On Mon, 2014-07-28 at 18:57 +0200, Peter Zijlstra wrote: > On Mon, Jul 28, 2014 at 09:53:58AM -0700, Davidlohr Bueso wrote: > > Well, it's not hard to see where the contention is when working on > > locking issues with perf. With mutexes there are only two sources, > > either the task is just spinning trying to get the lock, or its gone to > > the slowpath, and you can see a lot of contention on the wait_lock. > > > > So unless I'm missing something, I don't think we'd need to make this > > noinline again -- although I forget why it was changed in the first > > place. > > Not to mention that there's no actual caller of this function in the > entire kernel ;-) Currently its just 'documentation' describing what an > actual MCS lock looks like.
Yeah, we only use the cancellable version of the lock anyway and there's currently no benefit of changing the regular mcs lock to noinline. I was mainly thinking it could be helpful in potential later uses of the regular mcs lock (if anyone comes up with a need to use it) where it might not be as obvious where the contention is occurring. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
