diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2017-02-20 13:23:30 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2017-02-20 13:23:30 -0800 |
commit | 42e1b14b6e1455ece2ccbe474c25388d0230a590 (patch) | |
tree | 7f62d95f795a2ac5c183248dce39e75340ccfb76 /kernel | |
parent | 828cad8ea05d194d8a9452e0793261c2024c23a2 (diff) | |
parent | 95cb64c1fe61e70685a95f6260c8e9cd219fe08c (diff) |
Merge branch 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking updates from Ingo Molnar:
"The main changes in this cycle were:
- Implement wraparound-safe refcount_t and kref_t types based on
generic atomic primitives (Peter Zijlstra)
- Improve and fix the ww_mutex code (Nicolai Hähnle)
- Add self-tests to the ww_mutex code (Chris Wilson)
- Optimize percpu-rwsems with the 'rcuwait' mechanism (Davidlohr
Bueso)
- Micro-optimize the current-task logic all around the core kernel
(Davidlohr Bueso)
- Tidy up after recent optimizations: remove stale code and APIs,
clean up the code (Waiman Long)
- ... plus misc fixes, updates and cleanups"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (50 commits)
fork: Fix task_struct alignment
locking/spinlock/debug: Remove spinlock lockup detection code
lockdep: Fix incorrect condition to print bug msgs for MAX_LOCKDEP_CHAIN_HLOCKS
lkdtm: Convert to refcount_t testing
kref: Implement 'struct kref' using refcount_t
refcount_t: Introduce a special purpose refcount type
sched/wake_q: Clarify queue reinit comment
sched/wait, rcuwait: Fix typo in comment
locking/mutex: Fix lockdep_assert_held() fail
locking/rtmutex: Flip unlikely() branch to likely() in __rt_mutex_slowlock()
locking/rwsem: Reinit wake_q after use
locking/rwsem: Remove unnecessary atomic_long_t casts
jump_labels: Move header guard #endif down where it belongs
locking/atomic, kref: Implement kref_put_lock()
locking/ww_mutex: Turn off __must_check for now
locking/atomic, kref: Avoid more abuse
locking/atomic, kref: Use kref_get_unless_zero() more
locking/atomic, kref: Kill kref_sub()
locking/atomic, kref: Add kref_read()
locking/atomic, kref: Add KREF_INIT()
...
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/exit.c | 52 | ||||
-rw-r--r-- | kernel/fork.c | 6 | ||||
-rw-r--r-- | kernel/locking/Makefile | 1 | ||||
-rw-r--r-- | kernel/locking/lockdep.c | 2 | ||||
-rw-r--r-- | kernel/locking/locktorture.c | 73 | ||||
-rw-r--r-- | kernel/locking/mutex-debug.h | 17 | ||||
-rw-r--r-- | kernel/locking/mutex.c | 516 | ||||
-rw-r--r-- | kernel/locking/mutex.h | 4 | ||||
-rw-r--r-- | kernel/locking/percpu-rwsem.c | 7 | ||||
-rw-r--r-- | kernel/locking/qspinlock_paravirt.h | 2 | ||||
-rw-r--r-- | kernel/locking/rtmutex.c | 2 | ||||
-rw-r--r-- | kernel/locking/rwsem-spinlock.c | 18 | ||||
-rw-r--r-- | kernel/locking/rwsem-xadd.c | 14 | ||||
-rw-r--r-- | kernel/locking/semaphore.c | 7 | ||||
-rw-r--r-- | kernel/locking/spinlock.c | 8 | ||||
-rw-r--r-- | kernel/locking/spinlock_debug.c | 86 | ||||
-rw-r--r-- | kernel/locking/test-ww_mutex.c | 646 | ||||
-rw-r--r-- | kernel/pid.c | 4 |
18 files changed, 1127 insertions, 338 deletions
