Re: softlockups in multi_cpu_stop

From: Davidlohr Bueso
Date: Fri Mar 06 2015 - 23:44:53 EST


On Fri, 2015-03-06 at 20:31 -0800, Jason Low wrote:
> On Fri, 2015-03-06 at 13:12 -0800, Jason Low wrote:
>
> Just in case, here's the updated patch which addresses Linus's comments
> and with a changelog.
>
> Note: The changelog says that it fixes (locking/rwsem: Avoid deceiving
> lock spinners), though I still haven't seen full confirmation that it
> addresses all of the lockup reports.
>
> ------
> Subject: [PATCH] rwsem: Avoid spinning when owner is not running
>
> Fixes tip commmit b3fd4f03ca0b (locking/rwsem: Avoid deceiving lock spinners).
>
> When doing optimistic spinning in rwsem, threads should stop spinning when
> the lock owner is not running. While a thread is spinning on owner, if
> the owner reschedules, owner->on_cpu returns false and we stop spinning.
>
> However, commit b3fd4f03ca0b essentially caused the check to get ignored
> because when we break out of the spin loop due to !on_cpu, we continue
> spinning if sem->owner != NULL.

I would mention the actual effects of the bug, either just a "lockup"
and/or a fragment of the trace. But ultimately this comes down to
missing a need_resched() condition.

>
> This patch fixes this by making sure we stop spinning if the owner is not
> running. Furthermore, just like with mutexes, refactor the code such that
> we don't have separate checks for owner_running(). This makes it more
> straightforward in terms of why we exit the spin on owner loop and we
> would also avoid needing to "guess" why we broke out of the loop to make
> this more readable.
>
> Cc: Ming Lei <ming.lei@xxxxxxxxxxxxx>
> Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx>

Acked-by: Davidlohr Bueso <dave@xxxxxxxxxxxx>

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