Message-ID: <519DDDB2.80800@canonical.com>
Date:	Thu, 23 May 2013 11:13:22 +0200
From:	Maarten Lankhorst <maarten.lankhorst@...onical.com>
To:	Peter Zijlstra <peterz@...radead.org>
CC:	linux-kernel@...r.kernel.org, linux-arch@...r.kernel.org,
	x86@...nel.org, dri-devel@...ts.freedesktop.org,
	linaro-mm-sig@...ts.linaro.org, robclark@...il.com,
	rostedt@...dmis.org, tglx@...utronix.de, mingo@...e.hu,
	linux-media@...r.kernel.org, Dave Airlie <airlied@...hat.com>
Subject: Re: [PATCH v3 2/3] mutex: add support for wound/wait style locks,
 v3

Op 22-05-13 19:24, Maarten Lankhorst schreef:
> Hey,
>
> Op 22-05-13 18:18, Peter Zijlstra schreef:
>> On Wed, May 22, 2013 at 01:18:14PM +0200, Maarten Lankhorst wrote:
>>
>> Lacking the actual msg atm, I'm going to paste in here...
> Thanks for taking the time to review.
>>> Subject: [PATCH v3 2/3] mutex: add support for wound/wait style locks, v3
>>> From: Maarten Lankhorst <maarten.lankhorst@...xxxxxxxxxx>
>>>
>>> Changes since RFC patch v1:
>>>  - Updated to use atomic_long instead of atomic, since the reservation_id was a long.
>>>  - added mutex_reserve_lock_slow and mutex_reserve_lock_intr_slow
>>>  - removed mutex_locked_set_reservation_id (or w/e it was called)
>>> Changes since RFC patch v2:
>>>  - remove use of __mutex_lock_retval_arg, add warnings when using wrong combination of
>>>    mutex_(,reserve_)lock/unlock.
>>> Changes since v1:
>>>  - Add __always_inline to __mutex_lock_common, otherwise reservation paths can be
>>>    triggered from normal locks, because __builtin_constant_p might evaluate to false
>>>    for the constant 0 in that case. Tests for this have been added in the next patch.
>>>  - Updated documentation slightly.
>>> Changes since v2:
>>>  - Renamed everything to ww_mutex. (mlankhorst)
>>>  - Added ww_acquire_ctx and ww_class. (mlankhorst)
>>>  - Added a lot of checks for wrong api usage. (mlankhorst)
>>>  - Documentation updates. (danvet)
>>>
>>> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@...xxxxxxxxxx>
>>> Signed-off-by: Daniel Vetter <daniel.vetter@...xxxxx>
>>> ---
>>>  Documentation/ww-mutex-design.txt |  322 +++++++++++++++++++++++++++
>>>  include/linux/mutex-debug.h       |    1 
>>>  include/linux/mutex.h             |  257 +++++++++++++++++++++
>>>  kernel/mutex.c                    |  445 ++++++++++++++++++++++++++++++++++++-
>>>  lib/debug_locks.c                 |    2 
>>>  5 files changed, 1010 insertions(+), 17 deletions(-)
>>>  create mode 100644 Documentation/ww-mutex-design.txt
>>>
>>> diff --git a/Documentation/ww-mutex-design.txt b/Documentation/ww-mutex-design.txt
>>> new file mode 100644
>>> index 0000000..154bae3
>>> --- /dev/null
>>> +++ b/Documentation/ww-mutex-design.txt
>>> @@ -0,0 +1,322 @@
>>> +Wait/Wound Deadlock-Proof Mutex Design
>>> +======================================
>>> +
>>> +Please read mutex-design.txt first, as it applies to wait/wound mutexes too.
