[FFmpeg-devel] [PATCH] avcodec/svq1: zero initialize entries array

[James Almer]

On 4/9/2017 10:46 PM, Michael Niedermayer wrote:
> On Sun, Apr 09, 2017 at 07:51:10AM -0300, James Almer wrote:
>> Fixes valgrind warnings about "Use of uninitialised value of size 8"
> 
> how can this be reproduced ?

fate-svq1, fate-svq1-headerswap and fate-vsynth{1,2,3,_lena}-svq1 when
configured with --toolchain=valgrind-memcheck

A similar failure can be seen in all ffv1 vsynth tests as well, but i
couldn't find what caused them or where.

> is this a regression ?

Fate didn't always complain about this, so it's either something
introduced by a change in our tree, or a valgrind bug introduced in a
relatively recent version.

The reports in
http://fate.ffmpeg.org/history.cgi?slot=x86_64-archlinux-gcc-valgrindundef
are kinda broken and report a nonsense commit as the "last known good
ref", so i can't say when it started failing.

> 
> 
>>
>> Signed-off-by: James Almer <jamrial at gmail.com>
>> ---
> 
>> This probably just silences a bunch of false possitives.
> 
> if thats the case, then the fix is wrong
> valgrind bugs should be fixed in valgrind or the entries should be
> added to the valgrind suppression file

Assuming it is after all a false positive, zero initializing stack is
harmless and gets rid of the noise. But i agree it's not ideal.

> 
> [...]
> 
> 
> 
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>