diff --git a/kernel/exit.c b/kernel/exit.c index 8e5e21338b3a..b67c57faa705 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -55,6 +55,7 @@ #include <linux/shm.h> #include <linux/kcov.h> #include <linux/random.h> +#include <linux/rcuwait.h> #include <linux/uaccess.h> #include <asm/unistd.h> @@ -282,6 +283,35 @@ retry: return task; } +void rcuwait_wake_up(struct rcuwait *w) +{ + struct task_struct *task; + + rcu_read_lock(); + + /* + * Order condition vs @task, such that everything prior to the load + * of @task is visible. This is the condition as to why the user called + * rcuwait_trywake() in the first place. Pairs with set_current_state() + * barrier (A) in rcuwait_wait_event(). + * + * WAIT WAKE + * [S] tsk = current [S] cond = true + * MB (A) MB (B) + * [L] cond [L] tsk + */ + smp_rmb(); /* (B) */ + + /* + * Avoid using task_rcu_dereference() magic as long as we are careful, + * see comment in rcuwait_wait_event() regarding ->exit_state. + */ + task = rcu_dereference(w->task); + if (task) + wake_up_process(task); + rcu_read_unlock(); +} + struct task_struct *try_get_task_struct(struct task_struct **ptask) { struct task_struct *task; @@ -468,12 +498,12 @@ assign_new_owner: * Turn us into a lazy TLB process if we * aren't already.. */ -static void exit_mm(struct task_struct *tsk) +static void exit_mm(void) { - struct mm_struct *mm = tsk->mm; + struct mm_struct *mm = current->mm; struct core_state *core_state; - mm_release(tsk, mm); + mm_release(current, mm); if (!mm) return; sync_mm_rss(mm); @@ -491,7 +521,7 @@ static void exit_mm(struct task_struct *tsk) up_read(&mm->mmap_sem); - self.task = tsk; + self.task = current; self.next = xchg(&core_state->dumper.next, &self); /* * Implies mb(), the result of xchg() must be visible @@ -501,22 +531,22 @@ static void exit_mm(struct task_struct *tsk) complete(&core_state->startup); for (;;) { - set_task_state(tsk, TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE); if (!self.task) /* see coredump_finish() */ break; freezable_schedule(); } - __set_task_state(tsk, TASK_RUNNING); + __set_current_state(TASK_RUNNING); down_read(&mm->mmap_sem); } atomic_inc(&mm->mm_count); - BUG_ON(mm != tsk->active_mm); + BUG_ON(mm != current->active_mm); /* more a memory barrier than a real lock */ - task_lock(tsk); - tsk->mm = NULL; + task_lock(current); + current->mm = NULL; up_read(&mm->mmap_sem); enter_lazy_tlb(mm, current); - task_unlock(tsk); + task_unlock(current); mm_update_next_owner(mm); mmput(mm); if (test_thread_flag(TIF_MEMDIE)) @@ -823,7 +853,7 @@ void __noreturn do_exit(long code) tsk->exit_code = code; taskstats_exit(tsk, group_dead); - exit_mm(tsk); + exit_mm(); if (group_dead) acct_process(); diff --git a/kernel/fork.c b/kernel/fork.c index f6995cdfe714..ff82e24573b6 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -432,11 +432,13 @@ void __init fork_init(void) int i; #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR #ifndef ARCH_MIN_TASKALIGN -#define ARCH_MIN_TASKALIGN L1_CACHE_BYTES +#define ARCH_MIN_TASKALIGN 0 #endif + int align = max_t(int, L1_CACHE_BYTES, ARCH_MIN_TASKALIGN); + /* create a slab on which task_structs can be allocated */ task_struct_cachep = kmem_cache_create("task_struct", - arch_task_struct_size, ARCH_MIN_TASKALIGN, + arch_task_struct_size, align, SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, NULL); #endif diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile index 6f88e352cd4f..760158d9d98d 100644 --- a/kernel/locking/Makefile +++ b/kernel/locking/Makefile @@ -28,3 +28,4 @@ obj-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o obj-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem-xadd.o obj-$(CONFIG_QUEUED_RWLOCKS) += qrwlock.o obj-$(CONFIG_LOCK_TORTURE_TEST) += locktorture.o +obj-$(CONFIG_WW_MUTEX_SELFTEST) += test-ww_mutex.o diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index d9a698e8458f..9812e5dd409e 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -2203,7 +2203,7 @@ cache_hit: * Important for check_no_collision(). */ if (unlikely(nr_chain_hlocks > MAX_LOCKDEP_CHAIN_HLOCKS)) { - if (debug_locks_off_graph_unlock()) + if (!debug_locks_off_graph_unlock()) return 0; print_lockdep_off("BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!"); diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c index d3de04b12f8c..28350dc8ecbb 100644 --- a/kernel/locking/locktorture.c +++ b/kernel/locking/locktorture.c @@ -372,6 +372,78 @@ static struct lock_torture_ops mutex_lock_ops = { .name = "mutex_lock" }; +#include <linux/ww_mutex.h> +static DEFINE_WW_CLASS(torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_0, &torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_1, &torture_ww_class); +static DEFINE_WW_MUTEX(torture_ww_mutex_2, &torture_ww_class); + +static int torture_ww_mutex_lock(void) +__acquires(torture_ww_mutex_0) +__acquires(torture_ww_mutex_1) +__acquires(torture_ww_mutex_2) +{ + LIST_HEAD(list); + struct reorder_lock { + struct list_head link; + struct ww_mutex *lock; + } locks[3], *ll, *ln; + struct ww_acquire_ctx ctx; + + locks[0].lock = &torture_ww_mutex_0; + list_add(&locks[0].link, &list); + + locks[1].lock = &torture_ww_mutex_1; + list_add(&locks[1].link, &list); + + locks[2].lock = &torture_ww_mutex_2; + list_add(&locks[2].link, &list); + + ww_acquire_init(&ctx, &torture_ww_class); + + list_for_each_entry(ll, &list, link) { + int err; + + err = ww_mutex_lock(ll->lock, &ctx); + if (!err) + continue; + + ln = ll; + list_for_each_entry_continue_reverse(ln, &list, link) + ww_mutex_unlock(ln->lock); + + if (err != -EDEADLK) + return err; + + ww_mutex_lock_slow(ll->lock, &ctx); + list_move(&ll->link, &list); + } + + ww_acquire_fini(&ctx); + return 0; +} + +static void torture_ww_mutex_unlock(void) +__releases(torture_ww_mutex_0) +__releases(torture_ww_mutex_1) +__releases(torture_ww_mutex_2) +{ + ww_mutex_unlock(&torture_ww_mutex_0); + ww_mutex_unlock(&torture_ww_mutex_1); + ww_mutex_unlock(&torture_ww_mutex_2); +} + +static struct lock_torture_ops ww_mutex_lock_ops = { + .writelock = torture_ww_mutex_lock, + .write_delay = torture_mutex_delay, + .task_boost = torture_boost_dummy, + .writeunlock = torture_ww_mutex_unlock, + .readlock = NULL, + .read_delay = NULL, + .readunlock = NULL, + .name = "ww_mutex_lock" +}; + #ifdef CONFIG_RT_MUTEXES static DEFINE_RT_MUTEX(torture_rtmutex); @@ -797,6 +869,7 @@ static int __init lock_torture_init(void) &spin_lock_ops, &spin_lock_irq_ops, &rw_lock_ops, &rw_lock_irq_ops, &mutex_lock_ops, + &ww_mutex_lock_ops, #ifdef CONFIG_RT_MUTEXES &rtmutex_lock_ops, #endif diff --git a/kernel/locking/mutex-debug.h b/kernel/locking/mutex-debug.h index a459faa48987..4174417d5309 100644 --- a/kernel/locking/mutex-debug.h +++ b/kernel/locking/mutex-debug.h @@ -26,20 +26,3 @@ extern void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, extern void debug_mutex_unlock(struct mutex *lock); extern void debug_mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key); - -#define spin_lock_mutex(lock, flags) \ - do { \ - struct mutex *l = container_of(lock, struct mutex, wait_lock); \ - \ - DEBUG_LOCKS_WARN_ON(in_interrupt()); \ - local_irq_save(flags); \ - arch_spin_lock(&(lock)->rlock.raw_lock);\ - DEBUG_LOCKS_WARN_ON(l->magic != l); \ - } while (0) - -#define spin_unlock_mutex(lock, flags) \ - do { \ - arch_spin_unlock(&(lock)->rlock.raw_lock); \ - local_irq_restore(flags); \ - preempt_check_resched(); \ - } while (0) diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c index 8464a5cbab97..ad2d9e22697b 100644 --- a/kernel/locking/mutex.c +++ b/kernel/locking/mutex.c @@ -50,16 +50,17 @@ EXPORT_SYMBOL(__mutex_init); /* * @owner: contains: 'struct task_struct *' to the current lock owner, * NULL means not owned. Since