From f06e8c584fa0d05312c11ea66194f3d2efb93c21 Mon Sep 17 00:00:00 2001 From: Dmitry Vyukov Date: Thu, 22 Jun 2017 16:14:17 +0200 Subject: kasan: Allow kasan_check_read/write() to accept pointers to volatiles Currently kasan_check_read/write() accept 'const void*', make them accept 'const volatile void*'. This is required for instrumentation of atomic operations and there is just no reason to not allow that. Signed-off-by: Dmitry Vyukov Reviewed-by: Andrey Ryabinin Acked-by: Mark Rutland Cc: Andrew Morton Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: kasan-dev@googlegroups.com Cc: linux-mm@kvack.org Cc: will.deacon@arm.com Link: http://lkml.kernel.org/r/33e5ec275c1ee89299245b2ebbccd63709c6021f.1498140838.git.dvyukov@google.com Signed-off-by: Ingo Molnar --- include/linux/kasan-checks.h | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kasan-checks.h b/include/linux/kasan-checks.h index b7f8aced7870..41960fecf783 100644 --- a/include/linux/kasan-checks.h +++ b/include/linux/kasan-checks.h @@ -2,11 +2,13 @@ #define _LINUX_KASAN_CHECKS_H #ifdef CONFIG_KASAN -void kasan_check_read(const void *p, unsigned int size); -void kasan_check_write(const void *p, unsigned int size); +void kasan_check_read(const volatile void *p, unsigned int size); +void kasan_check_write(const volatile void *p, unsigned int size); #else -static inline void kasan_check_read(const void *p, unsigned int size) { } -static inline void kasan_check_write(const void *p, unsigned int size) { } +static inline void kasan_check_read(const volatile void *p, unsigned int size) +{ } +static inline void kasan_check_write(const volatile void *p, unsigned int size) +{ } #endif #endif -- cgit v1.2.3 From bc2eecd7ecce40af43b6eb3d256b6076257df846 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 1 Aug 2017 00:31:32 -0400 Subject: futex: Allow for compiling out PI support This makes it possible to preserve basic futex support and compile out the PI support when RT mutexes are not available. Signed-off-by: Nicolas Pitre Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Darren Hart Link: http://lkml.kernel.org/r/alpine.LFD.2.20.1708010024190.5981@knanqh.ubzr --- include/linux/futex.h | 7 ++++++- init/Kconfig | 7 ++++++- kernel/futex.c | 22 ++++++++++++++++++++++ kernel/locking/rtmutex_common.h | 29 +++++++++++++++++++++++++++++ 4 files changed, 63 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/futex.h b/include/linux/futex.h index 7c5b694864cd..f36bfd26f998 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -54,7 +54,6 @@ union futex_key { #ifdef CONFIG_FUTEX extern void exit_robust_list(struct task_struct *curr); -extern void exit_pi_state_list(struct task_struct *curr); #ifdef CONFIG_HAVE_FUTEX_CMPXCHG #define futex_cmpxchg_enabled 1 #else @@ -64,8 +63,14 @@ extern int futex_cmpxchg_enabled; static inline void exit_robust_list(struct task_struct *curr) { } +#endif + +#ifdef CONFIG_FUTEX_PI +extern void exit_pi_state_list(struct task_struct *curr); +#else static inline void exit_pi_state_list(struct task_struct *curr) { } #endif + #endif diff --git a/init/Kconfig b/init/Kconfig index 8514b25db21c..5f0ef850e808 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1275,12 +1275,17 @@ config BASE_FULL config FUTEX bool "Enable futex support" if EXPERT default y - select RT_MUTEXES + imply RT_MUTEXES help Disabling this option will cause the kernel to be built without support for "fast userspace mutexes". The resulting kernel may not run glibc-based applications correctly. +config FUTEX_PI + bool + depends on FUTEX && RT_MUTEXES + default y + config HAVE_FUTEX_CMPXCHG bool depends on FUTEX diff --git a/kernel/futex.c b/kernel/futex.c index 16dbe4c93895..ad0af4df1b9d 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -875,6 +875,8 @@ static struct task_struct *futex_find_get_task(pid_t pid) return p; } +#ifdef CONFIG_FUTEX_PI + /* * This task is holding PI mutexes at exit time => bad. * Kernel cleans up PI-state, but userspace is likely hosed. @@ -932,6 +934,8 @@ void exit_pi_state_list(struct task_struct *curr) raw_spin_unlock_irq(&curr->pi_lock); } +#endif + /* * We need to check the following states: * @@ -1799,6 +1803,15 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags, struct futex_q *this, *next; DEFINE_WAKE_Q(wake_q); + /* + * When PI not supported: return -ENOSYS if requeue_pi is true, + * consequently the compiler knows requeue_pi is always false past + * this point which will optimize away all the conditional code + * further down. + */ + if (!IS_ENABLED(CONFIG_FUTEX_PI) && requeue_pi) + return -ENOSYS; + if (requeue_pi) { /* * Requeue PI only works on two distinct uaddrs. This @@ -2594,6 +2607,9 @@ static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, struct futex_q q = futex_q_init; int res, ret; + if (!IS_ENABLED(CONFIG_FUTEX_PI)) + return -ENOSYS; + if (refill_pi_state_cache()) return -ENOMEM; @@ -2773,6 +2789,9 @@ static int futex_unlock_pi(u32 __user *uaddr, unsigned int flags) struct futex_q *top_waiter; int ret; + if (!IS_ENABLED(CONFIG_FUTEX_PI)) + return -ENOSYS; + retry: if (get_user(uval, uaddr)) return -EFAULT; @@ -2983,6 +3002,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, struct futex_q q = futex_q_init; int res, ret; + if (!IS_ENABLED(CONFIG_FUTEX_PI)) + return -ENOSYS; + if (uaddr == uaddr2) return -EINVAL; diff --git a/kernel/locking/rtmutex_common.h b/kernel/locking/rtmutex_common.h index 72ad45a9a794..8d039b928d61 100644 --- a/kernel/locking/rtmutex_common.h +++ b/kernel/locking/rtmutex_common.h @@ -40,6 +40,9 @@ struct rt_mutex_waiter { /* * Various helpers to access the waiters-tree: */ + +#ifdef CONFIG_RT_MUTEXES + static inline int rt_mutex_has_waiters(struct rt_mutex *lock) { return !RB_EMPTY_ROOT(&lock->waiters); @@ -69,6 +72,32 @@ task_top_pi_waiter(struct task_struct *p) pi_tree_entry); } +#else + +static inline int rt_mutex_has_waiters(struct rt_mutex *lock) +{ + return false; +} + +static inline struct rt_mutex_waiter * +rt_mutex_top_waiter(struct rt_mutex *lock) +{ + return NULL; +} + +static inline int task_has_pi_waiters(struct task_struct *p) +{ + return false; +} + +static inline struct rt_mutex_waiter * +task_top_pi_waiter(struct task_struct *p) +{ + return NULL; +} + +#endif + /* * lock->owner state tracking: */ -- cgit v1.2.3 From 0aa1125fa8bc5e5f98317156728fa4d0293561a5 Mon Sep 17 00:00:00 2001 From: Kirill Tkhai Date: Mon, 19 Jun 2017 21:02:12 +0300 Subject: locking/rwsem-spinlock: Add killable versions of __down_read() Rename __down_read() in __down_read_common() and teach it to abort waiting in case of pending signals and killable state argument passed. Note, that we shouldn't wake anybody up in EINTR path, as: We check for signal_pending_state() after (!waiter.task) test and under spinlock. So, current task wasn't able to be woken up. It may be in two cases: a writer is owner of the sem, or a writer is a first waiter of the sem. If a writer is owner of the sem, no one else may work with it in parallel. It will wake somebody, when it call up_write() or downgrade_write(). If a writer is the first waiter, it will be woken up, when the last active reader releases the sem, and sem->count became 0. Also note, that set_current_state() may be moved down to schedule() (after !waiter.task check), as all assignments in this type of semaphore (including wake_up), occur under spinlock, so we can't miss anything. Signed-off-by: Kirill Tkhai Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: arnd@arndb.de Cc: avagin@virtuozzo.com Cc: davem@davemloft.net Cc: fenghua.yu@intel.com Cc: gorcunov@virtuozzo.com Cc: heiko.carstens@de.ibm.com Cc: hpa@zytor.com Cc: ink@jurassic.park.msu.ru Cc: mattst88@gmail.com Cc: rth@twiddle.net Cc: schwidefsky@de.ibm.com Cc: tony.luck@intel.com Link: http://lkml.kernel.org/r/149789533283.9059.9829416940494747182.stgit@localhost.localdomain Signed-off-by: Ingo Molnar --- include/linux/rwsem-spinlock.h | 1 + kernel/locking/rwsem-spinlock.c | 37 ++++++++++++++++++++++++++++--------- 2 files changed, 29 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rwsem-spinlock.h b/include/linux/rwsem-spinlock.h index ae0528b834cd..e784761a4443 100644 --- a/include/linux/rwsem-spinlock.h +++ b/include/linux/rwsem-spinlock.h @@ -32,6 +32,7 @@ struct rw_semaphore { #define RWSEM_UNLOCKED_VALUE 0x00000000 extern void __down_read(struct rw_semaphore *sem); +extern int __must_check __down_read_killable(struct rw_semaphore *sem); extern int __down_read_trylock(struct rw_semaphore *sem); extern void __down_write(struct rw_semaphore *sem); extern int __must_check __down_write_killable(struct rw_semaphore *sem); diff --git a/kernel/locking/rwsem-spinlock.c b/kernel/locking/rwsem-spinlock.c index 20819df98125..0848634c5512 100644 --- a/kernel/locking/rwsem-spinlock.c +++ b/kernel/locking/rwsem-spinlock.c @@ -126,7 +126,7 @@ __rwsem_wake_one_writer(struct rw_semaphore *sem) /* * get a read lock on the semaphore */ -void __sched __down_read(struct rw_semaphore *sem) +int __sched __down_read_common(struct rw_semaphore *sem, int state) { struct rwsem_waiter waiter; unsigned long flags; @@ -140,8 +140,6 @@ void __sched __down_read(struct rw_semaphore *sem) goto out; } - set_current_state(TASK_UNINTERRUPTIBLE); - /* set up my own style of waitqueue */ waiter.task = current; waiter.type = RWSEM_WAITING_FOR_READ; @@ -149,20 +147,41 @@ void __sched __down_read(struct rw_semaphore *sem) list_add_tail(&waiter.list, &sem->wait_list); - /* we don't need to touch the semaphore struct anymore */ - raw_spin_unlock_irqrestore(&sem->wait_lock, flags); - /* wait to be given the lock */ for (;;) { if (!waiter.task) break; + if (signal_pending_state(state, current)) + goto out_nolock; + set_current_state(state); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); schedule(); - set_current_state(TASK_UNINTERRUPTIBLE); + raw_spin_lock_irqsave(&sem->wait_lock, flags); } - __set_current_state(TASK_RUNNING); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); out: - ; + return 0; + +out_nolock: + /* + * We didn't take the lock, so that there is a writer, which + * is owner or the first waiter of the sem. If it's a waiter, + * it will be woken by current owner. Not need to wake anybody. + */ + list_del(&waiter.list); + raw_spin_unlock_irqrestore(&sem->wait_lock, flags); + return -EINTR; +} + +void __sched __down_read(struct rw_semaphore *sem) +{ + __down_read_common(sem, TASK_UNINTERRUPTIBLE); +} + +int __sched __down_read_killable(struct rw_semaphore *sem) +{ + return __down_read_common(sem, TASK_KILLABLE); } /* -- cgit v1.2.3 From 83ced169d9a01f22eb39f1fcc1f89ad9d223238f Mon Sep 17 00:00:00 2001 From: Kirill Tkhai Date: Mon, 19 Jun 2017 21:02:26 +0300 Subject: locking/rwsem-xadd: Add killable versions of rwsem_down_read_failed() Rename rwsem_down_read_failed() in __rwsem_down_read_failed_common() and teach it to abort waiting in case of pending signals and killable state argument passed. Note, that we shouldn't wake anybody up in EINTR path, as: We check for (waiter.task) under spinlock before we go to out_nolock path. Current task wasn't able to be woken up, so there are a writer, owning the sem, or a writer, which is the first waiter. In the both cases we shouldn't wake anybody. If there is a writer, owning the sem, and we were the only waiter, remove RWSEM_WAITING_BIAS, as there are no waiters anymore. Signed-off-by: Kirill Tkhai Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: arnd@arndb.de Cc: avagin@virtuozzo.com Cc: davem@davemloft.net Cc: fenghua.yu@intel.com Cc: gorcunov@virtuozzo.com Cc: heiko.carstens@de.ibm.com Cc: hpa@zytor.com Cc: ink@jurassic.park.msu.ru Cc: mattst88@gmail.com Cc: rth@twiddle.net Cc: schwidefsky@de.ibm.com Cc: tony.luck@intel.com Link: http://lkml.kernel.org/r/149789534632.9059.2901382369609922565.stgit@localhost.localdomain Signed-off-by: Ingo Molnar --- include/linux/rwsem.h | 1 + kernel/locking/rwsem-xadd.c | 33 ++++++++++++++++++++++++++++++--- 2 files changed, 31 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h index dd1d14250340..0ad7318ff299 100644 --- a/include/linux/rwsem.h +++ b/include/linux/rwsem.h @@ -44,6 +44,7 @@ struct rw_semaphore { }; extern struct rw_semaphore *rwsem_down_read_failed(struct rw_semaphore *sem); +extern struct rw_semaphore *rwsem_down_read_failed_killable(struct rw_semaphore *sem); extern struct rw_semaphore *rwsem_down_write_failed(struct rw_semaphore *sem); extern struct rw_semaphore *rwsem_down_write_failed_killable(struct rw_semaphore *sem); extern struct rw_semaphore *rwsem_wake(struct rw_semaphore *); diff --git a/kernel/locking/rwsem-xadd.c b/kernel/locking/rwsem-xadd.c index 34e727f18e49..02f660666ab8 100644 --- a/kernel/locking/rwsem-xadd.c +++ b/kernel/locking/rwsem-xadd.c @@ -221,8 +221,8 @@ static void __rwsem_mark_wake(struct rw_semaphore *sem, /* * Wait for the read lock to be granted */ -__visible -struct rw_semaphore __sched *rwsem_down_read_failed(struct rw_semaphore *sem) +static inline struct rw_semaphore __sched * +__rwsem_down_read_failed_common(struct rw_semaphore *sem, int state) { long count, adjustment = -RWSEM_ACTIVE_READ_BIAS; struct rwsem_waiter waiter; @@ -255,17 +255,44 @@ struct rw_semaphore __sched *rwsem_down_read_failed(struct rw_semaphore *sem) /* wait to be given the lock */ while (true) { - set_current_state(TASK_UNINTERRUPTIBLE); + set_current_state(state); if (!waiter.task) break; + if (signal_pending_state(state, current)) { + raw_spin_lock_irq(&sem->wait_lock); + if (waiter.task) + goto out_nolock; + raw_spin_unlock_irq(&sem->wait_lock); + break; + } schedule(); } __set_current_state(TASK_RUNNING); return sem; +out_nolock: + list_del(&waiter.list); + if (list_empty(&sem->wait_list)) + atomic_long_add(-RWSEM_WAITING_BIAS, &sem->count); + raw_spin_unlock_irq(&sem->wait_lock); + __set_current_state(TASK_RUNNING); + return ERR_PTR(-EINTR); +} + +__visible struct rw_semaphore * __sched +rwsem_down_read_failed(struct rw_semaphore *sem) +{ + return __rwsem_down_read_failed_common(sem, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(rwsem_down_read_failed); +__visible struct rw_semaphore * __sched +rwsem_down_read_failed_killable(struct rw_semaphore *sem) +{ + return __rwsem_down_read_failed_common(sem, TASK_KILLABLE); +} +EXPORT_SYMBOL(rwsem_down_read_failed_killable); + /* * This function must be called with the sem->wait_lock held to prevent * race conditions between checking the rwsem wait list and setting the -- cgit v1.2.3 From 1dbb6704de91b169a58d0c8221624afd6a95cfc7 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 1 Aug 2017 17:24:04 +0200 Subject: jump_label: Fix concurrent static_key_enable/disable() static_key_enable/disable are trying to cap the static key count to 0/1. However, their use of key->enabled is outside jump_label_lock so they do not really ensure that. Rewrite them to do a quick check for an already enabled (respectively, already disabled), and then recheck under the jump label lock. Unlike static_key_slow_inc/dec, a failed check under the jump label lock does not modify key->enabled. Signed-off-by: Paolo Bonzini Signed-off-by: Peter Zijlstra (Intel) Cc: Eric Dumazet Cc: Jason Baron Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1501601046-35683-2-git-send-email-pbonzini@redhat.com Signed-off-by: Ingo Molnar --- include/linux/jump_label.h | 22 +++++++++-------- kernel/jump_label.c | 59 +++++++++++++++++++++++++++++----------------- 2 files changed, 49 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 2afd74b9d844..740a42ea7f7f 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -234,22 +234,24 @@ static inline int jump_label_apply_nops(struct module *mod) static inline void static_key_enable(struct static_key *key) { - int count = static_key_count(key); - - WARN_ON_ONCE(count < 0 || count > 1); + STATIC_KEY_CHECK_USE(); - if (!count) - static_key_slow_inc(key); + if (atomic_read(&key->enabled) != 0) { + WARN_ON_ONCE(atomic_read(&key->enabled) != 1); + return; + } + atomic_set(&key->enabled, 1); } static inline void static_key_disable(struct static_key *key) { - int count = static_key_count(key); - - WARN_ON_ONCE(count < 0 || count > 1); + STATIC_KEY_CHECK_USE(); - if (count) - static_key_slow_dec(key); + if (atomic_read(&key->enabled) != 1) { + WARN_ON_ONCE(atomic_read(&key->enabled) != 0); + return; + } + atomic_set(&key->enabled, 0); } #define STATIC_KEY_INIT_TRUE { .enabled = ATOMIC_INIT(1) } diff --git a/kernel/jump_label.c b/kernel/jump_label.c index d11c506a6ac3..833eecae825e 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -79,28 +79,6 @@ int static_key_count(struct static_key *key) } EXPORT_SYMBOL_GPL(static_key_count); -void static_key_enable(struct static_key *key) -{ - int count = static_key_count(key); - - WARN_ON_ONCE(count < 0 || count > 1); - - if (!count) - static_key_slow_inc(key); -} -EXPORT_SYMBOL_GPL(static_key_enable); - -void static_key_disable(struct static_key *key) -{ - int count = static_key_count(key); - - WARN_ON_ONCE(count < 0 || count > 1); - - if (count) - static_key_slow_dec(key); -} -EXPORT_SYMBOL_GPL(static_key_disable); - void static_key_slow_inc(struct static_key *key) { int v, v1; @@ -139,6 +117,43 @@ void static_key_slow_inc(struct static_key *key) } EXPORT_SYMBOL_GPL(static_key_slow_inc); +void static_key_enable(struct static_key *key) +{ + STATIC_KEY_CHECK_USE(); + if (atomic_read(&key->enabled) > 0) { + WARN_ON_ONCE(atomic_read(&key->enabled) != 1); + return; + } + + cpus_read_lock(); + jump_label_lock(); + if (atomic_read(&key->enabled) == 0) { + atomic_set(&key->enabled, -1); + jump_label_update(key); + atomic_set(&key->enabled, 1); + } + jump_label_unlock(); + cpus_read_unlock(); +} +EXPORT_SYMBOL_GPL(static_key_enable); + +void static_key_disable(struct static_key *key) +{ + STATIC_KEY_CHECK_USE(); + if (atomic_read(&key->enabled) != 1) { + WARN_ON_ONCE(atomic_read(&key->enabled) != 0); + return; + } + + cpus_read_lock(); + jump_label_lock(); + if (atomic_cmpxchg(&key->enabled, 1, 0)) + jump_label_update(key); + jump_label_unlock(); + cpus_read_unlock(); +} +EXPORT_SYMBOL_GPL(static_key_disable); + static void __static_key_slow_dec(struct static_key *key, unsigned long rate_limit, struct delayed_work *work) { -- cgit v1.2.3 From be040bea9085a9c2b1700c9e60888777baeb96d5 Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 1 Aug 2017 17:24:06 +0200 Subject: cpuset: Make nr_cpusets private Any use of key->enabled (that is static_key_enabled and static_key_count) outside jump_label_lock should handle its own serialization. In the case of cpusets_enabled_key, the key is always incremented/decremented under cpuset_mutex, and hence the same rule applies to nr_cpusets. The rule *is* respected currently, but the mutex is static so nr_cpusets should be static too. Signed-off-by: Paolo Bonzini Signed-off-by: Peter Zijlstra (Intel) Acked-by: Zefan Li Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1501601046-35683-4-git-send-email-pbonzini@redhat.com Signed-off-by: Ingo Molnar --- include/linux/cpuset.h | 6 ------ kernel/cgroup/cpuset.c | 7 +++++++ 2 files changed, 7 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 898cfe2eeb42..e74655d941b7 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -37,12 +37,6 @@ static inline bool cpusets_enabled(void) return static_branch_unlikely(&cpusets_enabled_key); } -static inline int nr_cpusets(void) -{ - /* jump label reference count + the top-level cpuset */ - return static_key_count(&cpusets_enabled_key.key) + 1; -} - static inline void cpuset_inc(void) { static_branch_inc(&cpusets_pre_enable_key); diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 8d5151688504..9ed6a051a1b9 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -577,6 +577,13 @@ static void update_domain_attr_tree(struct sched_domain_attr *dattr, rcu_read_unlock(); } +/* Must be called with cpuset_mutex held. */ +static inline int nr_cpusets(void) +{ + /* jump label reference count + the top-level cpuset */ + return static_key_count(&cpusets_enabled_key.key) + 1; +} + /* * generate_sched_domains() * -- cgit v1.2.3 From 5a40527f8f0798553764fc8db4111d7d9c33ea51 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 1 Aug 2017 09:02:56 +0100 Subject: jump_label: Provide hotplug context variants As using the normal static key API under the hotplug lock is pretty much impossible, let's provide a variant of some of them that require the hotplug lock to have already been taken. These function are only meant to be used in CPU hotplug callbacks. Signed-off-by: Marc Zyngier Signed-off-by: Peter Zijlstra (Intel) Cc: Leo Yan Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20170801080257.5056-4-marc.zyngier@arm.com Signed-off-by: Ingo Molnar --- Documentation/static-keys.txt | 15 +++++++++++++++ include/linux/jump_label.h | 11 +++++++++-- kernel/jump_label.c | 22 ++++++++++++++++++---- 3 files changed, 42 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/Documentation/static-keys.txt b/Documentation/static-keys.txt index 870b4be3cb11..ab16efe0c79d 100644 --- a/Documentation/static-keys.txt +++ b/Documentation/static-keys.txt @@ -154,6 +154,21 @@ and 'static_key_count()'. In general, if you use these functions, they should be protected with the same mutex used around the enable/disable or increment/decrement function. +Note that switching branches results in some locks being taken, +particularly the CPU hotplug lock (in order to avoid races against +CPUs being brought in the kernel whilst the kernel is getting +patched). Calling the static key API from within a hotplug notifier is +thus a sure deadlock recipe. In order to still allow use of the +functionnality, the following functions are provided: + + static_key_enable_cpuslocked() + static_key_disable_cpuslocked() + static_branch_enable_cpuslocked() + static_branch_disable_cpuslocked() + +These functions are *not* general purpose, and must only be used when +you really know that you're in the above context, and no other. + Where an array of keys is required, it can be defined as:: DEFINE_STATIC_KEY_ARRAY_TRUE(keys, count); diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 740a42ea7f7f..cd5861651b17 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -163,6 +163,8 @@ extern void jump_label_apply_nops(struct module *mod); extern int static_key_count(struct static_key *key); extern void static_key_enable(struct static_key *key); extern void static_key_disable(struct static_key *key); +extern void static_key_enable_cpuslocked(struct static_key *key); +extern void static_key_disable_cpuslocked(struct static_key *key); /* * We should be using ATOMIC_INIT() for initializing .enabled, but @@ -254,6 +256,9 @@ static inline void static_key_disable(struct static_key *key) atomic_set(&key->enabled, 0); } +#define static_key_enable_cpuslocked(k) static_key_enable((k)) +#define static_key_disable_cpuslocked(k) static_key_disable((k)) + #define STATIC_KEY_INIT_TRUE { .enabled = ATOMIC_INIT(1) } #define STATIC_KEY_INIT_FALSE { .enabled = ATOMIC_INIT(0) } @@ -415,8 +420,10 @@ extern bool ____wrong_branch_error(void); * Normal usage; boolean enable/disable. */ -#define static_branch_enable(x) static_key_enable(&(x)->key) -#define static_branch_disable(x) static_key_disable(&(x)->key) +#define static_branch_enable(x) static_key_enable(&(x)->key) +#define static_branch_disable(x) static_key_disable(&(x)->key) +#define static_branch_enable_cpuslocked(x) static_key_enable_cpuslocked(&(x)->key) +#define static_branch_disable_cpuslocked(x) static_key_disable_cpuslocked(&(x)->key) #endif /* __ASSEMBLY__ */ diff --git a/kernel/jump_label.c b/kernel/jump_label.c index cc6d815c75ed..0bf2e8f5244a 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -126,15 +126,15 @@ void static_key_slow_inc(struct static_key *key) } EXPORT_SYMBOL_GPL(static_key_slow_inc); -void static_key_enable(struct static_key *key) +void static_key_enable_cpuslocked(struct static_key *key) { STATIC_KEY_CHECK_USE(); + if (atomic_read(&key->enabled) > 0) { WARN_ON_ONCE(atomic_read(&key->enabled) != 1); return; } - cpus_read_lock(); jump_label_lock(); if (atomic_read(&key->enabled) == 0) { atomic_set(&key->enabled, -1); @@ -145,23 +145,37 @@ void static_key_enable(struct static_key *key) atomic_set_release(&key->enabled, 1); } jump_label_unlock(); +} +EXPORT_SYMBOL_GPL(static_key_enable_cpuslocked); + +void static_key_enable(struct static_key *key) +{ + cpus_read_lock(); + static_key_enable_cpuslocked(key); cpus_read_unlock(); } EXPORT_SYMBOL_GPL(static_key_enable); -void static_key_disable(struct static_key *key) +void static_key_disable_cpuslocked(struct static_key *key) { STATIC_KEY_CHECK_USE(); + if (atomic_read(&key->enabled) != 1) { WARN_ON_ONCE(atomic_read(&key->enabled) != 0); return; } - cpus_read_lock(); jump_label_lock(); if (atomic_cmpxchg(&key->enabled, 1, 0)) jump_label_update(key); jump_label_unlock(); +} +EXPORT_SYMBOL_GPL(static_key_disable_cpuslocked); + +void static_key_disable(struct static_key *key) +{ + cpus_read_lock(); + static_key_disable_cpuslocked(key); cpus_read_unlock(); } EXPORT_SYMBOL_GPL(static_key_disable); -- cgit v1.2.3 From 8b1b436dd1ccc8a1770af6e56eec047ad4920659 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 7 Jun 2017 18:05:07 +0200 Subject: mm, locking: Rework {set,clear,mm}_tlb_flush_pending() Commit: af2c1401e6f9 ("mm: numa: guarantee that tlb_flush_pending updates are visible before page table updates") added smp_mb__before_spinlock() to set_tlb_flush_pending(). I think we can solve the same problem without this barrier. If instead we mandate that mm_tlb_flush_pending() is used while holding the PTL we're guaranteed to observe prior set_tlb_flush_pending() instances. For this to work we need to rework migrate_misplaced_transhuge_page() a little and move the test up into do_huge_pmd_numa_page(). NOTE: this relies on flush_tlb_range() to guarantee: (1) it ensures that prior page table updates are visible to the page table walker and (2) it ensures that subsequent memory accesses are only made visible after the invalidation has completed This is required for architectures that implement TRANSPARENT_HUGEPAGE (arc, arm, arm64, mips, powerpc, s390, sparc, x86) or otherwise use mm_tlb_flush_pending() in their page-table operations (arm, arm64, x86). This appears true for: - arm (DSB ISB before and after), - arm64 (DSB ISHST before, and DSB ISH after), - powerpc (PTESYNC before and after), - s390 and x86 TLB invalidate are serializing instructions But I failed to understand the situation for: - arc, mips, sparc Now SPARC64 is a wee bit special in that flush_tlb_range() is a no-op and it flushes the TLBs using arch_{enter,leave}_lazy_mmu_mode() inside the PTL. It still needs to guarantee the PTL unlock happens _after_ the invalidate completes. Vineet, Ralf and Dave could you guys please have a look? Signed-off-by: Peter Zijlstra (Intel) Acked-by: Will Deacon Cc: Andrew Morton Cc: Benjamin Herrenschmidt Cc: David S. Miller Cc: Heiko Carstens Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Ralf Baechle Cc: Rik van Riel Cc: Russell King Cc: Thomas Gleixner Cc: Vineet Gupta Signed-off-by: Ingo Molnar --- include/linux/mm_types.h | 33 +++++++++++++++++++++++++++------ mm/huge_memory.c | 20 ++++++++++++++++++++ mm/migrate.c | 6 ------ 3 files changed, 47 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index 7f384bb62d8e..36ea3cf7d85e 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -531,23 +531,44 @@ static inline cpumask_t *mm_cpumask(struct mm_struct *mm) */ static inline bool mm_tlb_flush_pending(struct mm_struct *mm) { - barrier(); + /* + * Must be called with PTL held; such that our PTL acquire will have + * observed the store from set_tlb_flush_pending(). + */ return mm->tlb_flush_pending; } static inline void set_tlb_flush_pending(struct mm_struct *mm) { mm->tlb_flush_pending = true; - /* - * Guarantee that the tlb_flush_pending store does not leak into the - * critical section updating the page tables + * The only time this value is relevant is when there are indeed pages + * to flush. And we'll only flush pages after changing them, which + * requires the PTL. + * + * So the ordering here is: + * + * mm->tlb_flush_pending = true; + * spin_lock(&ptl); + * ... + * set_pte_at(); + * spin_unlock(&ptl); + * + * spin_lock(&ptl) + * mm_tlb_flush_pending(); + * .... + * spin_unlock(&ptl); + * + * flush_tlb_range(); + * mm->tlb_flush_pending = false; + * + * So the =true store is constrained by the PTL unlock, and the =false + * store is constrained by the TLB invalidate. */ - smp_mb__before_spinlock(); } /* Clearing is done after a TLB flush, which also provides a barrier. */ static inline void clear_tlb_flush_pending(struct mm_struct *mm) { - barrier(); + /* see set_tlb_flush_pending */ mm->tlb_flush_pending = false; } #else diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 86975dec0ba1..c76a720b936b 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1410,6 +1410,7 @@ int do_huge_pmd_numa_page(struct vm_fault *vmf, pmd_t pmd) unsigned long haddr = vmf->address & HPAGE_PMD_MASK; int page_nid = -1, this_nid = numa_node_id(); int target_nid, last_cpupid = -1; + bool need_flush = false; bool page_locked; bool migrated = false; bool was_writable; @@ -1495,11 +1496,30 @@ int do_huge_pmd_numa_page(struct vm_fault *vmf, pmd_t pmd) goto clear_pmdnuma; } + /* + * Since we took the NUMA fault, we must have observed the !accessible + * bit. Make sure all other CPUs agree with that, to avoid them + * modifying the page we're about to migrate. + * + * Must be done under PTL such that we'll observe the relevant + * set_tlb_flush_pending(). + */ + if (mm_tlb_flush_pending(vma->vm_mm)) + need_flush = true; + /* * Migrate the THP to the requested node, returns with page unlocked * and access rights restored. */ spin_unlock(vmf->ptl); + + /* + * We are not sure a pending tlb flush here is for a huge page + * mapping or not. Hence use the tlb range variant + */ + if (need_flush) + flush_tlb_range(vma, haddr, haddr + HPAGE_PMD_SIZE); + migrated = migrate_misplaced_transhuge_page(vma->vm_mm, vma, vmf->pmd, pmd, vmf->address, page, target_nid); if (migrated) { diff --git a/mm/migrate.c b/mm/migrate.c index 627671551873..d68a41da6abb 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1937,12 +1937,6 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, put_page(new_page); goto out_fail; } - /* - * We are not sure a pending tlb flush here is for a huge page - * mapping or not. Hence use the tlb range variant - */ - if (mm_tlb_flush_pending(mm)) - flush_tlb_range(vma, mmun_start, mmun_end); /* Prepare a page as a migration target */ __SetPageLocked(new_page); -- cgit v1.2.3 From d89e588ca4081615216cc25f2489b0281ac0bfe9 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 5 Sep 2016 11:37:53 +0200 Subject: locking: Introduce smp_mb__after_spinlock() Since its inception, our understanding of ACQUIRE, esp. as applied to spinlocks, has changed somewhat. Also, I wonder if, with a simple change, we cannot make it provide more. The problem with the comment is that the STORE done by spin_lock isn't itself ordered by the ACQUIRE, and therefore a later LOAD can pass over it and cross with any prior STORE, rendering the default WMB insufficient (pointed out by Alan). Now, this is only really a problem on PowerPC and ARM64, both of which already defined smp_mb__before_spinlock() as a smp_mb(). At the same time, we can get a much stronger construct if we place that same barrier _inside_ the spin_lock(). In that case we upgrade the RCpc spinlock to an RCsc. That would make all schedule() calls fully transitive against one another. Signed-off-by: Peter Zijlstra (Intel) Acked-by: Will Deacon Cc: Alan Stern Cc: Benjamin Herrenschmidt Cc: Linus Torvalds Cc: Michael Ellerman Cc: Nicholas Piggin Cc: Oleg Nesterov Cc: Paul McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar --- arch/arm64/include/asm/spinlock.h | 2 ++ arch/powerpc/include/asm/spinlock.h | 3 +++ include/linux/atomic.h | 3 +++ include/linux/spinlock.h | 36 ++++++++++++++++++++++++++++++++++++ kernel/sched/core.c | 4 ++-- 5 files changed, 46 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/arch/arm64/include/asm/spinlock.h b/arch/arm64/include/asm/spinlock.h index cae331d553f8..b103888b694a 100644 --- a/arch/arm64/include/asm/spinlock.h +++ b/arch/arm64/include/asm/spinlock.h @@ -367,5 +367,7 @@ static inline int arch_read_trylock(arch_rwlock_t *rw) * smp_mb__before_spinlock() can restore the required ordering. */ #define smp_mb__before_spinlock() smp_mb() +/* See include/linux/spinlock.h */ +#define smp_mb__after_spinlock() smp_mb() #endif /* __ASM_SPINLOCK_H */ diff --git a/arch/powerpc/include/asm/spinlock.h b/arch/powerpc/include/asm/spinlock.h index 8c1b913de6d7..c1b1ec94b06c 100644 --- a/arch/powerpc/include/asm/spinlock.h +++ b/arch/powerpc/include/asm/spinlock.h @@ -342,5 +342,8 @@ static inline void arch_write_unlock(arch_rwlock_t *rw) #define arch_read_relax(lock) __rw_yield(lock) #define arch_write_relax(lock) __rw_yield(lock) +/* See include/linux/spinlock.h */ +#define smp_mb__after_spinlock() smp_mb() + #endif /* __KERNEL__ */ #endif /* __ASM_SPINLOCK_H */ diff --git a/include/linux/atomic.h b/include/linux/atomic.h index c56be7410130..40d6bfec0e0d 100644 --- a/include/linux/atomic.h +++ b/include/linux/atomic.h @@ -38,6 +38,9 @@ * Besides, if an arch has a special barrier for acquire/release, it could * implement its own __atomic_op_* and use the same framework for building * variants + * + * If an architecture overrides __atomic_op_acquire() it will probably want + * to define smp_mb__after_spinlock(). */ #ifndef __atomic_op_acquire #define __atomic_op_acquire(op, args...) \ diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index d9510e8522d4..840281095933 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -130,6 +130,42 @@ do { \ #define smp_mb__before_spinlock() smp_wmb() #endif +/* + * This barrier must provide two things: + * + * - it must guarantee a STORE before the spin_lock() is ordered against a + * LOAD after it, see the comments at its two usage sites. + * + * - it must ensure the critical section is RCsc. + * + * The latter is important for cases where we observe values written by other + * CPUs in spin-loops, without barriers, while being subject to scheduling. + * + * CPU0 CPU1 CPU2 + * + * for (;;) { + * if (READ_ONCE(X)) + * break; + * } + * X=1 + * + * + * r = X; + * + * without transitivity it could be that CPU1 observes X!=0 breaks the loop, + * we get migrated and CPU2 sees X==0. + * + * Since most load-store architectures implement ACQUIRE with an smp_mb() after + * the LL/SC loop, they need no further barriers. Similarly all our TSO + * architectures imply an smp_mb() for each atomic instruction and equally don't + * need more. + * + * Architectures that can implement ACQUIRE better need to take care. + */ +#ifndef smp_mb__after_spinlock +#define smp_mb__after_spinlock() do { } while (0) +#endif + /** * raw_spin_unlock_wait - wait until the spinlock gets unlocked * @lock: the spinlock in question. diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 0869b20fba81..9fece583a1f0 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1967,8 +1967,8 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) * reordered with p->state check below. This pairs with mb() in * set_current_state() the waiting thread does. */ - smp_mb__before_spinlock(); raw_spin_lock_irqsave(&p->pi_lock, flags); + smp_mb__after_spinlock(); if (!(p->state & state)) goto out; @@ -3281,8 +3281,8 @@ static void __sched notrace __schedule(bool preempt) * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) * done by the caller to avoid the race with signal_wake_up(). */ - smp_mb__before_spinlock(); rq_lock(rq, &rf); + smp_mb__after_spinlock(); /* Promote REQ to ACT */ rq->clock_update_flags <<= 1; -- cgit v1.2.3 From a9668cd6ee288c4838bc668880ac085be551cac2 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 7 Jun 2017 17:51:27 +0200 Subject: locking: Remove smp_mb__before_spinlock() Now that there are no users of smp_mb__before_spinlock() left, remove it entirely. Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar --- Documentation/memory-barriers.txt | 5 +---- .../translations/ko_KR/memory-barriers.txt | 5 +---- arch/arm64/include/asm/spinlock.h | 9 -------- arch/powerpc/include/asm/barrier.h | 7 ------ fs/userfaultfd.c | 25 ++++++++++------------ include/linux/spinlock.h | 13 ----------- 6 files changed, 13 insertions(+), 51 deletions(-) (limited to 'include/linux') diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt index 9f34364922c8..d1d1716f904b 100644 --- a/Documentation/memory-barriers.txt +++ b/Documentation/memory-barriers.txt @@ -1981,10 +1981,7 @@ for each construct. These operations all imply certain barriers: ACQUIRE operation has completed. Memory operations issued before the ACQUIRE may be completed after - the ACQUIRE operation has completed. An smp_mb__before_spinlock(), - combined with a following ACQUIRE, orders prior stores against - subsequent loads and stores. Note that this is weaker than smp_mb()! - The smp_mb__before_spinlock() primitive is free on many architectures. + the ACQUIRE operation has completed. (2) RELEASE operation implication: diff --git a/Documentation/translations/ko_KR/memory-barriers.txt b/Documentation/translations/ko_KR/memory-barriers.txt index 38310dcd6620..bc80fc0e210f 100644 --- a/Documentation/translations/ko_KR/memory-barriers.txt +++ b/Documentation/translations/ko_KR/memory-barriers.txt @@ -1956,10 +1956,7 @@ MMIO 쓰기 배리어 뒤에 완료됩니다. ACQUIRE 앞에서 요청된 메모리 오퍼레이션은 ACQUIRE 오퍼레이션이 완료된 후에 - 완료될 수 있습니다. smp_mb__before_spinlock() 뒤에 ACQUIRE 가 실행되는 - 코드 블록은 블록 앞의 스토어를 블록 뒤의 로드와 스토어에 대해 순서 - 맞춥니다. 이건 smp_mb() 보다 완화된 것임을 기억하세요! 많은 아키텍쳐에서 - smp_mb__before_spinlock() 은 사실 아무일도 하지 않습니다. + 완료될 수 있습니다. (2) RELEASE 오퍼레이션의 영향: diff --git a/arch/arm64/include/asm/spinlock.h b/arch/arm64/include/asm/spinlock.h index b103888b694a..ae4241ab19a8 100644 --- a/arch/arm64/include/asm/spinlock.h +++ b/arch/arm64/include/asm/spinlock.h @@ -358,15 +358,6 @@ static inline int arch_read_trylock(arch_rwlock_t *rw) #define arch_read_relax(lock) cpu_relax() #define arch_write_relax(lock) cpu_relax() -/* - * Accesses appearing in program order before a spin_lock() operation - * can be reordered with accesses inside the critical section, by virtue - * of arch_spin_lock being constructed using acquire semantics. - * - * In cases where this is problematic (e.g. try_to_wake_up), an - * smp_mb__before_spinlock() can restore the required ordering. - */ -#define smp_mb__before_spinlock() smp_mb() /* See include/linux/spinlock.h */ #define smp_mb__after_spinlock() smp_mb() diff --git a/arch/powerpc/include/asm/barrier.h b/arch/powerpc/include/asm/barrier.h index 25d42bd3f114..9c601adfc500 100644 --- a/arch/powerpc/include/asm/barrier.h +++ b/arch/powerpc/include/asm/barrier.h @@ -74,13 +74,6 @@ do { \ ___p1; \ }) -/* - * This must resolve to hwsync on SMP for the context switch path. - * See _switch, and core scheduler context switch memory ordering - * comments. - */ -#define smp_mb__before_spinlock() smp_mb() - #include #endif /* _ASM_POWERPC_BARRIER_H */ diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 06ea26b8c996..44fcbefd84a2 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -109,27 +109,24 @@ static int userfaultfd_wake_function(wait_queue_entry_t *wq, unsigned mode, goto out; WRITE_ONCE(uwq->waken, true); /* - * The implicit smp_mb__before_spinlock in try_to_wake_up() - * renders uwq->waken visible to other CPUs before the task is - * waken. + * The Program-Order guarantees provided by the scheduler + * ensure uwq->waken is visible before the task is woken. */ ret = wake_up_state(wq->private, mode); - if (ret) + if (ret) { /* * Wake only once, autoremove behavior. * - * After the effect of list_del_init is visible to the - * other CPUs, the waitqueue may disappear from under - * us, see the !list_empty_careful() in - * handle_userfault(). try_to_wake_up() has an - * implicit smp_mb__before_spinlock, and the - * wq->private is read before calling the extern - * function "wake_up_state" (which in turns calls - * try_to_wake_up). While the spin_lock;spin_unlock; - * wouldn't be enough, the smp_mb__before_spinlock is - * enough to avoid an explicit smp_mb() here. + * After the effect of list_del_init is visible to the other + * CPUs, the waitqueue may disappear from under us, see the + * !list_empty_careful() in handle_userfault(). + * + * try_to_wake_up() has an implicit smp_mb(), and the + * wq->private is read before calling the extern function + * "wake_up_state" (which in turns calls try_to_wake_up). */ list_del_init(&wq->entry); + } out: return ret; } diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index 840281095933..4e8cce19b507 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -117,19 +117,6 @@ do { \ #endif /*arch_spin_is_contended*/ #endif -/* - * Despite its name it doesn't necessarily has to be a full barrier. - * It should only guarantee that a STORE before the critical section - * can not be reordered with LOADs and STOREs inside this section. - * spin_lock() is the one-way barrier, this LOAD can not escape out - * of the region. So the default implementation simply ensures that - * a STORE can not move into the critical section, smp_wmb() should - * serialize it with another STORE done by spin_lock(). - */ -#ifndef smp_mb__before_spinlock -#define smp_mb__before_spinlock() smp_wmb() -#endif - /* * This barrier must provide two things: * -- cgit v1.2.3 From d92a8cfcb37ecd1315269dab741f073b63b3a8b6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 3 Mar 2017 10:13:38 +0100 Subject: locking/lockdep: Rework FS_RECLAIM annotation A while ago someone, and I cannot find the email just now, asked if we could not implement the RECLAIM_FS inversion stuff with a 'fake' lock like we use for other things like workqueues etc. I think this should be possible which allows reducing the 'irq' states and will reduce the amount of __bfs() lookups we do. Removing the 1 IRQ state results in 4 less __bfs() walks per dependency, improving lockdep performance. And by moving this annotation out of the lockdep code it becomes easier for the mm people to extend. Signed-off-by: Peter Zijlstra (Intel) Cc: Byungchul Park Cc: Linus Torvalds Cc: Mel Gorman Cc: Michal Hocko Cc: Nikolay Borisov Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: akpm@linux-foundation.org Cc: boqun.feng@gmail.com Cc: iamjoonsoo.kim@lge.com Cc: kernel-team@lge.com Cc: kirill@shutemov.name Cc: npiggin@gmail.com Cc: walken@google.com Signed-off-by: Ingo Molnar --- drivers/gpu/drm/i915/i915_debugfs.c | 5 +- include/linux/lockdep.h | 11 +---- include/linux/sched.h | 1 - include/linux/sched/mm.h | 8 ++++ kernel/locking/lockdep.c | 95 +------------------------------------ kernel/locking/lockdep_states.h | 1 - mm/page_alloc.c | 49 +++++++++++++++++-- mm/slab.h | 6 ++- mm/slob.c | 6 ++- mm/vmscan.c | 13 +++-- 10 files changed, 75 insertions(+), 120 deletions(-) (limited to 'include/linux') diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 00d8967c8512..0511fce5c947 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -28,6 +28,7 @@ #include #include +#include #include "intel_drv.h" static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node) @@ -4331,7 +4332,7 @@ i915_drop_caches_set(void *data, u64 val) mutex_unlock(&dev->struct_mutex); } - lockdep_set_current_reclaim_state(GFP_KERNEL); + fs_reclaim_acquire(GFP_KERNEL); if (val & DROP_BOUND) i915_gem_shrink(dev_priv, LONG_MAX, I915_SHRINK_BOUND); @@ -4340,7 +4341,7 @@ i915_drop_caches_set(void *data, u64 val) if (val & DROP_SHRINK_ALL) i915_gem_shrink_all(dev_priv); - lockdep_clear_current_reclaim_state(); + fs_reclaim_release(GFP_KERNEL); if (val & DROP_FREED) { synchronize_rcu(); diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index fffe49f188e6..0a4c02c2d7a2 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -29,7 +29,7 @@ extern int lock_stat; * We'd rather not expose kernel/lockdep_states.h this wide, but we do need * the total number of states... :-( */ -#define XXX_LOCK_USAGE_STATES (1+3*4) +#define XXX_LOCK_USAGE_STATES (1+2*4) /* * NR_LOCKDEP_CACHING_CLASSES ... Number of classes @@ -363,10 +363,6 @@ static inline void lock_set_subclass(struct lockdep_map *lock, extern void lock_downgrade(struct lockdep_map *lock, unsigned long ip); -extern void lockdep_set_current_reclaim_state(gfp_t gfp_mask); -extern void lockdep_clear_current_reclaim_state(void); -extern void lockdep_trace_alloc(gfp_t mask); - struct pin_cookie { unsigned int val; }; #define NIL_COOKIE (struct pin_cookie){ .val = 0U, } @@ -375,7 +371,7 @@ extern struct pin_cookie lock_pin_lock(struct lockdep_map *lock); extern void lock_repin_lock(struct lockdep_map *lock, struct pin_cookie); extern void lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie); -# define INIT_LOCKDEP .lockdep_recursion = 0, .lockdep_reclaim_gfp = 0, +# define INIT_LOCKDEP .lockdep_recursion = 0, #define lockdep_depth(tsk) (debug_locks ? (tsk)->lockdep_depth : 0) @@ -416,9 +412,6 @@ static inline void lockdep_on(void) # define lock_downgrade(l, i) do { } while (0) # define lock_set_class(l, n, k, s, i) do { } while (0) # define lock_set_subclass(l, s, i) do { } while (0) -# define lockdep_set_current_reclaim_state(g) do { } while (0) -# define lockdep_clear_current_reclaim_state() do { } while (0) -# define lockdep_trace_alloc(g) do { } while (0) # define lockdep_info() do { } while (0) # define lockdep_init_map(lock, name, key, sub) \ do { (void)(name); (void)(key); } while (0) diff --git a/include/linux/sched.h b/include/linux/sched.h index 8337e2db0bb2..57db70ec70d0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -846,7 +846,6 @@ struct task_struct { int lockdep_depth; unsigned int lockdep_recursion; struct held_lock held_locks[MAX_LOCK_DEPTH]; - gfp_t lockdep_reclaim_gfp; #endif #ifdef CONFIG_UBSAN diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h index 2b24a6974847..2b0a281f9d26 100644 --- a/include/linux/sched/mm.h +++ b/include/linux/sched/mm.h @@ -167,6 +167,14 @@ static inline gfp_t current_gfp_context(gfp_t flags) return flags; } +#ifdef CONFIG_LOCKDEP +extern void fs_reclaim_acquire(gfp_t gfp_mask); +extern void fs_reclaim_release(gfp_t gfp_mask); +#else +static inline void fs_reclaim_acquire(gfp_t gfp_mask) { } +static inline void fs_reclaim_release(gfp_t gfp_mask) { } +#endif + static inline unsigned int memalloc_noio_save(void) { unsigned int flags = current->flags & PF_MEMALLOC_NOIO; diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 7d2499bec5fe..986f2fa79dbb 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -344,14 +344,12 @@ EXPORT_SYMBOL(lockdep_on); #if VERBOSE # define HARDIRQ_VERBOSE 1 # define SOFTIRQ_VERBOSE 1 -# define RECLAIM_VERBOSE 1 #else # define HARDIRQ_VERBOSE 0 # define SOFTIRQ_VERBOSE 0 -# define RECLAIM_VERBOSE 0 #endif -#if VERBOSE || HARDIRQ_VERBOSE || SOFTIRQ_VERBOSE || RECLAIM_VERBOSE +#if VERBOSE || HARDIRQ_VERBOSE || SOFTIRQ_VERBOSE /* * Quick filtering for interesting events: */ @@ -2567,14 +2565,6 @@ static int SOFTIRQ_verbose(struct lock_class *class) return 0; } -static int RECLAIM_FS_verbose(struct lock_class *class) -{ -#if RECLAIM_VERBOSE - return class_filter(class); -#endif - return 0; -} - #define STRICT_READ_CHECKS 1 static int (*state_verbose_f[])(struct lock_class *class) = { @@ -2870,57 +2860,6 @@ void trace_softirqs_off(unsigned long ip) debug_atomic_inc(redundant_softirqs_off); } -static void __lockdep_trace_alloc(gfp_t gfp_mask, unsigned long flags) -{ - struct task_struct *curr = current; - - if (unlikely(!debug_locks)) - return; - - gfp_mask = current_gfp_context(gfp_mask); - - /* no reclaim without waiting on it */ - if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) - return; - - /* this guy won't enter reclaim */ - if ((curr->flags & PF_MEMALLOC) && !(gfp_mask & __GFP_NOMEMALLOC)) - return; - - /* We're only interested __GFP_FS allocations for now */ - if (!(gfp_mask & __GFP_FS) || (curr->flags & PF_MEMALLOC_NOFS)) - return; - - /* - * Oi! Can't be having __GFP_FS allocations with IRQs disabled. - */ - if (DEBUG_LOCKS_WARN_ON(irqs_disabled_flags(flags))) - return; - - /* Disable lockdep if explicitly requested */ - if (gfp_mask & __GFP_NOLOCKDEP) - return; - - mark_held_locks(curr, RECLAIM_FS); -} - -static void check_flags(unsigned long flags); - -void lockdep_trace_alloc(gfp_t gfp_mask) -{ - unsigned long flags; - - if (unlikely(current->lockdep_recursion)) - return; - - raw_local_irq_save(flags); - check_flags(flags); - current->lockdep_recursion = 1; - __lockdep_trace_alloc(gfp_mask, flags); - current->lockdep_recursion = 0; - raw_local_irq_restore(flags); -} - static int mark_irqflags(struct task_struct *curr, struct held_lock *hlock) { /* @@ -2966,22 +2905,6 @@ static int mark_irqflags(struct task_struct *curr, struct held_lock *hlock) } } - /* - * We reuse the irq context infrastructure more broadly as a general - * context checking code. This tests GFP_FS recursion (a lock taken - * during reclaim for a GFP_FS allocation is held over a GFP_FS - * allocation). - */ - if (!hlock->trylock && (curr->lockdep_reclaim_gfp & __GFP_FS)) { - if (hlock->read) { - if (!mark_lock(curr, hlock, LOCK_USED_IN_RECLAIM_FS_READ)) - return 0; - } else { - if (!mark_lock(curr, hlock, LOCK_USED_IN_RECLAIM_FS)) - return 0; - } - } - return 1; } @@ -3040,10 +2963,6 @@ static inline int separate_irq_context(struct task_struct *curr, return 0; } -void lockdep_trace_alloc(gfp_t gfp_mask) -{ -} - #endif /* defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_PROVE_LOCKING) */ /* @@ -3952,18 +3871,6 @@ void lock_unpin_lock(struct lockdep_map *lock, struct pin_cookie cookie) } EXPORT_SYMBOL_GPL(lock_unpin_lock); -void lockdep_set_current_reclaim_state(gfp_t gfp_mask) -{ - current->lockdep_reclaim_gfp = current_gfp_context(gfp_mask); -} -EXPORT_SYMBOL_GPL(lockdep_set_current_reclaim_state); - -void lockdep_clear_current_reclaim_state(void) -{ - current->lockdep_reclaim_gfp = 0; -} -EXPORT_SYMBOL_GPL(lockdep_clear_current_reclaim_state); - #ifdef CONFIG_LOCK_STAT static int print_lock_contention_bug(struct task_struct *curr, struct lockdep_map *lock, diff --git a/kernel/locking/lockdep_states.h b/kernel/locking/lockdep_states.h index 995b0cc2b84c..35ca09f2ed0b 100644 --- a/kernel/locking/lockdep_states.h +++ b/kernel/locking/lockdep_states.h @@ -6,4 +6,3 @@ */ LOCKDEP_STATE(HARDIRQ) LOCKDEP_STATE(SOFTIRQ) -LOCKDEP_STATE(RECLAIM_FS) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index fc32aa81f359..c20d89601802 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -66,6 +66,7 @@ #include #include #include +#include #include #include @@ -3490,6 +3491,47 @@ should_compact_retry(struct alloc_context *ac, unsigned int order, int alloc_fla } #endif /* CONFIG_COMPACTION */ +#ifdef CONFIG_LOCKDEP +struct lockdep_map __fs_reclaim_map = + STATIC_LOCKDEP_MAP_INIT("fs_reclaim", &__fs_reclaim_map); + +static bool __need_fs_reclaim(gfp_t gfp_mask) +{ + gfp_mask = current_gfp_context(gfp_mask); + + /* no reclaim without waiting on it */ + if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) + return false; + + /* this guy won't enter reclaim */ + if ((current->flags & PF_MEMALLOC) && !(gfp_mask & __GFP_NOMEMALLOC)) + return false; + + /* We're only interested __GFP_FS allocations for now */ + if (!(gfp_mask & __GFP_FS)) + return false; + + if (gfp_mask & __GFP_NOLOCKDEP) + return false; + + return true; +} + +void fs_reclaim_acquire(gfp_t gfp_mask) +{ + if (__need_fs_reclaim(gfp_mask)) + lock_map_acquire(&__fs_reclaim_map); +} +EXPORT_SYMBOL_GPL(fs_reclaim_acquire); + +void fs_reclaim_release(gfp_t gfp_mask) +{ + if (__need_fs_reclaim(gfp_mask)) + lock_map_release(&__fs_reclaim_map); +} +EXPORT_SYMBOL_GPL(fs_reclaim_release); +#endif + /* Perform direct synchronous page reclaim */ static int __perform_reclaim(gfp_t gfp_mask, unsigned int order, @@ -3504,7 +3546,7 @@ __perform_reclaim(gfp_t gfp_mask, unsigned int order, /* We now go into synchronous reclaim */ cpuset_memory_pressure_bump(); noreclaim_flag = memalloc_noreclaim_save(); - lockdep_set_current_reclaim_state(gfp_mask); + fs_reclaim_acquire(gfp_mask); reclaim_state.reclaimed_slab = 0; current->reclaim_state = &reclaim_state; @@ -3512,7 +3554,7 @@ __perform_reclaim(gfp_t gfp_mask, unsigned int order, ac->nodemask); current->reclaim_state = NULL; - lockdep_clear_current_reclaim_state(); + fs_reclaim_release(gfp_mask); memalloc_noreclaim_restore(noreclaim_flag); cond_resched(); @@ -4041,7 +4083,8 @@ static inline bool prepare_alloc_pages(gfp_t gfp_mask, unsigned int order, *alloc_flags |= ALLOC_CPUSET; } - lockdep_trace_alloc(gfp_mask); + fs_reclaim_acquire(gfp_mask); + fs_reclaim_release(gfp_mask); might_sleep_if(gfp_mask & __GFP_DIRECT_RECLAIM); diff --git a/mm/slab.h b/mm/slab.h index 6885e1192ec5..073362816acc 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -43,6 +43,7 @@ struct kmem_cache { #include #include #include +#include /* * State of the slab allocator. @@ -412,7 +413,10 @@ static inline struct kmem_cache *slab_pre_alloc_hook(struct kmem_cache *s, gfp_t flags) { flags &= gfp_allowed_mask; - lockdep_trace_alloc(flags); + + fs_reclaim_acquire(flags); + fs_reclaim_release(flags); + might_sleep_if(gfpflags_allow_blocking(flags)); if (should_failslab(s, flags)) diff --git a/mm/slob.c b/mm/slob.c index 1bae78d71096..a8bd6fa11a66 100644 --- a/mm/slob.c +++ b/mm/slob.c @@ -432,7 +432,8 @@ __do_kmalloc_node(size_t size, gfp_t gfp, int node, unsigned long caller) gfp &= gfp_allowed_mask; - lockdep_trace_alloc(gfp); + fs_reclaim_acquire(gfp); + fs_reclaim_release(gfp); if (size < PAGE_SIZE - align) { if (!size) @@ -538,7 +539,8 @@ static void *slob_alloc_node(struct kmem_cache *c, gfp_t flags, int node) flags &= gfp_allowed_mask; - lockdep_trace_alloc(flags); + fs_reclaim_acquire(flags); + fs_reclaim_release(flags); if (c->size < PAGE_SIZE) { b = slob_alloc(c->size, flags, c->align, node); diff --git a/mm/vmscan.c b/mm/vmscan.c index a1af041930a6..f957afe900ec 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -3525,8 +3525,6 @@ static int kswapd(void *p) }; const struct cpumask *cpumask = cpumask_of_node(pgdat->node_id); - lockdep_set_current_reclaim_state(GFP_KERNEL); - if (!cpumask_empty(cpumask)) set_cpus_allowed_ptr(tsk, cpumask); current->reclaim_state = &reclaim_state; @@ -3585,14 +3583,15 @@ kswapd_try_sleep: */ trace_mm_vmscan_kswapd_wake(pgdat->node_id, classzone_idx, alloc_order); + fs_reclaim_acquire(GFP_KERNEL); reclaim_order = balance_pgdat(pgdat, alloc_order, classzone_idx); + fs_reclaim_release(GFP_KERNEL); if (reclaim_order < alloc_order) goto kswapd_try_sleep; } tsk->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD); current->reclaim_state = NULL; - lockdep_clear_current_reclaim_state(); return 0; } @@ -3655,14 +3654,14 @@ unsigned long shrink_all_memory(unsigned long nr_to_reclaim) unsigned int noreclaim_flag; noreclaim_flag = memalloc_noreclaim_save(); - lockdep_set_current_reclaim_state(sc.gfp_mask); + fs_reclaim_acquire(sc.gfp_mask); reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; nr_reclaimed = do_try_to_free_pages(zonelist, &sc); p->reclaim_state = NULL; - lockdep_clear_current_reclaim_state(); + fs_reclaim_release(sc.gfp_mask); memalloc_noreclaim_restore(noreclaim_flag); return nr_reclaimed; @@ -3847,7 +3846,7 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in */ noreclaim_flag = memalloc_noreclaim_save(); p->flags |= PF_SWAPWRITE; - lockdep_set_current_reclaim_state(sc.gfp_mask); + fs_reclaim_acquire(sc.gfp_mask); reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; @@ -3862,9 +3861,9 @@ static int __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned in } p->reclaim_state = NULL; + fs_reclaim_release(gfp_mask); current->flags &= ~PF_SWAPWRITE; memalloc_noreclaim_restore(noreclaim_flag); - lockdep_clear_current_reclaim_state(); return sc.nr_reclaimed >= nr_pages; } -- cgit v1.2.3 From b09be676e0ff25bd6d2e7637e26d349f9109ad75 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Mon, 7 Aug 2017 16:12:52 +0900 Subject: locking/lockdep: Implement the 'crossrelease' feature Lockdep is a runtime locking correctness validator that detects and reports a deadlock or its possibility by checking dependencies between locks. It's useful since it does not report just an actual deadlock but also the possibility of a deadlock that has not actually happened yet. That enables problems to be fixed before they affect real systems. However, this facility is only applicable to typical locks, such as spinlocks and mutexes, which are normally released within the context in which they were acquired. However, synchronization primitives like page locks or completions, which are allowed to be released in any context, also create dependencies and can cause a deadlock. So lockdep should track these locks to do a better job. The 'crossrelease' implementation makes these primitives also be tracked. Signed-off-by: Byungchul Park Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: akpm@linux-foundation.org Cc: boqun.feng@gmail.com Cc: kernel-team@lge.com Cc: kirill@shutemov.name Cc: npiggin@gmail.com Cc: walken@google.com Cc: willy@infradead.org Link: http://lkml.kernel.org/r/1502089981-21272-6-git-send-email-byungchul.park@lge.com Signed-off-by: Ingo Molnar --- include/linux/irqflags.h | 24 ++- include/linux/lockdep.h | 110 +++++++++- include/linux/sched.h | 8 + kernel/exit.c | 1 + kernel/fork.c | 4 + kernel/locking/lockdep.c | 508 ++++++++++++++++++++++++++++++++++++++++++++--- kernel/workqueue.c | 2 + lib/Kconfig.debug | 12 ++ 8 files changed, 635 insertions(+), 34 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 5dd1272d1ab2..5fdd93bb9300 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -23,10 +23,26 @@ # define trace_softirq_context(p) ((p)->softirq_context) # define trace_hardirqs_enabled(p) ((p)->hardirqs_enabled) # define trace_softirqs_enabled(p) ((p)->softirqs_enabled) -# define trace_hardirq_enter() do { current->hardirq_context++; } while (0) -# define trace_hardirq_exit() do { current->hardirq_context--; } while (0) -# define lockdep_softirq_enter() do { current->softirq_context++; } while (0) -# define lockdep_softirq_exit() do { current->softirq_context--; } while (0) +# define trace_hardirq_enter() \ +do { \ + current->hardirq_context++; \ + crossrelease_hist_start(XHLOCK_HARD); \ +} while (0) +# define trace_hardirq_exit() \ +do { \ + current->hardirq_context--; \ + crossrelease_hist_end(XHLOCK_HARD); \ +} while (0) +# define lockdep_softirq_enter() \ +do { \ + current->softirq_context++; \ + crossrelease_hist_start(XHLOCK_SOFT); \ +} while (0) +# define lockdep_softirq_exit() \ +do { \ + current->softirq_context--; \ + crossrelease_hist_end(XHLOCK_SOFT); \ +} while (0) # define INIT_TRACE_IRQFLAGS .softirqs_enabled = 1, #else # define trace_hardirqs_on() do { } while (0) diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 0a4c02c2d7a2..e1e0fcd99613 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -155,6 +155,12 @@ struct lockdep_map { int cpu; unsigned long ip; #endif +#ifdef CONFIG_LOCKDEP_CROSSRELEASE + /* + * Whether it's a crosslock. + */ + int cross; +#endif }; static inline void lockdep_copy_map(struct lockdep_map *to, @@ -258,8 +264,62 @@ struct held_lock { unsigned int hardirqs_off:1; unsigned int references:12; /* 32 bits */ unsigned int pin_count; +#ifdef CONFIG_LOCKDEP_CROSSRELEASE + /* + * Generation id. + * + * A value of cross_gen_id will be stored when holding this, + * which is globally increased whenever each crosslock is held. + */ + unsigned int gen_id; +#endif +}; + +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +#define MAX_XHLOCK_TRACE_ENTRIES 5 + +/* + * This is for keeping locks waiting for commit so that true dependencies + * can be added at commit step. + */ +struct hist_lock { + /* + * Seperate stack_trace data. This will be used at commit step. + */ + struct stack_trace trace; + unsigned long trace_entries[MAX_XHLOCK_TRACE_ENTRIES]; + + /* + * Seperate hlock instance. This will be used at commit step. + * + * TODO: Use a smaller data structure containing only necessary + * data. However, we should make lockdep code able to handle the + * smaller one first. + */ + struct held_lock hlock; +}; + +/* + * To initialize a lock as crosslock, lockdep_init_map_crosslock() should + * be called instead of lockdep_init_map(). + */ +struct cross_lock { + /* + * Seperate hlock instance. This will be used at commit step. + * + * TODO: Use a smaller data structure containing only necessary + * data. However, we should make lockdep code able to handle the + * smaller one first. + */ + struct held_lock hlock; }; +struct lockdep_map_cross { + struct lockdep_map map; + struct cross_lock xlock; +}; +#endif + /* * Initialization, self-test and debugging-output methods: */ @@ -281,13 +341,6 @@ extern void lockdep_on(void); extern void lockdep_init_map(struct lockdep_map *lock, const char *name, struct lock_class_key *key, int subclass); -/* - * To initialize a lockdep_map statically use this macro. - * Note that _name must not be NULL. - */ -#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ - { .name = (_name), .key = (void *)(_key), } - /* * Reinitialize a lock key - for cases where there is special locking or * special initialization of locks so that the validator gets the scope @@ -460,6 +513,49 @@ struct pin_cookie { }; #endif /* !LOCKDEP */ +enum xhlock_context_t { + XHLOCK_HARD, + XHLOCK_SOFT, + XHLOCK_PROC, + XHLOCK_CTX_NR, +}; + +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +extern void lockdep_init_map_crosslock(struct lockdep_map *lock, + const char *name, + struct lock_class_key *key, + int subclass); +extern void lock_commit_crosslock(struct lockdep_map *lock); + +#define STATIC_CROSS_LOCKDEP_MAP_INIT(_name, _key) \ + { .map.name = (_name), .map.key = (void *)(_key), \ + .map.cross = 1, } + +/* + * To initialize a lockdep_map statically use this macro. + * Note that _name must not be NULL. + */ +#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ + { .name = (_name), .key = (void *)(_key), .cross = 0, } + +extern void crossrelease_hist_start(enum xhlock_context_t c); +extern void crossrelease_hist_end(enum xhlock_context_t c); +extern void lockdep_init_task(struct task_struct *task); +extern void lockdep_free_task(struct task_struct *task); +#else +/* + * To initialize a lockdep_map statically use this macro. + * Note that _name must not be NULL. + */ +#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ + { .name = (_name), .key = (void *)(_key), } + +static inline void crossrelease_hist_start(enum xhlock_context_t c) {} +static inline void crossrelease_hist_end(enum xhlock_context_t c) {} +static inline void lockdep_init_task(struct task_struct *task) {} +static inline void lockdep_free_task(struct task_struct *task) {} +#endif + #ifdef CONFIG_LOCK_STAT extern void lock_contended(struct lockdep_map *lock, unsigned long ip); diff --git a/include/linux/sched.h b/include/linux/sched.h index 57db70ec70d0..5235fba537fc 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -848,6 +848,14 @@ struct task_struct { struct held_lock held_locks[MAX_LOCK_DEPTH]; #endif +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +#define MAX_XHLOCKS_NR 64UL + struct hist_lock *xhlocks; /* Crossrelease history locks */ + unsigned int xhlock_idx; + /* For restoring at history boundaries */ + unsigned int xhlock_idx_hist[XHLOCK_CTX_NR]; +#endif + #ifdef CONFIG_UBSAN unsigned int in_ubsan; #endif diff --git a/kernel/exit.c b/kernel/exit.c index c5548faa9f37..fa72d57db747 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -920,6 +920,7 @@ void __noreturn do_exit(long code) exit_rcu(); TASKS_RCU(__srcu_read_unlock(&tasks_rcu_exit_srcu, tasks_rcu_i)); + lockdep_free_task(tsk); do_task_dead(); } EXPORT_SYMBOL_GPL(do_exit); diff --git a/kernel/fork.c b/kernel/fork.c index 17921b0390b4..cbf2221ee81a 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -484,6 +484,8 @@ void __init fork_init(void) cpuhp_setup_state(CPUHP_BP_PREPARE_DYN, "fork:vm_stack_cache", NULL, free_vm_stack_cache); #endif + + lockdep_init_task(&init_task); } int __weak arch_dup_task_struct(struct task_struct *dst, @@ -1691,6 +1693,7 @@ static __latent_entropy struct task_struct *copy_process( p->lockdep_depth = 0; /* no locks held yet */ p->curr_chain_key = 0; p->lockdep_recursion = 0; + lockdep_init_task(p); #endif #ifdef CONFIG_DEBUG_MUTEXES @@ -1949,6 +1952,7 @@ bad_fork_cleanup_audit: bad_fork_cleanup_perf: perf_event_free_task(p); bad_fork_cleanup_policy: + lockdep_free_task(p); #ifdef CONFIG_NUMA mpol_put(p->mempolicy); bad_fork_cleanup_threadgroup_lock: diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 841828ba35b9..56f69cc53ddc 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -58,6 +58,10 @@ #define CREATE_TRACE_POINTS #include +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +#include +#endif + #ifdef CONFIG_PROVE_LOCKING int prove_locking = 1; module_param(prove_locking, int, 0644); @@ -724,6 +728,18 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass) return is_static || static_obj(lock->key) ? NULL : ERR_PTR(-EINVAL); } +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +static void cross_init(struct lockdep_map *lock, int cross); +static int cross_lock(struct lockdep_map *lock); +static int lock_acquire_crosslock(struct held_lock *hlock); +static int lock_release_crosslock(struct lockdep_map *lock); +#else +static inline void cross_init(struct lockdep_map *lock, int cross) {} +static inline int cross_lock(struct lockdep_map *lock) { return 0; } +static inline int lock_acquire_crosslock(struct held_lock *hlock) { return 2; } +static inline int lock_release_crosslock(struct lockdep_map *lock) { return 2; } +#endif + /* * Register a lock's class in the hash-table, if the class is not present * yet. Otherwise we look it up. We cache the result in the lock object @@ -1795,6 +1811,9 @@ check_deadlock(struct task_struct *curr, struct held_lock *next, if (nest) return 2; + if (cross_lock(prev->instance)) + continue; + return print_deadlock_bug(curr, prev, next); } return 1; @@ -1962,30 +1981,36 @@ check_prevs_add(struct task_struct *curr, struct held_lock *next) int distance = curr->lockdep_depth - depth + 1; hlock = curr->held_locks + depth - 1; /* - * Only non-recursive-read entries get new dependencies - * added: + * Only non-crosslock entries get new dependencies added. + * Crosslock entries will be added by commit later: */ - if (hlock->read != 2 && hlock->check) { - int ret = check_prev_add(curr, hlock, next, - distance, &trace, save); - if (!ret) - return 0; - + if (!cross_lock(hlock->instance)) { /* - * Stop saving stack_trace if save_trace() was - * called at least once: + * Only non-recursive-read entries get new dependencies + * added: */ - if (save && ret == 2) - save = NULL; + if (hlock->read != 2 && hlock->check) { + int ret = check_prev_add(curr, hlock, next, + distance, &trace, save); + if (!ret) + return 0; - /* - * Stop after the first non-trylock entry, - * as non-trylock entries have added their - * own direct dependencies already, so this - * lock is connected to them indirectly: - */ - if (!hlock->trylock) - break; + /* + * Stop saving stack_trace if save_trace() was + * called at least once: + */ + if (save && ret == 2) + save = NULL; + + /* + * Stop after the first non-trylock entry, + * as non-trylock entries have added their + * own direct dependencies already, so this + * lock is connected to them indirectly: + */ + if (!hlock->trylock) + break; + } } depth--; /* @@ -3176,7 +3201,7 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, /* * Initialize a lock instance's lock-class mapping info: */ -void lockdep_init_map(struct lockdep_map *lock, const char *name, +static void __lockdep_init_map(struct lockdep_map *lock, const char *name, struct lock_class_key *key, int subclass) { int i; @@ -3234,8 +3259,25 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, raw_local_irq_restore(flags); } } + +void lockdep_init_map(struct lockdep_map *lock, const char *name, + struct lock_class_key *key, int subclass) +{ + cross_init(lock, 0); + __lockdep_init_map(lock, name, key, subclass); +} EXPORT_SYMBOL_GPL(lockdep_init_map); +#ifdef CONFIG_LOCKDEP_CROSSRELEASE +void lockdep_init_map_crosslock(struct lockdep_map *lock, const char *name, + struct lock_class_key *key, int subclass) +{ + cross_init(lock, 1); + __lockdep_init_map(lock, name, key, subclass); +} +EXPORT_SYMBOL_GPL(lockdep_init_map_crosslock); +#endif + struct lock_class_key __lockdep_no_validate__; EXPORT_SYMBOL_GPL(__lockdep_no_validate__); @@ -3291,6 +3333,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int chain_head = 0; int class_idx; u64 chain_key; + int ret; if (unlikely(!debug_locks)) return 0; @@ -3339,7 +3382,8 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, class_idx = class - lock_classes + 1; - if (depth) { + /* TODO: nest_lock is not implemented for crosslock yet. */ + if (depth && !cross_lock(lock)) { hlock = curr->held_locks + depth - 1; if (hlock->class_idx == class_idx && nest_lock) { if (hlock->references) { @@ -3427,6 +3471,14 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, if (!validate_chain(curr, lock, hlock, chain_head, chain_key)) return 0; + ret = lock_acquire_crosslock(hlock); + /* + * 2 means normal acquire operations are needed. Otherwise, it's + * ok just to return with '0:fail, 1:success'. + */ + if (ret != 2) + return ret; + curr->curr_chain_key = chain_key; curr->lockdep_depth++; check_chain_key(curr); @@ -3664,11 +3716,19 @@ __lock_release(struct lockdep_map *lock, int nested, unsigned long ip) struct task_struct *curr = current; struct held_lock *hlock; unsigned int depth; - int i; + int ret, i; if (unlikely(!debug_locks)) return 0; + ret = lock_release_crosslock(lock); + /* + * 2 means normal release operations are needed. Otherwise, it's + * ok just to return with '0:fail, 1:success'. + */ + if (ret != 2) + return ret; + depth = curr->lockdep_depth; /* * So we're all set to release this lock.. wait what lock? We don't @@ -4532,6 +4592,13 @@ asmlinkage __visible void lockdep_sys_exit(void) curr->comm, curr->pid); lockdep_print_held_locks(curr); } + + /* + * The lock history for each syscall should be independent. So wipe the + * slate clean on return to userspace. + */ + crossrelease_hist_end(XHLOCK_PROC); + crossrelease_hist_start(XHLOCK_PROC); } void lockdep_rcu_suspicious(const char *file, const int line, const char *s) @@ -4580,3 +4647,398 @@ void lockdep_rcu_suspicious(const char *file, const int line, const char *s) dump_stack(); } EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious); + +#ifdef CONFIG_LOCKDEP_CROSSRELEASE + +/* + * Crossrelease works by recording a lock history for each thread and + * connecting those historic locks that were taken after the + * wait_for_completion() in the complete() context. + * + * Task-A Task-B + * + * mutex_lock(&A); + * mutex_unlock(&A); + * + * wait_for_completion(&C); + * lock_acquire_crosslock(); + * atomic_inc_return(&cross_gen_id); + * | + * | mutex_lock(&B); + * | mutex_unlock(&B); + * | + * | complete(&C); + * `-- lock_commit_crosslock(); + * + * Which will then add a dependency between B and C. + */ + +#define xhlock(i) (current->xhlocks[(i) % MAX_XHLOCKS_NR]) + +/* + * Whenever a crosslock is held, cross_gen_id will be increased. + */ +static atomic_t cross_gen_id; /* Can be wrapped */ + +/* + * Lock history stacks; we have 3 nested lock history stacks: + * + * Hard IRQ + * Soft IRQ + * History / Task + * + * The thing is that once we complete a (Hard/Soft) IRQ the future task locks + * should not depend on any of the locks observed while running the IRQ. + * + * So what we do is rewind the history buffer and erase all our knowledge of + * that temporal event. + */ + +/* + * We need this to annotate lock history boundaries. Take for instance + * workqueues; each work is independent of the last. The completion of a future + * work does not depend on the completion of a past work (in general). + * Therefore we must not carry that (lock) dependency across works. + * + * This is true for many things; pretty much all kthreads fall into this + * pattern, where they have an 'idle' state and future completions do not + * depend on past completions. Its just that since they all have the 'same' + * form -- the kthread does the same over and over -- it doesn't typically + * matter. + * + * The same is true for system-calls, once a system call is completed (we've + * returned to userspace) the next system call does not depend on the lock + * history of the previous system call. + */ +void crossrelease_hist_start(enum xhlock_context_t c) +{ + if (current->xhlocks) + current->xhlock_idx_hist[c] = current->xhlock_idx; +} + +void crossrelease_hist_end(enum xhlock_context_t c) +{ + if (current->xhlocks) + current->xhlock_idx = current->xhlock_idx_hist[c]; +} + +static int cross_lock(struct lockdep_map *lock) +{ + return lock ? lock->cross : 0; +} + +/* + * This is needed to decide the relationship between wrapable variables. + */ +static inline int before(unsigned int a, unsigned int b) +{ + return (int)(a - b) < 0; +} + +static inline struct lock_class *xhlock_class(struct hist_lock *xhlock) +{ + return hlock_class(&xhlock->hlock); +} + +static inline struct lock_class *xlock_class(struct cross_lock *xlock) +{ + return hlock_class(&xlock->hlock); +} + +/* + * Should we check a dependency with previous one? + */ +static inline int depend_before(struct held_lock *hlock) +{ + return hlock->read != 2 && hlock->check && !hlock->trylock; +} + +/* + * Should we check a dependency with next one? + */ +static inline int depend_after(struct held_lock *hlock) +{ + return hlock->read != 2 && hlock->check; +} + +/* + * Check if the xhlock is valid, which would be false if, + * + * 1. Has not used after initializaion yet. + * + * Remind hist_lock is implemented as a ring buffer. + */ +static inline int xhlock_valid(struct hist_lock *xhlock) +{ + /* + * xhlock->hlock.instance must be !NULL. + */ + return !!xhlock->hlock.instance; +} + +/* + * Record a hist_lock entry. + * + * Irq disable is only required. + */ +static void add_xhlock(struct held_lock *hlock) +{ + unsigned int idx = ++current->xhlock_idx; + struct hist_lock *xhlock = &xhlock(idx); + +#ifdef CONFIG_DEBUG_LOCKDEP + /* + * This can be done locklessly because they are all task-local + * state, we must however ensure IRQs are disabled. + */ + WARN_ON_ONCE(!irqs_disabled()); +#endif + + /* Initialize hist_lock's members */ + xhlock->hlock = *hlock; + + xhlock->trace.nr_entries = 0; + xhlock->trace.max_entries = MAX_XHLOCK_TRACE_ENTRIES; + xhlock->trace.entries = xhlock->trace_entries; + xhlock->trace.skip = 3; + save_stack_trace(&xhlock->trace); +} + +static inline int same_context_xhlock(struct hist_lock *xhlock) +{ + return xhlock->hlock.irq_context == task_irq_context(current); +} + +/* + * This should be lockless as far as possible because this would be + * called very frequently. + */ +static void check_add_xhlock(struct held_lock *hlock) +{ + /* + * Record a hist_lock, only in case that acquisitions ahead + * could depend on the held_lock. For example, if the held_lock + * is trylock then acquisitions ahead never depends on that. + * In that case, we don't need to record it. Just return. + */ + if (!current->xhlocks || !depend_before(hlock)) + return; + + add_xhlock(hlock); +} + +/* + * For crosslock. + */ +static int add_xlock(struct held_lock *hlock) +{ + struct cross_lock *xlock; + unsigned int gen_id; + + if (!graph_lock()) + return 0; + + xlock = &((struct lockdep_map_cross *)hlock->instance)->xlock; + + gen_id = (unsigned int)atomic_inc_return(&cross_gen_id); + xlock->hlock = *hlock; + xlock->hlock.gen_id = gen_id; + graph_unlock(); + + return 1; +} + +/* + * Called for both normal and crosslock acquires. Normal locks will be + * pushed on the hist_lock queue. Cross locks will record state and + * stop regular lock_acquire() to avoid being placed on the held_lock + * stack. + * + * Return: 0 - failure; + * 1 - crosslock, done; + * 2 - normal lock, continue to held_lock[] ops. + */ +static int lock_acquire_crosslock(struct held_lock *hlock) +{ + /* + * CONTEXT 1 CONTEXT 2 + * --------- --------- + * lock A (cross) + * X = atomic_inc_return(&cross_gen_id) + * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + * Y = atomic_read_acquire(&cross_gen_id) + * lock B + * + * atomic_read_acquire() is for ordering between A and B, + * IOW, A happens before B, when CONTEXT 2 see Y >= X. + * + * Pairs with atomic_inc_return() in add_xlock(). + */ + hlock->gen_id = (unsigned int)atomic_read_acquire(&cross_gen_id); + + if (cross_lock(hlock->instance)) + return add_xlock(hlock); + + check_add_xhlock(hlock); + return 2; +} + +static int copy_trace(struct stack_trace *trace) +{ + unsigned long *buf = stack_trace + nr_stack_trace_entries; + unsigned int max_nr = MAX_STACK_TRACE_ENTRIES - nr_stack_trace_entries; + unsigned int nr = min(max_nr, trace->nr_entries); + + trace->nr_entries = nr; + memcpy(buf, trace->entries, nr * sizeof(trace->entries[0])); + trace->entries = buf; + nr_stack_trace_entries += nr; + + if (nr_stack_trace_entries >= MAX_STACK_TRACE_ENTRIES-1) { + if (!debug_locks_off_graph_unlock()) + return 0; + + print_lockdep_off("BUG: MAX_STACK_TRACE_ENTRIES too low!"); + dump_stack(); + + return 0; + } + + return 1; +} + +static int commit_xhlock(struct cross_lock *xlock, struct hist_lock *xhlock) +{ + unsigned int xid, pid; + u64 chain_key; + + xid = xlock_class(xlock) - lock_classes; + chain_key = iterate_chain_key((u64)0, xid); + pid = xhlock_class(xhlock) - lock_classes; + chain_key = iterate_chain_key(chain_key, pid); + + if (lookup_chain_cache(chain_key)) + return 1; + + if (!add_chain_cache_classes(xid, pid, xhlock->hlock.irq_context, + chain_key)) + return 0; + + if (!check_prev_add(current, &xlock->hlock, &xhlock->hlock, 1, + &xhlock->trace, copy_trace)) + return 0; + + return 1; +} + +static void commit_xhlocks(struct cross_lock *xlock) +{ + unsigned int cur = current->xhlock_idx; + unsigned int i; + + if (!graph_lock()) + return; + + for (i = 0; i < MAX_XHLOCKS_NR; i++) { + struct hist_lock *xhlock = &xhlock(cur - i); + + if (!xhlock_valid(xhlock)) + break; + + if (before(xhlock->hlock.gen_id, xlock->hlock.gen_id)) + break; + + if (!same_context_xhlock(xhlock)) + break; + + /* + * commit_xhlock() returns 0 with graph_lock already + * released if fail. + */ + if (!commit_xhlock(xlock, xhlock)) + return; + } + + graph_unlock(); +} + +void lock_commit_crosslock(struct lockdep_map *lock) +{ + struct cross_lock *xlock; + unsigned long flags; + + if (unlikely(!debug_locks || current->lockdep_recursion)) + return; + + if (!current->xhlocks) + return; + + /* + * Do commit hist_locks with the cross_lock, only in case that + * the cross_lock could depend on acquisitions after that. + * + * For example, if the cross_lock does not have the 'check' flag + * then we don't need to check dependencies and commit for that. + * Just skip it. In that case, of course, the cross_lock does + * not depend on acquisitions ahead, either. + * + * WARNING: Don't do that in add_xlock() in advance. When an + * acquisition context is different from the commit context, + * invalid(skipped) cross_lock might be accessed. + */ + if (!depend_after(&((struct lockdep_map_cross *)lock)->xlock.hlock)) + return; + + raw_local_irq_save(flags); + check_flags(flags); + current->lockdep_recursion = 1; + xlock = &((struct lockdep_map_cross *)lock)->xlock; + commit_xhlocks(xlock); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(lock_commit_crosslock); + +/* + * Return: 1 - crosslock, done; + * 2 - normal lock, continue to held_lock[] ops. + */ +static int lock_release_crosslock(struct lockdep_map *lock) +{ + return cross_lock(lock) ? 1 : 2; +} + +static void cross_init(struct lockdep_map *lock, int cross) +{ + lock->cross = cross; + + /* + * Crossrelease assumes that the ring buffer size of xhlocks + * is aligned with power of 2. So force it on build. + */ + BUILD_BUG_ON(MAX_XHLOCKS_NR & (MAX_XHLOCKS_NR - 1)); +} + +void lockdep_init_task(struct task_struct *task) +{ + int i; + + task->xhlock_idx = UINT_MAX; + + for (i = 0; i < XHLOCK_CTX_NR; i++) + task->xhlock_idx_hist[i] = UINT_MAX; + + task->xhlocks = kzalloc(sizeof(struct hist_lock) * MAX_XHLOCKS_NR, + GFP_KERNEL); +} + +void lockdep_free_task(struct task_struct *task) +{ + if (task->xhlocks) { + void *tmp = task->xhlocks; + /* Diable crossrelease for current */ + task->xhlocks = NULL; + kfree(tmp); + } +} +#endif diff --git a/kernel/workqueue.c b/kernel/workqueue.c index ca937b0c3a96..e86733a8b344 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2093,6 +2093,7 @@ __acquires(&pool->lock) lock_map_acquire_read(&pwq->wq->lockdep_map); lock_map_acquire(&lockdep_map); + crossrelease_hist_start(XHLOCK_PROC); trace_workqueue_execute_start(work); worker->current_func(work); /* @@ -2100,6 +2101,7 @@ __acquires(&pool->lock) * point will only record its address. */ trace_workqueue_execute_end(work); + crossrelease_hist_end(XHLOCK_PROC); lock_map_release(&lockdep_map); lock_map_release(&pwq->wq->lockdep_map); diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 98fe715522e8..c6038f23bb1a 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1150,6 +1150,18 @@ config LOCK_STAT CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. (CONFIG_LOCKDEP defines "acquire" and "release" events.) +config LOCKDEP_CROSSRELEASE + bool "Lock debugging: make lockdep work for crosslocks" + depends on PROVE_LOCKING + default n + help + This makes lockdep work for crosslock which is a lock allowed to + be released in a different context from the acquisition context. + Normally a lock must be released in the context acquiring the lock. + However, relexing this constraint helps synchronization primitives + such as page locks or completions can use the lock correctness + detector, lockdep. + config DEBUG_LOCKDEP bool "Lock dependency engine debugging" depends on DEBUG_KERNEL && LOCKDEP -- cgit v1.2.3 From 23f873d8f9526ed7e49a1a02a45f8afb9ae5fb84 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Mon, 7 Aug 2017 16:12:53 +0900 Subject: locking/lockdep: Detect and handle hist_lock ring buffer overwrite The ring buffer can be overwritten by hardirq/softirq/work contexts. That cases must be considered on rollback or commit. For example, |<------ hist_lock ring buffer size ----->| ppppppppppppiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii wrapped > iiiiiiiiiiiiiiiiiiiiiii.................... where 'p' represents an acquisition in process context, 'i' represents an acquisition in irq context. On irq exit, crossrelease tries to rollback idx to original position, but it should not because the entry already has been invalid by overwriting 'i'. Avoid rollback or commit for entries overwritten. Signed-off-by: Byungchul Park Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: akpm@linux-foundation.org Cc: boqun.feng@gmail.com Cc: kernel-team@lge.com Cc: kirill@shutemov.name Cc: npiggin@gmail.com Cc: walken@google.com Cc: willy@infradead.org Link: http://lkml.kernel.org/r/1502089981-21272-7-git-send-email-byungchul.park@lge.com Signed-off-by: Ingo Molnar --- include/linux/lockdep.h | 20 +++++++++++++++++++ include/linux/sched.h | 3 +++ kernel/locking/lockdep.c | 52 +++++++++++++++++++++++++++++++++++++++++++----- 3 files changed, 70 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index e1e0fcd99613..c75eedd55af5 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -283,6 +283,26 @@ struct held_lock { * can be added at commit step. */ struct hist_lock { + /* + * Id for each entry in the ring buffer. This is used to + * decide whether the ring buffer was overwritten or not. + * + * For example, + * + * |<----------- hist_lock ring buffer size ------->| + * pppppppppppppppppppppiiiiiiiiiiiiiiiiiiiiiiiiiiiii + * wrapped > iiiiiiiiiiiiiiiiiiiiiiiiiii....................... + * + * where 'p' represents an acquisition in process + * context, 'i' represents an acquisition in irq + * context. + * + * In this example, the ring buffer was overwritten by + * acquisitions in irq context, that should be detected on + * rollback or commit. + */ + unsigned int hist_id; + /* * Seperate stack_trace data. This will be used at commit step. */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 5235fba537fc..772c5f643764 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -854,6 +854,9 @@ struct task_struct { unsigned int xhlock_idx; /* For restoring at history boundaries */ unsigned int xhlock_idx_hist[XHLOCK_CTX_NR]; + unsigned int hist_id; + /* For overwrite check at each context exit */ + unsigned int hist_id_save[XHLOCK_CTX_NR]; #endif #ifdef CONFIG_UBSAN diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 56f69cc53ddc..eda8114ef793 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4680,6 +4680,17 @@ EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious); */ static atomic_t cross_gen_id; /* Can be wrapped */ +/* + * Make an entry of the ring buffer invalid. + */ +static inline void invalidate_xhlock(struct hist_lock *xhlock) +{ + /* + * Normally, xhlock->hlock.instance must be !NULL. + */ + xhlock->hlock.instance = NULL; +} + /* * Lock history stacks; we have 3 nested lock history stacks: * @@ -4712,14 +4723,28 @@ static atomic_t cross_gen_id; /* Can be wrapped */ */ void crossrelease_hist_start(enum xhlock_context_t c) { - if (current->xhlocks) - current->xhlock_idx_hist[c] = current->xhlock_idx; + struct task_struct *cur = current; + + if (cur->xhlocks) { + cur->xhlock_idx_hist[c] = cur->xhlock_idx; + cur->hist_id_save[c] = cur->hist_id; + } } void crossrelease_hist_end(enum xhlock_context_t c) { - if (current->xhlocks) - current->xhlock_idx = current->xhlock_idx_hist[c]; + struct task_struct *cur = current; + + if (cur->xhlocks) { + unsigned int idx = cur->xhlock_idx_hist[c]; + struct hist_lock *h = &xhlock(idx); + + cur->xhlock_idx = idx; + + /* Check if the ring was overwritten. */ + if (h->hist_id != cur->hist_id_save[c]) + invalidate_xhlock(h); + } } static int cross_lock(struct lockdep_map *lock) @@ -4765,6 +4790,7 @@ static inline int depend_after(struct held_lock *hlock) * Check if the xhlock is valid, which would be false if, * * 1. Has not used after initializaion yet. + * 2. Got invalidated. * * Remind hist_lock is implemented as a ring buffer. */ @@ -4796,6 +4822,7 @@ static void add_xhlock(struct held_lock *hlock) /* Initialize hist_lock's members */ xhlock->hlock = *hlock; + xhlock->hist_id = current->hist_id++; xhlock->trace.nr_entries = 0; xhlock->trace.max_entries = MAX_XHLOCK_TRACE_ENTRIES; @@ -4934,6 +4961,7 @@ static int commit_xhlock(struct cross_lock *xlock, struct hist_lock *xhlock) static void commit_xhlocks(struct cross_lock *xlock) { unsigned int cur = current->xhlock_idx; + unsigned int prev_hist_id = xhlock(cur).hist_id; unsigned int i; if (!graph_lock()) @@ -4951,6 +4979,17 @@ static void commit_xhlocks(struct cross_lock *xlock) if (!same_context_xhlock(xhlock)) break; + /* + * Filter out the cases that the ring buffer was + * overwritten and the previous entry has a bigger + * hist_id than the following one, which is impossible + * otherwise. + */ + if (unlikely(before(xhlock->hist_id, prev_hist_id))) + break; + + prev_hist_id = xhlock->hist_id; + /* * commit_xhlock() returns 0 with graph_lock already * released if fail. @@ -5024,9 +5063,12 @@ void lockdep_init_task(struct task_struct *task) int i; task->xhlock_idx = UINT_MAX; + task->hist_id = 0; - for (i = 0; i < XHLOCK_CTX_NR; i++) + for (i = 0; i < XHLOCK_CTX_NR; i++) { task->xhlock_idx_hist[i] = UINT_MAX; + task->hist_id_save[i] = 0; + } task->xhlocks = kzalloc(sizeof(struct hist_lock) * MAX_XHLOCKS_NR, GFP_KERNEL); -- cgit v1.2.3 From 28a903f63ec0811ead70ad0f8665e838d207a25e Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Mon, 7 Aug 2017 16:12:54 +0900 Subject: locking/lockdep: Handle non(or multi)-acquisition of a crosslock No acquisition might be in progress on commit of a crosslock. Completion operations enabling crossrelease are the case like: CONTEXT X CONTEXT Y --------- --------- trigger completion context complete AX commit AX wait_for_complete AX acquire AX wait where AX is a crosslock. When no acquisition is in progress, we should not perform commit because the lock does not exist, which might cause incorrect memory access. So we have to track the number of acquisitions of a crosslock to handle it. Moreover, in case that more than one acquisition of a crosslock are overlapped like: CONTEXT W CONTEXT X CONTEXT Y CONTEXT Z --------- --------- --------- --------- acquire AX (gen_id: 1) acquire A acquire AX (gen_id: 10) acquire B commit AX acquire C commit AX where A, B and C are typical locks and AX is a crosslock. Current crossrelease code performs commits in Y and Z with gen_id = 10. However, we can use gen_id = 1 to do it, since not only 'acquire AX in X' but 'acquire AX in W' also depends on each acquisition in Y and Z until their commits. So make it use gen_id = 1 instead of 10 on their commits, which adds an additional dependency 'AX -> A' in the example above. Signed-off-by: Byungchul Park Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: akpm@linux-foundation.org Cc: boqun.feng@gmail.com Cc: kernel-team@lge.com Cc: kirill@shutemov.name Cc: npiggin@gmail.com Cc: walken@google.com Cc: willy@infradead.org Link: http://lkml.kernel.org/r/1502089981-21272-8-git-send-email-byungchul.park@lge.com Signed-off-by: Ingo Molnar --- include/linux/lockdep.h | 22 ++++++++++++- kernel/locking/lockdep.c | 82 +++++++++++++++++++++++++++++++++--------------- 2 files changed, 77 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index c75eedd55af5..651cc61af041 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -324,6 +324,19 @@ struct hist_lock { * be called instead of lockdep_init_map(). */ struct cross_lock { + /* + * When more than one acquisition of crosslocks are overlapped, + * we have to perform commit for them based on cross_gen_id of + * the first acquisition, which allows us to add more true + * dependencies. + * + * Moreover, when no acquisition of a crosslock is in progress, + * we should not perform commit because the lock might not exist + * any more, which might cause incorrect memory access. So we + * have to track the number of acquisitions of a crosslock. + */ + int nr_acquire; + /* * Seperate hlock instance. This will be used at commit step. * @@ -547,9 +560,16 @@ extern void lockdep_init_map_crosslock(struct lockdep_map *lock, int subclass); extern void lock_commit_crosslock(struct lockdep_map *lock); +/* + * What we essencially have to initialize is 'nr_acquire'. Other members + * will be initialized in add_xlock(). + */ +#define STATIC_CROSS_LOCK_INIT() \ + { .nr_acquire = 0,} + #define STATIC_CROSS_LOCKDEP_MAP_INIT(_name, _key) \ { .map.name = (_name), .map.key = (void *)(_key), \ - .map.cross = 1, } + .map.cross = 1, .xlock = STATIC_CROSS_LOCK_INIT(), } /* * To initialize a lockdep_map statically use this macro. diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index eda8114ef793..7f97871d48d5 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4867,11 +4867,28 @@ static int add_xlock(struct held_lock *hlock) xlock = &((struct lockdep_map_cross *)hlock->instance)->xlock; + /* + * When acquisitions for a crosslock are overlapped, we use + * nr_acquire to perform commit for them, based on cross_gen_id + * of the first acquisition, which allows to add additional + * dependencies. + * + * Moreover, when no acquisition of a crosslock is in progress, + * we should not perform commit because the lock might not exist + * any more, which might cause incorrect memory access. So we + * have to track the number of acquisitions of a crosslock. + * + * depend_after() is necessary to initialize only the first + * valid xlock so that the xlock can be used on its commit. + */ + if (xlock->nr_acquire++ && depend_after(&xlock->hlock)) + goto unlock; + gen_id = (unsigned int)atomic_inc_return(&cross_gen_id); xlock->hlock = *hlock; xlock->hlock.gen_id = gen_id; +unlock: graph_unlock(); - return 1; } @@ -4967,35 +4984,37 @@ static void commit_xhlocks(struct cross_lock *xlock) if (!graph_lock()) return; - for (i = 0; i < MAX_XHLOCKS_NR; i++) { - struct hist_lock *xhlock = &xhlock(cur - i); + if (xlock->nr_acquire) { + for (i = 0; i < MAX_XHLOCKS_NR; i++) { + struct hist_lock *xhlock = &xhlock(cur - i); - if (!xhlock_valid(xhlock)) - break; + if (!xhlock_valid(xhlock)) + break; - if (before(xhlock->hlock.gen_id, xlock->hlock.gen_id)) - break; + if (before(xhlock->hlock.gen_id, xlock->hlock.gen_id)) + break; - if (!same_context_xhlock(xhlock)) - break; + if (!same_context_xhlock(xhlock)) + break; - /* - * Filter out the cases that the ring buffer was - * overwritten and the previous entry has a bigger - * hist_id than the following one, which is impossible - * otherwise. - */ - if (unlikely(before(xhlock->hist_id, prev_hist_id))) - break; + /* + * Filter out the cases that the ring buffer was + * overwritten and the previous entry has a bigger + * hist_id than the following one, which is impossible + * otherwise. + */ + if (unlikely(before(xhlock->hist_id, prev_hist_id))) + break; - prev_hist_id = xhlock->hist_id; + prev_hist_id = xhlock->hist_id; - /* - * commit_xhlock() returns 0 with graph_lock already - * released if fail. - */ - if (!commit_xhlock(xlock, xhlock)) - return; + /* + * commit_xhlock() returns 0 with graph_lock already + * released if fail. + */ + if (!commit_xhlock(xlock, xhlock)) + return; + } } graph_unlock(); @@ -5039,16 +5058,27 @@ void lock_commit_crosslock(struct lockdep_map *lock) EXPORT_SYMBOL_GPL(lock_commit_crosslock); /* - * Return: 1 - crosslock, done; + * Return: 0 - failure; + * 1 - crosslock, done; * 2 - normal lock, continue to held_lock[] ops. */ static int lock_release_crosslock(struct lockdep_map *lock) { - return cross_lock(lock) ? 1 : 2; + if (cross_lock(lock)) { + if (!graph_lock()) + return 0; + ((struct lockdep_map_cross *)lock)->xlock.nr_acquire--; + graph_unlock(); + return 1; + } + return 2; } static void cross_init(struct lockdep_map *lock, int cross) { + if (cross) + ((struct lockdep_map_cross *)lock)->xlock.nr_acquire = 0; + lock->cross = cross; /* -- cgit v1.2.3 From cd8084f91c02c1afd256a39aa833bff737631304 Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Mon, 7 Aug 2017 16:12:56 +0900 Subject: locking/lockdep: Apply crossrelease to completions Although wait_for_completion() and its family can cause deadlock, the lock correctness validator could not be applied to them until now, because things like complete() are usually called in a different context from the waiting context, which violates lockdep's assumption. Thanks to CONFIG_LOCKDEP_CROSSRELEASE, we can now apply the lockdep detector to those completion operations. Applied it. Signed-off-by: Byungchul Park Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: akpm@linux-foundation.org Cc: boqun.feng@gmail.com Cc: kernel-team@lge.com Cc: kirill@shutemov.name Cc: npiggin@gmail.com Cc: walken@google.com Cc: willy@infradead.org Link: http://lkml.kernel.org/r/1502089981-21272-10-git-send-email-byungchul.park@lge.com Signed-off-by: Ingo Molnar --- include/linux/completion.h | 45 ++++++++++++++++++++++++++++++++++++++++++++- kernel/sched/completion.c | 11 +++++++++++ lib/Kconfig.debug | 9 +++++++++ 3 files changed, 64 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/completion.h b/include/linux/completion.h index 5d5aaae3af43..9bcebf509b4d 100644 --- a/include/linux/completion.h +++ b/include/linux/completion.h @@ -9,6 +9,9 @@ */ #include +#ifdef CONFIG_LOCKDEP_COMPLETE +#include +#endif /* * struct completion - structure used to maintain state for a "completion" @@ -25,10 +28,50 @@ struct completion { unsigned int done; wait_queue_head_t wait; +#ifdef CONFIG_LOCKDEP_COMPLETE + struct lockdep_map_cross map; +#endif }; +#ifdef CONFIG_LOCKDEP_COMPLETE +static inline void complete_acquire(struct completion *x) +{ + lock_acquire_exclusive((struct lockdep_map *)&x->map, 0, 0, NULL, _RET_IP_); +} + +static inline void complete_release(struct completion *x) +{ + lock_release((struct lockdep_map *)&x->map, 0, _RET_IP_); +} + +static inline void complete_release_commit(struct completion *x) +{ + lock_commit_crosslock((struct lockdep_map *)&x->map); +} + +#define init_completion(x) \ +do { \ + static struct lock_class_key __key; \ + lockdep_init_map_crosslock((struct lockdep_map *)&(x)->map, \ + "(complete)" #x, \ + &__key, 0); \ + __init_completion(x); \ +} while (0) +#else +#define init_completion(x) __init_completion(x) +static inline void complete_acquire(struct completion *x) {} +static inline void complete_release(struct completion *x) {} +static inline void complete_release_commit(struct completion *x) {} +#endif + +#ifdef CONFIG_LOCKDEP_COMPLETE +#define COMPLETION_INITIALIZER(work) \ + { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait), \ + STATIC_CROSS_LOCKDEP_MAP_INIT("(complete)" #work, &(work)) } +#else #define COMPLETION_INITIALIZER(work) \ { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait) } +#endif #define COMPLETION_INITIALIZER_ONSTACK(work) \ ({ init_completion(&work); work; }) @@ -70,7 +113,7 @@ struct completion { * This inline function will initialize a dynamically created completion * structure. */ -static inline void init_completion(struct completion *x) +static inline void __init_completion(struct completion *x) { x->done = 0; init_waitqueue_head(&x->wait); diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index 13fc5ae9bf2f..566b6ec7b6fe 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -32,6 +32,12 @@ void complete(struct completion *x) unsigned long flags; spin_lock_irqsave(&x->wait.lock, flags); + + /* + * Perform commit of crossrelease here. + */ + complete_release_commit(x); + if (x->done != UINT_MAX) x->done++; __wake_up_locked(&x->wait, TASK_NORMAL, 1); @@ -92,9 +98,14 @@ __wait_for_common(struct completion *x, { might_sleep(); + complete_acquire(x); + spin_lock_irq(&x->wait.lock); timeout = do_wait_for_common(x, action, timeout, state); spin_unlock_irq(&x->wait.lock); + + complete_release(x); + return timeout; } diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index c6038f23bb1a..ebd40d3b0b28 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1162,6 +1162,15 @@ config LOCKDEP_CROSSRELEASE such as page locks or completions can use the lock correctness detector, lockdep. +config LOCKDEP_COMPLETE + bool "Lock debugging: allow completions to use deadlock detector" + depends on PROVE_LOCKING + select LOCKDEP_CROSSRELEASE + default n + help + A deadlock caused by wait_for_completion() and complete() can be + detected by lockdep using crossrelease feature. + config DEBUG_LOCKDEP bool "Lock dependency engine debugging" depends on DEBUG_KERNEL && LOCKDEP -- cgit v1.2.3 From 7a46ec0e2f4850407de5e1d19a44edee6efa58ec Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Tue, 15 Aug 2017 09:19:24 -0700 Subject: locking/refcounts, x86/asm: Implement fast refcount overflow protection This implements refcount_t overflow protection on x86 without a noticeable performance impact, though without the fuller checking of REFCOUNT_FULL. This is done by duplicating the existing atomic_t refcount implementation but with normally a single instruction added to detect if the refcount has gone negative (e.g. wrapped past INT_MAX or below zero). When detected, the handler saturates the refcount_t to INT_MIN / 2. With this overflow protection, the erroneous reference release that would follow a wrap back to zero is blocked from happening, avoiding the class of refcount-overflow use-after-free vulnerabilities entirely. Only the overflow case of refcounting can be perfectly protected, since it can be detected and stopped before the reference is freed and left to be abused by an attacker. There isn't a way to block early decrements, and while REFCOUNT_FULL stops increment-from-zero cases (which would be the state _after_ an early decrement and stops potential double-free conditions), this fast implementation does not, since it would require the more expensive cmpxchg loops. Since the overflow case is much more common (e.g. missing a "put" during an error path), this protection provides real-world protection. For example, the two public refcount overflow use-after-free exploits published in 2016 would have been rendered unexploitable: http://perception-point.io/2016/01/14/analysis-and-exploitation-of-a-linux-kernel-vulnerability-cve-2016-0728/ http://cyseclabs.com/page?n=02012016 This implementation does, however, notice an unchecked decrement to zero (i.e. caller used refcount_dec() instead of refcount_dec_and_test() and it resulted in a zero). Decrements under zero are noticed (since they will have resulted in a negative value), though this only indicates that a use-after-free may have already happened. Such notifications are likely avoidable by an attacker that has already exploited a use-after-free vulnerability, but it's better to have them reported than allow such conditions to remain universally silent. On first overflow detection, the refcount value is reset to INT_MIN / 2 (which serves as a saturation value) and a report and stack trace are produced. When operations detect only negative value results (such as changing an already saturated value), saturation still happens but no notification is performed (since the value was already saturated). On the matter of races, since the entire range beyond INT_MAX but before 0 is negative, every operation at INT_MIN / 2 will trap, leaving no overflow-only race condition. As for performance, this implementation adds a single "js" instruction to the regular execution flow of a copy of the standard atomic_t refcount operations. (The non-"and_test" refcount_dec() function, which is uncommon in regular refcount design patterns, has an additional "jz" instruction to detect reaching exactly zero.) Since this is a forward jump, it is by default the non-predicted path, which will be reinforced by dynamic branch prediction. The result is this protection having virtually no measurable change in performance over standard atomic_t operations. The error path, located in .text.unlikely, saves the refcount location and then uses UD0 to fire a refcount exception handler, which resets the refcount, handles reporting, and returns to regular execution. This keeps the changes to .text size minimal, avoiding return jumps and open-coded calls to the error reporting routine. Example assembly comparison: refcount_inc() before: .text: ffffffff81546149: f0 ff 45 f4 lock incl -0xc(%rbp) refcount_inc() after: .text: ffffffff81546149: f0 ff 45 f4 lock incl -0xc(%rbp) ffffffff8154614d: 0f 88 80 d5 17 00 js ffffffff816c36d3 ... .text.unlikely: ffffffff816c36d3: 48 8d 4d f4 lea -0xc(%rbp),%rcx ffffffff816c36d7: 0f ff (bad) These are the cycle counts comparing a loop of refcount_inc() from 1 to INT_MAX and back down to 0 (via refcount_dec_and_test()), between unprotected refcount_t (atomic_t), fully protected REFCOUNT_FULL (refcount_t-full), and this overflow-protected refcount (refcount_t-fast): 2147483646 refcount_inc()s and 2147483647 refcount_dec_and_test()s: cycles protections atomic_t 82249267387 none refcount_t-fast 82211446892 overflow, untested dec-to-zero refcount_t-full 144814735193 overflow, untested dec-to-zero, inc-from-zero This code is a modified version of the x86 PAX_REFCOUNT atomic_t overflow defense from the last public patch of PaX/grsecurity, based on my understanding of the code. Changes or omissions from the original code are mine and don't reflect the original grsecurity/PaX code. Thanks to PaX Team for various suggestions for improvement for repurposing this code to be a refcount-only protection. Signed-off-by: Kees Cook Reviewed-by: Josh Poimboeuf Cc: Alexey Dobriyan Cc: Andrew Morton Cc: Arnd Bergmann Cc: Christoph Hellwig Cc: David S. Miller Cc: Davidlohr Bueso Cc: Elena Reshetova Cc: Eric Biggers Cc: Eric W. Biederman Cc: Greg KH Cc: Hans Liljestrand Cc: James Bottomley Cc: Jann Horn Cc: Linus Torvalds Cc: Manfred Spraul Cc: Peter Zijlstra Cc: Rik van Riel Cc: Serge E. Hallyn Cc: Thomas Gleixner Cc: arozansk@redhat.com Cc: axboe@kernel.dk Cc: kernel-hardening@lists.openwall.com Cc: linux-arch Link: http://lkml.kernel.org/r/20170815161924.GA133115@beast Signed-off-by: Ingo Molnar --- arch/Kconfig | 12 +++++ arch/x86/Kconfig | 1 + arch/x86/include/asm/asm.h | 6 +++ arch/x86/include/asm/refcount.h | 109 ++++++++++++++++++++++++++++++++++++++++ arch/x86/mm/extable.c | 42 ++++++++++++++++ include/linux/kernel.h | 7 +++ include/linux/refcount.h | 4 ++ kernel/panic.c | 12 +++++ 8 files changed, 193 insertions(+) create mode 100644 arch/x86/include/asm/refcount.h (limited to 'include/linux') diff --git a/arch/Kconfig b/arch/Kconfig index 21d0089117fe..2520ca5b42eb 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -931,6 +931,18 @@ config STRICT_MODULE_RWX config ARCH_WANT_RELAX_ORDER bool +config ARCH_HAS_REFCOUNT + bool + help + An architecture selects this when it has implemented refcount_t + using open coded assembly primitives that provide an optimized + refcount_t implementation, possibly at the expense of some full + refcount state checks of CONFIG_REFCOUNT_FULL=y. + + The refcount overflow check behavior, however, must be retained. + Catching overflows is the primary security concern for protecting + against bugs in reference counts. + config REFCOUNT_FULL bool "Perform full reference count validation at the expense of speed" help diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 781521b7cf9e..73574c91e857 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -55,6 +55,7 @@ config X86 select ARCH_HAS_KCOV if X86_64 select ARCH_HAS_MMIO_FLUSH select ARCH_HAS_PMEM_API if X86_64 + select ARCH_HAS_REFCOUNT select ARCH_HAS_UACCESS_FLUSHCACHE if X86_64 select ARCH_HAS_SET_MEMORY select ARCH_HAS_SG_CHAIN diff --git a/arch/x86/include/asm/asm.h b/arch/x86/include/asm/asm.h index 7a9df3beb89b..676ee5807d86 100644 --- a/arch/x86/include/asm/asm.h +++ b/arch/x86/include/asm/asm.h @@ -74,6 +74,9 @@ # define _ASM_EXTABLE_EX(from, to) \ _ASM_EXTABLE_HANDLE(from, to, ex_handler_ext) +# define _ASM_EXTABLE_REFCOUNT(from, to) \ + _ASM_EXTABLE_HANDLE(from, to, ex_handler_refcount) + # define _ASM_NOKPROBE(entry) \ .pushsection "_kprobe_blacklist","aw" ; \ _ASM_ALIGN ; \ @@ -123,6 +126,9 @@ # define _ASM_EXTABLE_EX(from, to) \ _ASM_EXTABLE_HANDLE(from, to, ex_handler_ext) +# define _ASM_EXTABLE_REFCOUNT(from, to) \ + _ASM_EXTABLE_HANDLE(from, to, ex_handler_refcount) + /* For C file, we already have NOKPROBE_SYMBOL macro */ #endif diff --git a/arch/x86/include/asm/refcount.h b/arch/x86/include/asm/refcount.h new file mode 100644 index 000000000000..ff871210b9f2 --- /dev/null +++ b/arch/x86/include/asm/refcount.h @@ -0,0 +1,109 @@ +#ifndef __ASM_X86_REFCOUNT_H +#define __ASM_X86_REFCOUNT_H +/* + * x86-specific implementation of refcount_t. Based on PAX_REFCOUNT from + * PaX/grsecurity. + */ +#include + +/* + * This is the first portion of the refcount error handling, which lives in + * .text.unlikely, and is jumped to from the CPU flag check (in the + * following macros). This saves the refcount value location into CX for + * the exception handler to use (in mm/extable.c), and then triggers the + * central refcount exception. The fixup address for the exception points + * back to the regular execution flow in .text. + */ +#define _REFCOUNT_EXCEPTION \ + ".pushsection .text.unlikely\n" \ + "111:\tlea %[counter], %%" _ASM_CX "\n" \ + "112:\t" ASM_UD0 "\n" \ + ASM_UNREACHABLE \ + ".popsection\n" \ + "113:\n" \ + _ASM_EXTABLE_REFCOUNT(112b, 113b) + +/* Trigger refcount exception if refcount result is negative. */ +#define REFCOUNT_CHECK_LT_ZERO \ + "js 111f\n\t" \ + _REFCOUNT_EXCEPTION + +/* Trigger refcount exception if refcount result is zero or negative. */ +#define REFCOUNT_CHECK_LE_ZERO \ + "jz 111f\n\t" \ + REFCOUNT_CHECK_LT_ZERO + +/* Trigger refcount exception unconditionally. */ +#define REFCOUNT_ERROR \ + "jmp 111f\n\t" \ + _REFCOUNT_EXCEPTION + +static __always_inline void refcount_add(unsigned int i, refcount_t *r) +{ + asm volatile(LOCK_PREFIX "addl %1,%0\n\t" + REFCOUNT_CHECK_LT_ZERO + : [counter] "+m" (r->refs.counter) + : "ir" (i) + : "cc", "cx"); +} + +static __always_inline void refcount_inc(refcount_t *r) +{ + asm volatile(LOCK_PREFIX "incl %0\n\t" + REFCOUNT_CHECK_LT_ZERO + : [counter] "+m" (r->refs.counter) + : : "cc", "cx"); +} + +static __always_inline void refcount_dec(refcount_t *r) +{ + asm volatile(LOCK_PREFIX "decl %0\n\t" + REFCOUNT_CHECK_LE_ZERO + : [counter] "+m" (r->refs.counter) + : : "cc", "cx"); +} + +static __always_inline __must_check +bool refcount_sub_and_test(unsigned int i, refcount_t *r) +{ + GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl", REFCOUNT_CHECK_LT_ZERO, + r->refs.counter, "er", i, "%0", e); +} + +static __always_inline __must_check bool refcount_dec_and_test(refcount_t *r) +{ + GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl", REFCOUNT_CHECK_LT_ZERO, + r->refs.counter, "%0", e); +} + +static __always_inline __must_check +bool refcount_add_not_zero(unsigned int i, refcount_t *r) +{ + int c, result; + + c = atomic_read(&(r->refs)); + do { + if (unlikely(c == 0)) + return false; + + result = c + i; + + /* Did we try to increment from/to an undesirable state? */ + if (unlikely(c < 0 || c == INT_MAX || result < c)) { + asm volatile(REFCOUNT_ERROR + : : [counter] "m" (r->refs.counter) + : "cc", "cx"); + break; + } + + } while (!atomic_try_cmpxchg(&(r->refs), &c, result)); + + return c != 0; +} + +static __always_inline __must_check bool refcount_inc_not_zero(refcount_t *r) +{ + return refcount_add_not_zero(1, r); +} + +#endif diff --git a/arch/x86/mm/extable.c b/arch/x86/mm/extable.c index 0ea8afcb929c..761fc88cd820 100644 --- a/arch/x86/mm/extable.c +++ b/arch/x86/mm/extable.c @@ -36,6 +36,48 @@ bool ex_handler_fault(const struct exception_table_entry *fixup, } EXPORT_SYMBOL_GPL(ex_handler_fault); +/* + * Handler for UD0 exception following a failed test against the + * result of a refcount inc/dec/add/sub. + */ +bool ex_handler_refcount(const struct exception_table_entry *fixup, + struct pt_regs *regs, int trapnr) +{ + /* First unconditionally saturate the refcount. */ + *(int *)regs->cx = INT_MIN / 2; + + /* + * Strictly speaking, this reports the fixup destination, not + * the fault location, and not the actually overflowing + * instruction, which is the instruction before the "js", but + * since that instruction could be a variety of lengths, just + * report the location after the overflow, which should be close + * enough for finding the overflow, as it's at least back in + * the function, having returned from .text.unlikely. + */ + regs->ip = ex_fixup_addr(fixup); + + /* + * This function has been called because either a negative refcount + * value was seen by any of the refcount functions, or a zero + * refcount value was seen by refcount_dec(). + * + * If we crossed from INT_MAX to INT_MIN, OF (Overflow Flag: result + * wrapped around) will be set. Additionally, seeing the refcount + * reach 0 will set ZF (Zero Flag: result was zero). In each of + * these cases we want a report, since it's a boundary condition. + * + */ + if (regs->flags & (X86_EFLAGS_OF | X86_EFLAGS_ZF)) { + bool zero = regs->flags & X86_EFLAGS_ZF; + + refcount_error_report(regs, zero ? "hit zero" : "overflow"); + } + + return true; +} +EXPORT_SYMBOL_GPL(ex_handler_refcount); + bool ex_handler_ext(const struct exception_table_entry *fixup, struct pt_regs *regs, int trapnr) { diff --git a/include/linux/kernel.h b/include/linux/kernel.h index bd6d96cf80b1..6607225d0ea4 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -277,6 +277,13 @@ extern int oops_may_print(void); void do_exit(long error_code) __noreturn; void complete_and_exit(struct completion *, long) __noreturn; +#ifdef CONFIG_ARCH_HAS_REFCOUNT +void refcount_error_report(struct pt_regs *regs, const char *err); +#else +static inline void refcount_error_report(struct pt_regs *regs, const char *err) +{ } +#endif + /* Internal, do not use. */ int __must_check _kstrtoul(const char *s, unsigned int base, unsigned long *res); int __must_check _kstrtol(const char *s, unsigned int base, long *res); diff --git a/include/linux/refcount.h b/include/linux/refcount.h index 591792c8e5b0..48b7c9c68c4d 100644 --- a/include/linux/refcount.h +++ b/include/linux/refcount.h @@ -53,6 +53,9 @@ extern __must_check bool refcount_sub_and_test(unsigned int i, refcount_t *r); extern __must_check bool refcount_dec_and_test(refcount_t *r); extern void refcount_dec(refcount_t *r); #else +# ifdef CONFIG_ARCH_HAS_REFCOUNT +# include +# else static inline __must_check bool refcount_add_not_zero(unsigned int i, refcount_t *r) { return atomic_add_unless(&r->refs, i, 0); @@ -87,6 +90,7 @@ static inline void refcount_dec(refcount_t *r) { atomic_dec(&r->refs); } +# endif /* !CONFIG_ARCH_HAS_REFCOUNT */ #endif /* CONFIG_REFCOUNT_FULL */ extern __must_check bool refcount_dec_if_one(refcount_t *r); diff --git a/kernel/panic.c b/kernel/panic.c index a58932b41700..bdd18afa19a4 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -26,6 +26,7 @@ #include #include #include +#include #define PANIC_TIMER_STEP 100 #define PANIC_BLINK_SPD 18 @@ -601,6 +602,17 @@ EXPORT_SYMBOL(__stack_chk_fail); #endif +#ifdef CONFIG_ARCH_HAS_REFCOUNT +void refcount_error_report(struct pt_regs *regs, const char *err) +{ + WARN_RATELIMIT(1, "refcount_t %s at %pB in %s[%d], uid/euid: %u/%u\n", + err, (void *)instruction_pointer(regs), + current->comm, task_pid_nr(current), + from_kuid_munged(&init_user_ns, current_uid()), + from_kuid_munged(&init_user_ns, current_euid())); +} +#endif + core_param(panic, panic_timeout, int, 0644); core_param(pause_on_oops, pause_on_oops, int, 0644); core_param(panic_on_warn, panic_on_warn, int, 0644); -- cgit v1.2.3 From ea3f2c0fdfbb180f142cdbc0d1f055c7b9a5e63e Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Thu, 17 Aug 2017 17:57:41 +0900 Subject: locking/lockdep: Rename CONFIG_LOCKDEP_COMPLETE to CONFIG_LOCKDEP_COMPLETIONS 'complete' is an adjective and LOCKDEP_COMPLETE sounds like 'lockdep is complete', so pick a better name that uses a noun. Signed-off-by: Byungchul Park Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: kernel-team@lge.com Link: http://lkml.kernel.org/r/1502960261-16206-3-git-send-email-byungchul.park@lge.com Signed-off-by: Ingo Molnar --- include/linux/completion.h | 8 ++++---- lib/Kconfig.debug | 4 ++-- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/completion.h b/include/linux/completion.h index 9bcebf509b4d..791f053f28b7 100644 --- a/include/linux/completion.h +++ b/include/linux/completion.h @@ -9,7 +9,7 @@ */ #include -#ifdef CONFIG_LOCKDEP_COMPLETE +#ifdef CONFIG_LOCKDEP_COMPLETIONS #include #endif @@ -28,12 +28,12 @@ struct completion { unsigned int done; wait_queue_head_t wait; -#ifdef CONFIG_LOCKDEP_COMPLETE +#ifdef CONFIG_LOCKDEP_COMPLETIONS struct lockdep_map_cross map; #endif }; -#ifdef CONFIG_LOCKDEP_COMPLETE +#ifdef CONFIG_LOCKDEP_COMPLETIONS static inline void complete_acquire(struct completion *x) { lock_acquire_exclusive((struct lockdep_map *)&x->map, 0, 0, NULL, _RET_IP_); @@ -64,7 +64,7 @@ static inline void complete_release(struct completion *x) {} static inline void complete_release_commit(struct completion *x) {} #endif -#ifdef CONFIG_LOCKDEP_COMPLETE +#ifdef CONFIG_LOCKDEP_COMPLETIONS #define COMPLETION_INITIALIZER(work) \ { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait), \ STATIC_CROSS_LOCKDEP_MAP_INIT("(complete)" #work, &(work)) } diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 8fb8a206db12..a0e60d56303e 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1082,7 +1082,7 @@ config PROVE_LOCKING select DEBUG_RT_MUTEXES if RT_MUTEXES select DEBUG_LOCK_ALLOC select LOCKDEP_CROSSRELEASE - select LOCKDEP_COMPLETE + select LOCKDEP_COMPLETIONS select TRACE_IRQFLAGS default n help @@ -1162,7 +1162,7 @@ config LOCKDEP_CROSSRELEASE such as page locks or completions can use the lock correctness detector, lockdep. -config LOCKDEP_COMPLETE +config LOCKDEP_COMPLETIONS bool "Lock debugging: allow completions to use deadlock detector" help A deadlock caused by wait_for_completion() and complete() can be -- cgit v1.2.3 From 52fa5bc5cbba089f09bc2c372e3432f3f3e48051 Mon Sep 17 00:00:00 2001 From: Boqun Feng Date: Thu, 17 Aug 2017 17:46:12 +0800 Subject: locking/lockdep: Explicitly initialize wq_barrier::done::map With the new lockdep crossrelease feature, which checks completions usage, a false positive is reported in the workqueue code: > Worker A : acquired of wfc.work -> wait for cpu_hotplug_lock to be released > Task B : acquired of cpu_hotplug_lock -> wait for lock#3 to be released > Task C : acquired of lock#3 -> wait for completion of barr->done > (Task C is in lru_add_drain_all_cpuslocked()) > Worker D : wait for wfc.work to be released -> will complete barr->done Such a dead lock can not happen because Task C's barr->done and Worker D's barr->done can not be the same instance. The reason of this false positive is we initialize all wq_barrier::done at insert_wq_barrier() via init_completion(), which makes them belong to the same lock class, therefore, impossible circles are reported. To fix this, explicitly initialize the lockdep map for wq_barrier::done in insert_wq_barrier(), so that the lock class key of wq_barrier::done is a subkey of the corresponding work_struct, as a result we won't build a dependency between a wq_barrier with a unrelated work, and we can differ wq barriers based on the related works, so the false positive above is avoided. Also define the empty lockdep_init_map_crosslock() for !CROSSRELEASE to make the code simple and away from unnecessary #ifdefs. Reported-by: Ingo Molnar Signed-off-by: Boqun Feng Cc: Byungchul Park Cc: Lai Jiangshan Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Tejun Heo Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20170817094622.12915-1-boqun.feng@gmail.com Signed-off-by: Ingo Molnar --- include/linux/lockdep.h | 1 + kernel/workqueue.c | 11 ++++++++++- 2 files changed, 11 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 651cc61af041..fc827cab6d6e 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -583,6 +583,7 @@ extern void crossrelease_hist_end(enum xhlock_context_t c); extern void lockdep_init_task(struct task_struct *task); extern void lockdep_free_task(struct task_struct *task); #else +#define lockdep_init_map_crosslock(m, n, k, s) do {} while (0) /* * To initialize a lockdep_map statically use this macro. * Note that _name must not be NULL. diff --git a/kernel/workqueue.c b/kernel/workqueue.c index e86733a8b344..f128b3becfe1 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2476,7 +2476,16 @@ static void insert_wq_barrier(struct pool_workqueue *pwq, */ INIT_WORK_ONSTACK(&barr->work, wq_barrier_func); __set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(&barr->work)); - init_completion(&barr->done); + + /* + * Explicitly init the crosslock for wq_barrier::done, make its lock + * key a subkey of the corresponding work. As a result we won't + * build a dependency between wq_barrier::done and unrelated work. + */ + lockdep_init_map_crosslock((struct lockdep_map *)&barr->done.map, + "(complete)wq_barr::done", + target->lockdep_map.key, 1); + __init_completion(&barr->done); barr->task = current; /* -- cgit v1.2.3 From 0e709703af5bbc9ea6e75e1f99c2dd0dae261869 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 11 Aug 2017 16:04:50 +0200 Subject: mm, locking/barriers: Clarify tlb_flush_pending() barriers Better document the ordering around tlb_flush_pending(). Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar --- include/linux/mm_types.h | 78 ++++++++++++++++++++++++++++-------------------- 1 file changed, 45 insertions(+), 33 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index dc1edec05a3f..57378c7cb5f8 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -526,30 +526,6 @@ extern void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm, extern void tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end); -/* - * Memory barriers to keep this state in sync are graciously provided by - * the page table locks, outside of which no page table modifications happen. - * The barriers are used to ensure the order between tlb_flush_pending updates, - * which happen while the lock is not taken, and the PTE updates, which happen - * while the lock is taken, are serialized. - */ -static inline bool mm_tlb_flush_pending(struct mm_struct *mm) -{ - /* - * Must be called with PTL held; such that our PTL acquire will have - * observed the store from set_tlb_flush_pending(). - */ - return atomic_read(&mm->tlb_flush_pending) > 0; -} - -/* - * Returns true if there are two above TLB batching threads in parallel. - */ -static inline bool mm_tlb_flush_nested(struct mm_struct *mm) -{ - return atomic_read(&mm->tlb_flush_pending) > 1; -} - static inline void init_tlb_flush_pending(struct mm_struct *mm) { atomic_set(&mm->tlb_flush_pending, 0); @@ -558,7 +534,6 @@ static inline void init_tlb_flush_pending(struct mm_struct *mm) static inline void inc_tlb_flush_pending(struct mm_struct *mm) { atomic_inc(&mm->tlb_flush_pending); - /* * The only time this value is relevant is when there are indeed pages * to flush. And we'll only flush pages after changing them, which @@ -580,24 +555,61 @@ static inline void inc_tlb_flush_pending(struct mm_struct *mm) * flush_tlb_range(); * atomic_dec(&mm->tlb_flush_pending); * - * So the =true store is constrained by the PTL unlock, and the =false - * store is constrained by the TLB invalidate. + * Where the increment if constrained by the PTL unlock, it thus + * ensures that the increment is visible if the PTE modification is + * visible. After all, if there is no PTE modification, nobody cares + * about TLB flushes either. + * + * This very much relies on users (mm_tlb_flush_pending() and + * mm_tlb_flush_nested()) only caring about _specific_ PTEs (and + * therefore specific PTLs), because with SPLIT_PTE_PTLOCKS and RCpc + * locks (PPC) the unlock of one doesn't order against the lock of + * another PTL. + * + * The decrement is ordered by the flush_tlb_range(), such that + * mm_tlb_flush_pending() will not return false unless all flushes have + * completed. */ } -/* Clearing is done after a TLB flush, which also provides a barrier. */ static inline void dec_tlb_flush_pending(struct mm_struct *mm) { /* - * Guarantee that the tlb_flush_pending does not not leak into the - * critical section, since we must order the PTE change and changes to - * the pending TLB flush indication. We could have relied on TLB flush - * as a memory barrier, but this behavior is not clearly documented. + * See inc_tlb_flush_pending(). + * + * This cannot be smp_mb__before_atomic() because smp_mb() simply does + * not order against TLB invalidate completion, which is what we need. + * + * Therefore we must rely on tlb_flush_*() to guarantee order. */ - smp_mb__before_atomic(); atomic_dec(&mm->tlb_flush_pending); } +static inline bool mm_tlb_flush_pending(struct mm_struct *mm) +{ + /* + * Must be called after having acquired the PTL; orders against that + * PTLs release and therefore ensures that if we observe the modified + * PTE we must also observe the increment from inc_tlb_flush_pending(). + * + * That is, it only guarantees to return true if there is a flush + * pending for _this_ PTL. + */ + return atomic_read(&mm->tlb_flush_pending); +} + +static inline bool mm_tlb_flush_nested(struct mm_struct *mm) +{ + /* + * Similar to mm_tlb_flush_pending(), we must have acquired the PTL + * for which there is a TLB flush pending in order to guarantee + * we've seen both that PTE modification and the increment. + * + * (no requirement on actually still holding the PTL, that is irrelevant) + */ + return atomic_read(&mm->tlb_flush_pending) > 1; +} + struct vm_fault; struct vm_special_mapping { -- cgit v1.2.3 From e6f3faa734a00c606b7b06c6b9f15e5627d3245b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 23 Aug 2017 13:23:30 +0200 Subject: locking/lockdep: Fix workqueue crossrelease annotation The new completion/crossrelease annotations interact unfavourable with the extant flush_work()/flush_workqueue() annotations. The problem is that when a single work class does: wait_for_completion(&C) and complete(&C) in different executions, we'll build dependencies like: lock_map_acquire(W) complete_acquire(C) and lock_map_acquire(W) complete_release(C) which results in the dependency chain: W->C->W, which lockdep thinks spells deadlock, even though there is no deadlock potential since works are ran concurrently. One possibility would be to change the work 'lock' to recursive-read, but that would mean hitting a lockdep limitation on recursive locks. Also, unconditinoally switching to recursive-read here would fail to detect the actual deadlock on single-threaded workqueues, which do have a problem with this. For now, forcefully disregard these locks for crossrelease. Signed-off-by: Peter Zijlstra (Intel) Acked-by: Tejun Heo Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: boqun.feng@gmail.com Cc: byungchul.park@lge.com Cc: david@fromorbit.com Cc: johannes@sipsolutions.net Cc: oleg@redhat.com Signed-off-by: Ingo Molnar --- include/linux/irqflags.h | 4 ++-- include/linux/lockdep.h | 10 +++++---- kernel/locking/lockdep.c | 56 +++++++++++++++++++++++++++++++----------------- kernel/workqueue.c | 23 +++++++++++++++++++- 4 files changed, 66 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 5fdd93bb9300..9bc050bc81b2 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -26,7 +26,7 @@ # define trace_hardirq_enter() \ do { \ current->hardirq_context++; \ - crossrelease_hist_start(XHLOCK_HARD); \ + crossrelease_hist_start(XHLOCK_HARD, 0);\ } while (0) # define trace_hardirq_exit() \ do { \ @@ -36,7 +36,7 @@ do { \ # define lockdep_softirq_enter() \ do { \ current->softirq_context++; \ - crossrelease_hist_start(XHLOCK_SOFT); \ + crossrelease_hist_start(XHLOCK_SOFT, 0);\ } while (0) # define lockdep_softirq_exit() \ do { \ diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index fc827cab6d6e..78bb7133abed 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -18,6 +18,8 @@ extern int lock_stat; #define MAX_LOCKDEP_SUBCLASSES 8UL +#include + #ifdef CONFIG_LOCKDEP #include @@ -578,11 +580,11 @@ extern void lock_commit_crosslock(struct lockdep_map *lock); #define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ { .name = (_name), .key = (void *)(_key), .cross = 0, } -extern void crossrelease_hist_start(enum xhlock_context_t c); +extern void crossrelease_hist_start(enum xhlock_context_t c, bool force); extern void crossrelease_hist_end(enum xhlock_context_t c); extern void lockdep_init_task(struct task_struct *task); extern void lockdep_free_task(struct task_struct *task); -#else +#else /* !CROSSRELEASE */ #define lockdep_init_map_crosslock(m, n, k, s) do {} while (0) /* * To initialize a lockdep_map statically use this macro. @@ -591,11 +593,11 @@ extern void lockdep_free_task(struct task_struct *task); #define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ { .name = (_name), .key = (void *)(_key), } -static inline void crossrelease_hist_start(enum xhlock_context_t c) {} +static inline void crossrelease_hist_start(enum xhlock_context_t c, bool force) {} static inline void crossrelease_hist_end(enum xhlock_context_t c) {} static inline void lockdep_init_task(struct task_struct *task) {} static inline void lockdep_free_task(struct task_struct *task) {} -#endif +#endif /* CROSSRELEASE */ #ifdef CONFIG_LOCK_STAT diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 66011c9f5df3..f73ca595b81e 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4629,7 +4629,7 @@ asmlinkage __visible void lockdep_sys_exit(void) * the index to point to the last entry, which is already invalid. */ crossrelease_hist_end(XHLOCK_PROC); - crossrelease_hist_start(XHLOCK_PROC); + crossrelease_hist_start(XHLOCK_PROC, false); } void lockdep_rcu_suspicious(const char *file, const int line, const char *s) @@ -4725,25 +4725,25 @@ static inline void invalidate_xhlock(struct hist_lock *xhlock) /* * Lock history stacks; we have 3 nested lock history stacks: * - * Hard IRQ - * Soft IRQ - * History / Task + * HARD(IRQ) + * SOFT(IRQ) + * PROC(ess) * - * The thing is that once we complete a (Hard/Soft) IRQ the future task locks - * should not depend on any of the locks observed while running the IRQ. + * The thing is that once we complete a HARD/SOFT IRQ the future task locks + * should not depend on any of the locks observed while running the IRQ. So + * what we do is rewind the history buffer and erase all our knowledge of that + * temporal event. * - * So what we do is rewind the history buffer and erase all our knowledge of - * that temporal event. - */ - -/* - * We need this to annotate lock history boundaries. Take for instance - * workqueues; each work is independent of the last. The completion of a future - * work does not depend on the completion of a past work (in general). - * Therefore we must not carry that (lock) dependency across works. + * The PROCess one is special though; it is used to annotate independence + * inside a task. + * + * Take for instance workqueues; each work is independent of the last. The + * completion of a future work does not depend on the completion of a past work + * (in general). Therefore we must not carry that (lock) dependency across + * works. * * This is true for many things; pretty much all kthreads fall into this - * pattern, where they have an 'idle' state and future completions do not + * pattern, where they have an invariant state and future completions do not * depend on past completions. Its just that since they all have the 'same' * form -- the kthread does the same over and over -- it doesn't typically * matter. @@ -4751,15 +4751,31 @@ static inline void invalidate_xhlock(struct hist_lock *xhlock) * The same is true for system-calls, once a system call is completed (we've * returned to userspace) the next system call does not depend on the lock * history of the previous system call. + * + * They key property for independence, this invariant state, is that it must be + * a point where we hold no locks and have no history. Because if we were to + * hold locks, the restore at _end() would not necessarily recover it's history + * entry. Similarly, independence per-definition means it does not depend on + * prior state. */ -void crossrelease_hist_start(enum xhlock_context_t c) +void crossrelease_hist_start(enum xhlock_context_t c, bool force) { struct task_struct *cur = current; - if (cur->xhlocks) { - cur->xhlock_idx_hist[c] = cur->xhlock_idx; - cur->hist_id_save[c] = cur->hist_id; + if (!cur->xhlocks) + return; + + /* + * We call this at an invariant point, no current state, no history. + */ + if (c == XHLOCK_PROC) { + /* verified the former, ensure the latter */ + WARN_ON_ONCE(!force && cur->lockdep_depth); + invalidate_xhlock(&xhlock(cur->xhlock_idx)); } + + cur->xhlock_idx_hist[c] = cur->xhlock_idx; + cur->hist_id_save[c] = cur->hist_id; } void crossrelease_hist_end(enum xhlock_context_t c) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 8ad214dc15a9..c0331891dec1 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2093,7 +2093,28 @@ __acquires(&pool->lock) lock_map_acquire(&pwq->wq->lockdep_map); lock_map_acquire(&lockdep_map); - crossrelease_hist_start(XHLOCK_PROC); + /* + * Strictly speaking we should do start(PROC) without holding any + * locks, that is, before these two lock_map_acquire()'s. + * + * However, that would result in: + * + * A(W1) + * WFC(C) + * A(W1) + * C(C) + * + * Which would create W1->C->W1 dependencies, even though there is no + * actual deadlock possible. There are two solutions, using a + * read-recursive acquire on the work(queue) 'locks', but this will then + * hit the lockdep limitation on recursive locks, or simly discard + * these locks. + * + * AFAICT there is no possible deadlock scenario between the + * flush_work() and complete() primitives (except for single-threaded + * workqueues), so hiding them isn't a problem. + */ + crossrelease_hist_start(XHLOCK_PROC, true); trace_workqueue_execute_start(work); worker->current_func(work); /* -- cgit v1.2.3 From f52be5708076b75a045ac52c6fef3fffb8300525 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 29 Aug 2017 10:59:39 +0200 Subject: locking/lockdep: Untangle xhlock history save/restore from task independence Where XHLOCK_{SOFT,HARD} are save/restore points in the xhlocks[] to ensure the temporal IRQ events don't interact with task state, the XHLOCK_PROC is a fundament different beast that just happens to share the interface. The purpose of XHLOCK_PROC is to annotate independent execution inside one task. For example workqueues, each work should appear to run in its own 'pristine' 'task'. Remove XHLOCK_PROC in favour of its own interface to avoid confusion. Signed-off-by: Peter Zijlstra (Intel) Cc: Byungchul Park Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: boqun.feng@gmail.com Cc: david@fromorbit.com Cc: johannes@sipsolutions.net Cc: kernel-team@lge.com Cc: oleg@redhat.com Cc: tj@kernel.org Link: http://lkml.kernel.org/r/20170829085939.ggmb6xiohw67micb@hirez.programming.kicks-ass.net Signed-off-by: Ingo Molnar --- include/linux/irqflags.h | 4 +-- include/linux/lockdep.h | 7 +++-- kernel/locking/lockdep.c | 79 +++++++++++++++++++++++------------------------- kernel/workqueue.c | 9 +++--- 4 files changed, 48 insertions(+), 51 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 9bc050bc81b2..5fdd93bb9300 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -26,7 +26,7 @@ # define trace_hardirq_enter() \ do { \ current->hardirq_context++; \ - crossrelease_hist_start(XHLOCK_HARD, 0);\ + crossrelease_hist_start(XHLOCK_HARD); \ } while (0) # define trace_hardirq_exit() \ do { \ @@ -36,7 +36,7 @@ do { \ # define lockdep_softirq_enter() \ do { \ current->softirq_context++; \ - crossrelease_hist_start(XHLOCK_SOFT, 0);\ + crossrelease_hist_start(XHLOCK_SOFT); \ } while (0) # define lockdep_softirq_exit() \ do { \ diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 78bb7133abed..bfa8e0b0d6f1 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -551,7 +551,6 @@ struct pin_cookie { }; enum xhlock_context_t { XHLOCK_HARD, XHLOCK_SOFT, - XHLOCK_PROC, XHLOCK_CTX_NR, }; @@ -580,8 +579,9 @@ extern void lock_commit_crosslock(struct lockdep_map *lock); #define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ { .name = (_name), .key = (void *)(_key), .cross = 0, } -extern void crossrelease_hist_start(enum xhlock_context_t c, bool force); +extern void crossrelease_hist_start(enum xhlock_context_t c); extern void crossrelease_hist_end(enum xhlock_context_t c); +extern void lockdep_invariant_state(bool force); extern void lockdep_init_task(struct task_struct *task); extern void lockdep_free_task(struct task_struct *task); #else /* !CROSSRELEASE */ @@ -593,8 +593,9 @@ extern void lockdep_free_task(struct task_struct *task); #define STATIC_LOCKDEP_MAP_INIT(_name, _key) \ { .name = (_name), .key = (void *)(_key), } -static inline void crossrelease_hist_start(enum xhlock_context_t c, bool force) {} +static inline void crossrelease_hist_start(enum xhlock_context_t c) {} static inline void crossrelease_hist_end(enum xhlock_context_t c) {} +static inline void lockdep_invariant_state(bool force) {} static inline void lockdep_init_task(struct task_struct *task) {} static inline void lockdep_free_task(struct task_struct *task) {} #endif /* CROSSRELEASE */ diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index f73ca595b81e..44c8d0d17170 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4623,13 +4623,8 @@ asmlinkage __visible void lockdep_sys_exit(void) /* * The lock history for each syscall should be independent. So wipe the * slate clean on return to userspace. - * - * crossrelease_hist_end() works well here even when getting here - * without starting (i.e. just after forking), because it rolls back - * the index to point to the last entry, which is already invalid. */ - crossrelease_hist_end(XHLOCK_PROC); - crossrelease_hist_start(XHLOCK_PROC, false); + lockdep_invariant_state(false); } void lockdep_rcu_suspicious(const char *file, const int line, const char *s) @@ -4723,19 +4718,47 @@ static inline void invalidate_xhlock(struct hist_lock *xhlock) } /* - * Lock history stacks; we have 3 nested lock history stacks: + * Lock history stacks; we have 2 nested lock history stacks: * * HARD(IRQ) * SOFT(IRQ) - * PROC(ess) * * The thing is that once we complete a HARD/SOFT IRQ the future task locks * should not depend on any of the locks observed while running the IRQ. So * what we do is rewind the history buffer and erase all our knowledge of that * temporal event. - * - * The PROCess one is special though; it is used to annotate independence - * inside a task. + */ + +void crossrelease_hist_start(enum xhlock_context_t c) +{ + struct task_struct *cur = current; + + if (!cur->xhlocks) + return; + + cur->xhlock_idx_hist[c] = cur->xhlock_idx; + cur->hist_id_save[c] = cur->hist_id; +} + +void crossrelease_hist_end(enum xhlock_context_t c) +{ + struct task_struct *cur = current; + + if (cur->xhlocks) { + unsigned int idx = cur->xhlock_idx_hist[c]; + struct hist_lock *h = &xhlock(idx); + + cur->xhlock_idx = idx; + + /* Check if the ring was overwritten. */ + if (h->hist_id != cur->hist_id_save[c]) + invalidate_xhlock(h); + } +} + +/* + * lockdep_invariant_state() is used to annotate independence inside a task, to + * make one task look like multiple independent 'tasks'. * * Take for instance workqueues; each work is independent of the last. The * completion of a future work does not depend on the completion of a past work @@ -4758,40 +4781,14 @@ static inline void invalidate_xhlock(struct hist_lock *xhlock) * entry. Similarly, independence per-definition means it does not depend on * prior state. */ -void crossrelease_hist_start(enum xhlock_context_t c, bool force) +void lockdep_invariant_state(bool force) { - struct task_struct *cur = current; - - if (!cur->xhlocks) - return; - /* * We call this at an invariant point, no current state, no history. + * Verify the former, enforce the latter. */ - if (c == XHLOCK_PROC) { - /* verified the former, ensure the latter */ - WARN_ON_ONCE(!force && cur->lockdep_depth); - invalidate_xhlock(&xhlock(cur->xhlock_idx)); - } - - cur->xhlock_idx_hist[c] = cur->xhlock_idx; - cur->hist_id_save[c] = cur->hist_id; -} - -void crossrelease_hist_end(enum xhlock_context_t c) -{ - struct task_struct *cur = current; - - if (cur->xhlocks) { - unsigned int idx = cur->xhlock_idx_hist[c]; - struct hist_lock *h = &xhlock(idx); - - cur->xhlock_idx = idx; - - /* Check if the ring was overwritten. */ - if (h->hist_id != cur->hist_id_save[c]) - invalidate_xhlock(h); - } + WARN_ON_ONCE(!force && current->lockdep_depth); + invalidate_xhlock(&xhlock(current->xhlock_idx)); } static int cross_lock(struct lockdep_map *lock) diff --git a/kernel/workqueue.c b/kernel/workqueue.c index c0331891dec1..ab3c0dc8c7ed 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2094,8 +2094,8 @@ __acquires(&pool->lock) lock_map_acquire(&pwq->wq->lockdep_map); lock_map_acquire(&lockdep_map); /* - * Strictly speaking we should do start(PROC) without holding any - * locks, that is, before these two lock_map_acquire()'s. + * Strictly speaking we should mark the invariant state without holding + * any locks, that is, before these two lock_map_acquire()'s. * * However, that would result in: * @@ -2107,14 +2107,14 @@ __acquires(&pool->lock) * Which would create W1->C->W1 dependencies, even though there is no * actual deadlock possible. There are two solutions, using a * read-recursive acquire on the work(queue) 'locks', but this will then - * hit the lockdep limitation on recursive locks, or simly discard + * hit the lockdep limitation on recursive locks, or simply discard * these locks. * * AFAICT there is no possible deadlock scenario between the * flush_work() and complete() primitives (except for single-threaded * workqueues), so hiding them isn't a problem. */ - crossrelease_hist_start(XHLOCK_PROC, true); + lockdep_invariant_state(true); trace_workqueue_execute_start(work); worker->current_func(work); /* @@ -2122,7 +2122,6 @@ __acquires(&pool->lock) * point will only record its address. */ trace_workqueue_execute_end(work); - crossrelease_hist_end(XHLOCK_PROC); lock_map_release(&lockdep_map); lock_map_release(&pwq->wq->lockdep_map); -- cgit v1.2.3 From 966a967116e699762dbf4af7f9e0d1955c25aa37 Mon Sep 17 00:00:00 2001 From: Ying Huang Date: Tue, 8 Aug 2017 12:30:00 +0800 Subject: smp: Avoid using two cache lines for struct call_single_data struct call_single_data is used in IPIs to transfer information between CPUs. Its size is bigger than sizeof(unsigned long) and less than cache line size. Currently it is not allocated with any explicit alignment requirements. This makes it possible for allocated call_single_data to cross two cache lines, which results in double the number of the cache lines that need to be transferred among CPUs. This can be fixed by requiring call_single_data to be aligned with the size of call_single_data. Currently the size of call_single_data is the power of 2. If we add new fields to call_single_data, we may need to add padding to make sure the size of new definition is the power of 2 as well. Fortunately, this is enforced by GCC, which will report bad sizes. To set alignment requirements of call_single_data to the size of call_single_data, a struct definition and a typedef is used. To test the effect of the patch, I used the vm-scalability multiple thread swap test case (swap-w-seq-mt). The test will create multiple threads and each thread will eat memory until all RAM and part of swap is used, so that huge number of IPIs are triggered when unmapping memory. In the test, the throughput of memory writing improves ~5% compared with misaligned call_single_data, because of faster IPIs. Suggested-by: Peter Zijlstra Signed-off-by: Huang, Ying [ Add call_single_data_t and align with size of call_single_data. ] Signed-off-by: Peter Zijlstra (Intel) Cc: Aaron Lu Cc: Borislav Petkov Cc: Eric Dumazet Cc: Juergen Gross Cc: Linus Torvalds Cc: Michael Ellerman Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/87bmnqd6lz.fsf@yhuang-mobile.sh.intel.com Signed-off-by: Ingo Molnar --- arch/mips/kernel/smp.c | 6 ++-- block/blk-softirq.c | 2 +- drivers/block/null_blk.c | 2 +- drivers/cpuidle/coupled.c | 10 +++---- drivers/net/ethernet/cavium/liquidio/lio_main.c | 2 +- drivers/net/ethernet/cavium/liquidio/octeon_droq.h | 2 +- include/linux/blkdev.h | 2 +- include/linux/netdevice.h | 2 +- include/linux/smp.h | 8 ++++-- kernel/sched/sched.h | 2 +- kernel/smp.c | 32 ++++++++++++---------- kernel/up.c | 2 +- 12 files changed, 39 insertions(+), 33 deletions(-) (limited to 'include/linux') diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index 6bace7695788..c7cbddfcdc3b 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -648,12 +648,12 @@ EXPORT_SYMBOL(flush_tlb_one); #ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST static DEFINE_PER_CPU(atomic_t, tick_broadcast_count); -static DEFINE_PER_CPU(struct call_single_data, tick_broadcast_csd); +static DEFINE_PER_CPU(call_single_data_t, tick_broadcast_csd); void tick_broadcast(const struct cpumask *mask) { atomic_t *count; - struct call_single_data *csd; + call_single_data_t *csd; int cpu; for_each_cpu(cpu, mask) { @@ -674,7 +674,7 @@ static void tick_broadcast_callee(void *info) static int __init tick_broadcast_init(void) { - struct call_single_data *csd; + call_single_data_t *csd; int cpu; for (cpu = 0; cpu < NR_CPUS; cpu++) { diff --git a/block/blk-softirq.c b/block/blk-softirq.c index 87b7df4851bf..07125e7941f4 100644 --- a/block/blk-softirq.c +++ b/block/blk-softirq.c @@ -60,7 +60,7 @@ static void trigger_softirq(void *data) static int raise_blk_irq(int cpu, struct request *rq) { if (cpu_online(cpu)) { - struct call_single_data *data = &rq->csd; + call_single_data_t *data = &rq->csd; data->func = trigger_softirq; data->info = rq; diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index 85c24cace973..81142ce781da 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -13,7 +13,7 @@ struct nullb_cmd { struct list_head list; struct llist_node ll_list; - struct call_single_data csd; + call_single_data_t csd; struct request *rq; struct bio *bio; unsigned int tag; diff --git a/drivers/cpuidle/coupled.c b/drivers/cpuidle/coupled.c index 71e586d7df71..147f38ea0fcd 100644 --- a/drivers/cpuidle/coupled.c +++ b/drivers/cpuidle/coupled.c @@ -119,13 +119,13 @@ struct cpuidle_coupled { #define CPUIDLE_COUPLED_NOT_IDLE (-1) -static DEFINE_PER_CPU(struct call_single_data, cpuidle_coupled_poke_cb); +static DEFINE_PER_CPU(call_single_data_t, cpuidle_coupled_poke_cb); /* * The cpuidle_coupled_poke_pending mask is used to avoid calling - * __smp_call_function_single with the per cpu call_single_data struct already + * __smp_call_function_single with the per cpu call_single_data_t struct already * in use. This prevents a deadlock where two cpus are waiting for each others - * call_single_data struct to be available + * call_single_data_t struct to be available */ static cpumask_t cpuidle_coupled_poke_pending; @@ -339,7 +339,7 @@ static void cpuidle_coupled_handle_poke(void *info) */ static void cpuidle_coupled_poke(int cpu) { - struct call_single_data *csd = &per_cpu(cpuidle_coupled_poke_cb, cpu); + call_single_data_t *csd = &per_cpu(cpuidle_coupled_poke_cb, cpu); if (!cpumask_test_and_set_cpu(cpu, &cpuidle_coupled_poke_pending)) smp_call_function_single_async(cpu, csd); @@ -651,7 +651,7 @@ int cpuidle_coupled_register_device(struct cpuidle_device *dev) { int cpu; struct cpuidle_device *other_dev; - struct call_single_data *csd; + call_single_data_t *csd; struct cpuidle_coupled *coupled; if (cpumask_empty(&dev->coupled_cpus)) diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c index 51583ae4b1eb..120b6e537b28 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_main.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c @@ -2468,7 +2468,7 @@ static void liquidio_napi_drv_callback(void *arg) if (OCTEON_CN23XX_PF(oct) || droq->cpu_id == this_cpu) { napi_schedule_irqoff(&droq->napi); } else { - struct call_single_data *csd = &droq->csd; + call_single_data_t *csd = &droq->csd; csd->func = napi_schedule_wrapper; csd->info = &droq->napi; diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h index 6efd139b894d..f91bc84d1719 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h @@ -328,7 +328,7 @@ struct octeon_droq { u32 cpu_id; - struct call_single_data csd; + call_single_data_t csd; }; #define OCT_DROQ_SIZE (sizeof(struct octeon_droq)) diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 25f6a0cb27d3..006fa09a641e 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -134,7 +134,7 @@ typedef __u32 __bitwise req_flags_t; struct request { struct list_head queuelist; union { - struct call_single_data csd; + call_single_data_t csd; u64 fifo_time; }; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 779b23595596..6557f320b66e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2774,7 +2774,7 @@ struct softnet_data { unsigned int input_queue_head ____cacheline_aligned_in_smp; /* Elements below can be accessed between CPUs for RPS/RFS */ - struct call_single_data csd ____cacheline_aligned_in_smp; + call_single_data_t csd ____cacheline_aligned_in_smp; struct softnet_data *rps_ipi_next; unsigned int cpu; unsigned int input_queue_tail; diff --git a/include/linux/smp.h b/include/linux/smp.h index 68123c1fe549..98b1fe027fc9 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -14,13 +14,17 @@ #include typedef void (*smp_call_func_t)(void *info); -struct call_single_data { +struct __call_single_data { struct llist_node llist; smp_call_func_t func; void *info; unsigned int flags; }; +/* Use __aligned() to avoid to use 2 cache lines for 1 csd */ +typedef struct __call_single_data call_single_data_t + __aligned(sizeof(struct __call_single_data)); + /* total number of cpus in this system (may exceed NR_CPUS) */ extern unsigned int total_cpus; @@ -48,7 +52,7 @@ void on_each_cpu_cond(bool (*cond_func)(int cpu, void *info), smp_call_func_t func, void *info, bool wait, gfp_t gfp_flags); -int smp_call_function_single_async(int cpu, struct call_single_data *csd); +int smp_call_function_single_async(int cpu, call_single_data_t *csd); #ifdef CONFIG_SMP diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index eeef1a3086d1..f29a7d2b57e1 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -769,7 +769,7 @@ struct rq { #ifdef CONFIG_SCHED_HRTICK #ifdef CONFIG_SMP int hrtick_csd_pending; - struct call_single_data hrtick_csd; + call_single_data_t hrtick_csd; #endif struct hrtimer hrtick_timer; #endif diff --git a/kernel/smp.c b/kernel/smp.c index 3061483cb3ad..81cfca9b4cc3 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -28,7 +28,7 @@ enum { }; struct call_function_data { - struct call_single_data __percpu *csd; + call_single_data_t __percpu *csd; cpumask_var_t cpumask; cpumask_var_t cpumask_ipi; }; @@ -51,7 +51,7 @@ int smpcfd_prepare_cpu(unsigned int cpu) free_cpumask_var(cfd->cpumask); return -ENOMEM; } - cfd->csd = alloc_percpu(struct call_single_data); + cfd->csd = alloc_percpu(call_single_data_t); if (!cfd->csd) { free_cpumask_var(cfd->cpumask); free_cpumask_var(cfd->cpumask_ipi); @@ -103,12 +103,12 @@ void __init call_function_init(void) * previous function call. For multi-cpu calls its even more interesting * as we'll have to ensure no other cpu is observing our csd. */ -static __always_inline void csd_lock_wait(struct call_single_data *csd) +static __always_inline void csd_lock_wait(call_single_data_t *csd) { smp_cond_load_acquire(&csd->flags, !(VAL & CSD_FLAG_LOCK)); } -static __always_inline void csd_lock(struct call_single_data *csd) +static __always_inline void csd_lock(call_single_data_t *csd) { csd_lock_wait(csd); csd->flags |= CSD_FLAG_LOCK; @@ -116,12 +116,12 @@ static __always_inline void csd_lock(struct call_single_data *csd) /* * prevent CPU from reordering the above assignment * to ->flags with any subsequent assignments to other - * fields of the specified call_single_data structure: + * fields of the specified call_single_data_t structure: */ smp_wmb(); } -static __always_inline void csd_unlock(struct call_single_data *csd) +static __always_inline void csd_unlock(call_single_data_t *csd) { WARN_ON(!(csd->flags & CSD_FLAG_LOCK)); @@ -131,14 +131,14 @@ static __always_inline void csd_unlock(struct call_single_data *csd) smp_store_release(&csd->flags, 0); } -static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data); +static DEFINE_PER_CPU_SHARED_ALIGNED(call_single_data_t, csd_data); /* - * Insert a previously allocated call_single_data element + * Insert a previously allocated call_single_data_t element * for execution on the given CPU. data must already have * ->func, ->info, and ->flags set. */ -static int generic_exec_single(int cpu, struct call_single_data *csd, +static int generic_exec_single(int cpu, call_single_data_t *csd, smp_call_func_t func, void *info) { if (cpu == smp_processor_id()) { @@ -210,7 +210,7 @@ static void flush_smp_call_function_queue(bool warn_cpu_offline) { struct llist_head *head; struct llist_node *entry; - struct call_single_data *csd, *csd_next; + call_single_data_t *csd, *csd_next; static bool warned; WARN_ON(!irqs_disabled()); @@ -268,8 +268,10 @@ static void flush_smp_call_function_queue(bool warn_cpu_offline) int smp_call_function_single(int cpu, smp_call_func_t func, void *info, int wait) { - struct call_single_data *csd; - struct call_single_data csd_stack = { .flags = CSD_FLAG_LOCK | CSD_FLAG_SYNCHRONOUS }; + call_single_data_t *csd; + call_single_data_t csd_stack = { + .flags = CSD_FLAG_LOCK | CSD_FLAG_SYNCHRONOUS, + }; int this_cpu; int err; @@ -321,7 +323,7 @@ EXPORT_SYMBOL(smp_call_function_single); * NOTE: Be careful, there is unfortunately no current debugging facility to * validate the correctness of this serialization. */ -int smp_call_function_single_async(int cpu, struct call_single_data *csd) +int smp_call_function_single_async(int cpu, call_single_data_t *csd) { int err = 0; @@ -444,7 +446,7 @@ void smp_call_function_many(const struct cpumask *mask, cpumask_clear(cfd->cpumask_ipi); for_each_cpu(cpu, cfd->cpumask) { - struct call_single_data *csd = per_cpu_ptr(cfd->csd, cpu); + call_single_data_t *csd = per_cpu_ptr(cfd->csd, cpu); csd_lock(csd); if (wait) @@ -460,7 +462,7 @@ void smp_call_function_many(const struct cpumask *mask, if (wait) { for_each_cpu(cpu, cfd->cpumask) { - struct call_single_data *csd; + call_single_data_t *csd; csd = per_cpu_ptr(cfd->csd, cpu); csd_lock_wait(csd); diff --git a/kernel/up.c b/kernel/up.c index ee81ac9af4ca..42c46bf3e0a5 100644 --- a/kernel/up.c +++ b/kernel/up.c @@ -23,7 +23,7 @@ int smp_call_function_single(int cpu, void (*func) (void *info), void *info, } EXPORT_SYMBOL(smp_call_function_single); -int smp_call_function_single_async(int cpu, struct call_single_data *csd) +int smp_call_function_single_async(int cpu, call_single_data_t *csd) { unsigned long flags; -- cgit v1.2.3 From ec81048cc340bb03334e6ca62661ecc0a684897a Mon Sep 17 00:00:00 2001 From: Boqun Feng Date: Wed, 23 Aug 2017 23:25:38 +0800 Subject: sched/completion: Avoid unnecessary stack allocation for COMPLETION_INITIALIZER_ONSTACK() In theory, COMPLETION_INITIALIZER_ONSTACK() should never affect the stack allocation of the caller. However, on some compilers, a temporary structure was allocated for the return value of COMPLETION_INITIALIZER_ONSTACK(). For example in write_journal() with LOCKDEP_COMPLETIONS=y (GCC is 7.1.1): io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp); 2462: e8 00 00 00 00 callq 2467 2467: 48 8d 85 80 fd ff ff lea -0x280(%rbp),%rax 246e: 48 c7 c6 00 00 00 00 mov $0x0,%rsi 2475: 48 c7 c2 00 00 00 00 mov $0x0,%rdx x->done = 0; 247c: c7 85 90 fd ff ff 00 movl $0x0,-0x270(%rbp) 2483: 00 00 00 init_waitqueue_head(&x->wait); 2486: 48 8d 78 18 lea 0x18(%rax),%rdi 248a: e8 00 00 00 00 callq 248f if (commit_start + commit_sections <= ic->journal_sections) { 248f: 41 8b 87 a8 00 00 00 mov 0xa8(%r15),%eax io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp); 2496: 48 8d bd e8 f9 ff ff lea -0x618(%rbp),%rdi 249d: 48 8d b5 90 fd ff ff lea -0x270(%rbp),%rsi 24a4: b9 17 00 00 00 mov $0x17,%ecx 24a9: f3 48 a5 rep movsq %ds:(%rsi),%es:(%rdi) if (commit_start + commit_sections <= ic->journal_sections) { 24ac: 41 39 c6 cmp %eax,%r14d io_comp.comp = COMPLETION_INITIALIZER_ONSTACK(io_comp.comp); 24af: 48 8d bd 90 fd ff ff lea -0x270(%rbp),%rdi 24b6: 48 8d b5 e8 f9 ff ff lea -0x618(%rbp),%rsi 24bd: b9 17 00 00 00 mov $0x17,%ecx 24c2: f3 48 a5 rep movsq %ds:(%rsi),%es:(%rdi) We can obviously see the temporary structure allocated, and the compiler also does two meaningless memcpy with "rep movsq". And according to: https://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html#Statement-Exprs The return value of a statement expression is returned by value, so the temporary variable is created in COMPLETION_INITIALIZER_ONSTACK(), and that's why the temporary structures are allocted. To fix this, make the brace block in COMPLETION_INITIALIZER_ONSTACK() return a pointer and dereference it outside the block rather than return the whole structure, in this way, we are able to teach the compiler not to do the unnecessary stack allocation. This could also reduce the stack size even if !LOCKDEP, for example in write_journal(), compiled with gcc 7.1.1, the result of command: objdump -d drivers/md/dm-integrity.o | ./scripts/checkstack.pl x86 before: 0x0000246a write_journal [dm-integrity.o]: 696 after: 0x00002b7a write_journal [dm-integrity.o]: 296 Reported-by: Arnd Bergmann Signed-off-by: Boqun Feng Signed-off-by: Peter Zijlstra (Intel) Acked-by: Arnd Bergmann Cc: Andrew Morton Cc: Byungchul Park Cc: Linus Torvalds Cc: Nicholas Piggin Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: walken@google.com Cc: willy@infradead.org Link: http://lkml.kernel.org/r/20170823152542.5150-3-boqun.feng@gmail.com Signed-off-by: Ingo Molnar --- include/linux/completion.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/completion.h b/include/linux/completion.h index 791f053f28b7..cae5400022a3 100644 --- a/include/linux/completion.h +++ b/include/linux/completion.h @@ -74,7 +74,7 @@ static inline void complete_release_commit(struct completion *x) {} #endif #define COMPLETION_INITIALIZER_ONSTACK(work) \ - ({ init_completion(&work); work; }) + (*({ init_completion(&work); &work; })) /** * DECLARE_COMPLETION - declare and initialize a completion structure -- cgit v1.2.3