>>> +
>>> +Motivation for WW-Mutexes
>>> +-------------------------
>>> +
>>> +GPU's do operations that commonly involve many buffers.  Those buffers
>>> +can be shared across contexts/processes, exist in different memory
>>> +domains (for example VRAM vs system memory), and so on.  And with
>>> +PRIME / dmabuf, they can even be shared across devices.  So there are
>>> +a handful of situations where the driver needs to wait for buffers to
>>> +become ready.  If you think about this in terms of waiting on a buffer
>>> +mutex for it to become available, this presents a problem because
>>> +there is no way to guarantee that buffers appear in a execbuf/batch in
>>> +the same order in all contexts.  That is directly under control of
>>> +userspace, and a result of the sequence of GL calls that an application
>>> +makes.	Which results in the potential for deadlock.  The problem gets
>>> +more complex when you consider that the kernel may need to migrate the
>>> +buffer(s) into VRAM before the GPU operates on the buffer(s), which
>>> +may in turn require evicting some other buffers (and you don't want to
>>> +evict other buffers which are already queued up to the GPU), but for a
>>> +simplified understanding of the problem you can ignore this.
>>> +
>>> +The algorithm that TTM came up with for dealing with this problem is quite
>>> +simple.  For each group of buffers (execbuf) that need to be locked, the caller
>>> +would be assigned a unique reservation id/ticket, from a global counter.  In
>>> +case of deadlock while locking all the buffers associated with a execbuf, the
>>> +one with the lowest reservation ticket (i.e. the oldest task) wins, and the one
>>> +with the higher reservation id (i.e. the younger task) unlocks all of the
>>> +buffers that it has already locked, and then tries again.
>>> +
>>> +In the RDBMS literature this deadlock handling approach is called wait/wound:
>>> +The older tasks waits until it can acquire the contended lock. The younger tasks
>>> +needs to back off and drop all the locks it is currently holding, i.e. the
>>> +younger task is wounded.
>>> +
>>> +Concepts
>>> +--------
>>> +
>>> +Compared to normal mutexes two additional concepts/objects show up in the lock
>>> +interface for w/w mutexes:
>>> +
>>> +Acquire context: To ensure eventual forward progress it is important the a task
>>> +trying to acquire locks doesn't grab a new reservation id, but keeps the one it
>>> +acquired when starting the lock acquisition. This ticket is stored in the
>>> +acquire context. Furthermore the acquire context keeps track of debugging state
>>> +to catch w/w mutex interface abuse.
>>> +
>>> +W/w class: In contrast to normal mutexes the lock class needs to be explicit for
>>> +w/w mutexes, since it is required to initialize the acquire context.
>>> +
>>> +Furthermore there are three different classe of w/w lock acquire functions:
>>> +- Normal lock acquisition with a context, using ww_mutex_lock
>>> +- Slowpath lock acquisition on the contending lock, used by the wounded task
>>> +  after having dropped all already acquired locks. These functions have the
>>> +  _slow postfix.
>> See below, I don't see the need for this interface.
>>
>>> +- Functions to only acquire a single w/w mutex, which results in the exact same
>>> +  semantics as a normal mutex. These functions have the _single postfix.
>> This is missing rationale.
> trylock_single is useful when iterating over a list, and you want to evict a bo, but only the first one that can be acquired.
> lock_single is useful when only a single bo needs to be acquired, for example to lock a buffer during mmap.
>
>>> +
>>> +Of course, all the usual variants for handling wake-ups due to signals are also
>>> +provided.
>>> +
>>> +Usage
>>> +-----
>>> +
>>> +Three different ways to acquire locks within the same w/w class. Common
>>> +definitions for methods 1&2.
>>> +
>>> +static DEFINE_WW_CLASS(ww_class);
>>> +
>>> +struct obj {
>>> +	sct ww_mutex lock;
>>> +	/* obj data */
>>> +};
>>> +
>>> +struct obj_entry {
>>> +	struct list_head *list;
>>> +	struct obj *obj;
>>> +};
>>> +
>>> +Method 1, using a list in execbuf->buffers that's not allowed to be reordered.
>>> +This is useful if a list of required objects is already tracked somewhere.
>>> +Furthermore the lock helper can use propagate the -EALREADY return code back to
>>> +the caller as a signal that an object is twice on the list. This is useful if
>>> +the list is constructed from userspace input and the ABI requires userspace to
>>> +no have duplicate entries (e.g. for a gpu commandbuffer submission ioctl).
