waitqueue lockdep annotation
Hi all,
this series adds a strategic lockdep_assert_held to __wake_up_common to ensure callers really do hold the wait_queue_head lock when calling the unlocked wake_up variants. It turns out epoll did not do this for a fairly common path (hit all the time by systemd during bootup), so the second patch fixed this instance as well.
PATCH 1/2 epoll: use the waitqueue lock to protect ep->wq
The eoll code currently always uses the unlocked waitqueue helpers for ep->wq, but instead of holding the lock inside the waitqueue around these calls, as expected by the epoll code uses its own lock. Given that the waitqueue is not exposed to the rest of the kernel this actually works ok at the moment, but prevents the epoll locking rules from being enforced using lockdep. Remove ep->lock and use the waitqueue lock to not only reduce the size of struct eventpoll but also make sure we can assert locking invariations in the waitqueue code.
--- fs/eventpoll.c | 46 ++++++++++++++++++++++------------------------ 1 file changed, 22 insertions(+), 24 deletions(-)
fs/eventpoll.c +22 -24
@@ -182,11 +182,10 @@ struct epitem {* This structure is stored inside the "private_data" member of the file * structure and represents the main data structure for the eventpoll * interface. +* +* Access to it is protected by the lock inside wq. */ struct eventpoll { -/* Protect the access to this structure */ -spinlock_t lock; -/* * This mutex is used to ensure that files are not removed * while epoll is using them. This is held during the event
@@ -210,7 +209,7 @@ struct eventpoll {/* * This is a single linked list that chains all the "struct epitem" that * happened while transferring ready events to userspace w/out -* holding ->lock. +* holding ->wq.lock. */ struct epitem *ovflist;
@@ -686,17 +685,17 @@ static int ep_scan_ready_list(struct eventpoll *ep,* because we want the "sproc" callback to be able to do it * in a lockless way. */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);list_splice_init (&ep->rdllist, &txlist); ep->ovflist =NULL ; -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* * Now call the callback function. */ error = (*sproc)(ep, &txlist, priv); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);/* * During the time we spent inside the "sproc" callback, some * other events might have been queued by the poll callback.
@@ -738,7 +737,7 @@ static int ep_scan_ready_list(struct eventpoll *ep,if (waitqueue_active (&ep->poll_wait)) pwake++; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);if (!ep_locked)mutex_unlock (&ep->mtx);
@@ -782,10 +781,10 @@ static int ep_remove(struct eventpoll *ep, struct epitem *epi)rb_erase_cached (&epi->rbn, &ep->rbr); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (ep_is_linked (&epi->rdllink))list_del_init (&epi->rdllink); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);wakeup_source_unregister (ep_wakeup_source (epi));/*
@@ -1015,7 +1014,6 @@ static int ep_alloc(struct eventpoll **pep)if (unlikely (!ep)) goto free_uid; -spin_lock_init (&ep->lock);mutex_init (&ep->mtx);init_waitqueue_head (&ep->wq);init_waitqueue_head (&ep->poll_wait);
@@ -1119,7 +1117,7 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, vstruct eventpoll *ep = epi->ep;int ewake =0 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);ep_set_busy_poll_napi_id (epi);
@@ -1196,7 +1194,7 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v pwake++; out_unlock: -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* We have to call this outside the lock */ if (pwake)
@@ -1480,7 +1478,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, goto error_remove_epi;/* We have to drop the new item inside our item list to keep track of it */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);/* record NAPI ID of new item if present */ ep_set_busy_poll_napi_id (epi);
@@ -1497,7 +1495,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, pwake++; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);atomic_long_inc (&ep->user->epoll_watches);
@@ -1523,10 +1521,10 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event,* list, since that is used/cleaned only inside a section bound by "mtx". * And ep_insert() is called with "mtx" held. */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (ep_is_linked (&epi->rdllink))list_del_init (&epi->rdllink); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);wakeup_source_unregister (ep_wakeup_source (epi));
@@ -1593,7 +1591,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_even* list, push it inside. */ if (revents & event->events) { -spin_lock_irq (&ep->lock); +spin_lock_irq (&ep->wq.lock);if (!ep_is_linked (&epi->rdllink)) {list_add_tail (&epi->rdllink, &ep->rdllist);ep_pm_stay_awake (epi);
@@ -1604,7 +1602,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_evenif (waitqueue_active (&ep->poll_wait)) pwake++; } -spin_unlock_irq (&ep->lock); +spin_unlock_irq (&ep->wq.lock); }/* We have to call this outside the lock */
@@ -1754,7 +1752,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,* caller specified a non blocking operation. */ timed_out =1 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags); goto check_events; }
@@ -1763,7 +1761,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,if (!ep_events_available (ep))ep_busy_loop (ep, timed_out); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (!ep_events_available (ep)) {/*
@@ -1805,11 +1803,11 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,break ; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);if (!schedule_hrtimeout_range (to, slack,HRTIMER_MODE_ABS )) timed_out =1 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags); }__remove_wait_queue (&ep->wq, &wait);
@@ -1819,7 +1817,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,/* Is it worth to try to dig for events ? */ eavail =ep_events_available (ep); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* * Try to transfer events to user space. In case we get 0 events and
Signature · 1 line
2.14.2
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
Quoted 1 line · Christoph Hellwig
The eoll code currently always uses the unlocked waitqueue helpers for
s/eoll
/epoll
Quoted 2 lines
ep->wq, but instead of holding the lock inside the waitqueue around these calls, as expected by the epoll code uses its own lock.
