On Mon, Feb 04, 2013 at 08:28:29PM +0100, Luca Barbato wrote:
> On 31/01/13 01:07, Diego Biurrun wrote:
> > On Wed, Jan 30, 2013 at 03:51:20PM +0200, Martin Storsjö wrote:
> >> On Tue, 29 Jan 2013, Diego Biurrun wrote:
> >>
> >>> This avoids SIMD-optimized functions having to sign-extend their
> >>> line size argument manually to be able to do pointer arithmetic.
> >>> ---
> >>>
> >>> Extracted from the big hpel megapatch.
> >>>
> >>> This is FATE-tested on x86_64 and x86_32 in all sorts of configurations
> >>> and with different cpuflags.  All other arches need testing still.
> >>
> >> Passes fate on ARM.
> > 
> > Reportedly also passes FATE on sparc32, ppc32, ppc64; sparc64 fails due
> > to compiler bugs even before applying this patch apparently.
> 
> Strange, here it builds fine.
> 
> --cross-prefix=sparc64-unknown-linux-gnu- --target-os=linux --arch=sparc64
> /tmp/libav-sparc64-201302041024 ~
> 
> real  1m8.500s
> user  15m2.629s
> sys   0m41.092s
> sparc64 build succeeded

You're using a non-buggy compiler I guess.  Sean knows more details.

Diego
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to