task_struct pointers are aligned at - * ARCH_MIN_TASKALIGN (which is at least sizeof(void *)), we have low - * bits to store extra state. + * at least L1_CACHE_BYTES, we have low bits to store extra state. * * Bit0 indicates a non-empty waiter list; unlock must issue a wakeup. * Bit1 indicates unlock needs to hand the lock to the top-waiter + * Bit2 indicates handoff has been done and we're waiting for pickup. */ #define MUTEX_FLAG_WAITERS 0x01 #define MUTEX_FLAG_HANDOFF 0x02 +#define MUTEX_FLAG_PICKUP 0x04 -#define MUTEX_FLAGS 0x03 +#define MUTEX_FLAGS 0x07 static inline struct task_struct *__owner_task(unsigned long owner) { @@ -72,38 +73,29 @@ static inline unsigned long __owner_flags(unsigned long owner) } /* - * Actual trylock that will work on any unlocked state. - * - * When setting the owner field, we must preserve the low flag bits. - * - * Be careful with @handoff, only set that in a wait-loop (where you set - * HANDOFF) to avoid recursive lock attempts. + * Trylock variant that retuns the owning task on failure. */ -static inline bool __mutex_trylock(struct mutex *lock, const bool handoff) +static inline struct task_struct *__mutex_trylock_or_owner(struct mutex *lock) { unsigned long owner, curr = (unsigned long)current; owner = atomic_long_read(&lock->owner); for (;;) { /* must loop, can race against a flag */ unsigned long old, flags = __owner_flags(owner); + unsigned long task = owner & ~MUTEX_FLAGS; - if (__owner_task(owner)) { - if (handoff && unlikely(__owner_task(owner) == current)) { - /* - * Provide ACQUIRE semantics for the lock-handoff. - * - * We cannot easily use load-acquire here, since - * the actual load is a failed cmpxchg, which - * doesn't imply any barriers. - * - * Also, this is a fairly unlikely scenario, and - * this contains the cost. - */ - smp_mb(); /* ACQUIRE */ - return true; - } + if (task) { + if (likely(task != curr)) + break; - return false; + if (likely(!(flags & MUTEX_FLAG_PICKUP))) + break; + + flags &= ~MUTEX_FLAG_PICKUP; + } else { +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(flags & MUTEX_FLAG_PICKUP); +#endif } /* @@ -111,15 +103,24 @@ static inline bool __mutex_trylock(struct mutex *lock, const bool handoff) * past the point where we acquire it. This would be possible * if we (accidentally) set the bit on an unlocked mutex. */ - if (handoff) - flags &= ~MUTEX_FLAG_HANDOFF; + flags &= ~MUTEX_FLAG_HANDOFF; old = atomic_long_cmpxchg_acquire(&lock->owner, owner, curr | flags); if (old == owner) - return true; + return NULL; owner = old; } + + return __owner_task(owner); +} + +/* + * Actual trylock that will work on any unlocked state. + */ +static inline bool __mutex_trylock(struct mutex *lock) +{ + return !__mutex_trylock_or_owner(lock); } #ifndef CONFIG_DEBUG_LOCK_ALLOC @@ -171,9 +172,9 @@ static inline bool __mutex_waiter_is_first(struct mutex *lock, struct mutex_wait /* * Give up ownership to a specific task, when @task = NULL, this is equivalent - * to a regular unlock. Clears HANDOFF, preserves WAITERS. Provides RELEASE - * semantics like a regular unlock, the __mutex_trylock() provides matching - * ACQUIRE semantics for the handoff. + * to a regular unlock. Sets PICKUP on a handoff, clears HANDOF, preserves + * WAITERS. Provides RELEASE semantics like a regular unlock, the + * __mutex_trylock() provides a matching ACQUIRE semantics for the handoff. */ static void __mutex_handoff(struct mutex *lock, struct task_struct *task) { @@ -184,10 +185,13 @@ static void __mutex_handoff(struct mutex *lock, struct task_struct *task) #ifdef CONFIG_DEBUG_MUTEXES