[Ffmpeg-devel] [PATCH] fix build for --cpu=i686

[Michael Niedermayer]

Hi

On Tue, Mar 27, 2007 at 06:21:15PM -0300, Ramiro Ribeiro Polla wrote:
> Hello,
> 
> Attached patch makes regression tests succeed with gcc 4.x when 
> --cpu=i686 is passed to configure. gcc svn didn't work either.
> 
> I don't know why exactly it does work. Maybe it works around a bug in 
> gcc 4.x. I still didn't understand what an "earlyclobber" operand means.
> 
> Ramiro Polla

> Index: libavutil/internal.h
> ===================================================================
> --- libavutil/internal.h	(revision 8535)
> +++ libavutil/internal.h	(working copy)
> @@ -230,7 +230,7 @@
>      "cmovl %3, %0       \n\t"\
>      "cmovl %4, %1       \n\t"\
>      "cmovl %5, %2       \n\t"\
> -    : "+r" (x), "+r" (a), "+r" (c)\
> +    : "+r" (x), "+&r" (a), "+&r" (c)\
>      : "r" (y), "r" (b), "r" (d)\

hmm, could someone ask the gcc devels if "+r" are supposed to be able
to overlap "r" ?

though either way the change is wrong, the early clobbers would have to
be on the first and second not second and third

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The educated differ from the uneducated as much as the living from the
dead. -- Aristotle 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20070328/2237b99c/attachment.pgp>