>>> +
>>> +int lock_objs(struct list_head *list, struct ww_acquire_ctx *ctx)
>>> +{
>>> +	struct obj *res_obj = NULL;
>>> +	struct obj_entry *contended_entry = NULL;
>>> +	struct obj_entry *entry;
>>> +
>>> +	ww_acquire_init(ctx, &ww_class);
>>> +
>>> +retry:
>>> +	list_for_each_entry (list, entry) {
>>> +		if (entry == res_obj) {
>>> +			res_obj = NULL;
>>> +			continue;
>>> +		}
>>> +		ret = ww_mutex_lock(&entry->obj->lock, ctx);
>>> +		if (ret < 0) {
>>> +			contended_obj = entry;
>>> +			goto err;
>>> +		}
>>> +	}
>>> +
>>> +	ww_acquire_done(ctx);
>>> +	return 0;
>>> +
>>> +err:
>>> +	list_for_each_entry_continue_reverse (list, contended_entry, entry)
>>> +		ww_mutex_unlock(&entry->obj->lock);
>>> +
>>> +	if (res_obj)
>>> +		ww_mutex_unlock(&res_obj->lock);
>>> +
>>> +	if (ret == -EDEADLK) {
>>> +		/* we lost out in a seqno race, lock and retry.. */
>>> +		ww_mutex_lock_slow(&contended_entry->obj->lock, ctx);
>> I missing the need for ww_mutex_lock_slow(). AFAICT we should be able to tell
>> its the first lock in the ctx and thus we cannot possibly deadlock.
> Theoretically true, but that would require always setting ctx->acquired correctly.
> Plus that would weaken the checks. Without ww_mutex_lock_slow you can not
> say for sure all mutexes have been unlocked, and tell that what you say is really true.
>
>>> +		res_obj = contended_entry->obj;
>>> +		goto retry;
>>> +	}
>>> +	ww_acquire_fini(ctx);
>>> +
>>> +	return ret;
>>> +}
>>> +
>> ... you certainly went all out on documentation.
>>
>>> diff --git a/include/linux/mutex-debug.h b/include/linux/mutex-debug.h
>>> index 731d77d..4ac8b19 100644
>>> --- a/include/linux/mutex-debug.h
>>> +++ b/include/linux/mutex-debug.h
>>> @@ -3,6 +3,7 @@
>>>  
>>>  #include <linux/linkage.h>
>>>  #include <linux/lockdep.h>
>>> +#include <linux/debug_locks.h>
>>>  
>>>  /*
>>>   * Mutexes - debugging helpers:
>>> diff --git a/include/linux/mutex.h b/include/linux/mutex.h
>>> index 9121595..004f863 100644
>>> --- a/include/linux/mutex.h
>>> +++ b/include/linux/mutex.h
>>> @@ -74,6 +74,35 @@ struct mutex_waiter {
>>>  #endif
>>>  };
>>>  
>>> +struct ww_class {
>>> +	atomic_long_t stamp;
>>> +	struct lock_class_key acquire_key;
>>> +	struct lock_class_key mutex_key;
>>> +	const char *acquire_name;
>>> +	const char *mutex_name;
>>> +};
>>> +
>>> +struct ww_acquire_ctx {
>>> +	struct task_struct *task;
>>> +	unsigned long stamp;
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +	unsigned acquired, done_acquire;
>>> +	struct ww_class *ww_class;
>>> +	struct ww_mutex *contending_lock;
>>> +#endif
>>> +#ifdef CONFIG_DEBUG_LOCK_ALLOC
>>> +	struct lockdep_map dep_map;
>>> +#endif
>>> +};
>>> +
>>> +struct ww_mutex {
>>> +	struct mutex base;
>>> +	struct ww_acquire_ctx *ctx;
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +	struct ww_class *ww_class;
>>> +#endif
>>> +};
>>> +
>>> @@ -167,6 +236,192 @@ extern int __must_check mutex_lock_killable(struct mutex *lock);
>>>   */
>>>  extern int mutex_trylock(struct mutex *lock);
>>>  extern void mutex_unlock(struct mutex *lock);
>>> +
>>> +/**
>>> + * ww_acquire_init - initialize a w/w acquire context
>>> + * @ctx: w/w acquire context to initialize
>>> + * @ww_class: w/w class of the context
>>> + *
>>> + * Initializes an context to acquire multiple mutexes of the given w/w class.
