new_value = o->default_value; + else + new_value = value; + plist_node_init(&dep->list, new_value); + dep->pm_qos_class = pm_qos_class; + update_target(o, &dep->list, 0, PM_QOS_DEFAULT_VALUE); +} +EXPORT_SYMBOL_GPL(pm_qos_add_request); + +/** + * pm_qos_update_request - modifies an existing qos request + * @pm_qos_req : handle to list element holding a pm_qos request to use + * @value: defines the qos request + * + * Updates an existing qos request for the pm_qos_class of parameters along + * with updating the target pm_qos_class value. + * + * Attempts are made to make this code callable on hot code paths. + */ +void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, + s32 new_value) +{ + s32 temp; + struct pm_qos_object *o; + + if (!pm_qos_req) /*guard against callers passing in null */ + return; + + if (!pm_qos_request_active(pm_qos_req)) { + WARN(1, KERN_ERR "pm_qos_update_request() called for unknown object\n"); + return; + } + + o = pm_qos_array[pm_qos_req->pm_qos_class]; + + if (new_value == PM_QOS_DEFAULT_VALUE) + temp = o->default_value; + else + temp = new_value; + + if (temp != pm_qos_req->list.prio) + update_target(o, &pm_qos_req->list, 0, temp); +} +EXPORT_SYMBOL_GPL(pm_qos_update_request); + +/** + * pm_qos_remove_request - modifies an existing qos request + * @pm_qos_req: handle to request list element + * + * Will remove pm qos request from the list of requests and + * recompute the current target value for the pm_qos_class. Call this + * on slow code paths. + */ +void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) +{ + struct pm_qos_object *o; + + if (pm_qos_req == NULL) + return; + /* silent return to keep pcm code cleaner */ + + if (!pm_qos_request_active(pm_qos_req)) { + WARN(1, KERN_ERR "pm_qos_remove_request() called for unknown object\n"); + return; + } + + o = pm_qos_array[pm_qos_req->pm_qos_class]; + update_target(o, &pm_qos_req->list, 1, PM_QOS_DEFAULT_VALUE); + memset(pm_qos_req, 0, sizeof(*pm_qos_req)); +} +EXPORT_SYMBOL_GPL(pm_qos_remove_request); + +/** + * pm_qos_add_notifier - sets notification entry for changes to target value + * @pm_qos_class: identifies which qos target changes should be notified. + * @notifier: notifier block managed by caller. + * + * will register the notifier into a notification chain that gets called + * upon changes to the pm_qos_class target value. + */ +int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier) +{ + int retval; + + retval = blocking_notifier_chain_register( + pm_qos_array[pm_qos_class]->notifiers, notifier); + + return retval; +} +EXPORT_SYMBOL_GPL(pm_qos_add_notifier); + +/** + * pm_qos_remove_notifier - deletes notification entry from chain. + * @pm_qos_class: identifies which qos target changes are notified. + * @notifier: notifier block to be removed. + * + * will remove the notifier from the notification chain that gets called + * upon changes to the pm_qos_class target value. + */ +int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) +{ + int retval; + + retval = blocking_notifier_chain_unregister( + pm_qos_array[pm_qos_class]->notifiers, notifier); + + return retval; +} +EXPORT_SYMBOL_GPL(pm_qos_remove_notifier); + +static int pm_qos_power_open(struct inode *inode, struct file *filp) +{ + long pm_qos_class; + + pm_qos_class = find_pm_qos_object_by_minor(iminor(inode)); + if (pm_qos_class >= 0) { + struct pm_qos_request_list *req = kzalloc(sizeof(*req), GFP_KERNEL); + if (!req) + return -ENOMEM; + + pm_qos_add_request(req, pm_qos_class, PM_QOS_DEFAULT_VALUE); + filp->private_data = req; + + if (filp->private_data) + return 0; + } + return -EPERM; +} + +static int pm_qos_power_release(struct inode *inode, struct file *filp) +{ + struct pm_qos_request_list *req; + + req = filp->private_data; + pm_qos_remove_request(req); + kfree(req); + + return 0; +} + + +static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, + size_t count, loff_t *f_pos) +{ + s32 value; + unsigned long flags; + struct pm_qos_object *o; + struct pm_qos_request_list *pm_qos_req = filp->private_data; + + if (!pm_qos_req) + return -EINVAL; + if (!pm_qos_request_active(pm_qos_req)) + return -EINVAL; + + o = pm_qos_array[pm_qos_req->pm_qos_class]; + spin_lock_irqsave(&pm_qos_lock, flags); + value = pm_qos_get_value(o); + spin_unlock_irqrestore(&pm_qos_lock, flags); + + return simple_read_from_buffer(buf, count, f_pos, &value, sizeof(s32)); +} + +static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, + size_t count, loff_t *f_pos) +{ + s32 value; + struct pm_qos_request_list *pm_qos_req; + + if (count == sizeof(s32)) { + if (copy_from_user(&value, buf, sizeof(s32))) + return -EFAULT; + } else if (count <= 11) { /* ASCII perhaps? */ + char ascii_value[11]; + unsigned long int ulval; + int ret; + + if (copy_from_user(ascii_value, buf, count)) + return -EFAULT; + + if (count > 10) { + if (ascii_value[10] == '\n') + ascii_value[10] = '\0'; + else + return -EINVAL; + } else { + ascii_value[count] = '\0'; + } + ret = strict_strtoul(ascii_value, 16, &ulval); + if (ret) { + pr_debug("%s, 0x%lx, 0x%x\n", ascii_value, ulval, ret); + return -EINVAL; + } + value = (s32)lower_32_bits(ulval); + } else { + return -EINVAL; + } + + pm_qos_req = filp->private_data; + pm_qos_update_request(pm_qos_req, value); + + return count; +} + + +static int __init pm_qos_power_init(void) +{ + int ret = 0; + + ret = register_pm_qos_misc(&cpu_dma_pm_qos); + if (ret < 0) { + printk(KERN_ERR "pm_qos_param: cpu_dma_latency setup failed\n"); + return ret; + } + ret = register_pm_qos_misc(&network_lat_pm_qos); + if (ret < 0) { + printk(KERN_ERR "pm_qos_param: network_latency setup failed\n"); + return ret; + } + ret = register_pm_qos_misc(&network_throughput_pm_qos); + if (ret < 0) + printk(KERN_ERR + "pm_qos_param: network_throughput setup failed\n"); + + return ret; +} + +late_initcall(pm_qos_power_init); diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 866f269183cf..bb771e9f15bc 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -19,7 +19,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index d6470c7fd6ce..9604abc61a59 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 6f09eca01112..beefb0afefa1 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -14,7 +14,7 @@ #include #include -#include +#include #include #include #include diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index 1c6be91dfb98..c74e228731ed 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -23,7 +23,7 @@ #include #include #include -#include +#include #include #include #include -- cgit v1.3-8-gc7d7 From cc74998618a66d34651c784dd02412614c3e81cc Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:12 +0200 Subject: PM QoS: Minor clean-ups - Misc fixes to improve code readability: * rename struct pm_qos_request_list to struct pm_qos_request, * rename pm_qos_req parameter to req in internal code, consistenly use req in the API parameters, * update the in-kernel API callers to the new parameters names, * rename of fields names (requests, list, node, constraints) Signed-off-by: Jean Pihet Acked-by: markgross Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- drivers/media/video/via-camera.c | 2 +- drivers/net/wireless/ipw2x00/ipw2100.c | 2 +- include/linux/netdevice.h | 2 +- include/linux/pm_qos.h | 22 ++++----- include/sound/pcm.h | 2 +- kernel/power/qos.c | 88 +++++++++++++++++----------------- 6 files changed, 59 insertions(+), 59 deletions(-) (limited to 'kernel') diff --git a/drivers/media/video/via-camera.c b/drivers/media/video/via-camera.c index b3ca3893f3de..fba6c6458cf9 100644 --- a/drivers/media/video/via-camera.c +++ b/drivers/media/video/via-camera.c @@ -69,7 +69,7 @@ struct via_camera { struct mutex lock; enum viacam_opstate opstate; unsigned long flags; - struct pm_qos_request_list qos_request; + struct pm_qos_request qos_request; /* * GPIO info for power/reset management */ diff --git a/drivers/net/wireless/ipw2x00/ipw2100.c b/drivers/net/wireless/ipw2x00/ipw2100.c index aaab76ce6020..db35f99cac14 100644 --- a/drivers/net/wireless/ipw2x00/ipw2100.c +++ b/drivers/net/wireless/ipw2x00/ipw2100.c @@ -174,7 +174,7 @@ that only one external action is invoked at a time. #define DRV_DESCRIPTION "Intel(R) PRO/Wireless 2100 Network Driver" #define DRV_COPYRIGHT "Copyright(c) 2003-2006 Intel Corporation" -static struct pm_qos_request_list ipw2100_pm_qos_req; +static struct pm_qos_request ipw2100_pm_qos_req; /* Debugging stuff */ #ifdef CONFIG_IPW2100_DEBUG diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f72ac6b972b7..f38ab5b7e768 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -964,7 +964,7 @@ struct net_device { */ char name[IFNAMSIZ]; - struct pm_qos_request_list pm_qos_req; + struct pm_qos_request pm_qos_req; /* device name hash chain */ struct hlist_node name_hlist; diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h index 7ba675413b08..6b0968f8bc6e 100644 --- a/include/linux/pm_qos.h +++ b/include/linux/pm_qos.h @@ -20,30 +20,30 @@ #define PM_QOS_NETWORK_LAT_DEFAULT_VALUE (2000 * USEC_PER_SEC) #define PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE 0 -struct pm_qos_request_list { - struct plist_node list; +struct pm_qos_request { + struct plist_node node; int pm_qos_class; }; #ifdef CONFIG_PM -void pm_qos_add_request(struct pm_qos_request_list *l, - int pm_qos_class, s32 value); -void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, +void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, + s32 value); +void pm_qos_update_request(struct pm_qos_request *req, s32 new_value); -void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req); +void pm_qos_remove_request(struct pm_qos_request *req); int pm_qos_request(int pm_qos_class); int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier); int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier); -int pm_qos_request_active(struct pm_qos_request_list *req); +int pm_qos_request_active(struct pm_qos_request *req); #else -static inline void pm_qos_add_request(struct pm_qos_request_list *l, +static inline void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value) { return; } -static inline void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, +static inline void pm_qos_update_request(struct pm_qos_request *req, s32 new_value) { return; } -static inline void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) +static inline void pm_qos_remove_request(struct pm_qos_request *req) { return; } static inline int pm_qos_request(int pm_qos_class) @@ -54,7 +54,7 @@ static inline int pm_qos_add_notifier(int pm_qos_class, static inline int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) { return 0; } -static inline int pm_qos_request_active(struct pm_qos_request_list *req) +static inline int pm_qos_request_active(struct pm_qos_request *req) { return 0; } #endif diff --git a/include/sound/pcm.h b/include/sound/pcm.h index 666ee91e8a2e..54cb079b7bf1 100644 --- a/include/sound/pcm.h +++ b/include/sound/pcm.h @@ -373,7 +373,7 @@ struct snd_pcm_substream { int number; char name[32]; /* substream name */ int stream; /* stream (direction) */ - struct pm_qos_request_list latency_pm_qos_req; /* pm_qos request */ + struct pm_qos_request latency_pm_qos_req; /* pm_qos request */ size_t buffer_bytes_max; /* limit ring buffer size */ struct snd_dma_buffer dma_buffer; unsigned int dma_buf_id; diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 61b47384329e..aa52c44e6080 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -45,7 +45,7 @@ #include /* - * locking rule: all changes to requests or notifiers lists + * locking rule: all changes to constraints or notifiers lists * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock * held, taken with _irqsave. One lock to rule them all */ @@ -60,7 +60,7 @@ enum pm_qos_type { * types linux supports for 32 bit quantites */ struct pm_qos_object { - struct plist_head requests; + struct plist_head constraints; struct blocking_notifier_head *notifiers; struct miscdevice pm_qos_power_miscdev; char *name; @@ -74,7 +74,7 @@ static DEFINE_SPINLOCK(pm_qos_lock); static struct pm_qos_object null_pm_qos; static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier); static struct pm_qos_object cpu_dma_pm_qos = { - .requests = PLIST_HEAD_INIT(cpu_dma_pm_qos.requests), + .constraints = PLIST_HEAD_INIT(cpu_dma_pm_qos.constraints), .notifiers = &cpu_dma_lat_notifier, .name = "cpu_dma_latency", .target_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, @@ -84,7 +84,7 @@ static struct pm_qos_object cpu_dma_pm_qos = { static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); static struct pm_qos_object network_lat_pm_qos = { - .requests = PLIST_HEAD_INIT(network_lat_pm_qos.requests), + .constraints = PLIST_HEAD_INIT(network_lat_pm_qos.constraints), .notifiers = &network_lat_notifier, .name = "network_latency", .target_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, @@ -95,7 +95,7 @@ static struct pm_qos_object network_lat_pm_qos = { static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier); static struct pm_qos_object network_throughput_pm_qos = { - .requests = PLIST_HEAD_INIT(network_throughput_pm_qos.requests), + .constraints = PLIST_HEAD_INIT(network_throughput_pm_qos.constraints), .notifiers = &network_throughput_notifier, .name = "network_throughput", .target_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, @@ -129,15 +129,15 @@ static const struct file_operations pm_qos_power_fops = { /* unlocked internal variant */ static inline int pm_qos_get_value(struct pm_qos_object *o) { - if (plist_head_empty(&o->requests)) + if (plist_head_empty(&o->constraints)) return o->default_value; switch (o->type) { case PM_QOS_MIN: - return plist_first(&o->requests)->prio; + return plist_first(&o->constraints)->prio; case PM_QOS_MAX: - return plist_last(&o->requests)->prio; + return plist_last(&o->constraints)->prio; default: /* runtime check for not using enum */ @@ -170,13 +170,13 @@ static void update_target(struct pm_qos_object *o, struct plist_node *node, * with new value and add, then see if the extremal * changed */ - plist_del(node, &o->requests); + plist_del(node, &o->constraints); plist_node_init(node, value); - plist_add(node, &o->requests); + plist_add(node, &o->constraints); } else if (del) { - plist_del(node, &o->requests); + plist_del(node, &o->constraints); } else { - plist_add(node, &o->requests); + plist_add(node, &o->constraints); } curr_value = pm_qos_get_value(o); pm_qos_set_value(o, curr_value); @@ -222,7 +222,7 @@ int pm_qos_request(int pm_qos_class) } EXPORT_SYMBOL_GPL(pm_qos_request); -int pm_qos_request_active(struct pm_qos_request_list *req) +int pm_qos_request_active(struct pm_qos_request *req) { return req->pm_qos_class != 0; } @@ -230,24 +230,24 @@ EXPORT_SYMBOL_GPL(pm_qos_request_active); /** * pm_qos_add_request - inserts new qos request into the list - * @dep: pointer to a preallocated handle + * @req: pointer to a preallocated handle * @pm_qos_class: identifies which list of qos request to use * @value: defines the qos request * * This function inserts a new entry in the pm_qos_class list of requested qos * performance characteristics. It recomputes the aggregate QoS expectations - * for the pm_qos_class of parameters and initializes the pm_qos_request_list + * for the pm_qos_class of parameters and initializes the pm_qos_request * handle. Caller needs to save this handle for later use in updates and * removal. */ -void pm_qos_add_request(struct pm_qos_request_list *dep, +void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value) { struct pm_qos_object *o = pm_qos_array[pm_qos_class]; int new_value; - if (pm_qos_request_active(dep)) { + if (pm_qos_request_active(req)) { WARN(1, KERN_ERR "pm_qos_add_request() called for already added request\n"); return; } @@ -255,15 +255,15 @@ void pm_qos_add_request(struct pm_qos_request_list *dep, new_value = o->default_value; else new_value = value; - plist_node_init(&dep->list, new_value); - dep->pm_qos_class = pm_qos_class; - update_target(o, &dep->list, 0, PM_QOS_DEFAULT_VALUE); + plist_node_init(&req->node, new_value); + req->pm_qos_class = pm_qos_class; + update_target(o, &req->node, 0, PM_QOS_DEFAULT_VALUE); } EXPORT_SYMBOL_GPL(pm_qos_add_request); /** * pm_qos_update_request - modifies an existing qos request - * @pm_qos_req : handle to list element holding a pm_qos request to use + * @req : handle to list element holding a pm_qos request to use * @value: defines the qos request * * Updates an existing qos request for the pm_qos_class of parameters along @@ -271,56 +271,56 @@ EXPORT_SYMBOL_GPL(pm_qos_add_request); * * Attempts are made to make this code callable on hot code paths. */ -void pm_qos_update_request(struct pm_qos_request_list *pm_qos_req, +void pm_qos_update_request(struct pm_qos_request *req, s32 new_value) { s32 temp; struct pm_qos_object *o; - if (!pm_qos_req) /*guard against callers passing in null */ + if (!req) /*guard against callers passing in null */ return; - if (!pm_qos_request_active(pm_qos_req)) { + if (!pm_qos_request_active(req)) { WARN(1, KERN_ERR "pm_qos_update_request() called for unknown object\n"); return; } - o = pm_qos_array[pm_qos_req->pm_qos_class]; + o = pm_qos_array[req->pm_qos_class]; if (new_value == PM_QOS_DEFAULT_VALUE) temp = o->default_value; else temp = new_value; - if (temp != pm_qos_req->list.prio) - update_target(o, &pm_qos_req->list, 0, temp); + if (temp != req->node.prio) + update_target(o, &req->node, 0, temp); } EXPORT_SYMBOL_GPL(pm_qos_update_request); /** * pm_qos_remove_request - modifies an existing qos request - * @pm_qos_req: handle to request list element + * @req: handle to request list element * - * Will remove pm qos request from the list of requests and + * Will remove pm qos request from the list of constraints and * recompute the current target value for the pm_qos_class. Call this * on slow code paths. */ -void pm_qos_remove_request(struct pm_qos_request_list *pm_qos_req) +void pm_qos_remove_request(struct pm_qos_request *req) { struct pm_qos_object *o; - if (pm_qos_req == NULL) + if (req == NULL) return; /* silent return to keep pcm code cleaner */ - if (!pm_qos_request_active(pm_qos_req)) { + if (!pm_qos_request_active(req)) { WARN(1, KERN_ERR "pm_qos_remove_request() called for unknown object\n"); return; } - o = pm_qos_array[pm_qos_req->pm_qos_class]; - update_target(o, &pm_qos_req->list, 1, PM_QOS_DEFAULT_VALUE); - memset(pm_qos_req, 0, sizeof(*pm_qos_req)); + o = pm_qos_array[req->pm_qos_class]; + update_target(o, &req->node, 1, PM_QOS_DEFAULT_VALUE); + memset(req, 0, sizeof(*req)); } EXPORT_SYMBOL_GPL(pm_qos_remove_request); @@ -368,7 +368,7 @@ static int pm_qos_power_open(struct inode *inode, struct file *filp) pm_qos_class = find_pm_qos_object_by_minor(iminor(inode)); if (pm_qos_class >= 0) { - struct pm_qos_request_list *req = kzalloc(sizeof(*req), GFP_KERNEL); + struct pm_qos_request *req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return -ENOMEM; @@ -383,7 +383,7 @@ static int pm_qos_power_open(struct inode *inode, struct file *filp) static int pm_qos_power_release(struct inode *inode, struct file *filp) { - struct pm_qos_request_list *req; + struct pm_qos_request *req; req = filp->private_data; pm_qos_remove_request(req); @@ -399,14 +399,14 @@ static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, s32 value; unsigned long flags; struct pm_qos_object *o; - struct pm_qos_request_list *pm_qos_req = filp->private_data; + struct pm_qos_request *req = filp->private_data; - if (!pm_qos_req) + if (!req) return -EINVAL; - if (!pm_qos_request_active(pm_qos_req)) + if (!pm_qos_request_active(req)) return -EINVAL; - o = pm_qos_array[pm_qos_req->pm_qos_class]; + o = pm_qos_array[req->pm_qos_class]; spin_lock_irqsave(&pm_qos_lock, flags); value = pm_qos_get_value(o); spin_unlock_irqrestore(&pm_qos_lock, flags); @@ -418,7 +418,7 @@ static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos) { s32 value; - struct pm_qos_request_list *pm_qos_req; + struct pm_qos_request *req; if (count == sizeof(s32)) { if (copy_from_user(&value, buf, sizeof(s32))) @@ -449,8 +449,8 @@ static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, return -EINVAL; } - pm_qos_req = filp->private_data; - pm_qos_update_request(pm_qos_req, value); + req = filp->private_data; + pm_qos_update_request(req, value); return count; } -- cgit v1.3-8-gc7d7 From 4a31a33425a1eb92f6a0b9846f081842268361c8 Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:20 +0200 Subject: PM QoS: Code reorganization Move around the PM QoS misc devices management code for better readability. Signed-off-by: Jean Pihet Acked-by: markgross Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- kernel/power/qos.c | 45 +++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/power/qos.c b/kernel/power/qos.c index aa52c44e6080..788c4cfe0cdb 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -188,28 +188,6 @@ static void update_target(struct pm_qos_object *o, struct plist_node *node, NULL); } -static int register_pm_qos_misc(struct pm_qos_object *qos) -{ - qos->pm_qos_power_miscdev.minor = MISC_DYNAMIC_MINOR; - qos->pm_qos_power_miscdev.name = qos->name; - qos->pm_qos_power_miscdev.fops = &pm_qos_power_fops; - - return misc_register(&qos->pm_qos_power_miscdev); -} - -static int find_pm_qos_object_by_minor(int minor) -{ - int pm_qos_class; - - for (pm_qos_class = 0; - pm_qos_class < PM_QOS_NUM_CLASSES; pm_qos_class++) { - if (minor == - pm_qos_array[pm_qos_class]->pm_qos_power_miscdev.minor) - return pm_qos_class; - } - return -1; -} - /** * pm_qos_request - returns current system wide qos expectation * @pm_qos_class: identification of which qos value is requested @@ -362,6 +340,29 @@ int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) } EXPORT_SYMBOL_GPL(pm_qos_remove_notifier); +/* User space interface to PM QoS classes via misc devices */ +static int register_pm_qos_misc(struct pm_qos_object *qos) +{ + qos->pm_qos_power_miscdev.minor = MISC_DYNAMIC_MINOR; + qos->pm_qos_power_miscdev.name = qos->name; + qos->pm_qos_power_miscdev.fops = &pm_qos_power_fops; + + return misc_register(&qos->pm_qos_power_miscdev); +} + +static int find_pm_qos_object_by_minor(int minor) +{ + int pm_qos_class; + + for (pm_qos_class = 0; + pm_qos_class < PM_QOS_NUM_CLASSES; pm_qos_class++) { + if (minor == + pm_qos_array[pm_qos_class]->pm_qos_power_miscdev.minor) + return pm_qos_class; + } + return -1; +} + static int pm_qos_power_open(struct inode *inode, struct file *filp) { long pm_qos_class; -- cgit v1.3-8-gc7d7 From 4e1779baaa542c83b459b0a56585e0c1a04c7782 Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:27 +0200 Subject: PM QoS: Reorganize data structs In preparation for the per-device constratins support, re-organize the data strctures: - add a struct pm_qos_constraints which contains the constraints related data - update struct pm_qos_object contents to the PM QoS internal object data. Add a pointer to struct pm_qos_constraints - update the internal code to use the new data structs. Signed-off-by: Jean Pihet Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- include/linux/pm_qos.h | 19 +++++++++++ kernel/power/qos.c | 85 ++++++++++++++++++++++++-------------------------- 2 files changed, 60 insertions(+), 44 deletions(-) (limited to 'kernel') diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h index 6b0968f8bc6e..97723113ae92 100644 --- a/include/linux/pm_qos.h +++ b/include/linux/pm_qos.h @@ -25,6 +25,25 @@ struct pm_qos_request { int pm_qos_class; }; +enum pm_qos_type { + PM_QOS_UNITIALIZED, + PM_QOS_MAX, /* return the largest value */ + PM_QOS_MIN /* return the smallest value */ +}; + +/* + * Note: The lockless read path depends on the CPU accessing + * target_value atomically. Atomic access is only guaranteed on all CPU + * types linux supports for 32 bit quantites + */ +struct pm_qos_constraints { + struct plist_head list; + s32 target_value; /* Do not change to 64 bit */ + s32 default_value; + enum pm_qos_type type; + struct blocking_notifier_head *notifiers; +}; + #ifdef CONFIG_PM void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value); diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 788c4cfe0cdb..4a35fe50b777 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -49,58 +49,53 @@ * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock * held, taken with _irqsave. One lock to rule them all */ -enum pm_qos_type { - PM_QOS_MAX, /* return the largest value */ - PM_QOS_MIN /* return the smallest value */ -}; - -/* - * Note: The lockless read path depends on the CPU accessing - * target_value atomically. Atomic access is only guaranteed on all CPU - * types linux supports for 32 bit quantites - */ struct pm_qos_object { - struct plist_head constraints; - struct blocking_notifier_head *notifiers; + struct pm_qos_constraints *constraints; struct miscdevice pm_qos_power_miscdev; char *name; - s32 target_value; /* Do not change to 64 bit */ - s32 default_value; - enum pm_qos_type type; }; static DEFINE_SPINLOCK(pm_qos_lock); static struct pm_qos_object null_pm_qos; + static BLOCKING_NOTIFIER_HEAD(cpu_dma_lat_notifier); -static struct pm_qos_object cpu_dma_pm_qos = { - .constraints = PLIST_HEAD_INIT(cpu_dma_pm_qos.constraints), - .notifiers = &cpu_dma_lat_notifier, - .name = "cpu_dma_latency", +static struct pm_qos_constraints cpu_dma_constraints = { + .list = PLIST_HEAD_INIT(cpu_dma_constraints.list), .target_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, .default_value = PM_QOS_CPU_DMA_LAT_DEFAULT_VALUE, .type = PM_QOS_MIN, + .notifiers = &cpu_dma_lat_notifier, +}; +static struct pm_qos_object cpu_dma_pm_qos = { + .constraints = &cpu_dma_constraints, }; static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); -static struct pm_qos_object network_lat_pm_qos = { - .constraints = PLIST_HEAD_INIT(network_lat_pm_qos.constraints), - .notifiers = &network_lat_notifier, - .name = "network_latency", +static struct pm_qos_constraints network_lat_constraints = { + .list = PLIST_HEAD_INIT(network_lat_constraints.list), .target_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, .default_value = PM_QOS_NETWORK_LAT_DEFAULT_VALUE, - .type = PM_QOS_MIN + .type = PM_QOS_MIN, + .notifiers = &network_lat_notifier, +}; +static struct pm_qos_object network_lat_pm_qos = { + .constraints = &network_lat_constraints, + .name = "network_latency", }; static BLOCKING_NOTIFIER_HEAD(network_throughput_notifier); -static struct pm_qos_object network_throughput_pm_qos = { - .constraints = PLIST_HEAD_INIT(network_throughput_pm_qos.constraints), - .notifiers = &network_throughput_notifier, - .name = "network_throughput", +static struct pm_qos_constraints network_tput_constraints = { + .list = PLIST_HEAD_INIT(network_tput_constraints.list), .target_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, .default_value = PM_QOS_NETWORK_THROUGHPUT_DEFAULT_VALUE, .type = PM_QOS_MAX, + .notifiers = &network_throughput_notifier, +}; +static struct pm_qos_object network_throughput_pm_qos = { + .constraints = &network_tput_constraints, + .name = "network_throughput", }; @@ -129,15 +124,15 @@ static const struct file_operations pm_qos_power_fops = { /* unlocked internal variant */ static inline int pm_qos_get_value(struct pm_qos_object *o) { - if (plist_head_empty(&o->constraints)) - return o->default_value; + if (plist_head_empty(&o->constraints->list)) + return o->constraints->default_value; - switch (o->type) { + switch (o->constraints->type) { case PM_QOS_MIN: - return plist_first(&o->constraints)->prio; + return plist_first(&o->constraints->list)->prio; case PM_QOS_MAX: - return plist_last(&o->constraints)->prio; + return plist_last(&o->constraints->list)->prio; default: /* runtime check for not using enum */ @@ -147,12 +142,12 @@ static inline int pm_qos_get_value(struct pm_qos_object *o) static inline s32 pm_qos_read_value(struct pm_qos_object *o) { - return o->target_value; + return o->constraints->target_value; } static inline void pm_qos_set_value(struct pm_qos_object *o, s32 value) { - o->target_value = value; + o->constraints->target_value = value; } static void update_target(struct pm_qos_object *o, struct plist_node *node, @@ -170,20 +165,20 @@ static void update_target(struct pm_qos_object *o, struct plist_node *node, * with new value and add, then see if the extremal * changed */ - plist_del(node, &o->constraints); + plist_del(node, &o->constraints->list); plist_node_init(node, value); - plist_add(node, &o->constraints); + plist_add(node, &o->constraints->list); } else if (del) { - plist_del(node, &o->constraints); + plist_del(node, &o->constraints->list); } else { - plist_add(node, &o->constraints); + plist_add(node, &o->constraints->list); } curr_value = pm_qos_get_value(o); pm_qos_set_value(o, curr_value); spin_unlock_irqrestore(&pm_qos_lock, flags); if (prev_value != curr_value) - blocking_notifier_call_chain(o->notifiers, + blocking_notifier_call_chain(o->constraints->notifiers, (unsigned long)curr_value, NULL); } @@ -230,7 +225,7 @@ void pm_qos_add_request(struct pm_qos_request *req, return; } if (value == PM_QOS_DEFAULT_VALUE) - new_value = o->default_value; + new_value = o->constraints->default_value; else new_value = value; plist_node_init(&req->node, new_value); @@ -266,7 +261,7 @@ void pm_qos_update_request(struct pm_qos_request *req, o = pm_qos_array[req->pm_qos_class]; if (new_value == PM_QOS_DEFAULT_VALUE) - temp = o->default_value; + temp = o->constraints->default_value; else temp = new_value; @@ -315,7 +310,8 @@ int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier) int retval; retval = blocking_notifier_chain_register( - pm_qos_array[pm_qos_class]->notifiers, notifier); + pm_qos_array[pm_qos_class]->constraints->notifiers, + notifier); return retval; } @@ -334,7 +330,8 @@ int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) int retval; retval = blocking_notifier_chain_unregister( - pm_qos_array[pm_qos_class]->notifiers, notifier); + pm_qos_array[pm_qos_class]->constraints->notifiers, + notifier); return retval; } -- cgit v1.3-8-gc7d7 From abe98ec2d86279fe821c9051003a0abc43444f15 Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:34 +0200 Subject: PM QoS: Generalize and export constraints management code In preparation for the per-device constratins support: - rename update_target to pm_qos_update_target - generalize and export pm_qos_update_target for usage by the upcoming per-device latency constraints framework: * operate on struct pm_qos_constraints for constraints management, * introduce an 'action' parameter for constraints add/update/remove, * the return value indicates if the aggregated constraint value has changed, - update the internal code to operate on struct pm_qos_constraints - add a NULL pointer check in the API functions Signed-off-by: Jean Pihet Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- include/linux/pm_qos.h | 14 ++++++ kernel/power/qos.c | 123 +++++++++++++++++++++++++++---------------------- 2 files changed, 81 insertions(+), 56 deletions(-) (limited to 'kernel') diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h index 97723113ae92..84aa15089896 100644 --- a/include/linux/pm_qos.h +++ b/include/linux/pm_qos.h @@ -44,7 +44,16 @@ struct pm_qos_constraints { struct blocking_notifier_head *notifiers; }; +/* Action requested to pm_qos_update_target */ +enum pm_qos_req_action { + PM_QOS_ADD_REQ, /* Add a new request */ + PM_QOS_UPDATE_REQ, /* Update an existing request */ + PM_QOS_REMOVE_REQ /* Remove an existing request */ +}; + #ifdef CONFIG_PM +int pm_qos_update_target(struct pm_qos_constraints *c, struct plist_node *node, + enum pm_qos_req_action action, int value); void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value); void pm_qos_update_request(struct pm_qos_request *req, @@ -56,6 +65,11 @@ int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier); int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier); int pm_qos_request_active(struct pm_qos_request *req); #else +static inline int pm_qos_update_target(struct pm_qos_constraints *c, + struct plist_node *node, + enum pm_qos_req_action action, + int value) + { return 0; } static inline void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value) { return; } diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 4a35fe50b777..7c7cd181cabe 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -122,17 +122,17 @@ static const struct file_operations pm_qos_power_fops = { }; /* unlocked internal variant */ -static inline int pm_qos_get_value(struct pm_qos_object *o) +static inline int pm_qos_get_value(struct pm_qos_constraints *c) { - if (plist_head_empty(&o->constraints->list)) - return o->constraints->default_value; + if (plist_head_empty(&c->list)) + return c->default_value; - switch (o->constraints->type) { + switch (c->type) { case PM_QOS_MIN: - return plist_first(&o->constraints->list)->prio; + return plist_first(&c->list)->prio; case PM_QOS_MAX: - return plist_last(&o->constraints->list)->prio; + return plist_last(&c->list)->prio; default: /* runtime check for not using enum */ @@ -140,47 +140,73 @@ static inline int pm_qos_get_value(struct pm_qos_object *o) } } -static inline s32 pm_qos_read_value(struct pm_qos_object *o) +static inline s32 pm_qos_read_value(struct pm_qos_constraints *c) { - return o->constraints->target_value; + return c->target_value; } -static inline void pm_qos_set_value(struct pm_qos_object *o, s32 value) +static inline void pm_qos_set_value(struct pm_qos_constraints *c, s32 value) { - o->constraints->target_value = value; + c->target_value = value; } -static void update_target(struct pm_qos_object *o, struct plist_node *node, - int del, int value) +/** + * pm_qos_update_target - manages the constraints list and calls the notifiers + * if needed + * @c: constraints data struct + * @node: request to add to the list, to update or to remove + * @action: action to take on the constraints list + * @value: value of the request to add or update + * + * This function returns 1 if the aggregated constraint value has changed, 0 + * otherwise. + */ +int pm_qos_update_target(struct pm_qos_constraints *c, struct plist_node *node, + enum pm_qos_req_action action, int value) { unsigned long flags; - int prev_value, curr_value; + int prev_value, curr_value, new_value; spin_lock_irqsave(&pm_qos_lock, flags); - prev_value = pm_qos_get_value(o); - /* PM_QOS_DEFAULT_VALUE is a signal that the value is unchanged */ - if (value != PM_QOS_DEFAULT_VALUE) { + prev_value = pm_qos_get_value(c); + if (value == PM_QOS_DEFAULT_VALUE) + new_value = c->default_value; + else + new_value = value; + + switch (action) { + case PM_QOS_REMOVE_REQ: + plist_del(node, &c->list); + break; + case PM_QOS_UPDATE_REQ: /* * to change the list, we atomically remove, reinit * with new value and add, then see if the extremal * changed */ - plist_del(node, &o->constraints->list); - plist_node_init(node, value); - plist_add(node, &o->constraints->list); - } else if (del) { - plist_del(node, &o->constraints->list); - } else { - plist_add(node, &o->constraints->list); + plist_del(node, &c->list); + case PM_QOS_ADD_REQ: + plist_node_init(node, new_value); + plist_add(node, &c->list); + break; + default: + /* no action */ + ; } - curr_value = pm_qos_get_value(o); - pm_qos_set_value(o, curr_value); + + curr_value = pm_qos_get_value(c); + pm_qos_set_value(c, curr_value); + spin_unlock_irqrestore(&pm_qos_lock, flags); - if (prev_value != curr_value) - blocking_notifier_call_chain(o->constraints->notifiers, + if (prev_value != curr_value) { + blocking_notifier_call_chain(c->notifiers, (unsigned long)curr_value, NULL); + return 1; + } else { + return 0; + } } /** @@ -191,7 +217,7 @@ static void update_target(struct pm_qos_object *o, struct plist_node *node, */ int pm_qos_request(int pm_qos_class) { - return pm_qos_read_value(pm_qos_array[pm_qos_class]); + return pm_qos_read_value(pm_qos_array[pm_qos_class]->constraints); } EXPORT_SYMBOL_GPL(pm_qos_request); @@ -217,20 +243,16 @@ EXPORT_SYMBOL_GPL(pm_qos_request_active); void pm_qos_add_request(struct pm_qos_request *req, int pm_qos_class, s32 value) { - struct pm_qos_object *o = pm_qos_array[pm_qos_class]; - int new_value; + if (!req) /*guard against callers passing in null */ + return; if (pm_qos_request_active(req)) { WARN(1, KERN_ERR "pm_qos_add_request() called for already added request\n"); return; } - if (value == PM_QOS_DEFAULT_VALUE) - new_value = o->constraints->default_value; - else - new_value = value; - plist_node_init(&req->node, new_value); req->pm_qos_class = pm_qos_class; - update_target(o, &req->node, 0, PM_QOS_DEFAULT_VALUE); + pm_qos_update_target(pm_qos_array[pm_qos_class]->constraints, + &req->node, PM_QOS_ADD_REQ, value); } EXPORT_SYMBOL_GPL(pm_qos_add_request); @@ -247,9 +269,6 @@ EXPORT_SYMBOL_GPL(pm_qos_add_request); void pm_qos_update_request(struct pm_qos_request *req, s32 new_value) { - s32 temp; - struct pm_qos_object *o; - if (!req) /*guard against callers passing in null */ return; @@ -258,15 +277,10 @@ void pm_qos_update_request(struct pm_qos_request *req, return; } - o = pm_qos_array[req->pm_qos_class]; - - if (new_value == PM_QOS_DEFAULT_VALUE) - temp = o->constraints->default_value; - else - temp = new_value; - - if (temp != req->node.prio) - update_target(o, &req->node, 0, temp); + if (new_value != req->node.prio) + pm_qos_update_target( + pm_qos_array[req->pm_qos_class]->constraints, + &req->node, PM_QOS_UPDATE_REQ, new_value); } EXPORT_SYMBOL_GPL(pm_qos_update_request); @@ -280,9 +294,7 @@ EXPORT_SYMBOL_GPL(pm_qos_update_request); */ void pm_qos_remove_request(struct pm_qos_request *req) { - struct pm_qos_object *o; - - if (req == NULL) + if (!req) /*guard against callers passing in null */ return; /* silent return to keep pcm code cleaner */ @@ -291,8 +303,9 @@ void pm_qos_remove_request(struct pm_qos_request *req) return; } - o = pm_qos_array[req->pm_qos_class]; - update_target(o, &req->node, 1, PM_QOS_DEFAULT_VALUE); + pm_qos_update_target(pm_qos_array[req->pm_qos_class]->constraints, + &req->node, PM_QOS_REMOVE_REQ, + PM_QOS_DEFAULT_VALUE); memset(req, 0, sizeof(*req)); } EXPORT_SYMBOL_GPL(pm_qos_remove_request); @@ -396,7 +409,6 @@ static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, { s32 value; unsigned long flags; - struct pm_qos_object *o; struct pm_qos_request *req = filp->private_data; if (!req) @@ -404,9 +416,8 @@ static ssize_t pm_qos_power_read(struct file *filp, char __user *buf, if (!pm_qos_request_active(req)) return -EINVAL; - o = pm_qos_array[req->pm_qos_class]; spin_lock_irqsave(&pm_qos_lock, flags); - value = pm_qos_get_value(o); + value = pm_qos_get_value(pm_qos_array[req->pm_qos_class]->constraints); spin_unlock_irqrestore(&pm_qos_lock, flags); return simple_read_from_buffer(buf, count, f_pos, &value, sizeof(s32)); -- cgit v1.3-8-gc7d7 From b66213cdb002b08b29603d488c451dfe25e2ca20 Mon Sep 17 00:00:00 2001 From: Jean Pihet Date: Thu, 25 Aug 2011 15:35:47 +0200 Subject: PM QoS: Add global notification mechanism for device constraints Add a global notification chain that gets called upon changes to the aggregated constraint value for any device. The notification callbacks are passing the full constraint request data in order for the callees to have access to it. The current use is for the platform low-level code to access the target device of the constraint. Signed-off-by: Jean Pihet Reviewed-by: Kevin Hilman Signed-off-by: Rafael J. Wysocki --- drivers/base/power/qos.c | 89 ++++++++++++++++++++++++++++++++++++++++-------- include/linux/pm_qos.h | 11 ++++++ kernel/power/qos.c | 2 +- 3 files changed, 87 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c index cc4c541398aa..8d0b81151c14 100644 --- a/drivers/base/power/qos.c +++ b/drivers/base/power/qos.c @@ -17,6 +17,12 @@ * * This QoS design is best effort based. Dependents register their QoS needs. * Watchers register to keep track of the current QoS needs of the system. + * Watchers can register different types of notification callbacks: + * . a per-device notification callback using the dev_pm_qos_*_notifier API. + * The notification chain data is stored in the per-device constraint + * data struct. + * . a system-wide notification callback using the dev_pm_qos_*_global_notifier + * API. The notification chain data is stored in a static variable. * * Note about the per-device constraint data struct allocation: * . The per-device constraints data struct ptr is tored into the device @@ -45,6 +51,36 @@ static DEFINE_MUTEX(dev_pm_qos_mtx); +static BLOCKING_NOTIFIER_HEAD(dev_pm_notifiers); + +/* + * apply_constraint + * @req: constraint request to apply + * @action: action to perform add/update/remove, of type enum pm_qos_req_action + * @value: defines the qos request + * + * Internal function to update the constraints list using the PM QoS core + * code and if needed call the per-device and the global notification + * callbacks + */ +static int apply_constraint(struct dev_pm_qos_request *req, + enum pm_qos_req_action action, int value) +{ + int ret, curr_value; + + ret = pm_qos_update_target(req->dev->power.constraints, + &req->node, action, value); + + if (ret) { + /* Call the global callbacks if needed */ + curr_value = pm_qos_read_value(req->dev->power.constraints); + blocking_notifier_call_chain(&dev_pm_notifiers, + (unsigned long)curr_value, + req); + } + + return ret; +} /* * dev_pm_qos_constraints_allocate @@ -111,12 +147,11 @@ void dev_pm_qos_constraints_destroy(struct device *dev) &dev->power.constraints->list, node) { /* - * Update constraints list and call the per-device + * Update constraints list and call the notification * callbacks if needed */ - pm_qos_update_target(req->dev->power.constraints, - &req->node, PM_QOS_REMOVE_REQ, - PM_QOS_DEFAULT_VALUE); + apply_constraint(req, PM_QOS_REMOVE_REQ, + PM_QOS_DEFAULT_VALUE); memset(req, 0, sizeof(*req)); } @@ -147,7 +182,7 @@ void dev_pm_qos_constraints_destroy(struct device *dev) * removed from the system */ int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req, - s32 value) + s32 value) { int ret = 0; @@ -178,8 +213,7 @@ int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req, ret = dev_pm_qos_constraints_allocate(dev); if (!ret) - ret = pm_qos_update_target(dev->power.constraints, &req->node, - PM_QOS_ADD_REQ, value); + ret = apply_constraint(req, PM_QOS_ADD_REQ, value); out: mutex_unlock(&dev_pm_qos_mtx); @@ -220,10 +254,8 @@ int dev_pm_qos_update_request(struct dev_pm_qos_request *req, if (req->dev->power.constraints_state == DEV_PM_QOS_ALLOCATED) { if (new_value != req->node.prio) - ret = pm_qos_update_target(req->dev->power.constraints, - &req->node, - PM_QOS_UPDATE_REQ, - new_value); + ret = apply_constraint(req, PM_QOS_UPDATE_REQ, + new_value); } else { /* Return if the device has been removed */ ret = -ENODEV; @@ -262,9 +294,8 @@ int dev_pm_qos_remove_request(struct dev_pm_qos_request *req) mutex_lock(&dev_pm_qos_mtx); if (req->dev->power.constraints_state == DEV_PM_QOS_ALLOCATED) { - ret = pm_qos_update_target(req->dev->power.constraints, - &req->node, PM_QOS_REMOVE_REQ, - PM_QOS_DEFAULT_VALUE); + ret = apply_constraint(req, PM_QOS_REMOVE_REQ, + PM_QOS_DEFAULT_VALUE); memset(req, 0, sizeof(*req)); } else { /* Return if the device has been removed */ @@ -336,3 +367,33 @@ out: return retval; } EXPORT_SYMBOL_GPL(dev_pm_qos_remove_notifier); + +/** + * dev_pm_qos_add_global_notifier - sets notification entry for changes to + * target value of the PM QoS constraints for any device + * + * @notifier: notifier block managed by caller. + * + * Will register the notifier into a notification chain that gets called + * upon changes to the target value for any device. + */ +int dev_pm_qos_add_global_notifier(struct notifier_block *notifier) +{ + return blocking_notifier_chain_register(&dev_pm_notifiers, notifier); +} +EXPORT_SYMBOL_GPL(dev_pm_qos_add_global_notifier); + +/** + * dev_pm_qos_remove_global_notifier - deletes notification for changes to + * target value of PM QoS constraints for any device + * + * @notifier: notifier block to be removed. + * + * Will remove the notifier from the notification chain that gets called + * upon changes to the target value for any device. + */ +int dev_pm_qos_remove_global_notifier(struct notifier_block *notifier) +{ + return blocking_notifier_chain_unregister(&dev_pm_notifiers, notifier); +} +EXPORT_SYMBOL_GPL(dev_pm_qos_remove_global_notifier); diff --git a/include/linux/pm_qos.h b/include/linux/pm_qos.h index f75f74dd9b90..ca7bd3f98cb4 100644 --- a/include/linux/pm_qos.h +++ b/include/linux/pm_qos.h @@ -75,6 +75,7 @@ int pm_qos_request(int pm_qos_class); int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier); int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier); int pm_qos_request_active(struct pm_qos_request *req); +s32 pm_qos_read_value(struct pm_qos_constraints *c); int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req, s32 value); @@ -84,6 +85,8 @@ int dev_pm_qos_add_notifier(struct device *dev, struct notifier_block *notifier); int dev_pm_qos_remove_notifier(struct device *dev, struct notifier_block *notifier); +int dev_pm_qos_add_global_notifier(struct notifier_block *notifier); +int dev_pm_qos_remove_global_notifier(struct notifier_block *notifier); void dev_pm_qos_constraints_init(struct device *dev); void dev_pm_qos_constraints_destroy(struct device *dev); #else @@ -111,6 +114,8 @@ static inline int pm_qos_remove_notifier(int pm_qos_class, { return 0; } static inline int pm_qos_request_active(struct pm_qos_request *req) { return 0; } +static inline s32 pm_qos_read_value(struct pm_qos_constraints *c) + { return 0; } static inline int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req, @@ -127,6 +132,12 @@ static inline int dev_pm_qos_add_notifier(struct device *dev, static inline int dev_pm_qos_remove_notifier(struct device *dev, struct notifier_block *notifier) { return 0; } +static inline int dev_pm_qos_add_global_notifier( + struct notifier_block *notifier) + { return 0; } +static inline int dev_pm_qos_remove_global_notifier( + struct notifier_block *notifier) + { return 0; } static inline void dev_pm_qos_constraints_init(struct device *dev) { return; } static inline void dev_pm_qos_constraints_destroy(struct device *dev) diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 7c7cd181cabe..1c1797dd1d1d 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -140,7 +140,7 @@ static inline int pm_qos_get_value(struct pm_qos_constraints *c) } } -static inline s32 pm_qos_read_value(struct pm_qos_constraints *c) +s32 pm_qos_read_value(struct pm_qos_constraints *c) { return c->target_value; } -- cgit v1.3-8-gc7d7 From be27425dcc516fd08245b047ea57f83b8f6f0903 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 19 Aug 2011 16:15:10 -0700 Subject: Add a personality to report 2.6.x version numbers I ran into a couple of programs which broke with the new Linux 3.0 version. Some of those were binary only. I tried to use LD_PRELOAD to work around it, but it was quite difficult and in one case impossible because of a mix of 32bit and 64bit executables. For example, all kind of management software from HP doesnt work, unless we pretend to run a 2.6 kernel. $ uname -a Linux svivoipvnx001 3.0.0-08107-g97cd98f #1062 SMP Fri Aug 12 18:11:45 CEST 2011 i686 i686 i386 GNU/Linux $ hpacucli ctrl all show Error: No controllers detected. $ rpm -qf /usr/sbin/hpacucli hpacucli-8.75-12.0 Another notable case is that Python now reports "linux3" from sys.platform(); which in turn can break things that were checking sys.platform() == "linux2": https://bugzilla.mozilla.org/show_bug.cgi?id=664564 It seems pretty clear to me though it's a bug in the apps that are using '==' instead of .startswith(), but this allows us to unbreak broken programs. This patch adds a UNAME26 personality that makes the kernel report a 2.6.40+x version number instead. The x is the x in 3.x. I know this is somewhat ugly, but I didn't find a better workaround, and compatibility to existing programs is important. Some programs also read /proc/sys/kernel/osrelease. This can be worked around in user space with mount --bind (and a mount namespace) To use: wget ftp://ftp.kernel.org/pub/linux/kernel/people/ak/uname26/uname26.c gcc -o uname26 uname26.c ./uname26 program Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- include/linux/personality.h | 1 + kernel/sys.c | 38 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 39 insertions(+) (limited to 'kernel') diff --git a/include/linux/personality.h b/include/linux/personality.h index eec3bae164d4..8fc7dd1a57ff 100644 --- a/include/linux/personality.h +++ b/include/linux/personality.h @@ -22,6 +22,7 @@ extern int __set_personality(unsigned int); * These occupy the top three bytes. */ enum { + UNAME26 = 0x0020000, ADDR_NO_RANDOMIZE = 0x0040000, /* disable randomization of VA space */ FDPIC_FUNCPTRS = 0x0080000, /* userspace function ptrs point to descriptors * (signal handling) diff --git a/kernel/sys.c b/kernel/sys.c index dd948a1fca4c..18ee1d2f6474 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -37,6 +37,8 @@ #include #include #include +#include +#include #include #include @@ -44,6 +46,8 @@ #include #include +/* Move somewhere else to avoid recompiling? */ +#include #include #include @@ -1161,6 +1165,34 @@ DECLARE_RWSEM(uts_sem); #define override_architecture(name) 0 #endif +/* + * Work around broken programs that cannot handle "Linux 3.0". + * Instead we map 3.x to 2.6.40+x, so e.g. 3.0 would be 2.6.40 + */ +static int override_release(char __user *release, int len) +{ + int ret = 0; + char buf[len]; + + if (current->personality & UNAME26) { + char *rest = UTS_RELEASE; + int ndots = 0; + unsigned v; + + while (*rest) { + if (*rest == '.' && ++ndots >= 3) + break; + if (!isdigit(*rest) && *rest != '.') + break; + rest++; + } + v = ((LINUX_VERSION_CODE >> 8) & 0xff) + 40; + snprintf(buf, len, "2.6.%u%s", v, rest); + ret = copy_to_user(release, buf, len); + } + return ret; +} + SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name) { int errno = 0; @@ -1170,6 +1202,8 @@ SYSCALL_DEFINE1(newuname, struct new_utsname __user *, name) errno = -EFAULT; up_read(&uts_sem); + if (!errno && override_release(name->release, sizeof(name->release))) + errno = -EFAULT; if (!errno && override_architecture(name)) errno = -EFAULT; return errno; @@ -1191,6 +1225,8 @@ SYSCALL_DEFINE1(uname, struct old_utsname __user *, name) error = -EFAULT; up_read(&uts_sem); + if (!error && override_release(name->release, sizeof(name->release))) + error = -EFAULT; if (!error && override_architecture(name)) error = -EFAULT; return error; @@ -1225,6 +1261,8 @@ SYSCALL_DEFINE1(olduname, struct oldold_utsname __user *, name) if (!error && override_architecture(name)) error = -EFAULT; + if (!error && override_release(name->release, sizeof(name->release))) + error = -EFAULT; return error ? -EFAULT : 0; } #endif -- cgit v1.3-8-gc7d7 From 4c30c6f566c0989ddaee3407da44751e340a63ed Mon Sep 17 00:00:00 2001 From: Nishanth Aravamudan Date: Thu, 25 Aug 2011 15:59:11 -0700 Subject: kernel/printk: do not turn off bootconsole in printk_late_init() if keep_bootcon It seems that 7bf693951a8e ("console: allow to retain boot console via boot option keep_bootcon") doesn't always achieve what it aims, as when printk_late_init() runs it unconditionally turns off all boot consoles. With this patch, I am able to see more messages on the boot console in KVM guests than I can without, when keep_bootcon is specified. I think it is appropriate for the relevant -stable trees. However, it's more of an annoyance than a serious bug (ideally you don't need to keep the boot console around as console handover should be working -- I was encountering a situation where the console handover wasn't working and not having the boot console available meant I couldn't see why). Signed-off-by: Nishanth Aravamudan Cc: David S. Miller Cc: Alan Cox Cc: Greg KH Acked-by: Fabio M. Di Nitto Cc: [2.6.39.x, 3.0.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 836a2ae0ac31..28a40d8171b8 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1604,7 +1604,7 @@ static int __init printk_late_init(void) struct console *con; for_each_console(con) { - if (con->flags & CON_BOOT) { + if (!keep_bootcon && con->flags & CON_BOOT) { printk(KERN_INFO "turn off boot console %s%d\n", con->name, con->index); unregister_console(con); -- cgit v1.3-8-gc7d7 From f5b940997397229975ea073679b03967932a541b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 26 Aug 2011 18:03:11 -0400 Subject: All Arch: remove linkage for sys_nfsservctl system call The nfsservctl system call is now gone, so we should remove all linkage for it. Signed-off-by: NeilBrown Signed-off-by: J. Bruce Fields Signed-off-by: Linus Torvalds --- arch/alpha/kernel/systbls.S | 2 +- arch/arm/kernel/calls.S | 2 +- arch/avr32/kernel/syscall_table.S | 2 +- arch/blackfin/mach-common/entry.S | 2 +- arch/cris/arch-v10/kernel/entry.S | 2 +- arch/cris/arch-v32/kernel/entry.S | 2 +- arch/frv/kernel/entry.S | 2 +- arch/h8300/kernel/syscalls.S | 2 +- arch/ia64/kernel/entry.S | 2 +- arch/m32r/kernel/syscall_table.S | 2 +- arch/m68k/kernel/syscalltable.S | 2 +- arch/microblaze/kernel/syscall_table.S | 2 +- arch/mips/kernel/scall32-o32.S | 2 +- arch/mips/kernel/scall64-64.S | 2 +- arch/mips/kernel/scall64-n32.S | 2 +- arch/mips/kernel/scall64-o32.S | 2 +- arch/mn10300/kernel/entry.S | 2 +- arch/s390/kernel/compat_wrapper.S | 6 ------ arch/s390/kernel/syscalls.S | 2 +- arch/sh/kernel/syscalls_32.S | 2 +- arch/sh/kernel/syscalls_64.S | 2 +- arch/sparc/kernel/sys32.S | 1 - arch/sparc/kernel/systbls_32.S | 2 +- arch/sparc/kernel/systbls_64.S | 2 +- arch/x86/ia32/ia32entry.S | 2 +- arch/x86/include/asm/unistd_64.h | 2 +- arch/x86/kernel/syscall_table_32.S | 2 +- arch/xtensa/include/asm/unistd.h | 2 +- fs/compat.c | 5 ----- include/asm-generic/unistd.h | 2 +- include/linux/compat.h | 1 - include/linux/syscalls.h | 3 --- kernel/sys_ni.c | 1 - 33 files changed, 27 insertions(+), 44 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index b9c28f3f1956..6acea1f96de3 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -360,7 +360,7 @@ sys_call_table: .quad sys_newuname .quad sys_nanosleep /* 340 */ .quad sys_mremap - .quad sys_nfsservctl + .quad sys_ni_syscall /* old nfsservctl */ .quad sys_setresuid .quad sys_getresuid .quad sys_pciconfig_read /* 345 */ diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index 80f7896cc016..9943e9e74a1b 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -178,7 +178,7 @@ CALL(sys_ni_syscall) /* vm86 */ CALL(sys_ni_syscall) /* was sys_query_module */ CALL(sys_poll) - CALL(sys_nfsservctl) + CALL(sys_ni_syscall) /* was nfsservctl */ /* 170 */ CALL(sys_setresgid16) CALL(sys_getresgid16) CALL(sys_prctl) diff --git a/arch/avr32/kernel/syscall_table.S b/arch/avr32/kernel/syscall_table.S index c7fd394d28a4..6eba53530d1c 100644 --- a/arch/avr32/kernel/syscall_table.S +++ b/arch/avr32/kernel/syscall_table.S @@ -158,7 +158,7 @@ sys_call_table: .long sys_sched_rr_get_interval .long sys_nanosleep .long sys_poll - .long sys_nfsservctl /* 145 */ + .long sys_ni_syscall /* 145 was nfsservctl */ .long sys_setresgid .long sys_getresgid .long sys_prctl diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S index 225d311c9701..e4137297b790 100644 --- a/arch/blackfin/mach-common/entry.S +++ b/arch/blackfin/mach-common/entry.S @@ -1543,7 +1543,7 @@ ENTRY(_sys_call_table) .long _sys_ni_syscall /* for vm86 */ .long _sys_ni_syscall /* old "query_module" */ .long _sys_ni_syscall /* sys_poll */ - .long _sys_nfsservctl + .long _sys_ni_syscall /* old nfsservctl */ .long _sys_setresgid /* setresgid16 */ /* 170 */ .long _sys_getresgid /* getresgid16 */ .long _sys_prctl diff --git a/arch/cris/arch-v10/kernel/entry.S b/arch/cris/arch-v10/kernel/entry.S index 1161883eb582..592fbe9dfb62 100644 --- a/arch/cris/arch-v10/kernel/entry.S +++ b/arch/cris/arch-v10/kernel/entry.S @@ -771,7 +771,7 @@ sys_call_table: .long sys_ni_syscall /* sys_vm86 */ .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* old nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/cris/arch-v32/kernel/entry.S b/arch/cris/arch-v32/kernel/entry.S index 84fed7e91ada..c3ea4694fbaf 100644 --- a/arch/cris/arch-v32/kernel/entry.S +++ b/arch/cris/arch-v32/kernel/entry.S @@ -714,7 +714,7 @@ sys_call_table: .long sys_ni_syscall /* sys_vm86 */ .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* Old nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/frv/kernel/entry.S b/arch/frv/kernel/entry.S index 017d6d7b784f..5ba23f715ea5 100644 --- a/arch/frv/kernel/entry.S +++ b/arch/frv/kernel/entry.S @@ -1358,7 +1358,7 @@ sys_call_table: .long sys_ni_syscall /* for vm86 */ .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* Old nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/h8300/kernel/syscalls.S b/arch/h8300/kernel/syscalls.S index f4b2e67bcc34..4be2ea2fbe26 100644 --- a/arch/h8300/kernel/syscalls.S +++ b/arch/h8300/kernel/syscalls.S @@ -183,7 +183,7 @@ SYMBOL_NAME_LABEL(sys_call_table) .long SYMBOL_NAME(sys_ni_syscall) /* for vm86 */ .long SYMBOL_NAME(sys_ni_syscall) /* sys_query_module */ .long SYMBOL_NAME(sys_poll) - .long SYMBOL_NAME(sys_nfsservctl) + .long SYMBOL_NAME(sys_ni_syscall) /* old nfsservctl */ .long SYMBOL_NAME(sys_setresgid16) /* 170 */ .long SYMBOL_NAME(sys_getresgid16) .long SYMBOL_NAME(sys_prctl) diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 97dd2abdeb1a..198c753d1006 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1614,7 +1614,7 @@ sys_call_table: data8 sys_sched_get_priority_min data8 sys_sched_rr_get_interval data8 sys_nanosleep - data8 sys_nfsservctl + data8 sys_ni_syscall // old nfsservctl data8 sys_prctl // 1170 data8 sys_getpagesize data8 sys_mmap2 diff --git a/arch/m32r/kernel/syscall_table.S b/arch/m32r/kernel/syscall_table.S index 528f2e6ad064..f365c19795ef 100644 --- a/arch/m32r/kernel/syscall_table.S +++ b/arch/m32r/kernel/syscall_table.S @@ -168,7 +168,7 @@ ENTRY(sys_call_table) .long sys_tas /* vm86 syscall holder */ .long sys_ni_syscall /* query_module syscall holder */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* was nfsservctl */ .long sys_setresgid /* 170 */ .long sys_getresgid .long sys_prctl diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S index 00d1452f9571..c468f2edaa85 100644 --- a/arch/m68k/kernel/syscalltable.S +++ b/arch/m68k/kernel/syscalltable.S @@ -189,7 +189,7 @@ ENTRY(sys_call_table) .long sys_getpagesize .long sys_ni_syscall /* old "query_module" */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* old nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/microblaze/kernel/syscall_table.S b/arch/microblaze/kernel/syscall_table.S index d915a122c865..8789daa2a346 100644 --- a/arch/microblaze/kernel/syscall_table.S +++ b/arch/microblaze/kernel/syscall_table.S @@ -173,7 +173,7 @@ ENTRY(sys_call_table) .long sys_ni_syscall /* sys_vm86 */ .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* old nfsservctl */ .long sys_setresgid /* 170 */ .long sys_getresgid .long sys_prctl diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index e521420a45a5..865bc7a6f5a1 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -424,7 +424,7 @@ einval: li v0, -ENOSYS sys sys_getresuid 3 sys sys_ni_syscall 0 /* was sys_query_module */ sys sys_poll 3 - sys sys_nfsservctl 3 + sys sys_ni_syscall 0 /* was nfsservctl */ sys sys_setresgid 3 /* 4190 */ sys sys_getresgid 3 sys sys_prctl 5 diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index 85874d6a8a70..fb7334bea731 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -299,7 +299,7 @@ sys_call_table: PTR sys_ni_syscall /* 5170, was get_kernel_syms */ PTR sys_ni_syscall /* was query_module */ PTR sys_quotactl - PTR sys_nfsservctl + PTR sys_ni_syscall /* was nfsservctl */ PTR sys_ni_syscall /* res. for getpmsg */ PTR sys_ni_syscall /* 5175 for putpmsg */ PTR sys_ni_syscall /* res. for afs_syscall */ diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index b85842fc87ae..f9296e894e46 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -294,7 +294,7 @@ EXPORT(sysn32_call_table) PTR sys_ni_syscall /* 6170, was get_kernel_syms */ PTR sys_ni_syscall /* was query_module */ PTR sys_quotactl - PTR compat_sys_nfsservctl + PTR sys_ni_syscall /* was nfsservctl */ PTR sys_ni_syscall /* res. for getpmsg */ PTR sys_ni_syscall /* 6175 for putpmsg */ PTR sys_ni_syscall /* res. for afs_syscall */ diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 46c4763edf21..4d7c9827706f 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -392,7 +392,7 @@ sys_call_table: PTR sys_getresuid PTR sys_ni_syscall /* was query_module */ PTR sys_poll - PTR compat_sys_nfsservctl + PTR sys_ni_syscall /* was nfsservctl */ PTR sys_setresgid /* 4190 */ PTR sys_getresgid PTR sys_prctl diff --git a/arch/mn10300/kernel/entry.S b/arch/mn10300/kernel/entry.S index ae435e1d5669..3e3620d9fc45 100644 --- a/arch/mn10300/kernel/entry.S +++ b/arch/mn10300/kernel/entry.S @@ -589,7 +589,7 @@ ENTRY(sys_call_table) .long sys_ni_syscall /* vm86 */ .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* was nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 08ab9aa6a0d5..7526db6bf501 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -665,12 +665,6 @@ ENTRY(sys32_poll_wrapper) lgfr %r4,%r4 # long jg sys_poll # branch to system call -ENTRY(compat_sys_nfsservctl_wrapper) - lgfr %r2,%r2 # int - llgtr %r3,%r3 # struct compat_nfsctl_arg* - llgtr %r4,%r4 # union compat_nfsctl_res* - jg compat_sys_nfsservctl # branch to system call - ENTRY(sys32_setresgid16_wrapper) llgfr %r2,%r2 # __kernel_old_gid_emu31_t llgfr %r3,%r3 # __kernel_old_gid_emu31_t diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 6ee39ef8fe4a..73eb08c874fb 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -177,7 +177,7 @@ SYSCALL(sys_getresuid16,sys_ni_syscall,sys32_getresuid16_wrapper) /* 165 old get NI_SYSCALL /* for vm86 */ NI_SYSCALL /* old sys_query_module */ SYSCALL(sys_poll,sys_poll,sys32_poll_wrapper) -SYSCALL(sys_nfsservctl,sys_nfsservctl,compat_sys_nfsservctl_wrapper) +NI_SYSCALL /* old nfsservctl */ SYSCALL(sys_setresgid16,sys_ni_syscall,sys32_setresgid16_wrapper) /* 170 old setresgid16 syscall */ SYSCALL(sys_getresgid16,sys_ni_syscall,sys32_getresgid16_wrapper) /* old getresgid16 syscall */ SYSCALL(sys_prctl,sys_prctl,sys32_prctl_wrapper) diff --git a/arch/sh/kernel/syscalls_32.S b/arch/sh/kernel/syscalls_32.S index 39b051de4c7c..293e39c59c00 100644 --- a/arch/sh/kernel/syscalls_32.S +++ b/arch/sh/kernel/syscalls_32.S @@ -185,7 +185,7 @@ ENTRY(sys_call_table) .long sys_ni_syscall /* vm86 */ .long sys_ni_syscall /* old "query_module" */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* was nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/sh/kernel/syscalls_64.S b/arch/sh/kernel/syscalls_64.S index 089c4d825d08..ceb34b94afa9 100644 --- a/arch/sh/kernel/syscalls_64.S +++ b/arch/sh/kernel/syscalls_64.S @@ -189,7 +189,7 @@ sys_call_table: .long sys_ni_syscall /* vm86 */ .long sys_ni_syscall /* old "query_module" */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* was nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index 44e5faf1ad5f..d97f3eb72e06 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -81,7 +81,6 @@ SIGN2(sys32_fadvise64, compat_sys_fadvise64, %o0, %o4) SIGN2(sys32_fadvise64_64, compat_sys_fadvise64_64, %o0, %o5) SIGN2(sys32_bdflush, sys_bdflush, %o0, %o1) SIGN1(sys32_mlockall, sys_mlockall, %o0) -SIGN1(sys32_nfsservctl, compat_sys_nfsservctl, %o0) SIGN1(sys32_clock_nanosleep, compat_sys_clock_nanosleep, %o1) SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1) SIGN1(sys32_io_submit, compat_sys_io_submit, %o1) diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S index 6e492d59f6b1..09d8ec454450 100644 --- a/arch/sparc/kernel/systbls_32.S +++ b/arch/sparc/kernel/systbls_32.S @@ -67,7 +67,7 @@ sys_call_table: /*235*/ .long sys_fstatfs64, sys_llseek, sys_mlock, sys_munlock, sys_mlockall /*240*/ .long sys_munlockall, sys_sched_setparam, sys_sched_getparam, sys_sched_setscheduler, sys_sched_getscheduler /*245*/ .long sys_sched_yield, sys_sched_get_priority_max, sys_sched_get_priority_min, sys_sched_rr_get_interval, sys_nanosleep -/*250*/ .long sys_mremap, sys_sysctl, sys_getsid, sys_fdatasync, sys_nfsservctl +/*250*/ .long sys_mremap, sys_sysctl, sys_getsid, sys_fdatasync, sys_ni_syscall /*255*/ .long sys_sync_file_range, sys_clock_settime, sys_clock_gettime, sys_clock_getres, sys_clock_nanosleep /*260*/ .long sys_sched_getaffinity, sys_sched_setaffinity, sys_timer_settime, sys_timer_gettime, sys_timer_getoverrun /*265*/ .long sys_timer_delete, sys_timer_create, sys_nis_syscall, sys_io_setup, sys_io_destroy diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S index f566518483b5..c9296ab0b1f4 100644 --- a/arch/sparc/kernel/systbls_64.S +++ b/arch/sparc/kernel/systbls_64.S @@ -145,7 +145,7 @@ sys_call_table: .word sys_fstatfs64, sys_llseek, sys_mlock, sys_munlock, sys_mlockall /*240*/ .word sys_munlockall, sys_sched_setparam, sys_sched_getparam, sys_sched_setscheduler, sys_sched_getscheduler .word sys_sched_yield, sys_sched_get_priority_max, sys_sched_get_priority_min, sys_sched_rr_get_interval, sys_nanosleep -/*250*/ .word sys_64_mremap, sys_sysctl, sys_getsid, sys_fdatasync, sys_nfsservctl +/*250*/ .word sys_64_mremap, sys_sysctl, sys_getsid, sys_fdatasync, sys_nis_syscall .word sys_sync_file_range, sys_clock_settime, sys_clock_gettime, sys_clock_getres, sys_clock_nanosleep /*260*/ .word sys_sched_getaffinity, sys_sched_setaffinity, sys_timer_settime, sys_timer_gettime, sys_timer_getoverrun .word sys_timer_delete, sys_timer_create, sys_ni_syscall, sys_io_setup, sys_io_destroy diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index a0e866d233ee..54edb207ff3a 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -672,7 +672,7 @@ ia32_sys_call_table: .quad sys32_vm86_warning /* vm86 */ .quad quiet_ni_syscall /* query_module */ .quad sys_poll - .quad compat_sys_nfsservctl + .quad quiet_ni_syscall /* old nfsservctl */ .quad sys_setresgid16 /* 170 */ .quad sys_getresgid16 .quad sys_prctl diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index d92641cc7acc..201040573444 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -414,7 +414,7 @@ __SYSCALL(__NR_query_module, sys_ni_syscall) __SYSCALL(__NR_quotactl, sys_quotactl) #define __NR_nfsservctl 180 -__SYSCALL(__NR_nfsservctl, sys_nfsservctl) +__SYSCALL(__NR_nfsservctl, sys_ni_syscall) /* reserved for LiS/STREAMS */ #define __NR_getpmsg 181 diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index fbb0a045a1a2..bc19be332bc9 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -168,7 +168,7 @@ ENTRY(sys_call_table) .long ptregs_vm86 .long sys_ni_syscall /* Old sys_query_module */ .long sys_poll - .long sys_nfsservctl + .long sys_ni_syscall /* Old nfsservctl */ .long sys_setresgid16 /* 170 */ .long sys_getresgid16 .long sys_prctl diff --git a/arch/xtensa/include/asm/unistd.h b/arch/xtensa/include/asm/unistd.h index a6f934f37f1a..798ee6d285a1 100644 --- a/arch/xtensa/include/asm/unistd.h +++ b/arch/xtensa/include/asm/unistd.h @@ -455,7 +455,7 @@ __SYSCALL(203, sys_reboot, 3) #define __NR_quotactl 204 __SYSCALL(204, sys_quotactl, 4) #define __NR_nfsservctl 205 -__SYSCALL(205, sys_nfsservctl, 3) +__SYSCALL(205, sys_ni_syscall, 0) #define __NR__sysctl 206 __SYSCALL(206, sys_sysctl, 1) #define __NR_bdflush 207 diff --git a/fs/compat.c b/fs/compat.c index 0b48d018e38a..58b1da459893 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1675,11 +1675,6 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds, } #endif /* HAVE_SET_RESTORE_SIGMASK */ -long asmlinkage compat_sys_nfsservctl(int cmd, void *notused, void *notused2) -{ - return sys_ni_syscall(); -} - #ifdef CONFIG_EPOLL #ifdef HAVE_SET_RESTORE_SIGMASK diff --git a/include/asm-generic/unistd.h b/include/asm-generic/unistd.h index 4f76959397fa..f4c38d8c6674 100644 --- a/include/asm-generic/unistd.h +++ b/include/asm-generic/unistd.h @@ -143,7 +143,7 @@ __SYSCALL(__NR_pivot_root, sys_pivot_root) /* fs/nfsctl.c */ #define __NR_nfsservctl 42 -__SC_COMP(__NR_nfsservctl, sys_nfsservctl, compat_sys_nfsservctl) +__SYSCALL(__NR_nfsservctl, sys_ni_syscall) /* fs/open.c */ #define __NR3264_statfs 43 diff --git a/include/linux/compat.h b/include/linux/compat.h index 8779405e15a8..c6e7523bf765 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -438,7 +438,6 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds, struct compat_timespec __user *tsp, const compat_sigset_t __user *sigmask, compat_size_t sigsetsize); -asmlinkage long compat_sys_nfsservctl(int cmd, void *notused, void *notused2); asmlinkage long compat_sys_signalfd4(int ufd, const compat_sigset_t __user *sigmask, compat_size_t sigsetsize, int flags); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 8c03b98df5f9..1ff0ec2a5e8d 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -702,9 +702,6 @@ asmlinkage long sys_sysctl(struct __sysctl_args __user *args); asmlinkage long sys_sysinfo(struct sysinfo __user *info); asmlinkage long sys_sysfs(int option, unsigned long arg1, unsigned long arg2); -asmlinkage long sys_nfsservctl(int cmd, - struct nfsctl_arg __user *arg, - void __user *res); asmlinkage long sys_syslog(int type, char __user *buf, int len); asmlinkage long sys_uselib(const char __user *library); asmlinkage long sys_ni_syscall(void); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 62cbc8877fef..a9a5de07c4f1 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -16,7 +16,6 @@ asmlinkage long sys_ni_syscall(void) return -ENOSYS; } -cond_syscall(sys_nfsservctl); cond_syscall(sys_quotactl); cond_syscall(sys32_quotactl); cond_syscall(sys_acct); -- cgit v1.3-8-gc7d7 From c259e01a1ec90063042f758e409cd26b2a0963c8 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 22 Jun 2011 19:47:00 +0200 Subject: sched: Separate the scheduler entry for preemption Block-IO and workqueues call into notifier functions from the scheduler core code with interrupts and preemption disabled. These calls should be made before entering the scheduler core. To simplify this, separate the scheduler core code into __schedule(). __schedule() is directly called from the places which set PREEMPT_ACTIVE and from schedule(). This allows us to add the work checks into schedule(), so they are only called when a task voluntary goes to sleep. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Cc: Tejun Heo Cc: Jens Axboe Cc: Linus Torvalds Cc: stable@kernel.org # 2.6.39+ Link: http://lkml.kernel.org/r/20110622174918.813258321@linutronix.de Signed-off-by: Ingo Molnar --- kernel/sched.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..ec15e8129cf7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4279,9 +4279,9 @@ pick_next_task(struct rq *rq) } /* - * schedule() is the main scheduler function. + * __schedule() is the main scheduler function. */ -asmlinkage void __sched schedule(void) +static void __sched __schedule(void) { struct task_struct *prev, *next; unsigned long *switch_count; @@ -4369,6 +4369,11 @@ need_resched: if (need_resched()) goto need_resched; } + +asmlinkage void schedule(void) +{ + __schedule(); +} EXPORT_SYMBOL(schedule); #ifdef CONFIG_MUTEX_SPIN_ON_OWNER @@ -4435,7 +4440,7 @@ asmlinkage void __sched notrace preempt_schedule(void) do { add_preempt_count_notrace(PREEMPT_ACTIVE); - schedule(); + __schedule(); sub_preempt_count_notrace(PREEMPT_ACTIVE); /* @@ -4463,7 +4468,7 @@ asmlinkage void __sched preempt_schedule_irq(void) do { add_preempt_count(PREEMPT_ACTIVE); local_irq_enable(); - schedule(); + __schedule(); local_irq_disable(); sub_preempt_count(PREEMPT_ACTIVE); @@ -5588,7 +5593,7 @@ static inline int should_resched(void) static void __cond_resched(void) { add_preempt_count(PREEMPT_ACTIVE); - schedule(); + __schedule(); sub_preempt_count(PREEMPT_ACTIVE); } -- cgit v1.3-8-gc7d7 From 9c40cef2b799f9b5e7fa5de4d2ad3a0168ba118c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 22 Jun 2011 19:47:01 +0200 Subject: sched: Move blk_schedule_flush_plug() out of __schedule() There is no real reason to run blk_schedule_flush_plug() with interrupts and preemption disabled. Move it into schedule() and call it when the task is going voluntarily to sleep. There might be false positives when the task is woken between that call and actually scheduling, but that's not really different from being woken immediately after switching away. This fixes a deadlock in the scheduler where the blk_schedule_flush_plug() callchain enables interrupts and thereby allows a wakeup to happen of the task that's going to sleep. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Cc: Tejun Heo Cc: Jens Axboe Cc: Linus Torvalds Cc: stable@kernel.org # 2.6.39+ Link: http://lkml.kernel.org/n/tip-dwfxtra7yg1b5r65m32ywtct@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index ec15e8129cf7..511732c39b6e 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4322,16 +4322,6 @@ need_resched: if (to_wakeup) try_to_wake_up_local(to_wakeup); } - - /* - * If we are going to sleep and we have plugged IO - * queued, make sure to submit it to avoid deadlocks. - */ - if (blk_needs_flush_plug(prev)) { - raw_spin_unlock(&rq->lock); - blk_schedule_flush_plug(prev); - raw_spin_lock(&rq->lock); - } } switch_count = &prev->nvcsw; } @@ -4370,8 +4360,23 @@ need_resched: goto need_resched; } +static inline void sched_submit_work(struct task_struct *tsk) +{ + if (!tsk->state) + return; + /* + * If we are going to sleep and we have plugged IO queued, + * make sure to submit it to avoid deadlocks. + */ + if (blk_needs_flush_plug(tsk)) + blk_schedule_flush_plug(tsk); +} + asmlinkage void schedule(void) { + struct task_struct *tsk = current; + + sched_submit_work(tsk); __schedule(); } EXPORT_SYMBOL(schedule); -- cgit v1.3-8-gc7d7 From feff8fa0075bdfd43c841e9d689ed81adda988d6 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Thu, 18 Aug 2011 20:36:57 +0800 Subject: sched: Fix a memory leak in __sdt_free() This patch fixes the following memory leak: unreferenced object 0xffff880107266800 (size 512): comm "sched-powersave", pid 3718, jiffies 4323097853 (age 27495.450s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] create_object+0x187/0x28b [] kmemleak_alloc+0x73/0x98 [] __kmalloc_node+0x104/0x159 [] kzalloc_node.clone.97+0x15/0x17 [] build_sched_domains+0xb7/0x7f3 [] partition_sched_domains+0x1db/0x24a [] do_rebuild_sched_domains+0x3b/0x47 [] rebuild_sched_domains+0x10/0x12 [] sched_power_savings_store+0x6c/0x7b [] sched_mc_power_savings_store+0x16/0x18 [] sysdev_class_store+0x20/0x22 [] sysfs_write_file+0x108/0x144 [] vfs_write+0xaf/0x102 [] sys_write+0x4d/0x74 [] system_call_fastpath+0x16/0x1b [] 0xffffffffffffffff Signed-off-by: WANG Cong Signed-off-by: Peter Zijlstra Cc: stable@kernel.org # 3.0 Link: http://lkml.kernel.org/r/1313671017-4112-1-git-send-email-amwang@redhat.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 511732c39b6e..c79e7c63a4aa 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7453,6 +7453,7 @@ static void __sdt_free(const struct cpumask *cpu_map) struct sched_domain *sd = *per_cpu_ptr(sdd->sd, j); if (sd && (sd->flags & SD_OVERLAP)) free_sched_groups(sd->groups, 0); + kfree(*per_cpu_ptr(sdd->sd, j)); kfree(*per_cpu_ptr(sdd->sg, j)); kfree(*per_cpu_ptr(sdd->sgp, j)); } -- cgit v1.3-8-gc7d7 From a8d757ef076f0f95f13a918808824058de25b3eb Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 25 Aug 2011 15:58:03 +0200 Subject: perf events: Fix slow and broken cgroup context switch code The current cgroup context switch code was incorrect leading to bogus counts. Furthermore, as soon as there was an active cgroup event on a CPU, the context switch cost on that CPU would increase by a significant amount as demonstrated by a simple ping/pong example: $ ./pong Both processes pinned to CPU1, running for 10s 10684.51 ctxsw/s Now start a cgroup perf stat: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 100 $ ./pong Both processes pinned to CPU1, running for 10s 6674.61 ctxsw/s That's a 37% penalty. Note that pong is not even in the monitored cgroup. The results shown by perf stat are bogus: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 100 Performance counter stats for 'sleep 100': CPU1 cycles test CPU1 16,984,189,138 cycles # 0.000 GHz The second 'cycles' event should report a count @ CPU clock (here 2.4GHz) as it is counting across all cgroups. The patch below fixes the bogus accounting and bypasses any cgroup switches in case the outgoing and incoming tasks are in the same cgroup. With this patch the same test now yields: $ ./pong Both processes pinned to CPU1, running for 10s 10775.30 ctxsw/s Start perf stat with cgroup: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Run pong outside the cgroup: $ /pong Both processes pinned to CPU1, running for 10s 10687.80 ctxsw/s The penalty is now less than 2%. And the results for perf stat are correct: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Performance counter stats for 'sleep 10': CPU1 cycles test # 0.000 GHz CPU1 23,933,981,448 cycles # 0.000 GHz Now perf stat reports the correct counts for for the non cgroup event. If we run pong inside the cgroup, then we also get the correct counts: $ perf stat -e cycles,cycles -A -a -G test -C 1 -- sleep 10 Performance counter stats for 'sleep 10': CPU1 22,297,726,205 cycles test # 0.000 GHz CPU1 23,933,981,448 cycles # 0.000 GHz 10.001457237 seconds time elapsed Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110825135803.GA4697@quad Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 24 +++++++++++------- kernel/events/core.c | 63 ++++++++++++++++++++++++++++++++++++++-------- kernel/sched.c | 2 +- 3 files changed, 69 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 245bafdafd5e..c816075c01ce 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -944,8 +944,10 @@ extern void perf_pmu_unregister(struct pmu *pmu); extern int perf_num_counters(void); extern const char *perf_pmu_name(void); -extern void __perf_event_task_sched_in(struct task_struct *task); -extern void __perf_event_task_sched_out(struct task_struct *task, struct task_struct *next); +extern void __perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task); +extern void __perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next); extern int perf_event_init_task(struct task_struct *child); extern void perf_event_exit_task(struct task_struct *child); extern void perf_event_free_task(struct task_struct *task); @@ -1059,17 +1061,20 @@ perf_sw_event(u32 event_id, u64 nr, struct pt_regs *regs, u64 addr) extern struct jump_label_key perf_sched_events; -static inline void perf_event_task_sched_in(struct task_struct *task) +static inline void perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { if (static_branch(&perf_sched_events)) - __perf_event_task_sched_in(task); + __perf_event_task_sched_in(prev, task); } -static inline void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next) +static inline void perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next) { perf_sw_event(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, NULL, 0); - __perf_event_task_sched_out(task, next); + if (static_branch(&perf_sched_events)) + __perf_event_task_sched_out(prev, next); } extern void perf_event_mmap(struct vm_area_struct *vma); @@ -1139,10 +1144,11 @@ extern void perf_event_disable(struct perf_event *event); extern void perf_event_task_tick(void); #else static inline void -perf_event_task_sched_in(struct task_struct *task) { } +perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { } static inline void -perf_event_task_sched_out(struct task_struct *task, - struct task_struct *next) { } +perf_event_task_sched_out(struct task_struct *prev, + struct task_struct *next) { } static inline int perf_event_init_task(struct task_struct *child) { return 0; } static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } diff --git a/kernel/events/core.c b/kernel/events/core.c index b8785e26ee1c..45847fbb599a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -399,14 +399,54 @@ void perf_cgroup_switch(struct task_struct *task, int mode) local_irq_restore(flags); } -static inline void perf_cgroup_sched_out(struct task_struct *task) +static inline void perf_cgroup_sched_out(struct task_struct *task, + struct task_struct *next) { - perf_cgroup_switch(task, PERF_CGROUP_SWOUT); + struct perf_cgroup *cgrp1; + struct perf_cgroup *cgrp2 = NULL; + + /* + * we come here when we know perf_cgroup_events > 0 + */ + cgrp1 = perf_cgroup_from_task(task); + + /* + * next is NULL when called from perf_event_enable_on_exec() + * that will systematically cause a cgroup_switch() + */ + if (next) + cgrp2 = perf_cgroup_from_task(next); + + /* + * only schedule out current cgroup events if we know + * that we are switching to a different cgroup. Otherwise, + * do no touch the cgroup events. + */ + if (cgrp1 != cgrp2) + perf_cgroup_switch(task, PERF_CGROUP_SWOUT); } -static inline void perf_cgroup_sched_in(struct task_struct *task) +static inline void perf_cgroup_sched_in(struct task_struct *prev, + struct task_struct *task) { - perf_cgroup_switch(task, PERF_CGROUP_SWIN); + struct perf_cgroup *cgrp1; + struct perf_cgroup *cgrp2 = NULL; + + /* + * we come here when we know perf_cgroup_events > 0 + */ + cgrp1 = perf_cgroup_from_task(task); + + /* prev can never be NULL */ + cgrp2 = perf_cgroup_from_task(prev); + + /* + * only need to schedule in cgroup events if we are changing + * cgroup during ctxsw. Cgroup events were not scheduled + * out of ctxsw out if that was not the case. + */ + if (cgrp1 != cgrp2) + perf_cgroup_switch(task, PERF_CGROUP_SWIN); } static inline int perf_cgroup_connect(int fd, struct perf_event *event, @@ -518,11 +558,13 @@ static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx) { } -static inline void perf_cgroup_sched_out(struct task_struct *task) +static inline void perf_cgroup_sched_out(struct task_struct *task, + struct task_struct *next) { } -static inline void perf_cgroup_sched_in(struct task_struct *task) +static inline void perf_cgroup_sched_in(struct task_struct *prev, + struct task_struct *task) { } @@ -1988,7 +2030,7 @@ void __perf_event_task_sched_out(struct task_struct *task, * cgroup event are system-wide mode only */ if (atomic_read(&__get_cpu_var(perf_cgroup_events))) - perf_cgroup_sched_out(task); + perf_cgroup_sched_out(task, next); } static void task_ctx_sched_out(struct perf_event_context *ctx) @@ -2153,7 +2195,8 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx, * accessing the event control register. If a NMI hits, then it will * keep the event running. */ -void __perf_event_task_sched_in(struct task_struct *task) +void __perf_event_task_sched_in(struct task_struct *prev, + struct task_struct *task) { struct perf_event_context *ctx; int ctxn; @@ -2171,7 +2214,7 @@ void __perf_event_task_sched_in(struct task_struct *task) * cgroup event are system-wide mode only */ if (atomic_read(&__get_cpu_var(perf_cgroup_events))) - perf_cgroup_sched_in(task); + perf_cgroup_sched_in(prev, task); } static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count) @@ -2427,7 +2470,7 @@ static void perf_event_enable_on_exec(struct perf_event_context *ctx) * ctxswin cgroup events which are already scheduled * in. */ - perf_cgroup_sched_out(current); + perf_cgroup_sched_out(current, NULL); raw_spin_lock(&ctx->lock); task_ctx_sched_out(ctx); diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..0408cdc6d572 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3065,7 +3065,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_disable(); #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ - perf_event_task_sched_in(current); + perf_event_task_sched_in(prev, current); #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_enable(); #endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ -- cgit v1.3-8-gc7d7 From 86b6ef21b80ac6565d172cdab4384404de007eea Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 22 Aug 2011 09:41:46 -0400 Subject: tracing: Add preempt disable for filter self test The self testing for event filters does not really need preemption disabled as there are no races at the time of testing, but the functions it calls uses rcu_dereference_sched() which will complain if preemption is not disabled. Cc: Jiri Olsa Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 6a642e278241..816d3d074979 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -2101,6 +2101,11 @@ static __init int ftrace_test_event_filter(void) break; } + /* + * The preemption disabling is not really needed for self + * tests, but the rcu dereference will complain without it. + */ + preempt_disable(); if (*d->not_visited) walk_pred_tree(filter->preds, filter->root, test_walk_pred_cb, @@ -2108,6 +2113,7 @@ static __init int ftrace_test_event_filter(void) test_pred_visited = 0; err = filter_match_preds(filter, &d->rec); + preempt_enable(); __free_filter(filter); -- cgit v1.3-8-gc7d7 From f81ab074c30234b07c8309c542cafd07bed721f7 Mon Sep 17 00:00:00 2001 From: Vaibhav Nagarnaik Date: Tue, 16 Aug 2011 14:46:15 -0700 Subject: trace: Add a new readonly entry to report total buffer size The current file "buffer_size_kb" reports the size of per-cpu buffer and not the overall memory allocated which could be misleading. A new file "buffer_total_size_kb" adds up all the enabled CPU buffer sizes and reports it. This is only a readonly entry. Signed-off-by: Vaibhav Nagarnaik Cc: Michael Rubin Cc: David Sharp Link: http://lkml.kernel.org/r/1313531179-9323-2-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index e5df02c69b1d..01176788c387 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3568,6 +3568,30 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, return cnt; } +static ssize_t +tracing_total_entries_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + struct trace_array *tr = filp->private_data; + char buf[64]; + int r, cpu; + unsigned long size = 0, expanded_size = 0; + + mutex_lock(&trace_types_lock); + for_each_tracing_cpu(cpu) { + size += tr->entries >> 10; + if (!ring_buffer_expanded) + expanded_size += trace_buf_size >> 10; + } + if (ring_buffer_expanded) + r = sprintf(buf, "%lu\n", size); + else + r = sprintf(buf, "%lu (expanded: %lu)\n", size, expanded_size); + mutex_unlock(&trace_types_lock); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + static ssize_t tracing_free_buffer_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) @@ -3739,6 +3763,12 @@ static const struct file_operations tracing_entries_fops = { .llseek = generic_file_llseek, }; +static const struct file_operations tracing_total_entries_fops = { + .open = tracing_open_generic, + .read = tracing_total_entries_read, + .llseek = generic_file_llseek, +}; + static const struct file_operations tracing_free_buffer_fops = { .write = tracing_free_buffer_write, .release = tracing_free_buffer_release, @@ -4450,6 +4480,9 @@ static __init int tracer_init_debugfs(void) trace_create_file("buffer_size_kb", 0644, d_tracer, &global_trace, &tracing_entries_fops); + trace_create_file("buffer_total_size_kb", 0444, d_tracer, + &global_trace, &tracing_total_entries_fops); + trace_create_file("free_buffer", 0644, d_tracer, &global_trace, &tracing_free_buffer_fops); -- cgit v1.3-8-gc7d7 From c64e148a3be3cb786534ad38298c25c833116c26 Mon Sep 17 00:00:00 2001 From: Vaibhav Nagarnaik Date: Tue, 16 Aug 2011 14:46:16 -0700 Subject: trace: Add ring buffer stats to measure rate of events The stats file under per_cpu folder provides the number of entries, overruns and other statistics about the CPU ring buffer. However, the numbers do not provide any indication of how full the ring buffer is in bytes compared to the overall size in bytes. Also, it is helpful to know the rate at which the cpu buffer is filling up. This patch adds an entry "bytes: " in printed stats for per_cpu ring buffer which provides the actual bytes consumed in the ring buffer. This field includes the number of bytes used by recorded events and the padding bytes added when moving the tail pointer to next page. It also adds the following time stamps: "oldest event ts:" - the oldest timestamp in the ring buffer "now ts:" - the timestamp at the time of reading The field "now ts" provides a consistent time snapshot to the userspace when being read. This is read from the same trace clock used by tracing event timestamps. Together, these values provide the rate at which the buffer is filling up, from the formula: bytes / (now_ts - oldest_event_ts) Signed-off-by: Vaibhav Nagarnaik Cc: Michael Rubin Cc: David Sharp Link: http://lkml.kernel.org/r/1313531179-9323-3-git-send-email-vnagarnaik@google.com Signed-off-by: Steven Rostedt --- include/linux/ring_buffer.h | 2 ++ kernel/trace/ring_buffer.c | 70 ++++++++++++++++++++++++++++++++++++++++++++- kernel/trace/trace.c | 13 +++++++++ 3 files changed, 84 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index b891de96000f..67be0376d8e3 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -154,6 +154,8 @@ void ring_buffer_record_enable(struct ring_buffer *buffer); void ring_buffer_record_disable_cpu(struct ring_buffer *buffer, int cpu); void ring_buffer_record_enable_cpu(struct ring_buffer *buffer, int cpu); +unsigned long ring_buffer_oldest_event_ts(struct ring_buffer *buffer, int cpu); +unsigned long ring_buffer_bytes_cpu(struct ring_buffer *buffer, int cpu); unsigned long ring_buffer_entries(struct ring_buffer *buffer); unsigned long ring_buffer_overruns(struct ring_buffer *buffer); unsigned long ring_buffer_entries_cpu(struct ring_buffer *buffer, int cpu); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 731201bf4acc..acf6b68dc4a8 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -488,12 +488,14 @@ struct ring_buffer_per_cpu { struct buffer_page *reader_page; unsigned long lost_events; unsigned long last_overrun; + local_t entries_bytes; local_t commit_overrun; local_t overrun; local_t entries; local_t committing; local_t commits; unsigned long read; + unsigned long read_bytes; u64 write_stamp; u64 read_stamp; }; @@ -1708,6 +1710,7 @@ rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer, * the counters. */ local_add(entries, &cpu_buffer->overrun); + local_sub(BUF_PAGE_SIZE, &cpu_buffer->entries_bytes); /* * The entries will be zeroed out when we move the @@ -1863,6 +1866,9 @@ rb_reset_tail(struct ring_buffer_per_cpu *cpu_buffer, event = __rb_page_index(tail_page, tail); kmemcheck_annotate_bitfield(event, bitfield); + /* account for padding bytes */ + local_add(BUF_PAGE_SIZE - tail, &cpu_buffer->entries_bytes); + /* * Save the original length to the meta data. * This will be used by the reader to add lost event @@ -2054,6 +2060,9 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, if (!tail) tail_page->page->time_stamp = ts; + /* account for these added bytes */ + local_add(length, &cpu_buffer->entries_bytes); + return event; } @@ -2076,6 +2085,7 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, if (bpage->page == (void *)addr && rb_page_write(bpage) == old_index) { unsigned long write_mask = local_read(&bpage->write) & ~RB_WRITE_MASK; + unsigned long event_length = rb_event_length(event); /* * This is on the tail page. It is possible that * a write could come in and move the tail page @@ -2085,8 +2095,11 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, old_index += write_mask; new_index += write_mask; index = local_cmpxchg(&bpage->write, old_index, new_index); - if (index == old_index) + if (index == old_index) { + /* update counters */ + local_sub(event_length, &cpu_buffer->entries_bytes); return 1; + } } /* could not discard */ @@ -2660,6 +2673,58 @@ rb_num_of_entries(struct ring_buffer_per_cpu *cpu_buffer) (local_read(&cpu_buffer->overrun) + cpu_buffer->read); } +/** + * ring_buffer_oldest_event_ts - get the oldest event timestamp from the buffer + * @buffer: The ring buffer + * @cpu: The per CPU buffer to read from. + */ +unsigned long ring_buffer_oldest_event_ts(struct ring_buffer *buffer, int cpu) +{ + unsigned long flags; + struct ring_buffer_per_cpu *cpu_buffer; + struct buffer_page *bpage; + unsigned long ret; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return 0; + + cpu_buffer = buffer->buffers[cpu]; + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + /* + * if the tail is on reader_page, oldest time stamp is on the reader + * page + */ + if (cpu_buffer->tail_page == cpu_buffer->reader_page) + bpage = cpu_buffer->reader_page; + else + bpage = rb_set_head_page(cpu_buffer); + ret = bpage->page->time_stamp; + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(ring_buffer_oldest_event_ts); + +/** + * ring_buffer_bytes_cpu - get the number of bytes consumed in a cpu buffer + * @buffer: The ring buffer + * @cpu: The per CPU buffer to read from. + */ +unsigned long ring_buffer_bytes_cpu(struct ring_buffer *buffer, int cpu) +{ + struct ring_buffer_per_cpu *cpu_buffer; + unsigned long ret; + + if (!cpumask_test_cpu(cpu, buffer->cpumask)) + return 0; + + cpu_buffer = buffer->buffers[cpu]; + ret = local_read(&cpu_buffer->entries_bytes) - cpu_buffer->read_bytes; + + return ret; +} +EXPORT_SYMBOL_GPL(ring_buffer_bytes_cpu); + /** * ring_buffer_entries_cpu - get the number of entries in a cpu buffer * @buffer: The ring buffer @@ -3527,11 +3592,13 @@ rb_reset_cpu(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->reader_page->read = 0; local_set(&cpu_buffer->commit_overrun, 0); + local_set(&cpu_buffer->entries_bytes, 0); local_set(&cpu_buffer->overrun, 0); local_set(&cpu_buffer->entries, 0); local_set(&cpu_buffer->committing, 0); local_set(&cpu_buffer->commits, 0); cpu_buffer->read = 0; + cpu_buffer->read_bytes = 0; cpu_buffer->write_stamp = 0; cpu_buffer->read_stamp = 0; @@ -3918,6 +3985,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer, } else { /* update the entry counter */ cpu_buffer->read += rb_page_entries(reader); + cpu_buffer->read_bytes += BUF_PAGE_SIZE; /* swap the pages */ rb_init_page(bpage); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 01176788c387..b41907000bc6 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -4056,6 +4056,8 @@ tracing_stats_read(struct file *filp, char __user *ubuf, struct trace_array *tr = &global_trace; struct trace_seq *s; unsigned long cnt; + unsigned long long t; + unsigned long usec_rem; s = kmalloc(sizeof(*s), GFP_KERNEL); if (!s) @@ -4072,6 +4074,17 @@ tracing_stats_read(struct file *filp, char __user *ubuf, cnt = ring_buffer_commit_overrun_cpu(tr->buffer, cpu); trace_seq_printf(s, "commit overrun: %ld\n", cnt); + cnt = ring_buffer_bytes_cpu(tr->buffer, cpu); + trace_seq_printf(s, "bytes: %ld\n", cnt); + + t = ns2usecs(ring_buffer_oldest_event_ts(tr->buffer, cpu)); + usec_rem = do_div(t, USEC_PER_SEC); + trace_seq_printf(s, "oldest event ts: %5llu.