diff mbox series

[FFmpeg-devel] FATE Errors

Message ID DM8P223MB0365E695970BBD2E9B7FEDE3BAFF9@DM8P223MB0365.NAMP223.PROD.OUTLOOK.COM
State New
Headers show
Series [FFmpeg-devel] FATE Errors | expand

Checks

Context Check Description
andriy/configure_armv7_RPi4 warning Failed to apply patch
andriy/configure_x86 warning Failed to apply patch
yinshiyou/configure_loongarch64 warning Failed to apply patch

Commit Message

Soft Works April 30, 2022, 6:54 p.m. UTC 
Hi,

is it a known issue that the current head of the master branch has
FATE errors?

I get the same locally as well as on the automated GitHub build.
One is this:

Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-mpeg2-422.err for details.
make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1


Is anybody seeing the same?

Thanks,
sw

Comments

Felix LeClair April 30, 2022, 7:01 p.m. UTC | #1 
I’ve been getting the same locally, assumed an issue had already been opened. 

-Fclc

> On Apr 30, 2022, at 14:54, Soft Works <softworkz@hotmail.com> wrote:
> 
> Hi,
> 
> is it a known issue that the current head of the master branch has
> FATE errors?
> 
> I get the same locally as well as on the automated GitHub build.
> One is this:
> 
> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30 14:23:44.330424058 +0000
> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201 +0000
> @@ -1,4 +1,4 @@
> -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-mpeg2-422.err for details.
> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> 
> 
> Is anybody seeing the same?
> 
> Thanks,
> sw
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
Andreas Rheinhardt April 30, 2022, 7:06 p.m. UTC | #2 
Soft Works:
> Hi,
> 
> is it a known issue that the current head of the master branch has
> FATE errors?
> 
> I get the same locally as well as on the automated GitHub build.
> One is this:
> 
> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30 14:23:44.330424058 +0000
> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201 +0000
> @@ -1,4 +1,4 @@
> -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-mpeg2-422.err for details.
> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> 
> 
> Is anybody seeing the same?
> 
> Thanks,
> sw

http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is fine
for me locally. I recently made changes to FATE (namely the test
requirements, vcodec.mak (where the vsynth-tests reside) among the files
affected), so I am interested in whether the failing tests are
concentrated on the files recently changed by me (it would obviously not
haved pushed them if I knew them to cause issues; also patchwork was fine).
Are these issues reproducible? Can you bisect them?

- Andreas
Soft Works April 30, 2022, 7:07 p.m. UTC | #3 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Felix LeClair
> Sent: Saturday, April 30, 2022 9:01 PM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> I’ve been getting the same locally, assumed an issue had already been
> opened.

The weird thing is that the patchwork builds don't show any error:
https://patchwork.ffmpeg.org/project/ffmpeg/list/

..which would be expected for a specific commit that might fix this,
but none of the commits is showing such error.
That's why I was wondering whether it's something on my side (2 sides).


Thanks for letting me know that you're seeing the same!

sw
James Almer April 30, 2022, 7:09 p.m. UTC | #4 
On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> Soft Works:
>> Hi,
>>
>> is it a known issue that the current head of the master branch has
>> FATE errors?
>>
>> I get the same locally as well as on the automated GitHub build.
>> One is this:
>>
>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30 14:23:44.330424058 +0000
>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201 +0000
>> @@ -1,4 +1,4 @@
>> -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>> -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
>> +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>> +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
>> Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-mpeg2-422.err for details.
>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
>>
>>
>> Is anybody seeing the same?
>>
>> Thanks,
>> sw
> 
> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is fine
> for me locally. I recently made changes to FATE (namely the test
> requirements, vcodec.mak (where the vsynth-tests reside) among the files
> affected), so I am interested in whether the failing tests are
> concentrated on the files recently changed by me (it would obviously not
> haved pushed them if I knew them to cause issues; also patchwork was fine).
> Are these issues reproducible? Can you bisect them?
> 
> - Andreas

This may be the alignment issue introduced in lavfi in 
https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00a680c6ebbaea58db95594d13d 
assuming it was not fixed.
I think it only affected targets where av_cpu_max_align() returned 64.
Soft Works April 30, 2022, 7:12 p.m. UTC | #5 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Andreas Rheinhardt
> Sent: Saturday, April 30, 2022 9:07 PM
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> Soft Works:
> > Hi,
> >
> > is it a known issue that the current head of the master branch has
> > FATE errors?
> >
> > I get the same locally as well as on the automated GitHub build.
> > One is this:
> >
> > --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> 14:23:44.330424058 +0000
> > +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201
> +0000
> > @@ -1,4 +1,4 @@
> > -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> > -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> > -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> > +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> > +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> > +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> > Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-
> mpeg2-422.err for details.
> > make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> >
> >
> > Is anybody seeing the same?
> >
> > Thanks,
> > sw
> 
> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is
> fine
> for me locally. 

That's weird. I got the same errors locally and from the GitHub
CI build. Here's the output:

https://dev.azure.com/githubsync/ffmpeg/_build/results?buildId=10000&view=logs&j=275f1d19-1bd8-5591-b06b-07d489ea915a&t=704ea7c9-8eb0-53b9-6b30-87d5a7befe2c


Kind regards,
softworkz
Soft Works April 30, 2022, 7:19 p.m. UTC | #6 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> James Almer
> Sent: Saturday, April 30, 2022 9:10 PM
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> 
> 
> On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > Soft Works:
> >> Hi,
> >>
> >> is it a known issue that the current head of the master branch has
> >> FATE errors?
> >>
> >> I get the same locally as well as on the automated GitHub build.
> >> One is this:
> >>
> >> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> 14:23:44.330424058 +0000
> >> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201
> +0000
> >> @@ -1,4 +1,4 @@
> >> -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> >> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >> -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> >> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> >> +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> >> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >> +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> >> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> >> Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-
> mpeg2-422.err for details.
> >> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> >>
> >>
> >> Is anybody seeing the same?
> >>
> >> Thanks,
> >> sw
> >
> > http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is
> fine
> > for me locally. I recently made changes to FATE (namely the test
> > requirements, vcodec.mak (where the vsynth-tests reside) among the
> files
> > affected), so I am interested in whether the failing tests are
> > concentrated on the files recently changed by me (it would obviously
> not
> > haved pushed them if I knew them to cause issues; also patchwork was
> fine).
> > Are these issues reproducible? Can you bisect them?
> >
> > - Andreas
> 
> This may be the alignment issue introduced in lavfi in
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00a
> 680c6ebbaea58db95594d13d
> assuming it was not fixed.
> I think it only affected targets where av_cpu_max_align() returned 64.

My local build is run on Ubuntu in Hyper-V VM on Windows. The GitHub
CI build runs on Azure DevOps which is also an Ubuntu VM (not sure 
about the host).

AFAIC, Patchwork runs in a Docker container only. Not sure whether
that could be the difference..

softworkz
Felix LeClair April 30, 2022, 7:22 p.m. UTC | #7 
Both times I’ve encountered were on PopOS (think Debian/Ubuntu meets arch like timing release) 

Currently pulling to a macOS machine and seeing if it can be replicated there.

-FCLC

> On Apr 30, 2022, at 15:19, Soft Works <softworkz@hotmail.com> wrote:
> 
> 
> 
>> -----Original Message-----
>> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
>> James Almer
>> Sent: Saturday, April 30, 2022 9:10 PM
>> To: ffmpeg-devel@ffmpeg.org
>> Subject: Re: [FFmpeg-devel] FATE Errors
>> 
>> 
>> 
>>> On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
>>> Soft Works:
>>>> Hi,
>>>> 
>>>> is it a known issue that the current head of the master branch has
>>>> FATE errors?
>>>> 
>>>> I get the same locally as well as on the automated GitHub build.
>>>> One is this:
>>>> 
>>>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
>> 14:23:44.330424058 +0000
>>>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201
>> +0000
>>>> @@ -1,4 +1,4 @@
>>>> -b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-
>> 422.mpeg2video
>>>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>>> -704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-
>> 422.out.rawvideo
>>>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
>>>> +8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-
>> 422.mpeg2video
>>>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>>> +0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-
>> 422.out.rawvideo
>>>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
>>>> Test vsynth2-mpeg2-422 failed. Look at tests/data/fate/vsynth2-
>> mpeg2-422.err for details.
>>>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
>>>> 
>>>> 
>>>> Is anybody seeing the same?
>>>> 
>>>> Thanks,
>>>> sw
>>> 
>>> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is
>> fine
>>> for me locally. I recently made changes to FATE (namely the test
>>> requirements, vcodec.mak (where the vsynth-tests reside) among the
>> files
>>> affected), so I am interested in whether the failing tests are
>>> concentrated on the files recently changed by me (it would obviously
>> not
>>> haved pushed them if I knew them to cause issues; also patchwork was
>> fine).
>>> Are these issues reproducible? Can you bisect them?
>>> 
>>> - Andreas
>> 
>> This may be the alignment issue introduced in lavfi in
>> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00a
>> 680c6ebbaea58db95594d13d
>> assuming it was not fixed.
>> I think it only affected targets where av_cpu_max_align() returned 64.
> 
> My local build is run on Ubuntu in Hyper-V VM on Windows. The GitHub
> CI build runs on Azure DevOps which is also an Ubuntu VM (not sure 
> about the host).
> 
> AFAIC, Patchwork runs in a Docker container only. Not sure whether
> that could be the difference..
> 
> softworkz
> 
> 
> 
> 
> 
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
Soft Works April 30, 2022, 7:25 p.m. UTC | #8 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Felix LeClair
> Sent: Saturday, April 30, 2022 9:22 PM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> Both times I’ve encountered were on PopOS (think Debian/Ubuntu meets
> arch like timing release)

VM or bare metal installation?

