A common pattern seen when wake_qs are used to defer a wakeup
until after a lock is released is something like:
preempt_disable();
raw_spin_unlock(lock);
wake_up_q(wake_q);
preempt_enable();
So create some raw_spin_unlock*_wake() helper functions to clean
this up.
Applies on top of the fix I submitted here:
https://lore.kernel.org/lkml/20241212222138.2400498-1-jstultz@google.com/
NOTE: I recognise the unlock()/unlock_irq()/unlock_irqrestore()
variants creates its own duplication, which we could use a macro
to generate the similar functions, but I often dislike how those
generation macros making finding the actual implementation
harder, so I left the three functions as is. If folks would
prefer otherwise, let me know and I'll switch it.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241217040803.243420-1-jstultz@google.com
Bert reported seeing occasional boot hangs when running with
PREEPT_RT and bisected it down to commit 894d1b3db4
("locking/mutex: Remove wakeups from under mutex::wait_lock").
It looks like I missed a few spots where we drop the wait_lock and
potentially call into schedule without waking up the tasks on the
wake_q structure. Since the tasks being woken are ww_mutex tasks
they need to be able to run to release the mutex and unblock the
task that currently is planning to wake them. Thus we can deadlock.
So make sure we wake the wake_q tasks when we unlock the wait_lock.
Closes: https://lore.kernel.org/lkml/20241211182502.2915-1-spasswolf@web.de
Fixes: 894d1b3db4 ("locking/mutex: Remove wakeups from under mutex::wait_lock")
Reported-by: Bert Karwatzki <spasswolf@web.de>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20241212222138.2400498-1-jstultz@google.com
Anders had bisected a crash using PREEMPT_RT with linux-next and
isolated it down to commit 894d1b3db4 ("locking/mutex: Remove
wakeups from under mutex::wait_lock"), where it seemed the
wake_q structure was somehow getting corrupted causing a null
pointer traversal.
I was able to easily repoduce this with PREEMPT_RT and managed
to isolate down that through various call stacks we were
actually calling wake_up_q() twice on the same wake_q.
I found that in the problematic commit, I had added the
wake_up_q() call in task_blocks_on_rt_mutex() around
__ww_mutex_add_waiter(), following a similar pattern in
__mutex_lock_common().
However, its just wrong. We haven't dropped the lock->wait_lock,
so its contrary to the point of the original patch. And it
didn't match the __mutex_lock_common() logic of re-initializing
the wake_q after calling it midway in the stack.
Looking at it now, the wake_up_q() call is incorrect and should
just be removed. So drop the erronious logic I had added.
Fixes: 894d1b3db4 ("locking/mutex: Remove wakeups from under mutex::wait_lock")
Closes: https://lore.kernel.org/lkml/6afb936f-17c7-43fa-90e0-b9e780866097@app.fastmail.com/
Reported-by: Anders Roxell <anders.roxell@linaro.org>
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: John Stultz <jstultz@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Juri Lelli <juri.lelli@redhat.com>
Tested-by: Anders Roxell <anders.roxell@linaro.org>
Tested-by: K Prateek Nayak <kprateek.nayak@amd.com>
Link: https://lore.kernel.org/r/20241114190051.552665-1-jstultz@google.com
Pull scheduler updates from Ingo Molnar:
- Implement the SCHED_DEADLINE server infrastructure - Daniel Bristot
de Oliveira's last major contribution to the kernel:
"SCHED_DEADLINE servers can help fixing starvation issues of low
priority tasks (e.g., SCHED_OTHER) when higher priority tasks
monopolize CPU cycles. Today we have RT Throttling; DEADLINE
servers should be able to replace and improve that."