%06lu\n", t, usec_rem); + + t = ns2usecs(ring_buffer_time_stamp(tr->buffer, cpu)); + usec_rem = do_div(t, USEC_PER_SEC); + trace_seq_printf(s, "now ts: %5llu.%06lu\n", t, usec_rem); + count = simple_read_from_buffer(ubuf, count, ppos, s->buffer, s->len); kfree(s); -- cgit v1.3-8-gc7d7 From 5f12a761937373d2f9b557d7519e6f1cf738b8f0 Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Thu, 11 Aug 2011 12:31:20 +0100 Subject: perf: provide PMU when initing events Currently, an event's 'pmu' field is set after pmu::event_init() is called. This means that pmu::event_init() must figure out which struct pmu the event was initialised from. This makes it difficult to consolidate common event initialisation code for similar PMUs, and very difficult to implement drivers for PMUs which can have multiple instances (e.g. a USB controller PMU, a GPU PMU, etc). This patch sets the 'pmu' field before initialising the event, allowing event init code to identify the struct pmu instance easily. In the event of failure to initialise an event, the event is destroyed via kfree() without calling perf_event::destroy(), so this shouldn't result in bad behaviour even if the destroy field was set before failure to initialise was noted. Signed-off-by: Mark Rutland Reviewed-by: Will Deacon Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1313062280-19123-1-git-send-email-mark.rutland@arm.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index b8785e26ee1c..68c8017de969 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5715,6 +5715,7 @@ struct pmu *perf_init_event(struct perf_event *event) pmu = idr_find(&pmu_idr, event->attr.type); rcu_read_unlock(); if (pmu) { + event->pmu = pmu; ret = pmu->event_init(event); if (ret) pmu = ERR_PTR(ret); @@ -5722,6 +5723,7 @@ struct pmu *perf_init_event(struct perf_event *event) } list_for_each_entry_rcu(pmu, &pmus, entry) { + event->pmu = pmu; ret = pmu->event_init(event); if (!ret) goto unlock; @@ -5848,8 +5850,6 @@ done: return ERR_PTR(err); } - event->pmu = pmu; - if (!event->parent) { if (event->attach_state & PERF_ATTACH_TASK) jump_label_inc(&perf_sched_events); -- cgit v1.3-8-gc7d7 From 7f310a5d4e8525ac0cc2f58c973d2100ce034410 Mon Sep 17 00:00:00 2001 From: Eric B Munson Date: Thu, 23 Jun 2011 16:34:38 -0400 Subject: perf_event: Fix broken calc_timer_values() We detected a serious issue with PERF_SAMPLE_READ and timing information when events were being multiplexing. Samples would have time_running > time_enabled. That was easy to reproduce with a libpfm4 example (ran 3 times to cause multiplexing on Core 2): $ syst_smpl -e uops_retired:freq=1 & $ syst_smpl -e uops_retired:freq=1 & $ syst_smpl -e uops_retired:freq=1 & IIP:0x0000000040062d ... PERIOD:2355332948 ENA=40144625315 RUN=60014875184 syst_smpl: WARNING: time_running > time_enabled 63277537998 uops_retired:freq=1 , scaled The bug was not present in kernel up to (and including) 3.0. It turns out the bug was introduced by the following commit: commit c4794295917ebeda8013b6cb9c8d71ab4f74a1fa events: Move lockless timer calculation into helper function The parameters of the function got reversed yet the call sites were not updated to reflect the change. That lead to time_running and time_enabled being swapped. That had no effect when there was no multiplexing because in that case time_running = time_enabled but it would show up in any other scenario. Signed-off-by: Stephane Eranian Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110829124112.GA4828@quad Signed-off-by: Ingo Molnar --- kernel/events/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 45847fbb599a..0f857782d06f 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3396,8 +3396,8 @@ static int perf_event_index(struct perf_event *event) } static void calc_timer_values(struct perf_event *event, - u64 *running, - u64 *enabled) + u64 *enabled, + u64 *running) { u64 now, ctx_time; -- cgit v1.3-8-gc7d7 From 6beea0cda8ce71c01354e688e5735c47e331e84f Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Wed, 24 Aug 2011 09:37:48 +0200 Subject: nohz: Fix update_ts_time_stat idle accounting update_ts_time_stat currently updates idle time even if we are in iowait loop at the moment. The only real users of the idle counter (via get_cpu_idle_time_us) are CPU governors and they expect to get cumulative time for both idle and iowait times. The value (idle_sleeptime) is also printed to userspace by print_cpu but it prints both idle and iowait times so the idle part is misleading. Let's clean this up and fix update_ts_time_stat to account both counters properly and update consumers of idle to consider iowait time as well. If we do this we might use get_cpu_{idle,iowait}_time_us from other contexts as well and we will get expected values. Signed-off-by: Michal Hocko Cc: Dave Jones Cc: Arnd Bergmann Cc: Alexey Dobriyan Link: http://lkml.kernel.org/r/e9c909c221a8da402c4da07e4cd968c3218f8eb1.1314172057.git.mhocko@suse.cz Signed-off-by: Thomas Gleixner --- drivers/cpufreq/cpufreq_conservative.c | 4 +++- drivers/cpufreq/cpufreq_ondemand.c | 4 +++- kernel/time/tick-sched.c | 8 ++++---- 3 files changed, 10 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index 33b56e5c5c14..c97b468ee9f7 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -120,10 +120,12 @@ static inline cputime64_t get_cpu_idle_time_jiffy(unsigned int cpu, static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall) { - u64 idle_time = get_cpu_idle_time_us(cpu, wall); + u64 idle_time = get_cpu_idle_time_us(cpu, NULL); if (idle_time == -1ULL) return get_cpu_idle_time_jiffy(cpu, wall); + else + idle_time += get_cpu_iowait_time_us(cpu, wall); return idle_time; } diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index 891360edecdd..07756bddedef 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -144,10 +144,12 @@ static inline cputime64_t get_cpu_idle_time_jiffy(unsigned int cpu, static inline cputime64_t get_cpu_idle_time(unsigned int cpu, cputime64_t *wall) { - u64 idle_time = get_cpu_idle_time_us(cpu, wall); + u64 idle_time = get_cpu_idle_time_us(cpu, NULL); if (idle_time == -1ULL) return get_cpu_idle_time_jiffy(cpu, wall); + else + idle_time += get_cpu_iowait_time_us(cpu, wall); return idle_time; } diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index d5097c44b407..7ab44bca6546 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -159,9 +159,10 @@ update_ts_time_stats(int cpu, struct tick_sched *ts, ktime_t now, u64 *last_upda if (ts->idle_active) { delta = ktime_sub(now, ts->idle_entrytime); - ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); if (nr_iowait_cpu(cpu) > 0) ts->iowait_sleeptime = ktime_add(ts->iowait_sleeptime, delta); + else + ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); ts->idle_entrytime = now; } @@ -200,8 +201,7 @@ static ktime_t tick_nohz_start_idle(int cpu, struct tick_sched *ts) * @last_update_time: variable to store update time in * * Return the cummulative idle time (since boot) for a given - * CPU, in microseconds. The idle time returned includes - * the iowait time (unlike what "top" and co report). + * CPU, in microseconds. * * This time is measured via accounting rather than sampling, * and is as accurate as ktime_get() is. @@ -221,7 +221,7 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time) } EXPORT_SYMBOL_GPL(get_cpu_idle_time_us); -/* +/** * get_cpu_iowait_time_us - get the total iowait time of a cpu * @cpu: CPU number to query * @last_update_time: variable to store update time in -- cgit v1.3-8-gc7d7 From 09a1d34f8535ecf9a347ea76f7597730c2bc0c8d Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Wed, 24 Aug 2011 09:39:30 +0200 Subject: nohz: Make idle/iowait counter update conditional get_cpu_{idle,iowait}_time_us update idle/iowait counters unconditionally if the given CPU is in the idle loop. This doesn't work well outside of CPU governors which are singletons so nobody (except for IRQ) can race with them. We will need to use both functions from /proc/stat handler to properly handle nohz idle/iowait times. Make the update depend on a non NULL last_update_time argument. Signed-off-by: Michal Hocko Cc: Dave Jones Cc: Arnd Bergmann Cc: Alexey Dobriyan Link: http://lkml.kernel.org/r/11f23179472635ce52e78921d47a20216b872f23.1314172057.git.mhocko@suse.cz Signed-off-by: Thomas Gleixner --- kernel/time/tick-sched.c | 41 +++++++++++++++++++++++++++++++++++------ 1 file changed, 35 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 7ab44bca6546..664c4a365439 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -198,7 +198,8 @@ static ktime_t tick_nohz_start_idle(int cpu, struct tick_sched *ts) /** * get_cpu_idle_time_us - get the total idle time of a cpu * @cpu: CPU number to query - * @last_update_time: variable to store update time in + * @last_update_time: variable to store update time in. Do not update + * counters if NULL. * * Return the cummulative idle time (since boot) for a given * CPU, in microseconds. @@ -211,20 +212,35 @@ static ktime_t tick_nohz_start_idle(int cpu, struct tick_sched *ts) u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time) { struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); + ktime_t now, idle; if (!tick_nohz_enabled) return -1; - update_ts_time_stats(cpu, ts, ktime_get(), last_update_time); + now = ktime_get(); + if (last_update_time) { + update_ts_time_stats(cpu, ts, now, last_update_time); + idle = ts->idle_sleeptime; + } else { + if (ts->idle_active && !nr_iowait_cpu(cpu)) { + ktime_t delta = ktime_sub(now, ts->idle_entrytime); + + idle = ktime_add(ts->idle_sleeptime, delta); + } else { + idle = ts->idle_sleeptime; + } + } + + return ktime_to_us(idle); - return ktime_to_us(ts->idle_sleeptime); } EXPORT_SYMBOL_GPL(get_cpu_idle_time_us); /** * get_cpu_iowait_time_us - get the total iowait time of a cpu * @cpu: CPU number to query - * @last_update_time: variable to store update time in + * @last_update_time: variable to store update time in. Do not update + * counters if NULL. * * Return the cummulative iowait time (since boot) for a given * CPU, in microseconds. @@ -237,13 +253,26 @@ EXPORT_SYMBOL_GPL(get_cpu_idle_time_us); u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time) { struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); + ktime_t now, iowait; if (!tick_nohz_enabled) return -1; - update_ts_time_stats(cpu, ts, ktime_get(), last_update_time); + now = ktime_get(); + if (last_update_time) { + update_ts_time_stats(cpu, ts, now, last_update_time); + iowait = ts->iowait_sleeptime; + } else { + if (ts->idle_active && nr_iowait_cpu(cpu) > 0) { + ktime_t delta = ktime_sub(now, ts->idle_entrytime); + + iowait = ktime_add(ts->iowait_sleeptime, delta); + } else { + iowait = ts->iowait_sleeptime; + } + } - return ktime_to_us(ts->iowait_sleeptime); + return ktime_to_us(iowait); } EXPORT_SYMBOL_GPL(get_cpu_iowait_time_us); -- cgit v1.3-8-gc7d7 From 29c158e81c733ac7d6a75c5ee929f34fb9f92983 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 23 Aug 2011 13:20:46 +0200 Subject: nohz: Remove "Switched to NOHz mode" debugging messages When performing cpu hotplug tests the kernel printk log buffer gets flooded with pointless "Switched to NOHz mode..." messages. Especially when afterwards analyzing a dump this might have removed more interesting stuff out of the buffer. Assuming that switching to NOHz mode simply works just remove the printk. Signed-off-by: Heiko Carstens Link: http://lkml.kernel.org/r/20110823112046.GB2540@osiris.boeblingen.de.ibm.com Signed-off-by: Thomas Gleixner --- kernel/time/tick-sched.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 664c4a365439..7e2e0817cbfb 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -669,8 +669,6 @@ static void tick_nohz_switch_to_nohz(void) next = ktime_add(next, tick_period); } local_irq_enable(); - - printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id()); } /* @@ -822,10 +820,8 @@ void tick_setup_sched_timer(void) } #ifdef CONFIG_NO_HZ - if (tick_nohz_enabled) { + if (tick_nohz_enabled) ts->nohz_mode = NOHZ_MODE_HIGHRES; - printk(KERN_INFO "Switched to NOHz mode on CPU #%d\n", smp_processor_id()); - } #endif } #endif /* HIGH_RES_TIMERS */ -- cgit v1.3-8-gc7d7 From d1748302f70be7469809809283fe164156a34231 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Tue, 23 Aug 2011 15:29:42 +0200 Subject: clockevents: Make minimum delay adjustments configurable The automatic increase of the min_delta_ns of a clockevents device should be done in the clockevents code as the minimum delay is an attribute of the clockevents device. In addition not all architectures want the automatic adjustment, on a massively virtualized system it can happen that the programming of a clock event fails several times in a row because the virtual cpu has been rescheduled quickly enough. In that case the minimum delay will erroneously be increased with no way back. The new config symbol GENERIC_CLOCKEVENTS_MIN_ADJUST is used to enable the automatic adjustment. The config option is selected only for x86. Signed-off-by: Martin Schwidefsky Cc: john stultz Link: http://lkml.kernel.org/r/20110823133142.494157493@de.ibm.com Signed-off-by: Thomas Gleixner --- arch/x86/Kconfig | 1 + include/linux/clockchips.h | 2 +- kernel/time/Kconfig | 2 + kernel/time/clockevents.c | 125 ++++++++++++++++++++++++++++++++++++++----- kernel/time/tick-broadcast.c | 4 +- kernel/time/tick-common.c | 4 +- kernel/time/tick-internal.h | 2 - kernel/time/tick-oneshot.c | 77 ++------------------------ 8 files changed, 123 insertions(+), 94 deletions(-) (limited to 'kernel') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 6a47bb22657f..a1609cd7e4c6 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -68,6 +68,7 @@ config X86 select GENERIC_IRQ_PROBE select GENERIC_PENDING_IRQ if SMP select GENERIC_IRQ_SHOW + select GENERIC_CLOCKEVENTS_MIN_ADJUST select IRQ_FORCED_THREADING select USE_GENERIC_SMP_HELPERS if SMP select HAVE_BPF_JIT if (X86_64 && NET) diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index d6733e27af34..39bb050bdbb2 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -140,7 +140,7 @@ extern void clockevents_set_mode(struct clock_event_device *dev, enum clock_event_mode mode); extern int clockevents_register_notifier(struct notifier_block *nb); extern int clockevents_program_event(struct clock_event_device *dev, - ktime_t expires, ktime_t now); + ktime_t expires, bool force); extern void clockevents_handle_noop(struct clock_event_device *dev); diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index f06a8a365648..b26c2228fe92 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -27,3 +27,5 @@ config GENERIC_CLOCKEVENTS_BUILD default y depends on GENERIC_CLOCKEVENTS || GENERIC_CLOCKEVENTS_MIGR +config GENERIC_CLOCKEVENTS_MIN_ADJUST + bool diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index e4c699dfa4e8..713ef94eceef 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -94,42 +94,139 @@ void clockevents_shutdown(struct clock_event_device *dev) dev->next_event.tv64 = KTIME_MAX; } +#ifdef CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST + +/* Limit min_delta to a jiffie */ +#define MIN_DELTA_LIMIT (NSEC_PER_SEC / HZ) + +/** + * clockevents_increase_min_delta - raise minimum delta of a clock event device + * @dev: device to increase the minimum delta + * + * Returns 0 on success, -ETIME when the minimum delta reached the limit. + */ +static int clockevents_increase_min_delta(struct clock_event_device *dev) +{ + /* Nothing to do if we already reached the limit */ + if (dev->min_delta_ns >= MIN_DELTA_LIMIT) { + printk(KERN_WARNING "CE: Reprogramming failure. Giving up\n"); + dev->next_event.tv64 = KTIME_MAX; + return -ETIME; + } + + if (dev->min_delta_ns < 5000) + dev->min_delta_ns = 5000; + else + dev->min_delta_ns += dev->min_delta_ns >> 1; + + if (dev->min_delta_ns > MIN_DELTA_LIMIT) + dev->min_delta_ns = MIN_DELTA_LIMIT; + + printk(KERN_WARNING "CE: %s increased min_delta_ns to %llu nsec\n", + dev->name ? dev->name : "?", + (unsigned long long) dev->min_delta_ns); + return 0; +} + +/** + * clockevents_program_min_delta - Set clock event device to the minimum delay. + * @dev: device to program + * + * Returns 0 on success, -ETIME when the retry loop failed. + */ +static int clockevents_program_min_delta(struct clock_event_device *dev) +{ + unsigned long long clc; + int64_t delta; + int i; + + for (i = 0;;) { + delta = dev->min_delta_ns; + dev->next_event = ktime_add_ns(ktime_get(), delta); + + if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) + return 0; + + dev->retries++; + clc = ((unsigned long long) delta * dev->mult) >> dev->shift; + if (dev->set_next_event((unsigned long) clc, dev) == 0) + return 0; + + if (++i > 2) { + /* + * We tried 3 times to program the device with the + * given min_delta_ns. Try to increase the minimum + * delta, if that fails as well get out of here. + */ + if (clockevents_increase_min_delta(dev)) + return -ETIME; + i = 0; + } + } +} + +#else /* CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST */ + +/** + * clockevents_program_min_delta - Set clock event device to the minimum delay. + * @dev: device to program + * + * Returns 0 on success, -ETIME when the retry loop failed. + */ +static int clockevents_program_min_delta(struct clock_event_device *dev) +{ + unsigned long long clc; + int64_t delta; + + delta = dev->min_delta_ns; + dev->next_event = ktime_add_ns(ktime_get(), delta); + + if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) + return 0; + + dev->retries++; + clc = ((unsigned long long) delta * dev->mult) >> dev->shift; + return dev->set_next_event((unsigned long) clc, dev); +} + +#endif /* CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST */ + /** * clockevents_program_event - Reprogram the clock event device. + * @dev: device to program * @expires: absolute expiry time (monotonic clock) + * @force: program minimum delay if expires can not be set * * Returns 0 on success, -ETIME when the event is in the past. */ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, - ktime_t now) + bool force) { unsigned long long clc; int64_t delta; + int rc; if (unlikely(expires.tv64 < 0)) { WARN_ON_ONCE(1); return -ETIME; } - delta = ktime_to_ns(ktime_sub(expires, now)); - - if (delta <= 0) - return -ETIME; - dev->next_event = expires; if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) return 0; - if (delta > dev->max_delta_ns) - delta = dev->max_delta_ns; - if (delta < dev->min_delta_ns) - delta = dev->min_delta_ns; + delta = ktime_to_ns(ktime_sub(expires, ktime_get())); + if (delta <= 0) + return force ? clockevents_program_min_delta(dev) : -ETIME; - clc = delta * dev->mult; - clc >>= dev->shift; + delta = min(delta, (int64_t) dev->max_delta_ns); + delta = max(delta, (int64_t) dev->min_delta_ns); - return dev->set_next_event((unsigned long) clc, dev); + clc = ((unsigned long long) delta * dev->mult) >> dev->shift; + rc = dev->set_next_event((unsigned long) clc, dev); + + return (rc && force) ? clockevents_program_min_delta(dev) : rc; } /** @@ -258,7 +355,7 @@ int clockevents_update_freq(struct clock_event_device *dev, u32 freq) if (dev->mode != CLOCK_EVT_MODE_ONESHOT) return 0; - return clockevents_program_event(dev, dev->next_event, ktime_get()); + return clockevents_program_event(dev, dev->next_event, false); } /* diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index c7218d132738..f954282d9a82 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -194,7 +194,7 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) for (next = dev->next_event; ;) { next = ktime_add(next, tick_period); - if (!clockevents_program_event(dev, next, ktime_get())) + if (!clockevents_program_event(dev, next, false)) return; tick_do_periodic_broadcast(); } @@ -373,7 +373,7 @@ static int tick_broadcast_set_event(ktime_t expires, int force) { struct clock_event_device *bc = tick_broadcast_device.evtdev; - return tick_dev_program_event(bc, expires, force); + return clockevents_program_event(bc, expires, force); } int tick_resume_broadcast_oneshot(struct clock_event_device *bc) diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 119528de8235..da6c9ecad4e4 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -94,7 +94,7 @@ void tick_handle_periodic(struct clock_event_device *dev) */ next = ktime_add(dev->next_event, tick_period); for (;;) { - if (!clockevents_program_event(dev, next, ktime_get())) + if (!clockevents_program_event(dev, next, false)) return; /* * Have to be careful here. If we're in oneshot mode, @@ -137,7 +137,7 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); for (;;) { - if (!clockevents_program_event(dev, next, ktime_get())) + if (!clockevents_program_event(dev, next, false)) return; next = ktime_add(next, tick_period); } diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index 1009b06d6f89..4e265b901fed 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -26,8 +26,6 @@ extern void clockevents_shutdown(struct clock_event_device *dev); extern void tick_setup_oneshot(struct clock_event_device *newdev, void (*handler)(struct clock_event_device *), ktime_t nextevt); -extern int tick_dev_program_event(struct clock_event_device *dev, - ktime_t expires, int force); extern int tick_program_event(ktime_t expires, int force); extern void tick_oneshot_notify(void); extern int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)); diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index 2d04411a5f05..824109060a33 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -21,74 +21,6 @@ #include "tick-internal.h" -/* Limit min_delta to a jiffie */ -#define MIN_DELTA_LIMIT (NSEC_PER_SEC / HZ) - -static int tick_increase_min_delta(struct clock_event_device *dev) -{ - /* Nothing to do if we already reached the limit */ - if (dev->min_delta_ns >= MIN_DELTA_LIMIT) - return -ETIME; - - if (dev->min_delta_ns < 5000) - dev->min_delta_ns = 5000; - else - dev->min_delta_ns += dev->min_delta_ns >> 1; - - if (dev->min_delta_ns > MIN_DELTA_LIMIT) - dev->min_delta_ns = MIN_DELTA_LIMIT; - - printk(KERN_WARNING "CE: %s increased min_delta_ns to %llu nsec\n", - dev->name ? dev->name : "?", - (unsigned long long) dev->min_delta_ns); - return 0; -} - -/** - * tick_program_event internal worker function - */ -int tick_dev_program_event(struct clock_event_device *dev, ktime_t expires, - int force) -{ - ktime_t now = ktime_get(); - int i; - - for (i = 0;;) { - int ret = clockevents_program_event(dev, expires, now); - - if (!ret || !force) - return ret; - - dev->retries++; - /* - * We tried 3 times to program the device with the given - * min_delta_ns. If that's not working then we increase it - * and emit a warning. - */ - if (++i > 2) { - /* Increase the min. delta and try again */ - if (tick_increase_min_delta(dev)) { - /* - * Get out of the loop if min_delta_ns - * hit the limit already. That's - * better than staying here forever. - * - * We clear next_event so we have a - * chance that the box survives. - */ - printk(KERN_WARNING - "CE: Reprogramming failure. Giving up\n"); - dev->next_event.tv64 = KTIME_MAX; - return -ETIME; - } - i = 0; - } - - now = ktime_get(); - expires = ktime_add_ns(now, dev->min_delta_ns); - } -} - /** * tick_program_event */ @@ -96,7 +28,7 @@ int tick_program_event(ktime_t expires, int force) { struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); - return tick_dev_program_event(dev, expires, force); + return clockevents_program_event(dev, expires, force); } /** @@ -104,11 +36,10 @@ int tick_program_event(ktime_t expires, int force) */ void tick_resume_oneshot(void) { - struct tick_device *td = &__get_cpu_var(tick_cpu_device); - struct clock_event_device *dev = td->evtdev; + struct clock_event_device *dev = __this_cpu_read(tick_cpu_device.evtdev); clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT); - tick_program_event(ktime_get(), 1); + clockevents_program_event(dev, ktime_get(), true); } /** @@ -120,7 +51,7 @@ void tick_setup_oneshot(struct clock_event_device *newdev, { newdev->event_handler = handler; clockevents_set_mode(newdev, CLOCK_EVT_MODE_ONESHOT); - tick_dev_program_event(newdev, next_event, 1); + clockevents_program_event(newdev, next_event, true); } /** -- cgit v1.3-8-gc7d7 From 65516f8a7c2028381f0dae4c16ddb621c96158cc Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Tue, 23 Aug 2011 15:29:43 +0200 Subject: clockevents: Add direct ktime programming function There is at least one architecture (s390) with a sane clockevent device that can be programmed with the equivalent of a ktime. No need to create a delta against the current time, the ktime can be used directly. A new clock device function 'set_next_ktime' is introduced that is called with the unmodified ktime for the timer if the clock event device has the CLOCK_EVT_FEAT_KTIME bit set. Signed-off-by: Martin Schwidefsky Cc: john stultz Link: http://lkml.kernel.org/r/20110823133142.815350967@de.ibm.com Signed-off-by: Thomas Gleixner --- include/linux/clockchips.h | 10 +++++++--- kernel/time/clockevents.c | 4 ++++ 2 files changed, 11 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index 39bb050bdbb2..81e803e90aa4 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -45,20 +45,22 @@ enum clock_event_nofitiers { */ #define CLOCK_EVT_FEAT_PERIODIC 0x000001 #define CLOCK_EVT_FEAT_ONESHOT 0x000002 +#define CLOCK_EVT_FEAT_KTIME 0x000004 /* * x86(64) specific misfeatures: * * - Clockevent source stops in C3 State and needs broadcast support. * - Local APIC timer is used as a dummy device. */ -#define CLOCK_EVT_FEAT_C3STOP 0x000004 -#define CLOCK_EVT_FEAT_DUMMY 0x000008 +#define CLOCK_EVT_FEAT_C3STOP 0x000008 +#define CLOCK_EVT_FEAT_DUMMY 0x000010 /** * struct clock_event_device - clock event device descriptor * @event_handler: Assigned by the framework to be called by the low * level handler of the event source - * @set_next_event: set next event function + * @set_next_event: set next event function using a clocksource delta + * @set_next_ktime: set next event function using a direct ktime value * @next_event: local storage for the next event in oneshot mode * @max_delta_ns: maximum delta value in ns * @min_delta_ns: minimum delta value in ns @@ -81,6 +83,8 @@ struct clock_event_device { void (*event_handler)(struct clock_event_device *); int (*set_next_event)(unsigned long evt, struct clock_event_device *); + int (*set_next_ktime)(ktime_t expires, + struct clock_event_device *); ktime_t next_event; u64 max_delta_ns; u64 min_delta_ns; diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 713ef94eceef..1ecd6ba36d6c 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -216,6 +216,10 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, if (dev->mode == CLOCK_EVT_MODE_SHUTDOWN) return 0; + /* Shortcut for clockevent devices that can deal with ktime. */ + if (dev->features & CLOCK_EVT_FEAT_KTIME) + return dev->set_next_ktime(expires, dev); + delta = ktime_to_ns(ktime_sub(expires, ktime_get())); if (delta <= 0) return force ? clockevents_program_min_delta(dev) : -ETIME; -- cgit v1.3-8-gc7d7 From e8abccb719377af63cb0f1fed289db405e3def16 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 1 Sep 2011 12:42:04 +0200 Subject: posix-cpu-timers: Cure SMP accounting oddities David reported: Attached below is a watered-down version of rt/tst-cpuclock2.c from GLIBC. Just build it with "gcc -o test test.c -lpthread -lrt" or similar. Run it several times, and you will see cases where the main thread will measure a process clock difference before and after the nanosleep which is smaller than the cpu-burner thread's individual thread clock difference. This doesn't make any sense since the cpu-burner thread is part of the top-level process's thread group. I've reproduced this on both x86-64 and sparc64 (using both 32-bit and 64-bit binaries). For example: [davem@boricha build-x86_64-linux]$ ./test process: before(0.001221967) after(0.498624371) diff(497402404) thread: before(0.000081692) after(0.498316431) diff(498234739) self: before(0.001223521) after(0.001240219) diff(16698) [davem@boricha build-x86_64-linux]$ The diff of 'process' should always be >= the diff of 'thread'. I make sure to wrap the 'thread' clock measurements the most tightly around the nanosleep() call, and that the 'process' clock measurements are the outer-most ones. --- #include #include #include #include #include #include #include #include static pthread_barrier_t barrier; static void *chew_cpu(void *arg) { pthread_barrier_wait(&barrier); while (1) __asm__ __volatile__("" : : : "memory"); return NULL; } int main(void) { clockid_t process_clock, my_thread_clock, th_clock; struct timespec process_before, process_after; struct timespec me_before, me_after; struct timespec th_before, th_after; struct timespec sleeptime; unsigned long diff; pthread_t th; int err; err = clock_getcpuclockid(0, &process_clock); if (err) return 1; err = pthread_getcpuclockid(pthread_self(), &my_thread_clock); if (err) return 1; pthread_barrier_init(&barrier, NULL, 2); err = pthread_create(&th, NULL, chew_cpu, NULL); if (err) return 1; err = pthread_getcpuclockid(th, &th_clock); if (err) return 1; pthread_barrier_wait(&barrier); err = clock_gettime(process_clock, &process_before); if (err) return 1; err = clock_gettime(my_thread_clock, &me_before); if (err) return 1; err = clock_gettime(th_clock, &th_before); if (err) return 1; sleeptime.tv_sec = 0; sleeptime.tv_nsec = 500000000; nanosleep(&sleeptime, NULL); err = clock_gettime(th_clock, &th_after); if (err) return 1; err = clock_gettime(my_thread_clock, &me_after); if (err) return 1; err = clock_gettime(process_clock, &process_after); if (err) return 1; diff = process_after.tv_nsec - process_before.tv_nsec; printf("process: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", process_before.tv_sec, process_before.tv_nsec, process_after.tv_sec, process_after.tv_nsec, diff); diff = th_after.tv_nsec - th_before.tv_nsec; printf("thread: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", th_before.tv_sec, th_before.tv_nsec, th_after.tv_sec, th_after.tv_nsec, diff); diff = me_after.tv_nsec - me_before.tv_nsec; printf("self: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", me_before.tv_sec, me_before.tv_nsec, me_after.tv_sec, me_after.tv_nsec, diff); return 0; } This is due to us using p->se.sum_exec_runtime in thread_group_cputime() where we iterate the thread group and sum all data. This does not take time since the last schedule operation (tick or otherwise) into account. We can cure this by using task_sched_runtime() at the cost of having to take locks. This also means we can (and must) do away with thread_group_sched_runtime() since the modified thread_group_cputime() is now more accurate and would deadlock when called from thread_group_sched_runtime(). Reported-by: David Miller Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1314874459.7945.22.camel@twins Cc: stable@kernel.org Signed-off-by: Thomas Gleixner --- include/linux/sched.h | 1 - kernel/posix-cpu-timers.c | 5 +++-- kernel/sched.c | 24 ------------------------ 3 files changed, 3 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 20b03bf94748..2909fe7a622d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1955,7 +1955,6 @@ static inline void disable_sched_clock_irqtime(void) {} extern unsigned long long task_sched_runtime(struct task_struct *task); -extern unsigned long long thread_group_sched_runtime(struct task_struct *task); /* sched_exec is called by processes performing an exec */ #ifdef CONFIG_SMP diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 58f405b581e7..c8008dd58ef2 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -250,7 +250,7 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) do { times->utime = cputime_add(times->utime, t->utime); times->stime = cputime_add(times->stime, t->stime); - times->sum_exec_runtime += t->se.sum_exec_runtime; + times->sum_exec_runtime += task_sched_runtime(t); } while_each_thread(tsk, t); out: rcu_read_unlock(); @@ -312,7 +312,8 @@ static int cpu_clock_sample_group(const clockid_t which_clock, cpu->cpu = cputime.utime; break; case CPUCLOCK_SCHED: - cpu->sched = thread_group_sched_runtime(p); + thread_group_cputime(p, &cputime); + cpu->sched = cputime.sum_exec_runtime; break; } return 0; diff --git a/kernel/sched.c b/kernel/sched.c index ccacdbdecf45..e1290ecee3c2 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3724,30 +3724,6 @@ unsigned long long task_sched_runtime(struct task_struct *p) return ns; } -/* - * Return sum_exec_runtime for the thread group. - * In case the task is currently running, return the sum plus current's - * pending runtime that have not been accounted yet. - * - * Note that the thread group might have other running tasks as well, - * so the return value not includes other pending runtime that other - * running tasks might have. - */ -unsigned long long thread_group_sched_runtime(struct task_struct *p) -{ - struct task_cputime totals; - unsigned long flags; - struct rq *rq; - u64 ns; - - rq = task_rq_lock(p, &flags); - thread_group_cputime(p, &totals); - ns = totals.sum_exec_runtime + do_task_delta_exec(p, rq); - task_rq_unlock(rq, p, &flags); - - return ns; -} - /* * Account user cpu time to a process. * @p: the process that the cpu time gets accounted to -- cgit v1.3-8-gc7d7 From ed585a651681e822089087b426e6ebfb6d3d9873 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Sun, 11 Sep 2011 13:59:27 +0200 Subject: genirq: Make irq_shutdown() symmetric vs. irq_startup again If an irq_chip provides .irq_shutdown(), but neither of .irq_disable() or .irq_mask(), free_irq() crashes when jumping to NULL. Fix this by only trying .irq_disable() and .irq_mask() if there's no .irq_shutdown() provided. This revives the symmetry with irq_startup(), which tries .irq_startup(), .irq_enable(), and irq_unmask(), and makes it consistent with the comment for irq_chip.irq_shutdown() in , which says: * @irq_shutdown: shut down the interrupt (defaults to ->disable if NULL) This is also how __free_irq() behaved before the big overhaul, cfr. e.g. 3b56f0585fd4c02d047dc406668cb40159b2d340 ("genirq: Remove bogus conditional"), where the core interrupt code always overrode .irq_shutdown() to .irq_disable() if .irq_shutdown() was NULL. Signed-off-by: Geert Uytterhoeven Cc: linux-m68k@lists.linux-m68k.org Link: http://lkml.kernel.org/r/1315742394-16036-2-git-send-email-geert@linux-m68k.org Cc: stable@kernel.org Signed-off-by: Thomas Gleixner --- kernel/irq/chip.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index d5a3009da71a..dc5114b4c16c 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -178,7 +178,7 @@ void irq_shutdown(struct irq_desc *desc) desc->depth = 1; if (desc->irq_data.chip->irq_shutdown) desc->irq_data.chip->irq_shutdown(&desc->irq_data); - if (desc->irq_data.chip->irq_disable) + else if (desc->irq_data.chip->irq_disable) desc->irq_data.chip->irq_disable(&desc->irq_data); else desc->irq_data.chip->irq_mask(&desc->irq_data); -- cgit v1.3-8-gc7d7 From 60f96b41f71d2a13d1c0a457b8b77958f77142d1 Mon Sep 17 00:00:00 2001 From: Santosh Shilimkar Date: Fri, 9 Sep 2011 13:59:35 +0530 Subject: genirq: Add IRQCHIP_SKIP_SET_WAKE flag Some irq chips need the irq_set_wake() functionality, but do not require a irq_set_wake() callback. Instead of forcing an empty callback to be implemented add a flag which notes this fact. Check for the flag in set_irq_wake_real() and return success when set. Signed-off-by: Santosh Shilimkar Cc: Thomas Gleixner --- include/linux/irq.h | 2 ++ kernel/irq/manage.c | 3 +++ 2 files changed, 5 insertions(+) (limited to 'kernel') diff --git a/include/linux/irq.h b/include/linux/irq.h index 59517300a315..73e31abeba1c 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -336,12 +336,14 @@ struct irq_chip { * IRQCHIP_MASK_ON_SUSPEND: Mask non wake irqs in the suspend path * IRQCHIP_ONOFFLINE_ENABLED: Only call irq_on/off_line callbacks * when irq enabled + * IRQCHIP_SKIP_SET_WAKE: Skip chip.irq_set_wake(), for this irq chip */ enum { IRQCHIP_SET_TYPE_MASKED = (1 << 0), IRQCHIP_EOI_IF_HANDLED = (1 << 1), IRQCHIP_MASK_ON_SUSPEND = (1 << 2), IRQCHIP_ONOFFLINE_ENABLED = (1 << 3), + IRQCHIP_SKIP_SET_WAKE = (1 << 4), }; /* This include will go away once we isolated irq_desc usage to core code */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 9b956fa20308..7e1a3ed1e61a 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -467,6 +467,9 @@ static int set_irq_wake_real(unsigned int irq, unsigned int on) struct irq_desc *desc = irq_to_desc(irq); int ret = -ENXIO; + if (irq_desc_get_chip(desc)->flags & IRQCHIP_SKIP_SET_WAKE) + return 0; + if (desc->irq_data.chip->irq_set_wake) ret = desc->irq_data.chip->irq_set_wake(&desc->irq_data, on); -- cgit v1.3-8-gc7d7 From 0fa914c632210e09940f0bf5b85d2329b75e04e0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 8 Jun 2011 09:58:38 +0200 Subject: rtmutex: Cleanup the debug code Use the existing lock debugging macros. Signed-off-by: Thomas Gleixner --- kernel/rtmutex-debug.c | 75 +++++++------------------------------------------- 1 file changed, 10 insertions(+), 65 deletions(-) (limited to 'kernel') diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c index 3c7cbc2c33be..cb1ced6967ad 100644 --- a/kernel/rtmutex-debug.c +++ b/kernel/rtmutex-debug.c @@ -29,61 +29,6 @@ #include "rtmutex_common.h" -# define TRACE_WARN_ON(x) WARN_ON(x) -# define TRACE_BUG_ON(x) BUG_ON(x) - -# define TRACE_OFF() \ -do { \ - if (rt_trace_on) { \ - rt_trace_on = 0; \ - console_verbose(); \ - if (raw_spin_is_locked(¤t->pi_lock)) \ - raw_spin_unlock(¤t->pi_lock); \ - } \ -} while (0) - -# define TRACE_OFF_NOLOCK() \ -do { \ - if (rt_trace_on) { \ - rt_trace_on = 0; \ - console_verbose(); \ - } \ -} while (0) - -# define TRACE_BUG_LOCKED() \ -do { \ - TRACE_OFF(); \ - BUG(); \ -} while (0) - -# define TRACE_WARN_ON_LOCKED(c) \ -do { \ - if (unlikely(c)) { \ - TRACE_OFF(); \ - WARN_ON(1); \ - } \ -} while (0) - -# define TRACE_BUG_ON_LOCKED(c) \ -do { \ - if (unlikely(c)) \ - TRACE_BUG_LOCKED(); \ -} while (0) - -#ifdef CONFIG_SMP -# define SMP_TRACE_BUG_ON_LOCKED(c) TRACE_BUG_ON_LOCKED(c) -#else -# define SMP_TRACE_BUG_ON_LOCKED(c) do { } while (0) -#endif - -/* - * deadlock detection 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(): - */ -static int rt_trace_on = 1; - static void printk_task(struct task_struct *p) { if (p) @@ -111,8 +56,8 @@ static void printk_lock(struct rt_mutex *lock, int print_owner) void rt_mutex_debug_task_free(struct task_struct *task) { - WARN_ON(!plist_head_empty(&task->pi_waiters)); - WARN_ON(task->pi_blocked_on); + DEBUG_LOCKS_WARN_ON(!plist_head_empty(&task->pi_waiters)); + DEBUG_LOCKS_WARN_ON(task->pi_blocked_on); } /* @@ -125,7 +70,7 @@ void debug_rt_mutex_deadlock(int detect, struct rt_mutex_waiter *act_waiter, { struct task_struct *task; - if (!rt_trace_on || detect || !act_waiter) + if (!debug_locks || detect || !act_waiter) return; task = rt_mutex_owner(act_waiter->lock); @@ -139,7 +84,7 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) { struct task_struct *task; - if (!waiter->deadlock_lock || !rt_trace_on) + if (!waiter->deadlock_lock || !debug_locks) return; rcu_read_lock(); @@ -149,7 +94,8 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) return; } - TRACE_OFF_NOLOCK(); + if (!debug_locks_off()) + return; printk("\n============================================\n"); printk( "[ BUG: circular locking deadlock detected! ]\n"); @@ -180,7 +126,6 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) printk("[ turning off deadlock detection." "Please report this trace. ]\n\n"); - local_irq_disable(); } void debug_rt_mutex_lock(struct rt_mutex *lock) @@ -189,7 +134,7 @@ void debug_rt_mutex_lock(struct rt_mutex *lock) void debug_rt_mutex_unlock(struct rt_mutex *lock) { - TRACE_WARN_ON_LOCKED(rt_mutex_owner(lock) != current); + DEBUG_LOCKS_WARN_ON(rt_mutex_owner(lock) != current); } void @@ -199,7 +144,7 @@ debug_rt_mutex_proxy_lock(struct rt_mutex *lock, struct task_struct *powner) void debug_rt_mutex_proxy_unlock(struct rt_mutex *lock) { - TRACE_WARN_ON_LOCKED(!rt_mutex_owner(lock)); + DEBUG_LOCKS_WARN_ON(!rt_mutex_owner(lock)); } void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) @@ -213,8 +158,8 @@ void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) { put_pid(waiter->deadlock_task_pid); - TRACE_WARN_ON(!plist_node_empty(&waiter->list_entry)); - TRACE_WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); + DEBUG_LOCKS_WARN_ON(!plist_node_empty(&waiter->list_entry)); + DEBUG_LOCKS_WARN_ON(!plist_node_empty(&waiter->pi_list_entry)); memset(waiter, 0x22, sizeof(*waiter)); } -- cgit v1.3-8-gc7d7 From 9fb60336253edf73dedc527b2aa2bf32eae0d6da Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 12 Sep 2011 13:32:23 +0200 Subject: clocksource: Make watchdog reset lockless KGDB needs to trylock watchdog_lock when trying to reset the clocksource watchdog after the system has been stopped to avoid a potential deadlock. When the trylock fails TSC usually becomes unstable. We can be more clever by using an atomic counter and checking it in the clocksource_watchdog callback. We restart the watchdog whenever the counter is > 0 and only decrement the counter when we ran through a full update cycle. Signed-off-by: Thomas Gleixner Cc: John Stultz Acked-by: Jason Wessel Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1109121326280.2723@ionos Signed-off-by: Thomas Gleixner --- kernel/time/clocksource.c | 38 ++++++++++++++++++-------------------- 1 file changed, 18 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index e0980f0d9a0a..cf52fda2e096 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -186,6 +186,7 @@ static struct timer_list watchdog_timer; static DECLARE_WORK(watchdog_work, clocksource_watchdog_work); static DEFINE_SPINLOCK(watchdog_lock); static int watchdog_running; +static atomic_t watchdog_reset_pending; static int clocksource_watchdog_kthread(void *data); static void __clocksource_change_rating(struct clocksource *cs, int rating); @@ -247,12 +248,14 @@ static void clocksource_watchdog(unsigned long data) struct clocksource *cs; cycle_t csnow, wdnow; int64_t wd_nsec, cs_nsec; - int next_cpu; + int next_cpu, reset_pending; spin_lock(&watchdog_lock); if (!watchdog_running) goto out; + reset_pending = atomic_read(&watchdog_reset_pending); + list_for_each_entry(cs, &watchdog_list, wd_list) { /* Clocksource already marked unstable? */ @@ -268,7 +271,8 @@ static void clocksource_watchdog(unsigned long data) local_irq_enable(); /* Clocksource initialized ? */ - if (!(cs->flags & CLOCK_SOURCE_WATCHDOG)) { + if (!(cs->flags & CLOCK_SOURCE_WATCHDOG) || + atomic_read(&watchdog_reset_pending)) { cs->flags |= CLOCK_SOURCE_WATCHDOG; cs->wd_last = wdnow; cs->cs_last = csnow; @@ -283,8 +287,11 @@ static void clocksource_watchdog(unsigned long data) cs->cs_last = csnow; cs->wd_last = wdnow; + if (atomic_read(&watchdog_reset_pending)) + continue; + /* Check the deviation from the watchdog clocksource. */ - if (abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) { + if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) { clocksource_unstable(cs, cs_nsec - wd_nsec); continue; } @@ -302,6 +309,13 @@ static void clocksource_watchdog(unsigned long data) } } + /* + * We only clear the watchdog_reset_pending, when we did a + * full cycle through all clocksources. + */ + if (reset_pending) + atomic_dec(&watchdog_reset_pending); + /* * Cycle through CPUs to check if the CPUs stay synchronized * to each other. @@ -344,23 +358,7 @@ static inline void clocksource_reset_watchdog(void) static void clocksource_resume_watchdog(void) { - unsigned long flags; - - /* - * We use trylock here to avoid a potential dead lock when - * kgdb calls this code after the kernel has been stopped with - * watchdog_lock held. When watchdog_lock is held we just - * return and accept, that the watchdog might trigger and mark - * the monitored clock source (usually TSC) unstable. - * - * This does not affect the other caller clocksource_resume() - * because at this point the kernel is UP, interrupts are - * disabled and nothing can hold watchdog_lock. - */ - if (!spin_trylock_irqsave(&watchdog_lock, flags)) - return; - clocksource_reset_watchdog(); - spin_unlock_irqrestore(&watchdog_lock, flags); + atomic_inc(&watchdog_reset_pending); } static void clocksource_enqueue_watchdog(struct clocksource *cs) -- cgit v1.3-8-gc7d7 From ec484608c5885931c432e99ecfd2772288cd993c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Jul 2009 16:09:17 +0200 Subject: locking, kprobes: Annotate the hash locks and kretprobe.lock as raw The kprobe locks can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/linux/kprobes.h | 2 +- kernel/kprobes.c | 34 +++++++++++++++++----------------- 2 files changed, 18 insertions(+), 18 deletions(-) (limited to 'kernel') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index dd7c12e875bc..dce6e4dbeda7 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -181,7 +181,7 @@ struct kretprobe { int nmissed; size_t data_size; struct hlist_head free_instances; - spinlock_t lock; + raw_spinlock_t lock; }; struct kretprobe_instance { diff --git a/kernel/kprobes.c b/kernel/kprobes.c index b30fd54eb985..2f193d0ba7f2 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -78,10 +78,10 @@ static bool kprobes_all_disarmed; static DEFINE_MUTEX(kprobe_mutex); static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL; static struct { - spinlock_t lock ____cacheline_aligned_in_smp; + raw_spinlock_t lock ____cacheline_aligned_in_smp; } kretprobe_table_locks[KPROBE_TABLE_SIZE]; -static spinlock_t *kretprobe_table_lock_ptr(unsigned long hash) +static raw_spinlock_t *kretprobe_table_lock_ptr(unsigned long hash) { return &(kretprobe_table_locks[hash].lock); } @@ -1013,9 +1013,9 @@ void __kprobes recycle_rp_inst(struct kretprobe_instance *ri, hlist_del(&ri->hlist); INIT_HLIST_NODE(&ri->hlist); if (likely(rp)) { - spin_lock(&rp->lock); + raw_spin_lock(&rp->lock); hlist_add_head(&ri->hlist, &rp->free_instances); - spin_unlock(&rp->lock); + raw_spin_unlock(&rp->lock); } else /* Unregistering */ hlist_add_head(&ri->hlist, head); @@ -1026,19 +1026,19 @@ void __kprobes kretprobe_hash_lock(struct task_struct *tsk, __acquires(hlist_lock) { unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS); - spinlock_t *hlist_lock; + raw_spinlock_t *hlist_lock; *head = &kretprobe_inst_table[hash]; hlist_lock = kretprobe_table_lock_ptr(hash); - spin_lock_irqsave(hlist_lock, *flags); + raw_spin_lock_irqsave(hlist_lock, *flags); } static void __kprobes kretprobe_table_lock(unsigned long hash, unsigned long *flags) __acquires(hlist_lock) { - spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); - spin_lock_irqsave(hlist_lock, *flags); + raw_spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); + raw_spin_lock_irqsave(hlist_lock, *flags); } void __kprobes kretprobe_hash_unlock(struct task_struct *tsk, @@ -1046,18 +1046,18 @@ void __kprobes kretprobe_hash_unlock(struct task_struct *tsk, __releases(hlist_lock) { unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS); - spinlock_t *hlist_lock; + raw_spinlock_t *hlist_lock; hlist_lock = kretprobe_table_lock_ptr(hash); - spin_unlock_irqrestore(hlist_lock, *flags); + raw_spin_unlock_irqrestore(hlist_lock, *flags); } static void __kprobes kretprobe_table_unlock(unsigned long hash, unsigned long *flags) __releases(hlist_lock) { - spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); - spin_unlock_irqrestore(hlist_lock, *flags); + raw_spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash); + raw_spin_unlock_irqrestore(hlist_lock, *flags); } /* @@ -1663,12 +1663,12 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p, /*TODO: consider to only swap the RA after the last pre_handler fired */ hash = hash_ptr(current, KPROBE_HASH_BITS); - spin_lock_irqsave(&rp->lock, flags); + raw_spin_lock_irqsave(&rp->lock, flags); if (!hlist_empty(&rp->free_instances)) { ri = hlist_entry(rp->free_instances.first, struct kretprobe_instance, hlist); hlist_del(&ri->hlist); - spin_unlock_irqrestore(&rp->lock, flags); + raw_spin_unlock_irqrestore(&rp->lock, flags); ri->rp = rp; ri->task = current; @@ -1685,7 +1685,7 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p, kretprobe_table_unlock(hash, &flags); } else { rp->nmissed++; - spin_unlock_irqrestore(&rp->lock, flags); + raw_spin_unlock_irqrestore(&rp->lock, flags); } return 0; } @@ -1721,7 +1721,7 @@ int __kprobes register_kretprobe(struct kretprobe *rp) rp->maxactive = num_possible_cpus(); #endif } - spin_lock_init(&rp->lock); + raw_spin_lock_init(&rp->lock); INIT_HLIST_HEAD(&rp->free_instances); for (i = 0; i < rp->maxactive; i++) { inst = kmalloc(sizeof(struct kretprobe_instance) + @@ -1959,7 +1959,7 @@ static int __init init_kprobes(void) for (i = 0; i < KPROBE_TABLE_SIZE; i++) { INIT_HLIST_HEAD(&kprobe_table[i]); INIT_HLIST_HEAD(&kretprobe_inst_table[i]); - spin_lock_init(&(kretprobe_table_locks[i].lock)); + raw_spin_lock_init(&(kretprobe_table_locks[i].lock)); } /* -- cgit v1.3-8-gc7d7 From cdcc136ffd264849a943acb42c36ffe9b458f811 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Jul 2009 16:47:45 +0200 Subject: locking, sched, cgroups: Annotate release_list_lock as raw The release_list_lock can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- kernel/cgroup.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 1d2b6ceea95d..453100a4159d 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -265,7 +265,7 @@ list_for_each_entry(_root, &roots, root_list) /* the list of cgroups eligible for automatic release. Protected by * release_list_lock */ static LIST_HEAD(release_list); -static DEFINE_SPINLOCK(release_list_lock); +static DEFINE_RAW_SPINLOCK(release_list_lock); static void cgroup_release_agent(struct work_struct *work); static DECLARE_WORK(release_agent_work, cgroup_release_agent); static void check_for_release(struct cgroup *cgrp); @@ -4014,11 +4014,11 @@ again: finish_wait(&cgroup_rmdir_waitq, &wait); clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); - spin_lock(&release_list_lock); + raw_spin_lock(&release_list_lock); set_bit(CGRP_REMOVED, &cgrp->flags); if (!list_empty(&cgrp->release_list)) list_del_init(&cgrp->release_list); - spin_unlock(&release_list_lock); + raw_spin_unlock(&release_list_lock); cgroup_lock_hierarchy(cgrp->root); /* delete this cgroup from parent->children */ @@ -4671,13 +4671,13 @@ static void check_for_release(struct cgroup *cgrp) * already queued for a userspace notification, queue * it now */ int need_schedule_work = 0; - spin_lock(&release_list_lock); + raw_spin_lock(&release_list_lock); if (!cgroup_is_removed(cgrp) && list_empty(&cgrp->release_list)) { list_add(&cgrp->release_list, &release_list); need_schedule_work = 1; } - spin_unlock(&release_list_lock); + raw_spin_unlock(&release_list_lock); if (need_schedule_work) schedule_work(&release_agent_work); } @@ -4729,7 +4729,7 @@ static void cgroup_release_agent(struct work_struct *work) { BUG_ON(work != &release_agent_work); mutex_lock(&cgroup_mutex); - spin_lock(&release_list_lock); + raw_spin_lock(&release_list_lock); while (!list_empty(&release_list)) { char *argv[3], *envp[3]; int i; @@ -4738,7 +4738,7 @@ static void cgroup_release_agent(struct work_struct *work) struct cgroup, release_list); list_del_init(&cgrp->release_list); - spin_unlock(&release_list_lock); + raw_spin_unlock(&release_list_lock); pathbuf = kmalloc(PAGE_SIZE, GFP_KERNEL); if (!pathbuf) goto continue_free; @@ -4768,9 +4768,9 @@ static void cgroup_release_agent(struct work_struct *work) continue_free: kfree(pathbuf); kfree(agentbuf); - spin_lock(&release_list_lock); + raw_spin_lock(&release_list_lock); } - spin_unlock(&release_list_lock); + raw_spin_unlock(&release_list_lock); mutex_unlock(&cgroup_mutex); } -- cgit v1.3-8-gc7d7 From 5389f6fad27019f2ba78f1b332f719ec05f12a42 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Jul 2009 17:13:33 +0200 Subject: locking, tracing: Annotate tracing locks as raw The tracing locks can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 52 ++++++++++++++++++++++---------------------- kernel/trace/trace.c | 10 ++++----- kernel/trace/trace_irqsoff.c | 6 ++--- 3 files changed, 34 insertions(+), 34 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 731201bf4acc..f2f821acc597 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -478,7 +478,7 @@ struct ring_buffer_per_cpu { int cpu; atomic_t record_disabled; struct ring_buffer *buffer; - spinlock_t reader_lock; /* serialize readers */ + raw_spinlock_t reader_lock; /* serialize readers */ arch_spinlock_t lock; struct lock_class_key lock_key; struct list_head *pages; @@ -1062,7 +1062,7 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) cpu_buffer->cpu = cpu; cpu_buffer->buffer = buffer; - spin_lock_init(&cpu_buffer->reader_lock); + raw_spin_lock_init(&cpu_buffer->reader_lock); lockdep_set_class(&cpu_buffer->reader_lock, buffer->reader_lock_key); cpu_buffer->lock = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED; @@ -1259,7 +1259,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) struct list_head *p; unsigned i; - spin_lock_irq(&cpu_buffer->reader_lock); + raw_spin_lock_irq(&cpu_buffer->reader_lock); rb_head_page_deactivate(cpu_buffer); for (i = 0; i < nr_pages; i++) { @@ -1277,7 +1277,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) rb_check_pages(cpu_buffer); out: - spin_unlock_irq(&cpu_buffer->reader_lock); + raw_spin_unlock_irq(&cpu_buffer->reader_lock); } static void @@ -1288,7 +1288,7 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, struct list_head *p; unsigned i; - spin_lock_irq(&cpu_buffer->reader_lock); + raw_spin_lock_irq(&cpu_buffer->reader_lock); rb_head_page_deactivate(cpu_buffer); for (i = 0; i < nr_pages; i++) { @@ -1303,7 +1303,7 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, rb_check_pages(cpu_buffer); out: - spin_unlock_irq(&cpu_buffer->reader_lock); + raw_spin_unlock_irq(&cpu_buffer->reader_lock); } /** @@ -2804,9 +2804,9 @@ void ring_buffer_iter_reset(struct ring_buffer_iter *iter) cpu_buffer = iter->cpu_buffer; - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); rb_iter_reset(iter); - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); } EXPORT_SYMBOL_GPL(ring_buffer_iter_reset); @@ -3265,12 +3265,12 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts, again: local_irq_save(flags); if (dolock) - spin_lock(&cpu_buffer->reader_lock); + raw_spin_lock(&cpu_buffer->reader_lock); event = rb_buffer_peek(cpu_buffer, ts, lost_events); if (event && event->type_len == RINGBUF_TYPE_PADDING) rb_advance_reader(cpu_buffer); if (dolock) - spin_unlock(&cpu_buffer->reader_lock); + raw_spin_unlock(&cpu_buffer->reader_lock); local_irq_restore(flags); if (event && event->type_len == RINGBUF_TYPE_PADDING) @@ -3295,9 +3295,9 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) unsigned long flags; again: - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); event = rb_iter_peek(iter, ts); - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); if (event && event->type_len == RINGBUF_TYPE_PADDING) goto again; @@ -3337,7 +3337,7 @@ ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); if (dolock) - spin_lock(&cpu_buffer->reader_lock); + raw_spin_lock(&cpu_buffer->reader_lock); event = rb_buffer_peek(cpu_buffer, ts, lost_events); if (event) { @@ -3346,7 +3346,7 @@ ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts, } if (dolock) - spin_unlock(&cpu_buffer->reader_lock); + raw_spin_unlock(&cpu_buffer->reader_lock); local_irq_restore(flags); out: @@ -3438,11 +3438,11 @@ ring_buffer_read_start(struct ring_buffer_iter *iter) cpu_buffer = iter->cpu_buffer; - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); arch_spin_lock(&cpu_buffer->lock); rb_iter_reset(iter); arch_spin_unlock(&cpu_buffer->lock); - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); } EXPORT_SYMBOL_GPL(ring_buffer_read_start); @@ -3477,7 +3477,7 @@ ring_buffer_read(struct ring_buffer_iter *iter, u64 *ts) struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; unsigned long flags; - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); again: event = rb_iter_peek(iter, ts); if (!event) @@ -3488,7 +3488,7 @@ ring_buffer_read(struct ring_buffer_iter *iter, u64 *ts) rb_advance_iter(iter); out: - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); return event; } @@ -3557,7 +3557,7 @@ void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) atomic_inc(&cpu_buffer->record_disabled); - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); if (RB_WARN_ON(cpu_buffer, local_read(&cpu_buffer->committing))) goto out; @@ -3569,7 +3569,7 @@ void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) arch_spin_unlock(&cpu_buffer->lock); out: - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); atomic_dec(&cpu_buffer->record_disabled); } @@ -3607,10 +3607,10 @@ int ring_buffer_empty(struct ring_buffer *buffer) cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); if (dolock) - spin_lock(&cpu_buffer->reader_lock); + raw_spin_lock(&cpu_buffer->reader_lock); ret = rb_per_cpu_empty(cpu_buffer); if (dolock) - spin_unlock(&cpu_buffer->reader_lock); + raw_spin_unlock(&cpu_buffer->reader_lock); local_irq_restore(flags); if (!ret) @@ -3641,10 +3641,10 @@ int ring_buffer_empty_cpu(struct ring_buffer *buffer, int cpu) cpu_buffer = buffer->buffers[cpu]; local_irq_save(flags); if (dolock) - spin_lock(&cpu_buffer->reader_lock); + raw_spin_lock(&cpu_buffer->reader_lock); ret = rb_per_cpu_empty(cpu_buffer); if (dolock) - spin_unlock(&cpu_buffer->reader_lock); + raw_spin_unlock(&cpu_buffer->reader_lock); local_irq_restore(flags); return ret; @@ -3841,7 +3841,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer, if (!bpage) goto out; - spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + raw_spin_lock_irqsave(&cpu_buffer->reader_lock, flags); reader = rb_get_reader_page(cpu_buffer); if (!reader) @@ -3964,7 +3964,7 @@ int ring_buffer_read_page(struct ring_buffer *buffer, memset(&bpage->data[commit], 0, BUF_PAGE_SIZE - commit); out_unlock: - spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + raw_spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); out: return ret; diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index e5df02c69b1d..0c8bdeeb358b 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -341,7 +341,7 @@ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE; static int trace_stop_count; -static DEFINE_SPINLOCK(tracing_start_lock); +static DEFINE_RAW_SPINLOCK(tracing_start_lock); static void wakeup_work_handler(struct work_struct *work) { @@ -960,7 +960,7 @@ void tracing_start(void) if (tracing_disabled) return; - spin_lock_irqsave(&tracing_start_lock, flags); + raw_spin_lock_irqsave(&tracing_start_lock, flags); if (--trace_stop_count) { if (trace_stop_count < 0) { /* Someone screwed up their debugging */ @@ -985,7 +985,7 @@ void tracing_start(void) ftrace_start(); out: - spin_unlock_irqrestore(&tracing_start_lock, flags); + raw_spin_unlock_irqrestore(&tracing_start_lock, flags); } /** @@ -1000,7 +1000,7 @@ void tracing_stop(void) unsigned long flags; ftrace_stop(); - spin_lock_irqsave(&tracing_start_lock, flags); + raw_spin_lock_irqsave(&tracing_start_lock, flags); if (trace_stop_count++) goto out; @@ -1018,7 +1018,7 @@ void tracing_stop(void) arch_spin_unlock(&ftrace_max_lock); out: - spin_unlock_irqrestore(&tracing_start_lock, flags); + raw_spin_unlock_irqrestore(&tracing_start_lock, flags); } void trace_stop_cmdline_recording(void); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 667aa8cc0cfc..11186212068c 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -23,7 +23,7 @@ static int tracer_enabled __read_mostly; static DEFINE_PER_CPU(int, tracing_cpu); -static DEFINE_SPINLOCK(max_trace_lock); +static DEFINE_RAW_SPINLOCK(max_trace_lock); enum { TRACER_IRQS_OFF = (1 << 1), @@ -321,7 +321,7 @@ check_critical_timing(struct trace_array *tr, if (!report_latency(delta)) goto out; - spin_lock_irqsave(&max_trace_lock, flags); + raw_spin_lock_irqsave(&max_trace_lock, flags); /* check if we are still the max latency */ if (!report_latency(delta)) @@ -344,7 +344,7 @@ check_critical_timing(struct trace_array *tr, max_sequence++; out_unlock: - spin_unlock_irqrestore(&max_trace_lock, flags); + raw_spin_unlock_irqrestore(&max_trace_lock, flags); out: data->critical_sequence = max_sequence; -- cgit v1.3-8-gc7d7 From 07354eb1a74d1e1ece29f8bafe0b46e8c77a95ef Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Jul 2009 17:50:36 +0200 Subject: locking, printk: Annotate logbuf_lock as raw The logbuf_lock lock can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner [ merged and fixed it ] Signed-off-by: Ingo Molnar --- include/linux/ratelimit.h | 6 +++--- kernel/printk.c | 46 +++++++++++++++++++++++----------------------- lib/ratelimit.c | 4 ++-- 3 files changed, 28 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/include/linux/ratelimit.h b/include/linux/ratelimit.h index 2f007157fab9..e11ccb4cf48d 100644 --- a/include/linux/ratelimit.h +++ b/include/linux/ratelimit.h @@ -8,7 +8,7 @@ #define DEFAULT_RATELIMIT_BURST 10 struct ratelimit_state { - spinlock_t lock; /* protect the state */ + raw_spinlock_t lock; /* protect the state */ int interval; int burst; @@ -20,7 +20,7 @@ struct ratelimit_state { #define DEFINE_RATELIMIT_STATE(name, interval_init, burst_init) \ \ struct ratelimit_state name = { \ - .lock = __SPIN_LOCK_UNLOCKED(name.lock), \ + .lock = __RAW_SPIN_LOCK_UNLOCKED(name.lock), \ .interval = interval_init, \ .burst = burst_init, \ } @@ -28,7 +28,7 @@ struct ratelimit_state { static inline void ratelimit_state_init(struct ratelimit_state *rs, int interval, int burst) { - spin_lock_init(&rs->lock); + raw_spin_lock_init(&rs->lock); rs->interval = interval; rs->burst = burst; rs->printed = 0; diff --git a/kernel/printk.c b/kernel/printk.c index 28a40d8171b8..b7da18391c38 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -100,7 +100,7 @@ static int console_locked, console_suspended; * It is also used in interesting ways to provide interlocking in * console_unlock();. */ -static DEFINE_SPINLOCK(logbuf_lock); +static DEFINE_RAW_SPINLOCK(logbuf_lock); #define LOG_BUF_MASK (log_buf_len-1) #define LOG_BUF(idx) (log_buf[(idx) & LOG_BUF_MASK]) @@ -212,7 +212,7 @@ void __init setup_log_buf(int early) return; } - spin_lock_irqsave(&logbuf_lock, flags); + raw_spin_lock_irqsave(&logbuf_lock, flags); log_buf_len = new_log_buf_len; log_buf = new_log_buf; new_log_buf_len = 0; @@ -230,7 +230,7 @@ void __init setup_log_buf(int early) log_start -= offset; con_start -= offset; log_end -= offset; - spin_unlock_irqrestore(&logbuf_lock, flags); + raw_spin_unlock_irqrestore(&logbuf_lock, flags); pr_info("log_buf_len: %d\n", log_buf_len); pr_info("early log buf free: %d(%d%%)\n", @@ -365,18 +365,18 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) if (error) goto out; i = 0; - spin_lock_irq(&logbuf_lock); + raw_spin_lock_irq(&logbuf_lock); while (!error && (log_start != log_end) && i < len) { c = LOG_BUF(log_start); log_start++; - spin_unlock_irq(&logbuf_lock); + raw_spin_unlock_irq(&logbuf_lock); error = __put_user(c,buf); buf++; i++; cond_resched(); - spin_lock_irq(&logbuf_lock); + raw_spin_lock_irq(&logbuf_lock); } - spin_unlock_irq(&logbuf_lock); + raw_spin_unlock_irq(&logbuf_lock); if (!error) error = i; break; @@ -399,7 +399,7 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) count = len; if (count > log_buf_len) count = log_buf_len; - spin_lock_irq(&logbuf_lock); + raw_spin_lock_irq(&logbuf_lock); if (count > logged_chars) count = logged_chars; if (do_clear) @@ -416,12 +416,12 @@ int do_syslog(int type, char __user *buf, int len, bool from_file) if (j + log_buf_len < log_end) break; c = LOG_BUF(j); - spin_unlock_irq(&logbuf_lock); + raw_spin_unlock_irq(&logbuf_lock); error = __put_user(c,&buf[count-1-i]); cond_resched(); - spin_lock_irq(&logbuf_lock); + raw_spin_lock_irq(&logbuf_lock); } - spin_unlock_irq(&logbuf_lock); + raw_spin_unlock_irq(&logbuf_lock); if (error) break; error = i; @@ -689,7 +689,7 @@ static void zap_locks(void) oops_timestamp = jiffies; /* If a crash is occurring, make sure we can't deadlock */ - spin_lock_init(&logbuf_lock); + raw_spin_lock_init(&logbuf_lock); /* And make sure that we print immediately */ sema_init(&console_sem, 1); } @@ -802,9 +802,9 @@ static int console_trylock_for_printk(unsigned int cpu) } } printk_cpu = UINT_MAX; - spin_unlock(&logbuf_lock); if (wake) up(&console_sem); + raw_spin_unlock(&logbuf_lock); return retval; } static const char recursion_bug_msg [] = @@ -864,7 +864,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) } lockdep_off(); - spin_lock(&logbuf_lock); + raw_spin_lock(&logbuf_lock); printk_cpu = this_cpu; if (recursion_bug) { @@ -1257,14 +1257,14 @@ void console_unlock(void) again: for ( ; ; ) { - spin_lock_irqsave(&logbuf_lock, flags); + raw_spin_lock_irqsave(&logbuf_lock, flags); wake_klogd |= log_start - log_end; if (con_start == log_end) break; /* Nothing to print */ _con_start = con_start; _log_end = log_end; con_start = log_end; /* Flush */ - spin_unlock(&logbuf_lock); + raw_spin_unlock(&logbuf_lock); stop_critical_timings(); /* don't trace print latency */ call_console_drivers(_con_start, _log_end); start_critical_timings(); @@ -1276,7 +1276,7 @@ again: if (unlikely(exclusive_console)) exclusive_console = NULL; - spin_unlock(&logbuf_lock); + raw_spin_unlock(&logbuf_lock); up(&console_sem); @@ -1286,13 +1286,13 @@ again: * there's a new owner and the console_unlock() from them will do the * flush, no worries. */ - spin_lock(&logbuf_lock); + raw_spin_lock(&logbuf_lock); if (con_start != log_end) retry = 1; - spin_unlock_irqrestore(&logbuf_lock, flags); if (retry && console_trylock()) goto again; + raw_spin_unlock_irqrestore(&logbuf_lock, flags); if (wake_klogd) wake_up_klogd(); } @@ -1522,9 +1522,9 @@ void register_console(struct console *newcon) * console_unlock(); will print out the buffered messages * for us. */ - spin_lock_irqsave(&logbuf_lock, flags); + raw_spin_lock_irqsave(&logbuf_lock, flags); con_start = log_start; - spin_unlock_irqrestore(&logbuf_lock, flags); + raw_spin_unlock_irqrestore(&logbuf_lock, flags); /* * We're about to replay the log buffer. Only do this to the * just-registered console to avoid excessive message spam to @@ -1731,10 +1731,10 @@ void kmsg_dump(enum kmsg_dump_reason reason) /* Theoretically, the log could move on after we do this, but there's not a lot we can do about that. The new messages will overwrite the start of what we dump. */ - spin_lock_irqsave(&logbuf_lock, flags); + raw_spin_lock_irqsave(&logbuf_lock, flags); end = log_end & LOG_BUF_MASK; chars = logged_chars; - spin_unlock_irqrestore(&logbuf_lock, flags); + raw_spin_unlock_irqrestore(&logbuf_lock, flags); if (chars > end) { s1 = log_buf + log_buf_len - chars + end; diff --git a/lib/ratelimit.c b/lib/ratelimit.c index 027a03f4c56d..c96d500577de 100644 --- a/lib/ratelimit.c +++ b/lib/ratelimit.c @@ -39,7 +39,7 @@ int ___ratelimit(struct ratelimit_state *rs, const char *func) * in addition to the one that will be printed by * the entity that is holding the lock already: */ - if (!spin_trylock_irqsave(&rs->lock, flags)) + if (!raw_spin_trylock_irqsave(&rs->lock, flags)) return 0; if (!rs->begin) @@ -60,7 +60,7 @@ int ___ratelimit(struct ratelimit_state *rs, const char *func) rs->missed++; ret = 0; } - spin_unlock_irqrestore(&rs->lock, flags); + raw_spin_unlock_irqrestore(&rs->lock, flags); return ret; } -- cgit v1.3-8-gc7d7 From ee30a7b2fc072f139dac44826860d2c1f422137c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Jul 2009 18:56:56 +0200 Subject: locking, sched: Annotate thread_group_cputimer as raw The thread_group_cputimer lock can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/linux/init_task.h | 2 +- include/linux/sched.h | 4 ++-- kernel/posix-cpu-timers.c | 12 ++++++------ kernel/sched_stats.h | 12 ++++++------ 4 files changed, 15 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index d14e058aaeed..08ffab01e76c 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -42,7 +42,7 @@ extern struct fs_struct init_fs; .cputimer = { \ .cputime = INIT_CPUTIME, \ .running = 0, \ - .lock = __SPIN_LOCK_UNLOCKED(sig.cputimer.lock), \ + .lock = __RAW_SPIN_LOCK_UNLOCKED(sig.cputimer.lock), \ }, \ .cred_guard_mutex = \ __MUTEX_INITIALIZER(sig.cred_guard_mutex), \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 4ac2c0578e0f..e672236c08e0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -510,7 +510,7 @@ struct task_cputime { struct thread_group_cputimer { struct task_cputime cputime; int running; - spinlock_t lock; + raw_spinlock_t lock; }; #include @@ -2566,7 +2566,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times); static inline void thread_group_cputime_init(struct signal_struct *sig) { - spin_lock_init(&sig->cputimer.lock); + raw_spin_lock_init(&sig->cputimer.lock); } /* diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 58f405b581e7..41440cca55a2 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -274,7 +274,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) struct task_cputime sum; unsigned long flags; - spin_lock_irqsave(&cputimer->lock, flags); + raw_spin_lock_irqsave(&cputimer->lock, flags); if (!cputimer->running) { cputimer->running = 1; /* @@ -287,7 +287,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) update_gt_cputime(&cputimer->cputime, &sum); } *times = cputimer->cputime; - spin_unlock_irqrestore(&cputimer->lock, flags); + raw_spin_unlock_irqrestore(&cputimer->lock, flags); } /* @@ -997,9 +997,9 @@ static void stop_process_timers(struct signal_struct *sig) struct thread_group_cputimer *cputimer = &sig->cputimer; unsigned long flags; - spin_lock_irqsave(&cputimer->lock, flags); + raw_spin_lock_irqsave(&cputimer->lock, flags); cputimer->running = 0; - spin_unlock_irqrestore(&cputimer->lock, flags); + raw_spin_unlock_irqrestore(&cputimer->lock, flags); } static u32 onecputick; @@ -1289,9 +1289,9 @@ static inline int fastpath_timer_check(struct task_struct *tsk) if (sig->cputimer.running) { struct task_cputime group_sample; - spin_lock(&sig->cputimer.lock); + raw_spin_lock(&sig->cputimer.lock); group_sample = sig->cputimer.cputime; - spin_unlock(&sig->cputimer.lock); + raw_spin_unlock(&sig->cputimer.lock); if (task_cputime_expired(&group_sample, &sig->cputime_expires)) return 1; diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 331e01bcd026..87f9e36ea56e 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -282,10 +282,10 @@ static inline void account_group_user_time(struct task_struct *tsk, if (!cputimer->running) return; - spin_lock(&cputimer->lock); + raw_spin_lock(&cputimer->lock); cputimer->cputime.utime = cputime_add(cputimer->cputime.utime, cputime); - spin_unlock(&cputimer->lock); + raw_spin_unlock(&cputimer->lock); } /** @@ -306,10 +306,10 @@ static inline void account_group_system_time(struct task_struct *tsk, if (!cputimer->running) return; - spin_lock(&cputimer->lock); + raw_spin_lock(&cputimer->lock); cputimer->cputime.stime = cputime_add(cputimer->cputime.stime, cputime); - spin_unlock(&cputimer->lock); + raw_spin_unlock(&cputimer->lock); } /** @@ -330,7 +330,7 @@ static inline void account_group_exec_runtime(struct task_struct *tsk, if (!cputimer->running) return; - spin_lock(&cputimer->lock); + raw_spin_lock(&cputimer->lock); cputimer->cputime.sum_exec_runtime += ns; - spin_unlock(&cputimer->lock); + raw_spin_unlock(&cputimer->lock); } -- cgit v1.3-8-gc7d7 From 8292c9e15c3b069459794a04f5e2cf0d5665ddc4 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 24 Feb 2010 09:50:22 +0100 Subject: locking, semaphores: Annotate inner lock as raw There is no reason to have the spin_lock protecting the semaphore preemptible on -rt. Annotate it as a raw_spinlock. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. ( On rt this also solves lockdep complaining about the rt_mutex.wait_lock being not initialized. ) Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/linux/semaphore.h | 4 ++-- kernel/semaphore.c | 28 ++++++++++++++-------------- 2 files changed, 16 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h index 39fa04966aa8..dc368b8ce215 100644 --- a/include/linux/semaphore.h +++ b/include/linux/semaphore.h @@ -14,14 +14,14 @@ /* Please don't access any members of this structure directly */ struct semaphore { - spinlock_t lock; + raw_spinlock_t lock; unsigned int count; struct list_head wait_list; }; #define __SEMAPHORE_INITIALIZER(name, n) \ { \ - .lock = __SPIN_LOCK_UNLOCKED((name).lock), \ + .lock = __RAW_SPIN_LOCK_UNLOCKED((name).lock), \ .count = n, \ .wait_list = LIST_HEAD_INIT((name).wait_list), \ } diff --git a/kernel/semaphore.c b/kernel/semaphore.c index 94a62c0d4ade..d831841e55a7 100644 --- a/kernel/semaphore.c +++ b/kernel/semaphore.c @@ -54,12 +54,12 @@ void down(struct semaphore *sem) { unsigned long flags; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); if (likely(sem->count > 0)) sem->count--; else __down(sem); - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); } EXPORT_SYMBOL(down); @@ -77,12 +77,12 @@ int down_interruptible(struct semaphore *sem) unsigned long flags; int result = 0; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); if (likely(sem->count > 0)) sem->count--; else result = __down_interruptible(sem); - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); return result; } @@ -103,12 +103,12 @@ int down_killable(struct semaphore *sem) unsigned long flags; int result = 0; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); if (likely(sem->count > 0)) sem->count--; else result = __down_killable(sem); - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); return result; } @@ -132,11 +132,11 @@ int down_trylock(struct semaphore *sem) unsigned long flags; int count; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); count = sem->count - 1; if (likely(count >= 0)) sem->count = count; - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); return (count < 0); } @@ -157,12 +157,12 @@ int down_timeout(struct semaphore *sem, long jiffies) unsigned long flags; int result = 0; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); if (likely(sem->count > 0)) sem->count--; else result = __down_timeout(sem, jiffies); - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); return result; } @@ -179,12 +179,12 @@ void up(struct semaphore *sem) { unsigned long flags; - spin_lock_irqsave(&sem->lock, flags); + raw_spin_lock_irqsave(&sem->lock, flags); if (likely(list_empty(&sem->wait_list))) sem->count++; else __up(sem); - spin_unlock_irqrestore(&sem->lock, flags); + raw_spin_unlock_irqrestore(&sem->lock, flags); } EXPORT_SYMBOL(up); @@ -217,9 +217,9 @@ static inline int __sched __down_common(struct semaphore *sem, long state, if (timeout <= 0) goto timed_out; __set_task_state(task, state); - spin_unlock_irq(&sem->lock); + raw_spin_unlock_irq(&sem->lock); timeout = schedule_timeout(timeout); - spin_lock_irq(&sem->lock); + raw_spin_lock_irq(&sem->lock); if (waiter.up) return 0; } -- cgit v1.3-8-gc7d7 From 2737c49f29a29f3d3645ba0778aa7a8798f32249 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 16 Jun 2010 16:58:34 +0200 Subject: locking, timer_stats: Annotate table_lock as raw The table_lock lock can be taken in atomic context and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Reported-by: Andreas Sundebo Signed-off-by: Thomas Gleixner Tested-by: Andreas Sundebo Signed-off-by: Ingo Molnar --- kernel/time/timer_stats.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index a5d0a3a85dd8..0b537f27b559 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -81,7 +81,7 @@ struct entry { /* * Spinlock protecting the tables - not taken during lookup: */ -static DEFINE_SPINLOCK(table_lock); +static DEFINE_RAW_SPINLOCK(table_lock); /* * Per-CPU lookup locks for fast hash lookup: @@ -188,7 +188,7 @@ static struct entry *tstat_lookup(struct entry *entry, char *comm) prev = NULL; curr = *head; - spin_lock(&table_lock); + raw_spin_lock(&table_lock); /* * Make sure we have not raced with another CPU: */ @@ -215,7 +215,7 @@ static struct entry *tstat_lookup(struct entry *entry, char *comm) *head = curr; } out_unlock: - spin_unlock(&table_lock); + raw_spin_unlock(&table_lock); return curr; } -- cgit v1.3-8-gc7d7 From 757455d41c390da4ae8e06468ad097fe42eaab41 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 17 Sep 2009 17:31:09 +0200 Subject: locking, latencytop: Annotate latency_lock as raw The latency_lock is lock can be taken in the guts of the scheduler code and therefore cannot be preempted on -rt - annotate it. In mainline this change documents the low level nature of the lock - otherwise there's no functional difference. Lockdep and Sparse checking will work as usual. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- kernel/latencytop.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 376066e10413..4ac8ebfcab59 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -58,7 +58,7 @@ #include #include -static DEFINE_SPINLOCK(latency_lock); +static DEFINE_RAW_SPINLOCK(latency_lock); #define MAXLR 128 static struct latency_record latency_record[MAXLR]; @@ -72,19 +72,19 @@ void clear_all_latency_tracing(struct task_struct *p) if (!latencytop_enabled) return; - spin_lock_irqsave(&latency_lock, flags); + raw_spin_lock_irqsave(&latency_lock, flags); memset(&p->latency_record, 0, sizeof(p->latency_record)); p->latency_record_count = 0; - spin_unlock_irqrestore(&latency_lock, flags); + raw_spin_unlock_irqrestore(&latency_lock, flags); } static void clear_global_latency_tracing(void) { unsigned long flags; - spin_lock_irqsave(&latency_lock, flags); + raw_spin_lock_irqsave(&latency_lock, flags); memset(&latency_record, 0, sizeof(latency_record)); - spin_unlock_irqrestore(&latency_lock, flags); + raw_spin_unlock_irqrestore(&latency_lock, flags); } static void __sched @@ -190,7 +190,7 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) lat.max = usecs; store_stacktrace(tsk, &lat); - spin_lock_irqsave(&latency_lock, flags); + raw_spin_lock_irqsave(&latency_lock, flags); account_global_scheduler_latency(tsk, &lat); @@ -231,7 +231,7 @@ __account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record)); out_unlock: - spin_unlock_irqrestore(&latency_lock, flags); + raw_spin_unlock_irqrestore(&latency_lock, flags); } static int lstats_show(struct seq_file *m, void *v) -- cgit v1.3-8-gc7d7 From 4523f6ada86853750565c68e17126af2e3df9b8a Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 14 Sep 2011 10:54:29 +0200 Subject: alarmtimers: Fix error handling commit 8bc0daf (alarmtimers: Rework RTC device selection using class interface) did not implement required error checks. Add them. Signed-off-by: Thomas Gleixner --- kernel/time/alarmtimer.c | 41 +++++++++++++++++++++++++++++++---------- 1 file changed, 31 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 154d5563ab1b..c436e790b21b 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -100,19 +100,25 @@ static struct class_interface alarmtimer_rtc_interface = { .add_dev = &alarmtimer_rtc_add_device, }; -static void alarmtimer_rtc_interface_setup(void) +static int alarmtimer_rtc_interface_setup(void) { alarmtimer_rtc_interface.class = rtc_class; - class_interface_register(&alarmtimer_rtc_interface); + return class_interface_register(&alarmtimer_rtc_interface); +} +static void alarmtimer_rtc_interface_remove(void) +{ + class_interface_unregister(&alarmtimer_rtc_interface); } #else -#define alarmtimer_get_rtcdev() (0) -#define rtcdev (0) -#define alarmtimer_rtc_interface_setup() +static inline struct rtc_device *alarmtimer_get_rtcdev(void) +{ + return NULL; +} +#define rtcdev (NULL) +static inline int alarmtimer_rtc_interface_setup(void) { return 0; } +static inline void alarmtimer_rtc_interface_remove(void) { } #endif - - /** * alarmtimer_enqueue - Adds an alarm timer to an alarm_base timerqueue * @base: pointer to the base where the timer is being run @@ -764,6 +770,7 @@ static struct platform_driver alarmtimer_driver = { */ static int __init alarmtimer_init(void) { + struct platform_device *pdev; int error = 0; int i; struct k_clock alarm_clock = { @@ -793,11 +800,25 @@ static int __init alarmtimer_init(void) alarm_bases[i].timer.function = alarmtimer_fired; } - alarmtimer_rtc_interface_setup(); + error = alarmtimer_rtc_interface_setup(); + if (error) + return error; + error = platform_driver_register(&alarmtimer_driver); - platform_device_register_simple("alarmtimer", -1, NULL, 0); + if (error) + goto out_if; + pdev = platform_device_register_simple("alarmtimer", -1, NULL, 0); + if (IS_ERR(pdev)) { + error = PTR_ERR(pdev); + goto out_drv; + } + return 0; + +out_drv: + platform_driver_unregister(&alarmtimer_driver); +out_if: + alarmtimer_rtc_interface_remove(); return error; } device_initcall(alarmtimer_init); - -- cgit v1.3-8-gc7d7 From fa2563e41c3d6d6e8af437643981ed28ae0cb56d Mon Sep 17 00:00:00 2001 From: Thomas Tuttle Date: Wed, 14 Sep 2011 16:22:28 -0700 Subject: workqueue: lock cwq access in drain_workqueue Take cwq->gcwq->lock to avoid racing between drain_workqueue checking to make sure the workqueues are empty and cwq_dec_nr_in_flight decrementing and then incrementing nr_active when it activates a delayed work. We discovered this when a corner case in one of our drivers resulted in us trying to destroy a workqueue in which the remaining work would always requeue itself again in the same workqueue. We would hit this race condition and trip the BUG_ON on workqueue.c:3080. Signed-off-by: Thomas Tuttle Acked-by: Tejun Heo Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/workqueue.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 25fb1b0e53fa..1783aabc6128 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2412,8 +2412,13 @@ reflush: for_each_cwq_cpu(cpu, wq) { struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); + bool drained; - if (!cwq->nr_active && list_empty(&cwq->delayed_works)) + spin_lock_irq(&cwq->gcwq->lock); + drained = !cwq->nr_active && list_empty(&cwq->delayed_works); + spin_unlock_irq(&cwq->gcwq->lock); + + if (drained) continue; if (++flush_cnt == 10 || -- cgit v1.3-8-gc7d7 From 124ff4e53ace98300bf7465e0285167a3df838d3 Mon Sep 17 00:00:00 2001 From: Vitaliy Ivanov Date: Thu, 7 Jul 2011 14:10:40 +0300 Subject: async: uninitialized warning corrections The variables here are really not used uninitialized. kernel/async.c: In function 'async_synchronize_cookie_domain': kernel/async.c:270:10: warning: 'starttime.tv64' may be used uninitialized in this function kernel/async.c: In function 'async_run_entry_fn': kernel/async.c:122:10: warning: 'calltime.tv64' may be used uninitialized in this function Signed-off-by: Vitaliy Ivanov Signed-off-by: Konstantin Khlebnikov Signed-off-by: Viresh Kumar Signed-off-by: Jiri Kosina --- kernel/async.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/async.c b/kernel/async.c index d5fe7af0de2e..4c2843c0043e 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -120,7 +120,7 @@ static void async_run_entry_fn(struct work_struct *work) struct async_entry *entry = container_of(work, struct async_entry, work); unsigned long flags; - ktime_t calltime, delta, rettime; + ktime_t uninitialized_var(calltime), delta, rettime; /* 1) move self to the running queue */ spin_lock_irqsave(&async_lock, flags); @@ -269,7 +269,7 @@ EXPORT_SYMBOL_GPL(async_synchronize_full_domain); void async_synchronize_cookie_domain(async_cookie_t cookie, struct list_head *running) { - ktime_t starttime, delta, endtime; + ktime_t uninitialized_var(starttime), delta, endtime; if (initcall_debug && system_state == SYSTEM_BOOTING) { printk(KERN_DEBUG "async_waiting @ %i\n", task_pid_nr(current)); -- cgit v1.3-8-gc7d7 From 7cfdaf38d4975b0266fbdef971e5ce37da7214fd Mon Sep 17 00:00:00 2001 From: Vitaliy Ivanov Date: Thu, 7 Jul 2011 15:10:31 +0300 Subject: futex: uninitialized warning corrections The variables here are really not used uninitialized. kernel/futex.c: In function 'fixup_pi_state_owner.clone.17': kernel/futex.c:1582:6: warning: 'curval' may be used uninitialized in this function kernel/futex.c: In function 'handle_futex_death': kernel/futex.c:2486:6: warning: 'nval' may be used uninitialized in this function kernel/futex.c: In function 'do_futex': kernel/futex.c:863:11: warning: 'curval' may be used uninitialized in this function kernel/futex.c:828:6: note: 'curval' was declared here kernel/futex.c:898:5: warning: 'oldval' may be used uninitialized in this function kernel/futex.c:890:6: note: 'oldval' was declared here Signed-off-by: Vitaliy Ivanov Acked-by: Darren Hart Signed-off-by: Jiri Kosina --- kernel/futex.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 0a308970c24a..f4e811c8623c 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -825,7 +825,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) { struct task_struct *new_owner; struct futex_pi_state *pi_state = this->pi_state; - u32 curval, newval; + u32 uninitialized_var(curval), newval; if (!pi_state) return -EINVAL; @@ -887,7 +887,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) static int unlock_futex_pi(u32 __user *uaddr, u32 uval) { - u32 oldval; + u32 uninitialized_var(oldval); /* * There is no waiter, so we unlock the futex. The owner died @@ -1546,7 +1546,7 @@ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; struct task_struct *oldowner = pi_state->owner; - u32 uval, curval, newval; + u32 uval, uninitialized_var(curval), newval; int ret; /* Owner died? */ @@ -2451,7 +2451,7 @@ err_unlock: */ int handle_futex_death(u32 __user *uaddr, struct task_struct *curr, int pi) { - u32 uval, nval, mval; + u32 uval, uninitialized_var(nval), mval; retry: if (get_user(uval, uaddr)) -- cgit v1.3-8-gc7d7 From ca4a04cf3dd0cecb5e7188ed7796cc55fc13aeb1 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Sun, 17 Jul 2011 09:01:00 +0200 Subject: futex: Fix spelling in a source code comment Change a single occurrence of "unlcoked" into "unlocked". Signed-off-by: Bart Van Assche Cc: Darren Hart Cc: Thomas Gleixner Signed-off-by: Jiri Kosina --- kernel/futex.