Hm, that reads a bit weirdly. How about:
The epoll code currently uses the unlocked waitqueue helpers for managing ep->wq, but instead of holding the waitqueue lock around these calls, it uses its own ep->lock spinlock.
Quoted 6 lines
Given that the
waitqueue is not exposed to the rest of the kernel this actually works ok at the moment, but prevents the epoll locking rules from being enforced using lockdep. Removeep->lockand use the waitqueue lock to not only reduce the size of struct eventpoll but also make sure we can assert locking invariations in the waitqueue code.
s/but also make sure
but also to make sure
s/invariations
/invariants
Quoted 24 lines
Signed-off-by Christoph Hellwig <[email protected]>--- fs/eventpoll.c | 46 ++++++++++++++++++++++------------------------ 1 file changed, 22 insertions(+), 24 deletions(-)fs/eventpoll.c +2
-3@@ -182,11 +182,10 @@ struct epitem {* This structure is stored inside the "private_data" member of the file * structure and represents the main data structure for the eventpoll * interface. +* +* Access to it is protected by the lock inside wq. */ struct eventpoll { -/* Protect the access to this structure */ -spinlock_t lock; -/* * This mutex is used to ensure that files are not removed * while epoll is using them. This is held during the event
fs/eventpoll.c struct eventpoll { · line 213
Quoted 6 lines
fs/eventpoll.c +1
-1@@ -210,7 +209,7 @@ struct eventpoll {/* * This is a single linked list that chains all the "struct epitem" that * happened while transferring ready events to userspace w/out -* holding ->lock. +* holding ->wq.lock. */
Neat trick!
This exposes some waitqueue internals, but AFAICS the FUSE code already does a similar trick with fiq->waitq.lock so there's precedent.
Peter, what do you think?
Thanks,
Ingo
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
Quoted 4 lines · Ingo Molnar
This exposes some waitqueue internals, but AFAICS the FUSE code already does a similar trick with
fiq->waitq.lockso there's precedent.
What about waitqueue_lock() and waitqueue_unlock() helpers that lock and unlock, to avoid exposing the internals? Or would that add confusion by making users think they need their own waitqueue locking?
Alternately, a helper that returns the pointer to the lock:
#define waitqueue_lockp(wq) &((wq)->lock)
Used like the following:
spin_lock_irqsave(waitqueue_lockp(&ep->wq), flags);
Cheers, Andreas
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
Quoted 8 lines · Andreas Dilger
Quoted 4 lines · Ingo Molnar
This exposes some waitqueue internals, but AFAICS the FUSE code already does a similar trick with
fiq->waitq.lockso there's precedent.What about
waitqueue_lock()andwaitqueue_unlock()helpers that lock and unlock, to avoid exposing the internals? Or would that add confusion by making users think they need their own waitqueue locking?
Right now there are just two users (FUSE and epoll), and both are well-maintained, essentially core kernel code - I'd rather prefer the readability of explicitly writing out the locking/unlocking pattern.
So while it's a mild layering violation, it's also a valid looking optimization.
Thanks,
Ingo
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
In reply to Ingo MolnarQuoted 5 lines
Hm, that reads a bit weirdly. How about:
The epoll code currently uses the unlocked waitqueue helpers for managing
ep->wq, but instead of holding the waitqueue lock around these calls, it uses its ownep->lockspinlock.
Thanks, fixed.
