Re: [RFC][PATCH 2/3] futex: Reduce hash bucket lock contention

[Peter Zijlstra]

On Mon, 2011-09-19 at 09:37 +0200, Peter Zijlstra wrote:
> On Sat, 2011-09-17 at 14:57 +0200, Manfred Spraul wrote:
> > How do we verify that everything is able to deal with spurious
> > wakeups?
> > 
> Well, I could go audit all 1400+ *schedule*() callsites in the kernel.
> Or we can rely on the fact that current code can already cause spurious
> wakeups due to signals.

Hrmm,. the sem code would have serialized on the IN_WAKER stuff, the
mutex code would serialize on the ->wait_lock, and the futex code would
have serialized on the hb lock.

So while it can issue multiple wakeups, those should not leak out of the
locking primitive.. crap.

Still wondering why we've got that many schedule() calls in the kernel
though, clearly we don't have enough blocking primitives or so..
--
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/