Re: [PATCH] sched: flush plug in schedule_preempt_disabled() to prevent deadlock

[Michael Wu]

The patch you mentioned also solves my problem; I've verified and tested 
it on my platform.

On 5/15/2026 2:18 PM, Xiaosen wrote:
> https://lore.kernel.org/lkml/20260427183848.698551-2-jstultz@xxxxxxxxxx/
> The above change can resolve the deadlock I reported before by setting
> task's state to TASK_RUNNING before switching context.
>
> There is the likely alternative fix below.
> https://lore.kernel.org/lkml/20260512025635.2840817-1-jstultz@xxxxxxxxxx/
>
> Regards,
> Xiaosen
>
> On 5/13/2026 4:08 PM, Ming Lei wrote:
> > On Wed, May 13, 2026 at 09:30:39AM +0200, Peter Zijlstra wrote:
> > > On Wed, May 13, 2026 at 10:07:03AM +0800, Ming Lei wrote:
> > > > On Tue, May 12, 2026 at 07:16:36AM -1000, Tejun Heo wrote:
> > > > > Hello, Ming.
> > > > >
> > > > > On Tue, May 12, 2026 at 11:45:14PM +0800, Ming Lei wrote:
> > > > > > On Tue, May 12, 2026 at 02:40:21PM +0200, Peter Zijlstra wrote:
> > > > > > > On Tue, May 12, 2026 at 02:04:32PM +0200, Peter Zijlstra wrote:
> > > > > > > > On Tue, May 12, 2026 at 04:59:39PM +0800, Ming Lei wrote:
> > > > > > > > > On preemptible kernels, a deadlock can occur when a task with plugged IO
> > > > > > > > > calls schedule_preempt_disabled():
> > > > > > > > >
> > > > > > > > >    schedule_preempt_disabled()
> > > > > > > > >      sched_preempt_enable_no_resched()  // preemption now enabled
> > > > > > > > >      schedule()                         // <-- preemption can happen here
> > > > > > > > >        sched_submit_work()
> > > > > > > > >          blk_flush_plug()
> > > > > > > > >
> > > > > > > > > After sched_preempt_enable_no_resched() re-enables preemption, the task
> > > > > > > > > can be preempted (e.g., by a higher-priority RT task) before reaching
> > > > > > > > > blk_flush_plug() in sched_submit_work(). Since the task's state is
> > > > > > > > > already TASK_UNINTERRUPTIBLE (set by the mutex/rwsem slowpath caller),
> > > > > > > > > requests in current->plug remain unflushed for an unbounded time.
> > > > > > > > >
> > > > > > > > > If another task depends on those plugged requests to make progress (e.g.,
> > > > > > > > > to release a lock the sleeping task needs), a deadlock results:
> > > > > > > > >
> > > > > > > > >    - Task A (writeback worker): holds plugged IO, preempted before
> > > > > > > > >      flushing, stuck on run queue behind higher-priority work
> > > > > > > > >    - Task B: waiting for IO completion from Task A's plug, holds a lock
> > > > > > > > >      that Task A needs to be woken up
> > > > >
> > > > > My memory is hazy around io_schedule but the above reads really weird to me.
> > > > > A task, regardless of its current state stays on the runqueue when
> > > > > preempted, so the condition is temporary. As soon as the preempted task can
> > > > > get CPU, it should unwind the situation. That's not a deadlock. Is the
> > > > > problem that there can be preemption-induced delay in flushing the plugs?
> > > >
> > > > IMO, preempting a `!TASK_RUNNING` task can be thought as effective sleep,
> > >
> > > No it cannot be. Preemption ignores task state.
> >
> > Yeah, I get similar conclusion too with AI's assistance.
> >
> > But both two reports show that the preempted task aren't switched back for
> > long enough time, can you share any idea for Michael & Xiaosen to investigate
> > further from scheduler side?
> >
> > https://lore.kernel.org/linux-block/20260417082744.30124-1-michael@xxxxxxxxxxxxxxxxx/
> >
> > https://lore.kernel.org/linux-block/5660795d-87de-46f5-add4-7729a02225ef@xxxxxxxxxxxxxxxx/
> >
> >
> > Thanks,
> > Ming

--
Regards,
Michael Wu