(Daniel Bristot de Oliveira, Peter Zijlstra, Joel Fernandes, Youssef
Esmat, Huang Shijie)
- Preparatory changes for sched_ext integration:
- Use set_next_task(.first) where required
- Fix up set_next_task() implementations
- Clean up DL server vs. core sched
- Split up put_prev_task_balance()
- Rework pick_next_task()
- Combine the last put_prev_task() and the first set_next_task()
- Rework dl_server
- Add put_prev_task(.next)
(Peter Zijlstra, with a fix by Tejun Heo)
- Complete the EEVDF transition and refine EEVDF scheduling:
- Implement delayed dequeue
- Allow shorter slices to wakeup-preempt
- Use sched_attr::sched_runtime to set request/slice suggestion
- Document the new feature flags
- Remove unused and duplicate-functionality fields
- Simplify & unify pick_next_task_fair()
- Misc debuggability enhancements
(Peter Zijlstra, with fixes/cleanups by Dietmar Eggemann, Valentin
Schneider and Chuyi Zhou)
- Initialize the vruntime of a new task when it is first enqueued,
resulting in significant decrease in latency of newly woken tasks
(Zhang Qiao)
- Introduce SM_IDLE and an idle re-entry fast-path in __schedule()
(K Prateek Nayak, Peter Zijlstra)
- Clean up and clarify the usage of Clean up usage of rt_task()
(Qais Yousef)
- Preempt SCHED_IDLE entities in strict cgroup hierarchies
(Tianchen Ding)
- Clarify the documentation of time units for deadline scheduler
parameters (Christian Loehle)
- Remove the HZ_BW chicken-bit feature flag introduced a year ago,
the original change seems to be working fine (Phil Auld)
- Misc fixes and cleanups (Chen Yu, Dan Carpenter, Huang Shijie,
Peilin He, Qais Yousefm and Vincent Guittot)
* tag 'sched-core-2024-09-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (64 commits)
sched/cpufreq: Use NSEC_PER_MSEC for deadline task
cpufreq/cppc: Use NSEC_PER_MSEC for deadline task
sched/deadline: Clarify nanoseconds in uapi
sched/deadline: Convert schedtool example to chrt
sched/debug: Fix the runnable tasks output
sched: Fix sched_delayed vs sched_core
kernel/sched: Fix util_est accounting for DELAY_DEQUEUE
kthread: Fix task state in kthread worker if being frozen
sched/pelt: Use rq_clock_task() for hw_pressure
sched/fair: Move effective_cpu_util() and effective_cpu_util() in fair.c
sched/core: Introduce SM_IDLE and an idle re-entry fast-path in __schedule()
sched: Add put_prev_task(.next)
sched: Rework dl_server
sched: Combine the last put_prev_task() and the first set_next_task()
sched: Rework pick_next_task()
sched: Split up put_prev_task_balance()
sched: Clean up DL server vs core sched
sched: Fixup set_next_task() implementations
sched: Use set_next_task(.first) where required
sched/fair: Properly deactivate sched_delayed task upon class change
...
rt_mutex_handle_deadlock() is called with rt_mutex::wait_lock held. In the
good case it returns with the lock held and in the deadlock case it emits a
warning and goes into an endless scheduling loop with the lock held, which
triggers the 'scheduling in atomic' warning.
Unlock rt_mutex::wait_lock in the dead lock case before issuing the warning
and dropping into the schedule for ever loop.
[ tglx: Moved unlock before the WARN(), removed the pointless comment,
massaged changelog, added Fixes tag ]
Fixes: 3d5c9340d1 ("rtmutex: Handle deadlock detection smarter")
Signed-off-by: Roland Xu <mu001999@outlook.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/all/ME0P300MB063599BEF0743B8FA339C2CECC802@ME0P300MB0635.AUSP300.PROD.OUTLOOK.COM
rt_task() checks if a task has RT priority. But depends on your
dictionary, this could mean it belongs to RT class, or is a 'realtime'
task, which includes RT and DL classes.
Since this has caused some confusion already on discussion [1], it
seemed a clean up is due.
I define the usage of rt_task() to be tasks that belong to RT class.
Make sure that it returns true only for RT class and audit the users and
replace the ones required the old behavior with the new realtime_task()
which returns true for RT and DL classes. Introduce similar
realtime_prio() to create similar distinction to rt_prio() and update
the users that required the old behavior to use the new function.