>>> + *
>>> + * Context-based w/w mutex acquiring can be done in any order whatsoever within
>>> + * a given lock class. Deadlocks will be detected and handled with the
>>> + * wait/wound logic.
>>> + *
>>> + * Mixing of context-based w/w mutex acquiring and single w/w mutex locking can
>>> + * result in undetected deadlocks and is so forbidden. Mixing different contexts
>>> + * for the same w/w class when acquiring mutexes can also result in undetected
>>> + * deadlocks, and is hence also forbidden.
>>> + *
>>> + * Nesting of acquire contexts for _different_ w/w classes is possible, subject
>>> + * to the usual locking rules between different lock classes.
>>> + *
>>> + * An acquire context must be release by the same task before the memory is
>>> + * freed with ww_acquire_fini. It is recommended to allocate the context itself
>>> + * on the stack.
>>> + */
>>> +static inline void ww_acquire_init(struct ww_acquire_ctx *ctx,
>>> +				   struct ww_class *ww_class)
>>> +{
>>> +	ctx->task = current;
>>> +	do {
>>> +		ctx->stamp = atomic_long_inc_return(&ww_class->stamp);
>>> +	} while (unlikely(!ctx->stamp));
>> I suppose we'll figure something out when this becomes a bottleneck. Ideally
>> we'd do something like:
>>
>>  ctx->stamp = local_clock();
>>
>> but for now we cannot guarantee that's not jiffies, and I suppose that's a tad
>> too coarse to work for this.
> This might mess up when 2 cores happen to return exactly the same time, how do you choose a winner in that case?
> EDIT: Using pointer address like you suggested below is fine with me. ctx pointer would be static enough.
>
>> Also, why is 0 special?
> Oops, 0 is no longer special.
>
> I used to set the samp directly on the lock, so 0 used to mean no ctx set.
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +	ctx->ww_class = ww_class;
>>> +	ctx->acquired = ctx->done_acquire = 0;
>>> +	ctx->contending_lock = NULL;
>>> +#endif
>>> +#ifdef CONFIG_DEBUG_LOCK_ALLOC
>>> +	debug_check_no_locks_freed((void *)ctx, sizeof(*ctx));
>>> +	lockdep_init_map(&ctx->dep_map, ww_class->acquire_name,
>>> +			 &ww_class->acquire_key, 0);
>>> +	mutex_acquire(&ctx->dep_map, 0, 0, _RET_IP_);
>>> +#endif
>>> +}
>>> +/**
>>> + * ww_mutex_trylock_single - tries to acquire the w/w mutex without acquire context
>>> + * @lock: mutex to lock
>>> + *
>>> + * Trylocks a mutex without acquire context, so no deadlock detection is
>>> + * possible. Returns 0 if the mutex has been acquired.
>>> + *
>>> + * Unlocking the mutex must happen with a call to ww_mutex_unlock_single.
>>> + */
>>> +static inline int __must_check ww_mutex_trylock_single(struct ww_mutex *lock)
>>> +{
>>> +	return mutex_trylock(&lock->base);
>>> +}
>> trylocks can never deadlock they don't block per definition, I don't see the
>> point of the _single() thing here.
> I called it single because they weren't annotated into any ctx. I can drop the _single suffix though,
> but you'd still need to unlock with unlock_single, or we need to remove that distinction altogether,
> lose a few lockdep checks and only have a one unlock function.
>
>>> +/**
>>> + * ww_mutex_lock_single - acquire the w/w mutex without acquire context
>>> + * @lock: mutex to lock
>>> + *
>>> + * Locks a mutex without acquire context, so no deadlock detection is
>>> + * possible.
