[FFmpeg-devel] [PATCH V2] Fixes compiler bug - replace vec_lvsl/vec_perm with vec_xl

Carl Eugen Hoyos ceffmpeg at gmail.com
Wed Apr 15 18:16:59 EEST 2020


Am Fr., 2. Aug. 2019 um 17:24 Uhr schrieb ckerchne
<ckerchne at linux.vnet.ibm.com>:
>
> A bug exist with the gcc compilers for Power in versions 6.x and 7.x
> (verified with 6.3 and 7.4). It was fixed in version 8.x (verified with
> 8.3). I was using a Power 9 ppc64le machine for building and testing.
> This is to address ticket #7124.
>
> It appears the compiler is generating the wrong code for little endian
> machines for the vec_lvsl/vec_perm instruction combos in some cases. If
> these instructions are replaced with vec_xl, the problem goes away for
> all versions of the compilers.

This patch broke compilation on ppc64be with gcc 9.3 (Debian sid):
src/libswscale/ppc/yuv2rgb_altivec.c: In function ‘altivec_yuv2_abgr’:
src/libswscale/ppc/yuv2rgb_altivec.c:338:18: error: implicit
declaration of function ‘vec_xl’; did you mean ‘vec_rl’?
[-Werror=implicit-function-declaration]
  338 |             y0 = vec_xl(0, y1i);

Carl Eugen


More information about the ffmpeg-devel mailing list