Move MAX_DL_PRIO to prio.h so it can be used in the new definitions.
Document the functions to make it more obvious what is the difference
between them. PI-boosted tasks is a factor that must be taken into
account when choosing which function to use.
Rename task_is_realtime() to realtime_task_policy() as the old name is
confusing against the new realtime_task().
No functional changes were intended.
[1] https://lore.kernel.org/lkml/20240506100509.GL40213@noisy.programming.kicks-ass.net/
Signed-off-by: Qais Yousef <qyousef@layalina.io>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Phil Auld <pauld@redhat.com>
Reviewed-by: "Steven Rostedt (Google)" <rostedt@goodmis.org>
Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20240610192018.1567075-2-qyousef@layalina.io
Use try_cmpxchg() instead of cmpxchg(*ptr, old, new) == old.
The x86 CMPXCHG instruction returns success in the ZF flag, so this change
saves a compare after CMPXCHG (and related move instruction in front of CMPXCHG).
Also, try_cmpxchg() implicitly assigns old *ptr value to "old" when CMPXCHG
fails. There is no need to re-read the value in the loop.
Note that the value from *ptr should be read using READ_ONCE() to prevent
the compiler from merging, refetching or reordering the read.
No functional change intended.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@kernel.org>
Link: https://lore.kernel.org/r/20240124104953.612063-1-ubizjak@gmail.com
With DEBUG_RT_MUTEXES enabled the fast-path rt_mutex_cmpxchg_acquire()
always fails and all lock operations take the slow path.
Provide a new helper inline rt_mutex_try_acquire() which maps to
rt_mutex_cmpxchg_acquire() in the non-debug case. For the debug case
it invokes rt_mutex_slowtrylock() which can acquire a non-contended
rtmutex under full debug coverage.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20230908162254.999499-3-bigeasy@linutronix.de
Henry reported that rt_mutex_adjust_prio_check() has an ordering
problem and puts the lie to the comment in [7]. Sharing the sort key
between lock->waiters and owner->pi_waiters *does* create problems,
since unlike what the comment claims, holding [L] is insufficient.
Notably, consider:
A
/ \
M1 M2
| |
B C
That is, task A owns both M1 and M2, B and C block on them. In this
case a concurrent chain walk (B & C) will modify their resp. sort keys
in [7] while holding M1->wait_lock and M2->wait_lock. So holding [L]
is meaningless, they're different Ls.
This then gives rise to a race condition between [7] and [11], where
the requeue of pi_waiters will observe an inconsistent tree order.
B C
(holds M1->wait_lock, (holds M2->wait_lock,
holds B->pi_lock) holds A->pi_lock)
[7]
waiter_update_prio();
...
[8]
raw_spin_unlock(B->pi_lock);
...
[10]
raw_spin_lock(A->pi_lock);
[11]
rt_mutex_enqueue_pi();
// observes inconsistent A->pi_waiters
// tree order
Fixing this means either extending the range of the owner lock from
[10-13] to [6-13], with the immediate problem that this means [6-8]
hold both blocked and owner locks, or duplicating the sort key.
Since the locking in chain walk is horrible enough without having to
consider pi_lock nesting rules, duplicate the sort key instead.
By giving each tree their own sort key, the above race becomes
harmless, if C sees B at the old location, then B will correct things
(if they need correcting) when it walks up the chain and reaches A.
Fixes: fb00aca474 ("rtmutex: Turn the plist into an rb-tree")
Reported-by: Henry Wu <triangletrap12@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Henry Wu <triangletrap12@gmail.com>
Link: https://lkml.kernel.org/r/20230707161052.GF2883469%40hirez.programming.kicks-ass.net
Let L1 and L2 be two spinlocks.
Let T1 be a task holding L1 and blocked on L2. T1, currently, is the top
waiter of L2.
Let T2 be the task holding L2.
Let T3 be a task trying to acquire L1.
The following events will lead to a state in which the wait queue of L2
isn't empty, but no task actually holds the lock.