DEBUG_LOCKS_WARN_ON(__owner_task(owner) != current); + DEBUG_LOCKS_WARN_ON(owner & MUTEX_FLAG_PICKUP); #endif new = (owner & MUTEX_FLAG_WAITERS); new |= (unsigned long)task; + if (task) + new |= MUTEX_FLAG_PICKUP; old = atomic_long_cmpxchg_release(&lock->owner, owner, new); if (old == owner) @@ -237,8 +241,8 @@ void __sched mutex_lock(struct mutex *lock) EXPORT_SYMBOL(mutex_lock); #endif -static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww, - struct ww_acquire_ctx *ww_ctx) +static __always_inline void +ww_mutex_lock_acquired(struct ww_mutex *ww, struct ww_acquire_ctx *ww_ctx) { #ifdef CONFIG_DEBUG_MUTEXES /* @@ -277,17 +281,50 @@ static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww, ww_ctx->acquired++; } +static inline bool __sched +__ww_ctx_stamp_after(struct ww_acquire_ctx *a, struct ww_acquire_ctx *b) +{ + return a->stamp - b->stamp <= LONG_MAX && + (a->stamp != b->stamp || a > b); +} + +/* + * Wake up any waiters that may have to back off when the lock is held by the + * given context. + * + * Due to the invariants on the wait list, this can only affect the first + * waiter with a context. + * + * The current task must not be on the wait list. + */ +static void __sched +__ww_mutex_wakeup_for_backoff(struct mutex *lock, struct ww_acquire_ctx *ww_ctx) +{ + struct mutex_waiter *cur; + + lockdep_assert_held(&lock->wait_lock); + + list_for_each_entry(cur, &lock->wait_list, list) { + if (!cur->ww_ctx) + continue; + + if (cur->ww_ctx->acquired > 0 && + __ww_ctx_stamp_after(cur->ww_ctx, ww_ctx)) { + debug_mutex_wake_waiter(lock, cur); + wake_up_process(cur->task); + } + + break; + } +} + /* * After acquiring lock with fastpath or when we lost out in contested * slowpath, set ctx and wake up any waiters so they can recheck. */ static __always_inline void -ww_mutex_set_context_fastpath(struct ww_mutex *lock, - struct ww_acquire_ctx *ctx) +ww_mutex_set_context_fastpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - unsigned long flags; - struct mutex_waiter *cur; - ww_mutex_lock_acquired(lock, ctx); lock->ctx = ctx; @@ -311,46 +348,79 @@ ww_mutex_set_context_fastpath(struct ww_mutex *lock, * Uh oh, we raced in fastpath, wake up everyone in this case, * so they can see the new lock->ctx. */ - spin_lock_mutex(&lock->base.wait_lock, flags); - list_for_each_entry(cur, &lock->base.wait_list, list) { - debug_mutex_wake_waiter(&lock->base, cur); - wake_up_process(cur->task); - } - spin_unlock_mutex(&lock->base.wait_lock, flags); + spin_lock(&lock->base.wait_lock); + __ww_mutex_wakeup_for_backoff(&lock->base, ctx); + spin_unlock(&lock->base.wait_lock); } /* - * After acquiring lock in the slowpath set ctx and wake up any - * waiters so they can recheck. + * After acquiring lock in the slowpath set ctx. + * + * Unlike for the fast path, the caller ensures that waiters are woken up where + * necessary. * * Callers must hold the mutex wait_lock. */ static __always_inline void -ww_mutex_set_context_slowpath(struct ww_mutex *lock, - struct ww_acquire_ctx *ctx) +ww_mutex_set_context_slowpath(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) { - struct mutex_waiter *cur; - ww_mutex_lock_acquired(lock, ctx); lock->ctx = ctx; +} + +#ifdef CONFIG_MUTEX_SPIN_ON_OWNER + +static inline +bool ww_mutex_spin_on_owner(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + struct mutex_waiter *waiter) +{ + struct ww_mutex *ww; + + ww = container_of(lock, struct ww_mutex, base); /* - * Give any possible sleeping processes the chance to wake up, - * so they can recheck if they have to back off. + * If ww->ctx is set the contents are undefined, only + * by acquiring