Re: [PATCH 1/2] ipc/sem.c: Fix complex_count vs. simple op race

From: Andrew Morton
Date: Mon Jun 20 2016 - 19:05:17 EST


On Sat, 18 Jun 2016 22:02:21 +0200 Manfred Spraul <manfred@xxxxxxxxxxxxxxxx> wrote:

> Commit 6d07b68ce16a ("ipc/sem.c: optimize sem_lock()") introduced a race:
>
> sem_lock has a fast path that allows parallel simple operations.
> There are two reasons why a simple operation cannot run in parallel:
> - a non-simple operations is ongoing (sma->sem_perm.lock held)
> - a complex operation is sleeping (sma->complex_count != 0)
>
> As both facts are stored independently, a thread can bypass the current
> checks by sleeping in the right positions. See below for more details
> (or kernel bugzilla 105651).
>
> The patch fixes that by creating one variable (complex_mode)
> that tracks both reasons why parallel operations are not possible.
>
> The patch also updates stale documentation regarding the locking.
>
> With regards to stable kernels:
> The patch is required for all kernels that include the commit 6d07b68ce16a
> ("ipc/sem.c: optimize sem_lock()") (3.10?)

I've had this in -mm (and -next) since January 4, without issues. I
put it on hold because Davidlohr expressed concern about performance
regressions.

Your [2/2] should prevent those regressions (yes?) so I assume that any
kernel which has [1/2] really should have [2/2] as well. But without
any quantitative information, this is all mad guesswork.

What to do?

(The [2/2] changelog should explain that it is the cure to [1/2]'s
regressions, btw).