T1 T2 T3
== == ==
spin_lock(L1)
| raw_spin_lock(L1->wait_lock)
| rtlock_slowlock_locked(L1)
| | task_blocks_on_rt_mutex(L1, T3)
| | | orig_waiter->lock = L1
| | | orig_waiter->task = T3
| | | raw_spin_unlock(L1->wait_lock)
| | | rt_mutex_adjust_prio_chain(T1, L1, L2, orig_waiter, T3)
spin_unlock(L2) | | | |
| rt_mutex_slowunlock(L2) | | | |
| | raw_spin_lock(L2->wait_lock) | | | |
| | wakeup(T1) | | | |
| | raw_spin_unlock(L2->wait_lock) | | | |
| | | | waiter = T1->pi_blocked_on
| | | | waiter == rt_mutex_top_waiter(L2)
| | | | waiter->task == T1
| | | | raw_spin_lock(L2->wait_lock)
| | | | dequeue(L2, waiter)
| | | | update_prio(waiter, T1)
| | | | enqueue(L2, waiter)
| | | | waiter != rt_mutex_top_waiter(L2)
| | | | L2->owner == NULL
| | | | wakeup(T1)
| | | | raw_spin_unlock(L2->wait_lock)
T1 wakes up
T1 != top_waiter(L2)
schedule_rtlock()
If the deadline of T1 is updated before the call to update_prio(), and the
new deadline is greater than the deadline of the second top waiter, then
after the requeue, T1 is no longer the top waiter, and the wrong task is
woken up which will then go back to sleep because it is not the top waiter.
This can be reproduced in PREEMPT_RT with stress-ng:
while true; do
stress-ng --sched deadline --sched-period 1000000000 \
--sched-runtime 800000000 --sched-deadline \
1000000000 --mmapfork 23 -t 20
done
A similar issue was pointed out by Thomas versus the cases where the top
waiter drops out early due to a signal or timeout, which is a general issue
for all regular rtmutex use cases, e.g. futex.
The problematic code is in rt_mutex_adjust_prio_chain():
// Save the top waiter before dequeue/enqueue
prerequeue_top_waiter = rt_mutex_top_waiter(lock);
rt_mutex_dequeue(lock, waiter);
waiter_update_prio(waiter, task);
rt_mutex_enqueue(lock, waiter);
// Lock has no owner?
if (!rt_mutex_owner(lock)) {
// Top waiter changed
----> if (prerequeue_top_waiter != rt_mutex_top_waiter(lock))
----> wake_up_state(waiter->task, waiter->wake_state);
This only takes the case into account where @waiter is the new top waiter
due to the requeue operation.
But it fails to handle the case where @waiter is not longer the top
waiter due to the requeue operation.
Ensure that the new top waiter is woken up so in all cases so it can take
over the ownerless lock.
[ tglx: Amend changelog, add Fixes tag ]
Fixes: c014ef69b3 ("locking/rtmutex: Add wake_state to rt_mutex_waiter")
Signed-off-by: Wander Lairson Costa <wander@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20230117172649.52465-1-wander@redhat.com
Link: https://lore.kernel.org/r/20230202123020.14844-1-wander@redhat.com
Jan Kara reported the following bug triggering on 6.0.5-rt14 running dbench
on XFS on arm64.
kernel BUG at fs/inode.c:625!
Internal error: Oops - BUG: 0 [#1] PREEMPT_RT SMP
CPU: 11 PID: 6611 Comm: dbench Tainted: G E 6.0.0-rt14-rt+ #1
pc : clear_inode+0xa0/0xc0
lr : clear_inode+0x38/0xc0
Call trace:
clear_inode+0xa0/0xc0
evict+0x160/0x180
iput+0x154/0x240
do_unlinkat+0x184/0x300
__arm64_sys_unlinkat+0x48/0xc0
el0_svc_common.constprop.4+0xe4/0x2c0
do_el0_svc+0xac/0x100
el0_svc+0x78/0x200
el0t_64_sync_handler+0x9c/0xc0
el0t_64_sync+0x19c/0x1a0
It also affects 6.1-rc7-rt5 and affects a preempt-rt fork of 5.14 so this
is likely a bug that existed forever and only became visible when ARM
support was added to preempt-rt. The same problem does not occur on x86-64
and he also reported that converting sb->s_inode_wblist_lock to
raw_spinlock_t makes the problem disappear indicating that the RT spinlock
variant is the problem.
