[asterisk-dev] Warning and/or question (Zaptel MMX)

[Steve Davies]

On 8/28/07, Nic Bellamy <nicb-lists at vadacom.co.nz> wrote:
> Tzafrir Cohen wrote:
> > Hi
> >
> > Long ago, in a far away galaxy, on Tue, Sep 12, 2006 at 09:50:17AM +0100,
> > Steve Davies wrote:

Hehe, made me laugh, getting a response to an email a year after it
was posted :)

> >> Firstly, a note of caution for those compiling MMX into Zaptel... I
> >> use an EPIA processor, which may be part of the problem, but enabling
> >> the Zaptel MMX code caused sufficient corruption that it affected
> >> other userspace processes! I suddenly started getting page-errors in a
> >> web application, and this only happened when Asterisk was up and
> >> running, using Zaptel.
> >
> > Do you happen to remember which CPU exactly it was? Is it a C2 (e.g:
> > Nehemia, Esther)? I believe those should have a better MMX
> > implementation than some of the older VIA CPUs (e.g: Ezra).
> >
> We're successfully using MMX cancellers on VIA Nehemia, VIA Esther and
> various other Athlon/P4 CPUs. Before putting this anywhere near
> production, I did quite extensive testing - using floating point ops in
> multiple processes while feeding multiple streams of audio through the
> in-kernel MMX-enabled cancellers, all the while comparing expected vs.
> actual. Not a glitch - and the speed increases are well worth it for the
> Via processors. Oh for an MMX-optimized HPEC :-)

We were using the C2 and newer processors (M10000 boards), but perhaps
a slightly old 2.6.10 kernel. In hindsight, I suspect that the issue
was with that kernel rather that with MMX itself.

> I am using CLOBBER_MMX though, so the compiler knows I've been twiddling
> bits.
> >> Recompile Zaptel with No-MMX, and all is well again.
> >>
> >> Secondly a question - I notice the "CLOBBER_MMX" macro in the Zaptel
> >> code. This is not documented anywhere... Might it help with the
> >> corruption?
> >>

So that is what CLOBBER_MMX does. Thanks.

> > I also heard from jsmith yesterday on #asterisk-dev that 4 or 5 years
> > ago he had bad luck using CONFIG_ZAPTEL_MMX with on Athlon and Athlon XP
> > CPUs then.
> >
> I suspect a considerable amount of the problems may have been with
> either buggy CPUs or with the kernel not invalidating/restoring the FPU
> state correctly after using MMX instructions. All my testing has been
> done on recent 2.6.x kernels.

I agree. I might consider looking at this again, but so much of our
equipment is HW echo-can these days that I am not sure it is worth the
effort.

Thanks for the revival of the thread though :)
Steve