[Ffmpeg-devel] [PATCH] protect cmov asm sections with HAVE_CMOV

[Michael Niedermayer]

Hi

On Fri, Oct 20, 2006 at 03:21:57PM +0200, Guillaume Poirier wrote:
> Hi,
> 
> Diego Biurrun wrote:
> > On Fri, Oct 20, 2006 at 10:26:45AM +0200, Guillaume Poirier wrote:
> > 
> >>Diego Biurrun wrote:
> >>
> >>>On Fri, Oct 20, 2006 at 09:49:24AM +0200, Guillaume POIRIER wrote:
> >>>
> >>>Patch looks fine except for
> >>>
> >>>
> >>>>--- libavcodec/cabac.h	(revision 6742)
> >>>>+++ libavcodec/cabac.h	(working copy)
> >>>>@@ -32,7 +32,9 @@
> >>>>#define CABAC_BITS 16
> >>>>#define CABAC_MASK ((1<<CABAC_BITS)-1)
> >>>>#define BRANCHLESS_CABAC_DECODER 1
> >>>>+#ifdef HAVE_CMOV
> >>>>#define CMOV_IS_FAST 1
> >>>>+#endif
> >>>>//#define ARCH_X86_DISABLED 1
> >>>>
> >>>>@@ -454,7 +456,7 @@
> >>>>
> >>>>-#if (defined CMOV_IS_FAST  && __CPU__ >= 686)
> >>>>+#if defined CMOV_IS_FAST
> >>>
> >>>This is rather silly IMO, I'd say just replace it by HAVE_CMOV.
> >>>Are there benchmarks proving that this really is slower on some
> >>>processors?  Then configue should deal with this as well.
> >>
> >>Yes, the benchmarks I'm run a few weeks ago showed that Netburst-based
> >>processors (P4/P-D) have a pretty slow implementation of CMOV.
> >>I'll run some new benchmarks with latest optimizations of CABAC, but
> >>I'm pretty confident that they will only confirm that on them, CMOV is
> >>too slow to be beneficial.
> 
> After reading a few tech docs, and some micro benchs, I can state out
> loud that CMOV isn't fast on Netburst-based processors.
> Since we need to both know if a cpu supports CMOV and if it's fast,
> with attached patch, configure (which attached patch) will set both
> defines.
> 
> 
> > OK.  My point was just that currently CMOV_IS_FAST is defined when
> > HAVE_CMOV is defined, so they are equal.  This should be done in
> > configure.  At least add a FIXME note so this is not forgotten.
> 
> Yes, I agree. I mainly wanted to split patches to the minimum, and let
> configure define CMOV_IS_FAST in an upcoming patch.

hmm, what use is it to know that CMOV is supprted but so slow that its not
useable? or is there any cmov based code which is faster then the alternative
on P4*

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

In the past you could go to a library and read, borrow or copy any book
Today you'd get arrested for mere telling someone where the library is