Which in turn means that RT mutexes on ARM64 and any other weakly ordered
architecture are affected by this independent of RT.
Will Deacon observed:
"I'd be more inclined to be suspicious of the slowpath tbh, as we need to
make sure that we have acquire semantics on all paths where the lock can
be taken. Looking at the rtmutex code, this really isn't obvious to me
-- for example, try_to_take_rt_mutex() appears to be able to return via
the 'takeit' label without acquire semantics and it looks like we might
be relying on the caller's subsequent _unlock_ of the wait_lock for
ordering, but that will give us release semantics which aren't correct."
Sebastian Andrzej Siewior prototyped a fix that does work based on that
comment but it was a little bit overkill and added some fences that should
not be necessary.
The lock owner is updated with an IRQ-safe raw spinlock held, but the
spin_unlock does not provide acquire semantics which are needed when
acquiring a mutex.
Adds the necessary acquire semantics for lock owner updates in the slow path
acquisition and the waiter bit logic.
It successfully completed 10 iterations of the dbench workload while the
vanilla kernel fails on the first iteration.
[ bigeasy@linutronix.de: Initial prototype fix ]
Fixes: 700318d1d7 ("locking/rtmutex: Use acquire/release semantics")
Fixes: 23f78d4a03 ("[PATCH] pi-futex: rt mutex core")
Reported-by: Jan Kara <jack@suse.cz>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20221202100223.6mevpbl7i6x5udfd@techsingularity.net
Similar to the issues in commits:
6467822b8c ("locking/rtmutex: Prevent spurious EDEADLK return caused by ww_mutexes")
a055fcc132 ("locking/rtmutex: Return success on deadlock for ww_mutex waiters")
ww_rt_mutex_lock() should not return EDEADLK without first going through
the __ww_mutex logic to set the required state. In fact, the chain-walk
can deal with the spurious cycles (per the above commits) this check
warns about and is trying to avoid.
Therefore ignore this test for ww_rt_mutex and simply let things fall
in place.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211129174654.668506-4-bigeasy@linutronix.de
The rw_semaphore and rwlock_t implementation both wake the waiter while
holding the rt_mutex_base::wait_lock acquired.
This can be optimized by waking the waiter lockless outside of the
locked section to avoid a needless contention on the
rt_mutex_base::wait_lock lock.
Extend rt_mutex_wake_q_add() to also accept task and state and use it in
__rwbase_read_unlock().
Suggested-by: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210928150006.597310-3-bigeasy@linutronix.de
rt_mutex_wake_q_add() needs to need to distiguish between sleeping
locks (TASK_RTLOCK_WAIT) and normal locks which use TASK_NORMAL to use
the proper wake mechanism.
Instead of checking for != TASK_NORMAL make it more robust and check
explicit for TASK_RTLOCK_WAIT which is the reason why a different wake
mechanism is used.
No functional change.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210928150006.597310-2-bigeasy@linutronix.de
Dan reported that rt_mutex_adjust_prio_chain() can be called with
.orig_waiter == NULL however commit a055fcc132 ("locking/rtmutex: Return
success on deadlock for ww_mutex waiters") unconditionally dereferences it.
Since both call-sites that have .orig_waiter == NULL don't care for the
return value, simply disable the deadlock squash by adding the NULL check.
Notably, both callers use the deadlock condition as a termination condition
for the iteration; once detected, it is sure that (de)boosting is done.
Arguably step [3] would be a more natural termination point, but it's
dubious whether adding a third deadlock detection state would improve the
code.
Fixes: a055fcc132 ("locking/rtmutex: Return success on deadlock for ww_mutex waiters")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/YS9La56fHMiCCo75@hirez.programming.kicks-ass.net
