Re: [PATCH v3 -mm -next] ipc,sem: fix lockdep false positive

2013-03-29 Thread Michel Lespinasse
On Fri, Mar 29, 2013 at 6:55 AM, Rik van Riel wrote: > On Thu, 28 Mar 2013 19:50:47 -0700 > Michel Lespinasse wrote: > >> This is IMO where the spin_unlock_wait(&sma->sem_perm.lock) would >> belong - right before the goto again. > > Here is the slightly more optimistic (and probably more readable

[PATCH v3 -mm -next] ipc,sem: fix lockdep false positive

2013-03-29 Thread Rik van Riel
On Thu, 28 Mar 2013 19:50:47 -0700 Michel Lespinasse wrote: > This is IMO where the spin_unlock_wait(&sma->sem_perm.lock) would > belong - right before the goto again. Here is the slightly more optimistic (and probably more readable) version of the patch: ---8<--- Unfortunately the locking sche