PS: The people here don't like top-posting

Thanks,
softworkz
Felix LeClair April 30, 2022, 7:31 p.m. UTC | #9 
> On Apr 30, 2022, at 15:25, Soft Works <softworkz@hotmail.com> wrote:
> 
> 
> 
>> -----Original Message-----
>> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
>> Felix LeClair
>> Sent: Saturday, April 30, 2022 9:22 PM
>> To: FFmpeg development discussions and patches <ffmpeg-
>> devel@ffmpeg.org>
>> Subject: Re: [FFmpeg-devel] FATE Errors
>> 
>> Both times I’ve encountered were on PopOS (think Debian/Ubuntu meets
>> arch like timing release)
> 
> VM or bare metal installation?
> 
> PS: The people here don't like top-posting
> 
> Thanks,
> softworkz
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

It’s a bare metal installation,  intel x86 12th gen, ecores disabled, AVX512 enabled (instructions include ice lake + FP16), kernel 5.18rc3. 

I don’t believe it matters in this case, but one was with a bare config, another with a fair few external libs enabled. 

Both resulted in the same. 


Re top posting: My bad! Currently on mobile, using an ssh app to get to my Mac.
James Almer April 30, 2022, 7:33 p.m. UTC | #10 
On 4/30/2022 4:31 PM, Felix LeClair wrote:
> 
> 
>> On Apr 30, 2022, at 15:25, Soft Works <softworkz@hotmail.com> wrote:
>>
>> 
>>
>>> -----Original Message-----
>>> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
>>> Felix LeClair
>>> Sent: Saturday, April 30, 2022 9:22 PM
>>> To: FFmpeg development discussions and patches <ffmpeg-
>>> devel@ffmpeg.org>
>>> Subject: Re: [FFmpeg-devel] FATE Errors
>>>
>>> Both times I’ve encountered were on PopOS (think Debian/Ubuntu meets
>>> arch like timing release)
>>
>> VM or bare metal installation?
>>
>> PS: The people here don't like top-posting
>>
>> Thanks,
>> softworkz
>>
>> _______________________________________________
>> ffmpeg-devel mailing list
>> ffmpeg-devel@ffmpeg.org
>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>> To unsubscribe, visit link above, or email
>> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
> 
> It’s a bare metal installation,  intel x86 12th gen, ecores disabled, AVX512 enabled (instructions include ice lake + FP16), kernel 5.18rc3.

Then i was right. AVX512 support means av_cpu_max_align() returns 64.

> 
> I don’t believe it matters in this case, but one was with a bare config, another with a fair few external libs enabled.
> 
> Both resulted in the same.
> 
> 
> Re top posting: My bad! Currently on mobile, using an ssh app to get to my Mac.
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> 
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
Soft Works April 30, 2022, 7:39 p.m. UTC | #11 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> James Almer
> Sent: Saturday, April 30, 2022 9:33 PM
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> 
> 
> On 4/30/2022 4:31 PM, Felix LeClair wrote:
> >
> >
> >> On Apr 30, 2022, at 15:25, Soft Works <softworkz@hotmail.com>
> wrote:
> >>
> >> 
> >>
> >>> -----Original Message-----
> >>> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> >>> Felix LeClair
> >>> Sent: Saturday, April 30, 2022 9:22 PM
> >>> To: FFmpeg development discussions and patches <ffmpeg-
> >>> devel@ffmpeg.org>
> >>> Subject: Re: [FFmpeg-devel] FATE Errors
> >>>
> >>> Both times I’ve encountered were on PopOS (think Debian/Ubuntu
> meets
> >>> arch like timing release)
> >>
> >> VM or bare metal installation?
> >>
> >> PS: The people here don't like top-posting
> >>
> >> Thanks,
> >> softworkz
> >>
> >> _______________________________________________
> >> ffmpeg-devel mailing list
> >> ffmpeg-devel@ffmpeg.org
> >> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
> >>
> >> To unsubscribe, visit link above, or email
> >> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
> >
> > It’s a bare metal installation,  intel x86 12th gen, ecores
> disabled, AVX512 enabled (instructions include ice lake + FP16),
> kernel 5.18rc3.
> 
> Then i was right. AVX512 support means av_cpu_max_align() returns 64.

Makes sense. Mine is 11 gen and I looked at the CI build history
(https://dev.azure.com/githubsync/ffmpeg/_build?definitionId=9&_a=summary)
which shows some runs have passed and some failed during the
recent past - probably depending on the host machine type
that gets the job.

softworkz
Andreas Rheinhardt April 30, 2022, 8:02 p.m. UTC | #12 
James Almer:
> 
> 
> On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
>> Soft Works:
>>> Hi,
>>>
>>> is it a known issue that the current head of the master branch has
>>> FATE errors?
>>>
>>> I get the same locally as well as on the automated GitHub build.
>>> One is this:
>>>
>>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
>>> 14:23:44.330424058 +0000
>>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201
>>> +0000
>>> @@ -1,4 +1,4 @@
>>> -b2fa9b73c3547191ecc01b8163abd4e5
>>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>> -704f6a96f93c2409219bd48b74169041
>>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
>>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
>>> +8f6d565723ccf879ab2b5aa910b7ce21
>>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
>>> +0797fddea4835687dedddebbbe98fa8f
>>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
>>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
>>> Test vsynth2-mpeg2-422 failed. Look at
>>> tests/data/fate/vsynth2-mpeg2-422.err for details.
>>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
>>>
>>>
>>> Is anybody seeing the same?
>>>
>>> Thanks,
>>> sw
>>
>> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE is fine
>> for me locally. I recently made changes to FATE (namely the test
>> requirements, vcodec.mak (where the vsynth-tests reside) among the files
>> affected), so I am interested in whether the failing tests are
>> concentrated on the files recently changed by me (it would obviously not
>> haved pushed them if I knew them to cause issues; also patchwork was
>> fine).
>> Are these issues reproducible? Can you bisect them?
>>
>> - Andreas
> 
> This may be the alignment issue introduced in lavfi in
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00a680c6ebbaea58db95594d13d
> assuming it was not fixed.
> I think it only affected targets where av_cpu_max_align() returned 64.

You are completely right: Just making av_cpu_max_align return 64 allows
to reproduce the issue. And it has nothing to do with my recent FATE
patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be the only
contender for changes).

- Andreas
Soft Works May 27, 2022, 9:10 p.m. UTC | #13 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Andreas Rheinhardt
> Sent: Saturday, April 30, 2022 10:02 PM
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> James Almer:
> >
> >
> > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> >> Soft Works:
> >>> Hi,
> >>>
> >>> is it a known issue that the current head of the master branch
> has
> >>> FATE errors?
> >>>
> >>> I get the same locally as well as on the automated GitHub build.
> >>> One is this:
> >>>
> >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> >>> 14:23:44.330424058 +0000
> >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> 14:38:41.071678201
> >>> +0000
> >>> @@ -1,4 +1,4 @@
> >>> -b2fa9b73c3547191ecc01b8163abd4e5
> >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >>> -704f6a96f93c2409219bd48b74169041
> >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> >>> +8f6d565723ccf879ab2b5aa910b7ce21
> >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> >>> +0797fddea4835687dedddebbbe98fa8f
> >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> >>> Test vsynth2-mpeg2-422 failed. Look at
> >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> >>>
> >>>
> >>> Is anybody seeing the same?
> >>>
> >>> Thanks,
> >>> sw
> >>
> >> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE
> is fine
> >> for me locally. I recently made changes to FATE (namely the test
> >> requirements, vcodec.mak (where the vsynth-tests reside) among the
> files
> >> affected), so I am interested in whether the failing tests are
> >> concentrated on the files recently changed by me (it would
> obviously not
> >> haved pushed them if I knew them to cause issues; also patchwork
> was
> >> fine).
> >> Are these issues reproducible? Can you bisect them?
> >>
> >> - Andreas
> >
> > This may be the alignment issue introduced in lavfi in
> >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> a680c6ebbaea58db95594d13d
> > assuming it was not fixed.
> > I think it only affected targets where av_cpu_max_align() returned
> 64.
> 
> You are completely right: Just making av_cpu_max_align return 64
> allows
> to reproduce the issue. And it has nothing to do with my recent FATE
> patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be the only
> contender for changes).
> 


I found that the FATE errors can be avoided by using

./configure --disable-avx512

Yet, I hope it will be fixed at some time..


softworkz
Paul B Mahol May 28, 2022, 7:25 a.m. UTC | #14 
On Fri, May 27, 2022 at 11:11 PM Soft Works <softworkz@hotmail.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Andreas Rheinhardt
> > Sent: Saturday, April 30, 2022 10:02 PM
> > To: ffmpeg-devel@ffmpeg.org
> > Subject: Re: [FFmpeg-devel] FATE Errors
> >
> > James Almer:
> > >
> > >
> > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > >> Soft Works:
> > >>> Hi,
> > >>>
> > >>> is it a known issue that the current head of the master branch
> > has
> > >>> FATE errors?
> > >>>
> > >>> I get the same locally as well as on the automated GitHub build.
> > >>> One is this:
> > >>>
> > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > >>> 14:23:44.330424058 +0000
> > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > 14:38:41.071678201
> > >>> +0000
> > >>> @@ -1,4 +1,4 @@
> > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > >>> -704f6a96f93c2409219bd48b74169041
> > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > >>> +0797fddea4835687dedddebbbe98fa8f
> > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> > >>> Test vsynth2-mpeg2-422 failed. Look at
> > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error 1
> > >>>
> > >>>
> > >>> Is anybody seeing the same?
> > >>>
> > >>> Thanks,
> > >>> sw
> > >>
> > >> http://fate.ffmpeg.org/ doesn't show recent regressions and FATE
> > is fine
> > >> for me locally. I recently made changes to FATE (namely the test
> > >> requirements, vcodec.mak (where the vsynth-tests reside) among the
> > files
> > >> affected), so I am interested in whether the failing tests are
> > >> concentrated on the files recently changed by me (it would
> > obviously not
> > >> haved pushed them if I knew them to cause issues; also patchwork
> > was
> > >> fine).
> > >> Are these issues reproducible? Can you bisect them?
> > >>
> > >> - Andreas
> > >
> > > This may be the alignment issue introduced in lavfi in
> > >
> > https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > a680c6ebbaea58db95594d13d
> > > assuming it was not fixed.
> > > I think it only affected targets where av_cpu_max_align() returned
> > 64.
> >
> > You are completely right: Just making av_cpu_max_align return 64
> > allows
> > to reproduce the issue. And it has nothing to do with my recent FATE
> > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be the only
> > contender for changes).
> >
>
>
> I found that the FATE errors can be avoided by using
>
> ./configure --disable-avx512
>
> Yet, I hope it will be fixed at some time..
>

I hope you will post a fix for it soon.


>
>
> softworkz
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
Soft Works May 28, 2022, 7:26 a.m. UTC | #15 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Paul B Mahol
> Sent: Saturday, May 28, 2022 9:26 AM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> On Fri, May 27, 2022 at 11:11 PM Soft Works <softworkz@hotmail.com>
> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Andreas Rheinhardt
> > > Sent: Saturday, April 30, 2022 10:02 PM
> > > To: ffmpeg-devel@ffmpeg.org
> > > Subject: Re: [FFmpeg-devel] FATE Errors
> > >
> > > James Almer:
> > > >
> > > >
> > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > >> Soft Works:
> > > >>> Hi,
> > > >>>
> > > >>> is it a known issue that the current head of the master
> branch
> > > has
> > > >>> FATE errors?
> > > >>>
> > > >>> I get the same locally as well as on the automated GitHub
> build.
> > > >>> One is this:
> > > >>>
> > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > >>> 14:23:44.330424058 +0000
> > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > 14:38:41.071678201
> > > >>> +0000
> > > >>> @@ -1,4 +1,4 @@
> > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > >>> -704f6a96f93c2409219bd48b74169041
> > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error
> 1
> > > >>>
> > > >>>
> > > >>> Is anybody seeing the same?
> > > >>>
> > > >>> Thanks,
> > > >>> sw
> > > >>
> > > >> http://fate.ffmpeg.org/ doesn't show recent regressions and
> FATE
> > > is fine
> > > >> for me locally. I recently made changes to FATE (namely the
> test
> > > >> requirements, vcodec.mak (where the vsynth-tests reside) among
> the
> > > files
> > > >> affected), so I am interested in whether the failing tests are
> > > >> concentrated on the files recently changed by me (it would
> > > obviously not
> > > >> haved pushed them if I knew them to cause issues; also
> patchwork
> > > was
> > > >> fine).
> > > >> Are these issues reproducible? Can you bisect them?
> > > >>
> > > >> - Andreas
> > > >
> > > > This may be the alignment issue introduced in lavfi in
> > > >
> > >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > a680c6ebbaea58db95594d13d
> > > > assuming it was not fixed.
> > > > I think it only affected targets where av_cpu_max_align()
> returned
> > > 64.
> > >
> > > You are completely right: Just making av_cpu_max_align return 64
> > > allows
> > > to reproduce the issue. And it has nothing to do with my recent
> FATE
> > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be the
> only
> > > contender for changes).
> > >
> >
> >
> > I found that the FATE errors can be avoided by using
> >
> > ./configure --disable-avx512
> >
> > Yet, I hope it will be fixed at some time..
> >
> 
> I hope you will post a fix for it soon.

That depends on whether we want alignment values > 32bit..?
Paul B Mahol May 28, 2022, 7:34 a.m. UTC | #16 
On Sat, May 28, 2022 at 9:26 AM Soft Works <softworkz@hotmail.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Paul B Mahol
> > Sent: Saturday, May 28, 2022 9:26 AM
> > To: FFmpeg development discussions and patches <ffmpeg-
> > devel@ffmpeg.org>
> > Subject: Re: [FFmpeg-devel] FATE Errors
> >
> > On Fri, May 27, 2022 at 11:11 PM Soft Works <softworkz@hotmail.com>
> > wrote:
> >
> > >
> > >
> > > > -----Original Message-----
> > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > > Andreas Rheinhardt
> > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > To: ffmpeg-devel@ffmpeg.org
> > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > >
> > > > James Almer:
> > > > >
> > > > >
> > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > >> Soft Works:
> > > > >>> Hi,
> > > > >>>
> > > > >>> is it a known issue that the current head of the master
> > branch
> > > > has
> > > > >>> FATE errors?
> > > > >>>
> > > > >>> I get the same locally as well as on the automated GitHub
> > build.
> > > > >>> One is this:
> > > > >>>
> > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > > >>> 14:23:44.330424058 +0000
> > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > 14:38:41.071678201
> > > > >>> +0000
> > > > >>> @@ -1,4 +1,4 @@
> > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
> > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200
> > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422] Error
> > 1
> > > > >>>
> > > > >>>
> > > > >>> Is anybody seeing the same?
> > > > >>>
> > > > >>> Thanks,
> > > > >>> sw
> > > > >>
> > > > >> http://fate.ffmpeg.org/ doesn't show recent regressions and
> > FATE
> > > > is fine
> > > > >> for me locally. I recently made changes to FATE (namely the
> > test
> > > > >> requirements, vcodec.mak (where the vsynth-tests reside) among
> > the
> > > > files
> > > > >> affected), so I am interested in whether the failing tests are
> > > > >> concentrated on the files recently changed by me (it would
> > > > obviously not
> > > > >> haved pushed them if I knew them to cause issues; also
> > patchwork
> > > > was
> > > > >> fine).
> > > > >> Are these issues reproducible? Can you bisect them?
> > > > >>
> > > > >> - Andreas
> > > > >
> > > > > This may be the alignment issue introduced in lavfi in
> > > > >
> > > >
> > https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > a680c6ebbaea58db95594d13d
> > > > > assuming it was not fixed.
> > > > > I think it only affected targets where av_cpu_max_align()
> > returned
> > > > 64.
> > > >
> > > > You are completely right: Just making av_cpu_max_align return 64
> > > > allows
> > > > to reproduce the issue. And it has nothing to do with my recent
> > FATE
> > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be the
> > only
> > > > contender for changes).
> > > >
> > >
> > >
> > > I found that the FATE errors can be avoided by using
> > >
> > > ./configure --disable-avx512
> > >
> > > Yet, I hope it will be fixed at some time..
> > >
> >
> > I hope you will post a fix for it soon.
>
> That depends on whether we want alignment values > 32bit..?
>

Nope, its about mpeg video encoder in lavc that does funny things from old
days when linesize alignment was hardcoded.


> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
Soft Works May 28, 2022, 7:47 a.m. UTC | #17 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Paul B Mahol
> Sent: Saturday, May 28, 2022 9:35 AM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> On Sat, May 28, 2022 at 9:26 AM Soft Works <softworkz@hotmail.com>
> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Paul B Mahol
> > > Sent: Saturday, May 28, 2022 9:26 AM
> > > To: FFmpeg development discussions and patches <ffmpeg-
> > > devel@ffmpeg.org>
> > > Subject: Re: [FFmpeg-devel] FATE Errors
> > >
> > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> <softworkz@hotmail.com>
> > > wrote:
> > >
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> Behalf Of
> > > > > Andreas Rheinhardt
> > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > >
> > > > > James Almer:
> > > > > >
> > > > > >
> > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > >> Soft Works:
> > > > > >>> Hi,
> > > > > >>>
> > > > > >>> is it a known issue that the current head of the master
> > > branch
> > > > > has
> > > > > >>> FATE errors?
> > > > > >>>
> > > > > >>> I get the same locally as well as on the automated GitHub
> > > build.
> > > > > >>> One is this:
> > > > > >>>
> > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > > > >>> 14:23:44.330424058 +0000
> > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > 14:38:41.071678201
> > > > > >>> +0000
> > > > > >>> @@ -1,4 +1,4 @@
> > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/
> 7603200
> > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/
> 7603200
> > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422]
> Error
> > > 1
> > > > > >>>
> > > > > >>>
> > > > > >>> Is anybody seeing the same?
> > > > > >>>
> > > > > >>> Thanks,
> > > > > >>> sw
> > > > > >>
> > > > > >> http://fate.ffmpeg.org/ doesn't show recent regressions
> and
> > > FATE
> > > > > is fine
> > > > > >> for me locally. I recently made changes to FATE (namely
> the
> > > test
> > > > > >> requirements, vcodec.mak (where the vsynth-tests reside)
> among
> > > the
> > > > > files
> > > > > >> affected), so I am interested in whether the failing tests
> are
> > > > > >> concentrated on the files recently changed by me (it would
> > > > > obviously not
> > > > > >> haved pushed them if I knew them to cause issues; also
> > > patchwork
> > > > > was
> > > > > >> fine).
> > > > > >> Are these issues reproducible? Can you bisect them?
> > > > > >>
> > > > > >> - Andreas
> > > > > >
> > > > > > This may be the alignment issue introduced in lavfi in
> > > > > >
> > > > >
> > >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > a680c6ebbaea58db95594d13d
> > > > > > assuming it was not fixed.
> > > > > > I think it only affected targets where av_cpu_max_align()
> > > returned
> > > > > 64.
> > > > >
> > > > > You are completely right: Just making av_cpu_max_align return
> 64
> > > > > allows
> > > > > to reproduce the issue. And it has nothing to do with my
> recent
> > > FATE
> > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be
> the
> > > only
> > > > > contender for changes).
> > > > >
> > > >
> > > >
> > > > I found that the FATE errors can be avoided by using
> > > >
> > > > ./configure --disable-avx512
> > > >
> > > > Yet, I hope it will be fixed at some time..
> > > >
> > >
> > > I hope you will post a fix for it soon.
> >
> > That depends on whether we want alignment values > 32bit..?
> >
> 
> Nope, its about mpeg video encoder in lavc that does funny things
> from old
> days when linesize alignment was hardcoded.

Adding this 

   align = FFMIN(align, 32);

to ff_default_get_video_buffer2() resolves the issue.

That's why I had written:

> That depends on whether we want alignment values > 32bit..?


Thanks,
sw
Paul B Mahol May 28, 2022, 8:04 a.m. UTC | #18 
On Sat, May 28, 2022 at 9:47 AM Soft Works <softworkz@hotmail.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Paul B Mahol
> > Sent: Saturday, May 28, 2022 9:35 AM
> > To: FFmpeg development discussions and patches <ffmpeg-
> > devel@ffmpeg.org>
> > Subject: Re: [FFmpeg-devel] FATE Errors
> >
> > On Sat, May 28, 2022 at 9:26 AM Soft Works <softworkz@hotmail.com>
> > wrote:
> >
> > >
> > >
> > > > -----Original Message-----
> > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > > Paul B Mahol
> > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > devel@ffmpeg.org>
> > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > >
> > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > <softworkz@hotmail.com>
> > > > wrote:
> > > >
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > Behalf Of
> > > > > > Andreas Rheinhardt
> > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > >
> > > > > > James Almer:
> > > > > > >
> > > > > > >
> > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > >> Soft Works:
> > > > > > >>> Hi,
> > > > > > >>>
> > > > > > >>> is it a known issue that the current head of the master
> > > > branch
> > > > > > has
> > > > > > >>> FATE errors?
> > > > > > >>>
> > > > > > >>> I get the same locally as well as on the automated GitHub
> > > > build.
> > > > > > >>> One is this:
> > > > > > >>>
> > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > > > > >>> 14:23:44.330424058 +0000
> > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > > 14:38:41.071678201
> > > > > > >>> +0000
> > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/
> > 7603200
> > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/
> > 7603200
> > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-422]
> > Error
> > > > 1
> > > > > > >>>
> > > > > > >>>
> > > > > > >>> Is anybody seeing the same?
> > > > > > >>>
> > > > > > >>> Thanks,
> > > > > > >>> sw
> > > > > > >>
> > > > > > >> http://fate.ffmpeg.org/ doesn't show recent regressions
> > and
> > > > FATE
> > > > > > is fine
> > > > > > >> for me locally. I recently made changes to FATE (namely
> > the
> > > > test
> > > > > > >> requirements, vcodec.mak (where the vsynth-tests reside)
> > among
> > > > the
> > > > > > files
> > > > > > >> affected), so I am interested in whether the failing tests
> > are
> > > > > > >> concentrated on the files recently changed by me (it would
> > > > > > obviously not
> > > > > > >> haved pushed them if I knew them to cause issues; also
> > > > patchwork
> > > > > > was
> > > > > > >> fine).
> > > > > > >> Are these issues reproducible? Can you bisect them?
> > > > > > >>
> > > > > > >> - Andreas
> > > > > > >
> > > > > > > This may be the alignment issue introduced in lavfi in
> > > > > > >
> > > > > >
> > > >
> > https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > a680c6ebbaea58db95594d13d
> > > > > > > assuming it was not fixed.
> > > > > > > I think it only affected targets where av_cpu_max_align()
> > > > returned
> > > > > > 64.
> > > > > >
> > > > > > You are completely right: Just making av_cpu_max_align return
> > 64
> > > > > > allows
> > > > > > to reproduce the issue. And it has nothing to do with my
> > recent
> > > > FATE
> > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would be
> > the
> > > > only
> > > > > > contender for changes).
> > > > > >
> > > > >
> > > > >
> > > > > I found that the FATE errors can be avoided by using
> > > > >
> > > > > ./configure --disable-avx512
> > > > >
> > > > > Yet, I hope it will be fixed at some time..
> > > > >
> > > >
> > > > I hope you will post a fix for it soon.
> > >
> > > That depends on whether we want alignment values > 32bit..?
> > >
> >
> > Nope, its about mpeg video encoder in lavc that does funny things
> > from old
> > days when linesize alignment was hardcoded.
>
> Adding this
>
>    align = FFMIN(align, 32);
>
> to ff_default_get_video_buffer2() resolves the issue.
>

Maybe it resolves it for you but that does not mean in any way that that is
valid or correct solution.


>
> That's why I had written:
>
> > That depends on whether we want alignment values > 32bit..?
>

32bit is 4 integers. so I failed to follow whatever you tried to
communicate.

>
>
> Thanks,
> sw
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
Soft Works May 28, 2022, 8:12 a.m. UTC | #19 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Paul B Mahol
> Sent: Saturday, May 28, 2022 10:04 AM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> On Sat, May 28, 2022 at 9:47 AM Soft Works <softworkz@hotmail.com>
> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Paul B Mahol
> > > Sent: Saturday, May 28, 2022 9:35 AM
> > > To: FFmpeg development discussions and patches <ffmpeg-
> > > devel@ffmpeg.org>
> > > Subject: Re: [FFmpeg-devel] FATE Errors
> > >
> > > On Sat, May 28, 2022 at 9:26 AM Soft Works
> <softworkz@hotmail.com>
> > > wrote:
> > >
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> Behalf Of
> > > > > Paul B Mahol
> > > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > devel@ffmpeg.org>
> > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > >
> > > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > > <softworkz@hotmail.com>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > > Behalf Of
> > > > > > > Andreas Rheinhardt
> > > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > >
> > > > > > > James Almer:
> > > > > > > >
> > > > > > > >
> > > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > > >> Soft Works:
> > > > > > > >>> Hi,
> > > > > > > >>>
> > > > > > > >>> is it a known issue that the current head of the
> master
> > > > > branch
> > > > > > > has
> > > > > > > >>> FATE errors?
> > > > > > > >>>
> > > > > > > >>> I get the same locally as well as on the automated
> GitHub
> > > > > build.
> > > > > > > >>> One is this:
> > > > > > > >>>
> > > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > > > > > >>> 14:23:44.330424058 +0000
> > > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > > > 14:38:41.071678201
> > > > > > > >>> +0000
> > > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/
> > > 7603200
> > > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/
> > > 7603200
> > > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-
> 422]
> > > Error
> > > > > 1
> > > > > > > >>>
> > > > > > > >>>
> > > > > > > >>> Is anybody seeing the same?
> > > > > > > >>>
> > > > > > > >>> Thanks,
> > > > > > > >>> sw
> > > > > > > >>
> > > > > > > >> http://fate.ffmpeg.org/ doesn't show recent
> regressions
> > > and
> > > > > FATE
> > > > > > > is fine
> > > > > > > >> for me locally. I recently made changes to FATE
> (namely
> > > the
> > > > > test
> > > > > > > >> requirements, vcodec.mak (where the vsynth-tests
> reside)
> > > among
> > > > > the
> > > > > > > files
> > > > > > > >> affected), so I am interested in whether the failing
> tests
> > > are
> > > > > > > >> concentrated on the files recently changed by me (it
> would
> > > > > > > obviously not
> > > > > > > >> haved pushed them if I knew them to cause issues; also
> > > > > patchwork
> > > > > > > was
> > > > > > > >> fine).
> > > > > > > >> Are these issues reproducible? Can you bisect them?
> > > > > > > >>
> > > > > > > >> - Andreas
> > > > > > > >
> > > > > > > > This may be the alignment issue introduced in lavfi in
> > > > > > > >
> > > > > > >
> > > > >
> > >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > > a680c6ebbaea58db95594d13d
> > > > > > > > assuming it was not fixed.
> > > > > > > > I think it only affected targets where
> av_cpu_max_align()
> > > > > returned
> > > > > > > 64.
> > > > > > >
> > > > > > > You are completely right: Just making av_cpu_max_align
> return
> > > 64
> > > > > > > allows
> > > > > > > to reproduce the issue. And it has nothing to do with my
> > > recent
> > > > > FATE
> > > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would
> be
> > > the
> > > > > only
> > > > > > > contender for changes).
> > > > > > >
> > > > > >
> > > > > >
> > > > > > I found that the FATE errors can be avoided by using
> > > > > >
> > > > > > ./configure --disable-avx512
> > > > > >
> > > > > > Yet, I hope it will be fixed at some time..
> > > > > >
> > > > >
> > > > > I hope you will post a fix for it soon.
> > > >
> > > > That depends on whether we want alignment values > 32bit..?
> > > >
> > >
> > > Nope, its about mpeg video encoder in lavc that does funny things
> > > from old
> > > days when linesize alignment was hardcoded.
> >
> > Adding this
> >
> >    align = FFMIN(align, 32);
> >
> > to ff_default_get_video_buffer2() resolves the issue.
> >
> 
> Maybe it resolves it for you but that does not mean in any way that
> that is
> valid or correct solution.
> 
> 
> >
> > That's why I had written:
> >
> > > That depends on whether we want alignment values > 32bit..?
> >
> 
> 32bit is 4 integers. so I failed to follow whatever you tried to
> communicate.

Sorry for being unclear. What I meant to ask is this:

In cases where av_cpu_max_align() returns something > 32 - for 
example 64, do we have to use that max-align value of 64 for alignment
in video.c and framepool,c or could we just limit this to 32?
(hence the FFMIN).

Limiting the alignment to 32 appears to fix the issue, meaning that
the three fate tests mentioned above are producing the expected ref
output and don't fail anymore.

softworkz
Paul B Mahol May 28, 2022, 8:33 a.m. UTC | #20 
On Sat, May 28, 2022 at 10:12 AM Soft Works <softworkz@hotmail.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Paul B Mahol
> > Sent: Saturday, May 28, 2022 10:04 AM
> > To: FFmpeg development discussions and patches <ffmpeg-
> > devel@ffmpeg.org>
> > Subject: Re: [FFmpeg-devel] FATE Errors
> >
> > On Sat, May 28, 2022 at 9:47 AM Soft Works <softworkz@hotmail.com>
> > wrote:
> >
> > >
> > >
> > > > -----Original Message-----
> > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > > Paul B Mahol
> > > > Sent: Saturday, May 28, 2022 9:35 AM
> > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > devel@ffmpeg.org>
> > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > >
> > > > On Sat, May 28, 2022 at 9:26 AM Soft Works
> > <softworkz@hotmail.com>
> > > > wrote:
> > > >
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > Behalf Of
> > > > > > Paul B Mahol
> > > > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > > devel@ffmpeg.org>
> > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > >
> > > > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > > > <softworkz@hotmail.com>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > > > Behalf Of
> > > > > > > > Andreas Rheinhardt
> > > > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > >
> > > > > > > > James Almer:
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > > > >> Soft Works:
> > > > > > > > >>> Hi,
> > > > > > > > >>>
> > > > > > > > >>> is it a known issue that the current head of the
> > master
> > > > > > branch
> > > > > > > > has
> > > > > > > > >>> FATE errors?
> > > > > > > > >>>
> > > > > > > > >>> I get the same locally as well as on the automated
> > GitHub
> > > > > > build.
> > > > > > > > >>> One is this:
> > > > > > > > >>>
> > > > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30
> > > > > > > > >>> 14:23:44.330424058 +0000
> > > > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > > > > 14:38:41.071678201
> > > > > > > > >>> +0000
> > > > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/
> > > > 7603200
> > > > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/
> > > > 7603200
> > > > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for details.
> > > > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-mpeg2-
> > 422]
> > > > Error
> > > > > > 1
> > > > > > > > >>>
> > > > > > > > >>>
> > > > > > > > >>> Is anybody seeing the same?
> > > > > > > > >>>
> > > > > > > > >>> Thanks,
> > > > > > > > >>> sw
> > > > > > > > >>
> > > > > > > > >> http://fate.ffmpeg.org/ doesn't show recent
> > regressions
> > > > and
> > > > > > FATE
> > > > > > > > is fine
> > > > > > > > >> for me locally. I recently made changes to FATE
> > (namely
> > > > the
> > > > > > test
> > > > > > > > >> requirements, vcodec.mak (where the vsynth-tests
> > reside)
> > > > among
> > > > > > the
> > > > > > > > files
> > > > > > > > >> affected), so I am interested in whether the failing
> > tests
> > > > are
> > > > > > > > >> concentrated on the files recently changed by me (it
> > would
> > > > > > > > obviously not
> > > > > > > > >> haved pushed them if I knew them to cause issues; also
> > > > > > patchwork
> > > > > > > > was
> > > > > > > > >> fine).
> > > > > > > > >> Are these issues reproducible? Can you bisect them?
> > > > > > > > >>
> > > > > > > > >> - Andreas
> > > > > > > > >
> > > > > > > > > This may be the alignment issue introduced in lavfi in
> > > > > > > > >
> > > > > > > >
> > > > > >
> > > >
> > https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > > > a680c6ebbaea58db95594d13d
> > > > > > > > > assuming it was not fixed.
> > > > > > > > > I think it only affected targets where
> > av_cpu_max_align()
> > > > > > returned
> > > > > > > > 64.
> > > > > > > >
> > > > > > > > You are completely right: Just making av_cpu_max_align
> > return
> > > > 64
> > > > > > > > allows
> > > > > > > > to reproduce the issue. And it has nothing to do with my
> > > > recent
> > > > > > FATE
> > > > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1 would
> > be
> > > > the
> > > > > > only
> > > > > > > > contender for changes).
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > I found that the FATE errors can be avoided by using
> > > > > > >
> > > > > > > ./configure --disable-avx512
> > > > > > >
> > > > > > > Yet, I hope it will be fixed at some time..
> > > > > > >
> > > > > >
> > > > > > I hope you will post a fix for it soon.
> > > > >
> > > > > That depends on whether we want alignment values > 32bit..?
> > > > >
> > > >
> > > > Nope, its about mpeg video encoder in lavc that does funny things
> > > > from old
> > > > days when linesize alignment was hardcoded.
> > >
> > > Adding this
> > >
> > >    align = FFMIN(align, 32);
> > >
> > > to ff_default_get_video_buffer2() resolves the issue.
> > >
> >
> > Maybe it resolves it for you but that does not mean in any way that
> > that is
> > valid or correct solution.
> >
> >
> > >
> > > That's why I had written:
> > >
> > > > That depends on whether we want alignment values > 32bit..?
> > >
> >
> > 32bit is 4 integers. so I failed to follow whatever you tried to
> > communicate.
>
> Sorry for being unclear. What I meant to ask is this:
>
> In cases where av_cpu_max_align() returns something > 32 - for
> example 64, do we have to use that max-align value of 64 for alignment
> in video.c and framepool,c or could we just limit this to 32?
> (hence the FFMIN).
>
> Limiting the alignment to 32 appears to fix the issue, meaning that
> the three fate tests mentioned above are producing the expected ref
> output and don't fail anymore.
>

That is hack.


