[PATCH] iWMMXt support was Re: [Ffmpeg-devel] iWMMXt support

Reimar Döffinger Reimar.Doeffinger
Thu Mar 15 11:21:57 CET 2007

On Thu, Mar 15, 2007 at 10:52:30AM +0100, Guillaume Poirier wrote:
> Reimar D?ffinger wrote:
> > And that one is still broken. It sent a patch to make it restore at
> > least the signal handler again to its original value, but it was
> > suggested to remove the whole hack instead.
> Maybe it's a hack, but I'd like to see someone coming up with a proper
> way to test for Altivec support. I mean, there's gotta be a proper
> way, it's just that Google isn't that informative :-(

IMO there is not, because half of the people working on PPC (whether
working on gcc, kernel or the CPU itself) obviously consider with and
without Altivec a different architecture, leaving as the only solution
to deal with it like this, i.e. requiring a different binary.
To take gcc as an example: -maltivec or so is needed to use the
intrinsics (which unfortunately is preferred by the Altivec
maintainers), but it will also enable Altivec everywhere, including even
the most simple C code. Thus runtime CPU-detection just can't work.

Reimar D?ffinger

More information about the ffmpeg-devel mailing list