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
In reply to Christoph HellwigQuoted 10 lines · Christoph Hellwig
The eoll code currently always uses the unlocked waitqueue helpers for
ep->wq, but instead of holding the lock inside the waitqueue around these calls, as expected by the epoll code uses its own lock. Given that the waitqueue is not exposed to the rest of the kernel this actually works ok at the moment, but prevents the epoll locking rules from being enforced using lockdep. Removeep->lockand use the waitqueue lock to not only reduce the size of struct eventpoll but also make sure we can assert locking invariations in the waitqueue code.Signed-off-by Christoph Hellwig <[email protected]>
Probably should also fix the locking comments at the top of fs/eventpoll.c that refer to ep->lock...
The rest looks good.
Thanks,
-Jason
Quoted 191 lines
--- fs/eventpoll.c | 46 ++++++++++++++++++++++------------------------ 1 file changed, 22 insertions(+), 24 deletions(-)fs/eventpoll.c +22
-24@@ -182,11 +182,10 @@ struct epitem {* This structure is stored inside the "private_data" member of the file * structure and represents the main data structure for the eventpoll * interface. +* +* Access to it is protected by the lock inside wq. */ struct eventpoll { -/* Protect the access to this structure */ -spinlock_t lock; -/* * This mutex is used to ensure that files are not removed * while epoll is using them. This is held during the event @@ -210,7 +209,7 @@ struct eventpoll {/* * This is a single linked list that chains all the "struct epitem" that * happened while transferring ready events to userspace w/out -* holding ->lock. +* holding ->wq.lock. */ struct epitem *ovflist;@@ -686,17 +685,17 @@ static int ep_scan_ready_list(struct eventpoll *ep,* because we want the "sproc" callback to be able to do it * in a lockless way. */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);list_splice_init (&ep->rdllist, &txlist); ep->ovflist =NULL ; -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* * Now call the callback function. */ error = (*sproc)(ep, &txlist, priv); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);/* * During the time we spent inside the "sproc" callback, some * other events might have been queued by the poll callback. @@ -738,7 +737,7 @@ static int ep_scan_ready_list(struct eventpoll *ep,if (waitqueue_active (&ep->poll_wait)) pwake++; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);if (!ep_locked)mutex_unlock (&ep->mtx);@@ -782,10 +781,10 @@ static int ep_remove(struct eventpoll *ep, struct epitem *epi)rb_erase_cached (&epi->rbn, &ep->rbr); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (ep_is_linked (&epi->rdllink))list_del_init (&epi->rdllink); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);wakeup_source_unregister (ep_wakeup_source (epi));/* @@ -1015,7 +1014,6 @@ static int ep_alloc(struct eventpoll **pep)if (unlikely (!ep)) goto free_uid; -spin_lock_init (&ep->lock);mutex_init (&ep->mtx);init_waitqueue_head (&ep->wq);init_waitqueue_head (&ep->poll_wait);@@ -1119,7 +1117,7 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, vstruct eventpoll *ep = epi->ep;int ewake =0 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);ep_set_busy_poll_napi_id (epi);@@ -1196,7 +1194,7 @@ static int ep_poll_callback(wait_queue_entry_t *wait, unsigned mode, int sync, v pwake++; out_unlock: -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* We have to call this outside the lock */ if (pwake)@@ -1480,7 +1478,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, goto error_remove_epi;/* We have to drop the new item inside our item list to keep track of it */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);/* record NAPI ID of new item if present */ ep_set_busy_poll_napi_id (epi);@@ -1497,7 +1495,7 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event, pwake++; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);atomic_long_inc (&ep->user->epoll_watches);@@ -1523,10 +1521,10 @@ static int ep_insert(struct eventpoll *ep, struct epoll_event *event,* list, since that is used/cleaned only inside a section bound by "mtx". * And ep_insert() is called with "mtx" held. */ -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (ep_is_linked (&epi->rdllink))list_del_init (&epi->rdllink); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);wakeup_source_unregister (ep_wakeup_source (epi));@@ -1593,7 +1591,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_even* list, push it inside. */ if (revents & event->events) { -spin_lock_irq (&ep->lock); +spin_lock_irq (&ep->wq.lock);if (!ep_is_linked (&epi->rdllink)) {list_add_tail (&epi->rdllink, &ep->rdllist);ep_pm_stay_awake (epi);@@ -1604,7 +1602,7 @@ static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_evenif (waitqueue_active (&ep->poll_wait)) pwake++; } -spin_unlock_irq (&ep->lock); +spin_unlock_irq (&ep->wq.lock); }/* We have to call this outside the lock */ @@ -1754,7 +1752,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,* caller specified a non blocking operation. */ timed_out =1 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags); goto check_events; }@@ -1763,7 +1761,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,if (!ep_events_available (ep))ep_busy_loop (ep, timed_out); -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags);if (!ep_events_available (ep)) {/* @@ -1805,11 +1803,11 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,break ; } -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);if (!schedule_hrtimeout_range (to, slack,HRTIMER_MODE_ABS )) timed_out =1 ; -spin_lock_irqsave (&ep->lock, flags); +spin_lock_irqsave (&ep->wq.lock, flags); }__remove_wait_queue (&ep->wq, &wait);@@ -1819,7 +1817,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,/* Is it worth to try to dig for events ? */ eavail =ep_events_available (ep); -spin_unlock_irqrestore (&ep->lock, flags); +spin_unlock_irqrestore (&ep->wq.lock, flags);/* * Try to transfer events to user space. In case we get 0 events and
Re: [PATCH 1/2] epoll: use the waitqueue lock to protect ep->wq
Quoted 10 lines · Jason Baron
Quoted 9 lines · Christoph Hellwig
The eoll code currently always uses the unlocked waitqueue helpers for
ep->wq, but instead of holding the lock inside the waitqueue around these calls, as expected by the epoll code uses its own lock. Given that the waitqueue is not exposed to the rest of the kernel this actually works ok at the moment, but prevents the epoll locking rules from being enforced using lockdep. Removeep->lockand use the waitqueue lock to not only reduce the size of struct eventpoll but also make sure we can assert locking invariations in the waitqueue code.
Quoted 4 lines
Quoted 1 line
Signed-off-by Christoph Hellwig <[email protected]>Probably should also fix the locking comments at the top of
fs/eventpoll.cthat refer toep->lock...
Done. Note that while doing this I noticed that the epoll code seems to have sketchy workarounds for the fact that it abused ep->poll as the waitqueue lock that might be able to be removed now. But I don't really dare to touch the guts of this code.
PATCH 2/2 sched/wait: assert the wait_queue_head lock is held in __wake_up_common
In reply to Christoph HellwigBetter ensure we actually hold the lock using lockdep than just commenting on it. Due to the various exported _locked interfaces it is far too easy to get the locking wrong.
--- kernel/sched/wait.c | 2 ++ 1 file changed, 2 insertions(+)
kernel/sched/wait.c +2 -0
@@ -76,6 +76,8 @@ static int __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode,wait_queue_entry_t *curr, *next;int cnt =0 ; +lockdep_assert_held (&wq_head->lock); +if (bookmark && (bookmark->flags &WQ_FLAG_BOOKMARK )) { curr =list_next_entry (bookmark, entry);
Signature · 1 line
2.14.2
Re: [PATCH 2/2] sched/wait: assert the wait_queue_head lock is held in __wake_up_common
Quoted 14 lines · Christoph Hellwig
Better ensure we actually hold the lock using lockdep than just commenting on it. Due to the various exported
_lockedinterfaces it is far too easy to get the locking wrong.Signed-off-by Christoph Hellwig <[email protected]>--- kernel/sched/wait.c | 2 ++ 1 file changed, 2 insertions(+)kernel/sched/wait.c +0
-0
kernel/sched/wait.c static int __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode, · line 82
Quoted 7 lines
kernel/sched/wait.c +2
-0@@ -76,6 +76,8 @@ static int __wake_up_common(struct wait_queue_head *wq_head, unsigned int mode,wait_queue_entry_t *curr, *next;int cnt =0 ; +lockdep_assert_held (&wq_head->lock); +if (bookmark && (bookmark->flags &WQ_FLAG_BOOKMARK )) { curr =list_next_entry (bookmark, entry);
Makes sense. Would you like to carry this patch together with the epoll patch, to be able to test them both? If yes then:
... otherwise I can pick this up into the scheduler tree as well.
Thanks,
Ingo
Re: [PATCH 2/2] sched/wait: assert the wait_queue_head lock is held in __wake_up_common
Quoted 2 lines · Ingo Molnar
Makes sense. Would you like to carry this patch together with the epoll patch, to be able to test them both? If yes then:
It would be good to merge them all together. I still thing the sched tree might be the best place for all of them, but I'll let you fight that out with Andrew once the next iteration is posted :)