>
> softworkz
>
>
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
Soft Works May 28, 2022, 8:41 a.m. UTC | #21 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Paul B Mahol
> Sent: Saturday, May 28, 2022 10:33 AM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> On Sat, May 28, 2022 at 10:12 AM Soft Works <softworkz@hotmail.com>
> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Paul B Mahol
> > > Sent: Saturday, May 28, 2022 10:04 AM
> > > To: FFmpeg development discussions and patches <ffmpeg-
> > > devel@ffmpeg.org>
> > > Subject: Re: [FFmpeg-devel] FATE Errors
> > >
> > > On Sat, May 28, 2022 at 9:47 AM Soft Works
> <softworkz@hotmail.com>
> > > wrote:
> > >
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> Behalf Of
> > > > > Paul B Mahol
> > > > > Sent: Saturday, May 28, 2022 9:35 AM
> > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > devel@ffmpeg.org>
> > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > >
> > > > > On Sat, May 28, 2022 at 9:26 AM Soft Works
> > > <softworkz@hotmail.com>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > > Behalf Of
> > > > > > > Paul B Mahol
> > > > > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > > > devel@ffmpeg.org>
> > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > >
> > > > > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > > > > <softworkz@hotmail.com>
> > > > > > > wrote:
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org>
> On
> > > > > Behalf Of
> > > > > > > > > Andreas Rheinhardt
> > > > > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > > >
> > > > > > > > > James Almer:
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > > > > >> Soft Works:
> > > > > > > > > >>> Hi,
> > > > > > > > > >>>
> > > > > > > > > >>> is it a known issue that the current head of the
> > > master
> > > > > > > branch
> > > > > > > > > has
> > > > > > > > > >>> FATE errors?
> > > > > > > > > >>>
> > > > > > > > > >>> I get the same locally as well as on the
> automated
> > > GitHub
> > > > > > > build.
> > > > > > > > > >>> One is this:
> > > > > > > > > >>>
> > > > > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-
> 30
> > > > > > > > > >>> 14:23:44.330424058 +0000
> > > > > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > > > > > 14:38:41.071678201
> > > > > > > > > >>> +0000
> > > > > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> > > > > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes:
> 7603200/
> > > > > 7603200
> > > > > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-
> 422.mpeg2video
> > > > > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes:
> 7603200/
> > > > > 7603200
> > > > > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for
> details.
> > > > > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-
> mpeg2-
> > > 422]
> > > > > Error
> > > > > > > 1
> > > > > > > > > >>>
> > > > > > > > > >>>
> > > > > > > > > >>> Is anybody seeing the same?
> > > > > > > > > >>>
> > > > > > > > > >>> Thanks,
> > > > > > > > > >>> sw
> > > > > > > > > >>
> > > > > > > > > >> http://fate.ffmpeg.org/ doesn't show recent
> > > regressions
> > > > > and
> > > > > > > FATE
> > > > > > > > > is fine
> > > > > > > > > >> for me locally. I recently made changes to FATE
> > > (namely
> > > > > the
> > > > > > > test
> > > > > > > > > >> requirements, vcodec.mak (where the vsynth-tests
> > > reside)
> > > > > among
> > > > > > > the
> > > > > > > > > files
> > > > > > > > > >> affected), so I am interested in whether the
> failing
> > > tests
> > > > > are
> > > > > > > > > >> concentrated on the files recently changed by me
> (it
> > > would
> > > > > > > > > obviously not
> > > > > > > > > >> haved pushed them if I knew them to cause issues;
> also
> > > > > > > patchwork
> > > > > > > > > was
> > > > > > > > > >> fine).
> > > > > > > > > >> Are these issues reproducible? Can you bisect
> them?
> > > > > > > > > >>
> > > > > > > > > >> - Andreas
> > > > > > > > > >
> > > > > > > > > > This may be the alignment issue introduced in lavfi
> in
> > > > > > > > > >
> > > > > > > > >
> > > > > > >
> > > > >
> > >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > > > > a680c6ebbaea58db95594d13d
> > > > > > > > > > assuming it was not fixed.
> > > > > > > > > > I think it only affected targets where
> > > av_cpu_max_align()
> > > > > > > returned
> > > > > > > > > 64.
> > > > > > > > >
> > > > > > > > > You are completely right: Just making
> av_cpu_max_align
> > > return
> > > > > 64
> > > > > > > > > allows
> > > > > > > > > to reproduce the issue. And it has nothing to do with
> my
> > > > > recent
> > > > > > > FATE
> > > > > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1
> would
> > > be
> > > > > the
> > > > > > > only
> > > > > > > > > contender for changes).
> > > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > I found that the FATE errors can be avoided by using
> > > > > > > >
> > > > > > > > ./configure --disable-avx512
> > > > > > > >
> > > > > > > > Yet, I hope it will be fixed at some time..
> > > > > > > >
> > > > > > >
> > > > > > > I hope you will post a fix for it soon.
> > > > > >
> > > > > > That depends on whether we want alignment values > 32bit..?
> > > > > >
> > > > >
> > > > > Nope, its about mpeg video encoder in lavc that does funny
> things
> > > > > from old
> > > > > days when linesize alignment was hardcoded.
> > > >
> > > > Adding this
> > > >
> > > >    align = FFMIN(align, 32);
> > > >
> > > > to ff_default_get_video_buffer2() resolves the issue.
> > > >
> > >
> > > Maybe it resolves it for you but that does not mean in any way
> that
> > > that is
> > > valid or correct solution.
> > >
> > >
> > > >
> > > > That's why I had written:
> > > >
> > > > > That depends on whether we want alignment values > 32bit..?
> > > >
> > >
> > > 32bit is 4 integers. so I failed to follow whatever you tried to
> > > communicate.
> >
> > Sorry for being unclear. What I meant to ask is this:
> >
> > In cases where av_cpu_max_align() returns something > 32 - for
> > example 64, do we have to use that max-align value of 64 for
> alignment
> > in video.c and framepool,c or could we just limit this to 32?
> > (hence the FFMIN).
> >
> > Limiting the alignment to 32 appears to fix the issue, meaning that
> > the three fate tests mentioned above are producing the expected ref
> > output and don't fail anymore.
> >
> 
> That is hack.

Only when 64bit alignment would be required or beneficial in terms of 
performance (e.g. AVX512 operations). 
Which is the reason for my question. If yes, then this would be 
surely a bad idea.

sw
Paul B Mahol May 28, 2022, 8:50 a.m. UTC | #22 
On Sat, May 28, 2022 at 10:41 AM Soft Works <softworkz@hotmail.com> wrote:

