[FFmpeg-devel] releases and stuff

[Sascha Sommer]

Hi,

On Dienstag, 10. Februar 2009, Diego Biurrun wrote:
> On Tue, Feb 10, 2009 at 04:33:48PM +0100, Patrik Kullman wrote:
> > On Tue, 2009-02-10 at 16:18 +0100, Diego Biurrun wrote:
> > > On Tue, Feb 10, 2009 at 04:08:36PM +0100, Michael Niedermayer wrote:
> > > > On Tue, Feb 10, 2009 at 03:59:08PM +0100, Diego Biurrun wrote:
> > > > > On Tue, Feb 10, 2009 at 03:56:47PM +0100, Diego Biurrun wrote:
> > > > > > On Tue, Feb 10, 2009 at 01:17:07PM +0100, Dominik 'Rathann' 
Mierzejewski wrote:
> > > > > > > On Tuesday, 10 February 2009 at 00:32, Carl Eugen Hoyos wrote:
> > > > > > > > Roman V. Shaposhnik <rvs <at> sun.com> writes:
> > > > > > > > > -4435d87463cd6c5407bd88cca241ca56 *./tests/data/a-wmav1.asf
> > > > > > > > > +6f7f3116b801ea641ce14f827de05cce *./tests/data/a-wmav1.asf
> > > > > > > >
> > > > > > > > I remember posting an incredibly beautiful patch to fix it,
> > > > > > > > fortunately I can't find it now;-)
> > > > > > >
> > > > > > > This?
> > > > > > >
> > > > > > > Date: Sat, 22 Nov 2008 13:42:11 +0100
> > > > > > > Subject: Re: [FFmpeg-devel] [PATCH] correct make test failure
> > > > > > > from 15261 releaseuntil now (15899)
> > > > > >
> > > > > > Here is the patch from Carl Eugen, it does indeed fix the WMA
> > > > > > regression tests on PPC.  Is this acceptable?  Acceptable minus
> > > > > > the commented-out code I mean.
> > > > >
> > > > > *sigh*, here is the patch..
> > > >
> > > > if it works on x86 32&64 & ppc & all gcc versions we care about & icc
> > > > and you add a comment to the code saying why it exists
> > >
> > > I tested on ppc with gcc 4.3 and x86_32 with gcc 4.1.  What other gcc
> > > versions do you want?  Can somebody test with x86_64?
> >
> > ---
> > /home/patrik/Programming/ffmpeg-svn/tests/ffmpeg.regression.ref	2009-02-0
> >7 08:37:45.000000000 +0100 +++ tests/data/vsynth.regression	2009-02-10
> > 16:30:26.000000000 +0100 @@ -205,7 +205,7 @@
> >  stddev:    0.00 PSNR:999.99 bytes:  1058444/  1058444
> > -4435d87463cd6c5407bd88cca241ca56 *./tests/data/a-wmav1.asf
> > +6f7f3116b801ea641ce14f827de05cce *./tests/data/a-wmav1.asf
>

Hmm



>--target=x86_64-linux-gnu

original patch:

>+#if defined(__ICC) || defined (__APPLE__) || ARCH_PPC

new patch:

>+#if (defined(__ICC) || defined (__APPLE__) || ARCH_PPC) && !ARCH_X86_64


I doubt the new patch is better.

Wouldn't the attached patch work for everyone?

It does not change the md5sums for me on i386 and gcc 4.2.1

Besides that, the calculation should be close to what it originally was in 
wma.c.
Performance differences shouldn't matter to much as the whole function is 
declared with av_cold and only called during decoder init afaik.

Regards

Sascha



-------------- next part --------------
A non-text attachment was scrubbed...
Name: wma_regtest.patch
Type: text/x-diff
Size: 464 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090210/5b980cdc/attachment.patch>