git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19750 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

patch by  Lars Täuber, lars.taeuber gmx net


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19749 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Josh Harris, qt tateu net 


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19748 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Thilo Borgmann <thilo.borgmann _at_ googlemail.com>


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19747 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19746 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19745 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

architectures since stride is multiple of 4 and not of 8, so split
fill_rectangle() calls to operate on 32-bit words instead of 64-bit ones.



git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19744 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

inside it.



git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19743 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

It works the same as CHECKED_ALLOCZ except that it does not zero the allocated
memory.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19742 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19741 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19740 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19739 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19738 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

The DECLARE_ALIGNED_8 macro is defined to align to 16 bytes instead
the 8 suggested by the name on some CPUs.  None of the uses of this
macro ever need 16-byte alignment, cases which once did having been
changed to always specify 16 bytes explicitly.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19737 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

decoded.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19736 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19735 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

a packet of Speex frames as a single frame.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19734 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19733 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19732 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19731 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19730 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19729 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19728 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19727 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

1.3x and 3% on G4.
Though I think only part of this speedup is due to my optimizations per se;
some of it is that I got a better roll on the GCC random code generator.
Trivial reorderings of this function have a disproportionate effect on speed.



git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19726 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19725 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

The first member of struct ScanTable is a simple pointer, extra alignment
of which serves no purpose.  The alignment specifier was added along with
some Altivec optimisations also adding a 16-byte-aligned array at the end
of struct ScanTable.  Presumably the redundant, outer alignment was added
by mistake.  The inner one is clearly sufficient.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19724 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19723 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Christian Schmidt, schmidt digadd de 


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19722 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19721 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Lars Täuber, lars D taeuber A gmx D net 


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19720 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19719 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19718 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19717 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

These arrays are passed to functions for which we promise 16-byte
alignment, so make sure we fulfil that promise.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19716 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

None of these arrays are used in ways requiring extra alignment.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19715 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Bluray subtitle decoder instead of as small as possible.
This avoids completely unnecessary issues with e.g. libswscale.
(See r19673)

Patch by Stephen Backway, stev391 A exemail D com D au


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19714 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Fixes "dvd_subtitle too big" errors when encoding HD video subtitles
to DVD format.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19713 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19712 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

(which uses movdqa to xmm1 on inv_zigzag_direct16).


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@19711 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b