[FFmpeg-devel] [PATCH 1/6] tests/ffserver.regression.ref: Update ffserver checksums

[Michael Niedermayer]

On Fri, Dec 02, 2016 at 12:59:41AM +0100, Andreas Cadhalpun wrote:
> On 01.12.2016 17:37, Michael Niedermayer wrote:
> > They have changed due to 122190392b297b7cd9783641d880df887fd80d07
> > 
> > Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> > ---
> >  tests/ffserver.regression.ref | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/tests/ffserver.regression.ref b/tests/ffserver.regression.ref
> > index 3106f85..398c285 100644
> > --- a/tests/ffserver.regression.ref
> > +++ b/tests/ffserver.regression.ref
> > @@ -1,5 +1,5 @@
> > -233020d119085ba47535d5f2faf73cc0 *ff-test_h.avi
> > -431b75d1f12cb039acebad61a3d39225 *ff-test_l.avi
> > +0c9639f09decbc54c9f091dcf1ca0e8f *ff-test_h.avi
> > +e28ba75853caf975e06d92955c9f7f73 *ff-test_l.avi
> >  a767dbdf5d1bded3450279f812f97b37 *ff-test.swf
> >  dc16f607e13328a832e73801cd21ec98 *ff-test_h.asf
> >  69337d6c8cd7ac7e626338decdbf41d3 *ff-test_l.asf
> 
> The test is unfortunately not very reproducible.
> Right now I'm getting varying values like on one run:
> -0c9639f09decbc54c9f091dcf1ca0e8f *ff-test_h.avi
> -e28ba75853caf975e06d92955c9f7f73 *ff-test_l.avi
> -a767dbdf5d1bded3450279f812f97b37 *ff-test.swf
> +81c49d1bfcd43bcc43aa6aeed081126f *ff-test_h.avi
> +2d642fbb77276ed1d12bf235a6a6bd57 *ff-test_l.avi
> +daeee6c1418c5b2dcddded9b659e9fc2 *ff-test.swf
> 
> And on the next:
> -0c9639f09decbc54c9f091dcf1ca0e8f *ff-test_h.avi
> -e28ba75853caf975e06d92955c9f7f73 *ff-test_l.avi
> +5e7fc3d0b2f4866866ae4f268dc431ca *ff-test_h.avi
> +6cc80ab0caec9499d1b82cd01fe7a861 *ff-test_l.avi
> 
> I've no idea what the underlying problem is.

it works here almost always i remember just one failure

but you could try to use diff the ffprobe output showing as much as
ffprobe supports to see what differs.
binary compare (if they match in size) or ffmpeg -c copy -f framecrc -

also please upload some different files somewhere

it also could be some timeout issue if you have a slow box and maybe
run this under valgrind

otherwise printing checksums on packets at various points might
help (that is if previous stuff points at that changing)

...

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Observe your enemies, for they first find out your faults. -- Antisthenes
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20161202/da9e0874/attachment.sig>