[MPlayer-dev-eng] liba52 reorganization

[Michael Niedermayer]

Hi

On Fri, Sep 22, 2006 at 01:52:30PM -0700, Alex Izvorski wrote:
> Hi Michael,
> 
> On Fri, 2006-09-22 at 20:35 +0200, Michael Niedermayer wrote:
> > well it depends on what you want, if you want to get the changes into 
> > main liba52 then you should disscuss this with the liba52 developers not
> > with us mplayer developers, as we dont have write access to the upstream
> > liba52 and cant help you with getting patches accepted there
> > 
> > if OTOH the liba52 developer(s) reject your changes and you want us to
> > accept them in mplayer svn then you of course  will have to provide us 
> > with patches against what we currently have in svn, that could be
> > first a diff which updates our stuff to upstream liba52 and then several
> > which cleanup our optimizations
> 
> I'm shooting for both, ideally ;) but I'm not too picky as to the order
> in which those happen.  Yes, I have been discussing these with liba52
> developers for some time.
> 
> Looks like mplayer liba52 is already updated to the latest upstream
> (0.7.4).  This is relatively recent.  Big thank you! to whoever did
> that.  So nothing to do there.
> 
> The diffs between the major new files and the original files the code
> came from look like this (count of = same  + added  - removed lines,
> mplayer svn versus patched a52dec-0.7.4):
> 
> mplayer/liba52/downmix.c a52dec-0.7.4/liba52/downmix_sse.c          
>   = 328  +  32  -1494 lines
> mplayer/liba52/downmix.c a52dec-0.7.4/liba52/downmix_3dnow.c      
>   = 435  +  17  -1387 lines
> mplayer/liba52/imdct.c a52dec-0.7.4/liba52/imdct_sse.c              
>   = 475  +  19  - 837 lines
> mplayer/liba52/imdct.c a52dec-0.7.4/liba52/imdct_altivec.c          
>   = 368  +   9  - 944 lines
> mplayer/liba52/imdct_3dnow.h a52dec-0.7.4/liba52/imdct_3dnow.c      
>   = 541  +  91  -  25 lines
> 
> There are a few other odds and ends.  These are as you can see just
> copies of a chunk of code out of one file into a new file, the added
> stuff is headers, and the "removed" stuff is simply the code that didn't
> get copied.  I can send you a whole bunch of diffs for each of those if
> you want but if you apply the big patch to an upstream a52dec it is easy
> to diff them locally.   Let me know if you want those.

well if you want a patch accepted you have to post it here

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

In the past you could go to a library and read, borrow or copy any book
Today you'd get arrested for mere telling someone where the library is