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index f4e811c8623c..557c9c7cdd88 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1763,7 +1763,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * * Returns: * 0 - uaddr contains val and hb has been locked - * <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlcoked + * <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlocked */ static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags, struct futex_q *q, struct futex_hash_bucket **hb) -- cgit v1.3-8-gc7d7 From 3be209a8e22cedafc1b6945608b7bb8d9887ab61 Mon Sep 17 00:00:00 2001 From: Shawn Bohrer Date: Mon, 12 Sep 2011 09:28:04 -0500 Subject: sched/rt: Migrate equal priority tasks to available CPUs Commit 43fa5460fe60dea5c610490a1d263415419c60f6 ("sched: Try not to migrate higher priority RT tasks") also introduced a change in behavior which keeps RT tasks on the same CPU if there is an equal priority RT task currently running even if there are empty CPUs available. This can cause unnecessary wakeup latencies, and can prevent the scheduler from balancing all RT tasks across available CPUs. This change causes an RT task to search for a new CPU if an equal priority RT task is already running on wakeup. Lower priority tasks will still have to wait on higher priority tasks, but the system should still balance out because there is always the possibility that if there are both a high and low priority RT tasks on a given CPU that the high priority task could wakeup while the low priority task is running and force it to search for a better runqueue. Signed-off-by: Shawn Bohrer Acked-by: Steven Rostedt Tested-by: Steven Rostedt Signed-off-by: Peter Zijlstra Cc: stable@kernel.org # 37+ Link: http://lkml.kernel.org/r/1315837684-18733-1-git-send-email-sbohrer@rgmadvisors.com Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 97540f0c9e47..af1177858be3 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1050,7 +1050,7 @@ select_task_rq_rt(struct task_struct *p, int sd_flag, int flags) */ if (curr && unlikely(rt_task(curr)) && (curr->rt.nr_cpus_allowed < 2 || - curr->prio < p->prio) && + curr->prio <= p->prio) && (p->rt.nr_cpus_allowed > 1)) { int target = find_lowest_rq(p); @@ -1581,7 +1581,7 @@ static void task_woken_rt(struct rq *rq, struct task_struct *p) p->rt.nr_cpus_allowed > 1 && rt_task(rq->curr) && (rq->curr->rt.nr_cpus_allowed < 2 || - rq->curr->prio < p->prio)) + rq->curr->prio <= p->prio)) push_rt_tasks(rq); } -- cgit v1.3-8-gc7d7 From 0119fee449f501d55924914a90f152540dd4ef9a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 2 Sep 2011 01:30:29 +0200 Subject: lockdep: Comment all warnings Andrew requested I comment all the lockdep WARN()s to help other people figure out wth is wrong.. Requested-by: Andrew Morton Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1315301493.3191.9.camel@twins Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 156 +++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 147 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 91d67ce3a8d5..c081fa967c8f 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -96,8 +96,13 @@ static int graph_lock(void) static inline int graph_unlock(void) { - if (debug_locks && !arch_spin_is_locked(&lockdep_lock)) + if (debug_locks && !arch_spin_is_locked(&lockdep_lock)) { + /* + * The lockdep graph lock isn't locked while we expect it to + * be, we're confused now, bye! + */ return DEBUG_LOCKS_WARN_ON(1); + } current->lockdep_recursion--; arch_spin_unlock(&lockdep_lock); @@ -134,6 +139,9 @@ static struct lock_class lock_classes[MAX_LOCKDEP_KEYS]; static inline struct lock_class *hlock_class(struct held_lock *hlock) { if (!hlock->class_idx) { + /* + * Someone passed in garbage, we give up. + */ DEBUG_LOCKS_WARN_ON(1); return NULL; } @@ -687,6 +695,10 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) */ list_for_each_entry(class, hash_head, hash_entry) { if (class->key == key) { + /* + * Huh! same key, different name? Did someone trample + * on some memory? We're most confused. + */ WARN_ON_ONCE(class->name != lock->name); return class; } @@ -800,6 +812,10 @@ out_unlock_set: else if (subclass < NR_LOCKDEP_CACHING_CLASSES) lock->class_cache[subclass] = class; + /* + * Hash collision, did we smoke some? We found a class with a matching + * hash but the subclass -- which is hashed in -- didn't match. + */ if (DEBUG_LOCKS_WARN_ON(class->subclass != subclass)) return NULL; @@ -926,7 +942,7 @@ static inline void mark_lock_accessed(struct lock_list *lock, unsigned long nr; nr = lock - list_entries; - WARN_ON(nr >= nr_list_entries); + WARN_ON(nr >= nr_list_entries); /* Out-of-bounds, input fail */ lock->parent = parent; lock->class->dep_gen_id = lockdep_dependency_gen_id; } @@ -936,7 +952,7 @@ static inline unsigned long lock_accessed(struct lock_list *lock) unsigned long nr; nr = lock - list_entries; - WARN_ON(nr >= nr_list_entries); + WARN_ON(nr >= nr_list_entries); /* Out-of-bounds, input fail */ return lock->class->dep_gen_id == lockdep_dependency_gen_id; } @@ -1196,6 +1212,9 @@ static noinline int print_bfs_bug(int ret) if (!debug_locks_off_graph_unlock()) return 0; + /* + * Breadth-first-search failed, graph got corrupted? + */ WARN(1, "lockdep bfs error:%d\n", ret); return 0; @@ -1944,6 +1963,11 @@ out_bug: if (!debug_locks_off_graph_unlock()) return 0; + /* + * Clearly we all shouldn't be here, but since we made it we + * can reliable say we messed up our state. See the above two + * gotos for reasons why we could possibly end up here. + */ WARN_ON(1); return 0; @@ -1975,6 +1999,11 @@ static inline int lookup_chain_cache(struct task_struct *curr, struct held_lock *hlock_curr, *hlock_next; int i, j; + /* + * We might need to take the graph lock, ensure we've got IRQs + * disabled to make this an IRQ-safe lock.. for recursion reasons + * lockdep won't complain about its own locking errors. + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return 0; /* @@ -2126,6 +2155,10 @@ static void check_chain_key(struct task_struct *curr) hlock = curr->held_locks + i; if (chain_key != hlock->prev_chain_key) { debug_locks_off(); + /* + * We got mighty confused, our chain keys don't match + * with what we expect, someone trample on our task state? + */ WARN(1, "hm#1, depth: %u [%u], %016Lx != %016Lx\n", curr->lockdep_depth, i, (unsigned long long)chain_key, @@ -2133,6 +2166,9 @@ static void check_chain_key(struct task_struct *curr) return; } id = hlock->class_idx - 1; + /* + * Whoops ran out of static storage again? + */ if (DEBUG_LOCKS_WARN_ON(id >= MAX_LOCKDEP_KEYS)) return; @@ -2144,6 +2180,10 @@ static void check_chain_key(struct task_struct *curr) } if (chain_key != curr->curr_chain_key) { debug_locks_off(); + /* + * More smoking hash instead of calculating it, damn see these + * numbers float.. I bet that a pink elephant stepped on my memory. + */ WARN(1, "hm#2, depth: %u [%u], %016Lx != %016Lx\n", curr->lockdep_depth, i, (unsigned long long)chain_key, @@ -2525,12 +2565,24 @@ void trace_hardirqs_on_caller(unsigned long ip) return; } + /* + * We're enabling irqs and according to our state above irqs weren't + * already enabled, yet we find the hardware thinks they are in fact + * enabled.. someone messed up their IRQ state tracing. + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return; + /* + * See the fine text that goes along with this variable definition. + */ if (DEBUG_LOCKS_WARN_ON(unlikely(early_boot_irqs_disabled))) return; + /* + * Can't allow enabling interrupts while in an interrupt handler, + * that's general bad form and such. Recursion, limited stack etc.. + */ if (DEBUG_LOCKS_WARN_ON(current->hardirq_context)) return; @@ -2558,6 +2610,10 @@ void trace_hardirqs_off_caller(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; + /* + * So we're supposed to get called after you mask local IRQs, but for + * some reason the hardware doesn't quite think you did a proper job. + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return; @@ -2590,6 +2646,10 @@ void trace_softirqs_on(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; + /* + * We fancy IRQs being disabled here, see softirq.c, avoids + * funny state and nesting things. + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return; @@ -2626,6 +2686,9 @@ void trace_softirqs_off(unsigned long ip) if (unlikely(!debug_locks || current->lockdep_recursion)) return; + /* + * We fancy IRQs being disabled here, see softirq.c + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return; @@ -2637,6 +2700,9 @@ void trace_softirqs_off(unsigned long ip) curr->softirq_disable_ip = ip; curr->softirq_disable_event = ++curr->irq_events; debug_atomic_inc(softirqs_off_events); + /* + * Whoops, we wanted softirqs off, so why aren't they? + */ DEBUG_LOCKS_WARN_ON(!softirq_count()); } else debug_atomic_inc(redundant_softirqs_off); @@ -2661,6 +2727,9 @@ static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) if (!(gfp_mask & __GFP_FS)) return; + /* + * Oi! Can't be having __GFP_FS allocations with IRQs disabled. + */ if (DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))) return; @@ -2773,13 +2842,13 @@ static int separate_irq_context(struct task_struct *curr, return 0; } -#else +#else /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */ static inline int mark_lock_irq(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit new_bit) { - WARN_ON(1); + WARN_ON(1); /* Impossible innit? when we don't have TRACE_IRQFLAG */ return 1; } @@ -2799,7 +2868,7 @@ void lockdep_trace_alloc(gfp_t gfp_mask) { } -#endif +#endif /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */ /* * Mark a lock with a usage bit, and validate the state transition: @@ -2880,6 +2949,9 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, lock->cpu = raw_smp_processor_id(); #endif + /* + * Can't be having no nameless bastards around this place! + */ if (DEBUG_LOCKS_WARN_ON(!name)) { lock->name = "NULL"; return; @@ -2887,6 +2959,9 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, lock->name = name; + /* + * No key, no joy, we need to hash something. + */ if (DEBUG_LOCKS_WARN_ON(!key)) return; /* @@ -2894,6 +2969,9 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, */ if (!static_obj(key)) { printk("BUG: key %p not in .data!\n", key); + /* + * What it says above ^^^^^, I suggest you read it. + */ DEBUG_LOCKS_WARN_ON(1); return; } @@ -2932,6 +3010,11 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (unlikely(!debug_locks)) return 0; + /* + * Lockdep should run with IRQs disabled, otherwise we could + * get an interrupt which would want to take locks, which would + * end up in lockdep and have you got a head-ache already? + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return 0; @@ -2963,6 +3046,9 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, * dependency checks are done) */ depth = curr->lockdep_depth; + /* + * Ran out of static storage for our per-task lock stack again have we? + */ if (DEBUG_LOCKS_WARN_ON(depth >= MAX_LOCK_DEPTH)) return 0; @@ -2981,6 +3067,10 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, } hlock = curr->held_locks + depth; + /* + * Plain impossible, we just registered it and checked it weren't no + * NULL like.. I bet this mushroom I ate was good! + */ if (DEBUG_LOCKS_WARN_ON(!class)) return 0; hlock->class_idx = class_idx; @@ -3015,11 +3105,17 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, * the hash, not class->key. */ id = class - lock_classes; + /* + * Whoops, we did it again.. ran straight out of our static allocation. + */ if (DEBUG_LOCKS_WARN_ON(id >= MAX_LOCKDEP_KEYS)) return 0; chain_key = curr->curr_chain_key; if (!depth) { + /* + * How can we have a chain hash when we ain't got no keys?! + */ if (DEBUG_LOCKS_WARN_ON(chain_key != 0)) return 0; chain_head = 1; @@ -3091,6 +3187,9 @@ static int check_unlock(struct task_struct *curr, struct lockdep_map *lock, { if (unlikely(!debug_locks)) return 0; + /* + * Lockdep should run with IRQs disabled, recursion, head-ache, etc.. + */ if (DEBUG_LOCKS_WARN_ON(!irqs_disabled())) return 0; @@ -3120,6 +3219,11 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock) if (!class) return 0; + /* + * References, but not a lock we're actually ref-counting? + * State got messed up, follow the sites that change ->references + * and try to make sense of it. + */ if (DEBUG_LOCKS_WARN_ON(!hlock->nest_lock)) return 0; @@ -3142,6 +3246,10 @@ __lock_set_class(struct lockdep_map *lock, const char *name, int i; depth = curr->lockdep_depth; + /* + * This function is about (re)setting the class of a held lock, + * yet we're not actually holding any locks. Naughty user! + */ if (DEBUG_LOCKS_WARN_ON(!depth)) return 0; @@ -3177,6 +3285,10 @@ found_it: return 0; } + /* + * I took it apart and put it back together again, except now I have + * these 'spare' parts.. where shall I put them. + */ if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth)) return 0; return 1; @@ -3201,6 +3313,10 @@ lock_release_non_nested(struct task_struct *curr, * of held locks: */ depth = curr->lockdep_depth; + /* + * So we're all set to release this lock.. wait what lock? We don't + * own any locks, you've been drinking again? + */ if (DEBUG_LOCKS_WARN_ON(!depth)) return 0; @@ -3253,6 +3369,10 @@ found_it: return 0; } + /* + * We had N bottles of beer on the wall, we drank one, but now + * there's not N-1 bottles of beer left on the wall... + */ if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth - 1)) return 0; return 1; @@ -3283,6 +3403,9 @@ static int lock_release_nested(struct task_struct *curr, return lock_release_non_nested(curr, lock, ip); curr->lockdep_depth--; + /* + * No more locks, but somehow we've got hash left over, who left it? + */ if (DEBUG_LOCKS_WARN_ON(!depth && (hlock->prev_chain_key != 0))) return 0; @@ -3365,10 +3488,13 @@ static void check_flags(unsigned long flags) * check if not in hardirq contexts: */ if (!hardirq_count()) { - if (softirq_count()) + if (softirq_count()) { + /* like the above, but with softirqs */ DEBUG_LOCKS_WARN_ON(current->softirqs_enabled); - else + } else { + /* lick the above, does it taste good? */ DEBUG_LOCKS_WARN_ON(!current->softirqs_enabled); + } } if (!debug_locks) @@ -3506,6 +3632,10 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) int i, contention_point, contending_point; depth = curr->lockdep_depth; + /* + * Whee, we contended on this lock, except it seems we're not + * actually trying to acquire anything much at all.. + */ if (DEBUG_LOCKS_WARN_ON(!depth)) return; @@ -3555,6 +3685,10 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) int i, cpu; depth = curr->lockdep_depth; + /* + * Yay, we acquired ownership of this lock we didn't try to + * acquire, how the heck did that happen? + */ if (DEBUG_LOCKS_WARN_ON(!depth)) return; @@ -3759,8 +3893,12 @@ void lockdep_reset_lock(struct lockdep_map *lock) match |= class == lock->class_cache[j]; if (unlikely(match)) { - if (debug_locks_off_graph_unlock()) + if (debug_locks_off_graph_unlock()) { + /* + * We all just reset everything, how did it match? + */ WARN_ON(1); + } goto out_restore; } } -- cgit v1.3-8-gc7d7 From cba9bd22a5f8f857534b9a7f3fb3cafa0ac5fb75 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 12 Sep 2011 13:40:05 +0200 Subject: watchdog: Drop FIFO policy in exit path When the watchdog thread exits it runs through the exit path with FIFO priority. There is no point in doing so. Switch back to SCHED_NORMAL before exiting. Cc: Don Zickus Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1109121337461.2723@ionos Signed-off-by: Ingo Molnar --- kernel/watchdog.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index e952a1394d26..d680381b0e9c 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -321,7 +321,7 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) */ static int watchdog(void *unused) { - static struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; + struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; struct hrtimer *hrtimer = &__raw_get_cpu_var(watchdog_hrtimer); sched_setscheduler(current, SCHED_FIFO, ¶m); @@ -350,7 +350,8 @@ static int watchdog(void *unused) set_current_state(TASK_INTERRUPTIBLE); } __set_current_state(TASK_RUNNING); - + param.sched_priority = 0; + sched_setscheduler(current, SCHED_NORMAL, ¶m); return 0; } -- cgit v1.3-8-gc7d7 From 6249687f76b69cc0b2ad34636f4a18d693ef3262 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 19 Sep 2011 11:35:58 -0400 Subject: tracing: Add a counter clock for those that do not trust clocks When debugging tight race conditions, it can be helpful to have a synchronized tracing method. Although in most cases the global clock provides this functionality, if timings is not the issue, it is more comforting to know that the order of events really happened in a precise order. Instead of using a clock, add a "counter" that is simply an incrementing atomic 64bit counter that orders the events as they are perceived to happen. The trace_clock_counter() is added from the attempt by Peter Zijlstra trying to convert the trace_clock_global() to it. I took Peter's counter code and made trace_clock_counter() instead, and added it to the choice of clocks. Just echo counter > /debug/tracing/trace_clock to activate it. Requested-by: Thomas Gleixner Requested-by: Peter Zijlstra Reviewed-By: Valdis Kletnieks Signed-off-by: Steven Rostedt --- include/linux/trace_clock.h | 1 + kernel/trace/trace.c | 1 + kernel/trace/trace_clock.c | 12 ++++++++++++ 3 files changed, 14 insertions(+) (limited to 'kernel') diff --git a/include/linux/trace_clock.h b/include/linux/trace_clock.h index 7a8130384087..4eb490237d4c 100644 --- a/include/linux/trace_clock.h +++ b/include/linux/trace_clock.h @@ -15,5 +15,6 @@ extern u64 notrace trace_clock_local(void); extern u64 notrace trace_clock(void); extern u64 notrace trace_clock_global(void); +extern u64 notrace trace_clock_counter(void); #endif /* _LINUX_TRACE_CLOCK_H */ diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index b41907000bc6..4b8df0dc9358 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -435,6 +435,7 @@ static struct { } trace_clocks[] = { { trace_clock_local, "local" }, { trace_clock_global, "global" }, + { trace_clock_counter, "counter" }, }; int trace_clock_id; diff --git a/kernel/trace/trace_clock.c b/kernel/trace/trace_clock.c index 6302747a1398..394783531cbb 100644 --- a/kernel/trace/trace_clock.c +++ b/kernel/trace/trace_clock.c @@ -113,3 +113,15 @@ u64 notrace trace_clock_global(void) return now; } + +static atomic64_t trace_counter; + +/* + * trace_clock_counter(): simply an atomic counter. + * Use the trace_counter "counter" for cases where you do not care + * about timings, but are interested in strict ordering. + */ +u64 notrace trace_clock_counter(void) +{ + return atomic64_add_return(1, &trace_counter); +} -- cgit v1.3-8-gc7d7 From 1a51410abe7d0ee4b1d112780f46df87d3621043 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2011 17:04:37 -0700 Subject: Make TASKSTATS require root access Ok, this isn't optimal, since it means that 'iotop' needs admin capabilities, and we may have to work on this some more. But at the same time it is very much not acceptable to let anybody just read anybody elses IO statistics quite at this level. Use of the GENL_ADMIN_PERM suggested by Johannes Berg as an alternative to checking the capabilities by hand. Reported-by: Vasiliy Kulikov Cc: Johannes Berg Acked-by: Balbir Singh Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/taskstats.c b/kernel/taskstats.c index e19ce1454ee1..e66046456f4f 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -655,6 +655,7 @@ static struct genl_ops taskstats_ops = { .cmd = TASKSTATS_CMD_GET, .doit = taskstats_user_cmd, .policy = taskstats_cmd_get_policy, + .flags = GENL_ADMIN_PERM, }; static struct genl_ops cgroupstats_ops = { -- cgit v1.3-8-gc7d7 From 58c3c3aa01b455ecb99d61ce73f1444274af696b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 19 Sep 2011 17:10:57 -0700 Subject: Make taskstats round statistics down to nearest 1k bytes/events Even with just the interface limited to admin, there really is little to reason to give byte-per-byte counts for taskstats. So round it down to something less intrusive. Acked-by: Balbir Singh Signed-off-by: Linus Torvalds --- kernel/tsacct.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/tsacct.c b/kernel/tsacct.c index 24dc60d9fa1f..5bbfac85866e 100644 --- a/kernel/tsacct.c +++ b/kernel/tsacct.c @@ -78,6 +78,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk) #define KB 1024 #define MB (1024*KB) +#define KB_MASK (~(KB-1)) /* * fill in extended accounting fields */ @@ -95,14 +96,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p) stats->hiwater_vm = get_mm_hiwater_vm(mm) * PAGE_SIZE / KB; mmput(mm); } - stats->read_char = p->ioac.rchar; - stats->write_char = p->ioac.wchar; - stats->read_syscalls = p->ioac.syscr; - stats->write_syscalls = p->ioac.syscw; + stats->read_char = p->ioac.rchar & KB_MASK; + stats->write_char = p->ioac.wchar & KB_MASK; + stats->read_syscalls = p->ioac.syscr & KB_MASK; + stats->write_syscalls = p->ioac.syscw & KB_MASK; #ifdef CONFIG_TASK_IO_ACCOUNTING - stats->read_bytes = p->ioac.read_bytes; - stats->write_bytes = p->ioac.write_bytes; - stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes; + stats->read_bytes = p->ioac.read_bytes & KB_MASK; + stats->write_bytes = p->ioac.write_bytes & KB_MASK; + stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes & KB_MASK; #else stats->read_bytes = 0; stats->write_bytes = 0; -- cgit v1.3-8-gc7d7 From 590e4d857153c5d4cf86052cdfd42cf9b0779841 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Sun, 24 Jul 2011 16:33:13 +0000 Subject: sched: Allow SD_NODES_PER_DOMAIN to be overridden We want to override the default value of SD_NODES_PER_DOMAIN on ppc64, so move it into linux/topology.h. Signed-off-by: Anton Blanchard Acked-by: Peter Zijlstra Signed-off-by: Benjamin Herrenschmidt --- include/linux/topology.h | 4 ++++ kernel/sched.c | 2 -- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/topology.h b/include/linux/topology.h index fc839bfa7935..e26db031303b 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -201,6 +201,10 @@ int arch_update_cpu_topology(void); .balance_interval = 64, \ } +#ifndef SD_NODES_PER_DOMAIN +#define SD_NODES_PER_DOMAIN 16 +#endif + #ifdef CONFIG_SCHED_BOOK #ifndef SD_BOOK_INIT #error Please define an appropriate SD_BOOK_INIT in include/asm/topology.h!!! diff --git a/kernel/sched.c b/kernel/sched.c index ec5f472bc5b9..d258b2d9a66d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6947,8 +6947,6 @@ static int __init isolated_cpu_setup(char *str) __setup("isolcpus=", isolated_cpu_setup); -#define SD_NODES_PER_DOMAIN 16 - #ifdef CONFIG_NUMA /** -- cgit v1.3-8-gc7d7 From eef24afb28561a5a9f4be8f8da97735b7e6a826f Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Wed, 14 Sep 2011 11:31:37 -0500 Subject: irq: Fix check for already initialized irq_domain in irq_domain_add The sanity check in irq_domain_add() tests desc->irq_data != NULL or irq_data->domain != NULL. This prevents adding an irq_domain to a irq descriptor when irq_data exists, which true when the irq descriptor exists. This went unnoticed so far as the simple domain code did not enter this code path because domain->nr_irqs is always 0 for the simple domains. Split the check for irq_data == NULL out and have a separate warning for it. [ tglx: Made the check for irq_data == NULL separate ] Signed-off-by: Rob Herring Cc: Grant Likely Cc: marc.zyngier@arm.com Cc: thomas.abraham@linaro.org Cc: jamie@jamieiles.com Cc: b-cousson@ti.com Cc: shawn.guo@linaro.org Cc: linux-arm-kernel@lists.infradead.org Cc: devicetree-discuss@lists.ozlabs.org Link: http://lkml.kernel.org/r/1316017900-19918-3-git-send-email-robherring2@gmail.com Signed-off-by: Thomas Gleixner --- kernel/irq/irqdomain.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index d5828da3fd38..b57a3776de44 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -29,7 +29,11 @@ void irq_domain_add(struct irq_domain *domain) */ for (hwirq = 0; hwirq < domain->nr_irq; hwirq++) { d = irq_get_irq_data(irq_domain_to_irq(domain, hwirq)); - if (d || d->domain) { + if (!d) { + WARN(1, "error: assigning domain to non existant irq_desc"); + return; + } + if (d->domain) { /* things are broken; just report, don't clean up */ WARN(1, "error: irq_desc already assigned to a domain"); return; -- cgit v1.3-8-gc7d7 From cbbc719fccdb8cbd87350a05c0d33167c9b79365 Mon Sep 17 00:00:00 2001 From: hank Date: Tue, 20 Sep 2011 13:53:39 -0700 Subject: time: Change jiffies_to_clock_t() argument type to unsigned long The parameter's origin type is long. On an i386 architecture, it can easily be larger than 0x80000000, causing this function to convert it to a sign-extended u64 type. Change the type to unsigned long so we get the correct result. Signed-off-by: hank Cc: John Stultz Cc: [ build fix ] Signed-off-by: Andrew Morton Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/linux/jiffies.h | 2 +- kernel/time.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index f97672a36fa8..265e2c3cbd1c 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -303,7 +303,7 @@ extern void jiffies_to_timespec(const unsigned long jiffies, extern unsigned long timeval_to_jiffies(const struct timeval *value); extern void jiffies_to_timeval(const unsigned long jiffies, struct timeval *value); -extern clock_t jiffies_to_clock_t(long x); +extern clock_t jiffies_to_clock_t(unsigned long x); extern unsigned long clock_t_to_jiffies(unsigned long x); extern u64 jiffies_64_to_clock_t(u64 x); extern u64 nsec_to_clock_t(u64 x); diff --git a/kernel/time.c b/kernel/time.c index 8e8dc6d705c9..d77606214529 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -575,7 +575,7 @@ EXPORT_SYMBOL(jiffies_to_timeval); /* * Convert jiffies/jiffies_64 to clock_t and back. */ -clock_t jiffies_to_clock_t(long x) +clock_t jiffies_to_clock_t(unsigned long x) { #if (TICK_NSEC % (NSEC_PER_SEC / USER_HZ)) == 0 # if HZ < USER_HZ -- cgit v1.3-8-gc7d7 From e36de1de4a5f95b7cb3e5c37d10e6bbb91833ef0 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 22 Sep 2011 11:11:51 -0400 Subject: tracing: Fix preemptirqsoff tracer to not stop at preempt off If irqs are disabled when preemption count reaches zero, the preemptirqsoff tracer should not flag that as the end. When interrupts are enabled and preemption count is not zero the preemptirqsoff correctly continues its tracing. Signed-off-by: Steven Rostedt --- kernel/trace/trace_irqsoff.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 667aa8cc0cfc..a1a3359996a7 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -505,13 +505,13 @@ EXPORT_SYMBOL(trace_hardirqs_off_caller); #ifdef CONFIG_PREEMPT_TRACER void trace_preempt_on(unsigned long a0, unsigned long a1) { - if (preempt_trace()) + if (preempt_trace() && !irq_trace()) stop_critical_timing(a0, a1); } void trace_preempt_off(unsigned long a0, unsigned long a1) { - if (preempt_trace()) + if (preempt_trace() && !irq_trace()) start_critical_timing(a0, a1); } #endif /* CONFIG_PREEMPT_TRACER */ -- cgit v1.3-8-gc7d7 From ab10023e0088d5075354afc7cb9e72304757dddd Mon Sep 17 00:00:00 2001 From: Colin Cross Date: Thu, 10 Feb 2011 02:04:45 -0800 Subject: cpu_pm: Add cpu power management notifiers During some CPU power modes entered during idle, hotplug and suspend, peripherals located in the CPU power domain, such as the GIC, localtimers, and VFP, may be powered down. Add a notifier chain that allows drivers for those peripherals to be notified before and after they may be reset. Notified drivers can include VFP co-processor, interrupt controller and it's PM extensions, local CPU timers context save/restore which shouldn't be interrupted. Hence CPU PM event APIs must be called with interrupts disabled. Signed-off-by: Colin Cross Signed-off-by: Santosh Shilimkar Reviewed-by: Kevin Hilman Tested-and-Acked-by: Shawn Guo Tested-by: Kevin Hilman Tested-by: Vishwanath BS --- include/linux/cpu_pm.h | 109 +++++++++++++++++++++++++++ kernel/Makefile | 1 + kernel/cpu_pm.c | 200 +++++++++++++++++++++++++++++++++++++++++++++++++ kernel/power/Kconfig | 4 + 4 files changed, 314 insertions(+) create mode 100644 include/linux/cpu_pm.h create mode 100644 kernel/cpu_pm.c (limited to 'kernel') diff --git a/include/linux/cpu_pm.h b/include/linux/cpu_pm.h new file mode 100644 index 000000000000..455b233dd3b1 --- /dev/null +++ b/include/linux/cpu_pm.h @@ -0,0 +1,109 @@ +/* + * Copyright (C) 2011 Google, Inc. + * + * Author: + * Colin Cross + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#ifndef _LINUX_CPU_PM_H +#define _LINUX_CPU_PM_H + +#include +#include + +/* + * When a CPU goes to a low power state that turns off power to the CPU's + * power domain, the contents of some blocks (floating point coprocessors, + * interrupt controllers, caches, timers) in the same power domain can + * be lost. The cpm_pm notifiers provide a method for platform idle, suspend, + * and hotplug implementations to notify the drivers for these blocks that + * they may be reset. + * + * All cpu_pm notifications must be called with interrupts disabled. + * + * The notifications are split into two classes: CPU notifications and CPU + * cluster notifications. + * + * CPU notifications apply to a single CPU and must be called on the affected + * CPU. They are used to save per-cpu context for affected blocks. + * + * CPU cluster notifications apply to all CPUs in a single power domain. They + * are used to save any global context for affected blocks, and must be called + * after all the CPUs in the power domain have been notified of the low power + * state. + */ + +/* + * Event codes passed as unsigned long val to notifier calls + */ +enum cpu_pm_event { + /* A single cpu is entering a low power state */ + CPU_PM_ENTER, + + /* A single cpu failed to enter a low power state */ + CPU_PM_ENTER_FAILED, + + /* A single cpu is exiting a low power state */ + CPU_PM_EXIT, + + /* A cpu power domain is entering a low power state */ + CPU_CLUSTER_PM_ENTER, + + /* A cpu power domain failed to enter a low power state */ + CPU_CLUSTER_PM_ENTER_FAILED, + + /* A cpu power domain is exiting a low power state */ + CPU_CLUSTER_PM_EXIT, +}; + +#ifdef CONFIG_CPU_PM +int cpu_pm_register_notifier(struct notifier_block *nb); +int cpu_pm_unregister_notifier(struct notifier_block *nb); +int cpu_pm_enter(void); +int cpu_pm_exit(void); +int cpu_cluster_pm_enter(void); +int cpu_cluster_pm_exit(void); + +#else + +static inline int cpu_pm_register_notifier(struct notifier_block *nb) +{ + return 0; +} + +static inline int cpu_pm_unregister_notifier(struct notifier_block *nb) +{ + return 0; +} + +static inline int cpu_pm_enter(void) +{ + return 0; +} + +static inline int cpu_pm_exit(void) +{ + return 0; +} + +static inline int cpu_cluster_pm_enter(void) +{ + return 0; +} + +static inline int cpu_cluster_pm_exit(void) +{ + return 0; +} +#endif +#endif diff --git a/kernel/Makefile b/kernel/Makefile index eca595e2fd52..988cb3da7031 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -101,6 +101,7 @@ obj-$(CONFIG_RING_BUFFER) += trace/ obj-$(CONFIG_TRACEPOINTS) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o obj-$(CONFIG_IRQ_WORK) += irq_work.o +obj-$(CONFIG_CPU_PM) += cpu_pm.o obj-$(CONFIG_PERF_EVENTS) += events/ diff --git a/kernel/cpu_pm.c b/kernel/cpu_pm.c new file mode 100644 index 000000000000..4d1ff4acd04b --- /dev/null +++ b/kernel/cpu_pm.c @@ -0,0 +1,200 @@ +/* + * Copyright (C) 2011 Google, Inc. + * + * Author: + * Colin Cross + * + * This software is licensed under the terms of the GNU General Public + * License version 2, as published by the Free Software Foundation, and + * may be copied, distributed, and modified under those terms. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ + +#include +#include +#include +#include +#include + +static DEFINE_RWLOCK(cpu_pm_notifier_lock); +static RAW_NOTIFIER_HEAD(cpu_pm_notifier_chain); + +static int cpu_pm_notify(enum cpu_pm_event event, int nr_to_call, int *nr_calls) +{ + int ret; + + ret = __raw_notifier_call_chain(&cpu_pm_notifier_chain, event, NULL, + nr_to_call, nr_calls); + + return notifier_to_errno(ret); +} + +/** + * cpu_pm_register_notifier - register a driver with cpu_pm + * @nb: notifier block to register + * + * Add a driver to a list of drivers that are notified about + * CPU and CPU cluster low power entry and exit. + * + * This function may sleep, and has the same return conditions as + * raw_notifier_chain_register. + */ +int cpu_pm_register_notifier(struct notifier_block *nb) +{ + unsigned long flags; + int ret; + + write_lock_irqsave(&cpu_pm_notifier_lock, flags); + ret = raw_notifier_chain_register(&cpu_pm_notifier_chain, nb); + write_unlock_irqrestore(&cpu_pm_notifier_lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_pm_register_notifier); + +/** + * cpu_pm_unregister_notifier - unregister a driver with cpu_pm + * @nb: notifier block to be unregistered + * + * Remove a driver from the CPU PM notifier list. + * + * This function may sleep, and has the same return conditions as + * raw_notifier_chain_unregister. + */ +int cpu_pm_unregister_notifier(struct notifier_block *nb) +{ + unsigned long flags; + int ret; + + write_lock_irqsave(&cpu_pm_notifier_lock, flags); + ret = raw_notifier_chain_unregister(&cpu_pm_notifier_chain, nb); + write_unlock_irqrestore(&cpu_pm_notifier_lock, flags); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_pm_unregister_notifier); + +/** + * cpm_pm_enter - CPU low power entry notifier + * + * Notifies listeners that a single CPU is entering a low power state that may + * cause some blocks in the same power domain as the cpu to reset. + * + * Must be called on the affected CPU with interrupts disabled. Platform is + * responsible for ensuring that cpu_pm_enter is not called twice on the same + * CPU before cpu_pm_exit is called. Notified drivers can include VFP + * co-processor, interrupt controller and it's PM extensions, local CPU + * timers context save/restore which shouldn't be interrupted. Hence it + * must be called with interrupts disabled. + * + * Return conditions are same as __raw_notifier_call_chain. + */ +int cpu_pm_enter(void) +{ + int nr_calls; + int ret = 0; + + read_lock(&cpu_pm_notifier_lock); + ret = cpu_pm_notify(CPU_PM_ENTER, -1, &nr_calls); + if (ret) + /* + * Inform listeners (nr_calls - 1) about failure of CPU PM + * PM entry who are notified earlier to prepare for it. + */ + cpu_pm_notify(CPU_PM_ENTER_FAILED, nr_calls - 1, NULL); + read_unlock(&cpu_pm_notifier_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_pm_enter); + +/** + * cpm_pm_exit - CPU low power exit notifier + * + * Notifies listeners that a single CPU is exiting a low power state that may + * have caused some blocks in the same power domain as the cpu to reset. + * + * Notified drivers can include VFP co-processor, interrupt controller + * and it's PM extensions, local CPU timers context save/restore which + * shouldn't be interrupted. Hence it must be called with interrupts disabled. + * + * Return conditions are same as __raw_notifier_call_chain. + */ +int cpu_pm_exit(void) +{ + int ret; + + read_lock(&cpu_pm_notifier_lock); + ret = cpu_pm_notify(CPU_PM_EXIT, -1, NULL); + read_unlock(&cpu_pm_notifier_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_pm_exit); + +/** + * cpm_cluster_pm_enter - CPU cluster low power entry notifier + * + * Notifies listeners that all cpus in a power domain are entering a low power + * state that may cause some blocks in the same power domain to reset. + * + * Must be called after cpu_pm_enter has been called on all cpus in the power + * domain, and before cpu_pm_exit has been called on any cpu in the power + * domain. Notified drivers can include VFP co-processor, interrupt controller + * and it's PM extensions, local CPU timers context save/restore which + * shouldn't be interrupted. Hence it must be called with interrupts disabled. + * + * Must be called with interrupts disabled. + * + * Return conditions are same as __raw_notifier_call_chain. + */ +int cpu_cluster_pm_enter(void) +{ + int nr_calls; + int ret = 0; + + read_lock(&cpu_pm_notifier_lock); + ret = cpu_pm_notify(CPU_CLUSTER_PM_ENTER, -1, &nr_calls); + if (ret) + /* + * Inform listeners (nr_calls - 1) about failure of CPU cluster + * PM entry who are notified earlier to prepare for it. + */ + cpu_pm_notify(CPU_CLUSTER_PM_ENTER_FAILED, nr_calls - 1, NULL); + read_unlock(&cpu_pm_notifier_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_cluster_pm_enter); + +/** + * cpm_cluster_pm_exit - CPU cluster low power exit notifier + * + * Notifies listeners that all cpus in a power domain are exiting form a + * low power state that may have caused some blocks in the same power domain + * to reset. + * + * Must be called after cpu_pm_exit has been called on all cpus in the power + * domain, and before cpu_pm_exit has been called on any cpu in the power + * domain. Notified drivers can include VFP co-processor, interrupt controller + * and it's PM extensions, local CPU timers context save/restore which + * shouldn't be interrupted. Hence it must be called with interrupts disabled. + * + * Return conditions are same as __raw_notifier_call_chain. + */ +int cpu_cluster_pm_exit(void) +{ + int ret; + + read_lock(&cpu_pm_notifier_lock); + ret = cpu_pm_notify(CPU_CLUSTER_PM_EXIT, -1, NULL); + read_unlock(&cpu_pm_notifier_lock); + + return ret; +} +EXPORT_SYMBOL_GPL(cpu_cluster_pm_exit); diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 3744c594b19b..80a85971cf64 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -235,3 +235,7 @@ config PM_GENERIC_DOMAINS config PM_GENERIC_DOMAINS_RUNTIME def_bool y depends on PM_RUNTIME && PM_GENERIC_DOMAINS + +config CPU_PM + bool + depends on SUSPEND || CPU_IDLE -- cgit v1.3-8-gc7d7 From 6f3eaec87b6b17bfa49cb3b5b8d07fa84be18512 Mon Sep 17 00:00:00 2001 From: Colin Cross Date: Fri, 22 Jul 2011 14:57:09 -0700 Subject: cpu_pm: call notifiers during suspend Implements syscore_ops in cpu_pm to call the cpu and cpu cluster notifiers during suspend and resume, allowing drivers receiving the notifications to avoid implementing syscore_ops. Signed-off-by: Colin Cross Signed-off-by: Santosh Shilimkar Reviewed-by: Kevin Hilman Tested-and-Acked-by: Shawn Guo Tested-by: Vishwanath BS --- kernel/cpu_pm.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'kernel') diff --git a/kernel/cpu_pm.c b/kernel/cpu_pm.c index 4d1ff4acd04b..249152e15308 100644 --- a/kernel/cpu_pm.c +++ b/kernel/cpu_pm.c @@ -20,6 +20,7 @@ #include #include #include +#include static DEFINE_RWLOCK(cpu_pm_notifier_lock); static RAW_NOTIFIER_HEAD(cpu_pm_notifier_chain); @@ -198,3 +199,35 @@ int cpu_cluster_pm_exit(void) return ret; } EXPORT_SYMBOL_GPL(cpu_cluster_pm_exit); + +#ifdef CONFIG_PM +static int cpu_pm_suspend(void) +{ + int ret; + + ret = cpu_pm_enter(); + if (ret) + return ret; + + ret = cpu_cluster_pm_enter(); + return ret; +} + +static void cpu_pm_resume(void) +{ + cpu_cluster_pm_exit(); + cpu_pm_exit(); +} + +static struct syscore_ops cpu_pm_syscore_ops = { + .suspend = cpu_pm_suspend, + .resume = cpu_pm_resume, +}; + +static int cpu_pm_init(void) +{ + register_syscore_ops(&cpu_pm_syscore_ops); + return 0; +} +core_initcall(cpu_pm_init); +#endif -- cgit v1.3-8-gc7d7 From f9d81f61c84aca693bc353dfef4b8c36c2e5e1b5 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 25 Sep 2011 19:46:22 +0200 Subject: ptrace: PTRACE_LISTEN forgets to unlock ->siglock If PTRACE_LISTEN fails after lock_task_sighand() it doesn't drop ->siglock. Reported-by: Matt Fleming Signed-off-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 23 ++++++++++------------- 1 file changed, 10 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 9de3ecfd20f9..a70d2a5d8c7b 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -744,20 +744,17 @@ int ptrace_request(struct task_struct *child, long request, break; si = child->last_siginfo; - if (unlikely(!si || si->si_code >> 8 != PTRACE_EVENT_STOP)) - break; - - child->jobctl |= JOBCTL_LISTENING; - - /* - * If NOTIFY is set, it means event happened between start - * of this trap and now. Trigger re-trap immediately. - */ - if (child->jobctl & JOBCTL_TRAP_NOTIFY) - signal_wake_up(child, true); - + if (likely(si && (si->si_code >> 8) == PTRACE_EVENT_STOP)) { + child->jobctl |= JOBCTL_LISTENING; + /* + * If NOTIFY is set, it means event happened between + * start of this trap and now. Trigger re-trap. + */ + if (child->jobctl & JOBCTL_TRAP_NOTIFY) + signal_wake_up(child, true); + ret = 0; + } unlock_task_sighand(child, &flags); - ret = 0; break; case PTRACE_DETACH: /* detach a process that was attached. */ -- cgit v1.3-8-gc7d7 From 6ebbe7a07b3bc40b168d2afc569a6543c020d2e3 Mon Sep 17 00:00:00 2001 From: Simon Kirby Date: Thu, 22 Sep 2011 17:03:46 -0700 Subject: sched: Fix up wchan borkage Commit c259e01a1ec ("sched: Separate the scheduler entry for preemption") contained a boo-boo wrecking wchan output. It forgot to put the new schedule() function in the __sched section and thereby doesn't get properly ignored for things like wchan. Tested-by: Simon Kirby Cc: stable@kernel.org # 2.6.39+ Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20110923000346.GA25425@hostway.ca Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index ec5f472bc5b9..d249ea88428c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4372,7 +4372,7 @@ static inline void sched_submit_work(struct task_struct *tsk) blk_schedule_flush_plug(tsk); } -asmlinkage void schedule(void) +asmlinkage void __sched schedule(void) { struct task_struct *tsk = current; -- cgit v1.3-8-gc7d7 From f4cfb33ed980085b14a8f376281edb2e270f0244 Mon Sep 17 00:00:00 2001 From: Wang Xingchao Date: Fri, 16 Sep 2011 13:35:52 -0400 Subject: sched: Remove redundant test in check_preempt_tick() The caller already checks for nr_running > 1, therefore we don't have to do so again. Signed-off-by: Wang Xingchao Reviewed-by: Paul Turner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1316194552-12019-1-git-send-email-xingchao.wang@intel.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 1ca2cd44d64a..fef0bfde7c8c 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1106,6 +1106,8 @@ static void check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) { unsigned long ideal_runtime, delta_exec; + struct sched_entity *se; + s64 delta; ideal_runtime = sched_slice(cfs_rq, curr); delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime; @@ -1127,16 +1129,14 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) if (delta_exec < sysctl_sched_min_granularity) return; - if (cfs_rq->nr_running > 1) { - struct sched_entity *se = __pick_first_entity(cfs_rq); - s64 delta = curr->vruntime - se->vruntime; + se = __pick_first_entity(cfs_rq); + delta = curr->vruntime - se->vruntime; - if (delta < 0) - return; + if (delta < 0) + return; - if (delta > ideal_runtime) - resched_task(rq_of(cfs_rq)->curr); - } + if (delta > ideal_runtime) + resched_task(rq_of(cfs_rq)->curr); } static void -- cgit v1.3-8-gc7d7 From bfb9035c98906aafcd3cf22694fba2550997bf53 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 17 Aug 2011 06:58:04 -0700 Subject: treewide: Correct spelling of successfully in comments Signed-off-by: Joe Perches Signed-off-by: Jiri Kosina --- drivers/media/video/cx18/cx18-mailbox.h | 2 +- drivers/net/igb/e1000_mbx.c | 2 +- drivers/net/igbvf/mbx.c | 2 +- drivers/net/ixgbe/ixgbe_mbx.c | 2 +- drivers/net/ixgbevf/mbx.c | 2 +- drivers/net/phy/broadcom.c | 2 +- drivers/net/tile/tilepro.c | 2 +- drivers/target/iscsi/iscsi_target_nego.c | 2 +- drivers/target/target_core_tpg.c | 2 +- kernel/sched.c | 2 +- sound/core/memalloc.c | 4 ++-- 11 files changed, 12 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/drivers/media/video/cx18/cx18-mailbox.h b/drivers/media/video/cx18/cx18-mailbox.h index 05fe6bdbe062..b63fdfaac49e 100644 --- a/drivers/media/video/cx18/cx18-mailbox.h +++ b/drivers/media/video/cx18/cx18-mailbox.h @@ -69,7 +69,7 @@ struct cx18_mailbox { /* Each command can have up to 6 arguments */ u32 args[MAX_MB_ARGUMENTS]; /* The return code can be one of the codes in the file cx23418.h. If the - command is completed successfuly, the error will be ERR_SYS_SUCCESS. + command is completed successfully, the error will be ERR_SYS_SUCCESS. If it is pending, the code is ERR_SYS_PENDING. If it failed, the error code would indicate the task from which the error originated and will be one of the errors in cx23418.h. In that case, the following diff --git a/drivers/net/igb/e1000_mbx.c b/drivers/net/igb/e1000_mbx.c index 74f2f11ac290..469d95eaa154 100644 --- a/drivers/net/igb/e1000_mbx.c +++ b/drivers/net/igb/e1000_mbx.c @@ -34,7 +34,7 @@ * @size: Length of buffer * @mbx_id: id of mailbox to read * - * returns SUCCESS if it successfuly read message from buffer + * returns SUCCESS if it successfully read message from buffer **/ s32 igb_read_mbx(struct e1000_hw *hw, u32 *msg, u16 size, u16 mbx_id) { diff --git a/drivers/net/igbvf/mbx.c b/drivers/net/igbvf/mbx.c index 3d6f4cc3998a..048aae248d06 100644 --- a/drivers/net/igbvf/mbx.c +++ b/drivers/net/igbvf/mbx.c @@ -288,7 +288,7 @@ out_no_write: * @msg: The message buffer * @size: Length of buffer * - * returns SUCCESS if it successfuly read message from buffer + * returns SUCCESS if it successfully read message from buffer **/ static s32 e1000_read_mbx_vf(struct e1000_hw *hw, u32 *msg, u16 size) { diff --git a/drivers/net/ixgbe/ixgbe_mbx.c b/drivers/net/ixgbe/ixgbe_mbx.c index 1ff0eefcfd0a..3f725d48336d 100644 --- a/drivers/net/ixgbe/ixgbe_mbx.c +++ b/drivers/net/ixgbe/ixgbe_mbx.c @@ -38,7 +38,7 @@ * @size: Length of buffer * @mbx_id: id of mailbox to read * - * returns SUCCESS if it successfuly read message from buffer + * returns SUCCESS if it successfully read message from buffer **/ s32 ixgbe_read_mbx(struct ixgbe_hw *hw, u32 *msg, u16 size, u16 mbx_id) { diff --git a/drivers/net/ixgbevf/mbx.c b/drivers/net/ixgbevf/mbx.c index 7a8833125770..930fa83f2568 100644 --- a/drivers/net/ixgbevf/mbx.c +++ b/drivers/net/ixgbevf/mbx.c @@ -276,7 +276,7 @@ out_no_write: * @msg: The message buffer * @size: Length of buffer * - * returns 0 if it successfuly read message from buffer + * returns 0 if it successfully read message from buffer **/ static s32 ixgbevf_read_mbx_vf(struct ixgbe_hw *hw, u32 *msg, u16 size) { diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c index d84c4224dd12..e8be47d6d7d0 100644 --- a/drivers/net/phy/broadcom.c +++ b/drivers/net/phy/broadcom.c @@ -553,7 +553,7 @@ static int bcm5481_config_aneg(struct phy_device *phydev) /* * There is no BCM5481 specification available, so down * here is everything we know about "register 0x18". This - * at least helps BCM5481 to successfuly receive packets + * at least helps BCM5481 to successfully receive packets * on MPC8360E-RDK board. Peter Barada * says: "This sets delay between the RXD and RXC signals * instead of using trace lengths to achieve timing". diff --git a/drivers/net/tile/tilepro.c b/drivers/net/tile/tilepro.c index 1e2af96fc29c..7b46e75deb5a 100644 --- a/drivers/net/tile/tilepro.c +++ b/drivers/net/tile/tilepro.c @@ -177,7 +177,7 @@ struct tile_net_cpu { struct tile_net_stats_t stats; /* True iff NAPI is enabled. */ bool napi_enabled; - /* True if this tile has succcessfully registered with the IPP. */ + /* True if this tile has successfully registered with the IPP. */ bool registered; /* True if the link was down last time we tried to register. */ bool link_down; diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c index 4d087ac11067..426cd4bf6a9a 100644 --- a/drivers/target/iscsi/iscsi_target_nego.c +++ b/drivers/target/iscsi/iscsi_target_nego.c @@ -504,7 +504,7 @@ static int iscsi_target_do_authentication( break; case 1: pr_debug("iSCSI security negotiation" - " completed sucessfully.\n"); + " completed successfully.\n"); login->auth_complete = 1; if ((login_req->flags & ISCSI_FLAG_LOGIN_NEXT_STAGE1) && (login_req->flags & ISCSI_FLAG_LOGIN_TRANSIT)) { diff --git a/drivers/target/target_core_tpg.c b/drivers/target/target_core_tpg.c index 162b736c7342..49fd0a9b0a56 100644 --- a/drivers/target/target_core_tpg.c +++ b/drivers/target/target_core_tpg.c @@ -593,7 +593,7 @@ int core_tpg_set_initiator_node_queue_depth( if (init_sess) tpg->se_tpg_tfo->close_session(init_sess); - pr_debug("Successfuly changed queue depth to: %d for Initiator" + pr_debug("Successfully changed queue depth to: %d for Initiator" " Node: %s on %s Target Portal Group: %u\n", queue_depth, initiatorname, tpg->se_tpg_tfo->get_fabric_name(), tpg->se_tpg_tfo->tpg_get_tag(tpg)); diff --git a/kernel/sched.c b/kernel/sched.c index ec5f472bc5b9..aecd83b4ffe8 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1739,7 +1739,7 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) #ifdef CONFIG_SMP /* * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be - * successfuly executed on another CPU. We must ensure that updates of + * successfully executed on another CPU. We must ensure that updates of * per-task data have been completed by this moment. */ smp_wmb(); diff --git a/sound/core/memalloc.c b/sound/core/memalloc.c index 16bd9c03679b..691569238435 100644 --- a/sound/core/memalloc.c +++ b/sound/core/memalloc.c @@ -176,7 +176,7 @@ static void snd_free_dev_pages(struct device *dev, size_t size, void *ptr, * Calls the memory-allocator function for the corresponding * buffer type. * - * Returns zero if the buffer with the given size is allocated successfuly, + * Returns zero if the buffer with the given size is allocated successfully, * other a negative value at error. */ int snd_dma_alloc_pages(int type, struct device *device, size_t size, @@ -230,7 +230,7 @@ int snd_dma_alloc_pages(int type, struct device *device, size_t size, * tries to allocate again. The size actually allocated is stored in * res_size argument. * - * Returns zero if the buffer with the given size is allocated successfuly, + * Returns zero if the buffer with the given size is allocated successfully, * other a negative value at error. */ int snd_dma_alloc_pages_fallback(int type, struct device *device, size_t size, -- cgit v1.3-8-gc7d7 From 53b615ccca567ada1931eb04ad0614ac150c14a3 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Tue, 27 Sep 2011 22:53:27 +0200 Subject: PM / Runtime: Introduce trace points for tracing rpm_* functions This patch introduces 3 trace points to prepare for tracing rpm_idle/rpm_suspend/rpm_resume functions, so we can use these trace points to replace the current dev_dbg(). Signed-off-by: Ming Lei Acked-by: Steven Rostedt Signed-off-by: Rafael J. Wysocki --- include/trace/events/rpm.h | 99 ++++++++++++++++++++++++++++++++++++++++++++++ kernel/trace/Makefile | 1 + kernel/trace/rpm-traces.c | 20 ++++++++++ 3 files changed, 120 insertions(+) create mode 100644 include/trace/events/rpm.h create mode 100644 kernel/trace/rpm-traces.c (limited to 'kernel') diff --git a/include/trace/events/rpm.h b/include/trace/events/rpm.h new file mode 100644 index 000000000000..d62c558bf64b --- /dev/null +++ b/include/trace/events/rpm.h @@ -0,0 +1,99 @@ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rpm + +#if !defined(_TRACE_RUNTIME_POWER_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_RUNTIME_POWER_H + +#include +#include +#include + +/* + * The rpm_internal events are used for tracing some important + * runtime pm internal functions. + */ +DECLARE_EVENT_CLASS(rpm_internal, + + TP_PROTO(struct device *dev, int flags), + + TP_ARGS(dev, flags), + + TP_STRUCT__entry( + __string( name, dev_name(dev) ) + __field( int, flags ) + __field( int , usage_count ) + __field( int , disable_depth ) + __field( int , runtime_auto ) + __field( int , request_pending ) + __field( int , irq_safe ) + __field( int , child_count ) + ), + + TP_fast_assign( + __assign_str(name, dev_name(dev)); + __entry->flags = flags; + __entry->usage_count = atomic_read( + &dev->power.usage_count); + __entry->disable_depth = dev->power.disable_depth; + __entry->runtime_auto = dev->power.runtime_auto; + __entry->request_pending = dev->power.request_pending; + __entry->irq_safe = dev->power.irq_safe; + __entry->child_count = atomic_read( + &dev->power.child_count); + ), + + TP_printk("%s flags-%x cnt-%-2d dep-%-2d auto-%-1d p-%-1d" + " irq-%-1d child-%d", + __get_str(name), __entry->flags, + __entry->usage_count, + __entry->disable_depth, + __entry->runtime_auto, + __entry->request_pending, + __entry->irq_safe, + __entry->child_count + ) +); +DEFINE_EVENT(rpm_internal, rpm_suspend, + + TP_PROTO(struct device *dev, int flags), + + TP_ARGS(dev, flags) +); +DEFINE_EVENT(rpm_internal, rpm_resume, + + TP_PROTO(struct device *dev, int flags), + + TP_ARGS(dev, flags) +); +DEFINE_EVENT(rpm_internal, rpm_idle, + + TP_PROTO(struct device *dev, int flags), + + TP_ARGS(dev, flags) +); + +TRACE_EVENT(rpm_return_int, + TP_PROTO(struct device *dev, unsigned long ip, int ret), + TP_ARGS(dev, ip, ret), + + TP_STRUCT__entry( + __string( name, dev_name(dev)) + __field( unsigned long, ip ) + __field( int, ret ) + ), + + TP_fast_assign( + __assign_str(name, dev_name(dev)); + __entry->ip = ip; + __entry->ret = ret; + ), + + TP_printk("%pS:%s ret=%d", (void *)__entry->ip, __get_str(name), + __entry->ret) +); + +#endif /* _TRACE_RUNTIME_POWER_H */ + +/* This part must be outside protection */ +#include diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 761c510a06c5..56bdab5b3793 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -53,6 +53,7 @@ endif obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o +obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o ifeq ($(CONFIG_TRACING),y) obj-$(CONFIG_KGDB_KDB) += trace_kdb.o endif diff --git a/kernel/trace/rpm-traces.c b/kernel/trace/rpm-traces.c new file mode 100644 index 000000000000..4b3b5eaf94d1 --- /dev/null +++ b/kernel/trace/rpm-traces.c @@ -0,0 +1,20 @@ +/* + * Power trace points + * + * Copyright (C) 2009 Ming Lei + */ + +#include +#include +#include +#include +#include +#include + +#define CREATE_TRACE_POINTS +#include + +EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_return_int); +EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_idle); +EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_suspend); +EXPORT_TRACEPOINT_SYMBOL_GPL(rpm_resume); -- cgit v1.3-8-gc7d7 From f786ecba4158880f8cdc0ebb93e7d78e6c125449 Mon Sep 17 00:00:00 2001 From: Vladimir Zapolskiy Date: Wed, 21 Sep 2011 09:26:44 +0000 Subject: connector: add comm change event report to proc connector Add an event to monitor comm value changes of tasks. Such an event becomes vital, if someone desires to control threads of a process in different manner. A natural characteristic of threads is its comm value, and helpfully application developers have an opportunity to change it in runtime. Reporting about such events via proc connector allows to fine-grain monitoring and control potentials, for instance a process control daemon listening to proc connector and following comm value policies can place specific threads to assigned cgroup partitions. It might be possible to achieve a pale partial one-shot likeness without this update, if an application changes comm value of a thread generator task beforehand, then a new thread is cloned, and after that proc connector listener gets the fork event and reads new thread's comm value from procfs stat file, but this change visibly simplifies and extends the matter. Signed-off-by: Vladimir Zapolskiy Acked-by: Evgeniy Polyakov Cc: David Miller Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- drivers/connector/cn_proc.c | 26 ++++++++++++++++++++++++++ include/linux/cn_proc.h | 11 +++++++++++ kernel/sys.c | 1 + 3 files changed, 38 insertions(+) (limited to 'kernel') diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c index e55814bc0d06..77e1e6cd66ce 100644 --- a/drivers/connector/cn_proc.c +++ b/drivers/connector/cn_proc.c @@ -205,6 +205,32 @@ void proc_ptrace_connector(struct task_struct *task, int ptrace_id) cn_netlink_send(msg, CN_IDX_PROC, GFP_KERNEL); } +void proc_comm_connector(struct task_struct *task) +{ + struct cn_msg *msg; + struct proc_event *ev; + struct timespec ts; + __u8 buffer[CN_PROC_MSG_SIZE]; + + if (atomic_read(&proc_event_num_listeners) < 1) + return; + + msg = (struct cn_msg *)buffer; + ev = (struct proc_event *)msg->data; + get_seq(&msg->seq, &ev->cpu); + ktime_get_ts(&ts); /* get high res monotonic timestamp */ + put_unaligned(timespec_to_ns(&ts), (__u64 *)&ev->timestamp_ns); + ev->what = PROC_EVENT_COMM; + ev->event_data.comm.process_pid = task->pid; + ev->event_data.comm.process_tgid = task->tgid; + get_task_comm(ev->event_data.comm.comm, task); + + memcpy(&msg->id, &cn_proc_event_id, sizeof(msg->id)); + msg->ack = 0; /* not used */ + msg->len = sizeof(*ev); + cn_netlink_send(msg, CN_IDX_PROC, GFP_KERNEL); +} + void proc_exit_connector(struct task_struct *task) { struct cn_msg *msg; diff --git a/include/linux/cn_proc.h b/include/linux/cn_proc.h index 12c517b51ca2..d03612b196e1 100644 --- a/include/linux/cn_proc.h +++ b/include/linux/cn_proc.h @@ -54,6 +54,7 @@ struct proc_event { PROC_EVENT_GID = 0x00000040, PROC_EVENT_SID = 0x00000080, PROC_EVENT_PTRACE = 0x00000100, + PROC_EVENT_COMM = 0x00000200, /* "next" should be 0x00000400 */ /* "last" is the last process event: exit */ PROC_EVENT_EXIT = 0x80000000 @@ -103,6 +104,12 @@ struct proc_event { __kernel_pid_t tracer_tgid; } ptrace; + struct comm_proc_event { + __kernel_pid_t process_pid; + __kernel_pid_t process_tgid; + char comm[16]; + } comm; + struct exit_proc_event { __kernel_pid_t process_pid; __kernel_pid_t process_tgid; @@ -118,6 +125,7 @@ void proc_exec_connector(struct task_struct *task); void proc_id_connector(struct task_struct *task, int which_id); void proc_sid_connector(struct task_struct *task); void proc_ptrace_connector(struct task_struct *task, int which_id); +void proc_comm_connector(struct task_struct *task); void proc_exit_connector(struct task_struct *task); #else static inline void proc_fork_connector(struct task_struct *task) @@ -133,6 +141,9 @@ static inline void proc_id_connector(struct task_struct *task, static inline void proc_sid_connector(struct task_struct *task) {} +static inline void proc_comm_connector(struct task_struct *task) +{} + static inline void proc_ptrace_connector(struct task_struct *task, int ptrace_id) {} diff --git a/kernel/sys.c b/kernel/sys.c index 18ee1d2f6474..b3dfb76f8073 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1759,6 +1759,7 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, sizeof(me->comm) - 1) < 0) return -EFAULT; set_task_comm(me, comm); + proc_comm_connector(me); return 0; case PR_GET_NAME: get_task_comm(comm, me); -- cgit v1.3-8-gc7d7 From 1f288094807861ec1e48c428d2c49ccf7aaf3767 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 16 Jun 2011 15:53:18 -0700 Subject: rcu: Use kthread_create_on_node() Commit a26ac2455ffc (move TREE_RCU from softirq to kthread) added per-CPU kthreads. However, kthread creation uses kthread_create(), which can put the kthread's stack and task struct on the wrong NUMA node. Therefore, use kthread_create_on_node() instead of kthread_create() so that the stacks and task structs are placed on the correct NUMA node. A similar change was carried out in commit 94dcf29a11b3 (kthread: use kthread_create_on_node()). Also change rcutorture's priority-boost-test kthread creation. Signed-off-by: Eric Dumazet CC: Tejun Heo CC: Rusty Russell CC: Andrew Morton CC: Andi Kleen CC: Ingo Molnar Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 5 +++-- kernel/rcutree_plugin.h | 5 ++++- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 98f51b13bb7e..1d2415046154 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -1280,8 +1280,9 @@ static int rcutorture_booster_init(int cpu) /* Don't allow time recalculation while creating a new task. */ mutex_lock(&boost_mutex); VERBOSE_PRINTK_STRING("Creating rcu_torture_boost task"); - boost_tasks[cpu] = kthread_create(rcu_torture_boost, NULL, - "rcu_torture_boost"); + boost_tasks[cpu] = kthread_create_on_node(rcu_torture_boost, NULL, + cpu_to_node(cpu), + "rcu_torture_boost"); if (IS_ERR(boost_tasks[cpu])) { retval = PTR_ERR(boost_tasks[cpu]); VERBOSE_PRINTK_STRING("rcu_torture_boost task create failed"); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 8aafbb80b8b0..7b850cdc0aee 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1560,7 +1560,10 @@ static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu) if (!rcu_scheduler_fully_active || per_cpu(rcu_cpu_kthread_task, cpu) != NULL) return 0; - t = kthread_create(rcu_cpu_kthread, (void *)(long)cpu, "rcuc%d", cpu); + t = kthread_create_on_node(rcu_cpu_kthread, + (void *)(long)cpu, + cpu_to_node(cpu), + "rcuc%d", cpu); if (IS_ERR(t)) return PTR_ERR(t); if (cpu_online(cpu)) -- cgit v1.3-8-gc7d7 From 1eb521210a8c9823038abe4ddfe8c69e713ec17d Mon Sep 17 00:00:00 2001 From: Shaohua Li Date: Thu, 16 Jun 2011 16:02:54 -0700 Subject: rcu: Avoid unnecessary self-wakeup of per-CPU kthreads There are a number of cases where the RCU can find additional work for the per-CPU kthread within the context of that per-CPU kthread. In such cases, the per-CPU kthread is already running, so attempting to wake itself up does nothing except waste CPU cycles. This commit therefore checks to see if it is in the per-CPU kthread context, omitting the wakeup in this case. Signed-off-by: Shaohua Li Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree_plugin.h | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 7b850cdc0aee..970329853dc5 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1291,11 +1291,9 @@ static void invoke_rcu_callbacks_kthread(void) local_irq_save(flags); __this_cpu_write(rcu_cpu_has_work, 1); - if (__this_cpu_read(rcu_cpu_kthread_task) == NULL) { - local_irq_restore(flags); - return; - } - wake_up_process(__this_cpu_read(rcu_cpu_kthread_task)); + if (__this_cpu_read(rcu_cpu_kthread_task) != NULL && + current != __this_cpu_read(rcu_cpu_kthread_task)) + wake_up_process(__this_cpu_read(rcu_cpu_kthread_task)); local_irq_restore(flags); } -- cgit v1.3-8-gc7d7 From b3fbab0571eb09746cc0283648165ec00efc8eb2 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 24 May 2011 08:31:09 -0700 Subject: rcu: Restore checks for blocking in RCU read-side critical sections Long ago, using TREE_RCU with PREEMPT would result in "scheduling while atomic" diagnostics if you blocked in an RCU read-side critical section. However, PREEMPT now implies TREE_PREEMPT_RCU, which defeats this diagnostic. This commit therefore adds a replacement diagnostic based on PROVE_RCU. Because rcu_lockdep_assert() and lockdep_rcu_dereference() are now being used for things that have nothing to do with rcu_dereference(), rename lockdep_rcu_dereference() to lockdep_rcu_suspicious() and add a third argument that is a string indicating what is suspicious. This third argument is passed in from a new third argument to rcu_lockdep_assert(). Update all calls to rcu_lockdep_assert() to add an informative third argument. Also, add a pair of rcu_lockdep_assert() calls from within rcu_note_context_switch(), one complaining if a context switch occurs in an RCU-bh read-side critical section and another complaining if a context switch occurs in an RCU-sched read-side critical section. These are present only if the PROVE_RCU kernel parameter is enabled. Finally, fix some checkpatch whitespace complaints in lockdep.c. Again, you must enable PROVE_RCU to see these new diagnostics. But you are enabling PROVE_RCU to check out new RCU uses in any case, aren't you? Signed-off-by: Paul E. McKenney --- include/linux/lockdep.h | 2 +- include/linux/rcupdate.h | 28 ++++++++++++---- kernel/lockdep.c | 84 +++++++++++++++++++++++++++--------------------- kernel/pid.c | 4 ++- kernel/sched.c | 2 ++ 5 files changed, 75 insertions(+), 45 deletions(-) (limited to 'kernel') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index ef820a3c378b..b6a56e37284c 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -548,7 +548,7 @@ do { \ #endif #ifdef CONFIG_PROVE_RCU -extern void lockdep_rcu_dereference(const char *file, const int line); +void lockdep_rcu_suspicious(const char *file, const int line, const char *s); #endif #endif /* __LINUX_LOCKDEP_H */ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 8f4f881a0ad8..8e7470d8b676 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -297,19 +297,31 @@ extern int rcu_my_thread_group_empty(void); /** * rcu_lockdep_assert - emit lockdep splat if specified condition not met * @c: condition to check + * @s: informative message */ -#define rcu_lockdep_assert(c) \ +#define rcu_lockdep_assert(c, s) \ do { \ static bool __warned; \ if (debug_lockdep_rcu_enabled() && !__warned && !(c)) { \ __warned = true; \ - lockdep_rcu_dereference(__FILE__, __LINE__); \ + lockdep_rcu_suspicious(__FILE__, __LINE__, s); \ } \ } while (0) +#define rcu_sleep_check() \ + do { \ + rcu_lockdep_assert(!lock_is_held(&rcu_bh_lock_map), \ + "Illegal context switch in RCU-bh" \ + " read-side critical section"); \ + rcu_lockdep_assert(!lock_is_held(&rcu_sched_lock_map), \ + "Illegal context switch in RCU-sched"\ + " read-side critical section"); \ + } while (0) + #else /* #ifdef CONFIG_PROVE_RCU */ -#define rcu_lockdep_assert(c) do { } while (0) +#define rcu_lockdep_assert(c, s) do { } while (0) +#define rcu_sleep_check() do { } while (0) #endif /* #else #ifdef CONFIG_PROVE_RCU */ @@ -338,14 +350,16 @@ extern int rcu_my_thread_group_empty(void); #define __rcu_dereference_check(p, c, space) \ ({ \ typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \ - rcu_lockdep_assert(c); \ + rcu_lockdep_assert(c, "suspicious rcu_dereference_check()" \ + " usage"); \ rcu_dereference_sparse(p, space); \ smp_read_barrier_depends(); \ ((typeof(*p) __force __kernel *)(_________p1)); \ }) #define __rcu_dereference_protected(p, c, space) \ ({ \ - rcu_lockdep_assert(c); \ + rcu_lockdep_assert(c, "suspicious rcu_dereference_protected()" \ + " usage"); \ rcu_dereference_sparse(p, space); \ ((typeof(*p) __force __kernel *)(p)); \ }) @@ -359,7 +373,9 @@ extern int rcu_my_thread_group_empty(void); #define __rcu_dereference_index_check(p, c) \ ({ \ typeof(p) _________p1 = ACCESS_ONCE(p); \ - rcu_lockdep_assert(c); \ + rcu_lockdep_assert(c, \ + "suspicious rcu_dereference_index_check()" \ + " usage"); \ smp_read_barrier_depends(); \ (_________p1); \ }) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 91d67ce3a8d5..1e48f1c3ea70 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -1129,10 +1129,11 @@ print_circular_bug_header(struct lock_list *entry, unsigned int depth, if (debug_locks_silent) return 0; - printk("\n=======================================================\n"); - printk( "[ INFO: possible circular locking dependency detected ]\n"); + printk("\n"); + printk("======================================================\n"); + printk("[ INFO: possible circular locking dependency detected ]\n"); print_kernel_version(); - printk( "-------------------------------------------------------\n"); + printk("-------------------------------------------------------\n"); printk("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); print_lock(check_src); @@ -1463,11 +1464,12 @@ print_bad_irq_dependency(struct task_struct *curr, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n======================================================\n"); - printk( "[ INFO: %s-safe -> %s-unsafe lock order detected ]\n", + printk("\n"); + printk("======================================================\n"); + printk("[ INFO: %s-safe -> %s-unsafe lock order detected ]\n", irqclass, irqclass); print_kernel_version(); - printk( "------------------------------------------------------\n"); + printk("------------------------------------------------------\n"); printk("%s/%d [HC%u[%lu]:SC%u[%lu]:HE%u:SE%u] is trying to acquire:\n", curr->comm, task_pid_nr(curr), curr->hardirq_context, hardirq_count() >> HARDIRQ_SHIFT, @@ -1692,10 +1694,11 @@ print_deadlock_bug(struct task_struct *curr, struct held_lock *prev, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n=============================================\n"); - printk( "[ INFO: possible recursive locking detected ]\n"); + printk("\n"); + printk("=============================================\n"); + printk("[ INFO: possible recursive locking detected ]\n"); print_kernel_version(); - printk( "---------------------------------------------\n"); + printk("---------------------------------------------\n"); printk("%s/%d is trying to acquire lock:\n", curr->comm, task_pid_nr(curr)); print_lock(next); @@ -2177,10 +2180,11 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n=================================\n"); - printk( "[ INFO: inconsistent lock state ]\n"); + printk("\n"); + printk("=================================\n"); + printk("[ INFO: inconsistent lock state ]\n"); print_kernel_version(); - printk( "---------------------------------\n"); + printk("---------------------------------\n"); printk("inconsistent {%s} -> {%s} usage.\n", usage_str[prev_bit], usage_str[new_bit]); @@ -2241,10 +2245,11 @@ print_irq_inversion_bug(struct task_struct *curr, if (!debug_locks_off_graph_unlock() || debug_locks_silent) return 0; - printk("\n=========================================================\n"); - printk( "[ INFO: possible irq lock inversion dependency detected ]\n"); + printk("\n"); + printk("=========================================================\n"); + printk("[ INFO: possible irq lock inversion dependency detected ]\n"); print_kernel_version(); - printk( "---------------------------------------------------------\n"); + printk("---------------------------------------------------------\n"); printk("%s/%d just changed the state of lock:\n", curr->comm, task_pid_nr(curr)); print_lock(this); @@ -3065,9 +3070,10 @@ print_unlock_inbalance_bug(struct task_struct *curr, struct lockdep_map *lock, if (debug_locks_silent) return 0; - printk("\n=====================================\n"); - printk( "[ BUG: bad unlock balance detected! ]\n"); - printk( "-------------------------------------\n"); + printk("\n"); + printk("=====================================\n"); + printk("[ BUG: bad unlock balance detected! ]\n"); + printk("-------------------------------------\n"); printk("%s/%d is trying to release lock (", curr->comm, task_pid_nr(curr)); print_lockdep_cache(lock); @@ -3478,9 +3484,10 @@ print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock, if (debug_locks_silent) return 0; - printk("\n=================================\n"); - printk( "[ BUG: bad contention detected! ]\n"); - printk( "---------------------------------\n"); + printk("\n"); + printk("=================================\n"); + printk("[ BUG: bad contention detected! ]\n"); + printk("---------------------------------\n"); printk("%s/%d is trying to contend lock (", curr->comm, task_pid_nr(curr)); print_lockdep_cache(lock); @@ -3839,9 +3846,10 @@ print_freed_lock_bug(struct task_struct *curr, const void *mem_from, if (debug_locks_silent) return; - printk("\n=========================\n"); - printk( "[ BUG: held lock freed! ]\n"); - printk( "-------------------------\n"); + printk("\n"); + printk("=========================\n"); + printk("[ BUG: held lock freed! ]\n"); + printk("-------------------------\n"); printk("%s/%d is freeing memory %p-%p, with a lock still held there!\n", curr->comm, task_pid_nr(curr), mem_from, mem_to-1); print_lock(hlock); @@ -3895,9 +3903,10 @@ static void print_held_locks_bug(struct task_struct *curr) if (debug_locks_silent) return; - printk("\n=====================================\n"); - printk( "[ BUG: lock held at task exit time! ]\n"); - printk( "-------------------------------------\n"); + printk("\n"); + printk("=====================================\n"); + printk("[ BUG: lock held at task exit time! ]\n"); + printk("-------------------------------------\n"); printk("%s/%d is exiting with locks still held!\n", curr->comm, task_pid_nr(curr)); lockdep_print_held_locks(curr); @@ -3991,16 +4000,17 @@ void lockdep_sys_exit(void) if (unlikely(curr->lockdep_depth)) { if (!debug_locks_off()) return; - printk("\n================================================\n"); - printk( "[ BUG: lock held when returning to user space! ]\n"); - printk( "------------------------------------------------\n"); + printk("\n"); + printk("================================================\n"); + printk("[ BUG: lock held when returning to user space! ]\n"); + printk("------------------------------------------------\n"); printk("%s/%d is leaving the kernel with locks still held!\n", curr->comm, curr->pid); lockdep_print_held_locks(curr); } } -void lockdep_rcu_dereference(const char *file, const int line) +void lockdep_rcu_suspicious(const char *file, const int line, const char *s) { struct task_struct *curr = current; @@ -4009,15 +4019,15 @@ void lockdep_rcu_dereference(const char *file, const int line) return; #endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */ /* Note: the following can be executed concurrently, so be careful. */ - printk("\n===================================================\n"); - printk( "[ INFO: suspicious rcu_dereference_check() usage. ]\n"); - printk( "---------------------------------------------------\n"); - printk("%s:%d invoked rcu_dereference_check() without protection!\n", - file, line); + printk("\n"); + printk("===============================\n"); + printk("[ INFO: suspicious RCU usage. ]\n"); + printk("-------------------------------\n"); + printk("%s:%d %s!\n", file, line, s); printk("\nother info that might help us debug this:\n\n"); printk("\nrcu_scheduler_active = %d, debug_locks = %d\n", rcu_scheduler_active, debug_locks); lockdep_print_held_locks(curr); printk("\nstack backtrace:\n"); dump_stack(); } -EXPORT_SYMBOL_GPL(lockdep_rcu_dereference); +EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious); diff --git a/kernel/pid.c b/kernel/pid.c index e432057f3b21..8cafe7e72ad2 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -418,7 +418,9 @@ EXPORT_SYMBOL(pid_task); */ struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns) { - rcu_lockdep_assert(rcu_read_lock_held()); + rcu_lockdep_assert(rcu_read_lock_held(), + "find_task_by_pid_ns() needs rcu_read_lock()" + " protection"); return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID); } diff --git a/kernel/sched.c b/kernel/sched.c index ec5f472bc5b9..e24cebe0e6cb 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4237,6 +4237,7 @@ static inline void schedule_debug(struct task_struct *prev) */ if (unlikely(in_atomic_preempt_off() && !prev->exit_state)) __schedule_bug(prev); + rcu_sleep_check(); profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -8230,6 +8231,7 @@ void __might_sleep(const char *file, int line, int preempt_offset) { static unsigned long prev_jiffy; /* ratelimiting */ + rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */ if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || system_state != SYSTEM_RUNNING || oops_in_progress) return; -- cgit v1.3-8-gc7d7 From f039d1f1884b2fe9c13d28f59d8330f0b0518fc4 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 7 Jun 2011 15:26:30 -0700 Subject: rcu: Fix mismatched variable in rcutree_trace.c rcutree.c defines rcu_cpu_kthread_cpu as int, not unsigned int, so the extern has to follow that. Signed-off-by: Andi Kleen Signed-off-by: Paul E. McKenney --- kernel/rcutree_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index 3b0c0986afc0..0a5a8bedc522 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -49,7 +49,7 @@ #ifdef CONFIG_RCU_BOOST DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status); -DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_cpu); +DECLARE_PER_CPU(int, rcu_cpu_kthread_cpu); DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_loops); DECLARE_PER_CPU(char, rcu_cpu_has_work); -- cgit v1.3-8-gc7d7 From 2c42818e962e2858334bf45bfc56662b3752df34 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 26 May 2011 22:14:36 -0700 Subject: rcu: Abstract common code for RCU grace-period-wait primitives Pull the code that waits for an RCU grace period into a single function, which is then called by synchronize_rcu() and friends in the case of TREE_RCU and TREE_PREEMPT_RCU, and from rcu_barrier() and friends in the case of TINY_RCU and TINY_PREEMPT_RCU. Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 130 ++++++++++++++++++++++++++--------------------- include/linux/rcutiny.h | 16 +++++- include/linux/rcutree.h | 2 + kernel/rcupdate.c | 21 +++++++- kernel/rcutiny.c | 28 ---------- kernel/rcutiny_plugin.h | 14 ----- kernel/rcutree.c | 22 +------- kernel/rcutree_plugin.h | 11 +--- 8 files changed, 113 insertions(+), 131 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 87bd390df73f..ae5327de41aa 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -66,11 +66,73 @@ static inline void rcutorture_record_progress(unsigned long vernum) #define ULONG_CMP_LT(a, b) (ULONG_MAX / 2 < (a) - (b)) /* Exported common interfaces */ + +#ifdef CONFIG_PREEMPT_RCU + +/** + * call_rcu() - Queue an RCU callback for invocation after a grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all pre-existing RCU read-side + * critical sections have completed. However, the callback function + * might well execute concurrently with RCU read-side critical sections + * that started after call_rcu() was invoked. RCU read-side critical + * sections are delimited by rcu_read_lock() and rcu_read_unlock(), + * and may be nested. + */ +extern void call_rcu(struct rcu_head *head, + void (*func)(struct rcu_head *head)); + +#else /* #ifdef CONFIG_PREEMPT_RCU */ + +/* In classic RCU, call_rcu() is just call_rcu_sched(). */ +#define call_rcu call_rcu_sched + +#endif /* #else #ifdef CONFIG_PREEMPT_RCU */ + +/** + * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_bh() assumes + * that the read-side critical sections end on completion of a softirq + * handler. This means that read-side critical sections in process + * context must not be interrupted by softirqs. This interface is to be + * used when most of the read-side critical sections are in softirq context. + * RCU read-side critical sections are delimited by : + * - rcu_read_lock() and rcu_read_unlock(), if in interrupt context. + * OR + * - rcu_read_lock_bh() and rcu_read_unlock_bh(), if in process context. + * These may be nested. + */ +extern void call_rcu_bh(struct rcu_head *head, + void (*func)(struct rcu_head *head)); + +/** + * call_rcu_sched() - Queue an RCU for invocation after sched grace period. + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_sched() assumes + * that the read-side critical sections end on enabling of preemption + * or on voluntary preemption. + * RCU read-side critical sections are delimited by : + * - rcu_read_lock_sched() and rcu_read_unlock_sched(), + * OR + * anything that disables preemption. + * These may be nested. + */ extern void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); + extern void synchronize_sched(void); -extern void rcu_barrier_bh(void); -extern void rcu_barrier_sched(void); static inline void __rcu_read_lock_bh(void) { @@ -143,6 +205,15 @@ static inline void rcu_exit_nohz(void) #endif /* #else #ifdef CONFIG_NO_HZ */ +/* + * Infrastructure to implement the synchronize_() primitives in + * TREE_RCU and rcu_barrier_() primitives in TINY_RCU. + */ + +typedef void call_rcu_func_t(struct rcu_head *head, + void (*func)(struct rcu_head *head)); +void wait_rcu_gp(call_rcu_func_t crf); + #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU) #include #elif defined(CONFIG_TINY_RCU) || defined(CONFIG_TINY_PREEMPT_RCU) @@ -723,61 +794,6 @@ static inline notrace void rcu_read_unlock_sched_notrace(void) #define RCU_INIT_POINTER(p, v) \ p = (typeof(*v) __force __rcu *)(v) -/* Infrastructure to implement the synchronize_() primitives. */ - -struct rcu_synchronize { - struct rcu_head head; - struct completion completion; -}; - -extern void wakeme_after_rcu(struct rcu_head *head); - -#ifdef CONFIG_PREEMPT_RCU - -/** - * call_rcu() - Queue an RCU callback for invocation after a grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual callback function to be invoked after the grace period - * - * The callback function will be invoked some time after a full grace - * period elapses, in other words after all pre-existing RCU read-side - * critical sections have completed. However, the callback function - * might well execute concurrently with RCU read-side critical sections - * that started after call_rcu() was invoked. RCU read-side critical - * sections are delimited by rcu_read_lock() and rcu_read_unlock(), - * and may be nested. - */ -extern void call_rcu(struct rcu_head *head, - void (*func)(struct rcu_head *head)); - -#else /* #ifdef CONFIG_PREEMPT_RCU */ - -/* In classic RCU, call_rcu() is just call_rcu_sched(). */ -#define call_rcu call_rcu_sched - -#endif /* #else #ifdef CONFIG_PREEMPT_RCU */ - -/** - * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period. - * @head: structure to be used for queueing the RCU updates. - * @func: actual callback function to be invoked after the grace period - * - * The callback function will be invoked some time after a full grace - * period elapses, in other words after all currently executing RCU - * read-side critical sections have completed. call_rcu_bh() assumes - * that the read-side critical sections end on completion of a softirq - * handler. This means that read-side critical sections in process - * context must not be interrupted by softirqs. This interface is to be - * used when most of the read-side critical sections are in softirq context. - * RCU read-side critical sections are delimited by : - * - rcu_read_lock() and rcu_read_unlock(), if in interrupt context. - * OR - * - rcu_read_lock_bh() and rcu_read_unlock_bh(), if in process context. - * These may be nested. - */ -extern void call_rcu_bh(struct rcu_head *head, - void (*func)(struct rcu_head *head)); - /* * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally * by call_rcu() and rcu callback execution, and are therefore not part of the diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index 52b3e0281fd0..4eab233a00cd 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -31,6 +31,16 @@ static inline void rcu_init(void) { } +static inline void rcu_barrier_bh(void) +{ + wait_rcu_gp(call_rcu_bh); +} + +static inline void rcu_barrier_sched(void) +{ + wait_rcu_gp(call_rcu_sched); +} + #ifdef CONFIG_TINY_RCU static inline void synchronize_rcu_expedited(void) @@ -45,9 +55,13 @@ static inline void rcu_barrier(void) #else /* #ifdef CONFIG_TINY_RCU */ -void rcu_barrier(void); void synchronize_rcu_expedited(void); +static inline void rcu_barrier(void) +{ + wait_rcu_gp(call_rcu); +} + #endif /* #else #ifdef CONFIG_TINY_RCU */ static inline void synchronize_rcu_bh(void) diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index e65d06634dd8..67458468f1a8 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -67,6 +67,8 @@ static inline void synchronize_rcu_bh_expedited(void) } extern void rcu_barrier(void); +extern void rcu_barrier_bh(void); +extern void rcu_barrier_sched(void); extern unsigned long rcutorture_testseq; extern unsigned long rcutorture_vernum; diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index ddddb320be61..09b3b1b54e02 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -94,11 +94,16 @@ EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held); #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ +struct rcu_synchronize { + struct rcu_head head; + struct completion completion; +}; + /* * Awaken the corresponding synchronize_rcu() instance now that a * grace period has elapsed. */ -void wakeme_after_rcu(struct rcu_head *head) +static void wakeme_after_rcu(struct rcu_head *head) { struct rcu_synchronize *rcu; @@ -106,6 +111,20 @@ void wakeme_after_rcu(struct rcu_head *head) complete(&rcu->completion); } +void wait_rcu_gp(call_rcu_func_t crf) +{ + struct rcu_synchronize rcu; + + init_rcu_head_on_stack(&rcu.head); + init_completion(&rcu.completion); + /* Will wake me after RCU finished. */ + crf(&rcu.head, wakeme_after_rcu); + /* Wait for it. */ + wait_for_completion(&rcu.completion); + destroy_rcu_head_on_stack(&rcu.head); +} +EXPORT_SYMBOL_GPL(wait_rcu_gp); + #ifdef CONFIG_PROVE_RCU /* * wrapper function to avoid #include problems. diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 7bbac7d0f5ab..f544e343256a 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -281,34 +281,6 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) } EXPORT_SYMBOL_GPL(call_rcu_bh); -void rcu_barrier_bh(void) -{ - struct rcu_synchronize rcu; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu_bh(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); -} -EXPORT_SYMBOL_GPL(rcu_barrier_bh); - -void rcu_barrier_sched(void) -{ - struct rcu_synchronize rcu; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu_sched(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); -} -EXPORT_SYMBOL_GPL(rcu_barrier_sched); - /* * Spawn the kthread that invokes RCU callbacks. */ diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index f259c676195f..6b0cedb383e0 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -697,20 +697,6 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) } EXPORT_SYMBOL_GPL(call_rcu); -void rcu_barrier(void) -{ - struct rcu_synchronize rcu; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); -} -EXPORT_SYMBOL_GPL(rcu_barrier); - /* * synchronize_rcu - wait until a grace period has elapsed. * diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ba06207b1dd3..a7c6bce1af83 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1613,18 +1613,9 @@ EXPORT_SYMBOL_GPL(call_rcu_bh); */ void synchronize_sched(void) { - struct rcu_synchronize rcu; - if (rcu_blocking_is_gp()) return; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu_sched(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); + wait_rcu_gp(call_rcu_sched); } EXPORT_SYMBOL_GPL(synchronize_sched); @@ -1639,18 +1630,9 @@ EXPORT_SYMBOL_GPL(synchronize_sched); */ void synchronize_rcu_bh(void) { - struct rcu_synchronize rcu; - if (rcu_blocking_is_gp()) return; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu_bh(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); + wait_rcu_gp(call_rcu_bh); } EXPORT_SYMBOL_GPL(synchronize_rcu_bh); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 970329853dc5..43daa46bc6f2 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -656,18 +656,9 @@ EXPORT_SYMBOL_GPL(call_rcu); */ void synchronize_rcu(void) { - struct rcu_synchronize rcu; - if (!rcu_scheduler_active) return; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - /* Will wake me after RCU finished. */ - call_rcu(&rcu.head, wakeme_after_rcu); - /* Wait for it. */ - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); + wait_rcu_gp(call_rcu); } EXPORT_SYMBOL_GPL(synchronize_rcu); -- cgit v1.3-8-gc7d7 From bdf2a4364904d6cf2f59b16b0bd86fdc5a2c6152 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 7 Jun 2011 16:59:35 -0700 Subject: rcu: Catch rcutorture up to new RCU API additions Now that the RCU API contains synchronize_rcu_bh(), synchronize_sched(), call_rcu_sched(), and rcu_bh_expedited()... Make rcutorture test synchronize_rcu_bh(), getting rid of the old rcu_bh_torture_synchronize() workaround. Similarly, make rcutorture test synchronize_sched(), getting rid of the old sched_torture_synchronize() workaround. Make rcutorture test call_rcu_sched() instead of wrappering synchronize_sched(). Also add testing of rcu_bh_expedited(). Signed-off-by: Paul E. McKenney --- Documentation/RCU/torture.txt | 3 +++ kernel/rcutorture.c | 55 +++++++++++++++++-------------------------- 2 files changed, 24 insertions(+), 34 deletions(-) (limited to 'kernel') diff --git a/Documentation/RCU/torture.txt b/Documentation/RCU/torture.txt index 4205ed1f8adc..783d6c134d3f 100644 --- a/Documentation/RCU/torture.txt +++ b/Documentation/RCU/torture.txt @@ -122,6 +122,9 @@ torture_type The type of RCU to test, with string values as follows: "rcu_bh_sync": rcu_read_lock_bh(), rcu_read_unlock_bh(), and synchronize_rcu_bh(). + "rcu_bh_expedited": rcu_read_lock_bh(), rcu_read_unlock_bh(), + and synchronize_rcu_bh_expedited(). + "srcu": srcu_read_lock(), srcu_read_unlock() and synchronize_srcu(). diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 1d2415046154..75fca518888c 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -480,30 +480,6 @@ static void rcu_bh_torture_deferred_free(struct rcu_torture *p) call_rcu_bh(&p->rtort_rcu, rcu_torture_cb); } -struct rcu_bh_torture_synchronize { - struct rcu_head head; - struct completion completion; -}; - -static void rcu_bh_torture_wakeme_after_cb(struct rcu_head *head) -{ - struct rcu_bh_torture_synchronize *rcu; - - rcu = container_of(head, struct rcu_bh_torture_synchronize, head); - complete(&rcu->completion); -} - -static void rcu_bh_torture_synchronize(void) -{ - struct rcu_bh_torture_synchronize rcu; - - init_rcu_head_on_stack(&rcu.head); - init_completion(&rcu.completion); - call_rcu_bh(&rcu.head, rcu_bh_torture_wakeme_after_cb); - wait_for_completion(&rcu.completion); - destroy_rcu_head_on_stack(&rcu.head); -} - static struct rcu_torture_ops rcu_bh_ops = { .init = NULL, .cleanup = NULL, @@ -512,7 +488,7 @@ static struct rcu_torture_ops rcu_bh_ops = { .readunlock = rcu_bh_torture_read_unlock, .completed = rcu_bh_torture_completed, .deferred_free = rcu_bh_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, + .sync = synchronize_rcu_bh, .cb_barrier = rcu_barrier_bh, .fqs = rcu_bh_force_quiescent_state, .stats = NULL, @@ -528,7 +504,7 @@ static struct rcu_torture_ops rcu_bh_sync_ops = { .readunlock = rcu_bh_torture_read_unlock, .completed = rcu_bh_torture_completed, .deferred_free = rcu_sync_torture_deferred_free, - .sync = rcu_bh_torture_synchronize, + .sync = synchronize_rcu_bh, .cb_barrier = NULL, .fqs = rcu_bh_force_quiescent_state, .stats = NULL, @@ -536,6 +512,22 @@ static struct rcu_torture_ops rcu_bh_sync_ops = { .name = "rcu_bh_sync" }; +static struct rcu_torture_ops rcu_bh_expedited_ops = { + .init = rcu_sync_torture_init, + .cleanup = NULL, + .readlock = rcu_bh_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = rcu_bh_torture_read_unlock, + .completed = rcu_bh_torture_completed, + .deferred_free = rcu_sync_torture_deferred_free, + .sync = synchronize_rcu_bh_expedited, + .cb_barrier = NULL, + .fqs = rcu_bh_force_quiescent_state, + .stats = NULL, + .irq_capable = 1, + .name = "rcu_bh_expedited" +}; + /* * Definitions for srcu torture testing. */ @@ -659,11 +651,6 @@ static void rcu_sched_torture_deferred_free(struct rcu_torture *p) call_rcu_sched(&p->rtort_rcu, rcu_torture_cb); } -static void sched_torture_synchronize(void) -{ - synchronize_sched(); -} - static struct rcu_torture_ops sched_ops = { .init = rcu_sync_torture_init, .cleanup = NULL, @@ -672,7 +659,7 @@ static struct rcu_torture_ops sched_ops = { .readunlock = sched_torture_read_unlock, .completed = rcu_no_completed, .deferred_free = rcu_sched_torture_deferred_free, - .sync = sched_torture_synchronize, + .sync = synchronize_sched, .cb_barrier = rcu_barrier_sched, .fqs = rcu_sched_force_quiescent_state, .stats = NULL, @@ -688,7 +675,7 @@ static struct rcu_torture_ops sched_sync_ops = { .readunlock = sched_torture_read_unlock, .completed = rcu_no_completed, .deferred_free = rcu_sync_torture_deferred_free, - .sync = sched_torture_synchronize, + .sync = synchronize_sched, .cb_barrier = NULL, .fqs = rcu_sched_force_quiescent_state, .stats = NULL, @@ -1425,7 +1412,7 @@ rcu_torture_init(void) int firsterr = 0; static struct rcu_torture_ops *torture_ops[] = { &rcu_ops, &rcu_sync_ops, &rcu_expedited_ops, - &rcu_bh_ops, &rcu_bh_sync_ops, + &rcu_bh_ops, &rcu_bh_sync_ops, &rcu_bh_expedited_ops, &srcu_ops, &srcu_expedited_ops, &sched_ops, &sched_sync_ops, &sched_expedited_ops, }; -- cgit v1.3-8-gc7d7 From 9d68197c05201d8edc70d58bd1d5dad05d8455e8 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 01:48:03 -0700 Subject: rcu: Don't destroy rcu_torture_boost() callback until it is done The rcu_torture_boost() cleanup code destroyed debug-objects state before waiting for the last RCU callback to be invoked, resulting in rare but very real debug-objects warnings. Move the destruction to after the waiting to fix this problem. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 75fca518888c..6a30ea3f1d82 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -796,11 +796,11 @@ checkwait: rcu_stutter_wait("rcu_torture_boost"); /* Clean up and exit. */ VERBOSE_PRINTK_STRING("rcu_torture_boost task stopping"); - destroy_rcu_head_on_stack(&rbi.rcu); rcutorture_shutdown_absorb("rcu_torture_boost"); while (!kthread_should_stop() || rbi.inflight) schedule_timeout_uninterruptible(1); smp_mb(); /* order accesses to ->inflight before stack-frame death. */ + destroy_rcu_head_on_stack(&rbi.rcu); return 0; } -- cgit v1.3-8-gc7d7 From 29c00b4a1d9e277786120032aa8364631820d863 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Jun 2011 15:53:19 -0700 Subject: rcu: Add event-tracing for RCU callback invocation There was recently some controversy about the overhead of invoking RCU callbacks. Add TRACE_EVENT()s to obtain fine-grained timings for the start and stop of a batch of callbacks and also for each callback invoked. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 50 ----------------------- include/trace/events/rcu.h | 98 ++++++++++++++++++++++++++++++++++++++++++++++ kernel/rcu.h | 79 +++++++++++++++++++++++++++++++++++++ kernel/rcupdate.c | 5 +++ kernel/rcutiny.c | 26 +++++++++++- kernel/rcutree.c | 15 +++++-- 6 files changed, 219 insertions(+), 54 deletions(-) create mode 100644 include/trace/events/rcu.h create mode 100644 kernel/rcu.h (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index ae5327de41aa..dd2bc2c6a285 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -794,44 +794,6 @@ static inline notrace void rcu_read_unlock_sched_notrace(void) #define RCU_INIT_POINTER(p, v) \ p = (typeof(*v) __force __rcu *)(v) -/* - * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally - * by call_rcu() and rcu callback execution, and are therefore not part of the - * RCU API. Leaving in rcupdate.h because they are used by all RCU flavors. - */ - -#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD -# define STATE_RCU_HEAD_READY 0 -# define STATE_RCU_HEAD_QUEUED 1 - -extern struct debug_obj_descr rcuhead_debug_descr; - -static inline void debug_rcu_head_queue(struct rcu_head *head) -{ - WARN_ON_ONCE((unsigned long)head & 0x3); - debug_object_activate(head, &rcuhead_debug_descr); - debug_object_active_state(head, &rcuhead_debug_descr, - STATE_RCU_HEAD_READY, - STATE_RCU_HEAD_QUEUED); -} - -static inline void debug_rcu_head_unqueue(struct rcu_head *head) -{ - debug_object_active_state(head, &rcuhead_debug_descr, - STATE_RCU_HEAD_QUEUED, - STATE_RCU_HEAD_READY); - debug_object_deactivate(head, &rcuhead_debug_descr); -} -#else /* !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ -static inline void debug_rcu_head_queue(struct rcu_head *head) -{ -} - -static inline void debug_rcu_head_unqueue(struct rcu_head *head) -{ -} -#endif /* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ - static __always_inline bool __is_kfree_rcu_offset(unsigned long offset) { return offset < 4096; @@ -850,18 +812,6 @@ void __kfree_rcu(struct rcu_head *head, unsigned long offset) call_rcu(head, (rcu_callback)offset); } -extern void kfree(const void *); - -static inline void __rcu_reclaim(struct rcu_head *head) -{ - unsigned long offset = (unsigned long)head->func; - - if (__is_kfree_rcu_offset(offset)) - kfree((void *)head - offset); - else - head->func(head); -} - /** * kfree_rcu() - kfree an object after a grace period. * @ptr: pointer to kfree diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h new file mode 100644 index 000000000000..db3f6e9e63e6 --- /dev/null +++ b/include/trace/events/rcu.h @@ -0,0 +1,98 @@ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM rcu + +#if !defined(_TRACE_RCU_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_RCU_H + +#include + +/* + * Tracepoint for calling rcu_do_batch, performed to start callback invocation: + */ +TRACE_EVENT(rcu_batch_start, + + TP_PROTO(long callbacks_ready, int blimit), + + TP_ARGS(callbacks_ready, blimit), + + TP_STRUCT__entry( + __field( long, callbacks_ready ) + __field( int, blimit ) + ), + + TP_fast_assign( + __entry->callbacks_ready = callbacks_ready; + __entry->blimit = blimit; + ), + + TP_printk("CBs=%ld bl=%d", __entry->callbacks_ready, __entry->blimit) +); + +/* + * Tracepoint for the invocation of a single RCU callback + */ +TRACE_EVENT(rcu_invoke_callback, + + TP_PROTO(struct rcu_head *rhp), + + TP_ARGS(rhp), + + TP_STRUCT__entry( + __field( void *, rhp ) + __field( void *, func ) + ), + + TP_fast_assign( + __entry->rhp = rhp; + __entry->func = rhp->func; + ), + + TP_printk("rhp=%p func=%pf", __entry->rhp, __entry->func) +); + +/* + * Tracepoint for the invocation of a single RCU kfree callback + */ +TRACE_EVENT(rcu_invoke_kfree_callback, + + TP_PROTO(struct rcu_head *rhp, unsigned long offset), + + TP_ARGS(rhp, offset), + + TP_STRUCT__entry( + __field(void *, rhp ) + __field(unsigned long, offset ) + ), + + TP_fast_assign( + __entry->rhp = rhp; + __entry->offset = offset; + ), + + TP_printk("rhp=%p func=%ld", __entry->rhp, __entry->offset) +); + +/* + * Tracepoint for leaving rcu_do_batch, performed after callback invocation: + */ +TRACE_EVENT(rcu_batch_end, + + TP_PROTO(int callbacks_invoked), + + TP_ARGS(callbacks_invoked), + + TP_STRUCT__entry( + __field( int, callbacks_invoked ) + ), + + TP_fast_assign( + __entry->callbacks_invoked = callbacks_invoked; + ), + + TP_printk("CBs-invoked=%d", __entry->callbacks_invoked) +); + +#endif /* _TRACE_RCU_H */ + +/* This part must be outside protection */ +#include diff --git a/kernel/rcu.h b/kernel/rcu.h new file mode 100644 index 000000000000..7bc16436aba0 --- /dev/null +++ b/kernel/rcu.h @@ -0,0 +1,79 @@ +/* + * Read-Copy Update definitions shared among RCU implementations. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright IBM Corporation, 2011 + * + * Author: Paul E. McKenney + */ + +#ifndef __LINUX_RCU_H +#define __LINUX_RCU_H + +/* + * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally + * by call_rcu() and rcu callback execution, and are therefore not part of the + * RCU API. Leaving in rcupdate.h because they are used by all RCU flavors. + */ + +#ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD +# define STATE_RCU_HEAD_READY 0 +# define STATE_RCU_HEAD_QUEUED 1 + +extern struct debug_obj_descr rcuhead_debug_descr; + +static inline void debug_rcu_head_queue(struct rcu_head *head) +{ + WARN_ON_ONCE((unsigned long)head & 0x3); + debug_object_activate(head, &rcuhead_debug_descr); + debug_object_active_state(head, &rcuhead_debug_descr, + STATE_RCU_HEAD_READY, + STATE_RCU_HEAD_QUEUED); +} + +static inline void debug_rcu_head_unqueue(struct rcu_head *head) +{ + debug_object_active_state(head, &rcuhead_debug_descr, + STATE_RCU_HEAD_QUEUED, + STATE_RCU_HEAD_READY); + debug_object_deactivate(head, &rcuhead_debug_descr); +} +#else /* !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ +static inline void debug_rcu_head_queue(struct rcu_head *head) +{ +} + +static inline void debug_rcu_head_unqueue(struct rcu_head *head) +{ +} +#endif /* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ + +extern void kfree(const void *); + +static inline void __rcu_reclaim(struct rcu_head *head) +{ + unsigned long offset = (unsigned long)head->func; + + if (__is_kfree_rcu_offset(offset)) { + trace_rcu_invoke_kfree_callback(head, offset); + kfree((void *)head - offset); + } else { + trace_rcu_invoke_callback(head); + head->func(head); + } +} + +#endif /* __LINUX_RCU_H */ diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 09b3b1b54e02..ca0d23b6b3e8 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -46,6 +46,11 @@ #include #include +#define CREATE_TRACE_POINTS +#include + +#include "rcu.h" + #ifdef CONFIG_DEBUG_LOCK_ALLOC static struct lock_class_key rcu_lock_key; struct lockdep_map rcu_lock_map = diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index f544e343256a..19453ba1392e 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -37,6 +37,25 @@ #include #include +#ifdef CONFIG_RCU_TRACE + +#include + +#else /* #ifdef CONFIG_RCU_TRACE */ + +/* No by-default tracing in TINY_RCU: Keep TINY_RCU tiny! */ +static void trace_rcu_invoke_kfree_callback(struct rcu_head *rhp, + unsigned long offset) +{ +} +static void trace_rcu_invoke_callback(struct rcu_head *head) +{ +} + +#endif /* #else #ifdef CONFIG_RCU_TRACE */ + +#include "rcu.h" + /* Controls for rcu_kthread() kthread, replacing RCU_SOFTIRQ used previously. */ static struct task_struct *rcu_kthread_task; static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq); @@ -161,11 +180,15 @@ static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) RCU_TRACE(int cb_count = 0); /* If no RCU callbacks ready to invoke, just return. */ - if (&rcp->rcucblist == rcp->donetail) + if (&rcp->rcucblist == rcp->donetail) { + RCU_TRACE(trace_rcu_batch_start(0, -1)); + RCU_TRACE(trace_rcu_batch_end(0)); return; + } /* Move the ready-to-invoke callbacks to a local list. */ local_irq_save(flags); + RCU_TRACE(trace_rcu_batch_start(0, -1)); list = rcp->rcucblist; rcp->rcucblist = *rcp->donetail; *rcp->donetail = NULL; @@ -187,6 +210,7 @@ static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) RCU_TRACE(cb_count++); } RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count)); + RCU_TRACE(trace_rcu_batch_end(cb_count)); } /* diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a7c6bce1af83..45dcc2036a1e 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -52,6 +52,9 @@ #include #include "rcutree.h" +#include + +#include "rcu.h" /* Data structures. */ @@ -1190,17 +1193,22 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; struct rcu_head *next, *list, **tail; - int count; + int bl, count; /* If no callbacks are ready, just return.*/ - if (!cpu_has_callbacks_ready_to_invoke(rdp)) + if (!cpu_has_callbacks_ready_to_invoke(rdp)) { + trace_rcu_batch_start(0, 0); + trace_rcu_batch_end(0); return; + } /* * Extract the list of ready callbacks, disabling to prevent * races with call_rcu() from interrupt handlers. */ local_irq_save(flags); + bl = rdp->blimit; + trace_rcu_batch_start(rdp->qlen, bl); list = rdp->nxtlist; rdp->nxtlist = *rdp->nxttail[RCU_DONE_TAIL]; *rdp->nxttail[RCU_DONE_TAIL] = NULL; @@ -1218,11 +1226,12 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) debug_rcu_head_unqueue(list); __rcu_reclaim(list); list = next; - if (++count >= rdp->blimit) + if (++count >= bl) break; } local_irq_save(flags); + trace_rcu_batch_end(count); /* Update count, and requeue any remaining callbacks. */ rdp->qlen -= count; -- cgit v1.3-8-gc7d7 From 300df91ca9358f7f09298eec9503c12b32054ef7 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Jun 2011 22:26:31 -0700 Subject: rcu: Event-trace markers for computing RCU CPU utilization This commit adds the trace_rcu_utilization() marker that is to be used to allow postprocessing scripts compute RCU's CPU utilization, give or take event-trace overhead. Note that we do not include RCU's dyntick-idle interface because event tracing requires RCU protection, which is not available in dyntick-idle mode. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/trace/events/rcu.h | 73 +++++++++++++++++++++++++++++++++------------- kernel/rcutree.c | 16 +++++++++- 2 files changed, 68 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index db3f6e9e63e6..ab458eb689fb 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -7,29 +7,58 @@ #include /* - * Tracepoint for calling rcu_do_batch, performed to start callback invocation: + * Tracepoint for start/end markers used for utilization calculations. + * By convention, the string is of the following forms: + * + * "Start " -- Mark the start of the specified activity, + * such as "context switch". Nesting is permitted. + * "End " -- Mark the end of the specified activity. + */ +TRACE_EVENT(rcu_utilization, + + TP_PROTO(char *s), + + TP_ARGS(s), + + TP_STRUCT__entry( + __field(char *, s) + ), + + TP_fast_assign( + __entry->s = s; + ), + + TP_printk("%s", __entry->s) +); + +/* + * Tracepoint for marking the beginning rcu_do_batch, performed to start + * RCU callback invocation. The first argument is the total number of + * callbacks (including those that are not yet ready to be invoked), + * and the second argument is the current RCU-callback batch limit. */ TRACE_EVENT(rcu_batch_start, - TP_PROTO(long callbacks_ready, int blimit), + TP_PROTO(long qlen, int blimit), - TP_ARGS(callbacks_ready, blimit), + TP_ARGS(qlen, blimit), TP_STRUCT__entry( - __field( long, callbacks_ready ) - __field( int, blimit ) + __field(long, qlen) + __field(int, blimit) ), TP_fast_assign( - __entry->callbacks_ready = callbacks_ready; - __entry->blimit = blimit; + __entry->qlen = qlen; + __entry->blimit = blimit; ), - TP_printk("CBs=%ld bl=%d", __entry->callbacks_ready, __entry->blimit) + TP_printk("CBs=%ld bl=%d", __entry->qlen, __entry->blimit) ); /* - * Tracepoint for the invocation of a single RCU callback + * Tracepoint for the invocation of a single RCU callback function. + * The argument is a pointer to the RCU callback itself. */ TRACE_EVENT(rcu_invoke_callback, @@ -38,20 +67,23 @@ TRACE_EVENT(rcu_invoke_callback, TP_ARGS(rhp), TP_STRUCT__entry( - __field( void *, rhp ) - __field( void *, func ) + __field(void *, rhp) + __field(void *, func) ), TP_fast_assign( - __entry->rhp = rhp; - __entry->func = rhp->func; + __entry->rhp = rhp; + __entry->func = rhp->func; ), TP_printk("rhp=%p func=%pf", __entry->rhp, __entry->func) ); /* - * Tracepoint for the invocation of a single RCU kfree callback + * Tracepoint for the invocation of a single RCU callback of the special + * kfree() form. The first argument is a pointer to the RCU callback + * and the second argument is the offset of the callback within the + * enclosing RCU-protected data structure. */ TRACE_EVENT(rcu_invoke_kfree_callback, @@ -60,12 +92,12 @@ TRACE_EVENT(rcu_invoke_kfree_callback, TP_ARGS(rhp, offset), TP_STRUCT__entry( - __field(void *, rhp ) - __field(unsigned long, offset ) + __field(void *, rhp) + __field(unsigned long, offset) ), TP_fast_assign( - __entry->rhp = rhp; + __entry->rhp = rhp; __entry->offset = offset; ), @@ -73,7 +105,8 @@ TRACE_EVENT(rcu_invoke_kfree_callback, ); /* - * Tracepoint for leaving rcu_do_batch, performed after callback invocation: + * Tracepoint for exiting rcu_do_batch after RCU callbacks have been + * invoked. The first argument is the number of callbacks actually invoked. */ TRACE_EVENT(rcu_batch_end, @@ -82,11 +115,11 @@ TRACE_EVENT(rcu_batch_end, TP_ARGS(callbacks_invoked), TP_STRUCT__entry( - __field( int, callbacks_invoked ) + __field(int, callbacks_invoked) ), TP_fast_assign( - __entry->callbacks_invoked = callbacks_invoked; + __entry->callbacks_invoked = callbacks_invoked; ), TP_printk("CBs-invoked=%d", __entry->callbacks_invoked) diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 45dcc2036a1e..2a9643bd6ae9 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -184,8 +184,10 @@ void rcu_bh_qs(int cpu) */ void rcu_note_context_switch(int cpu) { + trace_rcu_utilization("Start context switch"); rcu_sched_qs(cpu); rcu_preempt_note_context_switch(cpu); + trace_rcu_utilization("End context switch"); } EXPORT_SYMBOL_GPL(rcu_note_context_switch); @@ -1275,6 +1277,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) */ void rcu_check_callbacks(int cpu, int user) { + trace_rcu_utilization("Start scheduler-tick"); if (user || (idle_cpu(cpu) && rcu_scheduler_active && !in_softirq() && hardirq_count() <= (1 << HARDIRQ_SHIFT))) { @@ -1308,6 +1311,7 @@ void rcu_check_callbacks(int cpu, int user) rcu_preempt_check_callbacks(cpu); if (rcu_pending(cpu)) invoke_rcu_core(); + trace_rcu_utilization("End scheduler-tick"); } #ifdef CONFIG_SMP @@ -1369,10 +1373,14 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) unsigned long flags; struct rcu_node *rnp = rcu_get_root(rsp); - if (!rcu_gp_in_progress(rsp)) + trace_rcu_utilization("Start fqs"); + if (!rcu_gp_in_progress(rsp)) { + trace_rcu_utilization("End fqs"); return; /* No grace period in progress, nothing to force. */ + } if (!raw_spin_trylock_irqsave(&rsp->fqslock, flags)) { rsp->n_force_qs_lh++; /* Inexact, can lose counts. Tough! */ + trace_rcu_utilization("End fqs"); return; /* Someone else is already on the job. */ } if (relaxed && ULONG_CMP_GE(rsp->jiffies_force_qs, jiffies)) @@ -1421,11 +1429,13 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) raw_spin_unlock(&rsp->fqslock); /* irqs remain disabled */ rsp->fqs_need_gp = 0; rcu_start_gp(rsp, flags); /* releases rnp->lock */ + trace_rcu_utilization("End fqs"); return; } raw_spin_unlock(&rnp->lock); /* irqs remain disabled */ unlock_fqs_ret: raw_spin_unlock_irqrestore(&rsp->fqslock, flags); + trace_rcu_utilization("End fqs"); } #else /* #ifdef CONFIG_SMP */ @@ -1481,6 +1491,7 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) */ static void rcu_process_callbacks(struct softirq_action *unused) { + trace_rcu_utilization("Start RCU core"); __rcu_process_callbacks(&rcu_sched_state, &__get_cpu_var(rcu_sched_data)); __rcu_process_callbacks(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); @@ -1488,6 +1499,7 @@ static void rcu_process_callbacks(struct softirq_action *unused) /* If we are last CPU on way to dyntick-idle mode, accelerate it. */ rcu_needs_cpu_flush(); + trace_rcu_utilization("End RCU core"); } /* @@ -1910,6 +1922,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu); struct rcu_node *rnp = rdp->mynode; + trace_rcu_utilization("Start CPU hotplug"); switch (action) { case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: @@ -1945,6 +1958,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, default: break; } + trace_rcu_utilization("End CPU hotplug"); return NOTIFY_OK; } -- cgit v1.3-8-gc7d7 From e99033c5c160f1f247c665923a66acec693a967c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 00:13:44 -0700 Subject: rcu: Put names into TINY_RCU structures under RCU_TRACE In order to allow event tracing to distinguish between flavors of RCU, we need those names in the relevant RCU data structures. TINY_RCU has avoided them for memory-footprint reasons, so add them only if CONFIG_RCU_TRACE=y. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcu.h | 10 ++++++++-- kernel/rcutiny.c | 13 ------------- kernel/rcutiny_plugin.h | 10 ++++------ kernel/rcutree.c | 10 +++++----- kernel/rcutree_plugin.h | 2 +- 5 files changed, 18 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/kernel/rcu.h b/kernel/rcu.h index 7bc16436aba0..d7f00ec8b47b 100644 --- a/kernel/rcu.h +++ b/kernel/rcu.h @@ -23,6 +23,12 @@ #ifndef __LINUX_RCU_H #define __LINUX_RCU_H +#ifdef CONFIG_RCU_TRACE +#define RCU_TRACE(stmt) stmt +#else /* #ifdef CONFIG_RCU_TRACE */ +#define RCU_TRACE(stmt) +#endif /* #else #ifdef CONFIG_RCU_TRACE */ + /* * debug_rcu_head_queue()/debug_rcu_head_unqueue() are used internally * by call_rcu() and rcu callback execution, and are therefore not part of the @@ -68,10 +74,10 @@ static inline void __rcu_reclaim(struct rcu_head *head) unsigned long offset = (unsigned long)head->func; if (__is_kfree_rcu_offset(offset)) { - trace_rcu_invoke_kfree_callback(head, offset); + RCU_TRACE(trace_rcu_invoke_kfree_callback(head, offset)); kfree((void *)head - offset); } else { - trace_rcu_invoke_callback(head); + RCU_TRACE(trace_rcu_invoke_callback(head)); head->func(head); } } diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 19453ba1392e..0d28974b78f4 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -38,20 +38,7 @@ #include #ifdef CONFIG_RCU_TRACE - #include - -#else /* #ifdef CONFIG_RCU_TRACE */ - -/* No by-default tracing in TINY_RCU: Keep TINY_RCU tiny! */ -static void trace_rcu_invoke_kfree_callback(struct rcu_head *rhp, - unsigned long offset) -{ -} -static void trace_rcu_invoke_callback(struct rcu_head *head) -{ -} - #endif /* #else #ifdef CONFIG_RCU_TRACE */ #include "rcu.h" diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 6b0cedb383e0..791ddf7c99ab 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -26,29 +26,26 @@ #include #include -#ifdef CONFIG_RCU_TRACE -#define RCU_TRACE(stmt) stmt -#else /* #ifdef CONFIG_RCU_TRACE */ -#define RCU_TRACE(stmt) -#endif /* #else #ifdef CONFIG_RCU_TRACE */ - /* Global control variables for rcupdate callback mechanism. */ struct rcu_ctrlblk { struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ struct rcu_head **curtail; /* ->next pointer of last CB. */ RCU_TRACE(long qlen); /* Number of pending CBs. */ + RCU_TRACE(char *name); /* Name of RCU type. */ }; /* Definition for rcupdate control block. */ static struct rcu_ctrlblk rcu_sched_ctrlblk = { .donetail = &rcu_sched_ctrlblk.rcucblist, .curtail = &rcu_sched_ctrlblk.rcucblist, + RCU_TRACE(.name = "rcu_sched") }; static struct rcu_ctrlblk rcu_bh_ctrlblk = { .donetail = &rcu_bh_ctrlblk.rcucblist, .curtail = &rcu_bh_ctrlblk.rcucblist, + RCU_TRACE(.name = "rcu_bh") }; #ifdef CONFIG_DEBUG_LOCK_ALLOC @@ -131,6 +128,7 @@ static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = { .rcb.curtail = &rcu_preempt_ctrlblk.rcb.rcucblist, .nexttail = &rcu_preempt_ctrlblk.rcb.rcucblist, .blkd_tasks = LIST_HEAD_INIT(rcu_preempt_ctrlblk.blkd_tasks), + RCU_TRACE(.rcb.name = "rcu_preempt") }; static int rcu_preempted_readers_exp(void); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 2a9643bd6ae9..b953e2c72e25 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -61,7 +61,7 @@ static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; #define RCU_STATE_INITIALIZER(structname) { \ - .level = { &structname.node[0] }, \ + .level = { &structname##_state.node[0] }, \ .levelcnt = { \ NUM_RCU_LVL_0, /* root of hierarchy. */ \ NUM_RCU_LVL_1, \ @@ -72,17 +72,17 @@ static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; .signaled = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ - .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&structname.onofflock), \ - .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&structname.fqslock), \ + .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&structname##_state.onofflock), \ + .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&structname##_state.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ .name = #structname, \ } -struct rcu_state rcu_sched_state = RCU_STATE_INITIALIZER(rcu_sched_state); +struct rcu_state rcu_sched_state = RCU_STATE_INITIALIZER(rcu_sched); DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); -struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); +struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); static struct rcu_state *rcu_state; diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 43daa46bc6f2..a90bf3c17492 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -64,7 +64,7 @@ static void __init rcu_bootup_announce_oddness(void) #ifdef CONFIG_TREE_PREEMPT_RCU -struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt_state); +struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt); DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); static struct rcu_state *rcu_state = &rcu_preempt_state; -- cgit v1.3-8-gc7d7 From 72fe701b70e6ced35d734b676c13efbc8fc769a9 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 01:14:54 -0700 Subject: rcu: Add RCU type to callback-invocation tracing Add a string to the rcu_batch_start() and rcu_batch_end() trace messages that indicates the RCU type ("rcu_sched", "rcu_bh", or "rcu_preempt"). The trace messages for the actual invocations themselves are not marked, as it should be clear from the rcu_batch_start() and rcu_batch_end() events before and after. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/trace/events/rcu.h | 28 ++++++++++++++++++---------- kernel/rcutiny.c | 8 ++++---- kernel/rcutree.c | 8 ++++---- 3 files changed, 26 insertions(+), 18 deletions(-) (limited to 'kernel') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index ab458eb689fb..508824e5a77d 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -33,27 +33,31 @@ TRACE_EVENT(rcu_utilization, /* * Tracepoint for marking the beginning rcu_do_batch, performed to start - * RCU callback invocation. The first argument is the total number of - * callbacks (including those that are not yet ready to be invoked), - * and the second argument is the current RCU-callback batch limit. + * RCU callback invocation. The first argument is the RCU flavor, + * the second is the total number of callbacks (including those that + * are not yet ready to be invoked), and the third argument is the + * current RCU-callback batch limit. */ TRACE_EVENT(rcu_batch_start, - TP_PROTO(long qlen, int blimit), + TP_PROTO(char *rcuname, long qlen, int blimit), - TP_ARGS(qlen, blimit), + TP_ARGS(rcuname, qlen, blimit), TP_STRUCT__entry( + __field(char *, rcuname) __field(long, qlen) __field(int, blimit) ), TP_fast_assign( + __entry->rcuname = rcuname; __entry->qlen = qlen; __entry->blimit = blimit; ), - TP_printk("CBs=%ld bl=%d", __entry->qlen, __entry->blimit) + TP_printk("%s CBs=%ld bl=%d", + __entry->rcuname, __entry->qlen, __entry->blimit) ); /* @@ -106,23 +110,27 @@ TRACE_EVENT(rcu_invoke_kfree_callback, /* * Tracepoint for exiting rcu_do_batch after RCU callbacks have been - * invoked. The first argument is the number of callbacks actually invoked. + * invoked. The first argument is the name of the RCU flavor and + * the second argument is number of callbacks actually invoked. */ TRACE_EVENT(rcu_batch_end, - TP_PROTO(int callbacks_invoked), + TP_PROTO(char *rcuname, int callbacks_invoked), - TP_ARGS(callbacks_invoked), + TP_ARGS(rcuname, callbacks_invoked), TP_STRUCT__entry( + __field(char *, rcuname) __field(int, callbacks_invoked) ), TP_fast_assign( + __entry->rcuname = rcuname; __entry->callbacks_invoked = callbacks_invoked; ), - TP_printk("CBs-invoked=%d", __entry->callbacks_invoked) + TP_printk("%s CBs-invoked=%d", + __entry->rcuname, __entry->callbacks_invoked) ); #endif /* _TRACE_RCU_H */ diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 0d28974b78f4..1c37bdd464f1 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -168,14 +168,14 @@ static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) /* If no RCU callbacks ready to invoke, just return. */ if (&rcp->rcucblist == rcp->donetail) { - RCU_TRACE(trace_rcu_batch_start(0, -1)); - RCU_TRACE(trace_rcu_batch_end(0)); + RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, -1)); + RCU_TRACE(trace_rcu_batch_end(rcp->name, 0)); return; } /* Move the ready-to-invoke callbacks to a local list. */ local_irq_save(flags); - RCU_TRACE(trace_rcu_batch_start(0, -1)); + RCU_TRACE(trace_rcu_batch_start(rcp->name, 0, -1)); list = rcp->rcucblist; rcp->rcucblist = *rcp->donetail; *rcp->donetail = NULL; @@ -197,7 +197,7 @@ static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) RCU_TRACE(cb_count++); } RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count)); - RCU_TRACE(trace_rcu_batch_end(cb_count)); + RCU_TRACE(trace_rcu_batch_end(rcp->name, cb_count)); } /* diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b953e2c72e25..eb6e731088a0 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1199,8 +1199,8 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) /* If no callbacks are ready, just return.*/ if (!cpu_has_callbacks_ready_to_invoke(rdp)) { - trace_rcu_batch_start(0, 0); - trace_rcu_batch_end(0); + trace_rcu_batch_start(rsp->name, 0, 0); + trace_rcu_batch_end(rsp->name, 0); return; } @@ -1210,7 +1210,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) */ local_irq_save(flags); bl = rdp->blimit; - trace_rcu_batch_start(rdp->qlen, bl); + trace_rcu_batch_start(rsp->name, rdp->qlen, bl); list = rdp->nxtlist; rdp->nxtlist = *rdp->nxttail[RCU_DONE_TAIL]; *rdp->nxttail[RCU_DONE_TAIL] = NULL; @@ -1233,7 +1233,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) } local_irq_save(flags); - trace_rcu_batch_end(count); + trace_rcu_batch_end(rsp->name, count); /* Update count, and requeue any remaining callbacks. */ rdp->qlen -= count; -- cgit v1.3-8-gc7d7 From e0f23060adfa3f27beaa7918eff70258b88471b6 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 01:29:39 -0700 Subject: rcu: Update comments to reflect softirqs vs. kthreads We now have kthreads only for flavors of RCU that support boosting, so update the now-misleading comments accordingly. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 23 ++++++++++++----------- kernel/rcutree_plugin.h | 3 ++- 2 files changed, 14 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index eb6e731088a0..4e24399cabcf 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -198,7 +198,7 @@ DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { }; #endif /* #ifdef CONFIG_NO_HZ */ -static int blimit = 10; /* Maximum callbacks per softirq. */ +static int blimit = 10; /* Maximum callbacks per rcu_do_batch. */ static int qhimark = 10000; /* If this many pending, ignore blimit. */ static int qlowmark = 100; /* Once only this many pending, use blimit. */ @@ -1261,7 +1261,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) local_irq_restore(flags); - /* Re-raise the RCU softirq if there are callbacks remaining. */ + /* Re-invoke RCU core processing if there are callbacks remaining. */ if (cpu_has_callbacks_ready_to_invoke(rdp)) invoke_rcu_core(); } @@ -1269,7 +1269,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) /* * Check to see if this CPU is in a non-context-switch quiescent state * (user mode or idle loop for rcu, non-softirq execution for rcu_bh). - * Also schedule the RCU softirq handler. + * Also schedule RCU core processing. * * This function must be called with hardirqs disabled. It is normally * invoked from the scheduling-clock interrupt. If rcu_pending returns @@ -1448,9 +1448,9 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) #endif /* #else #ifdef CONFIG_SMP */ /* - * This does the RCU processing work from softirq context for the - * specified rcu_state and rcu_data structures. This may be called - * only from the CPU to whom the rdp belongs. + * This does the RCU core processing work for the specified rcu_state + * and rcu_data structures. This may be called only from the CPU to + * whom the rdp belongs. */ static void __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) @@ -1487,7 +1487,7 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) } /* - * Do softirq processing for the current CPU. + * Do RCU core processing for the current CPU. */ static void rcu_process_callbacks(struct softirq_action *unused) { @@ -1503,10 +1503,11 @@ static void rcu_process_callbacks(struct softirq_action *unused) } /* - * Wake up the current CPU's kthread. This replaces raise_softirq() - * in earlier versions of RCU. Note that because we are running on - * the current CPU with interrupts disabled, the rcu_cpu_kthread_task - * cannot disappear out from under us. + * Schedule RCU callback invocation. If the specified type of RCU + * does not support RCU priority boosting, just do a direct call, + * otherwise wake up the per-CPU kernel kthread. Note that because we + * are running on the current CPU with interrupts disabled, the + * rcu_cpu_kthread_task cannot disappear out from under us. */ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) { diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index a90bf3c17492..ecd48a2e3eeb 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1478,7 +1478,8 @@ static int rcu_cpu_kthread_should_stop(int cpu) /* * Per-CPU kernel thread that invokes RCU callbacks. This replaces the - * earlier RCU softirq. + * RCU softirq used in flavors and configurations of RCU that do not + * support RCU priority boosting. */ static int rcu_cpu_kthread(void *arg) { -- cgit v1.3-8-gc7d7 From eab0993c7ba5c7d9b3613d6037e0f31f0ccbe181 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 01:59:33 -0700 Subject: rcu: Move RCU_BOOST declarations to allow compiler checking Andi Kleen noticed that one of the RCU_BOOST data declarations was out of sync with the definition. Move the declarations so that the compiler can do the checking in the future. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.h | 7 +++++++ kernel/rcutree_trace.c | 5 ----- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 01b2ccda26fb..eee6c9406b46 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -417,6 +417,13 @@ extern struct rcu_state rcu_preempt_state; DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data); #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifdef CONFIG_RCU_BOOST +DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status); +DECLARE_PER_CPU(int, rcu_cpu_kthread_cpu); +DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_loops); +DECLARE_PER_CPU(char, rcu_cpu_has_work); +#endif /* #ifdef CONFIG_RCU_BOOST */ + #ifndef RCU_TREE_NONCORE /* Forward declarations for rcutree_plugin.h */ diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index 0a5a8bedc522..f328ed1c6e46 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -48,11 +48,6 @@ #ifdef CONFIG_RCU_BOOST -DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status); -DECLARE_PER_CPU(int, rcu_cpu_kthread_cpu); -DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_loops); -DECLARE_PER_CPU(char, rcu_cpu_has_work); - static char convert_kthread_status(unsigned int kthread_status) { if (kthread_status > RCU_KTHREAD_MAX) -- cgit v1.3-8-gc7d7 From 385680a9487d2f85382ad6d74e2a15837e47bfd9 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 21 Jun 2011 22:43:26 -0700 Subject: rcu: Add event-trace markers to TREE_RCU kthreads Add event-trace markers to TREE_RCU kthreads to allow including these kthread's CPU time in the utilization calculations. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/trace/events/rcu.h | 3 +++ kernel/rcutree_plugin.h | 12 ++++++++++++ 2 files changed, 15 insertions(+) (limited to 'kernel') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index 508824e5a77d..ac52aba00a3e 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -13,6 +13,9 @@ * "Start " -- Mark the start of the specified activity, * such as "context switch". Nesting is permitted. * "End " -- Mark the end of the specified activity. + * + * An "@" character within "" is a comment character: Data + * reduction scripts will ignore the "@" and the remainder of the line. */ TRACE_EVENT(rcu_utilization, diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index ecd48a2e3eeb..94d9ca1e4061 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1219,9 +1219,12 @@ static int rcu_boost_kthread(void *arg) int spincnt = 0; int more2boost; + trace_rcu_utilization("Start boost kthread@init"); for (;;) { rnp->boost_kthread_status = RCU_KTHREAD_WAITING; + trace_rcu_utilization("End boost kthread@rcu_wait"); rcu_wait(rnp->boost_tasks || rnp->exp_tasks); + trace_rcu_utilization("Start boost kthread@rcu_wait"); rnp->boost_kthread_status = RCU_KTHREAD_RUNNING; more2boost = rcu_boost(rnp); if (more2boost) @@ -1229,11 +1232,14 @@ static int rcu_boost_kthread(void *arg) else spincnt = 0; if (spincnt > 10) { + trace_rcu_utilization("End boost kthread@rcu_yield"); rcu_yield(rcu_boost_kthread_timer, (unsigned long)rnp); + trace_rcu_utilization("Start boost kthread@rcu_yield"); spincnt = 0; } } /* NOTREACHED */ + trace_rcu_utilization("End boost kthread@notreached"); return 0; } @@ -1490,9 +1496,12 @@ static int rcu_cpu_kthread(void *arg) char work; char *workp = &per_cpu(rcu_cpu_has_work, cpu); + trace_rcu_utilization("Start CPU kthread@init"); for (;;) { *statusp = RCU_KTHREAD_WAITING; + trace_rcu_utilization("End CPU kthread@rcu_wait"); rcu_wait(*workp != 0 || kthread_should_stop()); + trace_rcu_utilization("Start CPU kthread@rcu_wait"); local_bh_disable(); if (rcu_cpu_kthread_should_stop(cpu)) { local_bh_enable(); @@ -1513,11 +1522,14 @@ static int rcu_cpu_kthread(void *arg) spincnt = 0; if (spincnt > 10) { *statusp = RCU_KTHREAD_YIELDING; + trace_rcu_utilization("End CPU kthread@rcu_yield"); rcu_yield(rcu_cpu_kthread_timer, (unsigned long)cpu); + trace_rcu_utilization("Start CPU kthread@rcu_yield"); spincnt = 0; } } *statusp = RCU_KTHREAD_STOPPED; + trace_rcu_utilization("End CPU kthread@term"); return 0; } -- cgit v1.3-8-gc7d7 From 965a002b4f1a458c5dcb334ec29f48a0046faa25 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Jun 2011 09:55:39 -0700 Subject: rcu: Make TINY_RCU also use softirq for RCU_BOOST=n This patch #ifdefs TINY_RCU kthreads out of the kernel unless RCU_BOOST=y, thus eliminating context-switch overhead if RCU priority boosting has not been configured. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/linux/rcutiny.h | 4 ++ kernel/rcutiny.c | 74 +++++--------------------------- kernel/rcutiny_plugin.h | 110 ++++++++++++++++++++++++++++++++++++------------ 3 files changed, 97 insertions(+), 91 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index 4eab233a00cd..00b7a5e493d2 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -27,9 +27,13 @@ #include +#ifdef CONFIG_RCU_BOOST static inline void rcu_init(void) { } +#else /* #ifdef CONFIG_RCU_BOOST */ +void rcu_init(void); +#endif /* #else #ifdef CONFIG_RCU_BOOST */ static inline void rcu_barrier_bh(void) { diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 1c37bdd464f1..c9321d86999b 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -43,16 +43,11 @@ #include "rcu.h" -/* Controls for rcu_kthread() kthread, replacing RCU_SOFTIRQ used previously. */ -static struct task_struct *rcu_kthread_task; -static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq); -static unsigned long have_rcu_kthread_work; - /* Forward declarations for rcutiny_plugin.h. */ struct rcu_ctrlblk; -static void invoke_rcu_kthread(void); -static void rcu_process_callbacks(struct rcu_ctrlblk *rcp); -static int rcu_kthread(void *arg); +static void invoke_rcu_callbacks(void); +static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp); +static void rcu_process_callbacks(struct softirq_action *unused); static void __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), struct rcu_ctrlblk *rcp); @@ -101,16 +96,6 @@ static int rcu_qsctr_help(struct rcu_ctrlblk *rcp) return 0; } -/* - * Wake up rcu_kthread() to process callbacks now eligible for invocation - * or to boost readers. - */ -static void invoke_rcu_kthread(void) -{ - have_rcu_kthread_work = 1; - wake_up(&rcu_kthread_wq); -} - /* * Record an rcu quiescent state. And an rcu_bh quiescent state while we * are at it, given that any rcu quiescent state is also an rcu_bh @@ -123,7 +108,7 @@ void rcu_sched_qs(int cpu) local_irq_save(flags); if (rcu_qsctr_help(&rcu_sched_ctrlblk) + rcu_qsctr_help(&rcu_bh_ctrlblk)) - invoke_rcu_kthread(); + invoke_rcu_callbacks(); local_irq_restore(flags); } @@ -136,7 +121,7 @@ void rcu_bh_qs(int cpu) local_irq_save(flags); if (rcu_qsctr_help(&rcu_bh_ctrlblk)) - invoke_rcu_kthread(); + invoke_rcu_callbacks(); local_irq_restore(flags); } @@ -160,7 +145,7 @@ void rcu_check_callbacks(int cpu, int user) * Invoke the RCU callbacks on the specified rcu_ctrlkblk structure * whose grace period has elapsed. */ -static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) +static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) { struct rcu_head *next, *list; unsigned long flags; @@ -200,36 +185,11 @@ static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) RCU_TRACE(trace_rcu_batch_end(rcp->name, cb_count)); } -/* - * This kthread invokes RCU callbacks whose grace periods have - * elapsed. It is awakened as needed, and takes the place of the - * RCU_SOFTIRQ that was used previously for this purpose. - * This is a kthread, but it is never stopped, at least not until - * the system goes down. - */ -static int rcu_kthread(void *arg) +static void rcu_process_callbacks(struct softirq_action *unused) { - unsigned long work; - unsigned long morework; - unsigned long flags; - - for (;;) { - wait_event_interruptible(rcu_kthread_wq, - have_rcu_kthread_work != 0); - morework = rcu_boost(); - local_irq_save(flags); - work = have_rcu_kthread_work; - have_rcu_kthread_work = morework; - local_irq_restore(flags); - if (work) { - rcu_process_callbacks(&rcu_sched_ctrlblk); - rcu_process_callbacks(&rcu_bh_ctrlblk); - rcu_preempt_process_callbacks(); - } - schedule_timeout_interruptible(1); /* Leave CPU for others. */ - } - - return 0; /* Not reached, but needed to shut gcc up. */ + __rcu_process_callbacks(&rcu_sched_ctrlblk); + __rcu_process_callbacks(&rcu_bh_ctrlblk); + rcu_preempt_process_callbacks(); } /* @@ -291,17 +251,3 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) __call_rcu(head, func, &rcu_bh_ctrlblk); } EXPORT_SYMBOL_GPL(call_rcu_bh); - -/* - * Spawn the kthread that invokes RCU callbacks. - */ -static int __init rcu_spawn_kthreads(void) -{ - struct sched_param sp; - - rcu_kthread_task = kthread_run(rcu_kthread, NULL, "rcu_kthread"); - sp.sched_priority = RCU_BOOST_PRIO; - sched_setscheduler_nocheck(rcu_kthread_task, SCHED_FIFO, &sp); - return 0; -} -early_initcall(rcu_spawn_kthreads); diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 791ddf7c99ab..02aa7139861c 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -245,6 +245,13 @@ static void show_tiny_preempt_stats(struct seq_file *m) #include "rtmutex_common.h" +#define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO + +/* Controls for rcu_kthread() kthread. */ +static struct task_struct *rcu_kthread_task; +static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq); +static unsigned long have_rcu_kthread_work; + /* * Carry out RCU priority boosting on the task indicated by ->boost_tasks, * and advance ->boost_tasks to the next task in the ->blkd_tasks list. @@ -332,7 +339,7 @@ static int rcu_initiate_boost(void) if (rcu_preempt_ctrlblk.exp_tasks == NULL) rcu_preempt_ctrlblk.boost_tasks = rcu_preempt_ctrlblk.gp_tasks; - invoke_rcu_kthread(); + invoke_rcu_callbacks(); } else RCU_TRACE(rcu_initiate_boost_trace()); return 1; @@ -350,14 +357,6 @@ static void rcu_preempt_boost_start_gp(void) #else /* #ifdef CONFIG_RCU_BOOST */ -/* - * If there is no RCU priority boosting, we don't boost. - */ -static int rcu_boost(void) -{ - return 0; -} - /* * If there is no RCU priority boosting, we don't initiate boosting, * but we do indicate whether there are blocked readers blocking the @@ -425,7 +424,7 @@ static void rcu_preempt_cpu_qs(void) /* If there are done callbacks, cause them to be invoked. */ if (*rcu_preempt_ctrlblk.rcb.donetail != NULL) - invoke_rcu_kthread(); + invoke_rcu_callbacks(); } /* @@ -646,7 +645,7 @@ static void rcu_preempt_check_callbacks(void) rcu_preempt_cpu_qs(); if (&rcu_preempt_ctrlblk.rcb.rcucblist != rcu_preempt_ctrlblk.rcb.donetail) - invoke_rcu_kthread(); + invoke_rcu_callbacks(); if (rcu_preempt_gp_in_progress() && rcu_cpu_blocking_cur_gp() && rcu_preempt_running_reader()) @@ -672,7 +671,7 @@ static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp) */ static void rcu_preempt_process_callbacks(void) { - rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb); + __rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb); } /* @@ -847,15 +846,6 @@ static void show_tiny_preempt_stats(struct seq_file *m) #endif /* #ifdef CONFIG_RCU_TRACE */ -/* - * Because preemptible RCU does not exist, it is never necessary to - * boost preempted RCU readers. - */ -static int rcu_boost(void) -{ - return 0; -} - /* * Because preemptible RCU does not exist, it never has any callbacks * to check. @@ -882,6 +872,78 @@ static void rcu_preempt_process_callbacks(void) #endif /* #else #ifdef CONFIG_TINY_PREEMPT_RCU */ +#ifdef CONFIG_RCU_BOOST + +/* + * Wake up rcu_kthread() to process callbacks now eligible for invocation + * or to boost readers. + */ +static void invoke_rcu_callbacks(void) +{ + have_rcu_kthread_work = 1; + wake_up(&rcu_kthread_wq); +} + +/* + * This kthread invokes RCU callbacks whose grace periods have + * elapsed. It is awakened as needed, and takes the place of the + * RCU_SOFTIRQ that is used for this purpose when boosting is disabled. + * This is a kthread, but it is never stopped, at least not until + * the system goes down. + */ +static int rcu_kthread(void *arg) +{ + unsigned long work; + unsigned long morework; + unsigned long flags; + + for (;;) { + wait_event_interruptible(rcu_kthread_wq, + have_rcu_kthread_work != 0); + morework = rcu_boost(); + local_irq_save(flags); + work = have_rcu_kthread_work; + have_rcu_kthread_work = morework; + local_irq_restore(flags); + if (work) + rcu_process_callbacks(NULL); + schedule_timeout_interruptible(1); /* Leave CPU for others. */ + } + + return 0; /* Not reached, but needed to shut gcc up. */ +} + +/* + * Spawn the kthread that invokes RCU callbacks. + */ +static int __init rcu_spawn_kthreads(void) +{ + struct sched_param sp; + + rcu_kthread_task = kthread_run(rcu_kthread, NULL, "rcu_kthread"); + sp.sched_priority = RCU_BOOST_PRIO; + sched_setscheduler_nocheck(rcu_kthread_task, SCHED_FIFO, &sp); + return 0; +} +early_initcall(rcu_spawn_kthreads); + +#else /* #ifdef CONFIG_RCU_BOOST */ + +/* + * Start up softirq processing of callbacks. + */ +void invoke_rcu_callbacks(void) +{ + raise_softirq(RCU_SOFTIRQ); +} + +void rcu_init(void) +{ + open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); +} + +#endif /* #else #ifdef CONFIG_RCU_BOOST */ + #ifdef CONFIG_DEBUG_LOCK_ALLOC #include @@ -897,12 +959,6 @@ void __init rcu_scheduler_starting(void) #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ -#ifdef CONFIG_RCU_BOOST -#define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO -#else /* #ifdef CONFIG_RCU_BOOST */ -#define RCU_BOOST_PRIO 1 -#endif /* #else #ifdef CONFIG_RCU_BOOST */ - #ifdef CONFIG_RCU_TRACE #ifdef CONFIG_RCU_BOOST -- cgit v1.3-8-gc7d7 From d4c08f2ac311a360230eef7e5395b0ec8d8f0670 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 25 Jun 2011 06:36:56 -0700 Subject: rcu: Add grace-period, quiescent-state, and call_rcu trace events Add trace events to record grace-period start and end, quiescent states, CPUs noticing grace-period start and end, grace-period initialization, call_rcu() invocation, tasks blocking in RCU read-side critical sections, tasks exiting those same critical sections, force_quiescent_state() detection of dyntick-idle and offline CPUs, CPUs entering and leaving dyntick-idle mode (except from NMIs), CPUs coming online and going offline, and CPUs being kicked for staying in dyntick-idle mode for too long (as in many weeks, even on 32-bit systems). Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney rcu: Add the rcu flavor to callback trace events The earlier trace events for registering RCU callbacks and for invoking them did not include the RCU flavor (rcu_bh, rcu_preempt, or rcu_sched). This commit adds the RCU flavor to those trace events. Signed-off-by: Paul E. McKenney --- include/trace/events/rcu.h | 345 +++++++++++++++++++++++++++++++++++++++++++-- kernel/rcu.h | 6 +- kernel/rcutiny.c | 4 +- kernel/rcutree.c | 45 +++++- kernel/rcutree.h | 1 + kernel/rcutree_plugin.h | 22 ++- 6 files changed, 399 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index ac52aba00a3e..669fbd62ec25 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -24,7 +24,7 @@ TRACE_EVENT(rcu_utilization, TP_ARGS(s), TP_STRUCT__entry( - __field(char *, s) + __field(char *, s) ), TP_fast_assign( @@ -34,6 +34,297 @@ TRACE_EVENT(rcu_utilization, TP_printk("%s", __entry->s) ); +#ifdef CONFIG_RCU_TRACE + +#if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU) + +/* + * Tracepoint for grace-period events: starting and ending a grace + * period ("start" and "end", respectively), a CPU noting the start + * of a new grace period or the end of an old grace period ("cpustart" + * and "cpuend", respectively), a CPU passing through a quiescent + * state ("cpuqs"), a CPU coming online or going offline ("cpuonl" + * and "cpuofl", respectively), and a CPU being kicked for being too + * long in dyntick-idle mode ("kick"). + */ +TRACE_EVENT(rcu_grace_period, + + TP_PROTO(char *rcuname, unsigned long gpnum, char *gpevent), + + TP_ARGS(rcuname, gpnum, gpevent), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(char *, gpevent) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->gpevent = gpevent; + ), + + TP_printk("%s %lu %s", + __entry->rcuname, __entry->gpnum, __entry->gpevent) +); + +/* + * Tracepoint for grace-period-initialization events. These are + * distinguished by the type of RCU, the new grace-period number, the + * rcu_node structure level, the starting and ending CPU covered by the + * rcu_node structure, and the mask of CPUs that will be waited for. + * All but the type of RCU are extracted from the rcu_node structure. + */ +TRACE_EVENT(rcu_grace_period_init, + + TP_PROTO(char *rcuname, unsigned long gpnum, u8 level, + int grplo, int grphi, unsigned long qsmask), + + TP_ARGS(rcuname, gpnum, level, grplo, grphi, qsmask), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(u8, level) + __field(int, grplo) + __field(int, grphi) + __field(unsigned long, qsmask) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->level = level; + __entry->grplo = grplo; + __entry->grphi = grphi; + __entry->qsmask = qsmask; + ), + + TP_printk("%s %lu %u %d %d %lx", + __entry->rcuname, __entry->gpnum, __entry->level, + __entry->grplo, __entry->grphi, __entry->qsmask) +); + +/* + * Tracepoint for tasks blocking within preemptible-RCU read-side + * critical sections. Track the type of RCU (which one day might + * include SRCU), the grace-period number that the task is blocking + * (the current or the next), and the task's PID. + */ +TRACE_EVENT(rcu_preempt_task, + + TP_PROTO(char *rcuname, int pid, unsigned long gpnum), + + TP_ARGS(rcuname, pid, gpnum), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(int, pid) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->pid = pid; + ), + + TP_printk("%s %lu %d", + __entry->rcuname, __entry->gpnum, __entry->pid) +); + +/* + * Tracepoint for tasks that blocked within a given preemptible-RCU + * read-side critical section exiting that critical section. Track the + * type of RCU (which one day might include SRCU) and the task's PID. + */ +TRACE_EVENT(rcu_unlock_preempted_task, + + TP_PROTO(char *rcuname, unsigned long gpnum, int pid), + + TP_ARGS(rcuname, gpnum, pid), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(int, pid) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->pid = pid; + ), + + TP_printk("%s %lu %d", __entry->rcuname, __entry->gpnum, __entry->pid) +); + +/* + * Tracepoint for quiescent-state-reporting events. These are + * distinguished by the type of RCU, the grace-period number, the + * mask of quiescent lower-level entities, the rcu_node structure level, + * the starting and ending CPU covered by the rcu_node structure, and + * whether there are any blocked tasks blocking the current grace period. + * All but the type of RCU are extracted from the rcu_node structure. + */ +TRACE_EVENT(rcu_quiescent_state_report, + + TP_PROTO(char *rcuname, unsigned long gpnum, + unsigned long mask, unsigned long qsmask, + u8 level, int grplo, int grphi, int gp_tasks), + + TP_ARGS(rcuname, gpnum, mask, qsmask, level, grplo, grphi, gp_tasks), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(unsigned long, mask) + __field(unsigned long, qsmask) + __field(u8, level) + __field(int, grplo) + __field(int, grphi) + __field(u8, gp_tasks) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->mask = mask; + __entry->qsmask = qsmask; + __entry->level = level; + __entry->grplo = grplo; + __entry->grphi = grphi; + __entry->gp_tasks = gp_tasks; + ), + + TP_printk("%s %lu %lx>%lx %u %d %d %u", + __entry->rcuname, __entry->gpnum, + __entry->mask, __entry->qsmask, __entry->level, + __entry->grplo, __entry->grphi, __entry->gp_tasks) +); + +/* + * Tracepoint for quiescent states detected by force_quiescent_state(). + * These trace events include the type of RCU, the grace-period number + * that was blocked by the CPU, the CPU itself, and the type of quiescent + * state, which can be "dti" for dyntick-idle mode, "ofl" for CPU offline, + * or "kick" when kicking a CPU that has been in dyntick-idle mode for + * too long. + */ +TRACE_EVENT(rcu_fqs, + + TP_PROTO(char *rcuname, unsigned long gpnum, int cpu, char *qsevent), + + TP_ARGS(rcuname, gpnum, cpu, qsevent), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(unsigned long, gpnum) + __field(int, cpu) + __field(char *, qsevent) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->gpnum = gpnum; + __entry->cpu = cpu; + __entry->qsevent = qsevent; + ), + + TP_printk("%s %lu %d %s", + __entry->rcuname, __entry->gpnum, + __entry->cpu, __entry->qsevent) +); + +#endif /* #if defined(CONFIG_TREE_RCU) || defined(CONFIG_TREE_PREEMPT_RCU) */ + +/* + * Tracepoint for dyntick-idle entry/exit events. These take a string + * as argument: "Start" for entering dyntick-idle mode and "End" for + * leaving it. + */ +TRACE_EVENT(rcu_dyntick, + + TP_PROTO(char *polarity), + + TP_ARGS(polarity), + + TP_STRUCT__entry( + __field(char *, polarity) + ), + + TP_fast_assign( + __entry->polarity = polarity; + ), + + TP_printk("%s", __entry->polarity) +); + +/* + * Tracepoint for the registration of a single RCU callback function. + * The first argument is the type of RCU, the second argument is + * a pointer to the RCU callback itself, and the third element is the + * new RCU callback queue length for the current CPU. + */ +TRACE_EVENT(rcu_callback, + + TP_PROTO(char *rcuname, struct rcu_head *rhp, long qlen), + + TP_ARGS(rcuname, rhp, qlen), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(void *, rhp) + __field(void *, func) + __field(long, qlen) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->rhp = rhp; + __entry->func = rhp->func; + __entry->qlen = qlen; + ), + + TP_printk("%s rhp=%p func=%pf %ld", + __entry->rcuname, __entry->rhp, __entry->func, __entry->qlen) +); + +/* + * Tracepoint for the registration of a single RCU callback of the special + * kfree() form. The first argument is the RCU type, the second argument + * is a pointer to the RCU callback, the third argument is the offset + * of the callback within the enclosing RCU-protected data structure, + * and the fourth argument is the new RCU callback queue length for the + * current CPU. + */ +TRACE_EVENT(rcu_kfree_callback, + + TP_PROTO(char *rcuname, struct rcu_head *rhp, unsigned long offset, + long qlen), + + TP_ARGS(rcuname, rhp, offset, qlen), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(void *, rhp) + __field(unsigned long, offset) + __field(long, qlen) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->rhp = rhp; + __entry->offset = offset; + __entry->qlen = qlen; + ), + + TP_printk("%s rhp=%p func=%ld %ld", + __entry->rcuname, __entry->rhp, __entry->offset, + __entry->qlen) +); + /* * Tracepoint for marking the beginning rcu_do_batch, performed to start * RCU callback invocation. The first argument is the RCU flavor, @@ -65,50 +356,58 @@ TRACE_EVENT(rcu_batch_start, /* * Tracepoint for the invocation of a single RCU callback function. - * The argument is a pointer to the RCU callback itself. + * The first argument is the type of RCU, and the second argument is + * a pointer to the RCU callback itself. */ TRACE_EVENT(rcu_invoke_callback, - TP_PROTO(struct rcu_head *rhp), + TP_PROTO(char *rcuname, struct rcu_head *rhp), - TP_ARGS(rhp), + TP_ARGS(rcuname, rhp), TP_STRUCT__entry( - __field(void *, rhp) - __field(void *, func) + __field(char *, rcuname) + __field(void *, rhp) + __field(void *, func) ), TP_fast_assign( + __entry->rcuname = rcuname; __entry->rhp = rhp; __entry->func = rhp->func; ), - TP_printk("rhp=%p func=%pf", __entry->rhp, __entry->func) + TP_printk("%s rhp=%p func=%pf", + __entry->rcuname, __entry->rhp, __entry->func) ); /* * Tracepoint for the invocation of a single RCU callback of the special - * kfree() form. The first argument is a pointer to the RCU callback - * and the second argument is the offset of the callback within the - * enclosing RCU-protected data structure. + * kfree() form. The first argument is the RCU flavor, the second + * argument is a pointer to the RCU callback, and the third argument + * is the offset of the callback within the enclosing RCU-protected + * data structure. */ TRACE_EVENT(rcu_invoke_kfree_callback, - TP_PROTO(struct rcu_head *rhp, unsigned long offset), + TP_PROTO(char *rcuname, struct rcu_head *rhp, unsigned long offset), - TP_ARGS(rhp, offset), + TP_ARGS(rcuname, rhp, offset), TP_STRUCT__entry( - __field(void *, rhp) + __field(char *, rcuname) + __field(void *, rhp) __field(unsigned long, offset) ), TP_fast_assign( + __entry->rcuname = rcuname; __entry->rhp = rhp; __entry->offset = offset; ), - TP_printk("rhp=%p func=%ld", __entry->rhp, __entry->offset) + TP_printk("%s rhp=%p func=%ld", + __entry->rcuname, __entry->rhp, __entry->offset) ); /* @@ -136,6 +435,24 @@ TRACE_EVENT(rcu_batch_end, __entry->rcuname, __entry->callbacks_invoked) ); +#else /* #ifdef CONFIG_RCU_TRACE */ + +#define trace_rcu_grace_period(rcuname, gpnum, gpevent) do { } while (0) +#define trace_rcu_grace_period_init(rcuname, gpnum, level, grplo, grphi, qsmask) do { } while (0) +#define trace_rcu_preempt_task(rcuname, pid, gpnum) do { } while (0) +#define trace_rcu_unlock_preempted_task(rcuname, gpnum, pid) do { } while (0) +#define trace_rcu_quiescent_state_report(rcuname, gpnum, mask, qsmask, level, grplo, grphi, gp_tasks) do { } while (0) +#define trace_rcu_fqs(rcuname, gpnum, cpu, qsevent) do { } while (0) +#define trace_rcu_dyntick(polarity) do { } while (0) +#define trace_rcu_callback(rcuname, rhp, qlen) do { } while (0) +#define trace_rcu_kfree_callback(rcuname, rhp, offset, qlen) do { } while (0) +#define trace_rcu_batch_start(rcuname, qlen, blimit) do { } while (0) +#define trace_rcu_invoke_callback(rcuname, rhp) do { } while (0) +#define trace_rcu_invoke_kfree_callback(rcuname, rhp, offset) do { } while (0) +#define trace_rcu_batch_end(rcuname, callbacks_invoked) do { } while (0) + +#endif /* #else #ifdef CONFIG_RCU_TRACE */ + #endif /* _TRACE_RCU_H */ /* This part must be outside protection */ diff --git a/kernel/rcu.h b/kernel/rcu.h index d7f00ec8b47b..f600868d550d 100644 --- a/kernel/rcu.h +++ b/kernel/rcu.h @@ -69,15 +69,15 @@ static inline void debug_rcu_head_unqueue(struct rcu_head *head) extern void kfree(const void *); -static inline void __rcu_reclaim(struct rcu_head *head) +static inline void __rcu_reclaim(char *rn, struct rcu_head *head) { unsigned long offset = (unsigned long)head->func; if (__is_kfree_rcu_offset(offset)) { - RCU_TRACE(trace_rcu_invoke_kfree_callback(head, offset)); + RCU_TRACE(trace_rcu_invoke_kfree_callback(rn, head, offset)); kfree((void *)head - offset); } else { - RCU_TRACE(trace_rcu_invoke_callback(head)); + RCU_TRACE(trace_rcu_invoke_callback(rn, head)); head->func(head); } } diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index c9321d86999b..da775c87f27f 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -147,6 +147,7 @@ void rcu_check_callbacks(int cpu, int user) */ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) { + char *rn = NULL; struct rcu_head *next, *list; unsigned long flags; RCU_TRACE(int cb_count = 0); @@ -171,12 +172,13 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) local_irq_restore(flags); /* Invoke the callbacks on the local list. */ + RCU_TRACE(rn = rcp->name); while (list) { next = list->next; prefetch(next); debug_rcu_head_unqueue(list); local_bh_disable(); - __rcu_reclaim(list); + __rcu_reclaim(rn, list); local_bh_enable(); list = next; RCU_TRACE(cb_count++); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 4e24399cabcf..7e0282949f8a 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -166,6 +166,8 @@ void rcu_sched_qs(int cpu) rdp->passed_quiesc_completed = rdp->gpnum - 1; barrier(); + if (rdp->passed_quiesc == 0) + trace_rcu_grace_period("rcu_sched", rdp->gpnum, "cpuqs"); rdp->passed_quiesc = 1; } @@ -175,6 +177,8 @@ void rcu_bh_qs(int cpu) rdp->passed_quiesc_completed = rdp->gpnum - 1; barrier(); + if (rdp->passed_quiesc == 0) + trace_rcu_grace_period("rcu_bh", rdp->gpnum, "cpuqs"); rdp->passed_quiesc = 1; } @@ -319,6 +323,7 @@ static int rcu_implicit_offline_qs(struct rcu_data *rdp) * trust its state not to change because interrupts are disabled. */ if (cpu_is_offline(rdp->cpu)) { + trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "ofl"); rdp->offline_fqs++; return 1; } @@ -359,6 +364,7 @@ void rcu_enter_nohz(void) local_irq_restore(flags); return; } + trace_rcu_dyntick("Start"); /* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */ smp_mb__before_atomic_inc(); /* See above. */ atomic_inc(&rdtp->dynticks); @@ -396,6 +402,7 @@ void rcu_exit_nohz(void) /* CPUs seeing atomic_inc() must see later RCU read-side crit sects */ smp_mb__after_atomic_inc(); /* See above. */ WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1)); + trace_rcu_dyntick("End"); local_irq_restore(flags); } @@ -501,6 +508,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) * of the current RCU grace period. */ if ((curr & 0x1) == 0 || ULONG_CMP_GE(curr, snap + 2)) { + trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "dti"); rdp->dynticks_fqs++; return 1; } @@ -683,6 +691,7 @@ static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct * go looking for one. */ rdp->gpnum = rnp->gpnum; + trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpustart"); if (rnp->qsmask & rdp->grpmask) { rdp->qs_pending = 1; rdp->passed_quiesc = 0; @@ -746,6 +755,7 @@ __rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_dat /* Remember that we saw this grace-period completion. */ rdp->completed = rnp->completed; + trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuend"); /* * If we were in an extended quiescent state, we may have @@ -856,6 +866,7 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) /* Advance to a new grace period and initialize state. */ rsp->gpnum++; + trace_rcu_grace_period(rsp->name, rsp->gpnum, "start"); WARN_ON_ONCE(rsp->signaled == RCU_GP_INIT); rsp->signaled = RCU_GP_INIT; /* Hold off force_quiescent_state. */ rsp->jiffies_force_qs = jiffies + RCU_JIFFIES_TILL_FORCE_QS; @@ -870,6 +881,9 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state OK. */ rcu_start_gp_per_cpu(rsp, rnp, rdp); rcu_preempt_boost_start_gp(rnp); + trace_rcu_grace_period_init(rsp->name, rnp->gpnum, + rnp->level, rnp->grplo, + rnp->grphi, rnp->qsmask); raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -906,6 +920,9 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) if (rnp == rdp->mynode) rcu_start_gp_per_cpu(rsp, rnp, rdp); rcu_preempt_boost_start_gp(rnp); + trace_rcu_grace_period_init(rsp->name, rnp->gpnum, + rnp->level, rnp->grplo, + rnp->grphi, rnp->qsmask); raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ } @@ -939,6 +956,7 @@ static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) if (gp_duration > rsp->gp_max) rsp->gp_max = gp_duration; rsp->completed = rsp->gpnum; + trace_rcu_grace_period(rsp->name, rsp->completed, "end"); rsp->signaled = RCU_GP_IDLE; rcu_start_gp(rsp, flags); /* releases root node's rnp->lock. */ } @@ -967,6 +985,10 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp, return; } rnp->qsmask &= ~mask; + trace_rcu_quiescent_state_report(rsp->name, rnp->gpnum, + mask, rnp->qsmask, rnp->level, + rnp->grplo, rnp->grphi, + !!rnp->gp_tasks); if (rnp->qsmask != 0 || rcu_preempt_blocked_readers_cgp(rnp)) { /* Other bits still set at this level, so done. */ @@ -1135,11 +1157,20 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) if (rnp->qsmaskinit != 0) { if (rnp != rdp->mynode) raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + else + trace_rcu_grace_period(rsp->name, + rnp->gpnum + 1 - + !!(rnp->qsmask & mask), + "cpuofl"); break; } - if (rnp == rdp->mynode) + if (rnp == rdp->mynode) { + trace_rcu_grace_period(rsp->name, + rnp->gpnum + 1 - + !!(rnp->qsmask & mask), + "cpuofl"); need_report = rcu_preempt_offline_tasks(rsp, rnp, rdp); - else + } else raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ mask = rnp->grpmask; rnp = rnp->parent; @@ -1226,7 +1257,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) next = list->next; prefetch(next); debug_rcu_head_unqueue(list); - __rcu_reclaim(list); + __rcu_reclaim(rsp->name, list); list = next; if (++count >= bl) break; @@ -1552,6 +1583,12 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rdp->nxttail[RCU_NEXT_TAIL] = &head->next; rdp->qlen++; + if (__is_kfree_rcu_offset((unsigned long)func)) + trace_rcu_kfree_callback(rsp->name, head, (unsigned long)func, + rdp->qlen); + else + trace_rcu_callback(rsp->name, head, rdp->qlen); + /* If interrupts were disabled, don't dive into RCU core. */ if (irqs_disabled_flags(flags)) { local_irq_restore(flags); @@ -1850,6 +1887,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->dynticks = &per_cpu(rcu_dynticks, cpu); #endif /* #ifdef CONFIG_NO_HZ */ rdp->cpu = cpu; + rdp->rsp = rsp; raw_spin_unlock_irqrestore(&rnp->lock, flags); } @@ -1898,6 +1936,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) rdp->gpnum = rnp->completed; /* if GP in progress... */ rdp->completed = rnp->completed; rdp->passed_quiesc_completed = rnp->completed - 1; + trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuonl"); } raw_spin_unlock(&rnp->lock); /* irqs already disabled. */ rnp = rnp->parent; diff --git a/kernel/rcutree.h b/kernel/rcutree.h index eee6c9406b46..d11a0065321c 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -299,6 +299,7 @@ struct rcu_data { unsigned long n_rp_need_nothing; int cpu; + struct rcu_state *rsp; }; /* Values for signaled field in struct rcu_state. */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 94d9ca1e4061..bdb2e82f78d3 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -124,6 +124,8 @@ static void rcu_preempt_qs(int cpu) rdp->passed_quiesc_completed = rdp->gpnum - 1; barrier(); + if (rdp->passed_quiesc == 0) + trace_rcu_grace_period("rcu_preempt", rdp->gpnum, "cpuqs"); rdp->passed_quiesc = 1; current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; } @@ -190,6 +192,11 @@ static void rcu_preempt_note_context_switch(int cpu) if (rnp->qsmask & rdp->grpmask) rnp->gp_tasks = &t->rcu_node_entry; } + trace_rcu_preempt_task(rdp->rsp->name, + t->pid, + (rnp->qsmask & rdp->grpmask) + ? rnp->gpnum + : rnp->gpnum + 1); raw_spin_unlock_irqrestore(&rnp->lock, flags); } else if (t->rcu_read_lock_nesting < 0 && t->rcu_read_unlock_special) { @@ -344,6 +351,8 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */ np = rcu_next_node_entry(t, rnp); list_del_init(&t->rcu_node_entry); + trace_rcu_unlock_preempted_task("rcu_preempt", + rnp->gpnum, t->pid); if (&t->rcu_node_entry == rnp->gp_tasks) rnp->gp_tasks = np; if (&t->rcu_node_entry == rnp->exp_tasks) @@ -364,10 +373,17 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) * we aren't waiting on any CPUs, report the quiescent state. * Note that rcu_report_unblock_qs_rnp() releases rnp->lock. */ - if (empty) - raw_spin_unlock_irqrestore(&rnp->lock, flags); - else + if (!empty && !rcu_preempt_blocked_readers_cgp(rnp)) { + trace_rcu_quiescent_state_report("preempt_rcu", + rnp->gpnum, + 0, rnp->qsmask, + rnp->level, + rnp->grplo, + rnp->grphi, + !!rnp->gp_tasks); rcu_report_unblock_qs_rnp(rnp, flags); + } else + raw_spin_unlock_irqrestore(&rnp->lock, flags); #ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ -- cgit v1.3-8-gc7d7 From e4cc1f22b2f4e9b0207a8cdb63e56dcf99e82d35 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 27 Jun 2011 00:17:43 -0700 Subject: rcu: Simplify quiescent-state accounting There is often a delay between the time that a CPU passes through a quiescent state and the time that this quiescent state is reported to the RCU core. It is quite possible that the grace period ended before the quiescent state could be reported, for example, some other CPU might have deduced that this CPU passed through dyntick-idle mode. It is critically important that quiescent state be counted only against the grace period that was in effect at the time that the quiescent state was detected. Previously, this was handled by recording the number of the last grace period to complete when passing through a quiescent state. The RCU core then checks this number against the current value, and rejects the quiescent state if there is a mismatch. However, one additional possibility must be accounted for, namely that the quiescent state was recorded after the prior grace period completed but before the current grace period started. In this case, the RCU core must reject the quiescent state, but the recorded number will match. This is handled when the CPU becomes aware of a new grace period -- at that point, it invalidates any prior quiescent state. This works, but is a bit indirect. The new approach records the current grace period, and the RCU core checks to see (1) that this is still the current grace period and (2) that this grace period has not yet ended. This approach simplifies reasoning about correctness, and this commit changes over to this new approach. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- Documentation/RCU/trace.txt | 34 +++++++++++++++++----------------- kernel/rcutree.c | 44 ++++++++++++++++++++++---------------------- kernel/rcutree.h | 6 +++--- kernel/rcutree_plugin.h | 6 +++--- kernel/rcutree_trace.c | 8 ++++---- 5 files changed, 49 insertions(+), 49 deletions(-) (limited to 'kernel') diff --git a/Documentation/RCU/trace.txt b/Documentation/RCU/trace.txt index a67af0a39ded..aaf65f6c6cd7 100644 --- a/Documentation/RCU/trace.txt +++ b/Documentation/RCU/trace.txt @@ -33,23 +33,23 @@ rcu/rcuboost: The output of "cat rcu/rcudata" looks as follows: rcu_sched: - 0 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=545/1/0 df=50 of=0 ri=0 ql=163 qs=NRW. kt=0/W/0 ktl=ebc3 b=10 ci=153737 co=0 ca=0 - 1 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=967/1/0 df=58 of=0 ri=0 ql=634 qs=NRW. kt=0/W/1 ktl=58c b=10 ci=191037 co=0 ca=0 - 2 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=1081/1/0 df=175 of=0 ri=0 ql=74 qs=N.W. kt=0/W/2 ktl=da94 b=10 ci=75991 co=0 ca=0 - 3 c=20942 g=20943 pq=1 pqc=20942 qp=1 dt=1846/0/0 df=404 of=0 ri=0 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=72261 co=0 ca=0 - 4 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=369/1/0 df=83 of=0 ri=0 ql=48 qs=N.W. kt=0/W/4 ktl=e0e7 b=10 ci=128365 co=0 ca=0 - 5 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=381/1/0 df=64 of=0 ri=0 ql=169 qs=NRW. kt=0/W/5 ktl=fb2f b=10 ci=164360 co=0 ca=0 - 6 c=20972 g=20973 pq=1 pqc=20972 qp=0 dt=1037/1/0 df=183 of=0 ri=0 ql=62 qs=N.W. kt=0/W/6 ktl=d2ad b=10 ci=65663 co=0 ca=0 - 7 c=20897 g=20897 pq=1 pqc=20896 qp=0 dt=1572/0/0 df=382 of=0 ri=0 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=75006 co=0 ca=0 + 0 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=545/1/0 df=50 of=0 ri=0 ql=163 qs=NRW. kt=0/W/0 ktl=ebc3 b=10 ci=153737 co=0 ca=0 + 1 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=967/1/0 df=58 of=0 ri=0 ql=634 qs=NRW. kt=0/W/1 ktl=58c b=10 ci=191037 co=0 ca=0 + 2 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=1081/1/0 df=175 of=0 ri=0 ql=74 qs=N.W. kt=0/W/2 ktl=da94 b=10 ci=75991 co=0 ca=0 + 3 c=20942 g=20943 pq=1 pgp=20942 qp=1 dt=1846/0/0 df=404 of=0 ri=0 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=72261 co=0 ca=0 + 4 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=369/1/0 df=83 of=0 ri=0 ql=48 qs=N.W. kt=0/W/4 ktl=e0e7 b=10 ci=128365 co=0 ca=0 + 5 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=381/1/0 df=64 of=0 ri=0 ql=169 qs=NRW. kt=0/W/5 ktl=fb2f b=10 ci=164360 co=0 ca=0 + 6 c=20972 g=20973 pq=1 pgp=20973 qp=0 dt=1037/1/0 df=183 of=0 ri=0 ql=62 qs=N.W. kt=0/W/6 ktl=d2ad b=10 ci=65663 co=0 ca=0 + 7 c=20897 g=20897 pq=1 pgp=20896 qp=0 dt=1572/0/0 df=382 of=0 ri=0 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=75006 co=0 ca=0 rcu_bh: - 0 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=545/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/0 ktl=ebc3 b=10 ci=0 co=0 ca=0 - 1 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=967/1/0 df=3 of=0 ri=1 ql=0 qs=.... kt=0/W/1 ktl=58c b=10 ci=151 co=0 ca=0 - 2 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1081/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/2 ktl=da94 b=10 ci=0 co=0 ca=0 - 3 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1846/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=0 co=0 ca=0 - 4 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=369/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/4 ktl=e0e7 b=10 ci=0 co=0 ca=0 - 5 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=381/1/0 df=4 of=0 ri=1 ql=0 qs=.... kt=0/W/5 ktl=fb2f b=10 ci=0 co=0 ca=0 - 6 c=1480 g=1480 pq=1 pqc=1479 qp=0 dt=1037/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/6 ktl=d2ad b=10 ci=0 co=0 ca=0 - 7 c=1474 g=1474 pq=1 pqc=1473 qp=0 dt=1572/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=0 co=0 ca=0 + 0 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=545/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/0 ktl=ebc3 b=10 ci=0 co=0 ca=0 + 1 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=967/1/0 df=3 of=0 ri=1 ql=0 qs=.... kt=0/W/1 ktl=58c b=10 ci=151 co=0 ca=0 + 2 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=1081/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/2 ktl=da94 b=10 ci=0 co=0 ca=0 + 3 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=1846/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/3 ktl=d1cd b=10 ci=0 co=0 ca=0 + 4 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=369/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/4 ktl=e0e7 b=10 ci=0 co=0 ca=0 + 5 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=381/1/0 df=4 of=0 ri=1 ql=0 qs=.... kt=0/W/5 ktl=fb2f b=10 ci=0 co=0 ca=0 + 6 c=1480 g=1480 pq=1 pgp=1480 qp=0 dt=1037/1/0 df=6 of=0 ri=1 ql=0 qs=.... kt=0/W/6 ktl=d2ad b=10 ci=0 co=0 ca=0 + 7 c=1474 g=1474 pq=1 pgp=1473 qp=0 dt=1572/0/0 df=8 of=0 ri=1 ql=0 qs=.... kt=0/W/7 ktl=cf15 b=10 ci=0 co=0 ca=0 The first section lists the rcu_data structures for rcu_sched, the second for rcu_bh. Note that CONFIG_TREE_PREEMPT_RCU kernels will have an @@ -84,7 +84,7 @@ o "pq" indicates that this CPU has passed through a quiescent state CPU has not yet reported that fact, (2) some other CPU has not yet reported for this grace period, or (3) both. -o "pqc" indicates which grace period the last-observed quiescent +o "pgp" indicates which grace period the last-observed quiescent state for this CPU corresponds to. This is important for handling the race between CPU 0 reporting an extended dynticks-idle quiescent state for CPU 1 and CPU 1 suddenly waking up and diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 7e0282949f8a..7e2f297aeec8 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -159,32 +159,34 @@ static int rcu_gp_in_progress(struct rcu_state *rsp) * Note a quiescent state. Because we do not need to know * how many quiescent states passed, just if there was at least * one since the start of the grace period, this just sets a flag. + * The caller must have disabled preemption. */ void rcu_sched_qs(int cpu) { struct rcu_data *rdp = &per_cpu(rcu_sched_data, cpu); - rdp->passed_quiesc_completed = rdp->gpnum - 1; + rdp->passed_quiesce_gpnum = rdp->gpnum; barrier(); - if (rdp->passed_quiesc == 0) + if (rdp->passed_quiesce == 0) trace_rcu_grace_period("rcu_sched", rdp->gpnum, "cpuqs"); - rdp->passed_quiesc = 1; + rdp->passed_quiesce = 1; } void rcu_bh_qs(int cpu) { struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu); - rdp->passed_quiesc_completed = rdp->gpnum - 1; + rdp->passed_quiesce_gpnum = rdp->gpnum; barrier(); - if (rdp->passed_quiesc == 0) + if (rdp->passed_quiesce == 0) trace_rcu_grace_period("rcu_bh", rdp->gpnum, "cpuqs"); - rdp->passed_quiesc = 1; + rdp->passed_quiesce = 1; } /* * Note a context switch. This is a quiescent state for RCU-sched, * and requires special handling for preemptible RCU. + * The caller must have disabled preemption. */ void rcu_note_context_switch(int cpu) { @@ -694,7 +696,7 @@ static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpustart"); if (rnp->qsmask & rdp->grpmask) { rdp->qs_pending = 1; - rdp->passed_quiesc = 0; + rdp->passed_quiesce = 0; } else rdp->qs_pending = 0; } @@ -1027,7 +1029,7 @@ rcu_report_qs_rnp(unsigned long mask, struct rcu_state *rsp, * based on quiescent states detected in an earlier grace period! */ static void -rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastcomp) +rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long lastgp) { unsigned long flags; unsigned long mask; @@ -1035,17 +1037,15 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp, long las rnp = rdp->mynode; raw_spin_lock_irqsave(&rnp->lock, flags); - if (lastcomp != rnp->completed) { + if (lastgp != rnp->gpnum || rnp->completed == rnp->gpnum) { /* - * Someone beat us to it for this grace period, so leave. - * The race with GP start is resolved by the fact that we - * hold the leaf rcu_node lock, so that the per-CPU bits - * cannot yet be initialized -- so we would simply find our - * CPU's bit already cleared in rcu_report_qs_rnp() if this - * race occurred. + * The grace period in which this quiescent state was + * recorded has ended, so don't report it upwards. + * We will instead need a new quiescent state that lies + * within the current grace period. */ - rdp->passed_quiesc = 0; /* try again later! */ + rdp->passed_quiesce = 0; /* need qs for new gp. */ raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -1089,14 +1089,14 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) * Was there a quiescent state since the beginning of the grace * period? If no, then exit and wait for the next call. */ - if (!rdp->passed_quiesc) + if (!rdp->passed_quiesce) return; /* * Tell RCU we are done (but rcu_report_qs_rdp() will be the * judge of that). */ - rcu_report_qs_rdp(rdp->cpu, rsp, rdp, rdp->passed_quiesc_completed); + rcu_report_qs_rdp(rdp->cpu, rsp, rdp, rdp->passed_quiesce_gpnum); } #ifdef CONFIG_HOTPLUG_CPU @@ -1712,7 +1712,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) check_cpu_stall(rsp, rdp); /* Is the RCU core waiting for a quiescent state from this CPU? */ - if (rdp->qs_pending && !rdp->passed_quiesc) { + if (rdp->qs_pending && !rdp->passed_quiesce) { /* * If force_quiescent_state() coming soon and this CPU @@ -1724,7 +1724,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) ULONG_CMP_LT(ACCESS_ONCE(rsp->jiffies_force_qs) - 1, jiffies)) set_need_resched(); - } else if (rdp->qs_pending && rdp->passed_quiesc) { + } else if (rdp->qs_pending && rdp->passed_quiesce) { rdp->n_rp_report_qs++; return 1; } @@ -1907,7 +1907,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) /* Set up local state, ensuring consistent view of global state. */ raw_spin_lock_irqsave(&rnp->lock, flags); - rdp->passed_quiesc = 0; /* We could be racing with new GP, */ + rdp->passed_quiesce = 0; /* We could be racing with new GP, */ rdp->qs_pending = 1; /* so set up to respond to current GP. */ rdp->beenonline = 1; /* We have now been online. */ rdp->preemptible = preemptible; @@ -1935,7 +1935,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) if (rnp == rdp->mynode) { rdp->gpnum = rnp->completed; /* if GP in progress... */ rdp->completed = rnp->completed; - rdp->passed_quiesc_completed = rnp->completed - 1; + rdp->passed_quiesce_gpnum = rnp->gpnum - 1; trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuonl"); } raw_spin_unlock(&rnp->lock); /* irqs already disabled. */ diff --git a/kernel/rcutree.h b/kernel/rcutree.h index d11a0065321c..51638b68b2dc 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -230,9 +230,9 @@ struct rcu_data { /* in order to detect GP end. */ unsigned long gpnum; /* Highest gp number that this CPU */ /* is aware of having started. */ - unsigned long passed_quiesc_completed; - /* Value of completed at time of qs. */ - bool passed_quiesc; /* User-mode/idle loop etc. */ + unsigned long passed_quiesce_gpnum; + /* gpnum at time of quiescent state. */ + bool passed_quiesce; /* User-mode/idle loop etc. */ bool qs_pending; /* Core waits for quiesc state. */ bool beenonline; /* CPU online at least once. */ bool preemptible; /* Preemptible RCU? */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index bdb2e82f78d3..4bac5a29fb69 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -122,11 +122,11 @@ static void rcu_preempt_qs(int cpu) { struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu); - rdp->passed_quiesc_completed = rdp->gpnum - 1; + rdp->passed_quiesce_gpnum = rdp->gpnum; barrier(); - if (rdp->passed_quiesc == 0) + if (rdp->passed_quiesce == 0) trace_rcu_grace_period("rcu_preempt", rdp->gpnum, "cpuqs"); - rdp->passed_quiesc = 1; + rdp->passed_quiesce = 1; current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; } diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index f328ed1c6e46..9feffa4c0695 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -61,11 +61,11 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) { if (!rdp->beenonline) return; - seq_printf(m, "%3d%cc=%lu g=%lu pq=%d pqc=%lu qp=%d", + seq_printf(m, "%3d%cc=%lu g=%lu pq=%d pgp=%lu qp=%d", rdp->cpu, cpu_is_offline(rdp->cpu) ? '!' : ' ', rdp->completed, rdp->gpnum, - rdp->passed_quiesc, rdp->passed_quiesc_completed, + rdp->passed_quiesce, rdp->passed_quiesce_gpnum, rdp->qs_pending); #ifdef CONFIG_NO_HZ seq_printf(m, " dt=%d/%d/%d df=%lu", @@ -139,7 +139,7 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) rdp->cpu, cpu_is_offline(rdp->cpu) ? "\"N\"" : "\"Y\"", rdp->completed, rdp->gpnum, - rdp->passed_quiesc, rdp->passed_quiesc_completed, + rdp->passed_quiesce, rdp->passed_quiesce_gpnum, rdp->qs_pending); #ifdef CONFIG_NO_HZ seq_printf(m, ",%d,%d,%d,%lu", @@ -170,7 +170,7 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) static int show_rcudata_csv(struct seq_file *m, void *unused) { - seq_puts(m, "\"CPU\",\"Online?\",\"c\",\"g\",\"pq\",\"pqc\",\"pq\","); + seq_puts(m, "\"CPU\",\"Online?\",\"c\",\"g\",\"pq\",\"pgp\",\"pq\","); #ifdef CONFIG_NO_HZ seq_puts(m, "\"dt\",\"dt nesting\",\"dt NMI nesting\",\"df\","); #endif /* #ifdef CONFIG_NO_HZ */ -- cgit v1.3-8-gc7d7 From 3721bc1d3ed9940b17791805b7ee3a4743295d12 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 21 Jul 2011 16:00:17 -0700 Subject: rcu: Allow rcutorture's stat_interval parameter to be changed at runtime When rcutorture is compiled directly into the kernel (instead of separately as a module), it is necessary to specify rcutorture.stat_interval as a kernel command-line parameter, otherwise, the rcu_torture_stats kthread is never started. However, when working with the system after it has booted, it is convenient to be able to change the time between statistic printing, particularly when logged into the console. This commit therefore allows the stat_interval parameter to be changed at runtime. Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 6a30ea3f1d82..ade7771b208a 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -73,7 +73,7 @@ module_param(nreaders, int, 0444); MODULE_PARM_DESC(nreaders, "Number of RCU reader threads"); module_param(nfakewriters, int, 0444); MODULE_PARM_DESC(nfakewriters, "Number of RCU fake writer threads"); -module_param(stat_interval, int, 0444); +module_param(stat_interval, int, 0644); MODULE_PARM_DESC(stat_interval, "Number of seconds between stats printk()s"); module_param(verbose, bool, 0444); MODULE_PARM_DESC(verbose, "Enable verbose debugging printk()s"); -- cgit v1.3-8-gc7d7 From fc0763f53e3ff6a6bfa66934662a3446b9ca6f16 Mon Sep 17 00:00:00 2001 From: "Shi, Alex" Date: Thu, 28 Jul 2011 14:56:12 +0800 Subject: nohz: Remove nohz_cpu_mask RCU no longer uses this global variable, nor does anyone else. This commit therefore removes this variable. This reduces memory footprint and also removes some atomic instructions and memory barriers from the dyntick-idle path. Signed-off-by: Alex Shi Signed-off-by: Paul E. McKenney --- include/linux/sched.h | 1 - kernel/sched.c | 11 ----------- kernel/time/tick-sched.c | 6 ------ 3 files changed, 18 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4ac2c0578e0f..6ee91e20353b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -270,7 +270,6 @@ extern void init_idle_bootup_task(struct task_struct *idle); extern int runqueue_is_locked(int cpu); -extern cpumask_var_t nohz_cpu_mask; #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ) extern void select_nohz_load_balancer(int stop_tick); extern int get_nohz_timer_target(void); diff --git a/kernel/sched.c b/kernel/sched.c index e24cebe0e6cb..3e5525630459 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5979,15 +5979,6 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) ftrace_graph_init_idle_task(idle, cpu); } -/* - * In a system that switches off the HZ timer nohz_cpu_mask - * indicates which cpus entered this state. This is used - * in the rcu update to wait only for active cpus. For system - * which do not switch off the HZ timer nohz_cpu_mask should - * always be CPU_BITS_NONE. - */ -cpumask_var_t nohz_cpu_mask; - /* * Increase the granularity value when there are more CPUs, * because with more CPUs the 'effective latency' as visible @@ -8200,8 +8191,6 @@ void __init sched_init(void) */ current->sched_class = &fair_sched_class; - /* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */ - zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT); #ifdef CONFIG_SMP zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT); #ifdef CONFIG_NO_HZ diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index d5097c44b407..eb98e55196b9 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -139,7 +139,6 @@ static void tick_nohz_update_jiffies(ktime_t now) struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); unsigned long flags; - cpumask_clear_cpu(cpu, nohz_cpu_mask); ts->idle_waketime = now; local_irq_save(flags); @@ -389,9 +388,6 @@ void tick_nohz_stop_sched_tick(int inidle) else expires.tv64 = KTIME_MAX; - if (delta_jiffies > 1) - cpumask_set_cpu(cpu, nohz_cpu_mask); - /* Skip reprogram of event if its not changed */ if (ts->tick_stopped && ktime_equal(expires, dev->next_event)) goto out; @@ -441,7 +437,6 @@ void tick_nohz_stop_sched_tick(int inidle) * softirq. */ tick_do_update_jiffies64(ktime_get()); - cpumask_clear_cpu(cpu, nohz_cpu_mask); } raise_softirq_irqoff(TIMER_SOFTIRQ); out: @@ -524,7 +519,6 @@ void tick_nohz_restart_sched_tick(void) /* Update jiffies first */ select_nohz_load_balancer(0); tick_do_update_jiffies64(now); - cpumask_clear_cpu(cpu, nohz_cpu_mask); #ifndef CONFIG_VIRT_CPU_ACCOUNTING /* -- cgit v1.3-8-gc7d7 From 717f98f0f87c16c1e5768ac8f67c27f3d3535a25 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 5 Aug 2011 22:39:02 -0700 Subject: rcu: Eliminate in_irq() checks in rcu_enter_nohz() The in_irq() check in rcu_enter_nohz() is redundant because if we really are in an interrupt, the attempt to re-enter dyntick-idle mode will invoke rcu_needs_cpu() in any case, which will force the check for RCU callbacks. So this commit removes the check along with the set_need_resched(). Suggested-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 7 ------- 1 file changed, 7 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 7e2f297aeec8..0c6c30dc6a7c 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -373,13 +373,6 @@ void rcu_enter_nohz(void) smp_mb__after_atomic_inc(); /* Force ordering with next sojourn. */ WARN_ON_ONCE(atomic_read(&rdtp->dynticks) & 0x1); local_irq_restore(flags); - - /* If the interrupt queued a callback, get out of dyntick mode. */ - if (in_irq() && - (__get_cpu_var(rcu_sched_data).nxtlist || - __get_cpu_var(rcu_bh_data).nxtlist || - rcu_preempt_needs_cpu(smp_processor_id()))) - set_need_resched(); } /* -- cgit v1.3-8-gc7d7 From 7eb4f4553ceaa6c64da83c8a71d5a991c0188655 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 30 Jul 2011 07:32:48 -0700 Subject: rcu: Make rcu_implicit_dynticks_qs() locals be correct size When the ->dynticks field in the rcu_dynticks structure changed to an atomic_t, its size on 64-bit systems changed from 64 bits to 32 bits. The local variables in rcu_implicit_dynticks_qs() need to change as well, hence this commit. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 0c6c30dc6a7c..ebd18e56947b 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -488,11 +488,11 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp) */ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) { - unsigned long curr; - unsigned long snap; + unsigned int curr; + unsigned int snap; - curr = (unsigned long)atomic_add_return(0, &rdp->dynticks->dynticks); - snap = (unsigned long)rdp->dynticks_snap; + curr = (unsigned int)atomic_add_return(0, &rdp->dynticks->dynticks); + snap = (unsigned int)rdp->dynticks_snap; /* * If the CPU passed through or entered a dynticks idle phase with @@ -502,7 +502,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp) * read-side critical section that started before the beginning * of the current RCU grace period. */ - if ((curr & 0x1) == 0 || ULONG_CMP_GE(curr, snap + 2)) { + if ((curr & 0x1) == 0 || UINT_CMP_GE(curr, snap + 2)) { trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, "dti"); rdp->dynticks_fqs++; return 1; -- cgit v1.3-8-gc7d7 From 6206ab9bab620fc0fbbed30ce20d145b0b3d1840 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 1 Aug 2011 06:22:11 -0700 Subject: rcu: Move __rcu_read_unlock()'s barrier() within if-statement We only need to constrain the compiler if we are actually exiting the top-level RCU read-side critical section. This commit therefore moves the first barrier() cal in __rcu_read_unlock() to inside the "if" statement, thus avoiding needless register flushes for inner rcu_read_unlock() calls. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 14 ++------------ kernel/rcutree_plugin.h | 2 +- 2 files changed, 3 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index af186e260c43..2cf4226ade7e 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -134,16 +134,6 @@ extern void call_rcu_sched(struct rcu_head *head, extern void synchronize_sched(void); -static inline void __rcu_read_lock_bh(void) -{ - local_bh_disable(); -} - -static inline void __rcu_read_unlock_bh(void) -{ - local_bh_enable(); -} - #ifdef CONFIG_PREEMPT_RCU extern void __rcu_read_lock(void); @@ -686,7 +676,7 @@ static inline void rcu_read_unlock(void) */ static inline void rcu_read_lock_bh(void) { - __rcu_read_lock_bh(); + local_bh_disable(); __acquire(RCU_BH); rcu_read_acquire_bh(); } @@ -700,7 +690,7 @@ static inline void rcu_read_unlock_bh(void) { rcu_read_release_bh(); __release(RCU_BH); - __rcu_read_unlock_bh(); + local_bh_enable(); } /** diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 4bac5a29fb69..ed70f6bf4c31 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -415,10 +415,10 @@ void __rcu_read_unlock(void) { struct task_struct *t = current; - barrier(); /* needed if we ever invoke rcu_read_unlock in rcutree.c */ if (t->rcu_read_lock_nesting != 1) --t->rcu_read_lock_nesting; else { + barrier(); /* critical section before exit code. */ t->rcu_read_lock_nesting = INT_MIN; barrier(); /* assign before ->rcu_read_unlock_special load */ if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) -- cgit v1.3-8-gc7d7 From 4627e240dfee4a0a46a58010b1b721b4ded1918f Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 3 Aug 2011 03:34:24 -0700 Subject: rcu: Dump local stack if cannot dump all CPUs' stacks The trigger_all_cpu_backtrace() function is a no-op in architectures that do not define arch_trigger_all_cpu_backtrace. On such architectures, RCU CPU stall warning messages contain no stack trace information, which makes debugging quite difficult. This commit therefore substitutes dump_stack() for architectures that do not define arch_trigger_all_cpu_backtrace, so that at least the local CPU's stack is dumped as part of the RCU CPU stall warning message. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ebd18e56947b..a07bf553e02a 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -583,7 +583,8 @@ static void print_other_cpu_stall(struct rcu_state *rsp) } printk("} (detected by %d, t=%ld jiffies)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start)); - trigger_all_cpu_backtrace(); + if (!trigger_all_cpu_backtrace()) + dump_stack(); /* If so configured, complain about tasks blocking the grace period. */ @@ -604,7 +605,8 @@ static void print_cpu_stall(struct rcu_state *rsp) */ printk(KERN_ERR "INFO: %s detected stall on CPU %d (t=%lu jiffies)\n", rsp->name, smp_processor_id(), jiffies - rsp->gp_start); - trigger_all_cpu_backtrace(); + if (!trigger_all_cpu_backtrace()) + dump_stack(); raw_spin_lock_irqsave(&rnp->lock, flags); if (ULONG_CMP_GE(jiffies, rsp->jiffies_stall)) -- cgit v1.3-8-gc7d7 From 5c51dd7349d4bb26f845f17f85daa168f5fa03f2 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 4 Aug 2011 06:59:03 -0700 Subject: rcu: Prevent early boot set_need_resched() from __rcu_pending() There isn't a whole lot of point in poking the scheduler before there are other tasks to switch to. This commit therefore adds a check for rcu_scheduler_fully_active in __rcu_pending() to suppress any pre-scheduler calls to set_need_resched(). The downside of this approach is additional runtime overhead in a reasonably hot code path. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a07bf553e02a..0051dbf6958e 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1707,7 +1707,8 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) check_cpu_stall(rsp, rdp); /* Is the RCU core waiting for a quiescent state from this CPU? */ - if (rdp->qs_pending && !rdp->passed_quiesce) { + if (rcu_scheduler_fully_active && + rdp->qs_pending && !rdp->passed_quiesce) { /* * If force_quiescent_state() coming soon and this CPU -- cgit v1.3-8-gc7d7 From 82e78d80fc392ac7e98326bc8beeb8a679913ffd Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 4 Aug 2011 07:55:34 -0700 Subject: rcu: Simplify unboosting checks Commit 7765be (Fix RCU_BOOST race handling current->rcu_read_unlock_special) introduced a new ->rcu_boosted field in the task structure. This is redundant because the existing ->rcu_boost_mutex will be non-NULL at any time that ->rcu_boosted is nonzero. Therefore, this commit removes ->rcu_boosted and tests ->rcu_boost_mutex instead. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- include/linux/sched.h | 3 --- kernel/rcutree_plugin.h | 20 ++++++++++---------- 2 files changed, 10 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6ee91e20353b..acca43560805 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1259,9 +1259,6 @@ struct task_struct { #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; char rcu_read_unlock_special; -#if defined(CONFIG_RCU_BOOST) && defined(CONFIG_TREE_PREEMPT_RCU) - int rcu_boosted; -#endif /* #if defined(CONFIG_RCU_BOOST) && defined(CONFIG_TREE_PREEMPT_RCU) */ struct list_head rcu_node_entry; #endif /* #ifdef CONFIG_PREEMPT_RCU */ #ifdef CONFIG_TREE_PREEMPT_RCU diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index ed70f6bf4c31..eeb38ee8ebba 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -306,6 +306,9 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) int empty_exp; unsigned long flags; struct list_head *np; +#ifdef CONFIG_RCU_BOOST + struct rt_mutex *rbmp = NULL; +#endif /* #ifdef CONFIG_RCU_BOOST */ struct rcu_node *rnp; int special; @@ -351,6 +354,7 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */ np = rcu_next_node_entry(t, rnp); list_del_init(&t->rcu_node_entry); + t->rcu_blocked_node = NULL; trace_rcu_unlock_preempted_task("rcu_preempt", rnp->gpnum, t->pid); if (&t->rcu_node_entry == rnp->gp_tasks) @@ -360,13 +364,12 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) #ifdef CONFIG_RCU_BOOST if (&t->rcu_node_entry == rnp->boost_tasks) rnp->boost_tasks = np; - /* Snapshot and clear ->rcu_boosted with rcu_node lock held. */ - if (t->rcu_boosted) { - special |= RCU_READ_UNLOCK_BOOSTED; - t->rcu_boosted = 0; + /* Snapshot/clear ->rcu_boost_mutex with rcu_node lock held. */ + if (t->rcu_boost_mutex) { + rbmp = t->rcu_boost_mutex; + t->rcu_boost_mutex = NULL; } #endif /* #ifdef CONFIG_RCU_BOOST */ - t->rcu_blocked_node = NULL; /* * If this was the last task on the current list, and if @@ -387,10 +390,8 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) #ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ - if (special & RCU_READ_UNLOCK_BOOSTED) { - rt_mutex_unlock(t->rcu_boost_mutex); - t->rcu_boost_mutex = NULL; - } + if (rbmp) + rt_mutex_unlock(rbmp); #endif /* #ifdef CONFIG_RCU_BOOST */ /* @@ -1206,7 +1207,6 @@ static int rcu_boost(struct rcu_node *rnp) t = container_of(tb, struct task_struct, rcu_node_entry); rt_mutex_init_proxy_locked(&mtx, t); t->rcu_boost_mutex = &mtx; - t->rcu_boosted = 1; raw_spin_unlock_irqrestore(&rnp->lock, flags); rt_mutex_lock(&mtx); /* Side effect: boosts task t's priority. */ rt_mutex_unlock(&mtx); /* Keep lockdep happy. */ -- cgit v1.3-8-gc7d7 From 037067a1b6f9a70f862f3ed9d59fe28b7cd55ac4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 7 Aug 2011 20:26:31 -0700 Subject: rcu: Prohibit grace periods during early boot Greater use of RCU during early boot (before the scheduler is operating) is causing RCU to attempt to start grace periods during that time, which in turn is resulting in both RCU and the callback functions attempting to use the scheduler before it is ready. This commit prevents these problems by prohibiting RCU grace periods until after the scheduler has spawned the first non-idle task. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 0051dbf6958e..9970116163ba 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -838,8 +838,11 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) struct rcu_data *rdp = this_cpu_ptr(rsp->rda); struct rcu_node *rnp = rcu_get_root(rsp); - if (!cpu_needs_another_gp(rsp, rdp) || rsp->fqs_active) { - if (cpu_needs_another_gp(rsp, rdp)) + if (!rcu_scheduler_fully_active || + !cpu_needs_another_gp(rsp, rdp) || + rsp->fqs_active) { + if (rcu_scheduler_fully_active && + cpu_needs_another_gp(rsp, rdp)) rsp->fqs_need_gp = 1; if (rnp->completed == rsp->completed) { raw_spin_unlock_irqrestore(&rnp->lock, flags); -- cgit v1.3-8-gc7d7 From 9bc8b5586f94be6391458074ecbba8827ba8ba9d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 13 Aug 2011 13:31:47 -0700 Subject: rcu: Suppress NMI backtraces when stall ends before dump It is possible for an RCU CPU stall to end just as it is detected, in which case the current code will uselessly dump all CPU's stacks. This commit therefore checks for this condition and refrains from sending needless NMIs. And yes, the stall might also end just after we checked all CPUs and tasks, but in that case we would at least have given some clue as to which CPU/task was at fault. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 13 +++++++++---- kernel/rcutree.h | 2 +- kernel/rcutree_plugin.h | 13 +++++++++---- 3 files changed, 19 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 9970116163ba..ade788320dd6 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -545,6 +545,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp) int cpu; long delta; unsigned long flags; + int ndetected; struct rcu_node *rnp = rcu_get_root(rsp); /* Only let one CPU complain about others per time interval. */ @@ -561,7 +562,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp) * Now rat on any tasks that got kicked up to the root rcu_node * due to CPU offlining. */ - rcu_print_task_stall(rnp); + ndetected = rcu_print_task_stall(rnp); raw_spin_unlock_irqrestore(&rnp->lock, flags); /* @@ -573,17 +574,21 @@ static void print_other_cpu_stall(struct rcu_state *rsp) rsp->name); rcu_for_each_leaf_node(rsp, rnp) { raw_spin_lock_irqsave(&rnp->lock, flags); - rcu_print_task_stall(rnp); + ndetected += rcu_print_task_stall(rnp); raw_spin_unlock_irqrestore(&rnp->lock, flags); if (rnp->qsmask == 0) continue; for (cpu = 0; cpu <= rnp->grphi - rnp->grplo; cpu++) - if (rnp->qsmask & (1UL << cpu)) + if (rnp->qsmask & (1UL << cpu)) { printk(" %d", rnp->grplo + cpu); + ndetected++; + } } printk("} (detected by %d, t=%ld jiffies)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start)); - if (!trigger_all_cpu_backtrace()) + if (ndetected == 0) + printk(KERN_ERR "INFO: Stall ended before state dump start\n"); + else if (!trigger_all_cpu_backtrace()) dump_stack(); /* If so configured, complain about tasks blocking the grace period. */ diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 51638b68b2dc..f509f728f9fb 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -438,7 +438,7 @@ static void rcu_report_unblock_qs_rnp(struct rcu_node *rnp, static void rcu_stop_cpu_kthread(int cpu); #endif /* #ifdef CONFIG_HOTPLUG_CPU */ static void rcu_print_detail_task_stall(struct rcu_state *rsp); -static void rcu_print_task_stall(struct rcu_node *rnp); +static int rcu_print_task_stall(struct rcu_node *rnp); static void rcu_preempt_stall_reset(void); static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); #ifdef CONFIG_HOTPLUG_CPU diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index eeb38ee8ebba..d3127e8764cb 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -483,16 +483,20 @@ static void rcu_print_detail_task_stall(struct rcu_state *rsp) * Scan the current list of tasks blocked within RCU read-side critical * sections, printing out the tid of each. */ -static void rcu_print_task_stall(struct rcu_node *rnp) +static int rcu_print_task_stall(struct rcu_node *rnp) { struct task_struct *t; + int ndetected = 0; if (!rcu_preempt_blocked_readers_cgp(rnp)) - return; + return 0; t = list_entry(rnp->gp_tasks, struct task_struct, rcu_node_entry); - list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) + list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) { printk(" P%d", t->pid); + ndetected++; + } + return ndetected; } /* @@ -976,8 +980,9 @@ static void rcu_print_detail_task_stall(struct rcu_state *rsp) * Because preemptible RCU does not exist, we never have to check for * tasks blocked within RCU read-side critical sections. */ -static void rcu_print_task_stall(struct rcu_node *rnp) +static int rcu_print_task_stall(struct rcu_node *rnp) { + return 0; } /* -- cgit v1.3-8-gc7d7 From 06ae115a1d551cd952d80df06eaf8b5153351875 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 14 Aug 2011 15:56:54 -0700 Subject: rcu: Avoid having just-onlined CPU resched itself when RCU is idle CPUs set rdp->qs_pending when coming online to resolve races with grace-period start. However, this means that if RCU is idle, the just-onlined CPU might needlessly send itself resched IPIs. Adjust the online-CPU initialization to avoid this, and also to correctly cause the CPU to respond to the current grace period if needed. Signed-off-by: Paul E. McKenney Tested-by: Josh Boyer Tested-by: Christian Hoffmann --- kernel/rcutree.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ade788320dd6..c95fa89ffef9 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1911,8 +1911,6 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) /* Set up local state, ensuring consistent view of global state. */ raw_spin_lock_irqsave(&rnp->lock, flags); - rdp->passed_quiesce = 0; /* We could be racing with new GP, */ - rdp->qs_pending = 1; /* so set up to respond to current GP. */ rdp->beenonline = 1; /* We have now been online. */ rdp->preemptible = preemptible; rdp->qlen_last_fqs_check = 0; @@ -1937,8 +1935,15 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) rnp->qsmaskinit |= mask; mask = rnp->grpmask; if (rnp == rdp->mynode) { - rdp->gpnum = rnp->completed; /* if GP in progress... */ + /* + * If there is a grace period in progress, we will + * set up to wait for it next time we run the + * RCU core code. + */ + rdp->gpnum = rnp->completed; rdp->completed = rnp->completed; + rdp->passed_quiesce = 0; + rdp->qs_pending = 0; rdp->passed_quiesce_gpnum = rnp->gpnum - 1; trace_rcu_grace_period(rsp->name, rdp->gpnum, "cpuonl"); } -- cgit v1.3-8-gc7d7 From 5342e269b2b58ee0b0b4168a94087faaa60d0567 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 16 Aug 2011 17:46:46 -0700 Subject: rcu: Permit rt_mutex_unlock() with irqs disabled Create a separate lockdep class for the rt_mutex used for RCU priority boosting and enable use of rt_mutex_lock() with irqs disabled. This prevents RCU priority boosting from falling prey to deadlocks when someone begins an RCU read-side critical section in preemptible state, but releases it with an irq-disabled lock held. Unfortunately, the scheduler's runqueue and priority-inheritance locks still must either completely enclose or be completely enclosed by any overlapping RCU read-side critical section. This version removes a redundant local_irq_restore() noted by Yong Zhang. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree_plugin.h | 5 +++++ kernel/rtmutex.c | 8 ++++++++ 2 files changed, 13 insertions(+) (limited to 'kernel') diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index d3127e8764cb..28422767d854 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1149,6 +1149,8 @@ static void rcu_initiate_boost_trace(struct rcu_node *rnp) #endif /* #else #ifdef CONFIG_RCU_TRACE */ +static struct lock_class_key rcu_boost_class; + /* * Carry out RCU priority boosting on the task indicated by ->exp_tasks * or ->boost_tasks, advancing the pointer to the next task in the @@ -1211,6 +1213,9 @@ static int rcu_boost(struct rcu_node *rnp) */ t = container_of(tb, struct task_struct, rcu_node_entry); rt_mutex_init_proxy_locked(&mtx, t); + /* Avoid lockdep false positives. This rt_mutex is its own thing. */ + lockdep_set_class_and_name(&mtx.wait_lock, &rcu_boost_class, + "rcu_boost_mutex"); t->rcu_boost_mutex = &mtx; raw_spin_unlock_irqrestore(&rnp->lock, flags); rt_mutex_lock(&mtx); /* Side effect: boosts task t's priority. */ diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 255e1662acdb..5e8d9cce7470 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -579,6 +579,7 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, struct rt_mutex_waiter *waiter) { int ret = 0; + int was_disabled; for (;;) { /* Try to acquire the lock: */ @@ -601,10 +602,17 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, raw_spin_unlock(&lock->wait_lock); + was_disabled = irqs_disabled(); + if (was_disabled) + local_irq_enable(); + debug_rt_mutex_print_deadlock(waiter); schedule_rt_mutex(lock); + if (was_disabled) + local_irq_disable(); + raw_spin_lock(&lock->wait_lock); set_current_state(state); } -- cgit v1.3-8-gc7d7 From 93898fb1a395d2a5a53db238c68036da2f8c64d1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 17 Aug 2011 12:39:34 -0700 Subject: rcu: Make rcu_torture_fqs() exit loops at end of test The rcu_torture_fqs() function can prevent the rcutorture tests from completing, resulting in a hang. This commit therefore ensures that rcu_torture_fqs() will exit its inner loops at the end of the test, and also applies the newish ULONG_CMP_LT() macro to time comparisons. Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index ade7771b208a..2431d576e9ca 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -741,7 +741,7 @@ static int rcu_torture_boost(void *arg) do { /* Wait for the next test interval. */ oldstarttime = boost_starttime; - while (jiffies - oldstarttime > ULONG_MAX / 2) { + while (ULONG_CMP_LT(jiffies, oldstarttime)) { schedule_timeout_uninterruptible(1); rcu_stutter_wait("rcu_torture_boost"); if (kthread_should_stop() || @@ -752,7 +752,7 @@ static int rcu_torture_boost(void *arg) /* Do one boost-test interval. */ endtime = oldstarttime + test_boost_duration * HZ; call_rcu_time = jiffies; - while (jiffies - endtime > ULONG_MAX / 2) { + while (ULONG_CMP_LT(jiffies, endtime)) { /* If we don't have a callback in flight, post one. */ if (!rbi.inflight) { smp_mb(); /* RCU core before ->inflight = 1. */ @@ -818,11 +818,13 @@ rcu_torture_fqs(void *arg) VERBOSE_PRINTK_STRING("rcu_torture_fqs task started"); do { fqs_resume_time = jiffies + fqs_stutter * HZ; - while (jiffies - fqs_resume_time > LONG_MAX) { + while (ULONG_CMP_LT(jiffies, fqs_resume_time) && + !kthread_should_stop()) { schedule_timeout_interruptible(1); } fqs_burst_remaining = fqs_duration; - while (fqs_burst_remaining > 0) { + while (fqs_burst_remaining > 0 && + !kthread_should_stop()) { cur_ops->fqs(); udelay(fqs_holdoff); fqs_burst_remaining -= fqs_holdoff; -- cgit v1.3-8-gc7d7 From ab8f11e5f6655861ad4758a7da76b2fc0e0dcc98 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 18 Aug 2011 09:30:32 -0700 Subject: rcu: Make rcu_torture_boost() exit loops at end of test One of the loops in rcu_torture_boost() fails to check kthread_should_stop(), and thus might be slowing or even stopping completion of rcutorture tests at rmmod time. This commit adds the kthread_should_stop() check to the offending loop. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 2431d576e9ca..764825c2685c 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -779,7 +779,8 @@ static int rcu_torture_boost(void *arg) * interval. Besides, we are running at RT priority, * so delays should be relatively rare. */ - while (oldstarttime == boost_starttime) { + while (oldstarttime == boost_starttime && + !kthread_should_stop()) { if (mutex_trylock(&boost_mutex)) { boost_starttime = jiffies + test_boost_interval * HZ; -- cgit v1.3-8-gc7d7 From 5b61b0baa9e80289c53413e573befc5790a04ac7 Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Fri, 19 Aug 2011 11:39:11 -0700 Subject: rcu: Wire up RCU_BOOST_PRIO for rcutree RCU boost threads start life at RCU_BOOST_PRIO, while others remain at RCU_KTHREAD_PRIO. While here, change thread names to match other kthreads, and adjust rcu_yield() to not override the priority set by the user. This last change sets the stage for runtime changes to priority in the -rt tree. Signed-off-by: Mike Galbraith Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 2 -- kernel/rcutree_plugin.h | 20 +++++++++++++++----- 2 files changed, 15 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index c95fa89ffef9..8455043c9250 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -131,8 +131,6 @@ static void rcu_node_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu); static void invoke_rcu_core(void); static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp); -#define RCU_KTHREAD_PRIO 1 /* RT priority for per-CPU kthreads. */ - /* * Track the rcutorture test sequence number and the update version * number within a given test. The rcutorture_testseq is incremented diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 28422767d854..b4cbe5bf2326 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -27,6 +27,14 @@ #include #include +#define RCU_KTHREAD_PRIO 1 + +#ifdef CONFIG_RCU_BOOST +#define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO +#else +#define RCU_BOOST_PRIO RCU_KTHREAD_PRIO +#endif + /* * Check the RCU kernel configuration parameters and print informative * messages about anything out of the ordinary. If you like #ifdef, you @@ -1364,13 +1372,13 @@ static int __cpuinit rcu_spawn_one_boost_kthread(struct rcu_state *rsp, if (rnp->boost_kthread_task != NULL) return 0; t = kthread_create(rcu_boost_kthread, (void *)rnp, - "rcub%d", rnp_index); + "rcub/%d", rnp_index); if (IS_ERR(t)) return PTR_ERR(t); raw_spin_lock_irqsave(&rnp->lock, flags); rnp->boost_kthread_task = t; raw_spin_unlock_irqrestore(&rnp->lock, flags); - sp.sched_priority = RCU_KTHREAD_PRIO; + sp.sched_priority = RCU_BOOST_PRIO; sched_setscheduler_nocheck(t, SCHED_FIFO, &sp); wake_up_process(t); /* get to TASK_INTERRUPTIBLE quickly. */ return 0; @@ -1465,6 +1473,7 @@ static void rcu_yield(void (*f)(unsigned long), unsigned long arg) { struct sched_param sp; struct timer_list yield_timer; + int prio = current->rt_priority; setup_timer_on_stack(&yield_timer, f, arg); mod_timer(&yield_timer, jiffies + 2); @@ -1472,7 +1481,8 @@ static void rcu_yield(void (*f)(unsigned long), unsigned long arg) sched_setscheduler_nocheck(current, SCHED_NORMAL, &sp); set_user_nice(current, 19); schedule(); - sp.sched_priority = RCU_KTHREAD_PRIO; + set_user_nice(current, 0); + sp.sched_priority = prio; sched_setscheduler_nocheck(current, SCHED_FIFO, &sp); del_timer(&yield_timer); } @@ -1591,7 +1601,7 @@ static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu) t = kthread_create_on_node(rcu_cpu_kthread, (void *)(long)cpu, cpu_to_node(cpu), - "rcuc%d", cpu); + "rcuc/%d", cpu); if (IS_ERR(t)) return PTR_ERR(t); if (cpu_online(cpu)) @@ -1700,7 +1710,7 @@ static int __cpuinit rcu_spawn_one_node_kthread(struct rcu_state *rsp, return 0; if (rnp->node_kthread_task == NULL) { t = kthread_create(rcu_node_kthread, (void *)rnp, - "rcun%d", rnp_index); + "rcun/%d", rnp_index); if (IS_ERR(t)) return PTR_ERR(t); raw_spin_lock_irqsave(&rnp->lock, flags); -- cgit v1.3-8-gc7d7 From e90c53d3e238dd0b7b02964370e8fece1778df96 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 20 Aug 2011 18:29:32 -0700 Subject: rcu: Remove rcu_needs_cpu_flush() to avoid false quiescent states The purpose of rcu_needs_cpu_flush() was to iterate on pushing the current grace period in order to help the current CPU enter dyntick-idle mode. However, this can result in failures if the CPU starts entering dyntick-idle mode, but then backs out. In this case, the call to rcu_pending() from rcu_needs_cpu_flush() might end up announcing a non-existing quiescent state. This commit therefore removes rcu_needs_cpu_flush() in favor of letting the dyntick-idle machinery at the end of the softirq handler push the loop along via its call to rcu_pending(). Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 3 --- kernel/rcutree.h | 1 - kernel/rcutree_plugin.h | 25 ------------------------- 3 files changed, 29 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 8455043c9250..e75df0c93abd 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1528,9 +1528,6 @@ static void rcu_process_callbacks(struct softirq_action *unused) &__get_cpu_var(rcu_sched_data)); __rcu_process_callbacks(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); rcu_preempt_process_callbacks(); - - /* If we are last CPU on way to dyntick-idle mode, accelerate it. */ - rcu_needs_cpu_flush(); trace_rcu_utilization("End RCU core"); } diff --git a/kernel/rcutree.h b/kernel/rcutree.h index f509f728f9fb..849ce9ec51fe 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -458,7 +458,6 @@ static int rcu_preempt_needs_cpu(int cpu); static void __cpuinit rcu_preempt_init_percpu_data(int cpu); static void rcu_preempt_send_cbs_to_online(void); static void __init __rcu_init_preempt(void); -static void rcu_needs_cpu_flush(void); static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags); static void rcu_preempt_boost_start_gp(struct rcu_node *rnp); static void invoke_rcu_callbacks_kthread(void); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index b4cbe5bf2326..4b9b9f8a4184 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1948,15 +1948,6 @@ int rcu_needs_cpu(int cpu) return rcu_needs_cpu_quick_check(cpu); } -/* - * Check to see if we need to continue a callback-flush operations to - * allow the last CPU to enter dyntick-idle mode. But fast dyntick-idle - * entry is not configured, so we never do need to. - */ -static void rcu_needs_cpu_flush(void) -{ -} - #else /* #if !defined(CONFIG_RCU_FAST_NO_HZ) */ #define RCU_NEEDS_CPU_FLUSHES 5 @@ -2032,20 +2023,4 @@ int rcu_needs_cpu(int cpu) return c; } -/* - * Check to see if we need to continue a callback-flush operations to - * allow the last CPU to enter dyntick-idle mode. - */ -static void rcu_needs_cpu_flush(void) -{ - int cpu = smp_processor_id(); - unsigned long flags; - - if (per_cpu(rcu_dyntick_drain, cpu) <= 0) - return; - local_irq_save(flags); - (void)rcu_needs_cpu(cpu); - local_irq_restore(flags); -} - #endif /* #else #if !defined(CONFIG_RCU_FAST_NO_HZ) */ -- cgit v1.3-8-gc7d7 From afe24b122eb6edb5f1cb942570ac8d766105c7fc Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 24 Aug 2011 16:52:09 -0700 Subject: rcu: Move propagation of ->completed from rcu_start_gp() to rcu_report_qs_rsp() It is possible for the CPU that noted the end of the prior grace period to not need a new one, and therefore to decide to propagate ->completed throughout the rcu_node tree without starting another grace period. However, in so doing, it releases the root rcu_node structure's lock, which can allow some other CPU to start another grace period. The first CPU will be propagating ->completed in parallel with the second CPU initializing the rcu_node tree for the new grace period. In theory this is harmless, but in practice we need to keep things simple. This commit therefore moves the propagation of ->completed to rcu_report_qs_rsp(), and refrains from marking the old grace period as having been completed until it has finished doing this. This prevents anyone from starting a new grace period concurrently with marking the old grace period as having been completed. Of course, the optimization where a CPU needing a new grace period doesn't bother marking the old one completed is still in effect: In that case, the marking happens implicitly as part of initializing the new grace period. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 71 ++++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 51 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index e75df0c93abd..e234eb92a177 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -842,28 +842,24 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) struct rcu_node *rnp = rcu_get_root(rsp); if (!rcu_scheduler_fully_active || - !cpu_needs_another_gp(rsp, rdp) || - rsp->fqs_active) { - if (rcu_scheduler_fully_active && - cpu_needs_another_gp(rsp, rdp)) - rsp->fqs_need_gp = 1; - if (rnp->completed == rsp->completed) { - raw_spin_unlock_irqrestore(&rnp->lock, flags); - return; - } - raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + !cpu_needs_another_gp(rsp, rdp)) { + /* + * Either the scheduler hasn't yet spawned the first + * non-idle task or this CPU does not need another + * grace period. Either way, don't start a new grace + * period. + */ + raw_spin_unlock_irqrestore(&rnp->lock, flags); + return; + } + if (rsp->fqs_active) { /* - * Propagate new ->completed value to rcu_node structures - * so that other CPUs don't have to wait until the start - * of the next grace period to process their callbacks. + * This CPU needs a grace period, but force_quiescent_state() + * is running. Tell it to start one on this CPU's behalf. */ - rcu_for_each_node_breadth_first(rsp, rnp) { - raw_spin_lock(&rnp->lock); /* irqs already disabled. */ - rnp->completed = rsp->completed; - raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ - } - local_irq_restore(flags); + rsp->fqs_need_gp = 1; + raw_spin_unlock_irqrestore(&rnp->lock, flags); return; } @@ -947,6 +943,8 @@ static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) __releases(rcu_get_root(rsp)->lock) { unsigned long gp_duration; + struct rcu_node *rnp = rcu_get_root(rsp); + struct rcu_data *rdp = this_cpu_ptr(rsp->rda); WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); @@ -958,7 +956,40 @@ static void rcu_report_qs_rsp(struct rcu_state *rsp, unsigned long flags) gp_duration = jiffies - rsp->gp_start; if (gp_duration > rsp->gp_max) rsp->gp_max = gp_duration; - rsp->completed = rsp->gpnum; + + /* + * We know the grace period is complete, but to everyone else + * it appears to still be ongoing. But it is also the case + * that to everyone else it looks like there is nothing that + * they can do to advance the grace period. It is therefore + * safe for us to drop the lock in order to mark the grace + * period as completed in all of the rcu_node structures. + * + * But if this CPU needs another grace period, it will take + * care of this while initializing the next grace period. + * We use RCU_WAIT_TAIL instead of the usual RCU_DONE_TAIL + * because the callbacks have not yet been advanced: Those + * callbacks are waiting on the grace period that just now + * completed. + */ + if (*rdp->nxttail[RCU_WAIT_TAIL] == NULL) { + raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + + /* + * Propagate new ->completed value to rcu_node structures + * so that other CPUs don't have to wait until the start + * of the next grace period to process their callbacks. + */ + rcu_for_each_node_breadth_first(rsp, rnp) { + raw_spin_lock(&rnp->lock); /* irqs already disabled. */ + rnp->completed = rsp->gpnum; + raw_spin_unlock(&rnp->lock); /* irqs remain disabled. */ + } + rnp = rcu_get_root(rsp); + raw_spin_lock(&rnp->lock); /* irqs already disabled. */ + } + + rsp->completed = rsp->gpnum; /* Declare the grace period complete. */ trace_rcu_grace_period(rsp->name, rsp->completed, "end"); rsp->signaled = RCU_GP_IDLE; rcu_start_gp(rsp, flags); /* releases root node's rnp->lock. */ -- cgit v1.3-8-gc7d7 From 2a5306cc5f383b0e7414c75e458111afd4a563a4 Mon Sep 17 00:00:00 2001 From: Ming Lei Date: Thu, 29 Sep 2011 22:07:23 +0200 Subject: PM / Tracing: build rpm-traces.c only if CONFIG_PM_RUNTIME is set Do not build kernel/trace/rpm-traces.c if CONFIG_PM_RUNTIME is not set, which avoids a build failure. [rjw: Added the changelog and modified the subject slightly.] Signed-off-by: Ming Lei Signed-off-by: Rafael J. Wysocki --- kernel/trace/Makefile | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 56bdab5b3793..f49405f842f4 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -53,7 +53,9 @@ endif obj-$(CONFIG_EVENT_TRACING) += trace_events_filter.o obj-$(CONFIG_KPROBE_EVENT) += trace_kprobe.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o +ifeq ($(CONFIG_PM_RUNTIME),y) obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o +endif ifeq ($(CONFIG_TRACING),y) obj-$(CONFIG_KGDB_KDB) += trace_kdb.o endif -- cgit v1.3-8-gc7d7 From d178bc3a708f39cbfefc3fab37032d3f2511b4ec Mon Sep 17 00:00:00 2001 From: Serge Hallyn Date: Mon, 26 Sep 2011 10:45:18 -0500 Subject: user namespace: usb: make usb urbs user namespace aware (v2) Add to the dev_state and alloc_async structures the user namespace corresponding to the uid and euid. Pass these to kill_pid_info_as_uid(), which can then implement a proper, user-namespace-aware uid check. Changelog: Sep 20: Per Oleg's suggestion: Instead of caching and passing user namespace, uid, and euid each separately, pass a struct cred. Sep 26: Address Alan Stern's comments: don't define a struct cred at usbdev_open(), and take and put a cred at async_completed() to ensure it lasts for the duration of kill_pid_info_as_cred(). Signed-off-by: Serge Hallyn Cc: Oleg Nesterov Cc: "Eric W. Biederman" Cc: Tejun Heo Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/devio.c | 30 +++++++++++++----------------- include/linux/sched.h | 3 ++- kernel/signal.c | 24 ++++++++++++++++-------- 3 files changed, 31 insertions(+), 26 deletions(-) (limited to 'kernel') diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index 0ca54e22d319..e3beaf229ee3 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -46,6 +46,7 @@ #include #include #include +#include #include #include #include @@ -68,7 +69,7 @@ struct dev_state { wait_queue_head_t wait; /* wake up if a request completed */ unsigned int discsignr; struct pid *disc_pid; - uid_t disc_uid, disc_euid; + const struct cred *cred; void __user *disccontext; unsigned long ifclaimed; u32 secid; @@ -79,7 +80,7 @@ struct async { struct list_head asynclist; struct dev_state *ps; struct pid *pid; - uid_t uid, euid; + const struct cred *cred; unsigned int signr; unsigned int ifnum; void __user *userbuffer; @@ -248,6 +249,7 @@ static struct async *alloc_async(unsigned int numisoframes) static void free_async(struct async *as) { put_pid(as->pid); + put_cred(as->cred); kfree(as->urb->transfer_buffer); kfree(as->urb->setup_packet); usb_free_urb(as->urb); @@ -393,9 +395,8 @@ static void async_completed(struct urb *urb) struct dev_state *ps = as->ps; struct siginfo sinfo; struct pid *pid = NULL; - uid_t uid = 0; - uid_t euid = 0; u32 secid = 0; + const struct cred *cred = NULL; int signr; spin_lock(&ps->lock); @@ -408,8 +409,7 @@ static void async_completed(struct urb *urb) sinfo.si_code = SI_ASYNCIO; sinfo.si_addr = as->userurb; pid = get_pid(as->pid); - uid = as->uid; - euid = as->euid; + cred = get_cred(as->cred); secid = as->secid; } snoop(&urb->dev->dev, "urb complete\n"); @@ -423,9 +423,9 @@ static void async_completed(struct urb *urb) spin_unlock(&ps->lock); if (signr) { - kill_pid_info_as_uid(sinfo.si_signo, &sinfo, pid, uid, - euid, secid); + kill_pid_info_as_cred(sinfo.si_signo, &sinfo, pid, cred, secid); put_pid(pid); + put_cred(cred); } wake_up(&ps->wait); @@ -672,7 +672,6 @@ static int usbdev_open(struct inode *inode, struct file *file) { struct usb_device *dev = NULL; struct dev_state *ps; - const struct cred *cred = current_cred(); int ret; ret = -ENOMEM; @@ -722,8 +721,7 @@ static int usbdev_open(struct inode *inode, struct file *file) init_waitqueue_head(&ps->wait); ps->discsignr = 0; ps->disc_pid = get_pid(task_pid(current)); - ps->disc_uid = cred->uid; - ps->disc_euid = cred->euid; + ps->cred = get_current_cred(); ps->disccontext = NULL; ps->ifclaimed = 0; security_task_getsecid(current, &ps->secid); @@ -765,6 +763,7 @@ static int usbdev_release(struct inode *inode, struct file *file) usb_unlock_device(dev); usb_put_dev(dev); put_pid(ps->disc_pid); + put_cred(ps->cred); as = async_getcompleted(ps); while (as) { @@ -1065,7 +1064,6 @@ static int proc_do_submiturb(struct dev_state *ps, struct usbdevfs_urb *uurb, struct usb_host_endpoint *ep; struct async *as; struct usb_ctrlrequest *dr = NULL; - const struct cred *cred = current_cred(); unsigned int u, totlen, isofrmlen; int ret, ifnum = -1; int is_in; @@ -1279,8 +1277,7 @@ static int proc_do_submiturb(struct dev_state *ps, struct usbdevfs_urb *uurb, as->signr = uurb->signr; as->ifnum = ifnum; as->pid = get_pid(task_pid(current)); - as->uid = cred->uid; - as->euid = cred->euid; + as->cred = get_current_cred(); security_task_getsecid(current, &as->secid); if (!is_in && uurb->buffer_length > 0) { if (copy_from_user(as->urb->transfer_buffer, uurb->buffer, @@ -1998,9 +1995,8 @@ static void usbdev_remove(struct usb_device *udev) sinfo.si_errno = EPIPE; sinfo.si_code = SI_ASYNCIO; sinfo.si_addr = ps->disccontext; - kill_pid_info_as_uid(ps->discsignr, &sinfo, - ps->disc_pid, ps->disc_uid, - ps->disc_euid, ps->secid); + kill_pid_info_as_cred(ps->discsignr, &sinfo, + ps->disc_pid, ps->cred, ps->secid); } } } diff --git a/include/linux/sched.h b/include/linux/sched.h index 4ac2c0578e0f..57ddb5283d9f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2166,7 +2166,8 @@ extern int force_sigsegv(int, struct task_struct *); extern int force_sig_info(int, struct siginfo *, struct task_struct *); extern int __kill_pgrp_info(int sig, struct siginfo *info, struct pid *pgrp); extern int kill_pid_info(int sig, struct siginfo *info, struct pid *pid); -extern int kill_pid_info_as_uid(int, struct siginfo *, struct pid *, uid_t, uid_t, u32); +extern int kill_pid_info_as_cred(int, struct siginfo *, struct pid *, + const struct cred *, u32); extern int kill_pgrp(struct pid *pid, int sig, int priv); extern int kill_pid(struct pid *pid, int sig, int priv); extern int kill_proc_info(int, struct siginfo *, pid_t); diff --git a/kernel/signal.c b/kernel/signal.c index 291c9700be75..d252be2d3de5 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1344,13 +1344,24 @@ int kill_proc_info(int sig, struct siginfo *info, pid_t pid) return error; } +static int kill_as_cred_perm(const struct cred *cred, + struct task_struct *target) +{ + const struct cred *pcred = __task_cred(target); + if (cred->user_ns != pcred->user_ns) + return 0; + if (cred->euid != pcred->suid && cred->euid != pcred->uid && + cred->uid != pcred->suid && cred->uid != pcred->uid) + return 0; + return 1; +} + /* like kill_pid_info(), but doesn't use uid/euid of "current" */ -int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid, - uid_t uid, uid_t euid, u32 secid) +int kill_pid_info_as_cred(int sig, struct siginfo *info, struct pid *pid, + const struct cred *cred, u32 secid) { int ret = -EINVAL; struct task_struct *p; - const struct cred *pcred; unsigned long flags; if (!valid_signal(sig)) @@ -1362,10 +1373,7 @@ int kill_pid_info_as_uid(int sig, struct siginfo *info, struct pid *pid, ret = -ESRCH; goto out_unlock; } - pcred = __task_cred(p); - if (si_fromuser(info) && - euid != pcred->suid && euid != pcred->uid && - uid != pcred->suid && uid != pcred->uid) { + if (si_fromuser(info) && !kill_as_cred_perm(cred, p)) { ret = -EPERM; goto out_unlock; } @@ -1384,7 +1392,7 @@ out_unlock: rcu_read_unlock(); return ret; } -EXPORT_SYMBOL_GPL(kill_pid_info_as_uid); +EXPORT_SYMBOL_GPL(kill_pid_info_as_cred); /* * kill_something_info() interprets pid in interesting ways just like kill(2). -- cgit v1.3-8-gc7d7 From 47ea91b4052d9e94b9dca5d7a3d947fbebd07ba9 Mon Sep 17 00:00:00 2001 From: Ram Pai Date: Thu, 22 Sep 2011 15:48:58 +0800 Subject: Resource: fix wrong resource window calculation __find_resource() incorrectly returns a resource window which overlaps an existing allocated window. This happens when the parent's resource-window spans 0x00000000 to 0xffffffff and is entirely allocated to all its children resource-windows. __find_resource() looks for gaps in resource allocation among the children resource windows. When it encounters the last child window it blindly tries the range next to one allocated to the last child. Since the last child's window ends at 0xffffffff the calculation overflows, leading the algorithm to believe that any window in the range 0x0000000 to 0xfffffff is available for allocation. This leads to a conflicting window allocation. Michal Ludvig reported this issue seen on his platform. The following patch fixes the problem and has been verified by Michal. I believe this bug has been there for ages. It got exposed by git commit 2bbc6942273b ("PCI : ability to relocate assigned pci-resources") Signed-off-by: Ram Pai Tested-by: Michal Ludvig Signed-off-by: Linus Torvalds --- kernel/resource.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/resource.c b/kernel/resource.c index 3b3cedc52592..c8dc249da5ce 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -419,6 +419,9 @@ static int __find_resource(struct resource *root, struct resource *old, else tmp.end = root->end; + if (tmp.end < tmp.start) + goto next; + resource_clip(&tmp, constraint->min, constraint->max); arch_remove_reservations(&tmp); @@ -436,8 +439,10 @@ static int __find_resource(struct resource *root, struct resource *old, return 0; } } - if (!this) + +next: if (!this || this->end == root->end) break; + if (this != old) tmp.start = this->end + 1; this = this->sibling; -- cgit v1.3-8-gc7d7 From d670ec13178d0fd8680e6742a2bc6e04f28f87d8 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 1 Sep 2011 12:42:04 +0200 Subject: posix-cpu-timers: Cure SMP wobbles David reported: Attached below is a watered-down version of rt/tst-cpuclock2.c from GLIBC. Just build it with "gcc -o test test.c -lpthread -lrt" or similar. Run it several times, and you will see cases where the main thread will measure a process clock difference before and after the nanosleep which is smaller than the cpu-burner thread's individual thread clock difference. This doesn't make any sense since the cpu-burner thread is part of the top-level process's thread group. I've reproduced this on both x86-64 and sparc64 (using both 32-bit and 64-bit binaries). For example: [davem@boricha build-x86_64-linux]$ ./test process: before(0.001221967) after(0.498624371) diff(497402404) thread: before(0.000081692) after(0.498316431) diff(498234739) self: before(0.001223521) after(0.001240219) diff(16698) [davem@boricha build-x86_64-linux]$ The diff of 'process' should always be >= the diff of 'thread'. I make sure to wrap the 'thread' clock measurements the most tightly around the nanosleep() call, and that the 'process' clock measurements are the outer-most ones. --- #include #include #include #include #include #include #include #include static pthread_barrier_t barrier; static void *chew_cpu(void *arg) { pthread_barrier_wait(&barrier); while (1) __asm__ __volatile__("" : : : "memory"); return NULL; } int main(void) { clockid_t process_clock, my_thread_clock, th_clock; struct timespec process_before, process_after; struct timespec me_before, me_after; struct timespec th_before, th_after; struct timespec sleeptime; unsigned long diff; pthread_t th; int err; err = clock_getcpuclockid(0, &process_clock); if (err) return 1; err = pthread_getcpuclockid(pthread_self(), &my_thread_clock); if (err) return 1; pthread_barrier_init(&barrier, NULL, 2); err = pthread_create(&th, NULL, chew_cpu, NULL); if (err) return 1; err = pthread_getcpuclockid(th, &th_clock); if (err) return 1; pthread_barrier_wait(&barrier); err = clock_gettime(process_clock, &process_before); if (err) return 1; err = clock_gettime(my_thread_clock, &me_before); if (err) return 1; err = clock_gettime(th_clock, &th_before); if (err) return 1; sleeptime.tv_sec = 0; sleeptime.tv_nsec = 500000000; nanosleep(&sleeptime, NULL); err = clock_gettime(th_clock, &th_after); if (err) return 1; err = clock_gettime(my_thread_clock, &me_after); if (err) return 1; err = clock_gettime(process_clock, &process_after); if (err) return 1; diff = process_after.tv_nsec - process_before.tv_nsec; printf("process: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", process_before.tv_sec, process_before.tv_nsec, process_after.tv_sec, process_after.tv_nsec, diff); diff = th_after.tv_nsec - th_before.tv_nsec; printf("thread: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", th_before.tv_sec, th_before.tv_nsec, th_after.tv_sec, th_after.tv_nsec, diff); diff = me_after.tv_nsec - me_before.tv_nsec; printf("self: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n", me_before.tv_sec, me_before.tv_nsec, me_after.tv_sec, me_after.tv_nsec, diff); return 0; } This is due to us using p->se.sum_exec_runtime in thread_group_cputime() where we iterate the thread group and sum all data. This does not take time since the last schedule operation (tick or otherwise) into account. We can cure this by using task_sched_runtime() at the cost of having to take locks. This also means we can (and must) do away with thread_group_sched_runtime() since the modified thread_group_cputime() is now more accurate and would deadlock when called from thread_group_sched_runtime(). Aside of that it makes the function safe on 32 bit systems. The old code added t->se.sum_exec_runtime unprotected. sum_exec_runtime is a 64bit value and could be changed on another cpu at the same time. Reported-by: David Miller Signed-off-by: Peter Zijlstra Cc: stable@kernel.org Link: http://lkml.kernel.org/r/1314874459.7945.22.camel@twins Tested-by: David Miller Signed-off-by: Thomas Gleixner --- include/linux/sched.h | 1 - kernel/posix-cpu-timers.c | 5 +++-- kernel/sched.c | 24 ------------------------ 3 files changed, 3 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4ac2c0578e0f..41d0237fd449 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1956,7 +1956,6 @@ static inline void disable_sched_clock_irqtime(void) {} extern unsigned long long task_sched_runtime(struct task_struct *task); -extern unsigned long long thread_group_sched_runtime(struct task_struct *task); /* sched_exec is called by processes performing an exec */ #ifdef CONFIG_SMP diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 58f405b581e7..c8008dd58ef2 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -250,7 +250,7 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) do { times->utime = cputime_add(times->utime, t->utime); times->stime = cputime_add(times->stime, t->stime); - times->sum_exec_runtime += t->se.sum_exec_runtime; + times->sum_exec_runtime += task_sched_runtime(t); } while_each_thread(tsk, t); out: rcu_read_unlock(); @@ -312,7 +312,8 @@ static int cpu_clock_sample_group(const clockid_t which_clock, cpu->cpu = cputime.utime; break; case CPUCLOCK_SCHED: - cpu->sched = thread_group_sched_runtime(p); + thread_group_cputime(p, &cputime); + cpu->sched = cputime.sum_exec_runtime; break; } return 0; diff --git a/kernel/sched.c b/kernel/sched.c index d249ea88428c..b50b0f0c9aa9 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -3724,30 +3724,6 @@ unsigned long long task_sched_runtime(struct task_struct *p) return ns; } -/* - * Return sum_exec_runtime for the thread group. - * In case the task is currently running, return the sum plus current's - * pending runtime that have not been accounted yet. - * - * Note that the thread group might have other running tasks as well, - * so the return value not includes other pending runtime that other - * running tasks might have. - */ -unsigned long long thread_group_sched_runtime(struct task_struct *p) -{ - struct task_cputime totals; - unsigned long flags; - struct rq *rq; - u64 ns; - - rq = task_rq_lock(p, &flags); - thread_group_cputime(p, &totals); - ns = totals.sum_exec_runtime + do_task_delta_exec(p, rq); - task_rq_unlock(rq, p, &flags); - - return ns; -} - /* * Account user cpu time to a process. * @p: the process that the cpu time gets accounted to -- cgit v1.3-8-gc7d7 From 9d823e8f6b1b7b39f952d7d1795f29162143a433 Mon Sep 17 00:00:00 2001 From: Wu Fengguang Date: Sat, 11 Jun 2011 18:10:12 -0600 Subject: writeback: per task dirty rate limit Add two fields to task_struct. 1) account dirtied pages in the individual tasks, for accuracy 2) per-task balance_dirty_pages() call intervals, for flexibility The balance_dirty_pages() call interval (ie. nr_dirtied_pause) will scale near-sqrt to the safety gap between dirty pages and threshold. The main problem of per-task nr_dirtied is, if 1k+ tasks start dirtying pages at exactly the same time, each task will be assigned a large initial nr_dirtied_pause, so that the dirty threshold will be exceeded long before each task reached its nr_dirtied_pause and hence call balance_dirty_pages(). The solution is to watch for the number of pages dirtied on each CPU in between the calls into balance_dirty_pages(). If it exceeds ratelimit_pages (3% dirty threshold), force call balance_dirty_pages() for a chance to set bdi->dirty_exceeded. In normal situations, this safeguarding condition is not expected to trigger at all. On the sqrt in dirty_poll_interval(): It will serve as an initial guess when dirty pages are still in the freerun area. When dirty pages are floating inside the dirty control scope [freerun, limit], a followup patch will use some refined dirty poll interval to get the desired pause time. thresh-dirty (MB) sqrt 1 16 2 22 4 32 8 45 16 64 32 90 64 128 128 181 256 256 512 362 1024 512 The above table means, given 1MB (or 1GB) gap and the dd tasks polling balance_dirty_pages() on every 16 (or 512) pages, the dirty limit won't be exceeded as long as there are less than 16 (or 512) concurrent dd's. So sqrt naturally leads to less overheads and more safe concurrent tasks for large memory servers, which have large (thresh-freerun) gaps. peter: keep the per-CPU ratelimit for safeguarding the 1k+ tasks case CC: Peter Zijlstra Reviewed-by: Andrea Righi Signed-off-by: Wu Fengguang --- include/linux/sched.h | 7 ++++ kernel/fork.c | 3 ++ mm/page-writeback.c | 89 +++++++++++++++++++++++++++++---------------------- 3 files changed, 60 insertions(+), 39 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 41d0237fd449..a4a5582dc618 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1525,6 +1525,13 @@ struct task_struct { int make_it_fail; #endif struct prop_local_single dirties; + /* + * when (nr_dirtied >= nr_dirtied_pause), it's time to call + * balance_dirty_pages() for some dirty throttling pause + */ + int nr_dirtied; + int nr_dirtied_pause; + #ifdef CONFIG_LATENCYTOP int latency_record_count; struct latency_record latency_record[LT_SAVECOUNT]; diff --git a/kernel/fork.c b/kernel/fork.c index 8e6b6f4fb272..cc0815df99f2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1302,6 +1302,9 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->pdeath_signal = 0; p->exit_state = 0; + p->nr_dirtied = 0; + p->nr_dirtied_pause = 128 >> (PAGE_SHIFT - 10); + /* * Ok, make it visible to the rest of the system. * We dont wake it up yet. diff --git a/mm/page-writeback.c b/mm/page-writeback.c index d4a6e91bd9e5..daff320d263f 100644 --- a/mm/page-writeback.c +++ b/mm/page-writeback.c @@ -54,20 +54,6 @@ */ static long ratelimit_pages = 32; -/* - * When balance_dirty_pages decides that the caller needs to perform some - * non-background writeback, this is how many pages it will attempt to write. - * It should be somewhat larger than dirtied pages to ensure that reasonably - * large amounts of I/O are submitted. - */ -static inline long sync_writeback_pages(unsigned long dirtied) -{ - if (dirtied < ratelimit_pages) - dirtied = ratelimit_pages; - - return dirtied + dirtied / 2; -} - /* The following parameters are exported via /proc/sys/vm */ /* @@ -169,6 +155,8 @@ static void update_completion_period(void) int shift = calc_period_shift(); prop_change_shift(&vm_completions, shift); prop_change_shift(&vm_dirties, shift); + + writeback_set_ratelimit(); } int dirty_background_ratio_handler(struct ctl_table *table, int write, @@ -978,6 +966,23 @@ static void bdi_update_bandwidth(struct backing_dev_info *bdi, spin_unlock(&bdi->wb.list_lock); } +/* + * After a task dirtied this many pages, balance_dirty_pages_ratelimited_nr() + * will look to see if it needs to start dirty throttling. + * + * If dirty_poll_interval is too low, big NUMA machines will call the expensive + * global_page_state() too often. So scale it near-sqrt to the safety margin + * (the number of pages we may dirty without exceeding the dirty limits). + */ +static unsigned long dirty_poll_interval(unsigned long dirty, + unsigned long thresh) +{ + if (thresh > dirty) + return 1UL << (ilog2(thresh - dirty) >> 1); + + return 1; +} + /* * balance_dirty_pages() must be called by processes which are generating dirty * data. It looks at the number of dirty pages in the machine and will force @@ -1112,6 +1117,9 @@ static void balance_dirty_pages(struct address_space *mapping, if (clear_dirty_exceeded && bdi->dirty_exceeded) bdi->dirty_exceeded = 0; + current->nr_dirtied = 0; + current->nr_dirtied_pause = dirty_poll_interval(nr_dirty, dirty_thresh); + if (writeback_in_progress(bdi)) return; @@ -1138,7 +1146,7 @@ void set_page_dirty_balance(struct page *page, int page_mkwrite) } } -static DEFINE_PER_CPU(unsigned long, bdp_ratelimits) = 0; +static DEFINE_PER_CPU(int, bdp_ratelimits); /** * balance_dirty_pages_ratelimited_nr - balance dirty memory state @@ -1158,31 +1166,39 @@ void balance_dirty_pages_ratelimited_nr(struct address_space *mapping, unsigned long nr_pages_dirtied) { struct backing_dev_info *bdi = mapping->backing_dev_info; - unsigned long ratelimit; - unsigned long *p; + int ratelimit; + int *p; if (!bdi_cap_account_dirty(bdi)) return; - ratelimit = ratelimit_pages; - if (mapping->backing_dev_info->dirty_exceeded) - ratelimit = 8; + ratelimit = current->nr_dirtied_pause; + if (bdi->dirty_exceeded) + ratelimit = min(ratelimit, 32 >> (PAGE_SHIFT - 10)); + + current->nr_dirtied += nr_pages_dirtied; + preempt_disable(); /* - * Check the rate limiting. Also, we do not want to throttle real-time - * tasks in balance_dirty_pages(). Period. + * This prevents one CPU to accumulate too many dirtied pages without + * calling into balance_dirty_pages(), which can happen when there are + * 1000+ tasks, all of them start dirtying pages at exactly the same + * time, hence all honoured too large initial task->nr_dirtied_pause. */ - preempt_disable(); p = &__get_cpu_var(bdp_ratelimits); - *p += nr_pages_dirtied; - if (unlikely(*p >= ratelimit)) { - ratelimit = sync_writeback_pages(*p); + if (unlikely(current->nr_dirtied >= ratelimit)) *p = 0; - preempt_enable(); - balance_dirty_pages(mapping, ratelimit); - return; + else { + *p += nr_pages_dirtied; + if (unlikely(*p >= ratelimit_pages)) { + *p = 0; + ratelimit = 0; + } } preempt_enable(); + + if (unlikely(current->nr_dirtied >= ratelimit)) + balance_dirty_pages(mapping, current->nr_dirtied); } EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr); @@ -1277,22 +1293,17 @@ void laptop_sync_completion(void) * * Here we set ratelimit_pages to a level which ensures that when all CPUs are * dirtying in parallel, we cannot go more than 3% (1/32) over the dirty memory - * thresholds before writeback cuts in. - * - * But the limit should not be set too high. Because it also controls the - * amount of memory which the balance_dirty_pages() caller has to write back. - * If this is too large then the caller will block on the IO queue all the - * time. So limit it to four megabytes - the balance_dirty_pages() caller - * will write six megabyte chunks, max. + * thresholds. */ void writeback_set_ratelimit(void) { - ratelimit_pages = vm_total_pages / (num_online_cpus() * 32); + unsigned long background_thresh; + unsigned long dirty_thresh; + global_dirty_limits(&background_thresh, &dirty_thresh); + ratelimit_pages = dirty_thresh / (num_online_cpus() * 32); if (ratelimit_pages < 16) ratelimit_pages = 16; - if (ratelimit_pages * PAGE_CACHE_SIZE > 4096 * 1024) - ratelimit_pages = (4096 * 1024) / PAGE_CACHE_SIZE; } static int __cpuinit -- cgit v1.3-8-gc7d7 From 31d9d9b6d83030f748d013e61502fa5477e2ac0e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 23 Sep 2011 17:03:06 +0100 Subject: genirq: Add support for per-cpu dev_id interrupts The ARM GIC interrupt controller offers per CPU interrupts (PPIs), which are usually used to connect local timers to each core. Each CPU has its own private interface to the GIC, and only sees the PPIs that are directly connect to it. While these timers are separate devices and have a separate interrupt line to a core, they all use the same IRQ number. For these devices, request_irq() is not the right API as it assumes that an IRQ number is visible by a number of CPUs (through the affinity setting), but makes it very awkward to express that an IRQ number can be handled by all CPUs, and yet be a different interrupt line on each CPU, requiring a different dev_id cookie to be passed back to the handler. The *_percpu_irq() functions is designed to overcome these limitations, by providing a per-cpu dev_id vector: int request_percpu_irq(unsigned int irq, irq_handler_t handler, const char *devname, void __percpu *percpu_dev_id); void free_percpu_irq(unsigned int, void __percpu *); int setup_percpu_irq(unsigned int irq, struct irqaction *new); void remove_percpu_irq(unsigned int irq, struct irqaction *act); void enable_percpu_irq(unsigned int irq); void disable_percpu_irq(unsigned int irq); The API has a number of limitations: - no interrupt sharing - no threading - common handler across all the CPUs Once the interrupt is requested using setup_percpu_irq() or request_percpu_irq(), it must be enabled by each core that wishes its local interrupt to be delivered. Based on an initial patch by Thomas Gleixner. Signed-off-by: Marc Zyngier Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/1316793788-14500-2-git-send-email-marc.zyngier@arm.com Signed-off-by: Thomas Gleixner --- include/linux/interrupt.h | 38 ++++++--- include/linux/irq.h | 16 +++- include/linux/irqdesc.h | 1 + kernel/irq/chip.c | 64 +++++++++++++-- kernel/irq/internals.h | 19 +++-- kernel/irq/irqdesc.c | 32 +++++++- kernel/irq/manage.c | 202 +++++++++++++++++++++++++++++++++++++++++++--- kernel/irq/settings.h | 7 ++ 8 files changed, 345 insertions(+), 34 deletions(-) (limited to 'kernel') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index a103732b7588..1cdfd09c8abc 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -95,6 +95,7 @@ typedef irqreturn_t (*irq_handler_t)(int, void *); * @flags: flags (see IRQF_* above) * @name: name of the device * @dev_id: cookie to identify the device + * @percpu_dev_id: cookie to identify the device * @next: pointer to the next irqaction for shared interrupts * @irq: interrupt number * @dir: pointer to the proc/irq/NN/name entry @@ -104,17 +105,18 @@ typedef irqreturn_t (*irq_handler_t)(int, void *); * @thread_mask: bitmask for keeping track of @thread activity */ struct irqaction { - irq_handler_t handler; - unsigned long flags; - void *dev_id; - struct irqaction *next; - int irq; - irq_handler_t thread_fn; - struct task_struct *thread; - unsigned long thread_flags; - unsigned long thread_mask; - const char *name; - struct proc_dir_entry *dir; + irq_handler_t handler; + unsigned long flags; + void *dev_id; + void __percpu *percpu_dev_id; + struct irqaction *next; + int irq; + irq_handler_t thread_fn; + struct task_struct *thread; + unsigned long thread_flags; + unsigned long thread_mask; + const char *name; + struct proc_dir_entry *dir; } ____cacheline_internodealigned_in_smp; extern irqreturn_t no_action(int cpl, void *dev_id); @@ -136,6 +138,10 @@ extern int __must_check request_any_context_irq(unsigned int irq, irq_handler_t handler, unsigned long flags, const char *name, void *dev_id); +extern int __must_check +request_percpu_irq(unsigned int irq, irq_handler_t handler, + const char *devname, void __percpu *percpu_dev_id); + extern void exit_irq_thread(void); #else @@ -164,10 +170,18 @@ request_any_context_irq(unsigned int irq, irq_handler_t handler, return request_irq(irq, handler, flags, name, dev_id); } +static inline int __must_check +request_percpu_irq(unsigned int irq, irq_handler_t handler, + const char *devname, void __percpu *percpu_dev_id) +{ + return request_irq(irq, handler, 0, devname, percpu_dev_id); +} + static inline void exit_irq_thread(void) { } #endif extern void free_irq(unsigned int, void *); +extern void free_percpu_irq(unsigned int, void __percpu *); struct device; @@ -207,7 +221,9 @@ extern void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id); extern void disable_irq_nosync(unsigned int irq); extern void disable_irq(unsigned int irq); +extern void disable_percpu_irq(unsigned int irq); extern void enable_irq(unsigned int irq); +extern void enable_percpu_irq(unsigned int irq); /* The following three functions are for the core kernel use only. */ #ifdef CONFIG_GENERIC_HARDIRQS diff --git a/include/linux/irq.h b/include/linux/irq.h index 73e31abeba1c..59e49c80cc2c 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -66,6 +66,7 @@ typedef void (*irq_preflow_handler_t)(struct irq_data *data); * IRQ_NO_BALANCING - Interrupt cannot be balanced (affinity set) * IRQ_MOVE_PCNTXT - Interrupt can be migrated from process context * IRQ_NESTED_TRHEAD - Interrupt nests into another thread + * IRQ_PER_CPU_DEVID - Dev_id is a per-cpu variable */ enum { IRQ_TYPE_NONE = 0x00000000, @@ -88,12 +89,13 @@ enum { IRQ_MOVE_PCNTXT = (1 << 14), IRQ_NESTED_THREAD = (1 << 15), IRQ_NOTHREAD = (1 << 16), + IRQ_PER_CPU_DEVID = (1 << 17), }; #define IRQF_MODIFY_MASK \ (IRQ_TYPE_SENSE_MASK | IRQ_NOPROBE | IRQ_NOREQUEST | \ IRQ_NOAUTOEN | IRQ_MOVE_PCNTXT | IRQ_LEVEL | IRQ_NO_BALANCING | \ - IRQ_PER_CPU | IRQ_NESTED_THREAD) + IRQ_PER_CPU | IRQ_NESTED_THREAD | IRQ_NOTHREAD | IRQ_PER_CPU_DEVID) #define IRQ_NO_BALANCING_MASK (IRQ_PER_CPU | IRQ_NO_BALANCING) @@ -367,6 +369,8 @@ enum { struct irqaction; extern int setup_irq(unsigned int irq, struct irqaction *new); extern void remove_irq(unsigned int irq, struct irqaction *act); +extern int setup_percpu_irq(unsigned int irq, struct irqaction *new); +extern void remove_percpu_irq(unsigned int irq, struct irqaction *act); extern void irq_cpu_online(void); extern void irq_cpu_offline(void); @@ -394,6 +398,7 @@ extern void handle_edge_irq(unsigned int irq, struct irq_desc *desc); extern void handle_edge_eoi_irq(unsigned int irq, struct irq_desc *desc); extern void handle_simple_irq(unsigned int irq, struct irq_desc *desc); extern void handle_percpu_irq(unsigned int irq, struct irq_desc *desc); +extern void handle_percpu_devid_irq(unsigned int irq, struct irq_desc *desc); extern void handle_bad_irq(unsigned int irq, struct irq_desc *desc); extern void handle_nested_irq(unsigned int irq); @@ -422,6 +427,8 @@ static inline void irq_set_chip_and_handler(unsigned int irq, struct irq_chip *c irq_set_chip_and_handler_name(irq, chip, handle, NULL); } +extern int irq_set_percpu_devid(unsigned int irq); + extern void __irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, const char *name); @@ -483,6 +490,13 @@ static inline void irq_set_nested_thread(unsigned int irq, bool nest) irq_clear_status_flags(irq, IRQ_NESTED_THREAD); } +static inline void irq_set_percpu_devid_flags(unsigned int irq) +{ + irq_set_status_flags(irq, + IRQ_NOAUTOEN | IRQ_PER_CPU | IRQ_NOTHREAD | + IRQ_NOPROBE | IRQ_PER_CPU_DEVID); +} + /* Handle dynamic irq creation and destruction */ extern unsigned int create_irq_nr(unsigned int irq_want, int node); extern int create_irq(void); diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 150134ac709a..6b69c2c9dff1 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -53,6 +53,7 @@ struct irq_desc { unsigned long last_unhandled; /* Aging timer for unhandled count */ unsigned int irqs_unhandled; raw_spinlock_t lock; + struct cpumask *percpu_enabled; #ifdef CONFIG_SMP const struct cpumask *affinity_hint; struct irq_affinity_notify *affinity_notify; diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c index dc5114b4c16c..f7c543a801d9 100644 --- a/kernel/irq/chip.c +++ b/kernel/irq/chip.c @@ -26,7 +26,7 @@ int irq_set_chip(unsigned int irq, struct irq_chip *chip) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); if (!desc) return -EINVAL; @@ -54,7 +54,7 @@ EXPORT_SYMBOL(irq_set_chip); int irq_set_irq_type(unsigned int irq, unsigned int type) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags); + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); int ret = 0; if (!desc) @@ -78,7 +78,7 @@ EXPORT_SYMBOL(irq_set_irq_type); int irq_set_handler_data(unsigned int irq, void *data) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); if (!desc) return -EINVAL; @@ -98,7 +98,7 @@ EXPORT_SYMBOL(irq_set_handler_data); int irq_set_msi_desc(unsigned int irq, struct msi_desc *entry) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); if (!desc) return -EINVAL; @@ -119,7 +119,7 @@ int irq_set_msi_desc(unsigned int irq, struct msi_desc *entry) int irq_set_chip_data(unsigned int irq, void *data) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); if (!desc) return -EINVAL; @@ -204,6 +204,24 @@ void irq_disable(struct irq_desc *desc) } } +void irq_percpu_enable(struct irq_desc *desc, unsigned int cpu) +{ + if (desc->irq_data.chip->irq_enable) + desc->irq_data.chip->irq_enable(&desc->irq_data); + else + desc->irq_data.chip->irq_unmask(&desc->irq_data); + cpumask_set_cpu(cpu, desc->percpu_enabled); +} + +void irq_percpu_disable(struct irq_desc *desc, unsigned int cpu) +{ + if (desc->irq_data.chip->irq_disable) + desc->irq_data.chip->irq_disable(&desc->irq_data); + else + desc->irq_data.chip->irq_mask(&desc->irq_data); + cpumask_clear_cpu(cpu, desc->percpu_enabled); +} + static inline void mask_ack_irq(struct irq_desc *desc) { if (desc->irq_data.chip->irq_mask_ack) @@ -544,12 +562,44 @@ handle_percpu_irq(unsigned int irq, struct irq_desc *desc) chip->irq_eoi(&desc->irq_data); } +/** + * handle_percpu_devid_irq - Per CPU local irq handler with per cpu dev ids + * @irq: the interrupt number + * @desc: the interrupt description structure for this irq + * + * Per CPU interrupts on SMP machines without locking requirements. Same as + * handle_percpu_irq() above but with the following extras: + * + * action->percpu_dev_id is a pointer to percpu variables which + * contain the real device id for the cpu on which this handler is + * called + */ +void handle_percpu_devid_irq(unsigned int irq, struct irq_desc *desc) +{ + struct irq_chip *chip = irq_desc_get_chip(desc); + struct irqaction *action = desc->action; + void *dev_id = __this_cpu_ptr(action->percpu_dev_id); + irqreturn_t res; + + kstat_incr_irqs_this_cpu(irq, desc); + + if (chip->irq_ack) + chip->irq_ack(&desc->irq_data); + + trace_irq_handler_entry(irq, action); + res = action->handler(irq, dev_id); + trace_irq_handler_exit(irq, action, res); + + if (chip->irq_eoi) + chip->irq_eoi(&desc->irq_data); +} + void __irq_set_handler(unsigned int irq, irq_flow_handler_t handle, int is_chained, const char *name) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags); + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, 0); if (!desc) return; @@ -593,7 +643,7 @@ irq_set_chip_and_handler_name(unsigned int irq, struct irq_chip *chip, void irq_modify_status(unsigned int irq, unsigned long clr, unsigned long set) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); if (!desc) return; diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index 6546431447d7..a73dd6c7372d 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -71,6 +71,8 @@ extern int irq_startup(struct irq_desc *desc); extern void irq_shutdown(struct irq_desc *desc); extern void irq_enable(struct irq_desc *desc); extern void irq_disable(struct irq_desc *desc); +extern void irq_percpu_enable(struct irq_desc *desc, unsigned int cpu); +extern void irq_percpu_disable(struct irq_desc *desc, unsigned int cpu); extern void mask_irq(struct irq_desc *desc); extern void unmask_irq(struct irq_desc *desc); @@ -114,14 +116,21 @@ static inline void chip_bus_sync_unlock(struct irq_desc *desc) desc->irq_data.chip->irq_bus_sync_unlock(&desc->irq_data); } +#define _IRQ_DESC_CHECK (1 << 0) +#define _IRQ_DESC_PERCPU (1 << 1) + +#define IRQ_GET_DESC_CHECK_GLOBAL (_IRQ_DESC_CHECK) +#define IRQ_GET_DESC_CHECK_PERCPU (_IRQ_DESC_CHECK | _IRQ_DESC_PERCPU) + struct irq_desc * -__irq_get_desc_lock(unsigned int irq, unsigned long *flags, bool bus); +__irq_get_desc_lock(unsigned int irq, unsigned long *flags, bool bus, + unsigned int check); void __irq_put_desc_unlock(struct irq_desc *desc, unsigned long flags, bool bus); static inline struct irq_desc * -irq_get_desc_buslock(unsigned int irq, unsigned long *flags) +irq_get_desc_buslock(unsigned int irq, unsigned long *flags, unsigned int check) { - return __irq_get_desc_lock(irq, flags, true); + return __irq_get_desc_lock(irq, flags, true, check); } static inline void @@ -131,9 +140,9 @@ irq_put_desc_busunlock(struct irq_desc *desc, unsigned long flags) } static inline struct irq_desc * -irq_get_desc_lock(unsigned int irq, unsigned long *flags) +irq_get_desc_lock(unsigned int irq, unsigned long *flags, unsigned int check) { - return __irq_get_desc_lock(irq, flags, false); + return __irq_get_desc_lock(irq, flags, false, check); } static inline void diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 039b889ea053..1550e8447a16 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -424,11 +424,22 @@ unsigned int irq_get_next_irq(unsigned int offset) } struct irq_desc * -__irq_get_desc_lock(unsigned int irq, unsigned long *flags, bool bus) +__irq_get_desc_lock(unsigned int irq, unsigned long *flags, bool bus, + unsigned int check) { struct irq_desc *desc = irq_to_desc(irq); if (desc) { + if (check & _IRQ_DESC_CHECK) { + if ((check & _IRQ_DESC_PERCPU) && + !irq_settings_is_per_cpu_devid(desc)) + return NULL; + + if (!(check & _IRQ_DESC_PERCPU) && + irq_settings_is_per_cpu_devid(desc)) + return NULL; + } + if (bus) chip_bus_lock(desc); raw_spin_lock_irqsave(&desc->lock, *flags); @@ -443,6 +454,25 @@ void __irq_put_desc_unlock(struct irq_desc *desc, unsigned long flags, bool bus) chip_bus_sync_unlock(desc); } +int irq_set_percpu_devid(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + + if (!desc) + return -EINVAL; + + if (desc->percpu_enabled) + return -EINVAL; + + desc->percpu_enabled = kzalloc(sizeof(*desc->percpu_enabled), GFP_KERNEL); + + if (!desc->percpu_enabled) + return -ENOMEM; + + irq_set_percpu_devid_flags(irq); + return 0; +} + /** * dynamic_irq_cleanup - cleanup a dynamically allocated irq * @irq: irq number to initialize diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 7e1a3ed1e61a..7b4b156d065c 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -195,7 +195,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *mask) int irq_set_affinity_hint(unsigned int irq, const struct cpumask *m) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); if (!desc) return -EINVAL; @@ -356,7 +356,7 @@ void __disable_irq(struct irq_desc *desc, unsigned int irq, bool suspend) static int __disable_irq_nosync(unsigned int irq) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags); + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); if (!desc) return -EINVAL; @@ -448,7 +448,7 @@ void __enable_irq(struct irq_desc *desc, unsigned int irq, bool resume) void enable_irq(unsigned int irq) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags); + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); if (!desc) return; @@ -491,7 +491,7 @@ static int set_irq_wake_real(unsigned int irq, unsigned int on) int irq_set_irq_wake(unsigned int irq, unsigned int on) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_buslock(irq, &flags); + struct irq_desc *desc = irq_get_desc_buslock(irq, &flags, IRQ_GET_DESC_CHECK_GLOBAL); int ret = 0; if (!desc) @@ -532,7 +532,7 @@ EXPORT_SYMBOL(irq_set_irq_wake); int can_request_irq(unsigned int irq, unsigned long irqflags) { unsigned long flags; - struct irq_desc *desc = irq_get_desc_lock(irq, &flags); + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, 0); int canrequest = 0; if (!desc) @@ -1121,6 +1121,8 @@ int setup_irq(unsigned int irq, struct irqaction *act) int retval; struct irq_desc *desc = irq_to_desc(irq); + if (WARN_ON(irq_settings_is_per_cpu_devid(desc))) + return -EINVAL; chip_bus_lock(desc); retval = __setup_irq(irq, desc, act); chip_bus_sync_unlock(desc); @@ -1129,7 +1131,7 @@ int setup_irq(unsigned int irq, struct irqaction *act) } EXPORT_SYMBOL_GPL(setup_irq); - /* +/* * Internal function to unregister an irqaction - used to free * regular and special interrupts that are part of the architecture. */ @@ -1227,7 +1229,10 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) */ void remove_irq(unsigned int irq, struct irqaction *act) { - __free_irq(irq, act->dev_id); + struct irq_desc *desc = irq_to_desc(irq); + + if (desc && !WARN_ON(irq_settings_is_per_cpu_devid(desc))) + __free_irq(irq, act->dev_id); } EXPORT_SYMBOL_GPL(remove_irq); @@ -1249,7 +1254,7 @@ void free_irq(unsigned int irq, void *dev_id) { struct irq_desc *desc = irq_to_desc(irq); - if (!desc) + if (!desc || WARN_ON(irq_settings_is_per_cpu_devid(desc))) return; #ifdef CONFIG_SMP @@ -1327,7 +1332,8 @@ int request_threaded_irq(unsigned int irq, irq_handler_t handler, if (!desc) return -EINVAL; - if (!irq_settings_can_request(desc)) + if (!irq_settings_can_request(desc) || + WARN_ON(irq_settings_is_per_cpu_devid(desc))) return -EINVAL; if (!handler) { @@ -1412,3 +1418,181 @@ int request_any_context_irq(unsigned int irq, irq_handler_t handler, return !ret ? IRQC_IS_HARDIRQ : ret; } EXPORT_SYMBOL_GPL(request_any_context_irq); + +void enable_percpu_irq(unsigned int irq) +{ + unsigned int cpu = smp_processor_id(); + unsigned long flags; + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, IRQ_GET_DESC_CHECK_PERCPU); + + if (!desc) + return; + + irq_percpu_enable(desc, cpu); + irq_put_desc_unlock(desc, flags); +} + +void disable_percpu_irq(unsigned int irq) +{ + unsigned int cpu = smp_processor_id(); + unsigned long flags; + struct irq_desc *desc = irq_get_desc_lock(irq, &flags, IRQ_GET_DESC_CHECK_PERCPU); + + if (!desc) + return; + + irq_percpu_disable(desc, cpu); + irq_put_desc_unlock(desc, flags); +} + +/* + * Internal function to unregister a percpu irqaction. + */ +static struct irqaction *__free_percpu_irq(unsigned int irq, void __percpu *dev_id) +{ + struct irq_desc *desc = irq_to_desc(irq); + struct irqaction *action; + unsigned long flags; + + WARN(in_interrupt(), "Trying to free IRQ %d from IRQ context!\n", irq); + + if (!desc) + return NULL; + + raw_spin_lock_irqsave(&desc->lock, flags); + + action = desc->action; + if (!action || action->percpu_dev_id != dev_id) { + WARN(1, "Trying to free already-free IRQ %d\n", irq); + goto bad; + } + + if (!cpumask_empty(desc->percpu_enabled)) { + WARN(1, "percpu IRQ %d still enabled on CPU%d!\n", + irq, cpumask_first(desc->percpu_enabled)); + goto bad; + } + + /* Found it - now remove it from the list of entries: */ + desc->action = NULL; + + raw_spin_unlock_irqrestore(&desc->lock, flags); + + unregister_handler_proc(irq, action); + + module_put(desc->owner); + return action; + +bad: + raw_spin_unlock_irqrestore(&desc->lock, flags); + return NULL; +} + +/** + * remove_percpu_irq - free a per-cpu interrupt + * @irq: Interrupt line to free + * @act: irqaction for the interrupt + * + * Used to remove interrupts statically setup by the early boot process. + */ +void remove_percpu_irq(unsigned int irq, struct irqaction *act) +{ + struct irq_desc *desc = irq_to_desc(irq); + + if (desc && irq_settings_is_per_cpu_devid(desc)) + __free_percpu_irq(irq, act->percpu_dev_id); +} + +/** + * free_percpu_irq - free an interrupt allocated with request_percpu_irq + * @irq: Interrupt line to free + * @dev_id: Device identity to free + * + * Remove a percpu interrupt handler. The handler is removed, but + * the interrupt line is not disabled. This must be done on each + * CPU before calling this function. The function does not return + * until any executing interrupts for this IRQ have completed. + * + * This function must not be called from interrupt context. + */ +void free_percpu_irq(unsigned int irq, void __percpu *dev_id) +{ + struct irq_desc *desc = irq_to_desc(irq); + + if (!desc || !irq_settings_is_per_cpu_devid(desc)) + return; + + chip_bus_lock(desc); + kfree(__free_percpu_irq(irq, dev_id)); + chip_bus_sync_unlock(desc); +} + +/** + * setup_percpu_irq - setup a per-cpu interrupt + * @irq: Interrupt line to setup + * @act: irqaction for the interrupt + * + * Used to statically setup per-cpu interrupts in the early boot process. + */ +int setup_percpu_irq(unsigned int irq, struct irqaction *act) +{ + struct irq_desc *desc = irq_to_desc(irq); + int retval; + + if (!desc || !irq_settings_is_per_cpu_devid(desc)) + return -EINVAL; + chip_bus_lock(desc); + retval = __setup_irq(irq, desc, act); + chip_bus_sync_unlock(desc); + + return retval; +} + +/** + * request_percpu_irq - allocate a percpu interrupt line + * @irq: Interrupt line to allocate + * @handler: Function to be called when the IRQ occurs. + * @devname: An ascii name for the claiming device + * @dev_id: A percpu cookie passed back to the handler function + * + * This call allocates interrupt resources, but doesn't + * automatically enable the interrupt. It has to be done on each + * CPU using enable_percpu_irq(). + * + * Dev_id must be globally unique. It is a per-cpu variable, and + * the handler gets called with the interrupted CPU's instance of + * that variable. + */ +int request_percpu_irq(unsigned int irq, irq_handler_t handler, + const char *devname, void __percpu *dev_id) +{ + struct irqaction *action; + struct irq_desc *desc; + int retval; + + if (!dev_id) + return -EINVAL; + + desc = irq_to_desc(irq); + if (!desc || !irq_settings_can_request(desc) || + !irq_settings_is_per_cpu_devid(desc)) + return -EINVAL; + + action = kzalloc(sizeof(struct irqaction), GFP_KERNEL); + if (!action) + return -ENOMEM; + + action->handler = handler; + action->flags = IRQF_PERCPU; + action->name = devname; + action->percpu_dev_id = dev_id; + + chip_bus_lock(desc); + retval = __setup_irq(irq, desc, action); + chip_bus_sync_unlock(desc); + + if (retval) + kfree(action); + + return retval; +} diff --git a/kernel/irq/settings.h b/kernel/irq/settings.h index f1667833d444..1162f1030f18 100644 --- a/kernel/irq/settings.h +++ b/kernel/irq/settings.h @@ -13,6 +13,7 @@ enum { _IRQ_MOVE_PCNTXT = IRQ_MOVE_PCNTXT, _IRQ_NO_BALANCING = IRQ_NO_BALANCING, _IRQ_NESTED_THREAD = IRQ_NESTED_THREAD, + _IRQ_PER_CPU_DEVID = IRQ_PER_CPU_DEVID, _IRQF_MODIFY_MASK = IRQF_MODIFY_MASK, }; @@ -24,6 +25,7 @@ enum { #define IRQ_NOTHREAD GOT_YOU_MORON #define IRQ_NOAUTOEN GOT_YOU_MORON #define IRQ_NESTED_THREAD GOT_YOU_MORON +#define IRQ_PER_CPU_DEVID GOT_YOU_MORON #undef IRQF_MODIFY_MASK #define IRQF_MODIFY_MASK GOT_YOU_MORON @@ -39,6 +41,11 @@ static inline bool irq_settings_is_per_cpu(struct irq_desc *desc) return desc->status_use_accessors & _IRQ_PER_CPU; } +static inline bool irq_settings_is_per_cpu_devid(struct irq_desc *desc) +{ + return desc->status_use_accessors & _IRQ_PER_CPU_DEVID; +} + static inline void irq_settings_set_per_cpu(struct irq_desc *desc) { desc->status_use_accessors |= _IRQ_PER_CPU; -- cgit v1.3-8-gc7d7 From 1e7c5fd29487ee88cb3abac945bafa60ae026146 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Fri, 30 Sep 2011 10:48:47 +0100 Subject: genirq: percpu: allow interrupt type to be set at enable time As request_percpu_irq() doesn't allow for a percpu interrupt to have its type configured (it is generally impossible to configure it on all CPUs at once), add a 'type' argument to enable_percpu_irq(). This allows some low-level, board specific init code to be switched to a generic API. [ tglx: Added WARN_ON argument ] Signed-off-by: Marc Zyngier Cc: Abhijeet Dharmapurikar Signed-off-by: Thomas Gleixner --- include/linux/interrupt.h | 2 +- kernel/irq/manage.c | 15 ++++++++++++++- 2 files changed, 15 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 1cdfd09c8abc..664544ff77d5 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -223,7 +223,7 @@ extern void disable_irq_nosync(unsigned int irq); extern void disable_irq(unsigned int irq); extern void disable_percpu_irq(unsigned int irq); extern void enable_irq(unsigned int irq); -extern void enable_percpu_irq(unsigned int irq); +extern void enable_percpu_irq(unsigned int irq, unsigned int type); /* The following three functions are for the core kernel use only. */ #ifdef CONFIG_GENERIC_HARDIRQS diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 7b4b156d065c..2bc86869859e 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1419,7 +1419,7 @@ int request_any_context_irq(unsigned int irq, irq_handler_t handler, } EXPORT_SYMBOL_GPL(request_any_context_irq); -void enable_percpu_irq(unsigned int irq) +void enable_percpu_irq(unsigned int irq, unsigned int type) { unsigned int cpu = smp_processor_id(); unsigned long flags; @@ -1428,7 +1428,20 @@ void enable_percpu_irq(unsigned int irq) if (!desc) return; + type &= IRQ_TYPE_SENSE_MASK; + if (type != IRQ_TYPE_NONE) { + int ret; + + ret = __irq_set_trigger(desc, irq, type); + + if (ret) { + WARN(1, "failed to set type for IRQ%d\n, irq"); + goto out; + } + } + irq_percpu_enable(desc, cpu); +out: irq_put_desc_unlock(desc, flags); } -- cgit v1.3-8-gc7d7 From 349d2895cc8b7db1f5be677cd685209a3805d2ed Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Fri, 30 Sep 2011 01:11:10 +0000 Subject: ipv4: NET_IPV4_ROUTE_GC_INTERVAL removal removing obsoleted sysctl, ip_rt_gc_interval variable no longer used since 2.6.38 Signed-off-by: Vasily Averin Signed-off-by: David S. Miller --- include/linux/sysctl.h | 2 +- kernel/sysctl_binary.c | 2 +- net/ipv4/route.c | 8 -------- 3 files changed, 2 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 11684d9e6bd2..9a1ec10fd504 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -435,7 +435,7 @@ enum { NET_IPV4_ROUTE_MAX_SIZE=5, NET_IPV4_ROUTE_GC_MIN_INTERVAL=6, NET_IPV4_ROUTE_GC_TIMEOUT=7, - NET_IPV4_ROUTE_GC_INTERVAL=8, + NET_IPV4_ROUTE_GC_INTERVAL=8, /* obsolete since 2.6.38 */ NET_IPV4_ROUTE_REDIRECT_LOAD=9, NET_IPV4_ROUTE_REDIRECT_NUMBER=10, NET_IPV4_ROUTE_REDIRECT_SILENCE=11, diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index e8bffbe2ba4b..6318b511afa1 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -214,7 +214,7 @@ static const struct bin_table bin_net_ipv4_route_table[] = { { CTL_INT, NET_IPV4_ROUTE_GC_MIN_INTERVAL, "gc_min_interval" }, { CTL_INT, NET_IPV4_ROUTE_GC_MIN_INTERVAL_MS, "gc_min_interval_ms" }, { CTL_INT, NET_IPV4_ROUTE_GC_TIMEOUT, "gc_timeout" }, - { CTL_INT, NET_IPV4_ROUTE_GC_INTERVAL, "gc_interval" }, + /* NET_IPV4_ROUTE_GC_INTERVAL "gc_interval" no longer used */ { CTL_INT, NET_IPV4_ROUTE_REDIRECT_LOAD, "redirect_load" }, { CTL_INT, NET_IPV4_ROUTE_REDIRECT_NUMBER, "redirect_number" }, { CTL_INT, NET_IPV4_ROUTE_REDIRECT_SILENCE, "redirect_silence" }, diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 2c21d3be891b..26c77e14395f 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -120,7 +120,6 @@ static int ip_rt_max_size; static int ip_rt_gc_timeout __read_mostly = RT_GC_TIMEOUT; -static int ip_rt_gc_interval __read_mostly = 60 * HZ; static int ip_rt_gc_min_interval __read_mostly = HZ / 2; static int ip_rt_redirect_number __read_mostly = 9; static int ip_rt_redirect_load __read_mostly = HZ / 50; @@ -3120,13 +3119,6 @@ static ctl_table ipv4_route_table[] = { .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { - .procname = "gc_interval", - .data = &ip_rt_gc_interval, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec_jiffies, - }, { .procname = "redirect_load", .data = &ip_rt_redirect_load, -- cgit v1.3-8-gc7d7 From 38aaf8090d34b623b7919d8c933f6e938c9bf44b Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Thu, 8 Sep 2011 14:00:46 +0800 Subject: irq_work: Use llist in the struct irq_work logic Use llist in irq_work instead of the lock-less linked list implementation in irq_work to avoid the code duplication. Signed-off-by: Huang Ying Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1315461646-1379-6-git-send-email-ying.huang@intel.com Signed-off-by: Ingo Molnar --- include/linux/irq_work.h | 15 ++++---- kernel/irq_work.c | 91 ++++++++++++++++++------------------------------ 2 files changed, 42 insertions(+), 64 deletions(-) (limited to 'kernel') diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h index 4fa09d4d0b71..6a9e8f5399e2 100644 --- a/include/linux/irq_work.h +++ b/include/linux/irq_work.h @@ -1,20 +1,23 @@ #ifndef _LINUX_IRQ_WORK_H #define _LINUX_IRQ_WORK_H +#include + struct irq_work { - struct irq_work *next; + unsigned long flags; + struct llist_node llnode; void (*func)(struct irq_work *); }; static inline -void init_irq_work(struct irq_work *entry, void (*func)(struct irq_work *)) +void init_irq_work(struct irq_work *work, void (*func)(struct irq_work *)) { - entry->next = NULL; - entry->func = func; + work->flags = 0; + work->func = func; } -bool irq_work_queue(struct irq_work *entry); +bool irq_work_queue(struct irq_work *work); void irq_work_run(void); -void irq_work_sync(struct irq_work *entry); +void irq_work_sync(struct irq_work *work); #endif /* _LINUX_IRQ_WORK_H */ diff --git a/kernel/irq_work.c b/kernel/irq_work.c index c58fa7da8aef..6f0a4310defd 100644 --- a/kernel/irq_work.c +++ b/kernel/irq_work.c @@ -17,54 +17,34 @@ * claimed NULL, 3 -> {pending} : claimed to be enqueued * pending next, 3 -> {busy} : queued, pending callback * busy NULL, 2 -> {free, claimed} : callback in progress, can be claimed - * - * We use the lower two bits of the next pointer to keep PENDING and BUSY - * flags. */ #define IRQ_WORK_PENDING 1UL #define IRQ_WORK_BUSY 2UL #define IRQ_WORK_FLAGS 3UL -static inline bool irq_work_is_set(struct irq_work *entry, int flags) -{ - return (unsigned long)entry->next & flags; -} - -static inline struct irq_work *irq_work_next(struct irq_work *entry) -{ - unsigned long next = (unsigned long)entry->next; - next &= ~IRQ_WORK_FLAGS; - return (struct irq_work *)next; -} - -static inline struct irq_work *next_flags(struct irq_work *entry, int flags) -{ - unsigned long next = (unsigned long)entry; - next |= flags; - return (struct irq_work *)next; -} - -static DEFINE_PER_CPU(struct irq_work *, irq_work_list); +static DEFINE_PER_CPU(struct llist_head, irq_work_list); /* * Claim the entry so that no one else will poke at it. */ -static bool irq_work_claim(struct irq_work *entry) +static bool irq_work_claim(struct irq_work *work) { - struct irq_work *next, *nflags; + unsigned long flags, nflags; - do { - next = entry->next; - if ((unsigned long)next & IRQ_WORK_PENDING) + for (;;) { + flags = work->flags; + if (flags & IRQ_WORK_PENDING) return false; - nflags = next_flags(next, IRQ_WORK_FLAGS); - } while (cmpxchg(&entry->next, next, nflags) != next); + nflags = flags | IRQ_WORK_FLAGS; + if (cmpxchg(&work->flags, flags, nflags) == flags) + break; + cpu_relax(); + } return true; } - void __weak arch_irq_work_raise(void) { /* @@ -75,20 +55,15 @@ void __weak arch_irq_work_raise(void) /* * Queue the entry and raise the IPI if needed. */ -static void __irq_work_queue(struct irq_work *entry) +static void __irq_work_queue(struct irq_work *work) { - struct irq_work *next; + bool empty; preempt_disable(); - do { - next = __this_cpu_read(irq_work_list); - /* Can assign non-atomic because we keep the flags set. */ - entry->next = next_flags(next, IRQ_WORK_FLAGS); - } while (this_cpu_cmpxchg(irq_work_list, next, entry) != next); - + empty = llist_add(&work->llnode, &__get_cpu_var(irq_work_list)); /* The list was empty, raise self-interrupt to start processing. */ - if (!irq_work_next(entry)) + if (empty) arch_irq_work_raise(); preempt_enable(); @@ -100,16 +75,16 @@ static void __irq_work_queue(struct irq_work *entry) * * Can be re-enqueued while the callback is still in progress. */ -bool irq_work_queue(struct irq_work *entry) +bool irq_work_queue(struct irq_work *work) { - if (!irq_work_claim(entry)) { + if (!irq_work_claim(work)) { /* * Already enqueued, can't do! */ return false; } - __irq_work_queue(entry); + __irq_work_queue(work); return true; } EXPORT_SYMBOL_GPL(irq_work_queue); @@ -120,34 +95,34 @@ EXPORT_SYMBOL_GPL(irq_work_queue); */ void irq_work_run(void) { - struct irq_work *list; + struct irq_work *work; + struct llist_head *this_list; + struct llist_node *llnode; - if (this_cpu_read(irq_work_list) == NULL) + this_list = &__get_cpu_var(irq_work_list); + if (llist_empty(this_list)) return; BUG_ON(!in_irq()); BUG_ON(!irqs_disabled()); - list = this_cpu_xchg(irq_work_list, NULL); - - while (list != NULL) { - struct irq_work *entry = list; + llnode = llist_del_all(this_list); + while (llnode != NULL) { + work = llist_entry(llnode, struct irq_work, llnode); - list = irq_work_next(list); + llnode = llnode->next; /* - * Clear the PENDING bit, after this point the @entry + * Clear the PENDING bit, after this point the @work * can be re-used. */ - entry->next = next_flags(NULL, IRQ_WORK_BUSY); - entry->func(entry); + work->flags = IRQ_WORK_BUSY; + work->func(work); /* * Clear the BUSY bit and return to the free state if * no-one else claimed it meanwhile. */ - (void)cmpxchg(&entry->next, - next_flags(NULL, IRQ_WORK_BUSY), - NULL); + (void)cmpxchg(&work->flags, IRQ_WORK_BUSY, 0); } } EXPORT_SYMBOL_GPL(irq_work_run); @@ -156,11 +131,11 @@ EXPORT_SYMBOL_GPL(irq_work_run); * Synchronize against the irq_work @entry, ensures the entry is not * currently in use. */ -void irq_work_sync(struct irq_work *entry) +void irq_work_sync(struct irq_work *work) { WARN_ON_ONCE(irqs_disabled()); - while (irq_work_is_set(entry, IRQ_WORK_BUSY)) + while (work->flags & IRQ_WORK_BUSY) cpu_relax(); } EXPORT_SYMBOL_GPL(irq_work_sync); -- cgit v1.3-8-gc7d7 From 924f8f5af31423529cc3940cb2ae9fee736b7517 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 12 Sep 2011 13:12:28 +0200 Subject: llist: Add llist_next() So we don't have to expose the struct list_node member. Cc: Huang Ying Cc: Andrew Morton Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1315836348.26517.41.camel@twins Signed-off-by: Ingo Molnar --- include/linux/llist.h | 5 +++++ kernel/irq_work.c | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/llist.h b/include/linux/llist.h index 27bbdf5ddf82..e2e96d04ee48 100644 --- a/include/linux/llist.h +++ b/include/linux/llist.h @@ -138,6 +138,11 @@ static inline bool llist_empty(const struct llist_head *head) return ACCESS_ONCE(head->first) == NULL; } +static inline struct llist_node *llist_next(struct llist_node *node) +{ + return node->next; +} + /** * llist_add - add a new entry * @new: new entry to be added diff --git a/kernel/irq_work.c b/kernel/irq_work.c index 6f0a4310defd..0e2cde4f380b 100644 --- a/kernel/irq_work.c +++ b/kernel/irq_work.c @@ -110,7 +110,7 @@ void irq_work_run(void) while (llnode != NULL) { work = llist_entry(llnode, struct irq_work, llnode); - llnode = llnode->next; + llnode = llist_next(llnode); /* * Clear the PENDING bit, after this point the @work -- cgit v1.3-8-gc7d7 From fa14ff4accfb24e59d2473f3d864d6648d80563b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 12 Sep 2011 13:06:17 +0200 Subject: sched: Convert to struct llist Use the generic llist primitives. We had a private lockless list implementation in the scheduler in the wake-list code, now that we have a generic llist implementation that provides all required operations, switch to it. This patch is not expected to change any behavior. Signed-off-by: Peter Zijlstra Cc: Huang Ying Cc: Andrew Morton Link: http://lkml.kernel.org/r/1315836353.26517.42.camel@twins Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 ++- kernel/sched.c | 48 ++++++++++-------------------------------------- 2 files changed, 12 insertions(+), 39 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 9fda2888a6ab..fc3e8911818a 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -90,6 +90,7 @@ struct sched_param { #include #include #include +#include #include @@ -1225,7 +1226,7 @@ struct task_struct { unsigned int ptrace; #ifdef CONFIG_SMP - struct task_struct *wake_entry; + struct llist_node wake_entry; int on_cpu; #endif int on_rq; diff --git a/kernel/sched.c b/kernel/sched.c index c5cf15e1eb57..1874c7418319 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -702,7 +702,7 @@ struct rq { #endif #ifdef CONFIG_SMP - struct task_struct *wake_list; + struct llist_head wake_list; #endif }; @@ -2698,42 +2698,26 @@ static int ttwu_remote(struct task_struct *p, int wake_flags) } #ifdef CONFIG_SMP -static void sched_ttwu_do_pending(struct task_struct *list) +static void sched_ttwu_pending(void) { struct rq *rq = this_rq(); + struct llist_node *llist = llist_del_all(&rq->wake_list); + struct task_struct *p; raw_spin_lock(&rq->lock); - while (list) { - struct task_struct *p = list; - list = list->wake_entry; + while (llist) { + p = llist_entry(llist, struct task_struct, wake_entry); + llist = llist_next(llist); ttwu_do_activate(rq, p, 0); } raw_spin_unlock(&rq->lock); } -#ifdef CONFIG_HOTPLUG_CPU - -static void sched_ttwu_pending(void) -{ - struct rq *rq = this_rq(); - struct task_struct *list = xchg(&rq->wake_list, NULL); - - if (!list) - return; - - sched_ttwu_do_pending(list); -} - -#endif /* CONFIG_HOTPLUG_CPU */ - void scheduler_ipi(void) { - struct rq *rq = this_rq(); - struct task_struct *list = xchg(&rq->wake_list, NULL); - - if (!list) + if (llist_empty(&this_rq()->wake_list)) return; /* @@ -2750,25 +2734,13 @@ void scheduler_ipi(void) * somewhat pessimize the simple resched case. */ irq_enter(); - sched_ttwu_do_pending(list); + sched_ttwu_pending(); irq_exit(); } static void ttwu_queue_remote(struct task_struct *p, int cpu) { - struct rq *rq = cpu_rq(cpu); - struct task_struct *next = rq->wake_list; - - for (;;) { - struct task_struct *old = next; - - p->wake_entry = next; - next = cmpxchg(&rq->wake_list, old, p); - if (next == old) - break; - } - - if (!next) + if (llist_add(&p->wake_entry, &cpu_rq(cpu)->wake_list)) smp_send_reschedule(cpu); } -- cgit v1.3-8-gc7d7 From 908a3283728d92df36e0c7cd63304fd35e93a8a9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 15 Sep 2011 15:32:06 +0200 Subject: sched: Fix idle_cpu() On -rt we observed hackbench waking all 400 tasks to a single cpu. This is because of select_idle_sibling()'s interaction with the new ipi based wakeup scheme. The existing idle_cpu() test only checks to see if the current task on that cpu is the idle task, it does not take already queued tasks into account, nor does it take queued to be woken tasks into account. If the remote wakeup IPIs come hard enough, there won't be time to schedule away from the idle task, and would thus keep thinking the cpu was in fact idle, regardless of the fact that there were already several hundred tasks runnable. We couldn't reproduce on mainline, but there's no reason it couldn't happen. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-3o30p18b2paswpc9ohy2gltp@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 1874c7418319..4cdc91cf48f6 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5138,7 +5138,20 @@ EXPORT_SYMBOL(task_nice); */ int idle_cpu(int cpu) { - return cpu_curr(cpu) == cpu_rq(cpu)->idle; + struct rq *rq = cpu_rq(cpu); + + if (rq->curr != rq->idle) + return 0; + + if (rq->nr_running) + return 0; + +#ifdef CONFIG_SMP + if (!llist_empty(&rq->wake_list)) + return 0; +#endif + + return 1; } /** -- cgit v1.3-8-gc7d7 From 32cffdde4a3ee6c2d9e0f0a94edecf1a9ce7586b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 4 Oct 2011 18:43:57 +0200 Subject: genirq: Fix fatfinered fixup really Putting the argument inside the quote does not really help. Signed-off-by: Thomas Gleixner --- kernel/irq/manage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 2bc86869859e..67ce837ae52c 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1435,7 +1435,7 @@ void enable_percpu_irq(unsigned int irq, unsigned int type) ret = __irq_set_trigger(desc, irq, type); if (ret) { - WARN(1, "failed to set type for IRQ%d\n, irq"); + WARN(1, "failed to set type for IRQ%d\n", irq); goto out; } } -- cgit v1.3-8-gc7d7 From 68cc3990a545dc0da221b4844dd8b9c06623a6c5 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 5 Oct 2011 13:20:24 +0200 Subject: rtmutex: Add missing rcu_read_unlock() in debug_rt_mutex_print_deadlock() Signed-off-by: Thomas Gleixner --- kernel/rtmutex-debug.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c index cb1ced6967ad..a2e7e7210f3e 100644 --- a/kernel/rtmutex-debug.c +++ b/kernel/rtmutex-debug.c @@ -94,8 +94,10 @@ void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) return; } - if (!debug_locks_off()) + if (!debug_locks_off()) { + rcu_read_unlock(); return; + } printk("\n============================================\n"); printk( "[ BUG: circular locking deadlock detected! ]\n"); -- cgit v1.3-8-gc7d7 From ca38062e57e97791c2f62e3dbd06caf3ebb5721c Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Mon, 3 Oct 2011 15:09:00 -0700 Subject: sched: Use resched IPI to kick off the nohz idle balance Current use of smp call function to kick the nohz idle balance can deadlock in this scenario. 1. cpu-A did a generic_exec_single() to cpu-B and after queuing its call single data (csd) to the call single queue, cpu-A took a timer interrupt. Actual IPI to cpu-B to process the call single queue is not yet sent. 2. As part of the timer interrupt handler, cpu-A decided to kick cpu-B for the idle load balancing (sets cpu-B's rq->nohz_balance_kick to 1) and __smp_call_function_single() with nowait will queue the csd to the cpu-B's queue. But the generic_exec_single() won't send an IPI to cpu-B as the call single queue was not empty. 3. cpu-A is busy with lot of interrupts 4. Meanwhile cpu-B is entering and exiting idle and noticed that it has it's rq->nohz_balance_kick set to '1'. So it will go ahead and do the idle load balancer and clear its rq->nohz_balance_kick. 5. At this point, csd queued as part of the step-2 above is still locked and waiting to be serviced on cpu-B. 6. cpu-A is still busy with interrupt load and now it got another timer interrupt and as part of it decided to kick cpu-B for another idle load balancing (as it finds cpu-B's rq->nohz_balance_kick cleared in step-4 above) and does __smp_call_function_single() with the same csd that is still locked. 7. And we get a deadlock waiting for the csd_lock() in the __smp_call_function_single(). Main issue here is that cpu-B can service the idle load balancer kick request from cpu-A even with out receiving the IPI and this lead to doing multiple __smp_call_function_single() on the same csd leading to deadlock. To kick a cpu, scheduler already has the reschedule vector reserved. Use that mechanism (kick_process()) instead of using the generic smp call function mechanism to kick off the nohz idle load balancing and avoid the deadlock. [ This issue is present from 2.6.35+ kernels, but marking it -stable only from v3.0+ as the proposed fix depends on the scheduler_ipi() that is introduced recently. ] Reported-by: Prarit Bhargava Signed-off-by: Suresh Siddha Cc: stable@kernel.org # v3.0+ Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111003220934.834943260@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 21 +++++++++++++++++++-- kernel/sched_fair.c | 29 +++++++++-------------------- 2 files changed, 28 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 4cdc91cf48f6..9e49af00ae3e 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1404,6 +1404,18 @@ void wake_up_idle_cpu(int cpu) smp_send_reschedule(cpu); } +static inline bool got_nohz_idle_kick(void) +{ + return idle_cpu(smp_processor_id()) && this_rq()->nohz_balance_kick; +} + +#else /* CONFIG_NO_HZ */ + +static inline bool got_nohz_idle_kick(void) +{ + return false; +} + #endif /* CONFIG_NO_HZ */ static u64 sched_avg_period(void) @@ -2717,7 +2729,7 @@ static void sched_ttwu_pending(void) void scheduler_ipi(void) { - if (llist_empty(&this_rq()->wake_list)) + if (llist_empty(&this_rq()->wake_list) && !got_nohz_idle_kick()) return; /* @@ -2735,6 +2747,12 @@ void scheduler_ipi(void) */ irq_enter(); sched_ttwu_pending(); + + /* + * Check if someone kicked us for doing the nohz idle load balance. + */ + if (unlikely(got_nohz_idle_kick() && !need_resched())) + raise_softirq_irqoff(SCHED_SOFTIRQ); irq_exit(); } @@ -8288,7 +8306,6 @@ void __init sched_init(void) rq_attach_root(rq, &def_root_domain); #ifdef CONFIG_NO_HZ rq->nohz_balance_kick = 0; - init_sched_softirq_csd(&per_cpu(remote_sched_softirq_cb, i)); #endif #endif init_rq_hrtick(rq); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index fef0bfde7c8c..6c5fa1099229 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -4269,22 +4269,6 @@ out_unlock: } #ifdef CONFIG_NO_HZ - -static DEFINE_PER_CPU(struct call_single_data, remote_sched_softirq_cb); - -static void trigger_sched_softirq(void *data) -{ - raise_softirq_irqoff(SCHED_SOFTIRQ); -} - -static inline void init_sched_softirq_csd(struct call_single_data *csd) -{ - csd->func = trigger_sched_softirq; - csd->info = NULL; - csd->flags = 0; - csd->priv = 0; -} - /* * idle load balancing details * - One of the idle CPUs nominates itself as idle load_balancer, while @@ -4450,11 +4434,16 @@ static void nohz_balancer_kick(int cpu) } if (!cpu_rq(ilb_cpu)->nohz_balance_kick) { - struct call_single_data *cp; - cpu_rq(ilb_cpu)->nohz_balance_kick = 1; - cp = &per_cpu(remote_sched_softirq_cb, cpu); - __smp_call_function_single(ilb_cpu, cp, 0); + + smp_mb(); + /* + * Use smp_send_reschedule() instead of resched_cpu(). + * This way we generate a sched IPI on the target cpu which + * is idle. And the softirq performing nohz idle load balance + * will be run before returning from the IPI. + */ + smp_send_reschedule(ilb_cpu); } return; } -- cgit v1.3-8-gc7d7 From 6eb57e0d65ebd99a71d435dc96d83e725752eef8 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Mon, 3 Oct 2011 15:09:01 -0700 Subject: sched: Request for idle balance during nohz idle load balance rq's idle_at_tick is set to idle/busy during the timer tick depending on the cpu was idle or not. This will be used later in the load balance that will be done in the softirq context (which is a process context in -RT kernels). For nohz kernels, for the cpu doing nohz idle load balance on behalf of all the idle cpu's, its rq->idle_at_tick might have a stale value (which is recorded when it got the timer tick presumably when it is busy). As the nohz idle load balancing is also being done at the same place as the regular load balancing, nohz idle load balancing was bailing out when it sees rq's idle_at_tick not set. Thus leading to poor system utilization. Rename rq's idle_at_tick to idle_balance and set it when someone requests for nohz idle balance on an idle cpu. Reported-by: Srivatsa Vaddagiri Signed-off-by: Suresh Siddha Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111003220934.892350549@sbsiddha-desk.sc.intel.com Signed-off-by: Ingo Molnar --- kernel/sched.c | 8 +++++--- kernel/sched_fair.c | 4 ++-- 2 files changed, 7 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 9e49af00ae3e..7bc9b0e84eb3 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -644,7 +644,7 @@ struct rq { unsigned long cpu_power; - unsigned char idle_at_tick; + unsigned char idle_balance; /* For active balancing */ int post_schedule; int active_balance; @@ -2751,8 +2751,10 @@ void scheduler_ipi(void) /* * Check if someone kicked us for doing the nohz idle load balance. */ - if (unlikely(got_nohz_idle_kick() && !need_resched())) + if (unlikely(got_nohz_idle_kick() && !need_resched())) { + this_rq()->idle_balance = 1; raise_softirq_irqoff(SCHED_SOFTIRQ); + } irq_exit(); } @@ -4247,7 +4249,7 @@ void scheduler_tick(void) perf_event_task_tick(); #ifdef CONFIG_SMP - rq->idle_at_tick = idle_cpu(cpu); + rq->idle_balance = idle_cpu(cpu); trigger_load_balance(rq, cpu); #endif } diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 6c5fa1099229..506db0966eb8 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -4676,7 +4676,7 @@ static inline int nohz_kick_needed(struct rq *rq, int cpu) if (time_before(now, nohz.next_balance)) return 0; - if (rq->idle_at_tick) + if (idle_cpu(cpu)) return 0; first_pick_cpu = atomic_read(&nohz.first_pick_cpu); @@ -4712,7 +4712,7 @@ static void run_rebalance_domains(struct softirq_action *h) { int this_cpu = smp_processor_id(); struct rq *this_rq = cpu_rq(this_cpu); - enum cpu_idle_type idle = this_rq->idle_at_tick ? + enum cpu_idle_type idle = this_rq->idle_balance ? CPU_IDLE : CPU_NOT_IDLE; rebalance_domains(this_cpu, idle); -- cgit v1.3-8-gc7d7 From fa17b507f142d37aeac322a95f6f7c6375f25601 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 16 Jun 2011 12:23:22 +0200 Subject: sched: Wrap scheduler p->cpus_allowed access This task is preparatory for the migrate_disable() implementation, but stands on its own and provides a cleanup. It currently only converts those sites required for task-placement. Kosaki-san once mentioned replacing cpus_allowed with a proper cpumask_t instead of the NR_CPUS sized array it currently is, that would also require something like this. Signed-off-by: Peter Zijlstra Acked-by: Thomas Gleixner Cc: KOSAKI Motohiro Link: http://lkml.kernel.org/n/tip-e42skvaddos99psip0vce41o@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 8 ++++---- kernel/sched_fair.c | 12 ++++++------ kernel/sched_rt.c | 4 ++-- lib/smp_processor_id.c | 2 +- 4 files changed, 13 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 7bc9b0e84eb3..45174ca5c8ea 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2544,11 +2544,11 @@ static int select_fallback_rq(int cpu, struct task_struct *p) /* Look for allowed, online CPU in same node. */ for_each_cpu_and(dest_cpu, nodemask, cpu_active_mask) - if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) + if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) return dest_cpu; /* Any allowed, online CPU? */ - dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); + dest_cpu = cpumask_any_and(tsk_cpus_allowed(p), cpu_active_mask); if (dest_cpu < nr_cpu_ids) return dest_cpu; @@ -2585,7 +2585,7 @@ int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) * [ this allows ->select_task() to simply return task_cpu(p) and * not worry about this generic constraint ] */ - if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) || + if (unlikely(!cpumask_test_cpu(cpu, tsk_cpus_allowed(p)) || !cpu_online(cpu))) cpu = select_fallback_rq(task_cpu(p), p); @@ -6262,7 +6262,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) if (task_cpu(p) != src_cpu) goto done; /* Affinity changed (again). */ - if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) + if (!cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) goto fail; /* diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 506db0966eb8..5c9e67923b7c 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -2183,7 +2183,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, /* Skip over this group if it has no CPUs allowed */ if (!cpumask_intersects(sched_group_cpus(group), - &p->cpus_allowed)) + tsk_cpus_allowed(p))) continue; local_group = cpumask_test_cpu(this_cpu, @@ -2229,7 +2229,7 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) int i; /* Traverse only the allowed CPUs */ - for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { + for_each_cpu_and(i, sched_group_cpus(group), tsk_cpus_allowed(p)) { load = weighted_cpuload(i); if (load < min_load || (load == min_load && i == this_cpu)) { @@ -2273,7 +2273,7 @@ static int select_idle_sibling(struct task_struct *p, int target) if (!(sd->flags & SD_SHARE_PKG_RESOURCES)) break; - for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) { + for_each_cpu_and(i, sched_domain_span(sd), tsk_cpus_allowed(p)) { if (idle_cpu(i)) { target = i; break; @@ -2316,7 +2316,7 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) int sync = wake_flags & WF_SYNC; if (sd_flag & SD_BALANCE_WAKE) { - if (cpumask_test_cpu(cpu, &p->cpus_allowed)) + if (cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) want_affine = 1; new_cpu = prev_cpu; } @@ -2697,7 +2697,7 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, * 2) cannot be migrated to this CPU due to cpus_allowed, or * 3) are cache-hot on their current CPU. */ - if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) { + if (!cpumask_test_cpu(this_cpu, tsk_cpus_allowed(p))) { schedstat_inc(p, se.statistics.nr_failed_migrations_affine); return 0; } @@ -4087,7 +4087,7 @@ redo: * moved to this_cpu */ if (!cpumask_test_cpu(this_cpu, - &busiest->curr->cpus_allowed)) { + tsk_cpus_allowed(busiest->curr))) { raw_spin_unlock_irqrestore(&busiest->lock, flags); all_pinned = 1; diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 0cc188cf7664..57a10842afa1 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1179,7 +1179,7 @@ static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep); static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) && + (cpu < 0 || cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) && (p->rt.nr_cpus_allowed > 1)) return 1; return 0; @@ -1324,7 +1324,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) */ if (unlikely(task_rq(task) != rq || !cpumask_test_cpu(lowest_rq->cpu, - &task->cpus_allowed) || + tsk_cpus_allowed(task)) || task_running(rq, task) || !task->on_rq)) { diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c index 4689cb073da4..503f087382a4 100644 --- a/lib/smp_processor_id.c +++ b/lib/smp_processor_id.c @@ -22,7 +22,7 @@ notrace unsigned int debug_smp_processor_id(void) * Kernel threads bound to a single CPU can safely use * smp_processor_id(): */ - if (cpumask_equal(¤t->cpus_allowed, cpumask_of(this_cpu))) + if (cpumask_equal(tsk_cpus_allowed(current), cpumask_of(this_cpu))) goto out; /* -- cgit v1.3-8-gc7d7 From 4939602a2441306008c6dca38216b741d4e09a42 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sat, 25 Jun 2011 15:45:46 +0200 Subject: sched: Unify the ->cpus_allowed mask copy Currently every sched_class::set_cpus_allowed() implementation has to copy the cpumask into task_struct::cpus_allowed, this is pointless, put this copy in the generic code. Signed-off-by: Peter Zijlstra Acked-by: Thomas Gleixner Link: http://lkml.kernel.org/n/tip-jhl5s9fckd9ptw1fzbqqlrd3@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 7 +++---- kernel/sched_rt.c | 3 --- 2 files changed, 3 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 45174ca5c8ea..ce9a9e7db116 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6161,10 +6161,9 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) { if (p->sched_class && p->sched_class->set_cpus_allowed) p->sched_class->set_cpus_allowed(p, new_mask); - else { - cpumask_copy(&p->cpus_allowed, new_mask); - p->rt.nr_cpus_allowed = cpumask_weight(new_mask); - } + + cpumask_copy(&p->cpus_allowed, new_mask); + p->rt.nr_cpus_allowed = cpumask_weight(new_mask); } /* diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 57a10842afa1..3023fd1ef364 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -1608,9 +1608,6 @@ static void set_cpus_allowed_rt(struct task_struct *p, update_rt_migration(&rq->rt); } - - cpumask_copy(&p->cpus_allowed, new_mask); - p->rt.nr_cpus_allowed = weight; } /* Assumes rq->lock is held */ -- cgit v1.3-8-gc7d7 From 1c83437e80186832a9a48dbb6b8868d28e40e562 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 5 Oct 2011 13:32:34 +0200 Subject: sched: Warn on rt throttling The default rt-throttling is a source of never ending questions. Warn once when we go into throttling so folks have that info in dmesg. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1110051331480.18778@ionos Signed-off-by: Ingo Molnar --- kernel/sched_rt.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 3023fd1ef364..056cbd2e2a27 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -655,6 +655,7 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq) if (rt_rq->rt_time > runtime) { rt_rq->rt_throttled = 1; + printk_once(KERN_WARNING "sched: RT throttling activated\n"); if (rt_rq_throttled(rt_rq)) { sched_rt_rq_dequeue(rt_rq); return 1; -- cgit v1.3-8-gc7d7 From 510f5acc4f4fb07f3f075900dc468d6e380beff6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 17 Jul 2011 20:47:54 +0200 Subject: sched: Don't use tasklist_lock for debug prints Avoid taking locks from debug prints, this avoids latencies on -rt, and improves reliability of the debug code. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/sched.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index ce9a9e7db116..24637c782002 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6021,7 +6021,7 @@ void show_state_filter(unsigned long state_filter) printk(KERN_INFO " task PC stack pid father\n"); #endif - read_lock(&tasklist_lock); + rcu_read_lock(); do_each_thread(g, p) { /* * reset the NMI-timeout, listing all files on a slow @@ -6037,7 +6037,7 @@ void show_state_filter(unsigned long state_filter) #ifdef CONFIG_SCHED_DEBUG sysrq_sched_debug_show(); #endif - read_unlock(&tasklist_lock); + rcu_read_unlock(); /* * Only show locks if all tasks are dumped: */ -- cgit v1.3-8-gc7d7 From 02ca1521ad404cf566e0075848f80d064c0a0503 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Tue, 4 Oct 2011 19:44:38 +0900 Subject: ftrace/kprobes: Fix not to delete probes if in use Fix kprobe-tracer not to delete a probe if the probe is in use. In that case, delete operation will return -EBUSY. This bug can cause a kernel panic if enabled probes are deleted during perf record. (Add some probes on functions) sh-4.2# perf probe --del probe:\* sh-4.2# exit (kernel panic) This is originally reported on the fedora bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=742383 I've also checked that this problem doesn't happen on tracepoints when module removing because perf event locks target module. $ sudo ./perf record -e xfs:\* -aR sh sh-4.2# rmmod xfs ERROR: Module xfs is in use sh-4.2# exit [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.203 MB perf.data (~8862 samples) ] Signed-off-by: Masami Hiramatsu Cc: Arnaldo Carvalho de Melo Cc: Ingo Molnar Cc: Frederic Weisbecker Cc: Frank Ch. Eigler Cc: stable@kernel.org Link: http://lkml.kernel.org/r/20111004104438.14591.6553.stgit@fedora15 Signed-off-by: Steven Rostedt --- kernel/trace/trace_kprobe.c | 58 ++++++++++++++++++++++++++++++++++++++------- 1 file changed, 49 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 5fb3697bf0e5..00d527c945a4 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -836,11 +836,17 @@ static void __unregister_trace_probe(struct trace_probe *tp) } /* Unregister a trace_probe and probe_event: call with locking probe_lock */ -static void unregister_trace_probe(struct trace_probe *tp) +static int unregister_trace_probe(struct trace_probe *tp) { + /* Enabled event can not be unregistered */ + if (trace_probe_is_enabled(tp)) + return -EBUSY; + __unregister_trace_probe(tp); list_del(&tp->list); unregister_probe_event(tp); + + return 0; } /* Register a trace_probe and probe_event */ @@ -854,7 +860,9 @@ static int register_trace_probe(struct trace_probe *tp) /* Delete old (same name) event if exist */ old_tp = find_trace_probe(tp->call.name, tp->call.class->system); if (old_tp) { - unregister_trace_probe(old_tp); + ret = unregister_trace_probe(old_tp); + if (ret < 0) + goto end; free_trace_probe(old_tp); } @@ -892,6 +900,7 @@ static int trace_probe_module_callback(struct notifier_block *nb, mutex_lock(&probe_lock); list_for_each_entry(tp, &probe_list, list) { if (trace_probe_within_module(tp, mod)) { + /* Don't need to check busy - this should have gone. */ __unregister_trace_probe(tp); ret = __register_trace_probe(tp); if (ret) @@ -1205,10 +1214,11 @@ static int create_trace_probe(int argc, char **argv) return -ENOENT; } /* delete an event */ - unregister_trace_probe(tp); - free_trace_probe(tp); + ret = unregister_trace_probe(tp); + if (ret == 0) + free_trace_probe(tp); mutex_unlock(&probe_lock); - return 0; + return ret; } if (argc < 2) { @@ -1317,18 +1327,29 @@ error: return ret; } -static void release_all_trace_probes(void) +static int release_all_trace_probes(void) { struct trace_probe *tp; + int ret = 0; mutex_lock(&probe_lock); + /* Ensure no probe is in use. */ + list_for_each_entry(tp, &probe_list, list) + if (trace_probe_is_enabled(tp)) { + ret = -EBUSY; + goto end; + } /* TODO: Use batch unregistration */ while (!list_empty(&probe_list)) { tp = list_entry(probe_list.next, struct trace_probe, list); unregister_trace_probe(tp); free_trace_probe(tp); } + +end: mutex_unlock(&probe_lock); + + return ret; } /* Probes listing interfaces */ @@ -1380,9 +1401,13 @@ static const struct seq_operations probes_seq_op = { static int probes_open(struct inode *inode, struct file *file) { - if ((file->f_mode & FMODE_WRITE) && - (file->f_flags & O_TRUNC)) - release_all_trace_probes(); + int ret; + + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { + ret = release_all_trace_probes(); + if (ret < 0) + return ret; + } return seq_open(file, &probes_seq_op); } @@ -2055,6 +2080,21 @@ static __init int kprobe_trace_self_tests_init(void) ret = target(1, 2, 3, 4, 5, 6); + /* Disable trace points before removing it */ + tp = find_trace_probe("testprobe", KPROBE_EVENT_SYSTEM); + if (WARN_ON_ONCE(tp == NULL)) { + pr_warning("error on getting test probe.\n"); + warn++; + } else + disable_trace_probe(tp, TP_FLAG_TRACE); + + tp = find_trace_probe("testprobe2", KPROBE_EVENT_SYSTEM); + if (WARN_ON_ONCE(tp == NULL)) { + pr_warning("error on getting 2nd test probe.\n"); + warn++; + } else + disable_trace_probe(tp, TP_FLAG_TRACE); + ret = command_trace_probe("-:testprobe"); if (WARN_ON_ONCE(ret)) { pr_warning("error on deleting a probe.\n"); -- cgit v1.3-8-gc7d7 From e0a413f619ef8bc366dafc6f8221674993b8d85f Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 29 Sep 2011 21:26:16 -0400 Subject: tracing: Warn on output if the function tracer was found corrupted As the function tracer is very intrusive, lots of self checks are performed on the tracer and if something is found to be strange it will shut itself down keeping it from corrupting the rest of the kernel. This shutdown may still allow functions to be traced, as the tracing only stops new modifications from happening. Trying to stop the function tracer itself can cause more harm as it requires code modification. Although a WARN_ON() is executed, a user may not notice it. To help the user see that something isn't right with the tracing of the system a big warning is added to the output of the tracer that lets the user know that their data may be incomplete. Reported-by: Thomas Gleixner Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 8 ++++++++ kernel/trace/trace.c | 15 +++++++++++++++ kernel/trace/trace.h | 2 ++ 3 files changed, 25 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index c3e4575e7829..077d85387908 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -3862,6 +3862,14 @@ void ftrace_kill(void) clear_ftrace_function(); } +/** + * Test if ftrace is dead or not. + */ +int ftrace_is_dead(void) +{ + return ftrace_disabled; +} + /** * register_ftrace_function - register a function for profiling * @ops - ops structure that holds the function for profiling. diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 4b8df0dc9358..13f2b8472fed 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2160,6 +2160,14 @@ void trace_default_header(struct seq_file *m) } } +static void test_ftrace_alive(struct seq_file *m) +{ + if (!ftrace_is_dead()) + return; + seq_printf(m, "# WARNING: FUNCTION TRACING IS CORRUPTED\n"); + seq_printf(m, "# MAY BE MISSING FUNCTION EVENTS\n"); +} + static int s_show(struct seq_file *m, void *v) { struct trace_iterator *iter = v; @@ -2169,6 +2177,7 @@ static int s_show(struct seq_file *m, void *v) if (iter->tr) { seq_printf(m, "# tracer: %s\n", iter->trace->name); seq_puts(m, "#\n"); + test_ftrace_alive(m); } if (iter->trace && iter->trace->print_header) iter->trace->print_header(m); @@ -4613,6 +4622,12 @@ __ftrace_dump(bool disable_tracing, enum ftrace_dump_mode oops_dump_mode) tracing_off(); + /* Did function tracer already get disabled? */ + if (ftrace_is_dead()) { + printk("# WARNING: FUNCTION TRACING IS CORRUPTED\n"); + printk("# MAY BE MISSING FUNCTION EVENTS\n"); + } + if (disable_tracing) ftrace_kill(); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 4c7540ad5dcb..092e1f8d18dc 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -579,11 +579,13 @@ static inline int ftrace_trace_task(struct task_struct *task) return test_tsk_trace_trace(task); } +extern int ftrace_is_dead(void); #else static inline int ftrace_trace_task(struct task_struct *task) { return 1; } +static inline int ftrace_is_dead(void) { return 0; } #endif /* -- cgit v1.3-8-gc7d7 From d696b58ca2c3ca76e784ef89a7e0453d9b7ab187 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 22 Sep 2011 11:50:27 -0400 Subject: tracing: Do not allocate buffer for trace_marker When doing intense tracing, the kmalloc inside trace_marker can introduce side effects to what is being traced. As trace_marker() is used by userspace to inject data into the kernel ring buffer, it needs to do so with the least amount of intrusion to the operations of the kernel or the user space application. As the ring buffer is designed to write directly into the buffer without the need to make a temporary buffer, and userspace already went through the hassle of knowing how big the write will be, we can simply pin the userspace pages and write the data directly into the buffer. This improves the impact of tracing via trace_marker tremendously! Thanks to Peter Zijlstra and Thomas Gleixner for pointing out the use of get_user_pages_fast() and kmap_atomic(). Suggested-by: Thomas Gleixner Suggested-by: Peter Zijlstra Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 111 ++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 83 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 13f2b8472fed..f86efe90ca45 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3628,22 +3628,24 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) return 0; } -static int mark_printk(const char *fmt, ...) -{ - int ret; - va_list args; - va_start(args, fmt); - ret = trace_vprintk(0, fmt, args); - va_end(args); - return ret; -} - static ssize_t tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { - char *buf; - size_t written; + unsigned long addr = (unsigned long)ubuf; + struct ring_buffer_event *event; + struct ring_buffer *buffer; + struct print_entry *entry; + unsigned long irq_flags; + struct page *pages[2]; + int nr_pages = 1; + ssize_t written; + void *page1; + void *page2; + int offset; + int size; + int len; + int ret; if (tracing_disabled) return -EINVAL; @@ -3651,28 +3653,81 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, if (cnt > TRACE_BUF_SIZE) cnt = TRACE_BUF_SIZE; - buf = kmalloc(cnt + 2, GFP_KERNEL); - if (buf == NULL) - return -ENOMEM; + /* + * Userspace is injecting traces into the kernel trace buffer. + * We want to be as non intrusive as possible. + * To do so, we do not want to allocate any special buffers + * or take any locks, but instead write the userspace data + * straight into the ring buffer. + * + * First we need to pin the userspace buffer into memory, + * which, most likely it is, because it just referenced it. + * But there's no guarantee that it is. By using get_user_pages_fast() + * and kmap_atomic/kunmap_atomic() we can get access to the + * pages directly. We then write the data directly into the + * ring buffer. + */ + BUILD_BUG_ON(TRACE_BUF_SIZE >= PAGE_SIZE); - if (copy_from_user(buf, ubuf, cnt)) { - kfree(buf); - return -EFAULT; + /* check if we cross pages */ + if ((addr & PAGE_MASK) != ((addr + cnt) & PAGE_MASK)) + nr_pages = 2; + + offset = addr & (PAGE_SIZE - 1); + addr &= PAGE_MASK; + + ret = get_user_pages_fast(addr, nr_pages, 0, pages); + if (ret < nr_pages) { + while (--ret >= 0) + put_page(pages[ret]); + written = -EFAULT; + goto out; } - if (buf[cnt-1] != '\n') { - buf[cnt] = '\n'; - buf[cnt+1] = '\0'; + + page1 = kmap_atomic(pages[0]); + if (nr_pages == 2) + page2 = kmap_atomic(pages[1]); + + local_save_flags(irq_flags); + size = sizeof(*entry) + cnt + 2; /* possible \n added */ + buffer = global_trace.buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, + irq_flags, preempt_count()); + if (!event) { + /* Ring buffer disabled, return as if not open for write */ + written = -EBADF; + goto out_unlock; + } + + entry = ring_buffer_event_data(event); + entry->ip = _THIS_IP_; + + if (nr_pages == 2) { + len = PAGE_SIZE - offset; + memcpy(&entry->buf, page1 + offset, len); + memcpy(&entry->buf[len], page2, cnt - len); } else - buf[cnt] = '\0'; + memcpy(&entry->buf, page1 + offset, cnt); - written = mark_printk("%s", buf); - kfree(buf); - *fpos += written; + if (entry->buf[cnt - 1] != '\n') { + entry->buf[cnt] = '\n'; + entry->buf[cnt + 1] = '\0'; + } else + entry->buf[cnt] = '\0'; + + ring_buffer_unlock_commit(buffer, event); - /* don't tell userspace we wrote more - it might confuse them */ - if (written > cnt) - written = cnt; + written = cnt; + *fpos += written; + + out_unlock: + if (nr_pages == 2) + kunmap_atomic(page2); + kunmap_atomic(page1); + while (nr_pages > 0) + put_page(pages[--nr_pages]); + out: return written; } -- cgit v1.3-8-gc7d7 From 9b5f8b31af57a8ce9e9f77864d9143b5e3304815 Mon Sep 17 00:00:00 2001 From: Geunsik Lim Date: Fri, 12 Aug 2011 14:30:22 +0900 Subject: ftrace: Fix README to state tracing_on to start/stop tracing tracing_enabled option is deprecated. To start/stop tracing, write to /sys/kernel/debug/tracing/tracing_on without tracing_enabled. This patch is based on Linux 3.1.0-rc1 Signed-off-by: Geunsik Lim Link: http://lkml.kernel.org/r/1313127022-23830-1-git-send-email-leemgs1@gmail.com Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index f86efe90ca45..cea16053f553 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2720,9 +2720,9 @@ static const char readme_msg[] = "# cat /sys/kernel/debug/tracing/trace_options\n" "noprint-parent nosym-offset nosym-addr noverbose\n" "# echo print-parent > /sys/kernel/debug/tracing/trace_options\n" - "# echo 1 > /sys/kernel/debug/tracing/tracing_enabled\n" + "# echo 1 > /sys/kernel/debug/tracing/tracing_on\n" "# cat /sys/kernel/debug/tracing/trace > /tmp/trace.txt\n" - "# echo 0 > /sys/kernel/debug/tracing/tracing_enabled\n" + "# echo 0 > /sys/kernel/debug/tracing/tracing_on\n" ; static ssize_t -- cgit v1.3-8-gc7d7 From 436fc280261dcfce5af38f08b89287750dc91cd2 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 14 Oct 2011 10:44:25 -0400 Subject: tracing: Fix returning of duplicate data after EOF in trace_pipe_raw The trace_pipe_raw handler holds a cached page from the time the file is opened to the time it is closed. The cached page is used to handle the case of the user space buffer being smaller than what was read from the ring buffer. The left over buffer is held in the cache so that the next read will continue where the data left off. After EOF is returned (no more data in the buffer), the index of the cached page is set to zero. If a user app reads the page again after EOF, the check in the buffer will see that the cached page is less than page size and will return the cached page again. This will cause reading the trace_pipe_raw again after EOF to return duplicate data, making the output look like the time went backwards but instead data is just repeated. The fix is to not reset the index right after all data is read from the cache, but to reset it after all data is read and more data exists in the ring buffer. Cc: stable Reported-by: Jeremy Eder Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index cea16053f553..b24a72d35008 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3903,8 +3903,6 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, if (info->read < PAGE_SIZE) goto read; - info->read = 0; - trace_access_lock(info->cpu); ret = ring_buffer_read_page(info->tr->buffer, &info->spare, @@ -3914,6 +3912,8 @@ tracing_buffers_read(struct file *filp, char __user *ubuf, if (ret < 0) return 0; + info->read = 0; + read: size = PAGE_SIZE - info->read; if (size > count) -- cgit v1.3-8-gc7d7 From 2a77c46de1e3dace73745015635ebbc648eca69c Mon Sep 17 00:00:00 2001 From: ShuoX Liu Date: Wed, 10 Aug 2011 23:01:26 +0200 Subject: PM / Suspend: Add statistics debugfs file for suspend to RAM Record S3 failure time about each reason and the latest two failed devices' names in S3 progress. We can check it through 'suspend_stats' entry in debugfs. The motivation of the patch: We are enabling power features on Medfield. Comparing with PC/notebook, a mobile enters/exits suspend-2-ram (we call it s3 on Medfield) far more frequently. If it can't enter suspend-2-ram in time, the power might be used up soon. We often find sometimes, a device suspend fails. Then, system retries s3 over and over again. As display is off, testers and developers don't know what happens. Some testers and developers complain they don't know if system tries suspend-2-ram, and what device fails to suspend. They need such info for a quick check. The patch adds suspend_stats under debugfs for users to check suspend to RAM statistics quickly. If not using this patch, we have other methods to get info about what device fails. One is to turn on CONFIG_PM_DEBUG, but users would get too much info and testers need recompile the system. In addition, dynamic debug is another good tool to dump debug info. But it still doesn't match our utilization scenario closely. 1) user need write a user space parser to process the syslog output; 2) Our testing scenario is we leave the mobile for at least hours. Then, check its status. No serial console available during the testing. One is because console would be suspended, and the other is serial console connecting with spi or HSU devices would consume power. These devices are powered off at suspend-2-ram. Signed-off-by: ShuoX Liu Signed-off-by: Rafael J. Wysocki --- Documentation/power/basic-pm-debugging.txt | 24 +++++++ drivers/base/power/main.c | 31 +++++++-- include/linux/suspend.h | 52 +++++++++++++++ kernel/power/main.c | 102 +++++++++++++++++++++++++++++ kernel/power/suspend.c | 17 ++++- 5 files changed, 218 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/Documentation/power/basic-pm-debugging.txt b/Documentation/power/basic-pm-debugging.txt index ddd78172ef73..62eca080a71b 100644 --- a/Documentation/power/basic-pm-debugging.txt +++ b/Documentation/power/basic-pm-debugging.txt @@ -201,3 +201,27 @@ case, you may be able to search for failing drivers by following the procedure analogous to the one described in section 1. If you find some failing drivers, you will have to unload them every time before an STR transition (ie. before you run s2ram), and please report the problems with them. + +There is a debugfs entry which shows the suspend to RAM statistics. Here is an +example of its output. + # mount -t debugfs none /sys/kernel/debug + # cat /sys/kernel/debug/suspend_stats + success: 20 + fail: 5 + failed_freeze: 0 + failed_prepare: 0 + failed_suspend: 5 + failed_suspend_noirq: 0 + failed_resume: 0 + failed_resume_noirq: 0 + failures: + last_failed_dev: alarm + adc + last_failed_errno: -16 + -16 + last_failed_step: suspend + suspend +Field success means the success number of suspend to RAM, and field fail means +the failure number. Others are the failure number of different steps of suspend +to RAM. suspend_stats just lists the last 2 failed devices, error number and +failed step of suspend. diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index c6291ab725a3..b1b58260b4ff 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -46,6 +46,7 @@ LIST_HEAD(dpm_prepared_list); LIST_HEAD(dpm_suspended_list); LIST_HEAD(dpm_noirq_list); +struct suspend_stats suspend_stats; static DEFINE_MUTEX(dpm_list_mtx); static pm_message_t pm_transition; @@ -467,8 +468,12 @@ void dpm_resume_noirq(pm_message_t state) mutex_unlock(&dpm_list_mtx); error = device_resume_noirq(dev, state); - if (error) + if (error) { + suspend_stats.failed_resume_noirq++; + dpm_save_failed_step(SUSPEND_RESUME_NOIRQ); + dpm_save_failed_dev(dev_name(dev)); pm_dev_err(dev, state, " early", error); + } mutex_lock(&dpm_list_mtx); put_device(dev); @@ -629,8 +634,12 @@ void dpm_resume(pm_message_t state) mutex_unlock(&dpm_list_mtx); error = device_resume(dev, state, false); - if (error) + if (error) { + suspend_stats.failed_resume++; + dpm_save_failed_step(SUSPEND_RESUME); + dpm_save_failed_dev(dev_name(dev)); pm_dev_err(dev, state, "", error); + } mutex_lock(&dpm_list_mtx); } @@ -805,6 +814,9 @@ int dpm_suspend_noirq(pm_message_t state) mutex_lock(&dpm_list_mtx); if (error) { pm_dev_err(dev, state, " late", error); + suspend_stats.failed_suspend_noirq++; + dpm_save_failed_step(SUSPEND_SUSPEND_NOIRQ); + dpm_save_failed_dev(dev_name(dev)); put_device(dev); break; } @@ -926,8 +938,10 @@ static void async_suspend(void *data, async_cookie_t cookie) int error; error = __device_suspend(dev, pm_transition, true); - if (error) + if (error) { + dpm_save_failed_dev(dev_name(dev)); pm_dev_err(dev, pm_transition, " async", error); + } put_device(dev); } @@ -970,6 +984,7 @@ int dpm_suspend(pm_message_t state) mutex_lock(&dpm_list_mtx); if (error) { pm_dev_err(dev, state, "", error); + dpm_save_failed_dev(dev_name(dev)); put_device(dev); break; } @@ -983,7 +998,10 @@ int dpm_suspend(pm_message_t state) async_synchronize_full(); if (!error) error = async_error; - if (!error) + if (error) { + suspend_stats.failed_suspend++; + dpm_save_failed_step(SUSPEND_SUSPEND); + } else dpm_show_time(starttime, state, NULL); return error; } @@ -1091,7 +1109,10 @@ int dpm_suspend_start(pm_message_t state) int error; error = dpm_prepare(state); - if (!error) + if (error) { + suspend_stats.failed_prepare++; + dpm_save_failed_step(SUSPEND_PREPARE); + } else error = dpm_suspend(state); return error; } diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 6bbcef22e105..76f42e49b72d 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -34,6 +34,58 @@ typedef int __bitwise suspend_state_t; #define PM_SUSPEND_MEM ((__force suspend_state_t) 3) #define PM_SUSPEND_MAX ((__force suspend_state_t) 4) +enum suspend_stat_step { + SUSPEND_FREEZE = 1, + SUSPEND_PREPARE, + SUSPEND_SUSPEND, + SUSPEND_SUSPEND_NOIRQ, + SUSPEND_RESUME_NOIRQ, + SUSPEND_RESUME +}; + +struct suspend_stats { + int success; + int fail; + int failed_freeze; + int failed_prepare; + int failed_suspend; + int failed_suspend_noirq; + int failed_resume; + int failed_resume_noirq; +#define REC_FAILED_NUM 2 + int last_failed_dev; + char failed_devs[REC_FAILED_NUM][40]; + int last_failed_errno; + int errno[REC_FAILED_NUM]; + int last_failed_step; + enum suspend_stat_step failed_steps[REC_FAILED_NUM]; +}; + +extern struct suspend_stats suspend_stats; + +static inline void dpm_save_failed_dev(const char *name) +{ + strlcpy(suspend_stats.failed_devs[suspend_stats.last_failed_dev], + name, + sizeof(suspend_stats.failed_devs[0])); + suspend_stats.last_failed_dev++; + suspend_stats.last_failed_dev %= REC_FAILED_NUM; +} + +static inline void dpm_save_failed_errno(int err) +{ + suspend_stats.errno[suspend_stats.last_failed_errno] = err; + suspend_stats.last_failed_errno++; + suspend_stats.last_failed_errno %= REC_FAILED_NUM; +} + +static inline void dpm_save_failed_step(enum suspend_stat_step step) +{ + suspend_stats.failed_steps[suspend_stats.last_failed_step] = step; + suspend_stats.last_failed_step++; + suspend_stats.last_failed_step %= REC_FAILED_NUM; +} + /** * struct platform_suspend_ops - Callbacks for managing platform dependent * system sleep states. diff --git a/kernel/power/main.c b/kernel/power/main.c index 6c601f871964..2757acba8e8a 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #include "power.h" @@ -133,6 +135,101 @@ power_attr(pm_test); #endif /* CONFIG_PM_SLEEP */ +#ifdef CONFIG_DEBUG_FS +static char *suspend_step_name(enum suspend_stat_step step) +{ + switch (step) { + case SUSPEND_FREEZE: + return "freeze"; + case SUSPEND_PREPARE: + return "prepare"; + case SUSPEND_SUSPEND: + return "suspend"; + case SUSPEND_SUSPEND_NOIRQ: + return "suspend_noirq"; + case SUSPEND_RESUME_NOIRQ: + return "resume_noirq"; + case SUSPEND_RESUME: + return "resume"; + default: + return ""; + } +} + +static int suspend_stats_show(struct seq_file *s, void *unused) +{ + int i, index, last_dev, last_errno, last_step; + + last_dev = suspend_stats.last_failed_dev + REC_FAILED_NUM - 1; + last_dev %= REC_FAILED_NUM; + last_errno = suspend_stats.last_failed_errno + REC_FAILED_NUM - 1; + last_errno %= REC_FAILED_NUM; + last_step = suspend_stats.last_failed_step + REC_FAILED_NUM - 1; + last_step %= REC_FAILED_NUM; + seq_printf(s, "%s: %d\n%s: %d\n%s: %d\n%s: %d\n" + "%s: %d\n%s: %d\n%s: %d\n%s: %d\n", + "success", suspend_stats.success, + "fail", suspend_stats.fail, + "failed_freeze", suspend_stats.failed_freeze, + "failed_prepare", suspend_stats.failed_prepare, + "failed_suspend", suspend_stats.failed_suspend, + "failed_suspend_noirq", + suspend_stats.failed_suspend_noirq, + "failed_resume", suspend_stats.failed_resume, + "failed_resume_noirq", + suspend_stats.failed_resume_noirq); + seq_printf(s, "failures:\n last_failed_dev:\t%-s\n", + suspend_stats.failed_devs[last_dev]); + for (i = 1; i < REC_FAILED_NUM; i++) { + index = last_dev + REC_FAILED_NUM - i; + index %= REC_FAILED_NUM; + seq_printf(s, "\t\t\t%-s\n", + suspend_stats.failed_devs[index]); + } + seq_printf(s, " last_failed_errno:\t%-d\n", + suspend_stats.errno[last_errno]); + for (i = 1; i < REC_FAILED_NUM; i++) { + index = last_errno + REC_FAILED_NUM - i; + index %= REC_FAILED_NUM; + seq_printf(s, "\t\t\t%-d\n", + suspend_stats.errno[index]); + } + seq_printf(s, " last_failed_step:\t%-s\n", + suspend_step_name( + suspend_stats.failed_steps[last_step])); + for (i = 1; i < REC_FAILED_NUM; i++) { + index = last_step + REC_FAILED_NUM - i; + index %= REC_FAILED_NUM; + seq_printf(s, "\t\t\t%-s\n", + suspend_step_name( + suspend_stats.failed_steps[index])); + } + + return 0; +} + +static int suspend_stats_open(struct inode *inode, struct file *file) +{ + return single_open(file, suspend_stats_show, NULL); +} + +static const struct file_operations suspend_stats_operations = { + .open = suspend_stats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init pm_debugfs_init(void) +{ + debugfs_create_file("suspend_stats", S_IFREG | S_IRUGO, + NULL, NULL, &suspend_stats_operations); + return 0; +} + +late_initcall(pm_debugfs_init); +#endif /* CONFIG_DEBUG_FS */ + struct kobject *power_kobj; /** @@ -194,6 +291,11 @@ static ssize_t state_store(struct kobject *kobj, struct kobj_attribute *attr, } if (state < PM_SUSPEND_MAX && *s) error = enter_state(state); + if (error) { + suspend_stats.fail++; + dpm_save_failed_errno(error); + } else + suspend_stats.success++; #endif Exit: diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index b6b71ad2208f..595a3dd56a8a 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -104,7 +104,10 @@ static int suspend_prepare(void) goto Finish; error = suspend_freeze_processes(); - if (!error) + if (error) { + suspend_stats.failed_freeze++; + dpm_save_failed_step(SUSPEND_FREEZE); + } else return 0; suspend_thaw_processes(); @@ -315,8 +318,16 @@ int enter_state(suspend_state_t state) */ int pm_suspend(suspend_state_t state) { - if (state > PM_SUSPEND_ON && state <= PM_SUSPEND_MAX) - return enter_state(state); + int ret; + if (state > PM_SUSPEND_ON && state <= PM_SUSPEND_MAX) { + ret = enter_state(state); + if (ret) { + suspend_stats.fail++; + dpm_save_failed_errno(ret); + } else + suspend_stats.success++; + return ret; + } return -EINVAL; } EXPORT_SYMBOL(pm_suspend); -- cgit v1.3-8-gc7d7 From ca123102f69fb260221502ade9bbc069290fae84 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 11 Aug 2011 22:38:12 +0200 Subject: PM: Fix build issue in main.c for CONFIG_PM_SLEEP unset Suspend statistics should depend on CONFIG_PM_SLEEP, so make that happen. Signed-off-by: Rafael J. Wysocki --- kernel/power/main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/power/main.c b/kernel/power/main.c index 2757acba8e8a..a52e88425a31 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -133,8 +133,6 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, power_attr(pm_test); #endif /* CONFIG_PM_DEBUG */ -#endif /* CONFIG_PM_SLEEP */ - #ifdef CONFIG_DEBUG_FS static char *suspend_step_name(enum suspend_stat_step step) { @@ -230,6 +228,8 @@ static int __init pm_debugfs_init(void) late_initcall(pm_debugfs_init); #endif /* CONFIG_DEBUG_FS */ +#endif /* CONFIG_PM_SLEEP */ + struct kobject *power_kobj; /** -- cgit v1.3-8-gc7d7 From 85055dd805f0822f13f736bee2a521e222c38293 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 17 Aug 2011 20:42:24 +0200 Subject: PM / Hibernate: Include storage keys in hibernation image on s390 For s390 there is one additional byte associated with each page, the storage key. This byte contains the referenced and changed bits and needs to be included into the hibernation image. If the storage keys are not restored to their previous state all original pages would appear to be dirty. This can cause inconsistencies e.g. with read-only filesystems. Signed-off-by: Martin Schwidefsky Signed-off-by: Rafael J. Wysocki --- arch/s390/Kconfig | 1 + arch/s390/kernel/suspend.c | 118 ++++++++++++++++++++++++++++++++++++++++ arch/s390/kernel/swsusp_asm64.S | 3 + include/linux/suspend.h | 34 ++++++++++++ kernel/power/Kconfig | 3 + kernel/power/snapshot.c | 18 ++++++ 6 files changed, 177 insertions(+) (limited to 'kernel') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index ed5cb5af5281..6b99fc3f9b63 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -91,6 +91,7 @@ config S390 select HAVE_ARCH_MUTEX_CPU_RELAX select HAVE_ARCH_JUMP_LABEL if !MARCH_G5 select HAVE_RCU_TABLE_FREE if SMP + select ARCH_SAVE_PAGE_KEYS if HIBERNATION select ARCH_INLINE_SPIN_TRYLOCK select ARCH_INLINE_SPIN_TRYLOCK_BH select ARCH_INLINE_SPIN_LOCK diff --git a/arch/s390/kernel/suspend.c b/arch/s390/kernel/suspend.c index cf9e5c6d5527..b6f9afed74ec 100644 --- a/arch/s390/kernel/suspend.c +++ b/arch/s390/kernel/suspend.c @@ -7,6 +7,7 @@ */ #include +#include #include /* @@ -14,6 +15,123 @@ */ extern const void __nosave_begin, __nosave_end; +/* + * The restore of the saved pages in an hibernation image will set + * the change and referenced bits in the storage key for each page. + * Overindication of the referenced bits after an hibernation cycle + * does not cause any harm but the overindication of the change bits + * would cause trouble. + * Use the ARCH_SAVE_PAGE_KEYS hooks to save the storage key of each + * page to the most significant byte of the associated page frame + * number in the hibernation image. + */ + +/* + * Key storage is allocated as a linked list of pages. + * The size of the keys array is (PAGE_SIZE - sizeof(long)) + */ +struct page_key_data { + struct page_key_data *next; + unsigned char data[]; +}; + +#define PAGE_KEY_DATA_SIZE (PAGE_SIZE - sizeof(struct page_key_data *)) + +static struct page_key_data *page_key_data; +static struct page_key_data *page_key_rp, *page_key_wp; +static unsigned long page_key_rx, page_key_wx; + +/* + * For each page in the hibernation image one additional byte is + * stored in the most significant byte of the page frame number. + * On suspend no additional memory is required but on resume the + * keys need to be memorized until the page data has been restored. + * Only then can the storage keys be set to their old state. + */ +unsigned long page_key_additional_pages(unsigned long pages) +{ + return DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE); +} + +/* + * Free page_key_data list of arrays. + */ +void page_key_free(void) +{ + struct page_key_data *pkd; + + while (page_key_data) { + pkd = page_key_data; + page_key_data = pkd->next; + free_page((unsigned long) pkd); + } +} + +/* + * Allocate page_key_data list of arrays with enough room to store + * one byte for each page in the hibernation image. + */ +int page_key_alloc(unsigned long pages) +{ + struct page_key_data *pk; + unsigned long size; + + size = DIV_ROUND_UP(pages, PAGE_KEY_DATA_SIZE); + while (size--) { + pk = (struct page_key_data *) get_zeroed_page(GFP_KERNEL); + if (!pk) { + page_key_free(); + return -ENOMEM; + } + pk->next = page_key_data; + page_key_data = pk; + } + page_key_rp = page_key_wp = page_key_data; + page_key_rx = page_key_wx = 0; + return 0; +} + +/* + * Save the storage key into the upper 8 bits of the page frame number. + */ +void page_key_read(unsigned long *pfn) +{ + unsigned long addr; + + addr = (unsigned long) page_address(pfn_to_page(*pfn)); + *(unsigned char *) pfn = (unsigned char) page_get_storage_key(addr); +} + +/* + * Extract the storage key from the upper 8 bits of the page frame number + * and store it in the page_key_data list of arrays. + */ +void page_key_memorize(unsigned long *pfn) +{ + page_key_wp->data[page_key_wx] = *(unsigned char *) pfn; + *(unsigned char *) pfn = 0; + if (++page_key_wx < PAGE_KEY_DATA_SIZE) + return; + page_key_wp = page_key_wp->next; + page_key_wx = 0; +} + +/* + * Get the next key from the page_key_data list of arrays and set the + * storage key of the page referred by @address. If @address refers to + * a "safe" page the swsusp_arch_resume code will transfer the storage + * key from the buffer page to the original page. + */ +void page_key_write(void *address) +{ + page_set_storage_key((unsigned long) address, + page_key_rp->data[page_key_rx], 0); + if (++page_key_rx >= PAGE_KEY_DATA_SIZE) + return; + page_key_rp = page_key_rp->next; + page_key_rx = 0; +} + int pfn_is_nosave(unsigned long pfn) { unsigned long nosave_begin_pfn = PFN_DOWN(__pa(&__nosave_begin)); diff --git a/arch/s390/kernel/swsusp_asm64.S b/arch/s390/kernel/swsusp_asm64.S index 51bcdb50a230..acb78cdee896 100644 --- a/arch/s390/kernel/swsusp_asm64.S +++ b/arch/s390/kernel/swsusp_asm64.S @@ -136,11 +136,14 @@ ENTRY(swsusp_arch_resume) 0: lg %r2,8(%r1) lg %r4,0(%r1) + iske %r0,%r4 lghi %r3,PAGE_SIZE lghi %r5,PAGE_SIZE 1: mvcle %r2,%r4,0 jo 1b + lg %r2,8(%r1) + sske %r0,%r2 lg %r1,16(%r1) ltgr %r1,%r1 jnz 0b diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 76f42e49b72d..46f3548aef2d 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -386,4 +386,38 @@ static inline void unlock_system_sleep(void) } #endif +#ifdef CONFIG_ARCH_SAVE_PAGE_KEYS +/* + * The ARCH_SAVE_PAGE_KEYS functions can be used by an architecture + * to save/restore additional information to/from the array of page + * frame numbers in the hibernation image. For s390 this is used to + * save and restore the storage key for each page that is included + * in the hibernation image. + */ +unsigned long page_key_additional_pages(unsigned long pages); +int page_key_alloc(unsigned long pages); +void page_key_free(void); +void page_key_read(unsigned long *pfn); +void page_key_memorize(unsigned long *pfn); +void page_key_write(void *address); + +#else /* !CONFIG_ARCH_SAVE_PAGE_KEYS */ + +static inline unsigned long page_key_additional_pages(unsigned long pages) +{ + return 0; +} + +static inline int page_key_alloc(unsigned long pages) +{ + return 0; +} + +static inline void page_key_free(void) {} +static inline void page_key_read(unsigned long *pfn) {} +static inline void page_key_memorize(unsigned long *pfn) {} +static inline void page_key_write(void *address) {} + +#endif /* !CONFIG_ARCH_SAVE_PAGE_KEYS */ + #endif /* _LINUX_SUSPEND_H */ diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 3744c594b19b..e01e6899592c 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -65,6 +65,9 @@ config HIBERNATION For more information take a look at . +config ARCH_SAVE_PAGE_KEYS + bool + config PM_STD_PARTITION string "Default resume partition" depends on HIBERNATION diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 06efa54f93d6..cbe2c1441392 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1339,6 +1339,9 @@ int hibernate_preallocate_memory(void) count += highmem; count -= totalreserve_pages; + /* Add number of pages required for page keys (s390 only). */ + size += page_key_additional_pages(saveable); + /* Compute the maximum number of saveable pages to leave in memory. */ max_size = (count - (size + PAGES_FOR_IO)) / 2 - 2 * DIV_ROUND_UP(reserved_size, PAGE_SIZE); @@ -1662,6 +1665,8 @@ pack_pfns(unsigned long *buf, struct memory_bitmap *bm) buf[j] = memory_bm_next_pfn(bm); if (unlikely(buf[j] == BM_END_OF_MAP)) break; + /* Save page key for data page (s390 only). */ + page_key_read(buf + j); } } @@ -1821,6 +1826,9 @@ static int unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) if (unlikely(buf[j] == BM_END_OF_MAP)) break; + /* Extract and buffer page key for data page (s390 only). */ + page_key_memorize(buf + j); + if (memory_bm_pfn_present(bm, buf[j])) memory_bm_set_bit(bm, buf[j]); else @@ -2223,6 +2231,11 @@ int snapshot_write_next(struct snapshot_handle *handle) if (error) return error; + /* Allocate buffer for page keys. */ + error = page_key_alloc(nr_copy_pages); + if (error) + return error; + } else if (handle->cur <= nr_meta_pages + 1) { error = unpack_orig_pfns(buffer, ©_bm); if (error) @@ -2243,6 +2256,8 @@ int snapshot_write_next(struct snapshot_handle *handle) } } else { copy_last_highmem_page(); + /* Restore page key for data page (s390 only). */ + page_key_write(handle->buffer); handle->buffer = get_buffer(&orig_bm, &ca); if (IS_ERR(handle->buffer)) return PTR_ERR(handle->buffer); @@ -2264,6 +2279,9 @@ int snapshot_write_next(struct snapshot_handle *handle) void snapshot_write_finalize(struct snapshot_handle *handle) { copy_last_highmem_page(); + /* Restore page key for data page (s390 only). */ + page_key_write(handle->buffer); + page_key_free(); /* Free only if we have loaded the image entirely */ if (handle->cur > 1 && handle->cur > nr_meta_pages + nr_copy_pages) { memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR); -- cgit v1.3-8-gc7d7 From 528f7ce6e439edeac38f6b3f8561f1be129b5e91 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 21 Sep 2011 20:55:04 +0200 Subject: PM / Suspend: Off by one in pm_suspend() In enter_state() we use "state" as an offset for the pm_states[] array. The pm_states[] array only has PM_SUSPEND_MAX elements so this test is off by one. Signed-off-by: Dan Carpenter Signed-off-by: Rafael J. Wysocki Cc: stable@kernel.org --- kernel/power/suspend.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 595a3dd56a8a..fdd4263b995d 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -319,7 +319,7 @@ int enter_state(suspend_state_t state) int pm_suspend(suspend_state_t state) { int ret; - if (state > PM_SUSPEND_ON && state <= PM_SUSPEND_MAX) { + if (state > PM_SUSPEND_ON && state < PM_SUSPEND_MAX) { ret = enter_state(state); if (ret) { suspend_stats.fail++; -- cgit v1.3-8-gc7d7 From 37cce26b32142f09a8967f6d238178af654b20de Mon Sep 17 00:00:00 2001 From: H Hartley Sweeten Date: Wed, 21 Sep 2011 22:47:55 +0200 Subject: PM / VT: Cleanup #if defined uglyness and fix compile error Introduce the config option CONFIG_VT_CONSOLE_SLEEP in order to cleanup the #if defined ugliness for the vt suspend support functions. Note that CONFIG_VT_CONSOLE is already dependant on CONFIG_VT. The function pm_set_vt_switch is actually dependant on CONFIG_VT and not CONFIG_PM_SLEEP. This fixes a compile error when CONFIG_PM_SLEEP is not set: drivers/tty/vt/vt_ioctl.c:1794: error: redefinition of 'pm_set_vt_switch' include/linux/suspend.h:17: error: previous definition of 'pm_set_vt_switch' was here Also, remove the incorrect path from the comment in console.c. [rjw: Replaced #if defined() with #ifdef in suspend.h.] Signed-off-by: H Hartley Sweeten Acked-by: Arnd Bergmann Signed-off-by: Rafael J. Wysocki --- drivers/tty/Kconfig | 4 ++++ include/linux/suspend.h | 9 ++++++--- kernel/power/Makefile | 2 +- kernel/power/console.c | 4 +--- 4 files changed, 12 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/drivers/tty/Kconfig b/drivers/tty/Kconfig index bd7cc0527999..a7188a0b68eb 100644 --- a/drivers/tty/Kconfig +++ b/drivers/tty/Kconfig @@ -60,6 +60,10 @@ config VT_CONSOLE If unsure, say Y. +config VT_CONSOLE_SLEEP + def_bool y + depends on VT_CONSOLE && PM_SLEEP + config HW_CONSOLE bool depends on VT && !S390 && !UML diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 46f3548aef2d..57a692432f8a 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -8,15 +8,18 @@ #include #include -#if defined(CONFIG_PM_SLEEP) && defined(CONFIG_VT) && defined(CONFIG_VT_CONSOLE) +#ifdef CONFIG_VT extern void pm_set_vt_switch(int); -extern int pm_prepare_console(void); -extern void pm_restore_console(void); #else static inline void pm_set_vt_switch(int do_switch) { } +#endif +#ifdef CONFIG_VT_CONSOLE_SLEEP +extern int pm_prepare_console(void); +extern void pm_restore_console(void); +#else static inline int pm_prepare_console(void) { return 0; diff --git a/kernel/power/Makefile b/kernel/power/Makefile index ad6bdd8b401a..07e0e28ffba7 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -2,7 +2,7 @@ ccflags-$(CONFIG_PM_DEBUG) := -DDEBUG obj-$(CONFIG_PM) += main.o qos.o -obj-$(CONFIG_PM_SLEEP) += console.o +obj-$(CONFIG_VT_CONSOLE_SLEEP) += console.o obj-$(CONFIG_FREEZER) += process.o obj-$(CONFIG_SUSPEND) += suspend.o obj-$(CONFIG_PM_TEST_SUSPEND) += suspend_test.o diff --git a/kernel/power/console.c b/kernel/power/console.c index 218e5af90156..b1dc456474b5 100644 --- a/kernel/power/console.c +++ b/kernel/power/console.c @@ -1,5 +1,5 @@ /* - * drivers/power/process.c - Functions for saving/restoring console. + * Functions for saving/restoring console. * * Originally from swsusp. */ @@ -10,7 +10,6 @@ #include #include "power.h" -#if defined(CONFIG_VT) && defined(CONFIG_VT_CONSOLE) #define SUSPEND_CONSOLE (MAX_NR_CONSOLES-1) static int orig_fgconsole, orig_kmsg; @@ -32,4 +31,3 @@ void pm_restore_console(void) vt_kmsg_redirect(orig_kmsg); } } -#endif -- cgit v1.3-8-gc7d7 From 2aede851ddf08666f68ffc17be446420e9d2a056 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 26 Sep 2011 20:32:27 +0200 Subject: PM / Hibernate: Freeze kernel threads after preallocating memory There is a problem with the current ordering of hibernate code which leads to deadlocks in some filesystems' memory shrinkers. Namely, some filesystems use freezable kernel threads that are inactive when the hibernate memory preallocation is carried out. Those same filesystems use memory shrinkers that may be triggered by the hibernate memory preallocation. If those memory shrinkers wait for the frozen kernel threads, the hibernate process deadlocks (this happens with XFS, for one example). Apparently, it is not technically viable to redesign the filesystems in question to avoid the situation described above, so the only possible solution of this issue is to defer the freezing of kernel threads until the hibernate memory preallocation is done, which is implemented by this change. Unfortunately, this requires the memory preallocation to be done before the "prepare" stage of device freeze, so after this change the only way drivers can allocate additional memory for their freeze routines in a clean way is to use PM notifiers. Reported-by: Christoph Signed-off-by: Rafael J. Wysocki --- Documentation/power/devices.txt | 4 ---- include/linux/freezer.h | 4 +++- kernel/power/hibernate.c | 12 ++++++++---- kernel/power/power.h | 3 ++- kernel/power/process.c | 30 ++++++++++++++++++++---------- 5 files changed, 33 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt index 29b7a9817f5a..646a89e0c07d 100644 --- a/Documentation/power/devices.txt +++ b/Documentation/power/devices.txt @@ -281,10 +281,6 @@ When the system goes into the standby or memory sleep state, the phases are: time.) Unlike the other suspend-related phases, during the prepare phase the device tree is traversed top-down. - In addition to that, if device drivers need to allocate additional - memory to be able to hadle device suspend correctly, that should be - done in the prepare phase. - After the prepare callback method returns, no new children may be registered below the device. The method may also prepare the device or driver in some way for the upcoming system power transition (for diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 1effc8b56b4e..aa56cf31f7ff 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -49,6 +49,7 @@ extern int thaw_process(struct task_struct *p); extern void refrigerator(void); extern int freeze_processes(void); +extern int freeze_kernel_threads(void); extern void thaw_processes(void); static inline int try_to_freeze(void) @@ -171,7 +172,8 @@ static inline void clear_freeze_flag(struct task_struct *p) {} static inline int thaw_process(struct task_struct *p) { return 1; } static inline void refrigerator(void) {} -static inline int freeze_processes(void) { BUG(); return 0; } +static inline int freeze_processes(void) { return -ENOSYS; } +static inline int freeze_kernel_threads(void) { return -ENOSYS; } static inline void thaw_processes(void) {} static inline int try_to_freeze(void) { return 0; } diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 8f7b1db1ece1..3a20466015f8 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -334,12 +334,16 @@ int hibernation_snapshot(int platform_mode) if (error) goto Close; - error = dpm_prepare(PMSG_FREEZE); - if (error) - goto Complete_devices; - /* Preallocate image memory before shutting down devices. */ error = hibernate_preallocate_memory(); + if (error) + goto Close; + + error = freeze_kernel_threads(); + if (error) + goto Close; + + error = dpm_prepare(PMSG_FREEZE); if (error) goto Complete_devices; diff --git a/kernel/power/power.h b/kernel/power/power.h index 9a00a0a26280..e6206397ce67 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -228,7 +228,8 @@ extern int pm_test_level; #ifdef CONFIG_SUSPEND_FREEZER static inline int suspend_freeze_processes(void) { - return freeze_processes(); + int error = freeze_processes(); + return error ? : freeze_kernel_threads(); } static inline void suspend_thaw_processes(void) diff --git a/kernel/power/process.c b/kernel/power/process.c index 0cf3a27a6c9d..addbbe5531bc 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -135,7 +135,7 @@ static int try_to_freeze_tasks(bool sig_only) } /** - * freeze_processes - tell processes to enter the refrigerator + * freeze_processes - Signal user space processes to enter the refrigerator. */ int freeze_processes(void) { @@ -143,20 +143,30 @@ int freeze_processes(void) printk("Freezing user space processes ... "); error = try_to_freeze_tasks(true); - if (error) - goto Exit; - printk("done.\n"); + if (!error) { + printk("done."); + oom_killer_disable(); + } + printk("\n"); + BUG_ON(in_atomic()); + + return error; +} + +/** + * freeze_kernel_threads - Make freezable kernel threads go to the refrigerator. + */ +int freeze_kernel_threads(void) +{ + int error; printk("Freezing remaining freezable tasks ... "); error = try_to_freeze_tasks(false); - if (error) - goto Exit; - printk("done."); + if (!error) + printk("done."); - oom_killer_disable(); - Exit: - BUG_ON(in_atomic()); printk("\n"); + BUG_ON(in_atomic()); return error; } -- cgit v1.3-8-gc7d7 From 21e82808fc465b66fedaac0f4e885cafb304e843 Mon Sep 17 00:00:00 2001 From: Barry Song Date: Tue, 27 Sep 2011 22:05:44 +0200 Subject: PM / Hibernate: Fix typo in a kerneldoc comment Fix a typo in a function name in the kerneldoc comment next to resume_target_kernel(). [rjw: Changed the subject slightly, added the changelog.] Signed-off-by: Barry Song Signed-off-by: Rafael J. Wysocki --- kernel/power/hibernate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 3a20466015f8..7f44e5c26971 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -467,7 +467,7 @@ static int resume_target_kernel(bool platform_mode) * @platform_mode: If set, use platform driver to prepare for the transition. * * This routine must be called with pm_mutex held. If it is successful, control - * reappears in the restored target kernel in hibernation_snaphot(). + * reappears in the restored target kernel in hibernation_snapshot(). */ int hibernation_restore(int platform_mode) { -- cgit v1.3-8-gc7d7 From 6f8d7022a842809aeb24db1d15669198ef02c131 Mon Sep 17 00:00:00 2001 From: Barry Song Date: Thu, 6 Oct 2011 20:34:46 +0200 Subject: PM / Hibernate: Add resumewait param to support MMC-like devices as resume file Some devices like MMC are async detected very slow. For example, drivers/mmc/host/sdhci.c launches a 200ms delayed work to detect MMC partitions then add disk. We have wait_for_device_probe() and scsi_complete_async_scans() before calling swsusp_check(), but it is not enough to wait for MMC. This patch adds resumewait kernel param just like rootwait so that we have enough time to wait until MMC is ready. The difference is that we wait for resume partition whereas rootwait waits for rootfs partition (which may be on a different device). This patch will make hibernation support many embedded products without SCSI devices, but with devices like MMC. [rjw: Modified the changelog slightly.] Signed-off-by: Barry Song Reviewed-by: Valdis Kletnieks Signed-off-by: Rafael J. Wysocki --- Documentation/kernel-parameters.txt | 4 ++++ kernel/power/hibernate.c | 16 ++++++++++++++++ 2 files changed, 20 insertions(+) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 854ed5ca7e3f..88a7b197a4eb 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2240,6 +2240,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted. in units (needed only for swap files). See Documentation/power/swsusp-and-swap-files.txt + resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. + Useful for devices that are detected asynchronously + (e.g. USB and MMC devices). + hibernate= [HIBERNATION] noresume Don't check if there's a hibernation image present during boot. diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 7f44e5c26971..0f8785080cde 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -31,6 +32,7 @@ static int nocompress = 0; static int noresume = 0; +static int resume_wait = 0; static char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; sector_t swsusp_resume_block; @@ -736,6 +738,13 @@ static int software_resume(void) * to wait for this to finish. */ wait_for_device_probe(); + + if (resume_wait) { + while ((swsusp_resume_device = name_to_dev_t(resume_file)) == 0) + msleep(10); + async_synchronize_full(); + } + /* * We can't depend on SCSI devices being available after loading * one of their modules until scsi_complete_async_scans() is @@ -1064,7 +1073,14 @@ static int __init noresume_setup(char *str) return 1; } +static int __init resumewait_setup(char *str) +{ + resume_wait = 1; + return 1; +} + __setup("noresume", noresume_setup); __setup("resume_offset=", resume_offset_setup); __setup("resume=", resume_setup); __setup("hibernate=", hibernate_setup); +__setup("resumewait", resumewait_setup); -- cgit v1.3-8-gc7d7 From f126f7334f72e2fd1b7a62bba20c488b86e6e7c4 Mon Sep 17 00:00:00 2001 From: Barry Song Date: Mon, 10 Oct 2011 23:38:41 +0200 Subject: PM / Hibernate: Add resumedelay kernel param in addition to resumewait Patch "PM / Hibernate: Add resumewait param to support MMC-like devices as resume file" added the resumewait kernel command line option. The present patch adds resumedelay so that resumewait/delay were analogous to rootwait/delay. [rjw: Modified the subject and changelog slightly.] Signed-off-by: Barry Song Acked-by: Pavel Machek Signed-off-by: Rafael J. Wysocki --- Documentation/kernel-parameters.txt | 3 +++ kernel/power/hibernate.c | 14 ++++++++++++++ 2 files changed, 17 insertions(+) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 88a7b197a4eb..831bde222bde 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2240,6 +2240,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted. in units (needed only for swap files). See Documentation/power/swsusp-and-swap-files.txt + resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to + read the resume files + resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. Useful for devices that are detected asynchronously (e.g. USB and MMC devices). diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 0f8785080cde..50f537953e70 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -33,6 +33,7 @@ static int nocompress = 0; static int noresume = 0; static int resume_wait = 0; +static int resume_delay = 0; static char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; sector_t swsusp_resume_block; @@ -730,6 +731,12 @@ static int software_resume(void) pr_debug("PM: Checking hibernation image partition %s\n", resume_file); + if (resume_delay) { + printk(KERN_INFO "Waiting %dsec before reading resume device...\n", + resume_delay); + ssleep(resume_delay); + } + /* Check if the device is there */ swsusp_resume_device = name_to_dev_t(resume_file); if (!swsusp_resume_device) { @@ -1079,8 +1086,15 @@ static int __init resumewait_setup(char *str) return 1; } +static int __init resumedelay_setup(char *str) +{ + resume_delay = simple_strtoul(str, NULL, 0); + return 1; +} + __setup("noresume", noresume_setup); __setup("resume_offset=", resume_offset_setup); __setup("resume=", resume_setup); __setup("hibernate=", hibernate_setup); __setup("resumewait", resumewait_setup); +__setup("resumedelay=", resumedelay_setup); -- cgit v1.3-8-gc7d7 From 27920651fe0d16a25632dc238e81b54f3a504869 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Tue, 11 Oct 2011 21:20:55 +0200 Subject: PM / Freezer: Make fake_signal_wake_up() wake TASK_KILLABLE tasks too TASK_KILLABLE is often used to put tasks to sleep for quite some time. One of the most common uses is to put tasks to sleep while waiting for replies from a server on a networked filesystem (such as CIFS or NFS). Unfortunately, fake_signal_wake_up does not currently wake up tasks that are sleeping in TASK_KILLABLE state. This means that even if the code were in place to allow them to freeze while in this sleep, it wouldn't work anyway. This patch changes this function to wake tasks in this state as well. This should be harmless -- if the code doing the sleeping doesn't have handling to deal with freezer events, it should just go back to sleep. If it does, then this will allow that code to do the right thing. Signed-off-by: Jeff Layton Signed-off-by: Rafael J. Wysocki --- kernel/freezer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/freezer.c b/kernel/freezer.c index 7b01de98bb6a..66a594e8ad2f 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -67,7 +67,7 @@ static void fake_signal_wake_up(struct task_struct *p) unsigned long flags; spin_lock_irqsave(&p->sighand->siglock, flags); - signal_wake_up(p, 0); + signal_wake_up(p, 1); spin_unlock_irqrestore(&p->sighand->siglock, flags); } -- cgit v1.3-8-gc7d7 From d231ff1af70a2df43d809173cf8c94e9c3beb853 Mon Sep 17 00:00:00 2001 From: Barry Song Date: Tue, 11 Oct 2011 23:29:18 -0700 Subject: PM / Hibernate: Do not initialize static and extern variables to 0 Static and extern variables in kernel/power/hibernate.c need not be initialized to 0 explicitly, so remove those initializations. [rjw: Modified subject, added changelog.] Signed-off-by: Barry Song Signed-off-by: Rafael J. Wysocki --- kernel/power/hibernate.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 50f537953e70..ea12c8f1bdfd 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -30,14 +30,14 @@ #include "power.h" -static int nocompress = 0; -static int noresume = 0; -static int resume_wait = 0; -static int resume_delay = 0; +static int nocompress; +static int noresume; +static int resume_wait; +static int resume_delay; static char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; sector_t swsusp_resume_block; -int in_suspend __nosavedata = 0; +int in_suspend __nosavedata; enum { HIBERNATION_INVALID, -- cgit v1.3-8-gc7d7 From 081a9d043c983f161b78fdc4671324d1342b86bc Mon Sep 17 00:00:00 2001 From: Bojan Smojver Date: Thu, 13 Oct 2011 23:58:07 +0200 Subject: PM / Hibernate: Improve performance of LZO/plain hibernation, checksum image Use threads for LZO compression/decompression on hibernate/thaw. Improve buffering on hibernate/thaw. Calculate/verify CRC32 of the image pages on hibernate/thaw. In my testing, this improved write/read speed by a factor of about two. Signed-off-by: Bojan Smojver Signed-off-by: Rafael J. Wysocki --- kernel/power/Kconfig | 1 + kernel/power/hibernate.c | 3 + kernel/power/power.h | 1 + kernel/power/swap.c | 818 ++++++++++++++++++++++++++++++++++++----------- 4 files changed, 645 insertions(+), 178 deletions(-) (limited to 'kernel') diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index e01e6899592c..cedd9982306a 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -27,6 +27,7 @@ config HIBERNATION select HIBERNATE_CALLBACKS select LZO_COMPRESS select LZO_DECOMPRESS + select CRC32 ---help--- Enable the suspend to disk (STD) functionality, which is usually called "hibernation" in user interfaces. STD checkpoints the diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index ea12c8f1bdfd..1c53f7fad5f7 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -657,6 +657,9 @@ int hibernate(void) flags |= SF_PLATFORM_MODE; if (nocompress) flags |= SF_NOCOMPRESS_MODE; + else + flags |= SF_CRC32_MODE; + pr_debug("PM: writing image.\n"); error = swsusp_write(flags); swsusp_free(); diff --git a/kernel/power/power.h b/kernel/power/power.h index e6206397ce67..23a2db1ec442 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -146,6 +146,7 @@ extern int swsusp_swap_in_use(void); */ #define SF_PLATFORM_MODE 1 #define SF_NOCOMPRESS_MODE 2 +#define SF_CRC32_MODE 4 /* kernel/power/hibernate.c */ extern int swsusp_check(void); diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 7c97c3a0eee3..11a594c4ba25 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -27,6 +27,10 @@ #include #include #include +#include +#include +#include +#include #include "power.h" @@ -43,8 +47,7 @@ * allocated and populated one at a time, so we only need one memory * page to set up the entire structure. * - * During resume we also only need to use one swap_map_page structure - * at a time. + * During resume we pick up all swap_map_page structures into a list. */ #define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(sector_t) - 1) @@ -54,6 +57,11 @@ struct swap_map_page { sector_t next_swap; }; +struct swap_map_page_list { + struct swap_map_page *map; + struct swap_map_page_list *next; +}; + /** * The swap_map_handle structure is used for handling swap in * a file-alike way @@ -61,13 +69,18 @@ struct swap_map_page { struct swap_map_handle { struct swap_map_page *cur; + struct swap_map_page_list *maps; sector_t cur_swap; sector_t first_sector; unsigned int k; + unsigned long nr_free_pages, written; + u32 crc32; }; struct swsusp_header { - char reserved[PAGE_SIZE - 20 - sizeof(sector_t) - sizeof(int)]; + char reserved[PAGE_SIZE - 20 - sizeof(sector_t) - sizeof(int) - + sizeof(u32)]; + u32 crc32; sector_t image; unsigned int flags; /* Flags to pass to the "boot" kernel */ char orig_sig[10]; @@ -199,6 +212,8 @@ static int mark_swapfiles(struct swap_map_handle *handle, unsigned int flags) memcpy(swsusp_header->sig, HIBERNATE_SIG, 10); swsusp_header->image = handle->first_sector; swsusp_header->flags = flags; + if (flags & SF_CRC32_MODE) + swsusp_header->crc32 = handle->crc32; error = hib_bio_write_page(swsusp_resume_block, swsusp_header, NULL); } else { @@ -245,6 +260,7 @@ static int swsusp_swap_check(void) static int write_page(void *buf, sector_t offset, struct bio **bio_chain) { void *src; + int ret; if (!offset) return -ENOSPC; @@ -254,9 +270,17 @@ static int write_page(void *buf, sector_t offset, struct bio **bio_chain) if (src) { copy_page(src, buf); } else { - WARN_ON_ONCE(1); - bio_chain = NULL; /* Go synchronous */ - src = buf; + ret = hib_wait_on_bio_chain(bio_chain); /* Free pages */ + if (ret) + return ret; + src = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); + if (src) { + copy_page(src, buf); + } else { + WARN_ON_ONCE(1); + bio_chain = NULL; /* Go synchronous */ + src = buf; + } } } else { src = buf; @@ -293,6 +317,8 @@ static int get_swap_writer(struct swap_map_handle *handle) goto err_rel; } handle->k = 0; + handle->nr_free_pages = nr_free_pages() >> 1; + handle->written = 0; handle->first_sector = handle->cur_swap; return 0; err_rel: @@ -316,20 +342,23 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, return error; handle->cur->entries[handle->k++] = offset; if (handle->k >= MAP_PAGE_ENTRIES) { - error = hib_wait_on_bio_chain(bio_chain); - if (error) - goto out; offset = alloc_swapdev_block(root_swap); if (!offset) return -ENOSPC; handle->cur->next_swap = offset; - error = write_page(handle->cur, handle->cur_swap, NULL); + error = write_page(handle->cur, handle->cur_swap, bio_chain); if (error) goto out; clear_page(handle->cur); handle->cur_swap = offset; handle->k = 0; } + if (bio_chain && ++handle->written > handle->nr_free_pages) { + error = hib_wait_on_bio_chain(bio_chain); + if (error) + goto out; + handle->written = 0; + } out: return error; } @@ -372,6 +401,13 @@ static int swap_writer_finish(struct swap_map_handle *handle, LZO_HEADER, PAGE_SIZE) #define LZO_CMP_SIZE (LZO_CMP_PAGES * PAGE_SIZE) +/* Maximum number of threads for compression/decompression. */ +#define LZO_THREADS 3 + +/* Maximum number of pages for read buffering. */ +#define LZO_READ_PAGES (MAP_PAGE_ENTRIES * 8) + + /** * save_image - save the suspend image data */ @@ -419,6 +455,92 @@ static int save_image(struct swap_map_handle *handle, return ret; } +/** + * Structure used for CRC32. + */ +struct crc_data { + struct task_struct *thr; /* thread */ + atomic_t ready; /* ready to start flag */ + atomic_t stop; /* ready to stop flag */ + unsigned run_threads; /* nr current threads */ + wait_queue_head_t go; /* start crc update */ + wait_queue_head_t done; /* crc update done */ + u32 *crc32; /* points to handle's crc32 */ + size_t *unc_len[LZO_THREADS]; /* uncompressed lengths */ + unsigned char *unc[LZO_THREADS]; /* uncompressed data */ +}; + +/** + * CRC32 update function that runs in its own thread. + */ +static int crc32_threadfn(void *data) +{ + struct crc_data *d = data; + unsigned i; + + while (1) { + wait_event(d->go, atomic_read(&d->ready) || + kthread_should_stop()); + if (kthread_should_stop()) { + d->thr = NULL; + atomic_set(&d->stop, 1); + wake_up(&d->done); + break; + } + atomic_set(&d->ready, 0); + + for (i = 0; i < d->run_threads; i++) + *d->crc32 = crc32_le(*d->crc32, + d->unc[i], *d->unc_len[i]); + atomic_set(&d->stop, 1); + wake_up(&d->done); + } + return 0; +} +/** + * Structure used for LZO data compression. + */ +struct cmp_data { + struct task_struct *thr; /* thread */ + atomic_t ready; /* ready to start flag */ + atomic_t stop; /* ready to stop flag */ + int ret; /* return code */ + wait_queue_head_t go; /* start compression */ + wait_queue_head_t done; /* compression done */ + size_t unc_len; /* uncompressed length */ + size_t cmp_len; /* compressed length */ + unsigned char unc[LZO_UNC_SIZE]; /* uncompressed buffer */ + unsigned char cmp[LZO_CMP_SIZE]; /* compressed buffer */ + unsigned char wrk[LZO1X_1_MEM_COMPRESS]; /* compression workspace */ +}; + +/** + * Compression function that runs in its own thread. + */ +static int lzo_compress_threadfn(void *data) +{ + struct cmp_data *d = data; + + while (1) { + wait_event(d->go, atomic_read(&d->ready) || + kthread_should_stop()); + if (kthread_should_stop()) { + d->thr = NULL; + d->ret = -1; + atomic_set(&d->stop, 1); + wake_up(&d->done); + break; + } + atomic_set(&d->ready, 0); + + d->ret = lzo1x_1_compress(d->unc, d->unc_len, + d->cmp + LZO_HEADER, &d->cmp_len, + d->wrk); + atomic_set(&d->stop, 1); + wake_up(&d->done); + } + return 0; +} /** * save_image_lzo - Save the suspend image data compressed with LZO. @@ -437,42 +559,93 @@ static int save_image_lzo(struct swap_map_handle *handle, struct bio *bio; struct timeval start; struct timeval stop; - size_t off, unc_len, cmp_len; - unsigned char *unc, *cmp, *wrk, *page; + size_t off; + unsigned thr, run_threads, nr_threads; + unsigned char *page = NULL; + struct cmp_data *data = NULL; + struct crc_data *crc = NULL; + + /* + * We'll limit the number of threads for compression to limit memory + * footprint. + */ + nr_threads = num_online_cpus() - 1; + nr_threads = clamp_val(nr_threads, 1, LZO_THREADS); page = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); if (!page) { printk(KERN_ERR "PM: Failed to allocate LZO page\n"); - return -ENOMEM; + ret = -ENOMEM; + goto out_clean; } - wrk = vmalloc(LZO1X_1_MEM_COMPRESS); - if (!wrk) { - printk(KERN_ERR "PM: Failed to allocate LZO workspace\n"); - free_page((unsigned long)page); - return -ENOMEM; + data = vmalloc(sizeof(*data) * nr_threads); + if (!data) { + printk(KERN_ERR "PM: Failed to allocate LZO data\n"); + ret = -ENOMEM; + goto out_clean; } + for (thr = 0; thr < nr_threads; thr++) + memset(&data[thr], 0, offsetof(struct cmp_data, go)); - unc = vmalloc(LZO_UNC_SIZE); - if (!unc) { - printk(KERN_ERR "PM: Failed to allocate LZO uncompressed\n"); - vfree(wrk); - free_page((unsigned long)page); - return -ENOMEM; + crc = kmalloc(sizeof(*crc), GFP_KERNEL); + if (!crc) { + printk(KERN_ERR "PM: Failed to allocate crc\n"); + ret = -ENOMEM; + goto out_clean; + } + memset(crc, 0, offsetof(struct crc_data, go)); + + /* + * Start the compression threads. + */ + for (thr = 0; thr < nr_threads; thr++) { + init_waitqueue_head(&data[thr].go); + init_waitqueue_head(&data[thr].done); + + data[thr].thr = kthread_run(lzo_compress_threadfn, + &data[thr], + "image_compress/%u", thr); + if (IS_ERR(data[thr].thr)) { + data[thr].thr = NULL; + printk(KERN_ERR + "PM: Cannot start compression threads\n"); + ret = -ENOMEM; + goto out_clean; + } } - cmp = vmalloc(LZO_CMP_SIZE); - if (!cmp) { - printk(KERN_ERR "PM: Failed to allocate LZO compressed\n"); - vfree(unc); - vfree(wrk); - free_page((unsigned long)page); - return -ENOMEM; + /* + * Adjust number of free pages after all allocations have been done. + * We don't want to run out of pages when writing. + */ + handle->nr_free_pages = nr_free_pages() >> 1; + + /* + * Start the CRC32 thread. + */ + init_waitqueue_head(&crc->go); + init_waitqueue_head(&crc->done); + + handle->crc32 = 0; + crc->crc32 = &handle->crc32; + for (thr = 0; thr < nr_threads; thr++) { + crc->unc[thr] = data[thr].unc; + crc->unc_len[thr] = &data[thr].unc_len; + } + + crc->thr = kthread_run(crc32_threadfn, crc, "image_crc32"); + if (IS_ERR(crc->thr)) { + crc->thr = NULL; + printk(KERN_ERR "PM: Cannot start CRC32 thread\n"); + ret = -ENOMEM; + goto out_clean; } printk(KERN_INFO + "PM: Using %u thread(s) for compression.\n" "PM: Compressing and saving image data (%u pages) ... ", - nr_to_write); + nr_threads, nr_to_write); m = nr_to_write / 100; if (!m) m = 1; @@ -480,55 +653,83 @@ static int save_image_lzo(struct swap_map_handle *handle, bio = NULL; do_gettimeofday(&start); for (;;) { - for (off = 0; off < LZO_UNC_SIZE; off += PAGE_SIZE) { - ret = snapshot_read_next(snapshot); - if (ret < 0) - goto out_finish; - - if (!ret) + for (thr = 0; thr < nr_threads; thr++) { + for (off = 0; off < LZO_UNC_SIZE; off += PAGE_SIZE) { + ret = snapshot_read_next(snapshot); + if (ret < 0) + goto out_finish; + + if (!ret) + break; + + memcpy(data[thr].unc + off, + data_of(*snapshot), PAGE_SIZE); + + if (!(nr_pages % m)) + printk(KERN_CONT "\b\b\b\b%3d%%", + nr_pages / m); + nr_pages++; + } + if (!off) break; - memcpy(unc + off, data_of(*snapshot), PAGE_SIZE); + data[thr].unc_len = off; - if (!(nr_pages % m)) - printk(KERN_CONT "\b\b\b\b%3d%%", nr_pages / m); - nr_pages++; + atomic_set(&data[thr].ready, 1); + wake_up(&data[thr].go); } - if (!off) + if (!thr) break; - unc_len = off; - ret = lzo1x_1_compress(unc, unc_len, - cmp + LZO_HEADER, &cmp_len, wrk); - if (ret < 0) { - printk(KERN_ERR "PM: LZO compression failed\n"); - break; - } + crc->run_threads = thr; + atomic_set(&crc->ready, 1); + wake_up(&crc->go); - if (unlikely(!cmp_len || - cmp_len > lzo1x_worst_compress(unc_len))) { - printk(KERN_ERR "PM: Invalid LZO compressed length\n"); - ret = -1; - break; - } + for (run_threads = thr, thr = 0; thr < run_threads; thr++) { + wait_event(data[thr].done, + atomic_read(&data[thr].stop)); + atomic_set(&data[thr].stop, 0); - *(size_t *)cmp = cmp_len; + ret = data[thr].ret; - /* - * Given we are writing one page at a time to disk, we copy - * that much from the buffer, although the last bit will likely - * be smaller than full page. This is OK - we saved the length - * of the compressed data, so any garbage at the end will be - * discarded when we read it. - */ - for (off = 0; off < LZO_HEADER + cmp_len; off += PAGE_SIZE) { - memcpy(page, cmp + off, PAGE_SIZE); + if (ret < 0) { + printk(KERN_ERR "PM: LZO compression failed\n"); + goto out_finish; + } - ret = swap_write_page(handle, page, &bio); - if (ret) + if (unlikely(!data[thr].cmp_len || + data[thr].cmp_len > + lzo1x_worst_compress(data[thr].unc_len))) { + printk(KERN_ERR + "PM: Invalid LZO compressed length\n"); + ret = -1; goto out_finish; + } + + *(size_t *)data[thr].cmp = data[thr].cmp_len; + + /* + * Given we are writing one page at a time to disk, we + * copy that much from the buffer, although the last + * bit will likely be smaller than full page. This is + * OK - we saved the length of the compressed data, so + * any garbage at the end will be discarded when we + * read it. + */ + for (off = 0; + off < LZO_HEADER + data[thr].cmp_len; + off += PAGE_SIZE) { + memcpy(page, data[thr].cmp + off, PAGE_SIZE); + + ret = swap_write_page(handle, page, &bio); + if (ret) + goto out_finish; + } } + + wait_event(crc->done, atomic_read(&crc->stop)); + atomic_set(&crc->stop, 0); } out_finish: @@ -536,16 +737,25 @@ out_finish: do_gettimeofday(&stop); if (!ret) ret = err2; - if (!ret) + if (!ret) { printk(KERN_CONT "\b\b\b\bdone\n"); - else + } else { printk(KERN_CONT "\n"); + } swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); - - vfree(cmp); - vfree(unc); - vfree(wrk); - free_page((unsigned long)page); +out_clean: + if (crc) { + if (crc->thr) + kthread_stop(crc->thr); + kfree(crc); + } + if (data) { + for (thr = 0; thr < nr_threads; thr++) + if (data[thr].thr) + kthread_stop(data[thr].thr); + vfree(data); + } + if (page) free_page((unsigned long)page); return ret; } @@ -625,8 +835,15 @@ out_finish: static void release_swap_reader(struct swap_map_handle *handle) { - if (handle->cur) - free_page((unsigned long)handle->cur); + struct swap_map_page_list *tmp; + + while (handle->maps) { + if (handle->maps->map) + free_page((unsigned long)handle->maps->map); + tmp = handle->maps; + handle->maps = handle->maps->next; + kfree(tmp); + } handle->cur = NULL; } @@ -634,22 +851,46 @@ static int get_swap_reader(struct swap_map_handle *handle, unsigned int *flags_p) { int error; + struct swap_map_page_list *tmp, *last; + sector_t offset; *flags_p = swsusp_header->flags; if (!swsusp_header->image) /* how can this happen? */ return -EINVAL; - handle->cur = (struct swap_map_page *)get_zeroed_page(__GFP_WAIT | __GFP_HIGH); - if (!handle->cur) - return -ENOMEM; + handle->cur = NULL; + last = handle->maps = NULL; + offset = swsusp_header->image; + while (offset) { + tmp = kmalloc(sizeof(*handle->maps), GFP_KERNEL); + if (!tmp) { + release_swap_reader(handle); + return -ENOMEM; + } + memset(tmp, 0, sizeof(*tmp)); + if (!handle->maps) + handle->maps = tmp; + if (last) + last->next = tmp; + last = tmp; + + tmp->map = (struct swap_map_page *) + __get_free_page(__GFP_WAIT | __GFP_HIGH); + if (!tmp->map) { + release_swap_reader(handle); + return -ENOMEM; + } - error = hib_bio_read_page(swsusp_header->image, handle->cur, NULL); - if (error) { - release_swap_reader(handle); - return error; + error = hib_bio_read_page(offset, tmp->map, NULL); + if (error) { + release_swap_reader(handle); + return error; + } + offset = tmp->map->next_swap; } handle->k = 0; + handle->cur = handle->maps->map; return 0; } @@ -658,6 +899,7 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf, { sector_t offset; int error; + struct swap_map_page_list *tmp; if (!handle->cur) return -EINVAL; @@ -668,13 +910,15 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf, if (error) return error; if (++handle->k >= MAP_PAGE_ENTRIES) { - error = hib_wait_on_bio_chain(bio_chain); handle->k = 0; - offset = handle->cur->next_swap; - if (!offset) + free_page((unsigned long)handle->maps->map); + tmp = handle->maps; + handle->maps = handle->maps->next; + kfree(tmp); + if (!handle->maps) release_swap_reader(handle); - else if (!error) - error = hib_bio_read_page(offset, handle->cur, NULL); + else + handle->cur = handle->maps->map; } return error; } @@ -697,7 +941,7 @@ static int load_image(struct swap_map_handle *handle, unsigned int nr_to_read) { unsigned int m; - int error = 0; + int ret = 0; struct timeval start; struct timeval stop; struct bio *bio; @@ -713,15 +957,15 @@ static int load_image(struct swap_map_handle *handle, bio = NULL; do_gettimeofday(&start); for ( ; ; ) { - error = snapshot_write_next(snapshot); - if (error <= 0) + ret = snapshot_write_next(snapshot); + if (ret <= 0) break; - error = swap_read_page(handle, data_of(*snapshot), &bio); - if (error) + ret = swap_read_page(handle, data_of(*snapshot), &bio); + if (ret) break; if (snapshot->sync_read) - error = hib_wait_on_bio_chain(&bio); - if (error) + ret = hib_wait_on_bio_chain(&bio); + if (ret) break; if (!