Most EBML masters are much smaller than IO_BUFFER_SIZE and thus the size
can be updated. This makes parsing the resulting files easier.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22197 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

This results in a 50% speedup on main profile with no increase in binary size.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22196 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22195 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Thanks to backwards compatible HE-AAC signalling these values are unreliable.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22194 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

1 cpu cycle faster


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22193 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22192 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

consistent with the allocation currently done for PIX_FMT_GRAY8
pictures.

No significant slow-downs have been measured.

See the thread:
Subject: [FFmpeg-devel] [PATCH] Is gray8 a paletted format?
Date: Sun, 15 Nov 2009 23:36:03 +0100

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22191 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

This is generally around 12% faster than the prior method of creating a
linked list for each block as tokens are read, but can be anywhere from
8% to 28% faster depending on file and CPU.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22190 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

This increases the slice size to 64 pixels, due to having to decode an
entire chroma superblock row per slice.

This can be up to 6% slower depending on clip and CPU, but is necessary
for future optimizations that gain significantly more than was lost.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22189 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22188 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Anton Khirnov < whyskas at gmail >

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22187 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22186 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22185 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

original patch by elupus _at_ ecce _dot_ se


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22184 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

5 cpu cycles faster.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22183 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

This fixes some valgrind warnings.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22182 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

to the end of the structure.
4 cpu cycles faster in 3k cpu cycles


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22181 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22180 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22179 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22178 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22177 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22176 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

a few months now). Fixes issue 1738.



git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22175 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22174 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Patch by Jai Menon.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22173 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22172 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22171 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Merge weight & offset tables.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22170 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22169 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22168 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22167 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22166 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22165 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

NetBSD is unlikely to fix their headers and the FATE box passes it
as -D_NETBSD_SOURCE on the command line anyway. In this case, it's
better to keep the hack well-contained within this file.
Closes issue 886.
patch by Jorge Acereda, jacereda brainstorm es


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22164 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22162 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

30 cpu cycles faster


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22161 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

Based on a patch by NVIDIA.


git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22160 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22159 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22158 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b

This test generates many output files, and keeping them separate
is convenient.

git-svn-id: file:///var/local/repositories/ffmpeg/trunk@22157 9553f0bf-9b14-0410-a0b8-cfaf0461ba5b