[FFmpeg-devel] [FFmpeg-cvslog] fate/pixfmts: extend the high bit depth test

[James Almer]

On 11/7/2024 6:02 AM, Martin Storsjö wrote:
> On Thu, 7 Nov 2024, James Almer wrote:
> 
>> ffmpeg | branch: master | James Almer <jamrial at gmail.com> | Sat Nov  2 
>> 20:06:24 2024 -0300| [271aea60a4cd211d92923c53b72cd074c3030897] | 
>> committer: James Almer
>>
>> fate/pixfmts: extend the high bit depth test
>>
>> Also test 8bit formats, and try bitdepth conversion paths.
>>
>> Signed-off-by: James Almer <jamrial at gmail.com>
>>
>>> http://git.videolan.org/gitweb.cgi/ffmpeg.git/? 
>>> a=commit;h=271aea60a4cd211d92923c53b72cd074c3030897
>> ---
> 
> Some of the new tests included here fail on a number of systems - in 
> practice, I don't see any fully passing configuration at http:// 
> fate.ffmpeg.org/?sort=descdate.
> 
> The ones that only have a few test failures fail pixfmt-yuv444p12-yuv444p.
> 
> All other (both x86, arm and aarch64) fail on these tests:
> 
> pixfmt-gbrp10-nv12
> pixfmt-gbrp10-nv24
> pixfmt-gbrp12-nv12
> pixfmt-gbrp12-nv24
> pixfmt-rgb48-nv12
> pixfmt-rgb48-nv24
> pixfmt-yuv444p10-nv12
> pixfmt-yuv444p10-nv24
> pixfmt-yuv444p12-nv12
> pixfmt-yuv444p12-nv24
> pixfmt-yuv444p12-yuv444p
> 
> (When I see failures in these tests x86, that's on an old machine that 
> lacks AVX2.)

I don't know what's going on. I tried on mingw-w64 and Ubuntu x86_64 
22.04 and i get no failures, with CPUFLAGS=0, CPUFLAGS=mmx+mmxext, or 
CPUFLAGS=all (avx2).

I see all x86 systems in FATE failing on fate-pixfmt-yuv444p12-yuv444p, 
but i can't reproduce it.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20241107/ad8c4c3e/attachment.sig>