[FFmpeg-devel] [PATCH] lavfi test for 1-1 filters pixel format output

[Stefano Sabatini]

On date Monday 2010-06-14 18:50:30 +0200, Michael Niedermayer encoded:
> On Mon, Jun 14, 2010 at 02:04:49PM +0100, M?ns Rullg?rd wrote:
> > Stefano Sabatini <stefano.sabatini-lala at poste.it> writes:
> > 
> > > On date Monday 2010-06-14 12:28:51 +0100, M?ns Rullg?rd encoded:
> > >> Stefano Sabatini <stefano.sabatini-lala at poste.it> writes:
> > >> 
> > >> > On date Sunday 2010-06-06 01:30:16 +0200, Stefano Sabatini encoded:
> > >> >> On date Wednesday 2010-06-02 23:42:32 +0200, Stefano Sabatini encoded:
> > >> >> > On date Monday 2010-05-24 22:12:25 +0100, M?ns Rullg?rd encoded:
> > >> >> > > Stefano Sabatini <stefano.sabatini-lala at poste.it> writes:
> > >> >> > [...]
> > >> >> > > >> Did I review this?  Of course not, or I would have told you it breaks
> > >> >> > > >> cross-builds (only if the test is run of course).
> > >> >> > > >
> > >> >> > > > Michael asked to commit, I did since the test is disabled so people
> > >> >> > > > can start to work out the lavfi bugs.
> > >> >> > > 
> > >> >> > > He also asked me to review it.
> > >> >> > > 
> > >> >> > > > Feel free to fix it yourself or point out the problem it has.
> > >> >> > > 
> > >> >> > > I'll get to it eventually, but the problem is that the executable you
> > >> >> > > build has to be run with $TARGET_EXEC and $TARGET_PATH.
> > >> >> > 
> > >> >> > Uhm, if that is true why tiny_psnr in regression-funcs.sh is not
> > >> >> > defined like that.
> > >> >> > 
> > >> >> > Anyway my main problem seems that the lavfi-showfiltfmts I'm using
> > >> >> > needs to include and link against the libav* libraries, in this being
> > >> >> > different from the other testprogs, so it is failing to compile and I
> > >> >> > don't even know if this crossbuilding thing is even able to support
> > >> >> > that requirement.
> > >> >> > 
> > >> >> > Your help is welcome, regards.
> > >> >> 
> > >> >> M?ns-ping.
> > >> >
> > >> > I'll apply this patch or a variant in three days if I get no reply.
> > >> 
> > >> You will do no such thing.  The patch is WRONG.
> > >
> > > I know that,
> > 
> > Then this discussion is pointless.
> 
> So how exactly will this be resolved?
> Can you elaborate on what is wrong?
> Can you document how the cross compilation system you designed works
> so other people than you can modify the scripts without breaking it?
> Do we have some volunteer who is less busy and who could help maintain
> the scripts in ffmpeg?

Patch updated, don't know if this is the correct solution, anyway it
works for me. It still misses the BE tests, I don't have such a
machine so I need either someone to try them either an an account on
such a machine.

I'm still waiting for a useful reply from Mans, what's sure is that
I'm not going to wait forever (note that this thread started in
March).

Regards.
-- 
FFmpeg = Frightening & Fancy Majestic Powered Exxagerate Game