On Fri, Oct 05, 2012 at 08:45:48AM -0700, Ronald S. Bultje wrote:
> On Fri, Oct 5, 2012 at 3:50 AM, Diego Biurrun <[email protected]> wrote:
> > ---
> > The test program is of doubtful utility, fails to compile on x86_32 and
> > was never part of our test suite.
> >  libavcodec/Makefile |    3 +-
> >  libavcodec/cabac.c  |  147 
> > ---------------------------------------------------
> >  2 files changed, 1 insertions(+), 149 deletions(-)
> 
> This is probably the 20th time you're submitting this patch. Can you
> please not do that? Are you just hoping someone isn't paying attention
> and accidently says yes?

It's been 10 months and nobody fixed it, converted it into a FATE test
or even touched it.  Now it does not even compile anymore.   I sent a
mail mentioning that and I asked around on IRC what to do with
cabac-test, reactions ranged from "don't care" to "remove".

> Why don't you fix the bug that prevents it from working.

Because I don't believe it's worth the trouble.  The fact that these
test programs are in a continuous state of disrepair and only I notice
when they break after months or years suggests that nobody uses them
or cares.

But if you disagree with that asessment, why don't you fix it?  You know
our H.264 code, I don't.  I'm already working on all the surrounding glue
that will keep the test programs working into the future.

I've hacked up the FATE scripts to set up an x86_64 build-only FATE slot
that builds all the test/etc. programs.  After noticing that cabac-test
was broken on x86_32 I set up a similar slot for that.  Today I have
implemented build-only FATE tests.  Once that is in all test programs
will be kept working on all platforms that run FATE.

So you can hardly say that I do too little; I'm the only person working
on or caring about these programs at all ...

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

Reply via email to