(nr_pages % m)) printk("\b\b\b\b%3d%%", nr_pages / m); @@ -729,17 +973,61 @@ static int load_image(struct swap_map_handle *handle, } err2 = hib_wait_on_bio_chain(&bio); do_gettimeofday(&stop); - if (!error) - error = err2; - if (!error) { + if (!ret) + ret = err2; + if (!ret) { printk("\b\b\b\bdone\n"); snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) - error = -ENODATA; + ret = -ENODATA; } else printk("\n"); swsusp_show_speed(&start, &stop, nr_to_read, "Read"); - return error; + return ret; +} + +/** + * Structure used for LZO data decompression. + */ +struct dec_data { + struct task_struct *thr; /* thread */ + atomic_t ready; /* ready to start flag */ + atomic_t stop; /* ready to stop flag */ + int ret; /* return code */ + wait_queue_head_t go; /* start decompression */ + wait_queue_head_t done; /* decompression done */ + size_t unc_len; /* uncompressed length */ + size_t cmp_len; /* compressed length */ + unsigned char unc[LZO_UNC_SIZE]; /* uncompressed buffer */ + unsigned char cmp[LZO_CMP_SIZE]; /* compressed buffer */ +}; + +/** + * Deompression function that runs in its own thread. + */ +static int lzo_decompress_threadfn(void *data) +{ + struct dec_data *d = data; + + while (1) { + wait_event(d->go, atomic_read(&d->ready) || + kthread_should_stop()); + if (kthread_should_stop()) { + d->thr = NULL; + d->ret = -1; + atomic_set(&d->stop, 1); + wake_up(&d->done); + break; + } + atomic_set(&d->ready, 0); + + d->unc_len = LZO_UNC_SIZE; + d->ret = lzo1x_decompress_safe(d->cmp + LZO_HEADER, d->cmp_len, + d->unc, &d->unc_len); + atomic_set(&d->stop, 1); + wake_up(&d->done); + } + return 0; } /** @@ -753,50 +1041,120 @@ static int load_image_lzo(struct swap_map_handle *handle, unsigned int nr_to_read) { unsigned int m; - int error = 0; + int ret = 0; + int eof = 0; struct bio *bio; struct timeval start; struct timeval stop; unsigned nr_pages; - size_t i, off, unc_len, cmp_len; - unsigned char *unc, *cmp, *page[LZO_CMP_PAGES]; - - for (i = 0; i < LZO_CMP_PAGES; i++) { - page[i] = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); - if (!page[i]) { - printk(KERN_ERR "PM: Failed to allocate LZO page\n"); + size_t off; + unsigned i, thr, run_threads, nr_threads; + unsigned ring = 0, pg = 0, ring_size = 0, + have = 0, want, need, asked = 0; + unsigned long read_pages; + unsigned char **page = NULL; + struct dec_data *data = NULL; + struct crc_data *crc = NULL; + + /* + * We'll limit the number of threads for decompression to limit memory + * footprint. + */ + nr_threads = num_online_cpus() - 1; + nr_threads = clamp_val(nr_threads, 1, LZO_THREADS); + + page = vmalloc(sizeof(*page) * LZO_READ_PAGES); + if (!page) { + printk(KERN_ERR "PM: Failed to allocate LZO page\n"); + ret = -ENOMEM; + goto out_clean; + } - while (i) - free_page((unsigned long)page[--i]); + data = vmalloc(sizeof(*data) * nr_threads); + if (!data) { + printk(KERN_ERR "PM: Failed to allocate LZO data\n"); + ret = -ENOMEM; + goto out_clean; + } + for (thr = 0; thr < nr_threads; thr++) + memset(&data[thr], 0, offsetof(struct dec_data, go)); - return -ENOMEM; + crc = kmalloc(sizeof(*crc), GFP_KERNEL); + if (!crc) { + printk(KERN_ERR "PM: Failed to allocate crc\n"); + ret = -ENOMEM; + goto out_clean; + } + memset(crc, 0, offsetof(struct crc_data, go)); + + /* + * Start the decompression threads. + */ + for (thr = 0; thr < nr_threads; thr++) { + init_waitqueue_head(&data[thr].go); + init_waitqueue_head(&data[thr].done); + + data[thr].thr = kthread_run(lzo_decompress_threadfn, + &data[thr], + "image_decompress/%u", thr); + if (IS_ERR(data[thr].thr)) { + data[thr].thr = NULL; + printk(KERN_ERR + "PM: Cannot start decompression threads\n"); + ret = -ENOMEM; + goto out_clean; } } - unc = vmalloc(LZO_UNC_SIZE); - if (!unc) { - printk(KERN_ERR "PM: Failed to allocate LZO uncompressed\n"); - - for (i = 0; i < LZO_CMP_PAGES; i++) - free_page((unsigned long)page[i]); - - return -ENOMEM; + /* + * Start the CRC32 thread. + */ + init_waitqueue_head(&crc->go); + init_waitqueue_head(&crc->done); + + handle->crc32 = 0; + crc->crc32 = &handle->crc32; + for (thr = 0; thr < nr_threads; thr++) { + crc->unc[thr] = data[thr].unc; + crc->unc_len[thr] = &data[thr].unc_len; } - cmp = vmalloc(LZO_CMP_SIZE); - if (!cmp) { - printk(KERN_ERR "PM: Failed to allocate LZO compressed\n"); + crc->thr = kthread_run(crc32_threadfn, crc, "image_crc32"); + if (IS_ERR(crc->thr)) { + crc->thr = NULL; + printk(KERN_ERR "PM: Cannot start CRC32 thread\n"); + ret = -ENOMEM; + goto out_clean; + } - vfree(unc); - for (i = 0; i < LZO_CMP_PAGES; i++) - free_page((unsigned long)page[i]); + /* + * Adjust number of pages for read buffering, in case we are short. + */ + read_pages = (nr_free_pages() - snapshot_get_image_size()) >> 1; + read_pages = clamp_val(read_pages, LZO_CMP_PAGES, LZO_READ_PAGES); - return -ENOMEM; + for (i = 0; i < read_pages; i++) { + page[i] = (void *)__get_free_page(i < LZO_CMP_PAGES ? + __GFP_WAIT | __GFP_HIGH : + __GFP_WAIT); + if (!page[i]) { + if (i < LZO_CMP_PAGES) { + ring_size = i; + printk(KERN_ERR + "PM: Failed to allocate LZO pages\n"); + ret = -ENOMEM; + goto out_clean; + } else { + break; + } + } } + want = ring_size = i; printk(KERN_INFO + "PM: Using %u thread(s) for decompression.\n" "PM: Loading and decompressing image data (%u pages) ... ", - nr_to_read); + nr_threads, nr_to_read); m = nr_to_read / 100; if (!m) m = 1; @@ -804,85 +1162,189 @@ static int load_image_lzo(struct swap_map_handle *handle, bio = NULL; do_gettimeofday(&start); - error = snapshot_write_next(snapshot); - if (error <= 0) + ret = snapshot_write_next(snapshot); + if (ret <= 0) goto out_finish; - for (;;) { - error = swap_read_page(handle, page[0], NULL); /* sync */ - if (error) - break; - - cmp_len = *(size_t *)page[0]; - if (unlikely(!cmp_len || - cmp_len > lzo1x_worst_compress(LZO_UNC_SIZE))) { - printk(KERN_ERR "PM: Invalid LZO compressed length\n"); - error = -1; - break; + for(;;) { + for (i = 0; !eof && i < want; i++) { + ret = swap_read_page(handle, page[ring], &bio); + if (ret) { + /* + * On real read error, finish. On end of data, + * set EOF flag and just exit the read loop. + */ + if (handle->cur && + handle->cur->entries[handle->k]) { + goto out_finish; + } else { + eof = 1; + break; + } + } + if (++ring >= ring_size) + ring = 0; } + asked += i; + want -= i; - for (off = PAGE_SIZE, i = 1; - off < LZO_HEADER + cmp_len; off += PAGE_SIZE, i++) { - error = swap_read_page(handle, page[i], &bio); - if (error) + /* + * We are out of data, wait for some more. + */ + if (!have) { + if (!asked) + break; + + ret = hib_wait_on_bio_chain(&bio); + if (ret) goto out_finish; + have += asked; + asked = 0; + if (eof) + eof = 2; } - error = hib_wait_on_bio_chain(&bio); /* need all data now */ - if (error) - goto out_finish; - - for (off = 0, i = 0; - off < LZO_HEADER + cmp_len; off += PAGE_SIZE, i++) { - memcpy(cmp + off, page[i], PAGE_SIZE); + if (crc->run_threads) { + wait_event(crc->done, atomic_read(&crc->stop)); + atomic_set(&crc->stop, 0); + crc->run_threads = 0; } - unc_len = LZO_UNC_SIZE; - error = lzo1x_decompress_safe(cmp + LZO_HEADER, cmp_len, - unc, &unc_len); - if (error < 0) { - printk(KERN_ERR "PM: LZO decompression failed\n"); - break; + for (thr = 0; have && thr < nr_threads; thr++) { + data[thr].cmp_len = *(size_t *)page[pg]; + if (unlikely(!data[thr].cmp_len || + data[thr].cmp_len > + lzo1x_worst_compress(LZO_UNC_SIZE))) { + printk(KERN_ERR + "PM: Invalid LZO compressed length\n"); + ret = -1; + goto out_finish; + } + + need = DIV_ROUND_UP(data[thr].cmp_len + LZO_HEADER, + PAGE_SIZE); + if (need > have) { + if (eof > 1) { + ret = -1; + goto out_finish; + } + break; + } + + for (off = 0; + off < LZO_HEADER + data[thr].cmp_len; + off += PAGE_SIZE) { + memcpy(data[thr].cmp + off, + page[pg], PAGE_SIZE); + have--; + want++; + if (++pg >= ring_size) + pg = 0; + } + + atomic_set(&data[thr].ready, 1); + wake_up(&data[thr].go); } - if (unlikely(!unc_len || - unc_len > LZO_UNC_SIZE || - unc_len & (PAGE_SIZE - 1))) { - printk(KERN_ERR "PM: Invalid LZO uncompressed length\n"); - error = -1; - break; + /* + * Wait for more data while we are decompressing. + */ + if (have < LZO_CMP_PAGES && asked) { + ret = hib_wait_on_bio_chain(&bio); + if (ret) + goto out_finish; + have += asked; + asked = 0; + if (eof) + eof = 2; } - for (off = 0; off < unc_len; off += PAGE_SIZE) { - memcpy(data_of(*snapshot), unc + off, PAGE_SIZE); + for (run_threads = thr, thr = 0; thr < run_threads; thr++) { + wait_event(data[thr].done, + atomic_read(&data[thr].stop)); + atomic_set(&data[thr].stop, 0); + + ret = data[thr].ret; - if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); - nr_pages++; + if (ret < 0) { + printk(KERN_ERR + "PM: LZO decompression failed\n"); + goto out_finish; + } - error = snapshot_write_next(snapshot); - if (error <= 0) + if (unlikely(!data[thr].unc_len || + data[thr].unc_len > LZO_UNC_SIZE || + data[thr].unc_len & (PAGE_SIZE - 1))) { + printk(KERN_ERR + "PM: Invalid LZO uncompressed length\n"); + ret = -1; goto out_finish; + } + + for (off = 0; + off < data[thr].unc_len; off += PAGE_SIZE) { + memcpy(data_of(*snapshot), + data[thr].unc + off, PAGE_SIZE); + + if (!(nr_pages % m)) + printk("\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + + ret = snapshot_write_next(snapshot); + if (ret <= 0) { + crc->run_threads = thr + 1; + atomic_set(&crc->ready, 1); + wake_up(&crc->go); + goto out_finish; + } + } } + + crc->run_threads = thr; + atomic_set(&crc->ready, 1); + wake_up(&crc->go); } out_finish: + if (crc->run_threads) { + wait_event(crc->done, atomic_read(&crc->stop)); + atomic_set(&crc->stop, 0); + } do_gettimeofday(&stop); - if (!error) { + if (!ret) { printk("\b\b\b\bdone\n"); snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) - error = -ENODATA; + ret = -ENODATA; + if (!ret) { + if (swsusp_header->flags & SF_CRC32_MODE) { + if(handle->crc32 != swsusp_header->crc32) { + printk(KERN_ERR + "PM: Invalid image CRC32!\n"); + ret = -ENODATA; + } + } + } } else printk("\n"); swsusp_show_speed(&start, &stop, nr_to_read, "Read"); - - vfree(cmp); - vfree(unc); - for (i = 0; i < LZO_CMP_PAGES; i++) +out_clean: + for (i = 0; i < ring_size; i++) free_page((unsigned long)page[i]); + if (crc) { + if (crc->thr) + kthread_stop(crc->thr); + kfree(crc); + } + if (data) { + for (thr = 0; thr < nr_threads; thr++) + if (data[thr].thr) + kthread_stop(data[thr].thr); + vfree(data); + } + if (page) vfree(page); - return error; + return ret; } /** -- cgit v1.3-8-gc7d7 From 9bab0b7fbaceec47d32db51cd9e59c82fb071f5a Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Mon, 3 Oct 2011 15:37:00 +0100 Subject: genirq: Add IRQF_RESUME_EARLY and resume such IRQs earlier This adds a mechanism to resume selected IRQs during syscore_resume instead of dpm_resume_noirq. Under Xen we need to resume IRQs associated with IPIs early enough that the resched IPI is unmasked and we can therefore schedule ourselves out of the stop_machine where the suspend/resume takes place. This issue was introduced by 676dc3cf5bc3 "xen: Use IRQF_FORCE_RESUME". Signed-off-by: Ian Campbell Cc: Rafael J. Wysocki Cc: Jeremy Fitzhardinge Cc: xen-devel Cc: Konrad Rzeszutek Wilk Link: http://lkml.kernel.org/r/1318713254.11016.52.camel@dagon.hellion.org.uk Cc: stable@kernel.org (at least to 2.6.32.y) Signed-off-by: Thomas Gleixner --- drivers/xen/events.c | 2 +- include/linux/interrupt.h | 3 +++ kernel/irq/pm.c | 48 ++++++++++++++++++++++++++++++++++++++++------- 3 files changed, 45 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/drivers/xen/events.c b/drivers/xen/events.c index da70f5c32eb9..a758bc14d0f3 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -1021,7 +1021,7 @@ int bind_ipi_to_irqhandler(enum ipi_vector ipi, if (irq < 0) return irq; - irqflags |= IRQF_NO_SUSPEND | IRQF_FORCE_RESUME; + irqflags |= IRQF_NO_SUSPEND | IRQF_FORCE_RESUME | IRQF_EARLY_RESUME; retval = request_irq(irq, handler, irqflags, devname, dev_id); if (retval != 0) { unbind_from_irq(irq); diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 664544ff77d5..a64b00e286f5 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -59,6 +59,8 @@ * IRQF_NO_SUSPEND - Do not disable this IRQ during suspend * IRQF_FORCE_RESUME - Force enable it on resume even if IRQF_NO_SUSPEND is set * IRQF_NO_THREAD - Interrupt cannot be threaded + * IRQF_EARLY_RESUME - Resume IRQ early during syscore instead of at device + * resume time. */ #define IRQF_DISABLED 0x00000020 #define IRQF_SAMPLE_RANDOM 0x00000040 @@ -72,6 +74,7 @@ #define IRQF_NO_SUSPEND 0x00004000 #define IRQF_FORCE_RESUME 0x00008000 #define IRQF_NO_THREAD 0x00010000 +#define IRQF_EARLY_RESUME 0x00020000 #define IRQF_TIMER (__IRQF_TIMER | IRQF_NO_SUSPEND | IRQF_NO_THREAD) diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index f76fc00c9877..15e53b1766a6 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -9,6 +9,7 @@ #include #include #include +#include #include "internals.h" @@ -39,25 +40,58 @@ void suspend_device_irqs(void) } EXPORT_SYMBOL_GPL(suspend_device_irqs); -/** - * resume_device_irqs - enable interrupt lines disabled by suspend_device_irqs() - * - * Enable all interrupt lines previously disabled by suspend_device_irqs() that - * have the IRQS_SUSPENDED flag set. - */ -void resume_device_irqs(void) +static void resume_irqs(bool want_early) { struct irq_desc *desc; int irq; for_each_irq_desc(irq, desc) { unsigned long flags; + bool is_early = desc->action && + desc->action->flags & IRQF_EARLY_RESUME; + + if (is_early != want_early) + continue; raw_spin_lock_irqsave(&desc->lock, flags); __enable_irq(desc, irq, true); raw_spin_unlock_irqrestore(&desc->lock, flags); } } + +/** + * irq_pm_syscore_ops - enable interrupt lines early + * + * Enable all interrupt lines with %IRQF_EARLY_RESUME set. + */ +static void irq_pm_syscore_resume(void) +{ + resume_irqs(true); +} + +static struct syscore_ops irq_pm_syscore_ops = { + .resume = irq_pm_syscore_resume, +}; + +static int __init irq_pm_init_ops(void) +{ + register_syscore_ops(&irq_pm_syscore_ops); + return 0; +} + +device_initcall(irq_pm_init_ops); + +/** + * resume_device_irqs - enable interrupt lines disabled by suspend_device_irqs() + * + * Enable all non-%IRQF_EARLY_RESUME interrupt lines previously + * disabled by suspend_device_irqs() that have the IRQS_SUSPENDED flag + * set as well as those with %IRQF_FORCE_RESUME. + */ +void resume_device_irqs(void) +{ + resume_irqs(false); +} EXPORT_SYMBOL_GPL(resume_device_irqs); /** -- cgit v1.3-8-gc7d7 From a84a79e4d369a73c0130b5858199e949432da4c6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 17 Oct 2011 08:24:24 -0700 Subject: Avoid using variable-length arrays in kernel/sys.c The size is always valid, but variable-length arrays generate worse code for no good reason (unless the function happens to be inlined and the compiler sees the length for the simple constant it is). Also, there seems to be some code generation problem on POWER, where Henrik Bakken reports that register r28 can get corrupted under some subtle circumstances (interrupt happening at the wrong time?). That all indicates some seriously broken compiler issues, but since variable length arrays are bad regardless, there's little point in trying to chase it down. "Just don't do that, then". Reported-by: Henrik Grindal Bakken Cc: Benjamin Herrenschmidt Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- kernel/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index 18ee1d2f6474..1dbbe695a5ef 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1172,7 +1172,7 @@ DECLARE_RWSEM(uts_sem); static int override_release(char __user *release, int len) { int ret = 0; - char buf[len]; + char buf[65]; if (current->personality & UNAME26) { char *rest = UTS_RELEASE; -- cgit v1.3-8-gc7d7 From bcd5cff7216f9b2de0a148cc355eac199dc6f1cf Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 17 Oct 2011 11:50:30 +0200 Subject: cputimer: Cure lock inversion There's a lock inversion between the cputimer->lock and rq->lock; notably the two callchains involved are: update_rlimit_cpu() sighand->siglock set_process_cpu_timer() cpu_timer_sample_group() thread_group_cputimer() cputimer->lock thread_group_cputime() task_sched_runtime() ->pi_lock rq->lock scheduler_tick() rq->lock task_tick_fair() update_curr() account_group_exec() cputimer->lock Where the first one is enabling a CLOCK_PROCESS_CPUTIME_ID timer, and the second one is keeping up-to-date. This problem was introduced by e8abccb7193 ("posix-cpu-timers: Cure SMP accounting oddities"). Cure the problem by removing the cputimer->lock and rq->lock nesting, this leaves concurrent enablers doing duplicate work, but the time wasted should be on the same order otherwise wasted spinning on the lock and the greater-than assignment filter should ensure we preserve monotonicity. Reported-by: Dave Jones Reported-by: Simon Kirby Signed-off-by: Peter Zijlstra Cc: stable@kernel.org Cc: Linus Torvalds Cc: Martin Schwidefsky Link: http://lkml.kernel.org/r/1318928713.21167.4.camel@twins Signed-off-by: Thomas Gleixner --- kernel/posix-cpu-timers.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index c8008dd58ef2..640ded8f5c48 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -274,9 +274,7 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) struct task_cputime sum; unsigned long flags; - spin_lock_irqsave(&cputimer->lock, flags); if (!cputimer->running) { - cputimer->running = 1; /* * The POSIX timer interface allows for absolute time expiry * values through the TIMER_ABSTIME flag, therefore we have @@ -284,8 +282,11 @@ void thread_group_cputimer(struct task_struct *tsk, struct task_cputime *times) * it. */ thread_group_cputime(tsk, &sum); + spin_lock_irqsave(&cputimer->lock, flags); + cputimer->running = 1; update_gt_cputime(&cputimer->cputime, &sum); - } + } else + spin_lock_irqsave(&cputimer->lock, flags); *times = cputimer->cputime; spin_unlock_irqrestore(&cputimer->lock, flags); } -- cgit v1.3-8-gc7d7 From 825de2e9007439977aed63771db570fc2235e8cd Mon Sep 17 00:00:00 2001 From: Nobuhiro Iwamatsu Date: Mon, 17 Oct 2011 11:08:46 +0900 Subject: irq: Add EXPORT_SYMBOL_GPL to function of irq generic-chip Some functions of irq generic-chip is undefined, because EXPORT_SYMBOL_GPL is not set to these. ERROR: "irq_setup_generic_chip" [drivers/gpio/gpio-pch.ko] undefined! ERROR: "irq_alloc_generic_chip" [drivers/gpio/gpio-pch.ko] undefined! ERROR: "irq_setup_generic_chip" [drivers/gpio/gpio-ml-ioh.ko] undefined! ERROR: "irq_alloc_generic_chip" [drivers/gpio/gpio-ml-ioh.ko] undefined! This is revised that EXPORT_SYMBOL_GPL can be added and referred to in functions. Signed-off-by: Nobuhiro Iwamatsu Acked-by: Thomas Gleixner Signed-off-by: Grant Likely --- kernel/irq/generic-chip.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index e38544dddb18..6cb7613e4bf4 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -211,6 +211,7 @@ irq_alloc_generic_chip(const char *name, int num_ct, unsigned int irq_base, } return gc; } +EXPORT_SYMBOL_GPL(irq_alloc_generic_chip); /* * Separate lockdep class for interrupt chip which can nest irq_desc @@ -258,6 +259,7 @@ void irq_setup_generic_chip(struct irq_chip_generic *gc, u32 msk, } gc->irq_cnt = i - gc->irq_base; } +EXPORT_SYMBOL_GPL(irq_setup_generic_chip); /** * irq_setup_alt_chip - Switch to alternative chip @@ -281,6 +283,7 @@ int irq_setup_alt_chip(struct irq_data *d, unsigned int type) } return -EINVAL; } +EXPORT_SYMBOL_GPL(irq_setup_alt_chip); /** * irq_remove_generic_chip - Remove a chip @@ -311,6 +314,7 @@ void irq_remove_generic_chip(struct irq_chip_generic *gc, u32 msk, irq_modify_status(i, clr, set); } } +EXPORT_SYMBOL_GPL(irq_remove_generic_chip); #ifdef CONFIG_PM static int irq_gc_suspend(void) -- cgit v1.3-8-gc7d7 From 189c3fd68c7016e37c1ffd7a00009e2c944a9d06 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Fri, 30 Sep 2011 09:10:54 -0700 Subject: stop_machine: make stop_machine safe and efficient to call early Make stop_machine() safe to call early in boot, before stop_machine() has been set up, by simply calling the callback function directly if there's only one CPU online. [ Fixes from AKPM: - add comment - local_irq_flags, not save_flags - also call hard_irq_disable() for systems which need it Tejun suggested using an explicit flag rather than just looking at the online cpu count. ] Signed-off-by: Jeremy Fitzhardinge Acked-by: Tejun Heo Acked-by: Peter Zijlstra Cc: Rusty Russell Cc: Peter Zijlstra Cc: Andrew Morton Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Steven Rostedt --- kernel/stop_machine.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'kernel') diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index ba5070ce5765..d3f960a6ca56 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -41,6 +41,7 @@ struct cpu_stopper { }; static DEFINE_PER_CPU(struct cpu_stopper, cpu_stopper); +static bool stop_machine_initialized = false; static void cpu_stop_init_done(struct cpu_stop_done *done, unsigned int nr_todo) { @@ -386,6 +387,8 @@ static int __init cpu_stop_init(void) cpu_stop_cpu_callback(&cpu_stop_cpu_notifier, CPU_ONLINE, bcpu); register_cpu_notifier(&cpu_stop_cpu_notifier); + stop_machine_initialized = true; + return 0; } early_initcall(cpu_stop_init); @@ -485,6 +488,25 @@ int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus) .num_threads = num_online_cpus(), .active_cpus = cpus }; + if (!stop_machine_initialized) { + /* + * Handle the case where stop_machine() is called + * early in boot before stop_machine() has been + * initialized. + */ + unsigned long flags; + int ret; + + WARN_ON_ONCE(smdata.num_threads != 1); + + local_irq_save(flags); + hard_irq_disable(); + ret = (*fn)(data); + local_irq_restore(flags); + + return ret; + } + /* Set the initial state and stop all online cpus. */ set_state(&smdata, STOPMACHINE_PREPARE); return stop_cpus(cpu_online_mask, stop_machine_cpu_stop, &smdata); -- cgit v1.3-8-gc7d7 From 37348804e0289087d21ae8bff4c0732030a3c6ac Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 29 Sep 2011 11:10:05 -0700 Subject: jump_label: if a key has already been initialized, don't nop it out If a key has been enabled before jump_label_init() is called, don't nop it out. This removes arch_jump_label_text_poke_early() (which can only nop out a site) and uses arch_jump_label_transform() instead. Signed-off-by: Jeremy Fitzhardinge Acked-by: Jason Baron Acked-by: Peter Zijlstra --- include/linux/jump_label.h | 3 +-- kernel/jump_label.c | 20 ++++++++------------ 2 files changed, 9 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 1213e9d63f79..12e804ea32ab 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -44,8 +44,7 @@ extern struct jump_entry __stop___jump_table[]; extern void jump_label_lock(void); extern void jump_label_unlock(void); extern void arch_jump_label_transform(struct jump_entry *entry, - enum jump_label_type type); -extern void arch_jump_label_text_poke_early(jump_label_t addr); + enum jump_label_type type); extern int jump_label_text_reserved(void *start, void *end); extern void jump_label_inc(struct jump_label_key *key); extern void jump_label_dec(struct jump_label_key *key); diff --git a/kernel/jump_label.c b/kernel/jump_label.c index a8ce45097f3d..059202d5b77a 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -121,13 +121,6 @@ static void __jump_label_update(struct jump_label_key *key, } } -/* - * Not all archs need this. - */ -void __weak arch_jump_label_text_poke_early(jump_label_t addr) -{ -} - static __init int jump_label_init(void) { struct jump_entry *iter_start = __start___jump_table; @@ -139,12 +132,15 @@ static __init int jump_label_init(void) jump_label_sort_entries(iter_start, iter_stop); for (iter = iter_start; iter < iter_stop; iter++) { - arch_jump_label_text_poke_early(iter->code); - if (iter->key == (jump_label_t)(unsigned long)key) + struct jump_label_key *iterk; + + iterk = (struct jump_label_key *)(unsigned long)iter->key; + arch_jump_label_transform(iter, jump_label_enabled(iterk) ? + JUMP_LABEL_ENABLE : JUMP_LABEL_DISABLE); + if (iterk == key) continue; - key = (struct jump_label_key *)(unsigned long)iter->key; - atomic_set(&key->enabled, 0); + key = iterk; key->entries = iter; #ifdef CONFIG_MODULES key->next = NULL; @@ -212,7 +208,7 @@ void jump_label_apply_nops(struct module *mod) return; for (iter = iter_start; iter < iter_stop; iter++) - arch_jump_label_text_poke_early(iter->code); + arch_jump_label_transform(iter, JUMP_LABEL_DISABLE); } static int jump_label_add_module(struct module *mod) -- cgit v1.3-8-gc7d7 From 20284aa77c0f6227da4783a920b72dc61d4bcc09 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 3 Oct 2011 11:01:46 -0700 Subject: jump_label: add arch_jump_label_transform_static() to optimise non-live code updates When updating a newly loaded module, the code is definitely not yet executing on any processor, so it can be updated with no need for any heavyweight synchronization. This patch adds arch_jump_label_static() which is implemented as arch_jump_label_transform() by default, but architectures can override it if it avoids, say, a call to stop_machine(). Signed-off-by: Jeremy Fitzhardinge Acked-by: Jason Baron Acked-by: Peter Zijlstra --- include/linux/jump_label.h | 2 ++ kernel/jump_label.c | 18 +++++++++++++++--- 2 files changed, 17 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 12e804ea32ab..56594e45b011 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -45,6 +45,8 @@ extern void jump_label_lock(void); extern void jump_label_unlock(void); extern void arch_jump_label_transform(struct jump_entry *entry, enum jump_label_type type); +extern void arch_jump_label_transform_static(struct jump_entry *entry, + enum jump_label_type type); extern int jump_label_text_reserved(void *start, void *end); extern void jump_label_inc(struct jump_label_key *key); extern void jump_label_dec(struct jump_label_key *key); diff --git a/kernel/jump_label.c b/kernel/jump_label.c index 059202d5b77a..ff2028f35aa8 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -104,6 +104,18 @@ static int __jump_label_text_reserved(struct jump_entry *iter_start, return 0; } +/* + * Update code which is definitely not currently executing. + * Architectures which need heavyweight synchronization to modify + * running code can override this to make the non-live update case + * cheaper. + */ +void __weak arch_jump_label_transform_static(struct jump_entry *entry, + enum jump_label_type type) +{ + arch_jump_label_transform(entry, type); +} + static void __jump_label_update(struct jump_label_key *key, struct jump_entry *entry, struct jump_entry *stop, int enable) @@ -135,8 +147,8 @@ static __init int jump_label_init(void) struct jump_label_key *iterk; iterk = (struct jump_label_key *)(unsigned long)iter->key; - arch_jump_label_transform(iter, jump_label_enabled(iterk) ? - JUMP_LABEL_ENABLE : JUMP_LABEL_DISABLE); + arch_jump_label_transform_static(iter, jump_label_enabled(iterk) ? + JUMP_LABEL_ENABLE : JUMP_LABEL_DISABLE); if (iterk == key) continue; @@ -208,7 +220,7 @@ void jump_label_apply_nops(struct module *mod) return; for (iter = iter_start; iter < iter_stop; iter++) - arch_jump_label_transform(iter, JUMP_LABEL_DISABLE); + arch_jump_label_transform_static(iter, JUMP_LABEL_DISABLE); } static int jump_label_add_module(struct module *mod) -- cgit v1.3-8-gc7d7 From 97ce2c88f9ad42e3c60a9beb9fca87abf3639faa Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 12 Oct 2011 16:17:54 -0700 Subject: jump-label: initialize jump-label subsystem much earlier Initialize jump_labels much, much earlier, so they're available for use during system setup. Signed-off-by: Jeremy Fitzhardinge Acked-by: Peter Zijlstra --- include/linux/jump_label.h | 14 +++++++++----- init/main.c | 3 +++ kernel/jump_label.c | 5 +---- 3 files changed, 13 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 56594e45b011..388b0d425b50 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -16,7 +16,7 @@ struct jump_label_key { # include # define HAVE_JUMP_LABEL -#endif +#endif /* CC_HAVE_ASM_GOTO && CONFIG_JUMP_LABEL */ enum jump_label_type { JUMP_LABEL_DISABLE = 0, @@ -41,6 +41,7 @@ static __always_inline bool static_branch(struct jump_label_key *key) extern struct jump_entry __start___jump_table[]; extern struct jump_entry __stop___jump_table[]; +extern void jump_label_init(void); extern void jump_label_lock(void); extern void jump_label_unlock(void); extern void arch_jump_label_transform(struct jump_entry *entry, @@ -53,7 +54,7 @@ extern void jump_label_dec(struct jump_label_key *key); extern bool jump_label_enabled(struct jump_label_key *key); extern void jump_label_apply_nops(struct module *mod); -#else +#else /* !HAVE_JUMP_LABEL */ #include @@ -63,6 +64,10 @@ struct jump_label_key { atomic_t enabled; }; +static __always_inline void jump_label_init(void) +{ +} + static __always_inline bool static_branch(struct jump_label_key *key) { if (unlikely(atomic_read(&key->enabled))) @@ -97,7 +102,6 @@ static inline int jump_label_apply_nops(struct module *mod) { return 0; } +#endif /* HAVE_JUMP_LABEL */ -#endif - -#endif +#endif /* _LINUX_JUMP_LABEL_H */ diff --git a/init/main.c b/init/main.c index 2a9b88aa5e76..29d8d847de94 100644 --- a/init/main.c +++ b/init/main.c @@ -515,6 +515,9 @@ asmlinkage void __init start_kernel(void) parse_args("Booting kernel", static_command_line, __start___param, __stop___param - __start___param, &unknown_bootoption); + + jump_label_init(); + /* * These use large bootmem allocations and must precede * kmem_cache_init() diff --git a/kernel/jump_label.c b/kernel/jump_label.c index ff2028f35aa8..bbdfe2a462a0 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -133,7 +133,7 @@ static void __jump_label_update(struct jump_label_key *key, } } -static __init int jump_label_init(void) +void __init jump_label_init(void) { struct jump_entry *iter_start = __start___jump_table; struct jump_entry *iter_stop = __stop___jump_table; @@ -159,10 +159,7 @@ static __init int jump_label_init(void) #endif } jump_label_unlock(); - - return 0; } -early_initcall(jump_label_init); #ifdef CONFIG_MODULES -- cgit v1.3-8-gc7d7 From 37252db6aa576c34fd794a5a54fb32d7a8b3a07a Mon Sep 17 00:00:00 2001 From: Jiri Kosina Date: Wed, 26 Oct 2011 13:10:39 +1030 Subject: kmod: prevent kmod_loop_msg overflow in __request_module() Due to post-increment in condition of kmod_loop_msg in __request_module(), the system log can be spammed by much more than 5 instances of the 'runaway loop' message if the number of events triggering it makes the kmod_loop_msg to overflow. Fix that by making sure we never increment it past the threshold. Signed-off-by: Jiri Kosina Signed-off-by: Rusty Russell CC: stable@kernel.org --- kernel/kmod.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/kmod.c b/kernel/kmod.c index ddc7644c1305..a4bea97c75b6 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -114,10 +114,12 @@ int __request_module(bool wait, const char *fmt, ...) atomic_inc(&kmod_concurrent); if (atomic_read(&kmod_concurrent) > max_modprobes) { /* We may be blaming an innocent here, but unlikely */ - if (kmod_loop_msg++ < 5) + if (kmod_loop_msg < 5) { printk(KERN_ERR "request_module: runaway loop modprobe %s\n", module_name); + kmod_loop_msg++; + } atomic_dec(&kmod_concurrent); return -ENOMEM; } -- cgit v1.3-8-gc7d7 From b1e4d20cbf2ef8e27515da032b95fdcbb5b06bf1 Mon Sep 17 00:00:00 2001 From: Michal Schmidt Date: Mon, 10 Oct 2011 00:03:37 +0200 Subject: params: make dashes and underscores in parameter names truly equal The user may use "foo-bar" for a kernel parameter defined as "foo_bar". Make sure it works the other way around too. Apply the equality of dashes and underscores on early_params and __setup params as well. The example given in Documentation/kernel-parameters.txt indicates that this is the intended behaviour. With the patch the kernel accepts "log-buf-len=1M" as expected. https://bugzilla.redhat.com/show_bug.cgi?id=744545 Signed-off-by: Michal Schmidt Signed-off-by: Rusty Russell (neatened implementations) --- include/linux/moduleparam.h | 20 ++++++++++++++++++++ init/main.c | 4 ++-- kernel/params.c | 21 ++++++++++++++------- 3 files changed, 36 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/moduleparam.h b/include/linux/moduleparam.h index ddaae98c53f9..fffb10bd5514 100644 --- a/include/linux/moduleparam.h +++ b/include/linux/moduleparam.h @@ -262,6 +262,26 @@ static inline void __kernel_param_unlock(void) .str = &__param_string_##name, 0, perm); \ __MODULE_PARM_TYPE(name, "string") +/** + * parameq - checks if two parameter names match + * @name1: parameter name 1 + * @name2: parameter name 2 + * + * Returns true if the two parameter names are equal. + * Dashes (-) are considered equal to underscores (_). + */ +extern bool parameq(const char *name1, const char *name2); + +/** + * parameqn - checks if two parameter names match + * @name1: parameter name 1 + * @name2: parameter name 2 + * @n: the length to compare + * + * Similar to parameq(), except it compares @n characters. + */ +extern bool parameqn(const char *name1, const char *name2, size_t n); + /* Called on module insert or kernel boot */ extern int parse_args(const char *name, char *args, diff --git a/init/main.c b/init/main.c index 03b408dff825..63f5f6f8dc3b 100644 --- a/init/main.c +++ b/init/main.c @@ -163,7 +163,7 @@ static int __init obsolete_checksetup(char *line) p = __setup_start; do { int n = strlen(p->str); - if (!strncmp(line, p->str, n)) { + if (parameqn(line, p->str, n)) { if (p->early) { /* Already done in parse_early_param? * (Needs exact match on param part). @@ -392,7 +392,7 @@ static int __init do_early_param(char *param, char *val) const struct obs_kernel_param *p; for (p = __setup_start; p < __setup_end; p++) { - if ((p->early && strcmp(param, p->str) == 0) || + if ((p->early && parameq(param, p->str)) || (strcmp(param, "console") == 0 && strcmp(p->str, "earlycon") == 0) ) { diff --git a/kernel/params.c b/kernel/params.c index 22df3e0d142a..821788947e40 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -67,20 +67,27 @@ static void maybe_kfree_parameter(void *param) } } -static inline char dash2underscore(char c) +static char dash2underscore(char c) { if (c == '-') return '_'; return c; } -static inline int parameq(const char *input, const char *paramname) +bool parameqn(const char *a, const char *b, size_t n) { - unsigned int i; - for (i = 0; dash2underscore(input[i]) == paramname[i]; i++) - if (input[i] == '\0') - return 1; - return 0; + size_t i; + + for (i = 0; i < n; i++) { + if (dash2underscore(a[i]) != dash2underscore(b[i])) + return false; + } + return true; +} + +bool parameq(const char *a, const char *b) +{ + return parameqn(a, b, strlen(a)+1); } static int parse_one(char *param, -- cgit v1.3-8-gc7d7 From 3d214faea6e4f9b6018bf8589f4b245126349c0a Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sun, 30 Oct 2011 15:16:36 +0100 Subject: [S390] kdump: Add KEXEC_CRASH_CONTROL_MEMORY_LIMIT On s390 there is a different KEXEC_CONTROL_MEMORY_LIMIT for the normal and the kdump kexec case. Therefore this patch introduces a new macro KEXEC_CRASH_CONTROL_MEMORY_LIMIT. This is set to KEXEC_CONTROL_MEMORY_LIMIT for all architectures that do not define KEXEC_CRASH_CONTROL_MEMORY_LIMIT. Acked-by: Vivek Goyal Acked-by: Andrew Morton Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- include/linux/kexec.h | 4 ++++ kernel/kexec.c | 2 +- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index c2478a342cd7..07d9aba75562 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -33,6 +33,10 @@ #error KEXEC_ARCH not defined #endif +#ifndef KEXEC_CRASH_CONTROL_MEMORY_LIMIT +#define KEXEC_CRASH_CONTROL_MEMORY_LIMIT KEXEC_CONTROL_MEMORY_LIMIT +#endif + #define KEXEC_NOTE_HEAD_BYTES ALIGN(sizeof(struct elf_note), 4) #define KEXEC_CORE_NOTE_NAME "CORE" #define KEXEC_CORE_NOTE_NAME_BYTES ALIGN(sizeof(KEXEC_CORE_NOTE_NAME), 4) diff --git a/kernel/kexec.c b/kernel/kexec.c index 296fbc84d659..7204fb982ed5 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -498,7 +498,7 @@ static struct page *kimage_alloc_crash_control_pages(struct kimage *image, while (hole_end <= crashk_res.end) { unsigned long i; - if (hole_end > KEXEC_CONTROL_MEMORY_LIMIT) + if (hole_end > KEXEC_CRASH_CONTROL_MEMORY_LIMIT) break; if (hole_end > crashk_res.end) break; -- cgit v1.3-8-gc7d7 From d3bf37955d46718ee1a7f1fc69f953d2328ba7c2 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sun, 30 Oct 2011 15:16:37 +0100 Subject: [S390] kdump: Add size to elfcorehdr kernel parameter Currently only the address of the pre-allocated ELF header is passed with the elfcorehdr= kernel parameter. In order to reserve memory for the header in the 2nd kernel also the size is required. Current kdump architecture backends use different methods to do that, e.g. x86 uses the memmap= kernel parameter. On s390 there is no easy way to transfer this information. Therefore the elfcorehdr kernel parameter is extended to also pass the size. This now can also be used as standard mechanism by all future kdump architecture backends. The syntax of the kernel parameter is extended as follows: elfcorehdr=[size[KMG]@]offset[KMG] This change is backward compatible because elfcorehdr=size is still allowed. Acked-by: Vivek Goyal Acked-by: Andrew Morton Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- Documentation/kernel-parameters.txt | 6 +++--- include/linux/crash_dump.h | 1 + kernel/crash_dump.c | 11 +++++++++++ 3 files changed, 15 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 93413ce96883..1fbe3625b2da 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -741,10 +741,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted. See Documentation/block/cfq-iosched.txt and Documentation/block/deadline-iosched.txt for details. - elfcorehdr= [IA-64,PPC,SH,X86] + elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] Specifies physical address of start of kernel core - image elf header. Generally kexec loader will - pass this option to capture kernel. + image elf header and optionally the size. Generally + kexec loader will pass this option to capture kernel. See Documentation/kdump/kdump.txt for details. enable_mtrr_cleanup [X86] diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h index 74054074e876..5c4abce94ad1 100644 --- a/include/linux/crash_dump.h +++ b/include/linux/crash_dump.h @@ -10,6 +10,7 @@ #define ELFCORE_ADDR_ERR (-2ULL) extern unsigned long long elfcorehdr_addr; +extern unsigned long long elfcorehdr_size; extern ssize_t copy_oldmem_page(unsigned long, char *, size_t, unsigned long, int); diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c index 5f85690285d4..69ebf3380bac 100644 --- a/kernel/crash_dump.c +++ b/kernel/crash_dump.c @@ -19,9 +19,16 @@ unsigned long saved_max_pfn; */ unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX; +/* + * stores the size of elf header of crash image + */ +unsigned long long elfcorehdr_size; + /* * elfcorehdr= specifies the location of elf core header stored by the crashed * kernel. This option will be passed by kexec loader to the capture kernel. + * + * Syntax: elfcorehdr=[size[KMG]@]offset[KMG] */ static int __init setup_elfcorehdr(char *arg) { @@ -29,6 +36,10 @@ static int __init setup_elfcorehdr(char *arg) if (!arg) return -EINVAL; elfcorehdr_addr = memparse(arg, &end); + if (*end == '@') { + elfcorehdr_size = elfcorehdr_addr; + elfcorehdr_addr = memparse(end + 1, &end); + } return end > arg ? 0 : -EINVAL; } early_param("elfcorehdr", setup_elfcorehdr); -- cgit v1.3-8-gc7d7 From fa8ff292bb4844cc0b66b7b24d611eb7177f7ded Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sun, 30 Oct 2011 15:16:41 +0100 Subject: [S390] kdump: Initialize vmcoreinfo note at startup Currently the vmcoreinfo note is only initialized in case of kdump. On s390 it is possible to create kernel dumps with other dump mechanisms than kdump (e.g. via hypervisor dump or stand-alone dump tools). For those dumps it would also be desirable to include the vmcoreinfo data. To accomplish this, with this patch the vmcoreinfo ELF note is always initialized, not only in case of a (kdump) crash. On s390 we will add an ABI defined pointer at a well known address to vmcoreinfo so that dump analysis tools are able to find this information. In particular on s390 we have a tool named zgetdump. With this tool it is possible to convert dump formats on the fly using fuse. E.g. you can mount a s390 stand-alone dump as ELF dump. When this is done, the tool finds the vmcoreinfo in the stand-alone dump via the well known ABI defined address and it creates the respective VMCOREINFO ELF note in the output ELF dump. This then can be used e.g. by makedumpfile for dump filtering. No more need for a vmlinux file with debug information. So this will look like the following: $ zgetdump --mount standalone.dump -f elf /mnt $ ls /mnt dump.elf $ readelf -n /mnt/dump.elf $ ... VMCOREINFO 0x00000474 Unknown note type: (0x00000000) $ makedumpfile -c -d 31 /mnt/dump.elf dump.kdump Cc: Andrew Morton Acked-by: Vivek Goyal Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- kernel/kexec.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/kexec.c b/kernel/kexec.c index 7204fb982ed5..d3b8a4ceb90b 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1380,24 +1380,23 @@ int __init parse_crashkernel(char *cmdline, } - -void crash_save_vmcoreinfo(void) +static void update_vmcoreinfo_note(void) { - u32 *buf; + u32 *buf = vmcoreinfo_note; if (!vmcoreinfo_size) return; - - vmcoreinfo_append_str("CRASHTIME=%ld", get_seconds()); - - buf = (u32 *)vmcoreinfo_note; - buf = append_elf_note(buf, VMCOREINFO_NOTE_NAME, 0, vmcoreinfo_data, vmcoreinfo_size); - final_note(buf); } +void crash_save_vmcoreinfo(void) +{ + vmcoreinfo_append_str("CRASHTIME=%ld", get_seconds()); + update_vmcoreinfo_note(); +} + void vmcoreinfo_append_str(const char *fmt, ...) { va_list args; @@ -1483,6 +1482,7 @@ static int __init crash_save_vmcoreinfo_init(void) VMCOREINFO_NUMBER(PG_swapcache); arch_crash_save_vmcoreinfo(); + update_vmcoreinfo_note(); return 0; } -- cgit v1.3-8-gc7d7 From 558df7209e7997275f6b8ad37737494cf2da1512 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Sun, 30 Oct 2011 15:16:43 +0100 Subject: [S390] kdump: Add infrastructure for unmapping crashkernel memory This patch introduces a mechanism that allows architecture backends to remove page tables for the crashkernel memory. This can protect the loaded kdump kernel from being overwritten by broken kernel code. Two new functions crash_map_reserved_pages() and crash_unmap_reserved_pages() are added that can be implemented by architecture code. The crash_map_reserved_pages() function is called before and crash_unmap_reserved_pages() after the crashkernel segments are loaded. The functions are also called in crash_shrink_memory() to create/remove page tables when the crashkernel memory size is reduced. To support architectures that have large pages this patch also introduces a new define KEXEC_CRASH_MEM_ALIGN. The crashkernel start and size must always be aligned with KEXEC_CRASH_MEM_ALIGN. Cc: Andrew Morton Acked-by: Vivek Goyal Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- include/linux/kexec.h | 6 ++++++ kernel/kexec.c | 21 +++++++++++++++++++-- 2 files changed, 25 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 07d9aba75562..fe45136b32cc 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -37,6 +37,10 @@ #define KEXEC_CRASH_CONTROL_MEMORY_LIMIT KEXEC_CONTROL_MEMORY_LIMIT #endif +#ifndef KEXEC_CRASH_MEM_ALIGN +#define KEXEC_CRASH_MEM_ALIGN PAGE_SIZE +#endif + #define KEXEC_NOTE_HEAD_BYTES ALIGN(sizeof(struct elf_note), 4) #define KEXEC_CORE_NOTE_NAME "CORE" #define KEXEC_CORE_NOTE_NAME_BYTES ALIGN(sizeof(KEXEC_CORE_NOTE_NAME), 4) @@ -133,6 +137,8 @@ extern void crash_kexec(struct pt_regs *); int kexec_should_crash(struct task_struct *); void crash_save_cpu(struct pt_regs *regs, int cpu); void crash_save_vmcoreinfo(void); +void crash_map_reserved_pages(void); +void crash_unmap_reserved_pages(void); void arch_crash_save_vmcoreinfo(void); void vmcoreinfo_append_str(const char *fmt, ...) __attribute__ ((format (printf, 1, 2))); diff --git a/kernel/kexec.c b/kernel/kexec.c index d3b8a4ceb90b..dc7bc0829286 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -999,6 +999,7 @@ SYSCALL_DEFINE4(kexec_load, unsigned long, entry, unsigned long, nr_segments, kimage_free(xchg(&kexec_crash_image, NULL)); result = kimage_crash_alloc(&image, entry, nr_segments, segments); + crash_map_reserved_pages(); } if (result) goto out; @@ -1015,6 +1016,8 @@ SYSCALL_DEFINE4(kexec_load, unsigned long, entry, unsigned long, nr_segments, goto out; } kimage_terminate(image); + if (flags & KEXEC_ON_CRASH) + crash_unmap_reserved_pages(); } /* Install the new kernel, and Uninstall the old */ image = xchg(dest_image, image); @@ -1026,6 +1029,18 @@ out: return result; } +/* + * Add and remove page tables for crashkernel memory + * + * Provide an empty default implementation here -- architecture + * code may override this + */ +void __weak crash_map_reserved_pages(void) +{} + +void __weak crash_unmap_reserved_pages(void) +{} + #ifdef CONFIG_COMPAT asmlinkage long compat_sys_kexec_load(unsigned long entry, unsigned long nr_segments, @@ -1134,14 +1149,16 @@ int crash_shrink_memory(unsigned long new_size) goto unlock; } - start = roundup(start, PAGE_SIZE); - end = roundup(start + new_size, PAGE_SIZE); + start = roundup(start, KEXEC_CRASH_MEM_ALIGN); + end = roundup(start + new_size, KEXEC_CRASH_MEM_ALIGN); + crash_map_reserved_pages(); crash_free_reserved_phys_range(end, crashk_res.end); if ((start == end) && (crashk_res.parent != NULL)) release_resource(&crashk_res); crashk_res.end = end - 1; + crash_unmap_reserved_pages(); unlock: mutex_unlock(&kexec_mutex); -- cgit v1.3-8-gc7d7 From 638ad34a8811119b32247b7722288ef8b90907d1 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Sun, 30 Oct 2011 15:17:13 +0100 Subject: [S390] sparse: fix sparse warnings about missing prototypes Add prototypes and includes for functions used in different modules. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/page.h | 1 + arch/s390/include/asm/processor.h | 2 ++ arch/s390/include/asm/spinlock.h | 2 ++ arch/s390/include/asm/system.h | 2 ++ arch/s390/kernel/entry.h | 13 +++++++++++-- arch/s390/kernel/ipl.c | 1 + arch/s390/kernel/process.c | 1 + arch/s390/kernel/suspend.c | 1 + arch/s390/kernel/time.c | 1 + arch/s390/mm/mmap.c | 1 + arch/s390/mm/pageattr.c | 1 + drivers/s390/cio/css.h | 2 ++ kernel/sysctl.c | 8 -------- 13 files changed, 26 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/arch/s390/include/asm/page.h b/arch/s390/include/asm/page.h index accb372ddc7e..f7ec548c2b9d 100644 --- a/arch/s390/include/asm/page.h +++ b/arch/s390/include/asm/page.h @@ -177,6 +177,7 @@ static inline int page_test_and_clear_young(unsigned long pfn) struct page; void arch_free_page(struct page *page, int order); void arch_alloc_page(struct page *page, int order); +void arch_set_page_states(int make_stable); static inline int devmem_is_allowed(unsigned long pfn) { diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h index 20ffb12d4ae9..5f33d37d032c 100644 --- a/arch/s390/include/asm/processor.h +++ b/arch/s390/include/asm/processor.h @@ -33,6 +33,8 @@ static inline void get_cpu_id(struct cpuid *ptr) extern void s390_adjust_jiffies(void); extern int get_cpu_capability(unsigned int *); +extern const struct seq_operations cpuinfo_op; +extern int sysctl_ieee_emulation_warnings; /* * User space process size: 2GB for 31 bit, 4TB or 8PT for 64 bit. diff --git a/arch/s390/include/asm/spinlock.h b/arch/s390/include/asm/spinlock.h index 56612fc8186e..fd94dfec8d08 100644 --- a/arch/s390/include/asm/spinlock.h +++ b/arch/s390/include/asm/spinlock.h @@ -13,6 +13,8 @@ #include +extern int spin_retry; + static inline int _raw_compare_and_swap(volatile unsigned int *lock, unsigned int old, unsigned int new) diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h index d73cc6b60000..ef573c1d71a7 100644 --- a/arch/s390/include/asm/system.h +++ b/arch/s390/include/asm/system.h @@ -20,6 +20,8 @@ struct task_struct; +extern int sysctl_userprocess_debug; + extern struct task_struct *__switch_to(void *, void *); extern void update_per_regs(struct task_struct *task); diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h index 66729eb7bbc5..ef8fb1d6e8d7 100644 --- a/arch/s390/kernel/entry.h +++ b/arch/s390/kernel/entry.h @@ -5,24 +5,33 @@ #include #include + +extern void (*pgm_check_table[128])(struct pt_regs *, long, unsigned long); +extern void *restart_stack; + +asmlinkage long do_syscall_trace_enter(struct pt_regs *regs); +asmlinkage void do_syscall_trace_exit(struct pt_regs *regs); + void do_protection_exception(struct pt_regs *, long, unsigned long); void do_dat_exception(struct pt_regs *, long, unsigned long); void do_asce_exception(struct pt_regs *, long, unsigned long); -extern int sysctl_userprocess_debug; - void do_per_trap(struct pt_regs *regs); void syscall_trace(struct pt_regs *regs, int entryexit); void kernel_stack_overflow(struct pt_regs * regs); void do_signal(struct pt_regs *regs); int handle_signal32(unsigned long sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *oldset, struct pt_regs *regs); +void do_notify_resume(struct pt_regs *regs); void do_extint(struct pt_regs *regs, unsigned int, unsigned int, unsigned long); +void do_restart(void); int __cpuinit start_secondary(void *cpuvoid); void __init startup_init(void); void die(const char * str, struct pt_regs * regs, long err); +void __init time_init(void); + struct s390_mmap_arg_struct; struct fadvise64_64_args; struct old_sigaction; diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c index 6296809c8e02..affa8e68124a 100644 --- a/arch/s390/kernel/ipl.c +++ b/arch/s390/kernel/ipl.c @@ -27,6 +27,7 @@ #include #include #include +#include "entry.h" #define IPL_PARM_BLOCK_VERSION 0 diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 5e64a9a29ea4..9451b210a1b4 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/s390/kernel/suspend.c b/arch/s390/kernel/suspend.c index b6f9afed74ec..47df775c844d 100644 --- a/arch/s390/kernel/suspend.c +++ b/arch/s390/kernel/suspend.c @@ -7,6 +7,7 @@ */ #include +#include #include #include diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index 8d65bd0383fc..ebbfab3c6e5a 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -48,6 +48,7 @@ #include #include #include +#include "entry.h" /* change this if you have some constant time drift */ #define USECS_PER_JIFFY ((unsigned long) 1000000/HZ) diff --git a/arch/s390/mm/mmap.c b/arch/s390/mm/mmap.c index c9a9f7f18188..f09c74881b7e 100644 --- a/arch/s390/mm/mmap.c +++ b/arch/s390/mm/mmap.c @@ -26,6 +26,7 @@ #include #include +#include #include #include #include diff --git a/arch/s390/mm/pageattr.c b/arch/s390/mm/pageattr.c index d013ed39743b..b36537a5f43e 100644 --- a/arch/s390/mm/pageattr.c +++ b/arch/s390/mm/pageattr.c @@ -5,6 +5,7 @@ #include #include #include +#include #include static void change_page_attr(unsigned long addr, int numpages, diff --git a/drivers/s390/cio/css.h b/drivers/s390/cio/css.h index 80ebdddf7747..33bb4d891e16 100644 --- a/drivers/s390/cio/css.h +++ b/drivers/s390/cio/css.h @@ -133,6 +133,8 @@ struct channel_subsystem { extern struct channel_subsystem *channel_subsystems[]; +void channel_subsystem_reinit(void); + /* Helper functions to build lists for the slow path. */ void css_schedule_eval(struct subchannel_id schid); void css_schedule_eval_all(void); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 2d2ecdcc8cdb..2fe2bc2a57ea 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -151,14 +151,6 @@ extern int pwrsw_enabled; extern int unaligned_enabled; #endif -#ifdef CONFIG_S390 -#ifdef CONFIG_MATHEMU -extern int sysctl_ieee_emulation_warnings; -#endif -extern int sysctl_userprocess_debug; -extern int spin_retry; -#endif - #ifdef CONFIG_IA64 extern int no_unaligned_warning; extern int unaligned_dump_stack; -- cgit v1.3-8-gc7d7 From 6d274309d0e64bdbdb6c50945ca2964596e8fa5a Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Fri, 30 Sep 2011 10:48:38 -0500 Subject: irq: support domains with non-zero hwirq base Interrupt controllers can have non-zero starting value for h/w irq numbers. Adding support in irq_domain allows the domain hwirq numbering to match the interrupt controllers' numbering. As this makes looping over irqs for a domain more complicated, add loop iterators to iterate over all hwirqs and irqs for a domain. Signed-off-by: Rob Herring Reviewed-by: Jamie Iles Tested-by: Thomas Abraham Acked-by: Grant Likely Acked-by: Thomas Gleixner --- include/linux/irqdomain.h | 16 +++++++++++++++- kernel/irq/irqdomain.c | 12 ++++++------ 2 files changed, 21 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h index 3ad553e8eae2..99834e581b9e 100644 --- a/include/linux/irqdomain.h +++ b/include/linux/irqdomain.h @@ -47,6 +47,7 @@ struct irq_domain_ops { * of the irq_domain is responsible for allocating the array of * irq_desc structures. * @nr_irq: Number of irqs managed by the irq domain + * @hwirq_base: Starting number for hwirqs managed by the irq domain * @ops: pointer to irq_domain methods * @priv: private data pointer for use by owner. Not touched by irq_domain * core code. @@ -57,6 +58,7 @@ struct irq_domain { struct list_head list; unsigned int irq_base; unsigned int nr_irq; + unsigned int hwirq_base; const struct irq_domain_ops *ops; void *priv; struct device_node *of_node; @@ -72,9 +74,21 @@ struct irq_domain { static inline unsigned int irq_domain_to_irq(struct irq_domain *d, unsigned long hwirq) { - return d->ops->to_irq ? d->ops->to_irq(d, hwirq) : d->irq_base + hwirq; + if (d->ops->to_irq) + return d->ops->to_irq(d, hwirq); + if (WARN_ON(hwirq < d->hwirq_base)) + return 0; + return d->irq_base + hwirq - d->hwirq_base; } +#define irq_domain_for_each_hwirq(d, hw) \ + for (hw = d->hwirq_base; hw < d->hwirq_base + d->nr_irq; hw++) + +#define irq_domain_for_each_irq(d, hw, irq) \ + for (hw = d->hwirq_base, irq = irq_domain_to_irq(d, hw); \ + hw < d->hwirq_base + d->nr_irq; \ + hw++, irq = irq_domain_to_irq(d, hw)) + extern void irq_domain_add(struct irq_domain *domain); extern void irq_domain_del(struct irq_domain *domain); #endif /* CONFIG_IRQ_DOMAIN */ diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index b57a3776de44..200ce832c585 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -20,15 +20,15 @@ static DEFINE_MUTEX(irq_domain_mutex); void irq_domain_add(struct irq_domain *domain) { struct irq_data *d; - int hwirq; + int hwirq, irq; /* * This assumes that the irq_domain owner has already allocated * the irq_descs. This block will be removed when support for dynamic * allocation of irq_descs is added to irq_domain. */ - for (hwirq = 0; hwirq < domain->nr_irq; hwirq++) { - d = irq_get_irq_data(irq_domain_to_irq(domain, hwirq)); + irq_domain_for_each_irq(domain, hwirq, irq) { + d = irq_get_irq_data(irq); if (!d) { WARN(1, "error: assigning domain to non existant irq_desc"); return; @@ -54,15 +54,15 @@ void irq_domain_add(struct irq_domain *domain) void irq_domain_del(struct irq_domain *domain) { struct irq_data *d; - int hwirq; + int hwirq, irq; mutex_lock(&irq_domain_mutex); list_del(&domain->list); mutex_unlock(&irq_domain_mutex); /* Clear the irq_domain assignments */ - for (hwirq = 0; hwirq < domain->nr_irq; hwirq++) { - d = irq_get_irq_data(irq_domain_to_irq(domain, hwirq)); + irq_domain_for_each_irq(domain, hwirq, irq) { + d = irq_get_irq_data(irq); d->domain = NULL; } } -- cgit v1.3-8-gc7d7 From 9a418455134f5dc23f124d2818b2e8e1cea997a1 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Wed, 25 May 2011 09:52:21 -0400 Subject: range: fix bogus misuse of module.h to get printk() This file isn't doing anything with modules and so it should not be including just to get basic stuff like printk() and min/max. Revector it to . Signed-off-by: Paul Gortmaker --- kernel/range.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/range.c b/kernel/range.c index 37fa9b99ad58..9b8ae2d6ed68 100644 --- a/kernel/range.c +++ b/kernel/range.c @@ -1,7 +1,7 @@ /* * Range add and subtract */ -#include +#include #include #include -- cgit v1.3-8-gc7d7 From 9984de1a5a8a96275fcab818f7419af5a3c86e71 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 23 May 2011 14:51:41 -0400 Subject: kernel: Map most files to use export.h instead of module.h The changed files were only including linux/module.h for the EXPORT_SYMBOL infrastructure, and nothing else. Revector them onto the isolated export header for faster compile times. Nothing to see here but a whole lot of instances of: -#include +#include This commit is only changing the kernel dir; next targets will probably be mm, fs, the arch dirs, etc. Signed-off-by: Paul Gortmaker --- kernel/async.c | 2 +- kernel/audit.c | 2 +- kernel/auditsc.c | 2 +- kernel/capability.c | 2 +- kernel/cgroup_freezer.c | 2 +- kernel/cpu.c | 2 +- kernel/cpuset.c | 2 +- kernel/crash_dump.c | 2 +- kernel/cred.c | 2 +- kernel/dma.c | 2 +- kernel/freezer.c | 2 +- kernel/futex.c | 2 +- kernel/groups.c | 2 +- kernel/hrtimer.c | 2 +- kernel/hung_task.c | 2 +- kernel/irq_work.c | 2 +- kernel/kfifo.c | 2 +- kernel/kprobes.c | 2 +- kernel/ksysfs.c | 2 +- kernel/kthread.c | 2 +- kernel/latencytop.c | 2 +- kernel/lockdep_proc.c | 2 +- kernel/module.c | 2 +- kernel/mutex-debug.c | 2 +- kernel/mutex.c | 2 +- kernel/notifier.c | 2 +- kernel/nsproxy.c | 2 +- kernel/padata.c | 2 +- kernel/pid.c | 2 +- kernel/posix-timers.c | 2 +- kernel/profile.c | 2 +- kernel/ptrace.c | 2 +- kernel/rcupdate.c | 2 +- kernel/rcutiny.c | 2 +- kernel/rcutree.c | 2 +- kernel/relay.c | 2 +- kernel/resource.c | 2 +- kernel/rtmutex-debug.c | 2 +- kernel/rtmutex-tester.c | 2 +- kernel/rtmutex.c | 2 +- kernel/rwsem.c | 2 +- kernel/sched_clock.c | 2 +- kernel/semaphore.c | 2 +- kernel/signal.c | 2 +- kernel/smp.c | 2 +- kernel/softirq.c | 2 +- kernel/spinlock.c | 2 +- kernel/srcu.c | 2 +- kernel/stacktrace.c | 2 +- kernel/stop_machine.c | 2 +- kernel/sys.c | 2 +- kernel/time.c | 2 +- kernel/timer.c | 2 +- kernel/up.c | 2 +- kernel/user-return-notifier.c | 2 +- kernel/user.c | 2 +- kernel/user_namespace.c | 2 +- kernel/utsname.c | 2 +- kernel/utsname_sysctl.c | 2 +- kernel/wait.c | 2 +- kernel/workqueue.c | 2 +- 61 files changed, 61 insertions(+), 61 deletions(-) (limited to 'kernel') diff --git a/kernel/async.c b/kernel/async.c index 4c2843c0043e..80b74b88fefe 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -51,7 +51,7 @@ asynchronous and synchronous parts of the kernel. #include #include #include -#include +#include #include #include #include diff --git a/kernel/audit.c b/kernel/audit.c index 0a1355ca3d79..09fae2677a45 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -45,7 +45,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/auditsc.c b/kernel/auditsc.c index ce4b054acee5..47b7fc1ea893 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -48,7 +48,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/capability.c b/kernel/capability.c index 283c529f8b1c..b463871a4e69 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index e691818d7e45..5e828a2ca8e6 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -14,7 +14,7 @@ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -#include +#include #include #include #include diff --git a/kernel/cpu.c b/kernel/cpu.c index 12b7458f23b1..6a81ca906a06 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 10131fdaff70..d970fb508e34 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -37,7 +37,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/crash_dump.c b/kernel/crash_dump.c index 5f85690285d4..20e699265cef 100644 --- a/kernel/crash_dump.c +++ b/kernel/crash_dump.c @@ -2,7 +2,7 @@ #include #include #include -#include +#include /* * If we have booted due to a crash, max_pfn will be a very low value. We need diff --git a/kernel/cred.c b/kernel/cred.c index bb55d052d858..5791612a4045 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -8,7 +8,7 @@ * as published by the Free Software Foundation; either version * 2 of the Licence, or (at your option) any later version. */ -#include +#include #include #include #include diff --git a/kernel/dma.c b/kernel/dma.c index f903189c5304..68a2306522c8 100644 --- a/kernel/dma.c +++ b/kernel/dma.c @@ -9,7 +9,7 @@ * [It also happened to remove the sizeof(char *) == sizeof(int) * assumption introduced because of those /proc/dma patches. -- Hennus] */ -#include +#include #include #include #include diff --git a/kernel/freezer.c b/kernel/freezer.c index 66a594e8ad2f..f24aa0005530 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -6,7 +6,7 @@ #include #include -#include +#include #include #include diff --git a/kernel/futex.c b/kernel/futex.c index 1511dff0cfd6..ea87f4d2f455 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -55,7 +55,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/groups.c b/kernel/groups.c index 1cc476d52dd3..99b53d1eb7ea 100644 --- a/kernel/groups.c +++ b/kernel/groups.c @@ -2,7 +2,7 @@ * Supplementary group IDs */ #include -#include +#include #include #include #include diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index a9205e32a059..422e567eecf6 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -32,7 +32,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/hung_task.c b/kernel/hung_task.c index ea640120ab86..8b1748d0172c 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include /* diff --git a/kernel/irq_work.c b/kernel/irq_work.c index 0e2cde4f380b..3e460ea44955 100644 --- a/kernel/irq_work.c +++ b/kernel/irq_work.c @@ -6,7 +6,7 @@ */ #include -#include +#include #include #include diff --git a/kernel/kfifo.c b/kernel/kfifo.c index 01a0700e873f..c744b88c44e2 100644 --- a/kernel/kfifo.c +++ b/kernel/kfifo.c @@ -20,7 +20,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 2f193d0ba7f2..e5d84644823b 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -36,7 +36,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index 3b053c04dd86..6771de3a655d 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/kthread.c b/kernel/kthread.c index 4ba7cccb4994..b6d216a92639 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 4ac8ebfcab59..a462b317f9a0 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -53,7 +53,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index 71edd2f60c02..91c32a0b612c 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -11,7 +11,7 @@ * Code for /proc/lockdep and /proc/lockdep_stats: * */ -#include +#include #include #include #include diff --git a/kernel/module.c b/kernel/module.c index 93342d992f34..84205ae1607a 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -16,7 +16,7 @@ along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include +#include #include #include #include diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 73da83aff418..7e3443fe1f48 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -14,7 +14,7 @@ */ #include #include -#include +#include #include #include #include diff --git a/kernel/mutex.c b/kernel/mutex.c index d607ed5dd441..89096dd8786f 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -19,7 +19,7 @@ */ #include #include -#include +#include #include #include #include diff --git a/kernel/notifier.c b/kernel/notifier.c index 8d7b435806c9..2d5cc4ccff7f 100644 --- a/kernel/notifier.c +++ b/kernel/notifier.c @@ -1,6 +1,6 @@ #include #include -#include +#include #include #include #include diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c index 9aeab4b98c64..b576f7f14bc6 100644 --- a/kernel/nsproxy.c +++ b/kernel/nsproxy.c @@ -14,7 +14,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/padata.c b/kernel/padata.c index b91941df5e63..b45259931512 100644 --- a/kernel/padata.c +++ b/kernel/padata.c @@ -18,7 +18,7 @@ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. */ -#include +#include #include #include #include diff --git a/kernel/pid.c b/kernel/pid.c index 8cafe7e72ad2..fa5f72227e5f 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -27,7 +27,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 4556182527f3..69185ae6b701 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -46,7 +46,7 @@ #include #include #include -#include +#include /* * Management arrays for POSIX timers. Timers are kept in slab memory diff --git a/kernel/profile.c b/kernel/profile.c index 961b389fe52f..76b8e77773ee 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -13,7 +13,7 @@ * to resolve timer interrupt livelocks, William Irwin, Oracle, 2004 */ -#include +#include #include #include #include diff --git a/kernel/ptrace.c b/kernel/ptrace.c index a70d2a5d8c7b..24d04477b257 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -8,7 +8,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index ca0d23b6b3e8..c5b98e565aee 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -43,7 +43,7 @@ #include #include #include -#include +#include #include #define CREATE_TRACE_POINTS diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index da775c87f27f..b5e525d67fe3 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -28,7 +28,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/rcutree.c b/kernel/rcutree.c index e234eb92a177..6b76d812740c 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -38,7 +38,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/relay.c b/kernel/relay.c index 859ea5a9605f..226fade4d727 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/resource.c b/kernel/resource.c index c8dc249da5ce..7640b3a947d0 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -7,7 +7,7 @@ * Arbitrary resource management. */ -#include +#include #include #include #include diff --git a/kernel/rtmutex-debug.c b/kernel/rtmutex-debug.c index a2e7e7210f3e..8eafd1bd273e 100644 --- a/kernel/rtmutex-debug.c +++ b/kernel/rtmutex-debug.c @@ -18,7 +18,7 @@ */ #include #include -#include +#include #include #include #include diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index 5c9ccd380966..3d9f31cd79e7 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -7,7 +7,7 @@ * */ #include -#include +#include #include #include #include diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c index 5e8d9cce7470..f9d8482dd487 100644 --- a/kernel/rtmutex.c +++ b/kernel/rtmutex.c @@ -11,7 +11,7 @@ * See Documentation/rt-mutex-design.txt for details. */ #include -#include +#include #include #include diff --git a/kernel/rwsem.c b/kernel/rwsem.c index 9f48f3d82e9b..b152f74f02de 100644 --- a/kernel/rwsem.c +++ b/kernel/rwsem.c @@ -7,7 +7,7 @@ #include #include #include -#include +#include #include #include diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 9d8af0b3fb64..c685e31492df 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -62,7 +62,7 @@ */ #include #include -#include +#include #include #include #include diff --git a/kernel/semaphore.c b/kernel/semaphore.c index d831841e55a7..60636a4e25c3 100644 --- a/kernel/semaphore.c +++ b/kernel/semaphore.c @@ -27,7 +27,7 @@ #include #include -#include +#include #include #include #include diff --git a/kernel/signal.c b/kernel/signal.c index d252be2d3de5..b3f78d09a105 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -11,7 +11,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/smp.c b/kernel/smp.c index fb67dfa8394e..db197d60489b 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -6,7 +6,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/softirq.c b/kernel/softirq.c index fca82c32042b..2c71d91efff0 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -10,7 +10,7 @@ * Remote softirq infrastructure is by Jens Axboe. */ -#include +#include #include #include #include diff --git a/kernel/spinlock.c b/kernel/spinlock.c index be6517fb9c14..84c7d96918bf 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -19,7 +19,7 @@ #include #include #include -#include +#include /* * If lockdep is enabled then we use the non-preemption spin-ops diff --git a/kernel/srcu.c b/kernel/srcu.c index 73ce23feaea9..0febf61e1aa3 100644 --- a/kernel/srcu.c +++ b/kernel/srcu.c @@ -24,7 +24,7 @@ * */ -#include +#include #include #include #include diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c index d20c6983aad9..00fe55cc5a82 100644 --- a/kernel/stacktrace.c +++ b/kernel/stacktrace.c @@ -7,7 +7,7 @@ */ #include #include -#include +#include #include #include diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index ba5070ce5765..e78db365fa83 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/kernel/sys.c b/kernel/sys.c index 58459509b14c..4a0286241829 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -4,7 +4,7 @@ * Copyright (C) 1991, 1992 Linus Torvalds */ -#include +#include #include #include #include diff --git a/kernel/time.c b/kernel/time.c index d77606214529..73e416db0a1e 100644 --- a/kernel/time.c +++ b/kernel/time.c @@ -27,7 +27,7 @@ * with nanosecond accuracy */ -#include +#include #include #include #include diff --git a/kernel/timer.c b/kernel/timer.c index 8cff36119e4d..dbaa62422b13 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -20,7 +20,7 @@ */ #include -#include +#include #include #include #include diff --git a/kernel/up.c b/kernel/up.c index 1ff27a28bb7d..c54c75e9faf7 100644 --- a/kernel/up.c +++ b/kernel/up.c @@ -4,7 +4,7 @@ #include #include -#include +#include #include int smp_call_function_single(int cpu, void (*func) (void *info), void *info, diff --git a/kernel/user-return-notifier.c b/kernel/user-return-notifier.c index 92cb706c7fc8..1744bb80f1fb 100644 --- a/kernel/user-return-notifier.c +++ b/kernel/user-return-notifier.c @@ -2,7 +2,7 @@ #include #include #include -#include +#include static DEFINE_PER_CPU(struct hlist_head, return_notifier_list); diff --git a/kernel/user.c b/kernel/user.c index 9e03e9c1df8d..71dd2363ab0f 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include /* diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index 9da289c34f22..3b906e98b1db 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -5,7 +5,7 @@ * License. */ -#include +#include #include #include #include diff --git a/kernel/utsname.c b/kernel/utsname.c index bff131b9510a..405caf91aad5 100644 --- a/kernel/utsname.c +++ b/kernel/utsname.c @@ -9,7 +9,7 @@ * License. */ -#include +#include #include #include #include diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index a2cd77e70d4d..5a709452ec19 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -9,7 +9,7 @@ * License. */ -#include +#include #include #include #include diff --git a/kernel/wait.c b/kernel/wait.c index f45ea8d2a1ce..26fa7797f90f 100644 --- a/kernel/wait.c +++ b/kernel/wait.c @@ -4,7 +4,7 @@ * (C) 2004 William Irwin, Oracle */ #include -#include +#include #include #include #include diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1783aabc6128..42fa9ad0a810 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -23,7 +23,7 @@ * Please read Documentation/workqueue.txt for details. */ -#include +#include #include #include #include -- cgit v1.3-8-gc7d7 From 56d82e000cdfb51aa92241d4682302f78c35cd92 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 26 May 2011 17:53:52 -0400 Subject: kernel: Add to files using it implicitly These files are doing things like module_put and try_module_get so they need to call out the module.h for explicit inclusion, rather than getting it via which we ideally want to remove the module.h inclusion from. Signed-off-by: Paul Gortmaker --- kernel/trace/ftrace.c | 1 + kernel/trace/trace_syscalls.c | 1 + 2 files changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 077d85387908..900b409543db 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index ee7b5a0bb9f8..cb654542c1a1 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -2,6 +2,7 @@ #include #include #include +#include /* for MODULE_NAME_LEN via KSYM_SYMBOL_LEN */ #include #include #include -- cgit v1.3-8-gc7d7 From 74da1ff71350f3638c51613085f89c0865d7fe08 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 26 May 2011 12:48:41 -0400 Subject: kernel: fix several implicit usasges of kmod.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit These files were implicitly relying on coming in via module.h, as without it we get things like: kernel/power/suspend.c:100: error: implicit declaration of function ‘usermodehelper_disable’ kernel/power/suspend.c:109: error: implicit declaration of function ‘usermodehelper_enable’ kernel/power/user.c:254: error: implicit declaration of function ‘usermodehelper_disable’ kernel/power/user.c:261: error: implicit declaration of function ‘usermodehelper_enable’ kernel/sys.c:317: error: implicit declaration of function ‘usermodehelper_disable’ kernel/sys.c:1816: error: implicit declaration of function ‘call_usermodehelper_setup’ kernel/sys.c:1822: error: implicit declaration of function ‘call_usermodehelper_setfns’ kernel/sys.c:1824: error: implicit declaration of function ‘call_usermodehelper_exec’ Signed-off-by: Paul Gortmaker --- kernel/power/suspend.c | 1 + kernel/power/user.c | 1 + kernel/sys.c | 1 + 3 files changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index fdd4263b995d..31aae3219f89 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/power/user.c b/kernel/power/user.c index 42ddbc6f0de6..6d8f535c2b88 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/sys.c b/kernel/sys.c index 4a0286241829..d2cb1cda823a 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include -- cgit v1.3-8-gc7d7 From 967d1f90625ed9c1ab205d3f738fedf9d852e1fd Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 18 Jul 2011 13:03:04 -0400 Subject: kernel: fix two implicit header assumptions in irq_work.c Up until now, this file was getting percpu.h because nearly every file was implicitly getting module.h (and all its sub-includes). But we want to clean that up, so call out percpu.h explicitly. Otherwise we'll get things like this on an ARM build: kernel/irq_work.c:48: error: expected declaration specifiers or '...' before 'irq_work_list' kernel/irq_work.c:48: warning: type defaults to 'int' in declaration of 'DEFINE_PER_CPU' The same thing was happening for builds on ARM for asm/processor.h kernel/irq_work.c: In function 'irq_work_sync': kernel/irq_work.c:166: error: implicit declaration of function 'cpu_relax' Signed-off-by: Paul Gortmaker --- kernel/irq_work.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/irq_work.c b/kernel/irq_work.c index 3e460ea44955..c3c46c72046e 100644 --- a/kernel/irq_work.c +++ b/kernel/irq_work.c @@ -8,7 +8,9 @@ #include #include #include +#include #include +#include /* * An entry can be in one of four states: -- cgit v1.3-8-gc7d7 From 1596425fd7545abccb7b5059376e4d56c3d6be4d Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 28 Jul 2011 14:22:29 -0400 Subject: kernel: ksysfs.c is implicitly using stat.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With the module.h usage cleanup, we'll get this: kernel/ksysfs.c:161: error: ‘S_IRUGO’ undeclared here (not in a function) make[2]: *** [kernel/ksysfs.o] Error 1 Signed-off-by: Paul Gortmaker --- kernel/ksysfs.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index 6771de3a655d..4e316e1acf58 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include -- cgit v1.3-8-gc7d7 From 72a59aaada499d9bbf19f2fb68daa37502e4a9bb Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 26 May 2011 21:07:10 -0400 Subject: kernel: params.c needs module.h not moduleparam.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Through various other implicit include paths, some files were getting the full module.h file, and hence living the illusion that they really only needed moduleparam.h -- but the reality is that once you remove the module.h presence, these show up: kernel/params.c:583: warning: ‘struct module_kobject’ declared inside parameter list Such files really require module.h so simply make it so. As the file module.h grabs moduleparam.h on the fly, all will be well. Signed-off-by: Paul Gortmaker --- kernel/params.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/params.c b/kernel/params.c index 821788947e40..65aae11eb93f 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -15,7 +15,7 @@ along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include +#include #include #include #include -- cgit v1.3-8-gc7d7 From bdfa97bf7263657b83bc5b68567a3a60dde84c5b Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Tue, 25 Oct 2011 13:13:57 -0400 Subject: kernel: fix up module header handling in rcutiny files The file rcutiny.c does not need moduleparam.h header, as there are no modparams in this file. However rcutiny_plugin.h does define a module_init() and a module_exit() and it uses the various MODULE_ macros, so it really does need module.h included. Signed-off-by: Paul Gortmaker --- kernel/rcutiny.c | 1 - kernel/rcutiny_plugin.h | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index b5e525d67fe3..636af6d9c6e5 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -22,7 +22,6 @@ * For detailed explanation of Read-Copy Update mechanism see - * Documentation/RCU */ -#include #include #include #include diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 02aa7139861c..2b0484a5dc28 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -23,6 +23,7 @@ */ #include +#include #include #include -- cgit v1.3-8-gc7d7 From 6e5fdeedca610df600aabc393c4b1f44b128fe49 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Thu, 26 May 2011 16:00:52 -0400 Subject: kernel: Fix files explicitly needing EXPORT_SYMBOL infrastructure These files were getting via an implicit non-obvious path, but we want to crush those out of existence since they cost time during compiles of processing thousands of lines of headers for no reason. Give them the lightweight header that just contains the EXPORT_SYMBOL infrastructure. Signed-off-by: Paul Gortmaker --- kernel/compat.c | 1 + kernel/debug/kdb/kdb_debugger.c | 1 + kernel/events/core.c | 1 + kernel/irq/generic-chip.c | 1 + kernel/power/hibernate.c | 1 + kernel/power/main.c | 1 + kernel/power/qos.c | 1 + kernel/power/suspend.c | 1 + kernel/time/posix-clock.c | 1 + kernel/trace/blktrace.c | 1 + 10 files changed, 10 insertions(+) (limited to 'kernel') diff --git a/kernel/compat.c b/kernel/compat.c index e2435ee9993a..f346cedfe24d 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -21,6 +21,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/debug/kdb/kdb_debugger.c b/kernel/debug/kdb/kdb_debugger.c index d9ca9aa481ec..8b68ce78ff17 100644 --- a/kernel/debug/kdb/kdb_debugger.c +++ b/kernel/debug/kdb/kdb_debugger.c @@ -11,6 +11,7 @@ #include #include #include +#include #include "kdb_private.h" #include "../debug_core.h" diff --git a/kernel/events/core.c b/kernel/events/core.c index d1a1bee35228..92b8811f2234 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c index 6cb7613e4bf4..c89295a8f668 100644 --- a/kernel/irq/generic-chip.c +++ b/kernel/irq/generic-chip.c @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 1c53f7fad5f7..b4511b6d3ef9 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -9,6 +9,7 @@ * This file is released under the GPLv2. */ +#include #include #include #include diff --git a/kernel/power/main.c b/kernel/power/main.c index a52e88425a31..71f49fe4377e 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -8,6 +8,7 @@ * */ +#include #include #include #include diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 1c1797dd1d1d..2c0a65e27626 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -43,6 +43,7 @@ #include #include +#include /* * locking rule: all changes to constraints or notifiers lists diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 31aae3219f89..4953dc054c53 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include diff --git a/kernel/time/posix-clock.c b/kernel/time/posix-clock.c index c340ca658f37..ce033c7aa2e8 100644 --- a/kernel/time/posix-clock.c +++ b/kernel/time/posix-clock.c @@ -18,6 +18,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #include +#include #include #include #include diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 7c910a5593a6..16fc34a0806f 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include -- cgit v1.3-8-gc7d7 From ec53cf23c0ddb0c29950b9a4ac46964c4c6c6c2f Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Mon, 19 Sep 2011 20:33:19 -0400 Subject: irq: don't put module.h into irq.h for tracking irqgen modules. Recent commit "irq: Track the owner of irq descriptor" in commit ID b6873807a7143b7 placed module.h into linux/irq.h but we are trying to limit module.h inclusion to just C files that really need it, due to its size and number of children includes. This targets just reversing that include. Add in the basic "struct module" since that is all we really need to ensure things compile. In theory, b687380 should have added the module.h include to the irqdesc.h header as well, but the implicit module.h everywhere presence masked this from showing up. So give it the "struct module" as well. As for the C files, irqdesc.c is only using THIS_MODULE, so it does not need module.h - give it export.h instead. The C file irq/manage.c is now (as of b687380) using try_module_get and module_put and so it needs module.h (which it already has). Also convert the irq_alloc_descs variants to macros, since all they really do is is call the __irq_alloc_descs primitive. This avoids including export.h and no debug info is lost. Signed-off-by: Paul Gortmaker --- include/linux/irq.h | 32 ++++++++++++-------------------- include/linux/irqdesc.h | 1 + kernel/irq/irqdesc.c | 2 +- 3 files changed, 14 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/include/linux/irq.h b/include/linux/irq.h index 59e49c80cc2c..bff29c58da23 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -23,13 +23,13 @@ #include #include #include -#include #include #include #include struct seq_file; +struct module; struct irq_desc; struct irq_data; typedef void (*irq_flow_handler_t)(unsigned int irq, @@ -567,29 +567,21 @@ static inline struct msi_desc *irq_data_get_msi(struct irq_data *d) int __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, struct module *owner); -static inline int irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, - int node) -{ - return __irq_alloc_descs(irq, from, cnt, node, THIS_MODULE); -} +/* use macros to avoid needing export.h for THIS_MODULE */ +#define irq_alloc_descs(irq, from, cnt, node) \ + __irq_alloc_descs(irq, from, cnt, node, THIS_MODULE) -void irq_free_descs(unsigned int irq, unsigned int cnt); -int irq_reserve_irqs(unsigned int from, unsigned int cnt); +#define irq_alloc_desc(node) \ + irq_alloc_descs(-1, 0, 1, node) -static inline int irq_alloc_desc(int node) -{ - return irq_alloc_descs(-1, 0, 1, node); -} +#define irq_alloc_desc_at(at, node) \ + irq_alloc_descs(at, at, 1, node) -static inline int irq_alloc_desc_at(unsigned int at, int node) -{ - return irq_alloc_descs(at, at, 1, node); -} +#define irq_alloc_desc_from(from, node) \ + irq_alloc_descs(-1, from, 1, node) -static inline int irq_alloc_desc_from(unsigned int from, int node) -{ - return irq_alloc_descs(-1, from, 1, node); -} +void irq_free_descs(unsigned int irq, unsigned int cnt); +int irq_reserve_irqs(unsigned int from, unsigned int cnt); static inline void irq_free_desc(unsigned int irq) { diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 6b69c2c9dff1..f1e2527006bd 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -11,6 +11,7 @@ struct irq_affinity_notify; struct proc_dir_entry; struct timer_rand_state; +struct module; /** * struct irq_desc - interrupt descriptor * @irq_data: per irq and chip data passed down to chip functions diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 1550e8447a16..d86e254b95eb 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -9,7 +9,7 @@ */ #include #include -#include +#include #include #include #include -- cgit v1.3-8-gc7d7 From fcf634098c00dd9cd247447368495f0b79be12d1 Mon Sep 17 00:00:00 2001 From: Christopher Yeoh Date: Mon, 31 Oct 2011 17:06:39 -0700 Subject: Cross Memory Attach The basic idea behind cross memory attach is to allow MPI programs doing intra-node communication to do a single copy of the message rather than a double copy of the message via shared memory. The following patch attempts to achieve this by allowing a destination process, given an address and size from a source process, to copy memory directly from the source process into its own address space via a system call. There is also a symmetrical ability to copy from the current process's address space into a destination process's address space. - Use of /proc/pid/mem has been considered, but there are issues with using it: - Does not allow for specifying iovecs for both src and dest, assuming preadv or pwritev was implemented either the area read from or written to would need to be contiguous. - Currently mem_read allows only processes who are currently ptrace'ing the target and are still able to ptrace the target to read from the target. This check could possibly be moved to the open call, but its not clear exactly what race this restriction is stopping (reason appears to have been lost) - Having to send the fd of /proc/self/mem via SCM_RIGHTS on unix domain socket is a bit ugly from a userspace point of view, especially when you may have hundreds if not (eventually) thousands of processes that all need to do this with each other - Doesn't allow for some future use of the interface we would like to consider adding in the future (see below) - Interestingly reading from /proc/pid/mem currently actually involves two copies! (But this could be fixed pretty easily) As mentioned previously use of vmsplice instead was considered, but has problems. Since you need the reader and writer working co-operatively if the pipe is not drained then you block. Which requires some wrapping to do non blocking on the send side or polling on the receive. In all to all communication it requires ordering otherwise you can deadlock. And in the example of many MPI tasks writing to one MPI task vmsplice serialises the copying. There are some cases of MPI collectives where even a single copy interface does not get us the performance gain we could. For example in an MPI_Reduce rather than copy the data from the source we would like to instead use it directly in a mathops (say the reduce is doing a sum) as this would save us doing a copy. We don't need to keep a copy of the data from the source. I haven't implemented this, but I think this interface could in the future do all this through the use of the flags - eg could specify the math operation and type and the kernel rather than just copying the data would apply the specified operation between the source and destination and store it in the destination. Although we don't have a "second user" of the interface (though I've had some nibbles from people who may be interested in using it for intra process messaging which is not MPI). This interface is something which hardware vendors are already doing for their custom drivers to implement fast local communication. And so in addition to this being useful for OpenMPI it would mean the driver maintainers don't have to fix things up when the mm changes. There was some discussion about how much faster a true zero copy would go. Here's a link back to the email with some testing I did on that: http://marc.info/?l=linux-mm&m=130105930902915&w=2 There is a basic man page for the proposed interface here: http://ozlabs.org/~cyeoh/cma/process_vm_readv.txt This has been implemented for x86 and powerpc, other architecture should mainly (I think) just need to add syscall numbers for the process_vm_readv and process_vm_writev. There are 32 bit compatibility versions for 64-bit kernels. For arch maintainers there are some simple tests to be able to quickly verify that the syscalls are working correctly here: http://ozlabs.org/~cyeoh/cma/cma-test-20110718.tgz Signed-off-by: Chris Yeoh Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Thomas Gleixner Cc: Arnd Bergmann Cc: Paul Mackerras Cc: Benjamin Herrenschmidt Cc: David Howells Cc: James Morris Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/include/asm/systbl.h | 2 + arch/powerpc/include/asm/unistd.h | 4 +- arch/x86/ia32/ia32entry.S | 2 + arch/x86/include/asm/unistd_32.h | 4 +- arch/x86/include/asm/unistd_64.h | 4 + arch/x86/kernel/syscall_table_32.S | 2 + fs/aio.c | 4 +- fs/compat.c | 7 +- fs/read_write.c | 8 +- include/linux/compat.h | 3 +- include/linux/fs.h | 7 +- include/linux/syscalls.h | 13 + kernel/sys_ni.c | 4 + mm/Makefile | 3 +- mm/process_vm_access.c | 496 +++++++++++++++++++++++++++++++++++++ security/keys/compat.c | 2 +- security/keys/keyctl.c | 2 +- 17 files changed, 550 insertions(+), 17 deletions(-) create mode 100644 mm/process_vm_access.c (limited to 'kernel') diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h index fa0d27a400de..559ae1ee6706 100644 --- a/arch/powerpc/include/asm/systbl.h +++ b/arch/powerpc/include/asm/systbl.h @@ -354,3 +354,5 @@ COMPAT_SYS_SPU(clock_adjtime) SYSCALL_SPU(syncfs) COMPAT_SYS_SPU(sendmmsg) SYSCALL_SPU(setns) +COMPAT_SYS(process_vm_readv) +COMPAT_SYS(process_vm_writev) diff --git a/arch/powerpc/include/asm/unistd.h b/arch/powerpc/include/asm/unistd.h index b8b3f599362b..d3d1b5efd7eb 100644 --- a/arch/powerpc/include/asm/unistd.h +++ b/arch/powerpc/include/asm/unistd.h @@ -373,10 +373,12 @@ #define __NR_syncfs 348 #define __NR_sendmmsg 349 #define __NR_setns 350 +#define __NR_process_vm_readv 351 +#define __NR_process_vm_writev 352 #ifdef __KERNEL__ -#define __NR_syscalls 351 +#define __NR_syscalls 353 #define __NR__exit __NR_exit #define NR_syscalls __NR_syscalls diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 54edb207ff3a..a6253ec1b284 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -850,4 +850,6 @@ ia32_sys_call_table: .quad sys_syncfs .quad compat_sys_sendmmsg /* 345 */ .quad sys_setns + .quad compat_sys_process_vm_readv + .quad compat_sys_process_vm_writev ia32_syscall_end: diff --git a/arch/x86/include/asm/unistd_32.h b/arch/x86/include/asm/unistd_32.h index 593485b38ab3..599c77d38f33 100644 --- a/arch/x86/include/asm/unistd_32.h +++ b/arch/x86/include/asm/unistd_32.h @@ -352,10 +352,12 @@ #define __NR_syncfs 344 #define __NR_sendmmsg 345 #define __NR_setns 346 +#define __NR_process_vm_readv 347 +#define __NR_process_vm_writev 348 #ifdef __KERNEL__ -#define NR_syscalls 347 +#define NR_syscalls 349 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 0a6ba337a2eb..0431f193c3f2 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -682,6 +682,10 @@ __SYSCALL(__NR_sendmmsg, sys_sendmmsg) __SYSCALL(__NR_setns, sys_setns) #define __NR_getcpu 309 __SYSCALL(__NR_getcpu, sys_getcpu) +#define __NR_process_vm_readv 310 +__SYSCALL(__NR_process_vm_readv, sys_process_vm_readv) +#define __NR_process_vm_writev 311 +__SYSCALL(__NR_process_vm_writev, sys_process_vm_writev) #ifndef __NO_STUBS #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index bc19be332bc9..9a0e31293920 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -346,3 +346,5 @@ ENTRY(sys_call_table) .long sys_syncfs .long sys_sendmmsg /* 345 */ .long sys_setns + .long sys_process_vm_readv + .long sys_process_vm_writev diff --git a/fs/aio.c b/fs/aio.c index e29ec485af25..632b235f4fbe 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -1387,13 +1387,13 @@ static ssize_t aio_setup_vectored_rw(int type, struct kiocb *kiocb, bool compat) ret = compat_rw_copy_check_uvector(type, (struct compat_iovec __user *)kiocb->ki_buf, kiocb->ki_nbytes, 1, &kiocb->ki_inline_vec, - &kiocb->ki_iovec); + &kiocb->ki_iovec, 1); else #endif ret = rw_copy_check_uvector(type, (struct iovec __user *)kiocb->ki_buf, kiocb->ki_nbytes, 1, &kiocb->ki_inline_vec, - &kiocb->ki_iovec); + &kiocb->ki_iovec, 1); if (ret < 0) goto out; diff --git a/fs/compat.c b/fs/compat.c index 302e761bd0aa..c98787536bb8 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -546,7 +546,7 @@ out: ssize_t compat_rw_copy_check_uvector(int type, const struct compat_iovec __user *uvector, unsigned long nr_segs, unsigned long fast_segs, struct iovec *fast_pointer, - struct iovec **ret_pointer) + struct iovec **ret_pointer, int check_access) { compat_ssize_t tot_len; struct iovec *iov = *ret_pointer = fast_pointer; @@ -593,7 +593,8 @@ ssize_t compat_rw_copy_check_uvector(int type, } if (len < 0) /* size_t not fitting in compat_ssize_t .. */ goto out; - if (!access_ok(vrfy_dir(type), compat_ptr(buf), len)) { + if (check_access && + !access_ok(vrfy_dir(type), compat_ptr(buf), len)) { ret = -EFAULT; goto out; } @@ -1107,7 +1108,7 @@ static ssize_t compat_do_readv_writev(int type, struct file *file, goto out; tot_len = compat_rw_copy_check_uvector(type, uvector, nr_segs, - UIO_FASTIOV, iovstack, &iov); + UIO_FASTIOV, iovstack, &iov, 1); if (tot_len == 0) { ret = 0; goto out; diff --git a/fs/read_write.c b/fs/read_write.c index dfd125798791..5ad4248b0cd8 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -633,7 +633,8 @@ ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov, ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, unsigned long nr_segs, unsigned long fast_segs, struct iovec *fast_pointer, - struct iovec **ret_pointer) + struct iovec **ret_pointer, + int check_access) { unsigned long seg; ssize_t ret; @@ -689,7 +690,8 @@ ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, ret = -EINVAL; goto out; } - if (unlikely(!access_ok(vrfy_dir(type), buf, len))) { + if (check_access + && unlikely(!access_ok(vrfy_dir(type), buf, len))) { ret = -EFAULT; goto out; } @@ -721,7 +723,7 @@ static ssize_t do_readv_writev(int type, struct file *file, } ret = rw_copy_check_uvector(type, uvector, nr_segs, - ARRAY_SIZE(iovstack), iovstack, &iov); + ARRAY_SIZE(iovstack), iovstack, &iov, 1); if (ret <= 0) goto out; diff --git a/include/linux/compat.h b/include/linux/compat.h index c6e7523bf765..154bf5683015 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -547,7 +547,8 @@ extern ssize_t compat_rw_copy_check_uvector(int type, const struct compat_iovec __user *uvector, unsigned long nr_segs, unsigned long fast_segs, struct iovec *fast_pointer, - struct iovec **ret_pointer); + struct iovec **ret_pointer, + int check_access); extern void __user *compat_alloc_user_space(unsigned long len); diff --git a/include/linux/fs.h b/include/linux/fs.h index 14493a2d5a03..87b4c6b9692d 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1633,9 +1633,10 @@ struct inode_operations { struct seq_file; ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, - unsigned long nr_segs, unsigned long fast_segs, - struct iovec *fast_pointer, - struct iovec **ret_pointer); + unsigned long nr_segs, unsigned long fast_segs, + struct iovec *fast_pointer, + struct iovec **ret_pointer, + int check_access); extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *); extern ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 1ff0ec2a5e8d..86a24b1166d1 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -844,4 +844,17 @@ asmlinkage long sys_open_by_handle_at(int mountdirfd, struct file_handle __user *handle, int flags); asmlinkage long sys_setns(int fd, int nstype); +asmlinkage long sys_process_vm_readv(pid_t pid, + const struct iovec __user *lvec, + unsigned long liovcnt, + const struct iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags); +asmlinkage long sys_process_vm_writev(pid_t pid, + const struct iovec __user *lvec, + unsigned long liovcnt, + const struct iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags); + #endif diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index a9a5de07c4f1..47bfa16430d7 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -145,6 +145,10 @@ cond_syscall(sys_io_submit); cond_syscall(sys_io_cancel); cond_syscall(sys_io_getevents); cond_syscall(sys_syslog); +cond_syscall(sys_process_vm_readv); +cond_syscall(sys_process_vm_writev); +cond_syscall(compat_sys_process_vm_readv); +cond_syscall(compat_sys_process_vm_writev); /* arch-specific weak syscall entries */ cond_syscall(sys_pciconfig_read); diff --git a/mm/Makefile b/mm/Makefile index 836e4163c1bf..50ec00ef2a0e 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -5,7 +5,8 @@ mmu-y := nommu.o mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \ mlock.o mmap.o mprotect.o mremap.o msync.o rmap.o \ - vmalloc.o pagewalk.o pgtable-generic.o + vmalloc.o pagewalk.o pgtable-generic.o \ + process_vm_access.o obj-y := filemap.o mempool.o oom_kill.o fadvise.o \ maccess.o page_alloc.o page-writeback.o \ diff --git a/mm/process_vm_access.c b/mm/process_vm_access.c new file mode 100644 index 000000000000..e920aa3ce104 --- /dev/null +++ b/mm/process_vm_access.c @@ -0,0 +1,496 @@ +/* + * linux/mm/process_vm_access.c + * + * Copyright (C) 2010-2011 Christopher Yeoh , IBM Corp. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_COMPAT +#include +#endif + +/** + * process_vm_rw_pages - read/write pages from task specified + * @task: task to read/write from + * @mm: mm for task + * @process_pages: struct pages area that can store at least + * nr_pages_to_copy struct page pointers + * @pa: address of page in task to start copying from/to + * @start_offset: offset in page to start copying from/to + * @len: number of bytes to copy + * @lvec: iovec array specifying where to copy to/from + * @lvec_cnt: number of elements in iovec array + * @lvec_current: index in iovec array we are up to + * @lvec_offset: offset in bytes from current iovec iov_base we are up to + * @vm_write: 0 means copy from, 1 means copy to + * @nr_pages_to_copy: number of pages to copy + * @bytes_copied: returns number of bytes successfully copied + * Returns 0 on success, error code otherwise + */ +static int process_vm_rw_pages(struct task_struct *task, + struct mm_struct *mm, + struct page **process_pages, + unsigned long pa, + unsigned long start_offset, + unsigned long len, + const struct iovec *lvec, + unsigned long lvec_cnt, + unsigned long *lvec_current, + size_t *lvec_offset, + int vm_write, + unsigned int nr_pages_to_copy, + ssize_t *bytes_copied) +{ + int pages_pinned; + void *target_kaddr; + int pgs_copied = 0; + int j; + int ret; + ssize_t bytes_to_copy; + ssize_t rc = 0; + + *bytes_copied = 0; + + /* Get the pages we're interested in */ + down_read(&mm->mmap_sem); + pages_pinned = get_user_pages(task, mm, pa, + nr_pages_to_copy, + vm_write, 0, process_pages, NULL); + up_read(&mm->mmap_sem); + + if (pages_pinned != nr_pages_to_copy) { + rc = -EFAULT; + goto end; + } + + /* Do the copy for each page */ + for (pgs_copied = 0; + (pgs_copied < nr_pages_to_copy) && (*lvec_current < lvec_cnt); + pgs_copied++) { + /* Make sure we have a non zero length iovec */ + while (*lvec_current < lvec_cnt + && lvec[*lvec_current].iov_len == 0) + (*lvec_current)++; + if (*lvec_current == lvec_cnt) + break; + + /* + * Will copy smallest of: + * - bytes remaining in page + * - bytes remaining in destination iovec + */ + bytes_to_copy = min_t(ssize_t, PAGE_SIZE - start_offset, + len - *bytes_copied); + bytes_to_copy = min_t(ssize_t, bytes_to_copy, + lvec[*lvec_current].iov_len + - *lvec_offset); + + target_kaddr = kmap(process_pages[pgs_copied]) + start_offset; + + if (vm_write) + ret = copy_from_user(target_kaddr, + lvec[*lvec_current].iov_base + + *lvec_offset, + bytes_to_copy); + else + ret = copy_to_user(lvec[*lvec_current].iov_base + + *lvec_offset, + target_kaddr, bytes_to_copy); + kunmap(process_pages[pgs_copied]); + if (ret) { + *bytes_copied += bytes_to_copy - ret; + pgs_copied++; + rc = -EFAULT; + goto end; + } + *bytes_copied += bytes_to_copy; + *lvec_offset += bytes_to_copy; + if (*lvec_offset == lvec[*lvec_current].iov_len) { + /* + * Need to copy remaining part of page into the + * next iovec if there are any bytes left in page + */ + (*lvec_current)++; + *lvec_offset = 0; + start_offset = (start_offset + bytes_to_copy) + % PAGE_SIZE; + if (start_offset) + pgs_copied--; + } else { + start_offset = 0; + } + } + +end: + if (vm_write) { + for (j = 0; j < pages_pinned; j++) { + if (j < pgs_copied) + set_page_dirty_lock(process_pages[j]); + put_page(process_pages[j]); + } + } else { + for (j = 0; j < pages_pinned; j++) + put_page(process_pages[j]); + } + + return rc; +} + +/* Maximum number of pages kmalloc'd to hold struct page's during copy */ +#define PVM_MAX_KMALLOC_PAGES (PAGE_SIZE * 2) + +/** + * process_vm_rw_single_vec - read/write pages from task specified + * @addr: start memory address of target process + * @len: size of area to copy to/from + * @lvec: iovec array specifying where to copy to/from locally + * @lvec_cnt: number of elements in iovec array + * @lvec_current: index in iovec array we are up to + * @lvec_offset: offset in bytes from current iovec iov_base we are up to + * @process_pages: struct pages area that can store at least + * nr_pages_to_copy struct page pointers + * @mm: mm for task + * @task: task to read/write from + * @vm_write: 0 means copy from, 1 means copy to + * @bytes_copied: returns number of bytes successfully copied + * Returns 0 on success or on failure error code + */ +static int process_vm_rw_single_vec(unsigned long addr, + unsigned long len, + const struct iovec *lvec, + unsigned long lvec_cnt, + unsigned long *lvec_current, + size_t *lvec_offset, + struct page **process_pages, + struct mm_struct *mm, + struct task_struct *task, + int vm_write, + ssize_t *bytes_copied) +{ + unsigned long pa = addr & PAGE_MASK; + unsigned long start_offset = addr - pa; + unsigned long nr_pages; + ssize_t bytes_copied_loop; + ssize_t rc = 0; + unsigned long nr_pages_copied = 0; + unsigned long nr_pages_to_copy; + unsigned long max_pages_per_loop = PVM_MAX_KMALLOC_PAGES + / sizeof(struct pages *); + + *bytes_copied = 0; + + /* Work out address and page range required */ + if (len == 0) + return 0; + nr_pages = (addr + len - 1) / PAGE_SIZE - addr / PAGE_SIZE + 1; + + while ((nr_pages_copied < nr_pages) && (*lvec_current < lvec_cnt)) { + nr_pages_to_copy = min(nr_pages - nr_pages_copied, + max_pages_per_loop); + + rc = process_vm_rw_pages(task, mm, process_pages, pa, + start_offset, len, + lvec, lvec_cnt, + lvec_current, lvec_offset, + vm_write, nr_pages_to_copy, + &bytes_copied_loop); + start_offset = 0; + *bytes_copied += bytes_copied_loop; + + if (rc < 0) { + return rc; + } else { + len -= bytes_copied_loop; + nr_pages_copied += nr_pages_to_copy; + pa += nr_pages_to_copy * PAGE_SIZE; + } + } + + return rc; +} + +/* Maximum number of entries for process pages array + which lives on stack */ +#define PVM_MAX_PP_ARRAY_COUNT 16 + +/** + * process_vm_rw_core - core of reading/writing pages from task specified + * @pid: PID of process to read/write from/to + * @lvec: iovec array specifying where to copy to/from locally + * @liovcnt: size of lvec array + * @rvec: iovec array specifying where to copy to/from in the other process + * @riovcnt: size of rvec array + * @flags: currently unused + * @vm_write: 0 if reading from other process, 1 if writing to other process + * Returns the number of bytes read/written or error code. May + * return less bytes than expected if an error occurs during the copying + * process. + */ +static ssize_t process_vm_rw_core(pid_t pid, const struct iovec *lvec, + unsigned long liovcnt, + const struct iovec *rvec, + unsigned long riovcnt, + unsigned long flags, int vm_write) +{ + struct task_struct *task; + struct page *pp_stack[PVM_MAX_PP_ARRAY_COUNT]; + struct page **process_pages = pp_stack; + struct mm_struct *mm; + unsigned long i; + ssize_t rc = 0; + ssize_t bytes_copied_loop; + ssize_t bytes_copied = 0; + unsigned long nr_pages = 0; + unsigned long nr_pages_iov; + unsigned long iov_l_curr_idx = 0; + size_t iov_l_curr_offset = 0; + ssize_t iov_len; + + /* + * Work out how many pages of struct pages we're going to need + * when eventually calling get_user_pages + */ + for (i = 0; i < riovcnt; i++) { + iov_len = rvec[i].iov_len; + if (iov_len > 0) { + nr_pages_iov = ((unsigned long)rvec[i].iov_base + + iov_len) + / PAGE_SIZE - (unsigned long)rvec[i].iov_base + / PAGE_SIZE + 1; + nr_pages = max(nr_pages, nr_pages_iov); + } + } + + if (nr_pages == 0) + return 0; + + if (nr_pages > PVM_MAX_PP_ARRAY_COUNT) { + /* For reliability don't try to kmalloc more than + 2 pages worth */ + process_pages = kmalloc(min_t(size_t, PVM_MAX_KMALLOC_PAGES, + sizeof(struct pages *)*nr_pages), + GFP_KERNEL); + + if (!process_pages) + return -ENOMEM; + } + + /* Get process information */ + rcu_read_lock(); + task = find_task_by_vpid(pid); + if (task) + get_task_struct(task); + rcu_read_unlock(); + if (!task) { + rc = -ESRCH; + goto free_proc_pages; + } + + task_lock(task); + if (__ptrace_may_access(task, PTRACE_MODE_ATTACH)) { + task_unlock(task); + rc = -EPERM; + goto put_task_struct; + } + mm = task->mm; + + if (!mm || (task->flags & PF_KTHREAD)) { + task_unlock(task); + rc = -EINVAL; + goto put_task_struct; + } + + atomic_inc(&mm->mm_users); + task_unlock(task); + + for (i = 0; i < riovcnt && iov_l_curr_idx < liovcnt; i++) { + rc = process_vm_rw_single_vec( + (unsigned long)rvec[i].iov_base, rvec[i].iov_len, + lvec, liovcnt, &iov_l_curr_idx, &iov_l_curr_offset, + process_pages, mm, task, vm_write, &bytes_copied_loop); + bytes_copied += bytes_copied_loop; + if (rc != 0) { + /* If we have managed to copy any data at all then + we return the number of bytes copied. Otherwise + we return the error code */ + if (bytes_copied) + rc = bytes_copied; + goto put_mm; + } + } + + rc = bytes_copied; +put_mm: + mmput(mm); + +put_task_struct: + put_task_struct(task); + +free_proc_pages: + if (process_pages != pp_stack) + kfree(process_pages); + return rc; +} + +/** + * process_vm_rw - check iovecs before calling core routine + * @pid: PID of process to read/write from/to + * @lvec: iovec array specifying where to copy to/from locally + * @liovcnt: size of lvec array + * @rvec: iovec array specifying where to copy to/from in the other process + * @riovcnt: size of rvec array + * @flags: currently unused + * @vm_write: 0 if reading from other process, 1 if writing to other process + * Returns the number of bytes read/written or error code. May + * return less bytes than expected if an error occurs during the copying + * process. + */ +static ssize_t process_vm_rw(pid_t pid, + const struct iovec __user *lvec, + unsigned long liovcnt, + const struct iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags, int vm_write) +{ + struct iovec iovstack_l[UIO_FASTIOV]; + struct iovec iovstack_r[UIO_FASTIOV]; + struct iovec *iov_l = iovstack_l; + struct iovec *iov_r = iovstack_r; + ssize_t rc; + + if (flags != 0) + return -EINVAL; + + /* Check iovecs */ + if (vm_write) + rc = rw_copy_check_uvector(WRITE, lvec, liovcnt, UIO_FASTIOV, + iovstack_l, &iov_l, 1); + else + rc = rw_copy_check_uvector(READ, lvec, liovcnt, UIO_FASTIOV, + iovstack_l, &iov_l, 1); + if (rc <= 0) + goto free_iovecs; + + rc = rw_copy_check_uvector(READ, rvec, riovcnt, UIO_FASTIOV, + iovstack_r, &iov_r, 0); + if (rc <= 0) + goto free_iovecs; + + rc = process_vm_rw_core(pid, iov_l, liovcnt, iov_r, riovcnt, flags, + vm_write); + +free_iovecs: + if (iov_r != iovstack_r) + kfree(iov_r); + if (iov_l != iovstack_l) + kfree(iov_l); + + return rc; +} + +SYSCALL_DEFINE6(process_vm_readv, pid_t, pid, const struct iovec __user *, lvec, + unsigned long, liovcnt, const struct iovec __user *, rvec, + unsigned long, riovcnt, unsigned long, flags) +{ + return process_vm_rw(pid, lvec, liovcnt, rvec, riovcnt, flags, 0); +} + +SYSCALL_DEFINE6(process_vm_writev, pid_t, pid, + const struct iovec __user *, lvec, + unsigned long, liovcnt, const struct iovec __user *, rvec, + unsigned long, riovcnt, unsigned long, flags) +{ + return process_vm_rw(pid, lvec, liovcnt, rvec, riovcnt, flags, 1); +} + +#ifdef CONFIG_COMPAT + +asmlinkage ssize_t +compat_process_vm_rw(compat_pid_t pid, + const struct compat_iovec __user *lvec, + unsigned long liovcnt, + const struct compat_iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags, int vm_write) +{ + struct iovec iovstack_l[UIO_FASTIOV]; + struct iovec iovstack_r[UIO_FASTIOV]; + struct iovec *iov_l = iovstack_l; + struct iovec *iov_r = iovstack_r; + ssize_t rc = -EFAULT; + + if (flags != 0) + return -EINVAL; + + if (!access_ok(VERIFY_READ, lvec, liovcnt * sizeof(*lvec))) + goto out; + + if (!access_ok(VERIFY_READ, rvec, riovcnt * sizeof(*rvec))) + goto out; + + if (vm_write) + rc = compat_rw_copy_check_uvector(WRITE, lvec, liovcnt, + UIO_FASTIOV, iovstack_l, + &iov_l, 1); + else + rc = compat_rw_copy_check_uvector(READ, lvec, liovcnt, + UIO_FASTIOV, iovstack_l, + &iov_l, 1); + if (rc <= 0) + goto free_iovecs; + rc = compat_rw_copy_check_uvector(READ, rvec, riovcnt, + UIO_FASTIOV, iovstack_r, + &iov_r, 0); + if (rc <= 0) + goto free_iovecs; + + rc = process_vm_rw_core(pid, iov_l, liovcnt, iov_r, riovcnt, flags, + vm_write); + +free_iovecs: + if (iov_r != iovstack_r) + kfree(iov_r); + if (iov_l != iovstack_l) + kfree(iov_l); + +out: + return rc; +} + +asmlinkage ssize_t +compat_sys_process_vm_readv(compat_pid_t pid, + const struct compat_iovec __user *lvec, + unsigned long liovcnt, + const struct compat_iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags) +{ + return compat_process_vm_rw(pid, lvec, liovcnt, rvec, + riovcnt, flags, 0); +} + +asmlinkage ssize_t +compat_sys_process_vm_writev(compat_pid_t pid, + const struct compat_iovec __user *lvec, + unsigned long liovcnt, + const struct compat_iovec __user *rvec, + unsigned long riovcnt, + unsigned long flags) +{ + return compat_process_vm_rw(pid, lvec, liovcnt, rvec, + riovcnt, flags, 1); +} + +#endif diff --git a/security/keys/compat.c b/security/keys/compat.c index 338b510e9027..4c48e13448f8 100644 --- a/security/keys/compat.c +++ b/security/keys/compat.c @@ -38,7 +38,7 @@ long compat_keyctl_instantiate_key_iov( ret = compat_rw_copy_check_uvector(WRITE, _payload_iov, ioc, ARRAY_SIZE(iovstack), - iovstack, &iov); + iovstack, &iov, 1); if (ret < 0) return ret; if (ret == 0) diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index eca51918c951..0b3f5d72af1c 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1065,7 +1065,7 @@ long keyctl_instantiate_key_iov(key_serial_t id, goto no_payload; ret = rw_copy_check_uvector(WRITE, _payload_iov, ioc, - ARRAY_SIZE(iovstack), iovstack, &iov); + ARRAY_SIZE(iovstack), iovstack, &iov, 1); if (ret < 0) return ret; if (ret == 0) -- cgit v1.3-8-gc7d7 From c9f01245b6a7d77d17deaa71af10f6aca14fa24e Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 31 Oct 2011 17:07:15 -0700 Subject: oom: remove oom_disable_count This removes mm->oom_disable_count entirely since it's unnecessary and currently buggy. The counter was intended to be per-process but it's currently decremented in the exit path for each thread that exits, causing it to underflow. The count was originally intended to prevent oom killing threads that share memory with threads that cannot be killed since it doesn't lead to future memory freeing. The counter could be fixed to represent all threads sharing the same mm, but it's better to remove the count since: - it is possible that the OOM_DISABLE thread sharing memory with the victim is waiting on that thread to exit and will actually cause future memory freeing, and - there is no guarantee that a thread is disabled from oom killing just because another thread sharing its mm is oom disabled. Signed-off-by: David Rientjes Reported-by: Oleg Nesterov Reviewed-by: Oleg Nesterov Cc: Ying Han Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 4 ---- fs/proc/base.c | 13 ------------- include/linux/mm_types.h | 3 --- kernel/exit.c | 2 -- kernel/fork.c | 10 +--------- mm/oom_kill.c | 23 +++++------------------ 6 files changed, 6 insertions(+), 49 deletions(-) (limited to 'kernel') diff --git a/fs/exec.c b/fs/exec.c index 25dcbe5fc356..36254645b7cc 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -841,10 +841,6 @@ static int exec_mmap(struct mm_struct *mm) tsk->mm = mm; tsk->active_mm = mm; activate_mm(active_mm, mm); - if (old_mm && tsk->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) { - atomic_dec(&old_mm->oom_disable_count); - atomic_inc(&tsk->mm->oom_disable_count); - } task_unlock(tsk); arch_pick_mmap_layout(mm); if (old_mm) { diff --git a/fs/proc/base.c b/fs/proc/base.c index 5eb02069e1b8..8f0087e20e16 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1107,13 +1107,6 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, goto err_sighand; } - if (oom_adjust != task->signal->oom_adj) { - if (oom_adjust == OOM_DISABLE) - atomic_inc(&task->mm->oom_disable_count); - if (task->signal->oom_adj == OOM_DISABLE) - atomic_dec(&task->mm->oom_disable_count); - } - /* * Warn that /proc/pid/oom_adj is deprecated, see * Documentation/feature-removal-schedule.txt. @@ -1215,12 +1208,6 @@ static ssize_t oom_score_adj_write(struct file *file, const char __user *buf, goto err_sighand; } - if (oom_score_adj != task->signal->oom_score_adj) { - if (oom_score_adj == OOM_SCORE_ADJ_MIN) - atomic_inc(&task->mm->oom_disable_count); - if (task->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) - atomic_dec(&task->mm->oom_disable_count); - } task->signal->oom_score_adj = oom_score_adj; if (has_capability_noaudit(current, CAP_SYS_RESOURCE)) task->signal->oom_score_adj_min = oom_score_adj; diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index c93d00a6e95d..6456624aa964 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -336,9 +336,6 @@ struct mm_struct { unsigned int token_priority; unsigned int last_interval; - /* How many tasks sharing this mm are OOM_DISABLE */ - atomic_t oom_disable_count; - unsigned long flags; /* Must use atomic bitops to access the bits */ struct core_state *core_state; /* coredumping support */ diff --git a/kernel/exit.c b/kernel/exit.c index 2913b3509d42..d0b7d988f873 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -681,8 +681,6 @@ static void exit_mm(struct task_struct * tsk) enter_lazy_tlb(mm, current); /* We don't want this task to be frozen prematurely */ clear_freeze_flag(tsk); - if (tsk->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) - atomic_dec(&mm->oom_disable_count); task_unlock(tsk); mm_update_next_owner(mm); mmput(mm); diff --git a/kernel/fork.c b/kernel/fork.c index 8e6b6f4fb272..70d76191afb9 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -501,7 +501,6 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p) mm->cached_hole_size = ~0UL; mm_init_aio(mm); mm_init_owner(mm, p); - atomic_set(&mm->oom_disable_count, 0); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; @@ -816,8 +815,6 @@ good_mm: /* Initializing for Swap token stuff */ mm->token_priority = 0; mm->last_interval = 0; - if (tsk->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) - atomic_inc(&mm->oom_disable_count); tsk->mm = mm; tsk->active_mm = mm; @@ -1391,13 +1388,8 @@ bad_fork_cleanup_io: bad_fork_cleanup_namespaces: exit_task_namespaces(p); bad_fork_cleanup_mm: - if (p->mm) { - task_lock(p); - if (p->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) - atomic_dec(&p->mm->oom_disable_count); - task_unlock(p); + if (p->mm) mmput(p->mm); - } bad_fork_cleanup_signal: if (!(clone_flags & CLONE_THREAD)) free_signal_struct(p->signal); diff --git a/mm/oom_kill.c b/mm/oom_kill.c index b0d8943bc9fd..2b97e8f04607 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -54,13 +54,7 @@ int test_set_oom_score_adj(int new_val) spin_lock_irq(&sighand->siglock); old_val = current->signal->oom_score_adj; - if (new_val != old_val) { - if (new_val == OOM_SCORE_ADJ_MIN) - atomic_inc(¤t->mm->oom_disable_count); - else if (old_val == OOM_SCORE_ADJ_MIN) - atomic_dec(¤t->mm->oom_disable_count); - current->signal->oom_score_adj = new_val; - } + current->signal->oom_score_adj = new_val; spin_unlock_irq(&sighand->siglock); return old_val; @@ -172,16 +166,6 @@ unsigned int oom_badness(struct task_struct *p, struct mem_cgroup *mem, if (!p) return 0; - /* - * Shortcut check for a thread sharing p->mm that is OOM_SCORE_ADJ_MIN - * so the entire heuristic doesn't need to be executed for something - * that cannot be killed. - */ - if (atomic_read(&p->mm->oom_disable_count)) { - task_unlock(p); - return 0; - } - /* * The memory controller may have a limit of 0 bytes, so avoid a divide * by zero, if necessary. @@ -451,6 +435,9 @@ static int oom_kill_task(struct task_struct *p, struct mem_cgroup *mem) for_each_process(q) if (q->mm == mm && !same_thread_group(q, p) && !(q->flags & PF_KTHREAD)) { + if (q->signal->oom_score_adj == OOM_SCORE_ADJ_MIN) + continue; + task_lock(q); /* Protect ->comm from prctl() */ pr_err("Kill process %d (%s) sharing same memory\n", task_pid_nr(q), q->comm); @@ -727,7 +714,7 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, read_lock(&tasklist_lock); if (sysctl_oom_kill_allocating_task && !oom_unkillable_task(current, NULL, nodemask) && - current->mm && !atomic_read(¤t->mm->oom_disable_count)) { + current->mm) { /* * oom_kill_process() needs tasklist_lock held. If it returns * non-zero, current could not be killed so we must fallback to -- cgit v1.3-8-gc7d7 From bc3e53f682d93df677dbd5006a404722b3adfe18 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 31 Oct 2011 17:07:30 -0700 Subject: mm: distinguish between mlocked and pinned pages Some kernel components pin user space memory (infiniband and perf) (by increasing the page count) and account that memory as "mlocked". The difference between mlocking and pinning is: A. mlocked pages are marked with PG_mlocked and are exempt from swapping. Page migration may move them around though. They are kept on a special LRU list. B. Pinned pages cannot be moved because something needs to directly access physical memory. They may not be on any LRU list. I recently saw an mlockalled process where mm->locked_vm became bigger than the virtual size of the process (!) because some memory was accounted for twice: Once when the page was mlocked and once when the Infiniband layer increased the refcount because it needt to pin the RDMA memory. This patch introduces a separate counter for pinned pages and accounts them seperately. Signed-off-by: Christoph Lameter Cc: Mike Marciniszyn Cc: Roland Dreier Cc: Sean Hefty Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/infiniband/core/umem.c | 6 +++--- drivers/infiniband/hw/ipath/ipath_user_pages.c | 6 +++--- drivers/infiniband/hw/qib/qib_user_pages.c | 4 ++-- fs/proc/task_mmu.c | 2 ++ include/linux/mm_types.h | 2 +- kernel/events/core.c | 6 +++--- 6 files changed, 14 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index b645e558876f..9155f91d66bf 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -136,7 +136,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, down_write(¤t->mm->mmap_sem); - locked = npages + current->mm->locked_vm; + locked = npages + current->mm->pinned_vm; lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) { @@ -206,7 +206,7 @@ out: __ib_umem_release(context->device, umem, 0); kfree(umem); } else - current->mm->locked_vm = locked; + current->mm->pinned_vm = locked; up_write(¤t->mm->mmap_sem); if (vma_list) @@ -222,7 +222,7 @@ static void ib_umem_account(struct work_struct *work) struct ib_umem *umem = container_of(work, struct ib_umem, work); down_write(&umem->mm->mmap_sem); - umem->mm->locked_vm -= umem->diff; + umem->mm->pinned_vm -= umem->diff; up_write(&umem->mm->mmap_sem); mmput(umem->mm); kfree(umem); diff --git a/drivers/infiniband/hw/ipath/ipath_user_pages.c b/drivers/infiniband/hw/ipath/ipath_user_pages.c index cfed5399f074..dc66c4506916 100644 --- a/drivers/infiniband/hw/ipath/ipath_user_pages.c +++ b/drivers/infiniband/hw/ipath/ipath_user_pages.c @@ -79,7 +79,7 @@ static int __ipath_get_user_pages(unsigned long start_page, size_t num_pages, goto bail_release; } - current->mm->locked_vm += num_pages; + current->mm->pinned_vm += num_pages; ret = 0; goto bail; @@ -178,7 +178,7 @@ void ipath_release_user_pages(struct page **p, size_t num_pages) __ipath_release_user_pages(p, num_pages, 1); - current->mm->locked_vm -= num_pages; + current->mm->pinned_vm -= num_pages; up_write(¤t->mm->mmap_sem); } @@ -195,7 +195,7 @@ static void user_pages_account(struct work_struct *_work) container_of(_work, struct ipath_user_pages_work, work); down_write(&work->mm->mmap_sem); - work->mm->locked_vm -= work->num_pages; + work->mm->pinned_vm -= work->num_pages; up_write(&work->mm->mmap_sem); mmput(work->mm); kfree(work); diff --git a/drivers/infiniband/hw/qib/qib_user_pages.c b/drivers/infiniband/hw/qib/qib_user_pages.c index 7689e49c13c9..2bc1d2b96298 100644 --- a/drivers/infiniband/hw/qib/qib_user_pages.c +++ b/drivers/infiniband/hw/qib/qib_user_pages.c @@ -74,7 +74,7 @@ static int __qib_get_user_pages(unsigned long start_page, size_t num_pages, goto bail_release; } - current->mm->locked_vm += num_pages; + current->mm->pinned_vm += num_pages; ret = 0; goto bail; @@ -151,7 +151,7 @@ void qib_release_user_pages(struct page **p, size_t num_pages) __qib_release_user_pages(p, num_pages, 1); if (current->mm) { - current->mm->locked_vm -= num_pages; + current->mm->pinned_vm -= num_pages; up_write(¤t->mm->mmap_sem); } } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index c7d4ee663f14..e418c5abdb0e 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -44,6 +44,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) "VmPeak:\t%8lu kB\n" "VmSize:\t%8lu kB\n" "VmLck:\t%8lu kB\n" + "VmPin:\t%8lu kB\n" "VmHWM:\t%8lu kB\n" "VmRSS:\t%8lu kB\n" "VmData:\t%8lu kB\n" @@ -55,6 +56,7 @@ void task_mem(struct seq_file *m, struct mm_struct *mm) hiwater_vm << (PAGE_SHIFT-10), (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), mm->locked_vm << (PAGE_SHIFT-10), + mm->pinned_vm << (PAGE_SHIFT-10), hiwater_rss << (PAGE_SHIFT-10), total_rss << (PAGE_SHIFT-10), data << (PAGE_SHIFT-10), diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 6456624aa964..f3175830cc73 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -304,7 +304,7 @@ struct mm_struct { unsigned long hiwater_rss; /* High-watermark of RSS usage */ unsigned long hiwater_vm; /* High-water virtual memory usage */ - unsigned long total_vm, locked_vm, shared_vm, exec_vm; + unsigned long total_vm, locked_vm, pinned_vm, shared_vm, exec_vm; unsigned long stack_vm, reserved_vm, def_flags, nr_ptes; unsigned long start_code, end_code, start_data, end_data; unsigned long start_brk, brk, start_stack; diff --git a/kernel/events/core.c b/kernel/events/core.c index d1a1bee35228..12a0287e0358 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -3544,7 +3544,7 @@ static void perf_mmap_close(struct vm_area_struct *vma) struct ring_buffer *rb = event->rb; atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm); - vma->vm_mm->locked_vm -= event->mmap_locked; + vma->vm_mm->pinned_vm -= event->mmap_locked; rcu_assign_pointer(event->rb, NULL); mutex_unlock(&event->mmap_mutex); @@ -3625,7 +3625,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) lock_limit = rlimit(RLIMIT_MEMLOCK); lock_limit >>= PAGE_SHIFT; - locked = vma->vm_mm->locked_vm + extra; + locked = vma->vm_mm->pinned_vm + extra; if ((locked > lock_limit) && perf_paranoid_tracepoint_raw() && !capable(CAP_IPC_LOCK)) { @@ -3651,7 +3651,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) atomic_long_add(user_extra, &user->locked_vm); event->mmap_locked = extra; event->mmap_user = get_current_user(); - vma->vm_mm->locked_vm += event->mmap_locked; + vma->vm_mm->pinned_vm += event->mmap_locked; unlock: if (!ret) -- cgit v1.3-8-gc7d7 From f445027e4e692bd885118460b292d08027fd5501 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 31 Oct 2011 17:11:15 -0700 Subject: stop_machine: make stop_machine safe and efficient to call early Make stop_machine() safe to call early in boot, before SMP has been set up, by simply calling the callback function directly if there's only one CPU online. [ Fixes from AKPM: - add comment - local_irq_flags, not save_flags - also call hard_irq_disable() for systems which need it Tejun suggested using an explicit flag rather than just looking at the online cpu count. ] Cc: Tejun Heo Acked-by: Rusty Russell Cc: Peter Zijlstra Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Steven Rostedt Acked-by: Tejun Heo Cc: Konrad Rzeszutek Wilk Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/stop_machine.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'kernel') diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index ba5070ce5765..5b0951aa0496 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -41,6 +41,7 @@ struct cpu_stopper { }; static DEFINE_PER_CPU(struct cpu_stopper, cpu_stopper); +static bool stop_machine_initialized = false; static void cpu_stop_init_done(struct cpu_stop_done *done, unsigned int nr_todo) { @@ -386,6 +387,8 @@ static int __init cpu_stop_init(void) cpu_stop_cpu_callback(&cpu_stop_cpu_notifier, CPU_ONLINE, bcpu); register_cpu_notifier(&cpu_stop_cpu_notifier); + stop_machine_initialized = true; + return 0; } early_initcall(cpu_stop_init); @@ -485,6 +488,25 @@ int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus) .num_threads = num_online_cpus(), .active_cpus = cpus }; + if (!stop_machine_initialized) { + /* + * Handle the case where stop_machine() is called + * early in boot before stop_machine() has been + * initialized. + */ + unsigned long flags; + int ret; + + WARN_ON_ONCE(smdata.num_threads != 1); + + local_irq_save(flags); + hard_irq_disable(); + ret = (*fn)(data); + local_irq_restore(flags); + + return ret; + } + /* Set the initial state and stop all online cpus. */ set_state(&smdata, STOPMACHINE_PREPARE); return stop_cpus(cpu_online_mask, stop_machine_cpu_stop, &smdata); -- cgit v1.3-8-gc7d7 From 4ff819515b203f937cc6c8a0215a37a68d1ee71f Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Mon, 31 Oct 2011 17:11:18 -0700 Subject: watchdog: move watchdog_*_all_cpus under CONFIG_SYSCTL Fix compilation warnings for CONFIG_SYSCTL=n: fixed compilation warnings in case of disabled CONFIG_SYSCTL kernel/watchdog.c:483:13: warning: `watchdog_enable_all_cpus' defined but not used kernel/watchdog.c:500:13: warning: `watchdog_disable_all_cpus' defined but not used these functions are static and are used only in sysctl handler, so move them inside #ifdef CONFIG_SYSCTL too Signed-off-by: Vasily Averin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/watchdog.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/watchdog.c b/kernel/watchdog.c index d680381b0e9c..1d7bca7f4f52 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -481,6 +481,8 @@ static void watchdog_disable(int cpu) } } +/* sysctl functions */ +#ifdef CONFIG_SYSCTL static void watchdog_enable_all_cpus(void) { int cpu; @@ -510,8 +512,6 @@ static void watchdog_disable_all_cpus(void) } -/* sysctl functions */ -#ifdef CONFIG_SYSCTL /* * proc handler for /proc/sys/kernel/nmi_watchdog,watchdog_thresh */ -- cgit v1.3-8-gc7d7 From 73efc0394e148d0e15583e13712637831f926720 Mon Sep 17 00:00:00 2001 From: Dan Ballard Date: Mon, 31 Oct 2011 17:11:20 -0700 Subject: kernel/sysctl.c: add cap_last_cap to /proc/sys/kernel Userspace needs to know the highest valid capability of the running kernel, which right now cannot reliably be retrieved from the header files only. The fact that this value cannot be determined properly right now creates various problems for libraries compiled on newer header files which are run on older kernels. They assume capabilities are available which actually aren't. libcap-ng is one example. And we ran into the same problem with systemd too. Now the capability is exported in /proc/sys/kernel/cap_last_cap. [akpm@linux-foundation.org: make cap_last_cap const, per Ulrich] Signed-off-by: Dan Ballard Cc: Randy Dunlap Cc: Ingo Molnar Cc: Lennart Poettering Cc: Kay Sievers Cc: Ulrich Drepper Cc: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sysctl/kernel.txt | 8 ++++++++ kernel/sysctl.c | 9 +++++++++ 2 files changed, 17 insertions(+) (limited to 'kernel') diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt index 704e474a93df..1f2463671a1a 100644 --- a/Documentation/sysctl/kernel.txt +++ b/Documentation/sysctl/kernel.txt @@ -24,6 +24,7 @@ show up in /proc/sys/kernel: - bootloader_type [ X86 only ] - bootloader_version [ X86 only ] - callhome [ S390 only ] +- cap_last_cap - core_pattern - core_pipe_limit - core_uses_pid @@ -155,6 +156,13 @@ on has a service contract with IBM. ============================================================== +cap_last_cap + +Highest valid capability of the running kernel. Exports +CAP_LAST_CAP from the kernel. + +============================================================== + core_pattern: core_pattern is used to specify a core dumpfile pattern name. diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 2d2ecdcc8cdb..c49d66658ec0 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include @@ -134,6 +135,7 @@ static int minolduid; static int min_percpu_pagelist_fract = 8; static int ngroups_max = NGROUPS_MAX; +static const int cap_last_cap = CAP_LAST_CAP; #ifdef CONFIG_INOTIFY_USER #include @@ -740,6 +742,13 @@ static struct ctl_table kern_table[] = { .mode = 0444, .proc_handler = proc_dointvec, }, + { + .procname = "cap_last_cap", + .data = (void *)&cap_last_cap, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = proc_dointvec, + }, #if defined(CONFIG_LOCKUP_DETECTOR) { .procname = "watchdog", -- cgit v1.3-8-gc7d7 From 0eca6b7c78fd997e02bd9850e608102382b7822e Mon Sep 17 00:00:00 2001 From: Yanmin Zhang Date: Mon, 31 Oct 2011 17:11:25 -0700 Subject: printk: add module parameter ignore_loglevel to control ignore_loglevel We are enabling some power features on medfield. To test suspend-2-RAM conveniently, we need turn on/off ignore_loglevel frequently without rebooting. Add a module parameter, so users can change it by: /sys/module/printk/parameters/ignore_loglevel Signed-off-by: Yanmin Zhang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 3 +++ kernel/printk.c | 3 +++ 2 files changed, 6 insertions(+) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 27e0488d54d2..106efe1e2bab 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -973,6 +973,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted. ignore_loglevel [KNL] Ignore loglevel setting - this will print /all/ kernel messages to the console. Useful for debugging. + We also add it as printk module parameter, so users + could change it dynamically, usually by + /sys/module/printk/parameters/ignore_loglevel. ihash_entries= [KNL] Set number of hash buckets for inode cache. diff --git a/kernel/printk.c b/kernel/printk.c index b7da18391c38..e62f949ec140 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -532,6 +532,9 @@ static int __init ignore_loglevel_setup(char *str) } early_param("ignore_loglevel", ignore_loglevel_setup); +module_param_named(ignore_loglevel, ignore_loglevel, bool, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(ignore_loglevel, "ignore loglevel setting, to" + "print all kernel messages to the console."); /* * Write out chars from start to end - 1 inclusive -- cgit v1.3-8-gc7d7 From 134620f7a865b3bc9e3d56d460603592b70ede21 Mon Sep 17 00:00:00 2001 From: Yanmin Zhang Date: Mon, 31 Oct 2011 17:11:27 -0700 Subject: printk: add console_suspend module parameter We are enabling some power features on medfield. To test suspend-2-RAM conveniently, we need turn on/off console_suspend_enabled frequently. Add a module parameter, so users could change it by: /sys/module/printk/parameters/console_suspend Signed-off-by: Yanmin Zhang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 5 +++++ kernel/printk.c | 4 ++++ 2 files changed, 9 insertions(+) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 106efe1e2bab..c2efe285396c 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1669,6 +1669,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. debugging driver suspend/resume hooks). This may not work reliably with all consoles, but is known to work with serial and VGA consoles. + To facilitate more flexible debugging, we also add + console_suspend, a printk module parameter to control + it. Users could use console_suspend (usually + /sys/module/printk/parameters/console_suspend) to + turn on/off it dynamically. noaliencache [MM, NUMA, SLAB] Disables the allocation of alien caches in the slab allocator. Saves per-node memory, diff --git a/kernel/printk.c b/kernel/printk.c index e62f949ec140..6d9dedd11450 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1111,6 +1111,10 @@ static int __init console_suspend_disable(char *str) return 1; } __setup("no_console_suspend", console_suspend_disable); +module_param_named(console_suspend, console_suspend_enabled, + bool, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(console_suspend, "suspend console during suspend" + " and hibernate operations"); /** * suspend_console - suspend the console subsystem -- cgit v1.3-8-gc7d7 From 48e41899e4a3592746e5263c14681bf5c1393563 Mon Sep 17 00:00:00 2001 From: William Douglas Date: Mon, 31 Oct 2011 17:11:29 -0700 Subject: printk: fix bounds checking for log_prefix Currently log_prefix is testing that the first character of the log level and facility is less than '0' and greater than '9' (which is always false). It should be testing to see if the character less than '0' or greater than '9' instead. This patch makes that change. The code being changed worked because strtoul bombs out (endp isn't updated) and 0 is returned anyway. Signed-off-by: William Douglas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 6d9dedd11450..286d2c7be52c 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -595,7 +595,7 @@ static size_t log_prefix(const char *p, unsigned int *level, char *special) /* multi digit including the level and facility number */ char *endp = NULL; - if (p[1] < '0' && p[1] > '9') + if (p[1] < '0' || p[1] > '9') return 0; lev = (simple_strtoul(&p[1], &endp, 10) & 7); -- cgit v1.3-8-gc7d7 From ae29bc92da01a2e9d278a9a58c3b307d41cc0254 Mon Sep 17 00:00:00 2001 From: William Douglas Date: Mon, 31 Oct 2011 17:11:31 -0700 Subject: printk: remove bounds checking for log_prefix Currently log_prefix is testing that the first character of the log level and facility is less than '0' and greater than '9' (which is always false). Since the code being updated works because strtoul bombs out (endp isn't updated) and 0 is returned anyway just remove the check and don't change the behavior of the function. Signed-off-by: William Douglas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 286d2c7be52c..1455a0d4eedd 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -595,9 +595,6 @@ static size_t log_prefix(const char *p, unsigned int *level, char *special) /* multi digit including the level and facility number */ char *endp = NULL; - if (p[1] < '0' || p[1] > '9') - return 0; - lev = (simple_strtoul(&p[1], &endp, 10) & 7); if (endp == NULL || endp[0] != '>') return 0; -- cgit v1.3-8-gc7d7 From 50e1499f468fd74c6db95deb2e1e6bfee578ae70 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 31 Oct 2011 17:12:51 -0700 Subject: kgdb: follow rename pack_hex_byte() to hex_byte_pack() There is no functional change. Signed-off-by: Andy Shevchenko Acked-by: Jesper Nilsson Cc: David Howells Cc: Koichi Yasutake Cc: Jason Wessel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/arch-v10/kernel/kgdb.c | 6 ++--- arch/cris/arch-v32/kernel/kgdb.c | 14 +++++------ arch/frv/kernel/gdb-stub.c | 44 +++++++++++++++++----------------- arch/mn10300/kernel/gdb-stub.c | 52 ++++++++++++++++++++-------------------- kernel/debug/gdbstub.c | 12 +++++----- 5 files changed, 64 insertions(+), 64 deletions(-) (limited to 'kernel') diff --git a/arch/cris/arch-v10/kernel/kgdb.c b/arch/cris/arch-v10/kernel/kgdb.c index b9f9c8ce2169..b579dd02e098 100644 --- a/arch/cris/arch-v10/kernel/kgdb.c +++ b/arch/cris/arch-v10/kernel/kgdb.c @@ -694,7 +694,7 @@ mem2hex(char *buf, unsigned char *mem, int count) /* Valid mem address. */ for (i = 0; i < count; i++) { ch = *mem++; - buf = pack_hex_byte(buf, ch); + buf = hex_byte_pack(buf, ch); } } @@ -868,7 +868,7 @@ stub_is_stopped(int sigval) /* Send trap type (converted to signal) */ *ptr++ = 'T'; - ptr = pack_hex_byte(ptr, sigval); + ptr = hex_byte_pack(ptr, sigval); /* Send register contents. We probably only need to send the * PC, frame pointer and stack pointer here. Other registers will be @@ -881,7 +881,7 @@ stub_is_stopped(int sigval) status = read_register (regno, ®_cont); if (status == SUCCESS) { - ptr = pack_hex_byte(ptr, regno); + ptr = hex_byte_pack(ptr, regno); *ptr++ = ':'; ptr = mem2hex(ptr, (unsigned char *)®_cont, diff --git a/arch/cris/arch-v32/kernel/kgdb.c b/arch/cris/arch-v32/kernel/kgdb.c index c0343c3ea7f8..8c1d35cdf00a 100644 --- a/arch/cris/arch-v32/kernel/kgdb.c +++ b/arch/cris/arch-v32/kernel/kgdb.c @@ -677,7 +677,7 @@ mem2hex(char *buf, unsigned char *mem, int count) /* Valid mem address. */ for (i = 0; i < count; i++) { ch = *mem++; - buf = pack_hex_byte(buf, ch); + buf = hex_byte_pack(buf, ch); } } /* Terminate properly. */ @@ -695,7 +695,7 @@ mem2hex_nbo(char *buf, unsigned char *mem, int count) mem += count - 1; for (i = 0; i < count; i++) { ch = *mem--; - buf = pack_hex_byte(buf, ch); + buf = hex_byte_pack(buf, ch); } /* Terminate properly. */ @@ -880,7 +880,7 @@ stub_is_stopped(int sigval) /* Send trap type (converted to signal) */ *ptr++ = 'T'; - ptr = pack_hex_byte(ptr, sigval); + ptr = hex_byte_pack(ptr, sigval); if (((reg.exs & 0xff00) >> 8) == 0xc) { @@ -988,26 +988,26 @@ stub_is_stopped(int sigval) } /* Only send PC, frame and stack pointer. */ read_register(PC, ®_cont); - ptr = pack_hex_byte(ptr, PC); + ptr = hex_byte_pack(ptr, PC); *ptr++ = ':'; ptr = mem2hex(ptr, (unsigned char *)®_cont, register_size[PC]); *ptr++ = ';'; read_register(R8, ®_cont); - ptr = pack_hex_byte(ptr, R8); + ptr = hex_byte_pack(ptr, R8); *ptr++ = ':'; ptr = mem2hex(ptr, (unsigned char *)®_cont, register_size[R8]); *ptr++ = ';'; read_register(SP, ®_cont); - ptr = pack_hex_byte(ptr, SP); + ptr = hex_byte_pack(ptr, SP); *ptr++ = ':'; ptr = mem2hex(ptr, (unsigned char *)®_cont, register_size[SP]); *ptr++ = ';'; /* Send ERP as well; this will save us an entire register fetch in some cases. */ read_register(ERP, ®_cont); - ptr = pack_hex_byte(ptr, ERP); + ptr = hex_byte_pack(ptr, ERP); *ptr++ = ':'; ptr = mem2hex(ptr, (unsigned char *)®_cont, register_size[ERP]); *ptr++ = ';'; diff --git a/arch/frv/kernel/gdb-stub.c b/arch/frv/kernel/gdb-stub.c index a4dba6b20bd0..a6d5381c94fe 100644 --- a/arch/frv/kernel/gdb-stub.c +++ b/arch/frv/kernel/gdb-stub.c @@ -672,7 +672,7 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa if ((uint32_t)mem&1 && count>=1) { if (!gdbstub_read_byte(mem,ch)) return NULL; - buf = pack_hex_byte(buf, ch[0]); + buf = hex_byte_pack(buf, ch[0]); mem++; count--; } @@ -680,8 +680,8 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa if ((uint32_t)mem&3 && count>=2) { if (!gdbstub_read_word(mem,(uint16_t *)ch)) return NULL; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); mem += 2; count -= 2; } @@ -689,10 +689,10 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa while (count>=4) { if (!gdbstub_read_dword(mem,(uint32_t *)ch)) return NULL; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); - buf = pack_hex_byte(buf, ch[2]); - buf = pack_hex_byte(buf, ch[3]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); + buf = hex_byte_pack(buf, ch[2]); + buf = hex_byte_pack(buf, ch[3]); mem += 4; count -= 4; } @@ -700,8 +700,8 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa if (count>=2) { if (!gdbstub_read_word(mem,(uint16_t *)ch)) return NULL; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); mem += 2; count -= 2; } @@ -709,7 +709,7 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa if (count>=1) { if (!gdbstub_read_byte(mem,ch)) return NULL; - buf = pack_hex_byte(buf, ch[0]); + buf = hex_byte_pack(buf, ch[0]); } *buf = 0; @@ -1498,21 +1498,21 @@ void gdbstub(int sigval) ptr = mem2hex(title, ptr, sizeof(title) - 1,0); hx = hex_asc_hi(brr >> 24); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(brr >> 24); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(brr >> 16); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(brr >> 16); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(brr >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(brr >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(brr); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(brr); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0); *ptr = 0; @@ -1526,10 +1526,10 @@ void gdbstub(int sigval) /* Send trap type (converted to signal) */ *ptr++ = 'T'; - ptr = pack_hex_byte(ptr, sigval); + ptr = hex_byte_pack(ptr, sigval); /* Send Error PC */ - ptr = pack_hex_byte(ptr, GDB_REG_PC); + ptr = hex_byte_pack(ptr, GDB_REG_PC); *ptr++ = ':'; ptr = mem2hex(&__debug_frame->pc, ptr, 4, 0); *ptr++ = ';'; @@ -1537,7 +1537,7 @@ void gdbstub(int sigval) /* * Send frame pointer */ - ptr = pack_hex_byte(ptr, GDB_REG_FP); + ptr = hex_byte_pack(ptr, GDB_REG_FP); *ptr++ = ':'; ptr = mem2hex(&__debug_frame->fp, ptr, 4, 0); *ptr++ = ';'; @@ -1545,7 +1545,7 @@ void gdbstub(int sigval) /* * Send stack pointer */ - ptr = pack_hex_byte(ptr, GDB_REG_SP); + ptr = hex_byte_pack(ptr, GDB_REG_SP); *ptr++ = ':'; ptr = mem2hex(&__debug_frame->sp, ptr, 4, 0); *ptr++ = ';'; diff --git a/arch/mn10300/kernel/gdb-stub.c b/arch/mn10300/kernel/gdb-stub.c index 538266b2c9bc..522eb8a9b60d 100644 --- a/arch/mn10300/kernel/gdb-stub.c +++ b/arch/mn10300/kernel/gdb-stub.c @@ -798,7 +798,7 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault) if ((u32) mem & 1 && count >= 1) { if (gdbstub_read_byte(mem, ch) != 0) return 0; - buf = pack_hex_byte(buf, ch[0]); + buf = hex_byte_pack(buf, ch[0]); mem++; count--; } @@ -806,8 +806,8 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault) if ((u32) mem & 3 && count >= 2) { if (gdbstub_read_word(mem, ch) != 0) return 0; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); mem += 2; count -= 2; } @@ -815,10 +815,10 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault) while (count >= 4) { if (gdbstub_read_dword(mem, ch) != 0) return 0; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); - buf = pack_hex_byte(buf, ch[2]); - buf = pack_hex_byte(buf, ch[3]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); + buf = hex_byte_pack(buf, ch[2]); + buf = hex_byte_pack(buf, ch[3]); mem += 4; count -= 4; } @@ -826,8 +826,8 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault) if (count >= 2) { if (gdbstub_read_word(mem, ch) != 0) return 0; - buf = pack_hex_byte(buf, ch[0]); - buf = pack_hex_byte(buf, ch[1]); + buf = hex_byte_pack(buf, ch[0]); + buf = hex_byte_pack(buf, ch[1]); mem += 2; count -= 2; } @@ -835,7 +835,7 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault) if (count >= 1) { if (gdbstub_read_byte(mem, ch) != 0) return 0; - buf = pack_hex_byte(buf, ch[0]); + buf = hex_byte_pack(buf, ch[0]); } *buf = 0; @@ -1273,13 +1273,13 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep) ptr = mem2hex(title, ptr, sizeof(title) - 1, 0); hx = hex_asc_hi(excep >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(excep >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(excep); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(excep); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0); *ptr = 0; @@ -1291,21 +1291,21 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep) ptr = mem2hex(tbcberr, ptr, sizeof(tbcberr) - 1, 0); hx = hex_asc_hi(bcberr >> 24); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(bcberr >> 24); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(bcberr >> 16); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(bcberr >> 16); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(bcberr >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(bcberr >> 8); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_hi(bcberr); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); hx = hex_asc_lo(bcberr); - ptr = pack_hex_byte(ptr, hx); + ptr = hex_byte_pack(ptr, hx); ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0); *ptr = 0; @@ -1321,12 +1321,12 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep) * Send trap type (converted to signal) */ *ptr++ = 'T'; - ptr = pack_hex_byte(ptr, sigval); + ptr = hex_byte_pack(ptr, sigval); /* * Send Error PC */ - ptr = pack_hex_byte(ptr, GDB_REGID_PC); + ptr = hex_byte_pack(ptr, GDB_REGID_PC); *ptr++ = ':'; ptr = mem2hex(®s->pc, ptr, 4, 0); *ptr++ = ';'; @@ -1334,7 +1334,7 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep) /* * Send frame pointer */ - ptr = pack_hex_byte(ptr, GDB_REGID_FP); + ptr = hex_byte_pack(ptr, GDB_REGID_FP); *ptr++ = ':'; ptr = mem2hex(®s->a3, ptr, 4, 0); *ptr++ = ';'; @@ -1343,7 +1343,7 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep) * Send stack pointer */ ssp = (unsigned long) (regs + 1); - ptr = pack_hex_byte(ptr, GDB_REGID_SP); + ptr = hex_byte_pack(ptr, GDB_REGID_SP); *ptr++ = ':'; ptr = mem2hex(&ssp, ptr, 4, 0); *ptr++ = ';'; diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c index 34872482315e..c22d8c28ad84 100644 --- a/kernel/debug/gdbstub.c +++ b/kernel/debug/gdbstub.c @@ -217,7 +217,7 @@ void gdbstub_msg_write(const char *s, int len) /* Pack in hex chars */ for (i = 0; i < wcount; i++) - bufptr = pack_hex_byte(bufptr, s[i]); + bufptr = hex_byte_pack(bufptr, s[i]); *bufptr = '\0'; /* Move up */ @@ -249,7 +249,7 @@ char *kgdb_mem2hex(char *mem, char *buf, int count) if (err) return NULL; while (count > 0) { - buf = pack_hex_byte(buf, *tmp); + buf = hex_byte_pack(buf, *tmp); tmp++; count--; } @@ -411,14 +411,14 @@ static char *pack_threadid(char *pkt, unsigned char *id) limit = id + (BUF_THREAD_ID_SIZE / 2); while (id < limit) { if (!lzero || *id != 0) { - pkt = pack_hex_byte(pkt, *id); + pkt = hex_byte_pack(pkt, *id); lzero = 0; } id++; } if (lzero) - pkt = pack_hex_byte(pkt, 0); + pkt = hex_byte_pack(pkt, 0); return pkt; } @@ -486,7 +486,7 @@ static void gdb_cmd_status(struct kgdb_state *ks) dbg_remove_all_break(); remcom_out_buffer[0] = 'S'; - pack_hex_byte(&remcom_out_buffer[1], ks->signo); + hex_byte_pack(&remcom_out_buffer[1], ks->signo); } static void gdb_get_regs_helper(struct kgdb_state *ks) @@ -954,7 +954,7 @@ int gdb_serial_stub(struct kgdb_state *ks) /* Reply to host that an exception has occurred */ ptr = remcom_out_buffer; *ptr++ = 'T'; - ptr = pack_hex_byte(ptr, ks->signo); + ptr = hex_byte_pack(ptr, ks->signo); ptr += strlen(strcpy(ptr, "thread:")); int_to_threadref(thref, shadow_pid(current->pid)); ptr = pack_threadid(ptr, thref); -- cgit v1.3-8-gc7d7 From 33ef6b6984403a688189317ef46bb3caab3b70e0 Mon Sep 17 00:00:00 2001 From: Ben Blum Date: Wed, 2 Nov 2011 13:38:05 -0700 Subject: cgroups: more safe tasklist locking in cgroup_attach_proc Fix unstable tasklist locking in cgroup_attach_proc. According to this thread - https://lkml.org/lkml/2011/7/27/243 - RCU is not sufficient to guarantee the tasklist is stable w.r.t. de_thread and exit. Taking tasklist_lock for reading, instead of rcu_read_lock, ensures proper exclusion. Signed-off-by: Ben Blum Acked-by: Paul Menage Cc: Oleg Nesterov Cc: Frederic Weisbecker Cc: "Paul E. McKenney" Cc: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 453100a4159d..64b0e73402df 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2027,7 +2027,7 @@ int cgroup_attach_proc(struct cgroup *cgrp, struct task_struct *leader) goto out_free_group_list; /* prevent changes to the threadgroup list while we take a snapshot. */ - rcu_read_lock(); + read_lock(&tasklist_lock); if (!thread_group_leader(leader)) { /* * a race with de_thread from another thread's exec() may strip @@ -2036,7 +2036,7 @@ int cgroup_attach_proc(struct cgroup *cgrp, struct task_struct *leader) * throw this task away and try again (from cgroup_procs_write); * this is "double-double-toil-and-trouble-check locking". */ - rcu_read_unlock(); + read_unlock(&tasklist_lock); retval = -EAGAIN; goto out_free_group_list; } @@ -2057,7 +2057,7 @@ int cgroup_attach_proc(struct cgroup *cgrp, struct task_struct *leader) } while_each_thread(leader, tsk); /* remember the number of threads in the array for later. */ group_size = i; - rcu_read_unlock(); + read_unlock(&tasklist_lock); /* * step 1: check that we can legitimately attach to the cgroup. -- cgit v1.3-8-gc7d7 From 77ceab8ea590d7dc6c8f055ce43dfebd74428107 Mon Sep 17 00:00:00 2001 From: Ben Blum Date: Wed, 2 Nov 2011 13:38:07 -0700 Subject: cgroups: don't attach task to subsystem if migration failed If a task has exited to the point it has called cgroup_exit() already, then we can't migrate it to another cgroup anymore. This can happen when we are attaching a task to a new cgroup between the call to ->can_attach_task() on subsystems and the migration that is eventually tried in cgroup_task_migrate(). In this case cgroup_task_migrate() returns -ESRCH and we don't want to attach the task to the subsystems because the attachment to the new cgroup itself failed. Fix this by only calling ->attach_task() on the subsystems if the cgroup migration succeeded. Reported-by: Oleg Nesterov Signed-off-by: Ben Blum Acked-by: Paul Menage Cc: Li Zefan Cc: Tejun Heo Signed-off-by: Frederic Weisbecker Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 64b0e73402df..8386b21224ef 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2135,14 +2135,17 @@ int cgroup_attach_proc(struct cgroup *cgrp, struct task_struct *leader) oldcgrp = task_cgroup_from_root(tsk, root); if (cgrp == oldcgrp) continue; - /* attach each task to each subsystem */ - for_each_subsys(root, ss) { - if (ss->attach_task) - ss->attach_task(cgrp, tsk); - } /* if the thread is PF_EXITING, it can just get skipped. */ retval = cgroup_task_migrate(cgrp, oldcgrp, tsk, true); - BUG_ON(retval != 0 && retval != -ESRCH); + if (retval == 0) { + /* attach each task to each subsystem */ + for_each_subsys(root, ss) { + if (ss->attach_task) + ss->attach_task(cgrp, tsk); + } + } else { + BUG_ON(retval != -ESRCH); + } } /* nothing is sensitive to fork() after this point. */ -- cgit v1.3-8-gc7d7 From 89e8a244b97e48f1f30e898b6f32acca477f2a13 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Wed, 2 Nov 2011 13:38:39 -0700 Subject: cpusets: avoid looping when storing to mems_allowed if one node remains set {get,put}_mems_allowed() exist so that general kernel code may locklessly access a task's set of allowable nodes without having the chance that a concurrent write will cause the nodemask to be empty on configurations where MAX_NUMNODES > BITS_PER_LONG. This could incur a significant delay, however, especially in low memory conditions because the page allocator is blocking and reclaim requires get_mems_allowed() itself. It is not atypical to see writes to cpuset.mems take over 2 seconds to complete, for example. In low memory conditions, this is problematic because it's one of the most imporant times to change cpuset.mems in the first place! The only way a task's set of allowable nodes may change is through cpusets by writing to cpuset.mems and when attaching a task to a generic code is not reading the nodemask with get_mems_allowed() at the same time, and then clearing all the old nodes. This prevents the possibility that a reader will see an empty nodemask at the same time the writer is storing a new nodemask. If at least one node remains unchanged, though, it's possible to simply set all new nodes and then clear all the old nodes. Changing a task's nodemask is protected by cgroup_mutex so it's guaranteed that two threads are not changing the same task's nodemask at the same time, so the nodemask is guaranteed to be stored before another thread changes it and determines whether a node remains set or not. Signed-off-by: David Rientjes Cc: Miao Xie Cc: KOSAKI Motohiro Cc: Nick Piggin Cc: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cpuset.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 10131fdaff70..ed0ff443f036 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -949,6 +949,8 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, static void cpuset_change_task_nodemask(struct task_struct *tsk, nodemask_t *newmems) { + bool masks_disjoint = !nodes_intersects(*newmems, tsk->mems_allowed); + repeat: /* * Allow tasks that have access to memory reserves because they have @@ -963,7 +965,6 @@ repeat: nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); - /* * ensure checking ->mems_allowed_change_disable after setting all new * allowed nodes. @@ -980,9 +981,11 @@ repeat: /* * Allocation of memory is very fast, we needn't sleep when waiting - * for the read-side. + * for the read-side. No wait is necessary, however, if at least one + * node remains unchanged. */ - while (ACCESS_ONCE(tsk->mems_allowed_change_disable)) { + while (masks_disjoint && + ACCESS_ONCE(tsk->mems_allowed_change_disable)) { task_unlock(tsk); if (!task_curr(tsk)) yield(); -- cgit v1.3-8-gc7d7 From f1ecf06854a66ee663f4d4cf029c78cd62a15e04 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Wed, 2 Nov 2011 13:39:22 -0700 Subject: sysctl: add support for poll() Adding support for poll() in sysctl fs allows userspace to receive notifications of changes in sysctl entries. This adds a infrastructure to allow files in sysctl fs to be pollable and implements it for hostname and domainname. [akpm@linux-foundation.org: s/declare/define/ for definitions] Signed-off-by: Lucas De Marchi Cc: Greg KH Cc: Kay Sievers Cc: Al Viro Cc: "Eric W. Biederman" Cc: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/proc_sysctl.c | 45 +++++++++++++++++++++++++++++++++++++++++++++ include/linux/sysctl.h | 22 ++++++++++++++++++++++ include/linux/utsname.h | 16 ++++++++++++++++ kernel/sys.c | 2 ++ kernel/utsname_sysctl.c | 23 +++++++++++++++++++++++ 5 files changed, 108 insertions(+) (limited to 'kernel') diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index dacd840a675a..df594803f45a 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -3,6 +3,7 @@ */ #include #include +#include #include #include #include @@ -14,6 +15,15 @@ static const struct inode_operations proc_sys_inode_operations; static const struct file_operations proc_sys_dir_file_operations; static const struct inode_operations proc_sys_dir_operations; +void proc_sys_poll_notify(struct ctl_table_poll *poll) +{ + if (!poll) + return; + + atomic_inc(&poll->event); + wake_up_interruptible(&poll->wait); +} + static struct inode *proc_sys_make_inode(struct super_block *sb, struct ctl_table_header *head, struct ctl_table *table) { @@ -176,6 +186,39 @@ static ssize_t proc_sys_write(struct file *filp, const char __user *buf, return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1); } +static int proc_sys_open(struct inode *inode, struct file *filp) +{ + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + + if (table->poll) + filp->private_data = proc_sys_poll_event(table->poll); + + return 0; +} + +static unsigned int proc_sys_poll(struct file *filp, poll_table *wait) +{ + struct inode *inode = filp->f_path.dentry->d_inode; + struct ctl_table *table = PROC_I(inode)->sysctl_entry; + unsigned long event = (unsigned long)filp->private_data; + unsigned int ret = DEFAULT_POLLMASK; + + if (!table->proc_handler) + goto out; + + if (!table->poll) + goto out; + + poll_wait(filp, &table->poll->wait, wait); + + if (event != atomic_read(&table->poll->event)) { + filp->private_data = proc_sys_poll_event(table->poll); + ret = POLLIN | POLLRDNORM | POLLERR | POLLPRI; + } + +out: + return ret; +} static int proc_sys_fill_cache(struct file *filp, void *dirent, filldir_t filldir, @@ -364,6 +407,8 @@ static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct } static const struct file_operations proc_sys_file_operations = { + .open = proc_sys_open, + .poll = proc_sys_poll, .read = proc_sys_read, .write = proc_sys_write, .llseek = default_llseek, diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 9a1ec10fd504..703cfa33a3ca 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -931,6 +931,7 @@ enum #ifdef __KERNEL__ #include #include +#include /* For the /proc/sys support */ struct ctl_table; @@ -1011,6 +1012,26 @@ extern int proc_do_large_bitmap(struct ctl_table *, int, * cover common cases. */ +/* Support for userspace poll() to watch for changes */ +struct ctl_table_poll { + atomic_t event; + wait_queue_head_t wait; +}; + +static inline void *proc_sys_poll_event(struct ctl_table_poll *poll) +{ + return (void *)(unsigned long)atomic_read(&poll->event); +} + +void proc_sys_poll_notify(struct ctl_table_poll *poll); + +#define __CTL_TABLE_POLL_INITIALIZER(name) { \ + .event = ATOMIC_INIT(0), \ + .wait = __WAIT_QUEUE_HEAD_INITIALIZER(name.wait) } + +#define DEFINE_CTL_TABLE_POLL(name) \ + struct ctl_table_poll name = __CTL_TABLE_POLL_INITIALIZER(name) + /* A sysctl table is an array of struct ctl_table: */ struct ctl_table { @@ -1021,6 +1042,7 @@ struct ctl_table struct ctl_table *child; struct ctl_table *parent; /* Automatically set */ proc_handler *proc_handler; /* Callback for text formatting */ + struct ctl_table_poll *poll; void *extra1; void *extra2; }; diff --git a/include/linux/utsname.h b/include/linux/utsname.h index 4e5b0213fdc1..c714ed75eae2 100644 --- a/include/linux/utsname.h +++ b/include/linux/utsname.h @@ -37,6 +37,14 @@ struct new_utsname { #include #include +enum uts_proc { + UTS_PROC_OSTYPE, + UTS_PROC_OSRELEASE, + UTS_PROC_VERSION, + UTS_PROC_HOSTNAME, + UTS_PROC_DOMAINNAME, +}; + struct user_namespace; extern struct user_namespace init_user_ns; @@ -80,6 +88,14 @@ static inline struct uts_namespace *copy_utsname(unsigned long flags, } #endif +#ifdef CONFIG_PROC_SYSCTL +extern void uts_proc_notify(enum uts_proc proc); +#else +static inline void uts_proc_notify(enum uts_proc proc) +{ +} +#endif + static inline struct new_utsname *utsname(void) { return ¤t->nsproxy->uts_ns->name; diff --git a/kernel/sys.c b/kernel/sys.c index 58459509b14c..d06c091e0345 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1286,6 +1286,7 @@ SYSCALL_DEFINE2(sethostname, char __user *, name, int, len) memset(u->nodename + len, 0, sizeof(u->nodename) - len); errno = 0; } + uts_proc_notify(UTS_PROC_HOSTNAME); up_write(&uts_sem); return errno; } @@ -1336,6 +1337,7 @@ SYSCALL_DEFINE2(setdomainname, char __user *, name, int, len) memset(u->domainname + len, 0, sizeof(u->domainname) - len); errno = 0; } + uts_proc_notify(UTS_PROC_DOMAINNAME); up_write(&uts_sem); return errno; } diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c index a2cd77e70d4d..3b0d48ebf81d 100644 --- a/kernel/utsname_sysctl.c +++ b/kernel/utsname_sysctl.c @@ -13,6 +13,7 @@ #include #include #include +#include static void *get_uts(ctl_table *table, int write) { @@ -51,12 +52,19 @@ static int proc_do_uts_string(ctl_table *table, int write, uts_table.data = get_uts(table, write); r = proc_dostring(&uts_table,write,buffer,lenp, ppos); put_uts(table, write, uts_table.data); + + if (write) + proc_sys_poll_notify(table->poll); + return r; } #else #define proc_do_uts_string NULL #endif +static DEFINE_CTL_TABLE_POLL(hostname_poll); +static DEFINE_CTL_TABLE_POLL(domainname_poll); + static struct ctl_table uts_kern_table[] = { { .procname = "ostype", @@ -85,6 +93,7 @@ static struct ctl_table uts_kern_table[] = { .maxlen = sizeof(init_uts_ns.name.nodename), .mode = 0644, .proc_handler = proc_do_uts_string, + .poll = &hostname_poll, }, { .procname = "domainname", @@ -92,6 +101,7 @@ static struct ctl_table uts_kern_table[] = { .maxlen = sizeof(init_uts_ns.name.domainname), .mode = 0644, .proc_handler = proc_do_uts_string, + .poll = &domainname_poll, }, {} }; @@ -105,6 +115,19 @@ static struct ctl_table uts_root_table[] = { {} }; +#ifdef CONFIG_PROC_SYSCTL +/* + * Notify userspace about a change in a certain entry of uts_kern_table, + * identified by the parameter proc. + */ +void uts_proc_notify(enum uts_proc proc) +{ + struct ctl_table *table = &uts_kern_table[proc]; + + proc_sys_poll_notify(table->poll); +} +#endif + static int __init utsname_sysctl_init(void) { register_sysctl_table(uts_root_table); -- cgit v1.3-8-gc7d7 From c1e2ee2dc436574880758b3836fc96935b774c32 Mon Sep 17 00:00:00 2001 From: Andrew Bresticker Date: Wed, 2 Nov 2011 13:40:29 -0700 Subject: memcg: replace ss->id_lock with a rwlock While back-porting Johannes Weiner's patch "mm: memcg-aware global reclaim" for an internal effort, we noticed a significant performance regression during page-reclaim heavy workloads due to high contention of the ss->id_lock. This lock protects idr map, and serializes calls to idr_get_next() in css_get_next() (which is used during the memcg hierarchy walk). Since idr_get_next() is just doing a look up, we need only serialize it with respect to idr_remove()/idr_get_new(). By making the ss->id_lock a rwlock, contention is greatly reduced and performance improves. Tested: cat a 256m file from a ramdisk in a 128m container 50 times on each core (one file + container per core) in parallel on a NUMA machine. Result is the time for the test to complete in 1 of the containers. Both kernels included Johannes' memcg-aware global reclaim patches. Before rwlock patch: 1710.778s After rwlock patch: 152.227s Signed-off-by: Andrew Bresticker Cc: Paul Menage Cc: Li Zefan Acked-by: KAMEZAWA Hiroyuki Cc: Ying Han Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cgroup.h | 2 +- kernel/cgroup.c | 18 +++++++++--------- 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index da7e4bc34e8c..1b7f9d525013 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -516,7 +516,7 @@ struct cgroup_subsys { struct list_head sibling; /* used when use_id == true */ struct idr idr; - spinlock_t id_lock; + rwlock_t id_lock; /* should be defined only by modular subsystems */ struct module *module; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 8386b21224ef..d9d5648f3cdc 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -4883,9 +4883,9 @@ void free_css_id(struct cgroup_subsys *ss, struct cgroup_subsys_state *css) rcu_assign_pointer(id->css, NULL); rcu_assign_pointer(css->id, NULL); - spin_lock(&ss->id_lock); + write_lock(&ss->id_lock); idr_remove(&ss->idr, id->id); - spin_unlock(&ss->id_lock); + write_unlock(&ss->id_lock); kfree_rcu(id, rcu_head); } EXPORT_SYMBOL_GPL(free_css_id); @@ -4911,10 +4911,10 @@ static struct css_id *get_new_cssid(struct cgroup_subsys *ss, int depth) error = -ENOMEM; goto err_out; } - spin_lock(&ss->id_lock); + write_lock(&ss->id_lock); /* Don't use 0. allocates an ID of 1-65535 */ error = idr_get_new_above(&ss->idr, newid, 1, &myid); - spin_unlock(&ss->id_lock); + write_unlock(&ss->id_lock); /* Returns error when there are no free spaces for new ID.*/ if (error) { @@ -4929,9 +4929,9 @@ static struct css_id *get_new_cssid(struct cgroup_subsys *ss, int depth) return newid; remove_idr: error = -ENOSPC; - spin_lock(&ss->id_lock); + write_lock(&ss->id_lock); idr_remove(&ss->idr, myid); - spin_unlock(&ss->id_lock); + write_unlock(&ss->id_lock); err_out: kfree(newid); return ERR_PTR(error); @@ -4943,7 +4943,7 @@ static int __init_or_module cgroup_init_idr(struct cgroup_subsys *ss, { struct css_id *newid; - spin_lock_init(&ss->id_lock); + rwlock_init(&ss->id_lock); idr_init(&ss->idr); newid = get_new_cssid(ss, 0); @@ -5038,9 +5038,9 @@ css_get_next(struct cgroup_subsys *ss, int id, * scan next entry from bitmap(tree), tmpid is updated after * idr_get_next(). */ - spin_lock(&ss->id_lock); + read_lock(&ss->id_lock); tmp = idr_get_next(&ss->idr, &tmpid); - spin_unlock(&ss->id_lock); + read_unlock(&ss->id_lock); if (!tmp) break; -- cgit v1.3-8-gc7d7 From 4536e4d1d21c8172402a2217b0fa1880665ace36 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 3 Nov 2011 07:44:04 -0700 Subject: Revert "perf: Add PM notifiers to fix CPU hotplug races" This reverts commit 144060fee07e9c22e179d00819c83c86fbcbf82c. It causes a resume regression for Andi on his Acer Aspire 1830T post 3.1. The screen just stays black after wakeup. Also, it really looks like the wrong way to suspend and resume perf events: I think they should be done as part of the CPU suspend and resume, rather than as a notifier that does smp_call_function(). Reported-by: Andi Kleen Acked-by: Ingo Molnar Cc: Peter Zijlstra Cc: Rafael J. Wysocki Signed-off-by: Linus Torvalds --- kernel/events/core.c | 97 ++-------------------------------------------------- 1 file changed, 2 insertions(+), 95 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 12a0287e0358..e1253faa34dd 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -29,7 +29,6 @@ #include #include #include -#include #include #include #include @@ -6853,7 +6852,7 @@ static void __cpuinit perf_event_init_cpu(int cpu) struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); mutex_lock(&swhash->hlist_mutex); - if (swhash->hlist_refcount > 0 && !swhash->swevent_hlist) { + if (swhash->hlist_refcount > 0) { struct swevent_hlist *hlist; hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu)); @@ -6942,14 +6941,7 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; - /* - * Ignore suspend/resume action, the perf_pm_notifier will - * take care of that. - */ - if (action & CPU_TASKS_FROZEN) - return NOTIFY_OK; - - switch (action) { + switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: case CPU_DOWN_FAILED: @@ -6968,90 +6960,6 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) return NOTIFY_OK; } -static void perf_pm_resume_cpu(void *unused) -{ - struct perf_cpu_context *cpuctx; - struct perf_event_context *ctx; - struct pmu *pmu; - int idx; - - idx = srcu_read_lock(&pmus_srcu); - list_for_each_entry_rcu(pmu, &pmus, entry) { - cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); - ctx = cpuctx->task_ctx; - - perf_ctx_lock(cpuctx, ctx); - perf_pmu_disable(cpuctx->ctx.pmu); - - cpu_ctx_sched_out(cpuctx, EVENT_ALL); - if (ctx) - ctx_sched_out(ctx, cpuctx, EVENT_ALL); - - perf_pmu_enable(cpuctx->ctx.pmu); - perf_ctx_unlock(cpuctx, ctx); - } - srcu_read_unlock(&pmus_srcu, idx); -} - -static void perf_pm_suspend_cpu(void *unused) -{ - struct perf_cpu_context *cpuctx; - struct perf_event_context *ctx; - struct pmu *pmu; - int idx; - - idx = srcu_read_lock(&pmus_srcu); - list_for_each_entry_rcu(pmu, &pmus, entry) { - cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); - ctx = cpuctx->task_ctx; - - perf_ctx_lock(cpuctx, ctx); - perf_pmu_disable(cpuctx->ctx.pmu); - - perf_event_sched_in(cpuctx, ctx, current); - - perf_pmu_enable(cpuctx->ctx.pmu); - perf_ctx_unlock(cpuctx, ctx); - } - srcu_read_unlock(&pmus_srcu, idx); -} - -static int perf_resume(void) -{ - get_online_cpus(); - smp_call_function(perf_pm_resume_cpu, NULL, 1); - put_online_cpus(); - - return NOTIFY_OK; -} - -static int perf_suspend(void) -{ - get_online_cpus(); - smp_call_function(perf_pm_suspend_cpu, NULL, 1); - put_online_cpus(); - - return NOTIFY_OK; -} - -static int perf_pm(struct notifier_block *self, unsigned long action, void *ptr) -{ - switch (action) { - case PM_POST_HIBERNATION: - case PM_POST_SUSPEND: - return perf_resume(); - case PM_HIBERNATION_PREPARE: - case PM_SUSPEND_PREPARE: - return perf_suspend(); - default: - return NOTIFY_DONE; - } -} - -static struct notifier_block perf_pm_notifier = { - .notifier_call = perf_pm, -}; - void __init perf_event_init(void) { int ret; @@ -7066,7 +6974,6 @@ void __init perf_event_init(void) perf_tp_register(); perf_cpu_notifier(perf_cpu_notify); register_reboot_notifier(&perf_reboot_notifier); - register_pm_notifier(&perf_pm_notifier); ret = init_hw_breakpoint(); WARN(ret, "hw_breakpoint initialization failed with: %d", ret); -- cgit v1.3-8-gc7d7 From 79cfbdfa87e84992d509e6c1648a18e1d7e68c20 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 3 Nov 2011 00:59:25 +0100 Subject: PM / Sleep: Fix race between CPU hotplug and freezer The CPU hotplug notifications sent out by the _cpu_up() and _cpu_down() functions depend on the value of the 'tasks_frozen' argument passed to them (which indicates whether tasks have been frozen or not). (Examples for such CPU hotplug notifications: CPU_ONLINE, CPU_ONLINE_FROZEN, CPU_DEAD, CPU_DEAD_FROZEN). Thus, it is essential that while the callbacks for those notifications are running, the state of the system with respect to the tasks being frozen or not remains unchanged, *throughout that duration*. Hence there is a need for synchronizing the CPU hotplug code with the freezer subsystem. Since the freezer is involved only in the Suspend/Hibernate call paths, this patch hooks the CPU hotplug code to the suspend/hibernate notifiers PM_[SUSPEND|HIBERNATE]_PREPARE and PM_POST_[SUSPEND|HIBERNATE] to prevent the race between CPU hotplug and freezer, thus ensuring that CPU hotplug notifications will always be run with the state of the system really being what the notifications indicate, _throughout_ their execution time. Signed-off-by: Srivatsa S. Bhat Signed-off-by: Rafael J. Wysocki --- kernel/cpu.c | 74 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index 12b7458f23b1..aa39dd7a3846 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -15,6 +15,7 @@ #include #include #include +#include #ifdef CONFIG_SMP /* Serializes the updates to cpu_online_mask, cpu_present_mask */ @@ -476,6 +477,79 @@ static int alloc_frozen_cpus(void) return 0; } core_initcall(alloc_frozen_cpus); + +/* + * Prevent regular CPU hotplug from racing with the freezer, by disabling CPU + * hotplug when tasks are about to be frozen. Also, don't allow the freezer + * to continue until any currently running CPU hotplug operation gets + * completed. + * To modify the 'cpu_hotplug_disabled' flag, we need to acquire the + * 'cpu_add_remove_lock'. And this same lock is also taken by the regular + * CPU hotplug path and released only after it is complete. Thus, we + * (and hence the freezer) will block here until any currently running CPU + * hotplug operation gets completed. + */ +void cpu_hotplug_disable_before_freeze(void) +{ + cpu_maps_update_begin(); + cpu_hotplug_disabled = 1; + cpu_maps_update_done(); +} + + +/* + * When tasks have been thawed, re-enable regular CPU hotplug (which had been + * disabled while beginning to freeze tasks). + */ +void cpu_hotplug_enable_after_thaw(void) +{ + cpu_maps_update_begin(); + cpu_hotplug_disabled = 0; + cpu_maps_update_done(); +} + +/* + * When callbacks for CPU hotplug notifications are being executed, we must + * ensure that the state of the system with respect to the tasks being frozen + * or not, as reported by the notification, remains unchanged *throughout the + * duration* of the execution of the callbacks. + * Hence we need to prevent the freezer from racing with regular CPU hotplug. + * + * This synchronization is implemented by mutually excluding regular CPU + * hotplug and Suspend/Hibernate call paths by hooking onto the Suspend/ + * Hibernate notifications. + */ +static int +cpu_hotplug_pm_callback(struct notifier_block *nb, + unsigned long action, void *ptr) +{ + switch (action) { + + case PM_SUSPEND_PREPARE: + case PM_HIBERNATION_PREPARE: + cpu_hotplug_disable_before_freeze(); + break; + + case PM_POST_SUSPEND: + case PM_POST_HIBERNATION: + cpu_hotplug_enable_after_thaw(); + break; + + default: + return NOTIFY_DONE; + } + + return NOTIFY_OK; +} + + +int cpu_hotplug_pm_sync_init(void) +{ + pm_notifier(cpu_hotplug_pm_callback, 0); + return 0; +} +core_initcall(cpu_hotplug_pm_sync_init); + #endif /* CONFIG_PM_SLEEP_SMP */ /** -- cgit v1.3-8-gc7d7 From 6513fd6972f725291ee8ce62c7a39fb8a6c7391e Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Thu, 3 Nov 2011 10:12:36 +0100 Subject: PM / QoS: Remove redundant check Remove an "if" check, that repeats an equivalent one 6 lines above. Signed-off-by: Guennadi Liakhovetski Signed-off-by: Rafael J. Wysocki --- kernel/power/qos.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 1c1797dd1d1d..5167d996cd02 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -386,8 +386,7 @@ static int pm_qos_power_open(struct inode *inode, struct file *filp) pm_qos_add_request(req, pm_qos_class, PM_QOS_DEFAULT_VALUE); filp->private_data = req; - if (filp->private_data) - return 0; + return 0; } return -EPERM; } -- cgit v1.3-8-gc7d7 From d6cc76856d353a3a9c43bead33210b9216dce332 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 4 Nov 2011 01:04:52 +0100 Subject: PM / Freezer: Revert 27920651fe "PM / Freezer: Make fake_signal_wake_up() wake TASK_KILLABLE tasks too" Commit 27920651fe "PM / Freezer: Make fake_signal_wake_up() wake TASK_KILLABLE tasks too" updated fake_signal_wake_up() used by freezer to wake up KILLABLE tasks. Sending unsolicited wakeups to tasks in killable sleep is dangerous as there are code paths which depend on tasks not waking up spuriously from KILLABLE sleep. For example. sys_read() or page can sleep in TASK_KILLABLE assuming that wait/down/whatever _killable can only fail if we can not return to the usermode. TASK_TRACED is another obvious example. The previous patch updated wait_event_freezekillable() such that it doesn't depend on the spurious wakeup. This patch reverts the offending commit. Note that the spurious KILLABLE wakeup had other implicit effects in KILLABLE sleeps in nfs and cifs and those will need further updates to regain freezekillable behavior. Signed-off-by: Tejun Heo Signed-off-by: Rafael J. Wysocki --- kernel/freezer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/freezer.c b/kernel/freezer.c index 66a594e8ad2f..7b01de98bb6a 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -67,7 +67,7 @@ static void fake_signal_wake_up(struct task_struct *p) unsigned long flags; spin_lock_irqsave(&p->sighand->siglock, flags); - signal_wake_up(p, 1); + signal_wake_up(p, 0); spin_unlock_irqrestore(&p->sighand->siglock, flags); } -- cgit v1.3-8-gc7d7 From 1cd0d6c3021c8d76641b37203f504634b87fbabc Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 1 Nov 2011 03:59:33 +0000 Subject: module: Enable dynamic debugging regardless of taint Dynamic debugging is currently disabled for tainted modules, except for TAINT_CRAP. This prevents use of dynamic debugging for out-of-tree modules once the next patch is applied. This condition was apparently intended to avoid a crash if a force- loaded module has an incompatible definition of dynamic debug structures. However, a administrator that forces us to load a module is claiming that it *is* compatible even though it fails our version checks. If they are mistaken, there are any number of ways the module could crash the system. As a side-effect, proprietary and other tainted modules can now use dynamic_debug. Signed-off-by: Ben Hutchings Acked-by: Mathieu Desnoyers Signed-off-by: Rusty Russell --- kernel/module.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index 93342d992f34..3c5509642847 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2878,8 +2878,7 @@ static struct module *load_module(void __user *umod, } /* This has to be done once we're sure module name is unique. */ - if (!mod->taints || mod->taints == (1U<taints || mod->taints == (1U< Date: Mon, 24 Oct 2011 15:12:28 +0200 Subject: module,bug: Add TAINT_OOT_MODULE flag for modules not built in-tree Use of the GPL or a compatible licence doesn't necessarily make the code any good. We already consider staging modules to be suspect, and this should also be true for out-of-tree modules which may receive very little review. Signed-off-by: Ben Hutchings Reviewed-by: Dave Jones Acked-by: Greg Kroah-Hartman Signed-off-by: Rusty Russell (patched oops-tracing.txt) --- Documentation/oops-tracing.txt | 2 ++ include/linux/kernel.h | 1 + kernel/module.c | 5 +++++ kernel/panic.c | 2 ++ scripts/mod/modpost.c | 7 +++++++ 5 files changed, 17 insertions(+) (limited to 'kernel') diff --git a/Documentation/oops-tracing.txt b/Documentation/oops-tracing.txt index 6fe9001b9263..13032c0140d4 100644 --- a/Documentation/oops-tracing.txt +++ b/Documentation/oops-tracing.txt @@ -263,6 +263,8 @@ characters, each representing a particular tainted value. 12: 'I' if the kernel is working around a severe bug in the platform firmware (BIOS or similar). + 13: 'O' if an externally-built ("out-of-tree") module has been loaded. + The primary reason for the 'Tainted: ' string is to tell kernel debuggers if this is a clean kernel or if anything unusual has occurred. Tainting is permanent: even if an offending module is diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 4c0d3b2fd5fc..e8b1597b5cf2 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -371,6 +371,7 @@ extern enum system_states { #define TAINT_WARN 9 #define TAINT_CRAP 10 #define TAINT_FIRMWARE_WORKAROUND 11 +#define TAINT_OOT_MODULE 12 extern const char hex_asc[]; #define hex_asc_lo(x) hex_asc[((x) & 0x0f)] diff --git a/kernel/module.c b/kernel/module.c index 3c5509642847..ef8cb70c6996 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2487,6 +2487,9 @@ static int check_modinfo(struct module *mod, struct load_info *info) return -ENOEXEC; } + if (!get_modinfo(info, "intree")) + add_taint_module(mod, TAINT_OOT_MODULE); + if (get_modinfo(info, "staging")) { add_taint_module(mod, TAINT_CRAP); printk(KERN_WARNING "%s: module is from the staging directory," @@ -3255,6 +3258,8 @@ static char *module_flags(struct module *mod, char *buf) buf[bx++] = '('; if (mod->taints & (1 << TAINT_PROPRIETARY_MODULE)) buf[bx++] = 'P'; + else if (mod->taints & (1 << TAINT_OOT_MODULE)) + buf[bx++] = 'O'; if (mod->taints & (1 << TAINT_FORCED_MODULE)) buf[bx++] = 'F'; if (mod->taints & (1 << TAINT_CRAP)) diff --git a/kernel/panic.c b/kernel/panic.c index d7bb6974efb5..b26593604214 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -177,6 +177,7 @@ static const struct tnt tnts[] = { { TAINT_WARN, 'W', ' ' }, { TAINT_CRAP, 'C', ' ' }, { TAINT_FIRMWARE_WORKAROUND, 'I', ' ' }, + { TAINT_OOT_MODULE, 'O', ' ' }, }; /** @@ -194,6 +195,7 @@ static const struct tnt tnts[] = { * 'W' - Taint on warning. * 'C' - modules from drivers/staging are loaded. * 'I' - Working around severe firmware bug. + * 'O' - Out-of-tree module has been loaded. * * The string is overwritten by the next call to print_tainted(). */ diff --git a/scripts/mod/modpost.c b/scripts/mod/modpost.c index a509ff8f32fa..2bd594e6d1b4 100644 --- a/scripts/mod/modpost.c +++ b/scripts/mod/modpost.c @@ -1849,6 +1849,12 @@ static void add_header(struct buffer *b, struct module *mod) buf_printf(b, "};\n"); } +static void add_intree_flag(struct buffer *b, int is_intree) +{ + if (is_intree) + buf_printf(b, "\nMODULE_INFO(intree, \"Y\");\n"); +} + static void add_staging_flag(struct buffer *b, const char *name) { static const char *staging_dir = "drivers/staging"; @@ -2169,6 +2175,7 @@ int main(int argc, char **argv) buf.pos = 0; add_header(&buf, mod); + add_intree_flag(&buf, !external_module); add_staging_flag(&buf, mod->name); err |= add_versions(&buf, mod); add_depends(&buf, mod, modules); -- cgit v1.3-8-gc7d7 From a6f05b97d1ba87326bd96f3da9fef994830d6994 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Sun, 6 Nov 2011 21:54:12 +0100 Subject: PM / QoS: Set cpu_dma_pm_qos->name Since commit 4a31a334, the name of this misc device is not initialized, which leads to a funny device named /dev/(null) being created and /proc/misc containing an entry with just a number but no name. The latter leads to complaints by cryptsetup, which caused me to investigate this matter. Signed-off-by: Dominik Brodowski Signed-off-by: Rafael J. Wysocki --- kernel/power/qos.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 56db75147186..995e3bd3417b 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -70,6 +70,7 @@ static struct pm_qos_constraints cpu_dma_constraints = { }; static struct pm_qos_object cpu_dma_pm_qos = { .constraints = &cpu_dma_constraints, + .name = "cpu_dma_latency", }; static BLOCKING_NOTIFIER_HEAD(network_lat_notifier); -- cgit v1.3-8-gc7d7