Sean McGovern <[email protected]> writes:

> On 12/28/11, Michael Kostylev <[email protected]> wrote:
>>
>> On Wed Dec 28 11:10:55 2011
>> Sean McGovern wrote:
>>
>>>> Sometimes the vsynth1-dnxhd_720p_rd test fails on alpha, pa-risc
>>>> and sparc FATE machines. It explicitly uses '-threads 4' when
>>>> encoding, and for some reason vsynth1/dnxhd-720p-rd.dnxhd may be
>>>> spoiled (the first frame contains a small number of randomly placed
>>>> zero blocks). For comparison, the single-thread mode does not show
>>>> any issues.
>>> This sounds threading related. Feel like trying valgrind's Helgrind and/or
>>> DRD tools to examine this further?
>>
>> Thread debugging is another story. I meant that the hard-coded '-threads 4'
>> option looks rather harmful,

It is exposing a bug somewhere.  That does not make it harmful in itself.

>> it was added when fate did not provide any control over multithreaded
>> testing.
>
> I don't suppose the commonality between all the targets this is
> failing on is 'strict alignment', is it? I know SPARC is at least.

We have other targets with strict alignment requirement which do not
show this intermittent failure, e.g. MIPS.  An alignment error would
also result in a SIGBUS, not corrupt output.

-- 
Måns Rullgård
[email protected]
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to