>>> + *
>>> + * Unlocking the mutex must happen with a call to ww_mutex_unlock_single.
>>> + */
>>> +static inline void ww_mutex_lock_single(struct ww_mutex *lock)
>>> +{
>>> +	mutex_lock(&lock->base);
>>> +}
>> as per the above, I'm missing the rationale for having this.
>>
>>> diff --git a/kernel/mutex.c b/kernel/mutex.c
>>> index 84a5f07..66807c7 100644
>>> --- a/kernel/mutex.c
>>> +++ b/kernel/mutex.c
>>> @@ -127,16 +127,156 @@ void __sched mutex_unlock(struct mutex *lock)
>>>  
>>>  EXPORT_SYMBOL(mutex_unlock);
>>>  
>>> +/**
>>> + * ww_mutex_unlock - release the w/w mutex
>>> + * @lock: the mutex to be released
>>> + *
>>> + * Unlock a mutex that has been locked by this task previously
>>> + * with ww_mutex_lock* using an acquire context. It is forbidden to release the
>>> + * locks after releasing the acquire context.
>>> + *
>>> + * This function must not be used in interrupt context. Unlocking
>>> + * of a unlocked mutex is not allowed.
>>> + *
>>> + * Note that locks acquired with one of the ww_mutex_lock*single variant must be
>>> + * unlocked with ww_mutex_unlock_single.
>>> + */
>>> +void __sched ww_mutex_unlock(struct ww_mutex *lock)
>>> +{
>>> +	/*
>>> +	 * The unlocking fastpath is the 0->1 transition from 'locked'
>>> +	 * into 'unlocked' state:
>>> +	 */
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +	DEBUG_LOCKS_WARN_ON(!lock->ctx);
>>> +	if (lock->ctx) {
>>> +		DEBUG_LOCKS_WARN_ON(!lock->ctx->acquired);
>>> +		if (lock->ctx->acquired > 0)
>>> +			lock->ctx->acquired--;
>>> +	}
>>> +#endif
>>> +	lock->ctx = NULL;
>> barriers should always have a comment explaining the exact ordering and the
>> pairing barrier's location.
>>
>>> +	smp_mb__before_atomic_inc();
>>> +
>>> +#ifndef CONFIG_DEBUG_MUTEXES
>>> +	/*
>>> +	 * When debugging is enabled we must not clear the owner before time,
>>> +	 * the slow path will always be taken, and that clears the owner field
>>> +	 * after verifying that it was indeed current.
>>> +	 */
>>> +	mutex_clear_owner(&lock->base);
>>> +#endif
>>> +	__mutex_fastpath_unlock(&lock->base.count, __mutex_unlock_slowpath);
>>> +}
>>> +EXPORT_SYMBOL(ww_mutex_unlock);
>>> +
>>> +static inline int __sched
>>> +__mutex_lock_check_stamp(struct mutex *lock, struct ww_acquire_ctx *ctx)
>>> +{
>>> +	struct ww_mutex *ww = container_of(lock, struct ww_mutex, base);
>>> +	struct ww_acquire_ctx *hold_ctx = ACCESS_ONCE(ww->ctx);
>>> +
>>> +	if (!hold_ctx)
>>> +		return 0;
>>> +
>>> +	if (unlikely(ctx->stamp == hold_ctx->stamp))
>>> +		return -EALREADY;
>> Why compare stamps? I expected: ctx == hold_ctx here.
> Because the check just below it compares stamps too, having the same check
> would make it clear that when ctx->stamp - hold_ctx->stamp == 0 is not expected.
>
>>> +
>>> +	if (unlikely(ctx->stamp - hold_ctx->stamp <= LONG_MAX)) {
>> Why not simply write: ctx->stamp > hold_ctx->stamp ?
> To handle the wraparound case on 32-bits?
>> If we need to deal with equal stamps from different contexts we could tie-break
>> based on ctx address or so, but seeing its a global counter from the class,
>> that shouldn't happen for now.
> Sounds good enough to me.
