Re: [PATCH] mutex: have non-spinning mutexes on s390 by default

[Peter Zijlstra <peterz@infradead.org>]

On Thu, 2009-04-09 at 18:48 +0200, Heiko Carstens wrote:
> Updated Patch below:
> 
> Subject: [PATCH] mutex: have non-spinning mutexes on s390 by default
> 
> From: Heiko Carstens <heiko.carstens@de.ibm.com>
> 
> The adaptive spinning mutexes will not always do what one would expect on
> virtualized architectures like s390. Especially the cpu_relax() loop in
> mutex_spin_on_owner might hurt if the mutex holding cpu has been scheduled
> away by the hypervisor.
> We would end up in a cpu_relax() loop when there is no chance that the
> state of the mutex changes until the target cpu has been scheduled again by
> the hypervisor.
> For that reason we should change the default behaviour to no-spin on s390.
> 
> We do have an instruction which allows to yield the current cpu in favour of
> a different target cpu. Also we have an instruction which allows us to figure
> out if the target cpu is physically backed.
> 
> However we need to do some performance tests until we can come up with
> a solution that will do the right thing on s390.
> 
> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>

Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>

I was looking at how an monitor-wait could be used here, but that
appears non-trivial, there's two variables we're watching, lock->owner
and rq->curr, either could change.

Reducing that to 1 seems an interesting problem :-)