>
>
> > -----Original Message-----
> > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > Paul B Mahol
> > Sent: Saturday, May 28, 2022 10:33 AM
> > To: FFmpeg development discussions and patches <ffmpeg-
> > devel@ffmpeg.org>
> > Subject: Re: [FFmpeg-devel] FATE Errors
> >
> > On Sat, May 28, 2022 at 10:12 AM Soft Works <softworkz@hotmail.com>
> > wrote:
> >
> > >
> > >
> > > > -----Original Message-----
> > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > > Paul B Mahol
> > > > Sent: Saturday, May 28, 2022 10:04 AM
> > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > devel@ffmpeg.org>
> > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > >
> > > > On Sat, May 28, 2022 at 9:47 AM Soft Works
> > <softworkz@hotmail.com>
> > > > wrote:
> > > >
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > Behalf Of
> > > > > > Paul B Mahol
> > > > > > Sent: Saturday, May 28, 2022 9:35 AM
> > > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > > devel@ffmpeg.org>
> > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > >
> > > > > > On Sat, May 28, 2022 at 9:26 AM Soft Works
> > > > <softworkz@hotmail.com>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > > -----Original Message-----
> > > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > > > Behalf Of
> > > > > > > > Paul B Mahol
> > > > > > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > > > > devel@ffmpeg.org>
> > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > >
> > > > > > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > > > > > <softworkz@hotmail.com>
> > > > > > > > wrote:
> > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > > -----Original Message-----
> > > > > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org>
> > On
> > > > > > Behalf Of
> > > > > > > > > > Andreas Rheinhardt
> > > > > > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > > > >
> > > > > > > > > > James Almer:
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > > > > > >> Soft Works:
> > > > > > > > > > >>> Hi,
> > > > > > > > > > >>>
> > > > > > > > > > >>> is it a known issue that the current head of the
> > > > master
> > > > > > > > branch
> > > > > > > > > > has
> > > > > > > > > > >>> FATE errors?
> > > > > > > > > > >>>
> > > > > > > > > > >>> I get the same locally as well as on the
> > automated
> > > > GitHub
> > > > > > > > build.
> > > > > > > > > > >>> One is this:
> > > > > > > > > > >>>
> > > > > > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-
> > 30
> > > > > > > > > > >>> 14:23:44.330424058 +0000
> > > > > > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30
> > > > > > > > > > 14:38:41.071678201
> > > > > > > > > > >>> +0000
> > > > > > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-
> > 422.mpeg2video
> > > > > > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes:
> > 7603200/
> > > > > > 7603200
> > > > > > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-
> > 422.mpeg2video
> > > > > > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
> > > > > > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes:
> > 7603200/
> > > > > > 7603200
> > > > > > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for
> > details.
> > > > > > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-
> > mpeg2-
> > > > 422]
> > > > > > Error
> > > > > > > > 1
> > > > > > > > > > >>>
> > > > > > > > > > >>>
> > > > > > > > > > >>> Is anybody seeing the same?
> > > > > > > > > > >>>
> > > > > > > > > > >>> Thanks,
> > > > > > > > > > >>> sw
> > > > > > > > > > >>
> > > > > > > > > > >> http://fate.ffmpeg.org/ doesn't show recent
> > > > regressions
> > > > > > and
> > > > > > > > FATE
> > > > > > > > > > is fine
> > > > > > > > > > >> for me locally. I recently made changes to FATE
> > > > (namely
> > > > > > the
> > > > > > > > test
> > > > > > > > > > >> requirements, vcodec.mak (where the vsynth-tests
> > > > reside)
> > > > > > among
> > > > > > > > the
> > > > > > > > > > files
> > > > > > > > > > >> affected), so I am interested in whether the
> > failing
> > > > tests
> > > > > > are
> > > > > > > > > > >> concentrated on the files recently changed by me
> > (it
> > > > would
> > > > > > > > > > obviously not
> > > > > > > > > > >> haved pushed them if I knew them to cause issues;
> > also
> > > > > > > > patchwork
> > > > > > > > > > was
> > > > > > > > > > >> fine).
> > > > > > > > > > >> Are these issues reproducible? Can you bisect
> > them?
> > > > > > > > > > >>
> > > > > > > > > > >> - Andreas
> > > > > > > > > > >
> > > > > > > > > > > This may be the alignment issue introduced in lavfi
> > in
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > >
> > > > > >
> > > >
> > https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > > > > > a680c6ebbaea58db95594d13d
> > > > > > > > > > > assuming it was not fixed.
> > > > > > > > > > > I think it only affected targets where
> > > > av_cpu_max_align()
> > > > > > > > returned
> > > > > > > > > > 64.
> > > > > > > > > >
> > > > > > > > > > You are completely right: Just making
> > av_cpu_max_align
> > > > return
> > > > > > 64
> > > > > > > > > > allows
> > > > > > > > > > to reproduce the issue. And it has nothing to do with
> > my
> > > > > > recent
> > > > > > > > FATE
> > > > > > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1
> > would
> > > > be
> > > > > > the
> > > > > > > > only
> > > > > > > > > > contender for changes).
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > I found that the FATE errors can be avoided by using
> > > > > > > > >
> > > > > > > > > ./configure --disable-avx512
> > > > > > > > >
> > > > > > > > > Yet, I hope it will be fixed at some time..
> > > > > > > > >
> > > > > > > >
> > > > > > > > I hope you will post a fix for it soon.
> > > > > > >
> > > > > > > That depends on whether we want alignment values > 32bit..?
> > > > > > >
> > > > > >
> > > > > > Nope, its about mpeg video encoder in lavc that does funny
> > things
> > > > > > from old
> > > > > > days when linesize alignment was hardcoded.
> > > > >
> > > > > Adding this
> > > > >
> > > > >    align = FFMIN(align, 32);
> > > > >
> > > > > to ff_default_get_video_buffer2() resolves the issue.
> > > > >
> > > >
> > > > Maybe it resolves it for you but that does not mean in any way
> > that
> > > > that is
> > > > valid or correct solution.
> > > >
> > > >
> > > > >
> > > > > That's why I had written:
> > > > >
> > > > > > That depends on whether we want alignment values > 32bit..?
> > > > >
> > > >
> > > > 32bit is 4 integers. so I failed to follow whatever you tried to
> > > > communicate.
> > >
> > > Sorry for being unclear. What I meant to ask is this:
> > >
> > > In cases where av_cpu_max_align() returns something > 32 - for
> > > example 64, do we have to use that max-align value of 64 for
> > alignment
> > > in video.c and framepool,c or could we just limit this to 32?
> > > (hence the FFMIN).
> > >
> > > Limiting the alignment to 32 appears to fix the issue, meaning that
> > > the three fate tests mentioned above are producing the expected ref
> > > output and don't fail anymore.
> > >
> >
> > That is hack.
>
> Only when 64bit alignment would be required or beneficial in terms of
> performance (e.g. AVX512 operations).


AVX512 operations are not using 64bit alignment.


>
> Which is the reason for my question. If yes, then this would be
> surely a bad idea.
>

It is not Yes/No answer here.