>
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +		DEBUG_LOCKS_WARN_ON(ctx->contending_lock);
>>> +		ctx->contending_lock = ww;
>>> +#endif
>>> +		return -EDEADLK;
>>> +	}
>>> +
>>> +	return 0;
>>> +}
>>> +
>>> +static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww,
>>> +						   struct ww_acquire_ctx *ww_ctx,
>>> +						   bool ww_slow)
>>> +{
>>> +#ifdef CONFIG_DEBUG_MUTEXES
>>> +	/*
>>> +	 * If this WARN_ON triggers, you used mutex_lock to acquire,
>>> +	 * but released with ww_mutex_unlock in this call.
>>> +	 */
>>> +	DEBUG_LOCKS_WARN_ON(ww->ctx);
>>> +
>>> +	/*
>>> +	 * Not quite done after ww_acquire_done() ?
>>> +	 */
>>> +	DEBUG_LOCKS_WARN_ON(ww_ctx->done_acquire);
>>> +
>>> +	if (ww_slow) {
>> s/ww_slow/!ww_ctx->acquired/
>>
>>> +		DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock != ww);
>>> +		ww_ctx->contending_lock = NULL;
>>> +	} else
>>> +		DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock);
>>> +
>>> +
>>> +	/*
>>> +	 * Naughty, using a different class can lead to undefined behavior!
>>> +	 */
>>> +	DEBUG_LOCKS_WARN_ON(ww_ctx->ww_class != ww->ww_class);
>>> +
>>> +	if (ww_slow)
>>> +		DEBUG_LOCKS_WARN_ON(ww_ctx->acquired > 0);
>>> +
>>> +	ww_ctx->acquired++;
>>> +#endif
>>> +}
>>> +
>>> +/*
>>> + * after acquiring lock with fastpath or when we lost out in contested
>>> + * slowpath, set ctx and wake up any waiters so they can recheck.
>>> + *
>>> + * This function is never called when CONFIG_DEBUG_LOCK_ALLOC is set,
>>> + * as the fastpath and opportunistic spinning are disabled in that case.
>>> + */
>>> +static __always_inline void
>>> +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, false);
>>> +
>>> +	lock->ctx = ctx;
>>  missing comment
> Yeah, this was patched up as per danvet's command, moved the smp_mb__after upwards, and added a full smp_mb after setting lock->ctx.
>>> +	smp_mb__after_atomic_dec();
>>> +
>>> +	/*
>>> +	 * Check if lock is contended, if not there is nobody to wake up
>>> +	 */
>>> +	if (likely(atomic_read(&lock->base.count) == 0))
>>> +		return;
>>> +
>>> +	/*
>>> +	 * Uh oh, we raced in fastpath, wake up everyone in this case,
>>> +	 * so they can see the new 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);
>>> +}
>>> +
>>>  /*
>>>   * Lock a mutex (possibly interruptible), slowpath:
>>>   */
>>> -static inline int __sched
>>> +static __always_inline int __sched
>>>  __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
>>> -		    struct lockdep_map *nest_lock, unsigned long ip)
>>> +		    struct lockdep_map *nest_lock, unsigned long ip,
>>> +		    struct ww_acquire_ctx *ww_ctx, bool ww_slow)
>>>  {
>>>  	struct task_struct *task = current;
>>>  	struct mutex_waiter waiter;
>>>  	unsigned long flags;
>>> +	int ret;
>>>  
>>>  	preempt_disable();
>>>  	mutex_acquire_nest(&lock->dep_map, subclass, 0, nest_lock, ip);
>>> @@ -163,6 +303,14 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
>>>  	for (;;) {
>>>  		struct task_struct *owner;
>>>  
>>> +		if (!__builtin_constant_p(ww_ctx == NULL) && !ww_slow) {
>> Since we _know_ ww_ctx isn't NULL, we can trivially do: s/ww_slow/!ww_ctx->acquired/
>>
>>> +			struct ww_mutex *ww;
>>> +
>>> +			ww = container_of(lock, struct ww_mutex, base);
>>> +			if (ACCESS_ONCE(ww->ctx))
>> What's the point of this ACCESS_ONCE()?
