> Yes I suppose the branch can be executed "non speculatively" before the
> lwsync is completed. Wheras the larx/stcwx will have to complete before
> the branch outcome can be known. I suppose probably not worthwhile
> avoiding the full IO sync by adding yet more crap to make this work.
>
> Than
On Wed, Feb 17, 2010 at 08:43:14PM +1100, Anton Blanchard wrote:
>
> Hi Nick,
>
> > Ah, good to see this one come back. I also tested tbench over localhost
> > btw which actually did show some speedup on the G5.
> >
> > BTW. this was the last thing left:
> > http://www.mail-archive.com/linuxpp
Hi Nick,
> Ah, good to see this one come back. I also tested tbench over localhost
> btw which actually did show some speedup on the G5.
>
> BTW. this was the last thing left:
> http://www.mail-archive.com/linuxppc-dev@lists.ozlabs.org/msg29738.html
>
> Don't know if you took a look at that a
On Tue, Feb 16, 2010 at 03:19:03PM +1100, Benjamin Herrenschmidt wrote:
> On Mon, 2010-02-15 at 22:22 -0600, Olof Johansson wrote:
> >
> > Turns out this one hurts PA6T performance quite a bit, lwsync seems to be
> > significantly more expensive there. I see a 25% drop in the microbenchmark
> > do
On Mon, 2010-02-15 at 22:22 -0600, Olof Johansson wrote:
>
> Turns out this one hurts PA6T performance quite a bit, lwsync seems to be
> significantly more expensive there. I see a 25% drop in the microbenchmark
> doing pthread_lock/unlock loops on two cpus.
>
> Taking out the CPU_FTR_LWSYNC will
On Wed, Feb 10, 2010 at 10:10:25PM +1100, Anton Blanchard wrote:
>
> Nick Piggin discovered that lwsync barriers around locks were faster than
> isync
> on 970. That was a long time ago and I completely dropped the ball in testing
> his patches across other ppc64 processors.
>
> Turns out the id
On Wed, Feb 10, 2010 at 10:10:25PM +1100, Anton Blanchard wrote:
>
> Nick Piggin discovered that lwsync barriers around locks were faster than
> isync
> on 970. That was a long time ago and I completely dropped the ball in testing
> his patches across other ppc64 processors.
>
> Turns out the id
Nick Piggin discovered that lwsync barriers around locks were faster than isync
on 970. That was a long time ago and I completely dropped the ball in testing
his patches across other ppc64 processors.
Turns out the idea helps on other chips. Using a microbenchmark that
uses a lot of threads to co