>
> sw
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
Soft Works May 28, 2022, 8:57 a.m. UTC | #23 
> -----Original Message-----
> From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> Paul B Mahol
> Sent: Saturday, May 28, 2022 10:50 AM
> To: FFmpeg development discussions and patches <ffmpeg-
> devel@ffmpeg.org>
> Subject: Re: [FFmpeg-devel] FATE Errors
> 
> On Sat, May 28, 2022 at 10:41 AM Soft Works <softworkz@hotmail.com>
> wrote:
> 
> >
> >
> > > -----Original Message-----
> > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On Behalf Of
> > > Paul B Mahol
> > > Sent: Saturday, May 28, 2022 10:33 AM
> > > To: FFmpeg development discussions and patches <ffmpeg-
> > > devel@ffmpeg.org>
> > > Subject: Re: [FFmpeg-devel] FATE Errors
> > >
> > > On Sat, May 28, 2022 at 10:12 AM Soft Works
> <softworkz@hotmail.com>
> > > wrote:
> > >
> > > >
> > > >
> > > > > -----Original Message-----
> > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> Behalf Of
> > > > > Paul B Mahol
> > > > > Sent: Saturday, May 28, 2022 10:04 AM
> > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > devel@ffmpeg.org>
> > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > >
> > > > > On Sat, May 28, 2022 at 9:47 AM Soft Works
> > > <softworkz@hotmail.com>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > >
> > > > > > > -----Original Message-----
> > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org> On
> > > Behalf Of
> > > > > > > Paul B Mahol
> > > > > > > Sent: Saturday, May 28, 2022 9:35 AM
> > > > > > > To: FFmpeg development discussions and patches <ffmpeg-
> > > > > > > devel@ffmpeg.org>
> > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > >
> > > > > > > On Sat, May 28, 2022 at 9:26 AM Soft Works
> > > > > <softworkz@hotmail.com>
> > > > > > > wrote:
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > > -----Original Message-----
> > > > > > > > > From: ffmpeg-devel <ffmpeg-devel-bounces@ffmpeg.org>
> On
> > > > > Behalf Of
> > > > > > > > > Paul B Mahol
> > > > > > > > > Sent: Saturday, May 28, 2022 9:26 AM
> > > > > > > > > To: FFmpeg development discussions and patches
> <ffmpeg-
> > > > > > > > > devel@ffmpeg.org>
> > > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > > >
> > > > > > > > > On Fri, May 27, 2022 at 11:11 PM Soft Works
> > > > > > > <softworkz@hotmail.com>
> > > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > > -----Original Message-----
> > > > > > > > > > > From: ffmpeg-devel <ffmpeg-devel-
> bounces@ffmpeg.org>
> > > On
> > > > > > > Behalf Of
> > > > > > > > > > > Andreas Rheinhardt
> > > > > > > > > > > Sent: Saturday, April 30, 2022 10:02 PM
> > > > > > > > > > > To: ffmpeg-devel@ffmpeg.org
> > > > > > > > > > > Subject: Re: [FFmpeg-devel] FATE Errors
> > > > > > > > > > >
> > > > > > > > > > > James Almer:
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > On 4/30/2022 4:06 PM, Andreas Rheinhardt wrote:
> > > > > > > > > > > >> Soft Works:
> > > > > > > > > > > >>> Hi,
> > > > > > > > > > > >>>
> > > > > > > > > > > >>> is it a known issue that the current head of
> the
> > > > > master
> > > > > > > > > branch
> > > > > > > > > > > has
> > > > > > > > > > > >>> FATE errors?
> > > > > > > > > > > >>>
> > > > > > > > > > > >>> I get the same locally as well as on the
> > > automated
> > > > > GitHub
> > > > > > > > > build.
> > > > > > > > > > > >>> One is this:
> > > > > > > > > > > >>>
> > > > > > > > > > > >>> --- ./tests/ref/vsynth/vsynth2-mpeg2-422
> 2022-04-
> > > 30
> > > > > > > > > > > >>> 14:23:44.330424058 +0000
> > > > > > > > > > > >>> +++ tests/data/fate/vsynth2-mpeg2-422 2022-
> 04-30
> > > > > > > > > > > 14:38:41.071678201
> > > > > > > > > > > >>> +0000
> > > > > > > > > > > >>> @@ -1,4 +1,4 @@
> > > > > > > > > > > >>> -b2fa9b73c3547191ecc01b8163abd4e5
> > > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > > > >>> -379164 tests/data/fate/vsynth2-mpeg2-
> > > 422.mpeg2video
> > > > > > > > > > > >>> -704f6a96f93c2409219bd48b74169041
> > > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> > > > > > > > > > > >>> -stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes:
> > > 7603200/
> > > > > > > 7603200
> > > > > > > > > > > >>> +8f6d565723ccf879ab2b5aa910b7ce21
> > > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
> > > > > > > > > > > >>> +380544 tests/data/fate/vsynth2-mpeg2-
> > > 422.mpeg2video
> > > > > > > > > > > >>> +0797fddea4835687dedddebbbe98fa8f
> > > > > > > > > > > >>> *tests/data/fate/vsynth2-mpeg2-
> 422.out.rawvideo
> > > > > > > > > > > >>> +stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes:
> > > 7603200/
> > > > > > > 7603200
> > > > > > > > > > > >>> Test vsynth2-mpeg2-422 failed. Look at
> > > > > > > > > > > >>> tests/data/fate/vsynth2-mpeg2-422.err for
> > > details.
> > > > > > > > > > > >>> make: *** [tests/Makefile:277: fate-vsynth2-
> > > mpeg2-
> > > > > 422]
> > > > > > > Error
> > > > > > > > > 1
> > > > > > > > > > > >>>
> > > > > > > > > > > >>>
> > > > > > > > > > > >>> Is anybody seeing the same?
> > > > > > > > > > > >>>
> > > > > > > > > > > >>> Thanks,
> > > > > > > > > > > >>> sw
> > > > > > > > > > > >>
> > > > > > > > > > > >> http://fate.ffmpeg.org/ doesn't show recent
> > > > > regressions
> > > > > > > and
> > > > > > > > > FATE
> > > > > > > > > > > is fine
> > > > > > > > > > > >> for me locally. I recently made changes to
> FATE
> > > > > (namely
> > > > > > > the
> > > > > > > > > test
> > > > > > > > > > > >> requirements, vcodec.mak (where the vsynth-
> tests
> > > > > reside)
> > > > > > > among
> > > > > > > > > the
> > > > > > > > > > > files
> > > > > > > > > > > >> affected), so I am interested in whether the
> > > failing
> > > > > tests
> > > > > > > are
> > > > > > > > > > > >> concentrated on the files recently changed by
> me
> > > (it
> > > > > would
> > > > > > > > > > > obviously not
> > > > > > > > > > > >> haved pushed them if I knew them to cause
> issues;
> > > also
> > > > > > > > > patchwork
> > > > > > > > > > > was
> > > > > > > > > > > >> fine).
> > > > > > > > > > > >> Are these issues reproducible? Can you bisect
> > > them?
> > > > > > > > > > > >>
> > > > > > > > > > > >> - Andreas
> > > > > > > > > > > >
> > > > > > > > > > > > This may be the alignment issue introduced in
> lavfi
> > > in
> > > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > >
> > > > > > >
> > > > >
> > >
> https://git.videolan.org/?p=ffmpeg.git;a=commitdiff;h=17a59a634c39b00
> > > > > > > > > > > a680c6ebbaea58db95594d13d
> > > > > > > > > > > > assuming it was not fixed.
> > > > > > > > > > > > I think it only affected targets where
> > > > > av_cpu_max_align()
> > > > > > > > > returned
> > > > > > > > > > > 64.
> > > > > > > > > > >
> > > > > > > > > > > You are completely right: Just making
> > > av_cpu_max_align
> > > > > return
> > > > > > > 64
> > > > > > > > > > > allows
> > > > > > > > > > > to reproduce the issue. And it has nothing to do
> with
> > > my
> > > > > > > recent
> > > > > > > > > FATE
> > > > > > > > > > > patches (545e87f49dc9fa5b880e330fc4e1854df68cc0f1
> > > would
> > > > > be
> > > > > > > the
> > > > > > > > > only
> > > > > > > > > > > contender for changes).
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > I found that the FATE errors can be avoided by
> using
> > > > > > > > > >
> > > > > > > > > > ./configure --disable-avx512
> > > > > > > > > >
> > > > > > > > > > Yet, I hope it will be fixed at some time..
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > > I hope you will post a fix for it soon.
> > > > > > > >
> > > > > > > > That depends on whether we want alignment values >
> 32bit..?
> > > > > > > >
> > > > > > >
> > > > > > > Nope, its about mpeg video encoder in lavc that does
> funny
> > > things
> > > > > > > from old
> > > > > > > days when linesize alignment was hardcoded.
> > > > > >
> > > > > > Adding this
> > > > > >
> > > > > >    align = FFMIN(align, 32);
> > > > > >
> > > > > > to ff_default_get_video_buffer2() resolves the issue.
> > > > > >
> > > > >
> > > > > Maybe it resolves it for you but that does not mean in any
> way
> > > that
> > > > > that is
> > > > > valid or correct solution.
> > > > >
> > > > >
> > > > > >
> > > > > > That's why I had written:
> > > > > >
> > > > > > > That depends on whether we want alignment values >
> 32bit..?
> > > > > >
> > > > >
> > > > > 32bit is 4 integers. so I failed to follow whatever you tried
> to
> > > > > communicate.
> > > >
> > > > Sorry for being unclear. What I meant to ask is this:
> > > >
> > > > In cases where av_cpu_max_align() returns something > 32 - for
> > > > example 64, do we have to use that max-align value of 64 for
> > > alignment
> > > > in video.c and framepool,c or could we just limit this to 32?
> > > > (hence the FFMIN).
> > > >
> > > > Limiting the alignment to 32 appears to fix the issue, meaning
> that
> > > > the three fate tests mentioned above are producing the expected
> ref
> > > > output and don't fail anymore.
> > > >
> > >
> > > That is hack.
> >
> > Only when 64bit alignment would be required or beneficial in terms
> of
> > performance (e.g. AVX512 operations).
> 
> 
> AVX512 operations are not using 64bit alignment.


They don't require, but I thought they would benefit from it.
You may know better than I do, I had just read this article:

https://www.intel.com/content/www/us/en/developer/articles/technical/data-alignment-to-assist-vectorization.html



> > Which is the reason for my question. If yes, then this would be
> > surely a bad idea.
> >
> 
> It is not Yes/No answer here.

Got you, thanks. I have found a very targeted workaround now.
Will submit in a moment.

sw
diff mbox series

Patch

--- ./tests/ref/vsynth/vsynth2-mpeg2-422 2022-04-30 14:23:44.330424058 +0000
+++ tests/data/fate/vsynth2-mpeg2-422 2022-04-30 14:38:41.071678201 +0000
@@ -1,4 +1,4 @@ 
-b2fa9b73c3547191ecc01b8163abd4e5 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
-379164 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
-704f6a96f93c2409219bd48b74169041 *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
-stddev: 4.17 PSNR: 35.73 MAXDIFF: 70 bytes: 7603200/ 7603200
+8f6d565723ccf879ab2b5aa910b7ce21 *tests/data/fate/vsynth2-mpeg2-422.mpeg2video
+380544 tests/data/fate/vsynth2-mpeg2-422.mpeg2video
+0797fddea4835687dedddebbbe98fa8f *tests/data/fate/vsynth2-mpeg2-422.out.rawvideo
+stddev: 4.16 PSNR: 35.73 MAXDIFF: 75 bytes: 7603200/ 7603200