> Break out of the spin_on_owner loop. Without taking spin_lock_mutex there is no guarantee that the
> contents of ww->ctx are valid or sane in any way, so there's no way to check if we ought to back off or not.
>
>
>>> +				break;
>>> +		}
>>> +
>>>  		/*
>>>  		 * If there's an owner, wait for it to either
>>>  		 * release the lock or go to sleep.
>>> @@ -173,6 +321,13 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
>>>  
>>>  		if (atomic_cmpxchg(&lock->count, 1, 0) == 1) {
>>>  			lock_acquired(&lock->dep_map, ip);
>> Should this not also have a __builtin_constant_p(ww_ctx == NULL) ?
> ww_slow should not be set to non-zero when ww_ctx == NULL.
>
>>> +			if (ww_slow) {
>>> +				struct ww_mutex *ww;
>>> +				ww = container_of(lock, struct ww_mutex, base);
>>> +
>>> +				ww_mutex_set_context_fastpath(ww, ww_ctx);
>>> +			}
>>> +
>>>  			mutex_set_owner(lock);
>>>  			preempt_enable();
>>>  			return 0;
>>> @@ -228,15 +383,16 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
>>>  		 * TASK_UNINTERRUPTIBLE case.)
>>>  		 */
>>>  		if (unlikely(signal_pending_state(state, task))) {
>>> -			mutex_remove_waiter(lock, &waiter,
>>> -					    task_thread_info(task));
>>> -			mutex_release(&lock->dep_map, 1, ip);
>>> -			spin_unlock_mutex(&lock->wait_lock, flags);
>>> +			ret = -EINTR;
>>> +			goto err;
>>> +		}
>>>  
>>> -			debug_mutex_free_waiter(&waiter);
>>> -			preempt_enable();
>>> -			return -EINTR;
>>> +		if (!__builtin_constant_p(ww_ctx == NULL) && !ww_slow) {
>>> +			ret = __mutex_lock_check_stamp(lock, ww_ctx);
>>> +			if (ret)
>>> +				goto err;
>>>  		}
>>> +
>>>  		__set_task_state(task, state);
>>>  
>>>  		/* didn't get the lock, go to sleep: */
>>> @@ -251,6 +407,30 @@ done:
>>>  	mutex_remove_waiter(lock, &waiter, current_thread_info());
>>>  	mutex_set_owner(lock);
>>>  
>>> +	if (!__builtin_constant_p(ww_ctx == NULL)) {
>>> +		struct ww_mutex *ww = container_of(lock,
>>> +						      struct ww_mutex,
>>> +						      base);
>>> +		struct mutex_waiter *cur;
>>> +
>>> +		/*
>>> +		 * This branch gets optimized out for the common case,
>>> +		 * and is only important for ww_mutex_lock.
>>> +		 */
>>> +
>>> +		ww_mutex_lock_acquired(ww, ww_ctx, ww_slow);
>>> +		ww->ctx = ww_ctx;
>>> +
>>> +		/*
>>> +		 * Give any possible sleeping processes the chance to wake up,
>>> +		 * so they can recheck if they have to back off.