wait_lock there is a guarantee that + * they are not invalid when reading. + * + * As such, when deadlock detection needs to be + * performed the optimistic spinning cannot be done. + * + * Check this in every inner iteration because we may + * be racing against another thread's ww_mutex_lock. */ - list_for_each_entry(cur, &lock->base.wait_list, list) { - debug_mutex_wake_waiter(&lock->base, cur); - wake_up_process(cur->task); - } + if (ww_ctx->acquired > 0 && READ_ONCE(ww->ctx)) + return false; + + /* + * If we aren't on the wait list yet, cancel the spin + * if there are waiters. We want to avoid stealing the + * lock from a waiter with an earlier stamp, since the + * other thread may already own a lock that we also + * need. + */ + if (!waiter && (atomic_long_read(&lock->owner) & MUTEX_FLAG_WAITERS)) + return false; + + /* + * Similarly, stop spinning if we are no longer the + * first waiter. + */ + if (waiter && !__mutex_waiter_is_first(lock, waiter)) + return false; + + return true; } -#ifdef CONFIG_MUTEX_SPIN_ON_OWNER /* - * Look out! "owner" is an entirely speculative pointer - * access and not reliable. + * Look out! "owner" is an entirely speculative pointer access and not + * reliable. + * + * "noinline" so that this function shows up on perf profiles. */ static noinline -bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) +bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner, + struct ww_acquire_ctx *ww_ctx, struct mutex_waiter *waiter) { bool ret = true; @@ -373,6 +443,11 @@ bool mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner) break; } + if (ww_ctx && !ww_mutex_spin_on_owner(lock, ww_ctx, waiter)) { + ret = false; + break; + } + cpu_relax(); } rcu_read_unlock(); @@ -431,12 +506,10 @@ static inline int mutex_can_spin_on_owner(struct mutex *lock) * with the spinner at the head of the OSQ, if present, until the owner is * changed to itself. */ -static bool mutex_optimistic_spin(struct mutex *lock, - struct ww_acquire_ctx *ww_ctx, - const bool use_ww_ctx, const bool waiter) +static __always_inline bool +mutex_optimistic_spin(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + const bool use_ww_ctx, struct mutex_waiter *waiter) { - struct task_struct *task = current; - if (!waiter) { /* * The purpose of the mutex_can_spin_on_owner() function is @@ -460,40 +533,17 @@ static bool mutex_optimistic_spin(struct mutex *lock, for (;;) { struct task_struct *owner; - if (use_ww_ctx && ww_ctx->acquired > 0) { - struct ww_mutex *ww; - - ww = container_of(lock, struct ww_mutex, base); - /* - * If ww->ctx is set the contents are undefined, only - * by acquiring wait_lock there is a guarantee that - * they are not invalid when reading. - * - * As such, when deadlock detection needs to be - * performed the optimistic spinning cannot be done. - */ - if (READ_ONCE(ww->ctx)) - goto fail_unlock; - } + /* Try to acquire the mutex... */ + owner = __mutex_trylock_or_owner(lock); + if (!owner) + break; /* - * If there's an owner, wait for it to either + * There's an owner, wait for it to either * release the lock or go to sleep. */ - owner = __mutex_owner(lock); - if (owner) { - if (waiter && owner == task) { - smp_mb(); /* ACQUIRE */ - break; - } - - if (!mutex_spin_on_owner(lock, owner)) - goto fail_unlock; - } - - /* Try to acquire the mutex if it is unlocked. */ - if (__mutex_trylock(lock, waiter)) - break; + if (!mutex_spin_on_owner(lock, owner, ww_ctx, waiter)) + goto fail_unlock; /* * The cpu_relax() call is a compiler barrier which forces @@ -532,9 +582,9 @@ fail: return false; } #else -static bool mutex_optimistic_spin(struct mutex *lock, - struct ww_acquire_ctx *ww_ctx, - const bool use_ww_ctx, const bool waiter) +static __always_inline bool +mutex_optimistic_spin(struct mutex *lock, struct ww_acquire_ctx *ww_ctx, + const bool use_ww_ctx, struct mutex_waiter *waiter) { return false; } @@ -594,23 +644,88 @@ void __sched ww_mutex_unlock(struct ww_mutex *lock) EXPORT_SYMBOL(ww_mutex_unlock); static inline int __sched -__ww_mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx) +__ww_mutex_lock_check_stamp(struct mutex *lock, struct mutex_waiter *waiter, + struct ww_acquire_ctx *ctx) { struct ww_mutex *ww = container_of(lock, struct ww_mutex, base); struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx); + struct mutex_waiter *cur; + + if (hold_ctx && __ww_ctx_stamp_after(ctx, hold_ctx)) + goto deadlock; - if (!hold_ctx) + /* + * If there is a waiter in front of us that has a context, then its + * stamp is earlier than ours and we must back off. + */ + cur = waiter; + list_for_each_entry_continue_reverse(cur, &lock->wait_list, list) { + if (cur->ww_ctx) + goto deadlock; + } + + return 0; + +deadlock: +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(ctx->contending_lock); + ctx->contending_lock = ww; +#endif + return -EDEADLK; +} + +static inline int __sched +__ww_mutex_add_waiter(struct mutex_waiter *waiter, + struct mutex *lock, + struct ww_acquire_ctx *ww_ctx) +{ + struct mutex_waiter *cur; + struct list_head *pos; + + if (!ww_ctx) { + list_add_tail(&waiter->list, &lock->wait_list); return 0; + } - if (ctx->stamp - hold_ctx->stamp <= LONG_MAX && - (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) { + /* + * Add the waiter before the first waiter with a higher stamp. + * Waiters without a context are skipped to avoid starving + * them. + */ + pos = &lock->wait_list; + list_for_each_entry_reverse(cur, &lock->wait_list, list) { + if (!cur->ww_ctx) + continue; + + if (__ww_ctx_stamp_after(ww_ctx, cur->ww_ctx)) { + /* Back off immediately if necessary. */ + if (ww_ctx->acquired > 0) { #ifdef CONFIG_DEBUG_MUTEXES - DEBUG_LOCKS_WARN_ON(ctx->contending_lock); - ctx->contending_lock = ww; + struct ww_mutex *ww; + + ww = container_of(lock, struct ww_mutex, base); + DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock); + ww_ctx->contending_lock = ww; #endif - return -EDEADLK; + return -EDEADLK; + } + + break; + } + + pos = &cur->list; + + /* + * Wake up the waiter so that it gets a chance to back + * off. + */ + if (cur->ww_ctx->acquired > 0) { + debug_mutex_wake_waiter(lock, cur); + wake_up_process(cur->task); + } } + list_add_tail(&waiter->list, pos); return 0; } @@ -622,15 +737,15 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, struct lockdep_map *nest_lock, unsigned long ip, struct ww_acquire_ctx *ww_ctx, const bool use_ww_ctx) { - struct task_struct *task = current; struct mutex_waiter waiter; - unsigned long flags; bool first = false; struct ww_mutex *ww; int ret; - if (use_ww_ctx) { - ww = container_of(lock, struct ww_mutex, base); + might_sleep(); + + ww = container_of(lock, struct ww_mutex, base); + if (use_ww_ctx && ww_ctx) { if (unlikely(ww_ctx == READ_ONCE(ww->ctx))) return -EALREADY; } @@ -638,36 +753,54 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, preempt_disable(); mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip); - if (__mutex_trylock(lock, false) || - mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, false)) { + if (__mutex_trylock(lock) || + mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, NULL)) { /* got the lock, yay! */ lock_acquired(&lock->dep_map, ip); - if (use_ww_ctx) + if (use_ww_ctx && ww_ctx) ww_mutex_set_context_fastpath(ww, ww_ctx); preempt_enable(); return 0; } - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); /* * After waiting to acquire the wait_lock, try again. */ - if (__mutex_trylock(lock, false)) + if (__mutex_trylock(lock)) { + if (use_ww_ctx && ww_ctx) + __ww_mutex_wakeup_for_backoff(lock, ww_ctx); + goto skip_wait; + } debug_mutex_lock_common(lock, &waiter); - debug_mutex_add_waiter(lock, &waiter, task); + debug_mutex_add_waiter(lock, &waiter, current); - /* add waiting tasks to the end of the waitqueue (FIFO): */ - list_add_tail(&waiter.list, &lock->wait_list); - waiter.task = task; + lock_contended(&lock->dep_map, ip); + + if (!use_ww_ctx) { + /* add waiting tasks to the end of the waitqueue (FIFO): */ + list_add_tail(&waiter.list, &lock->wait_list); + +#ifdef CONFIG_DEBUG_MUTEXES + waiter.ww_ctx = MUTEX_POISON_WW_CTX; +#endif + } else { + /* Add in stamp order, waking up waiters that must back off. */ + ret = __ww_mutex_add_waiter(&waiter, lock, ww_ctx); + if (ret) + goto err_early_backoff; + + waiter.ww_ctx = ww_ctx; + } + + waiter.task = current; if (__mutex_waiter_is_first(lock, &waiter)) __mutex_set_flag(lock, MUTEX_FLAG_WAITERS); - lock_contended(&lock->dep_map, ip); - - set_task_state(task, state); + set_current_state(state); for (;;) { /* * Once we hold wait_lock, we're serialized against @@ -675,7 +808,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, * before testing the error conditions to make sure we pick up * the handoff. */ - if (__mutex_trylock(lock, first)) + if (__mutex_trylock(lock)) goto acquired; /* @@ -683,42 +816,47 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, * wait_lock. This ensures the lock cancellation is ordered * against mutex_unlock() and wake-ups do not go missing. */ - if (unlikely(signal_pending_state(state, task))) { + if (unlikely(signal_pending_state(state, current))) { ret = -EINTR; goto err; } - if (use_ww_ctx && ww_ctx->acquired > 0) { - ret = __ww_mutex_lock_check_stamp(lock, ww_ctx); + if (use_ww_ctx && ww_ctx && ww_ctx->acquired > 0) { + ret = __ww_mutex_lock_check_stamp(lock, &waiter, ww_ctx); if (ret) goto err; } - spin_unlock_mutex(&lock->wait_lock, flags); + spin_unlock(&lock->wait_lock); schedule_preempt_disabled(); - if (!first && __mutex_waiter_is_first(lock, &waiter)) { - first = true; - __mutex_set_flag(lock, MUTEX_FLAG_HANDOFF); + /* + * ww_mutex needs to always recheck its position since its waiter + * list is not FIFO ordered. + */ + if ((use_ww_ctx && ww_ctx) || !first) { + first = __mutex_waiter_is_first(lock, &waiter); + if (first) + __mutex_set_flag(lock, MUTEX_FLAG_HANDOFF); } - set_task_state(task, state); + set_current_state(state); /* * Here we order against unlock; we must either see it change * state back to RUNNING and fall through the next schedule(), * or we must see its unlock and acquire. */ - if ((first && mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, true)) || - __mutex_trylock(lock, first)) + if (__mutex_trylock(lock) || + (first && mutex_optimistic_spin(lock, ww_ctx, use_ww_ctx, &waiter))) break; - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); } - spin_lock_mutex(&lock->wait_lock, flags); + spin_lock(&lock->wait_lock); acquired: - __set_task_state(task, TASK_RUNNING); + __set_current_state(TASK_RUNNING); - mutex_remove_waiter(lock, &waiter, task); + mutex_remove_waiter(lock, &waiter, current); if (likely(list_empty(&lock->wait_list))) __mutex_clear_flag(lock, MUTEX_FLAGS); @@ -728,30 +866,44 @@ skip_wait: /* got the lock - cleanup and rejoice! */ lock_acquired(&a |