>>> +		 */
>>> +		list_for_each_entry(cur, &lock->wait_list, list) {
>>> +			debug_mutex_wake_waiter(lock, cur);
>>> +			wake_up_process(cur->task);
>>> +		}
>>> +	}
>>> +
>>>  	/* set it to 0 if there are no waiters left: */
>>>  	if (likely(list_empty(&lock->wait_list)))
>>>  		atomic_set(&lock->count, 0);
>>> @@ -261,6 +441,14 @@ done:
>>>  	preempt_enable();
>>>  
>>>  	return 0;
>>> +
>>> +err:
>>> +	mutex_remove_waiter(lock, &waiter, task_thread_info(task));
>>> +	spin_unlock_mutex(&lock->wait_lock, flags);
>>> +	debug_mutex_free_waiter(&waiter);
>>> +	mutex_release(&lock->dep_map, 1, ip);
>>> +	preempt_enable();
>>> +	return ret;
>>>  }
>>>  
>>>  #ifdef CONFIG_DEBUG_LOCK_ALLOC
>>> @@ -268,7 +456,8 @@ void __sched
>>>  mutex_lock_nested(struct mutex *lock, unsigned int subclass)
>>>  {
>>>  	might_sleep();
>>> -	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, subclass, NULL, _RET_IP_);
>>> +	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE,
>>> +			    subclass, NULL, _RET_IP_, 0, 0);
>>>  }
>> The pendant in me has to tell you 4x that NULL != 0 :-)
>>
>>> +EXPORT_SYMBOL_GPL(ww_mutex_lock);
>>> +EXPORT_SYMBOL_GPL(ww_mutex_lock_interruptible);
>>> +EXPORT_SYMBOL_GPL(ww_mutex_lock_slow);
>>> +EXPORT_SYMBOL_GPL(ww_mutex_lock_slow_interruptible);
>> Now having to do the _slow stuff saves lines and interface complexity!
> It will also reduce useful debugging information returned a little. Danvet answered it better than me.
>
>>> @@ -401,20 +738,39 @@ __mutex_lock_slowpath(atomic_t *lock_count)
>>>  {
>>>  	struct mutex *lock = container_of(lock_count, struct mutex, count);
>>>  
>>> -	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, 0, NULL, _RET_IP_);
>>> +	__mutex_lock_common(lock, TASK_UNINTERRUPTIBLE, 0,
>>> +			    NULL, _RET_IP_, 0, 0);
>>>  }
>>>  
>>>  static noinline int __sched
>>>  __mutex_lock_killable_slowpath(struct mutex *lock)
>>>  {
>>> -	return __mutex_lock_common(lock, TASK_KILLABLE, 0, NULL, _RET_IP_);
>>> +	return __mutex_lock_common(lock, TASK_KILLABLE, 0,
>>> +				   NULL, _RET_IP_, 0, 0);
>>>  }
>>>  
>>>  static noinline int __sched
>>>  __mutex_lock_interruptible_slowpath(struct mutex *lock)
>>>  {
>>> -	return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, 0, NULL, _RET_IP_);
>>> +	return __mutex_lock_common(lock, TASK_INTERRUPTIBLE, 0,
>>> +				   NULL, _RET_IP_, 0, 0);
>>>  }
>> A few more cases where NULL != 0 :-)
> But yeah, so open questions..
>
> 1. Do you still want to get rid of the _single variants, even though doing so would slightly reduce debugging?
This is not as disastrous as I originally thought.
lock_single needs to be kept, unlock_single can be removed without consequences.
trylock_single could then be renamed to trylock, since it's no longer paired with unlock_single.

It will prevent unlock and unlock_single from being mismatched, because the same code would get executed.
Calling the mutex_unlock function directly on a ww_mutex must then be forbidden, as it's only an implementation detail.


> 2. Do you really want to drop the *_slow variants?
> Doing so might reduce debugging slightly. I like method #2 in ww-mutex-design.txt, it makes it very clear why you
> would handle the *_slow case differently anyway.
As you pointed out, we wouldn't lose much debugging information.
The same checks could be done in the normal variant with
WARN_ON(ctx->lock && ctx->lock != lock);
WARN_ON(ctx->lock && ctx->acquired > 0);

But it boils down to ww_mutex_lock_slow returning void instead of int __must_check from ww_mutex_lock.

Maybe add inlines for *_slow, that use the ww_mutex_lock functions, and check ctx->lock == lock in debugging mode?
> 3. is a smp_mb needed to serialize lock->ctx with the atomic_read?
>
> (mutex locked in fastpath, which is typically an atomic_dec operation)
> smp_mb__after_atomic_dec();
> lock->ctx = ..;
> smp_mb();
> if (atomic_read(lock->count) == 0) return;
>
> feels a bit like overkill to me.
>
> ~Maarten

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists