Commit Graph

23189 Commits

Author SHA1 Message Date
Michael Niedermayer cf41a02b1b Only load the topleft mv/ref when the topright is unavailable.
8 cpu cycles faster.

Originally committed as revision 22079 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 15:26:11 +00:00
Michael Niedermayer cf7b67bc40 Remove some useless operations from the code setting left_cbp.
maybe 0.5 cpu cycles faster

Originally committed as revision 22078 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 15:03:00 +00:00
Francesco Cosoleto 6abda15f56 Prevent overflow of start_time + recording_time.
Patch by Francesco Cosoleto gmail($name)

Originally committed as revision 22077 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 14:46:20 +00:00
Peter Ross be707647e9 Add myself as maintainer of the bink demuxer and bink audio decoder
Originally committed as revision 22075 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 11:51:17 +00:00
Michael Niedermayer 59b5370f02 Simplify code to set cbp_*
this seems 1 cpu cycle slower even though we practically just remove code.
Speed loss seems caused by the merge of if(left_type), iam commiting this
anyway as i cant imagine this to be anything but compiler messup.

Originally committed as revision 22073 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 09:13:40 +00:00
Benoit Fouet 5818418758 Fix typos in ref_index documentation.
Originally committed as revision 22072 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 08:55:43 +00:00
Michael Niedermayer 747db4e31a Move init of right side of ref_cache from fill_caches() to init_the_darn_decoder().
Originally committed as revision 22071 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 03:27:52 +00:00
Michael Niedermayer 77c6edb846 Remove 3 mv_cache zeroing instructions that zeroed the right side.
This seems unneeded as nothing seems to ever set it to non zero values.

Originally committed as revision 22070 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 02:54:03 +00:00
Michael Niedermayer 8f8497ae78 Remove useless check of the 2 left MBs of a pair being in the same slice.
Originally committed as revision 22069 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 01:38:12 +00:00
Michael Niedermayer f57a5dcd4e Clarify ref_index.
Originally committed as revision 22068 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 00:51:26 +00:00
Michael Niedermayer 6e2fe0f20a Remove unneeded line of code from the neighbor setting code in h264.
Originally committed as revision 22067 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 00:10:35 +00:00
Michael Niedermayer aa5ee59eee Simplify code in mv_pred.
Not benchmarked as this is petty much just code removial.

Originally committed as revision 22066 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-26 00:08:27 +00:00
Michael Niedermayer 358b5b1a59 Get rid of mb2b8_xy and b8_stride, change arrays organized based on b8_stride to
ones based on mb_stride in h264.
about 20 cpu cycles faster overall per MB

Originally committed as revision 22065 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 23:44:42 +00:00
Måns Rullgård f3a572ca14 Disable suffix rules
Most of the make builtin rules, which we do not need, are suffix rules,
and we use only new-style pattern rules.  Disabling suffix rules saves
some time when building on slow systems.

Originally committed as revision 22064 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 23:27:54 +00:00
Stefano Sabatini 479e1e6170 Remove pointless empty line.
Originally committed as revision 30730 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2010-02-25 21:21:29 +00:00
Stefano Sabatini 8ebed38e7a Apply consistency nit.
Originally committed as revision 30729 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
2010-02-25 21:12:38 +00:00
Stefano Sabatini 11ab237e31 Add the graph2dot tools and document it.
Also link avfiltergraph.o and graphparser.o against libavfilter, as it
uses them.

Originally committed as revision 22063 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 19:55:01 +00:00
Stefano Sabatini 9c16add67b Replace not anymore valid syntax "-title X" with "-metadata title=X".
Originally committed as revision 22062 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 19:38:54 +00:00
Michael Niedermayer 3b606e71c4 svq3 now in working condition, at least vissually, ill let fate tell us
if the checksums match

Originally committed as revision 22061 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 18:41:48 +00:00
Michael Niedermayer fc7f7f171f fix compilation, sorry ive not checked cvslog for a while :(((
svq3 decoder does not work yet though but i didnt want to keep compilation
broken longer

Originally committed as revision 22060 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 18:39:01 +00:00
Anton Khirnov cba2f6d51b asfenc: write tags in proper UTF-16.
Patch by Anton Khirnov wyskas gmail com

Originally committed as revision 22059 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 16:59:55 +00:00
Anton Khirnov a677078e14 asfenc: simplify writing of comment header.
Patch by Anton Khirnov wyskas gmail com

Originally committed as revision 22058 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 16:50:26 +00:00
Anton Khirnov e4ea625ab0 asfdec: don't strip the "WM/" prefix, this should be done during conversion.
Patch by Anton Khirnov wyskas gmail com

Originally committed as revision 22057 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 16:38:23 +00:00
Michael Niedermayer 36b54927fd Dont allocate direct_table 8 times too large.
Originally committed as revision 22056 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 15:29:12 +00:00
Michael Niedermayer 5e350863cc Store data in direct_table interleaved.
seems 20cpu cycles faster

Originally committed as revision 22055 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 15:27:55 +00:00
Michael Niedermayer 013202d720 Simplify intra4x4_pred_mode_cache init.
Originally committed as revision 22054 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 14:54:31 +00:00
Michael Niedermayer 662a5b2370 Reorder intra4x4_pred_mode so that we can read/write 4 values at once.
3-7 cpu cycles faster

Originally committed as revision 22053 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 14:26:12 +00:00
Michael Niedermayer 5b0fb5244d Store intra4x4_pred_mode per row only.
about 5 cpu cycles slower in the local code but should be overall faster
due to reduced cache use. (my sample though has too few intra4x4 blocks
for this to be meassureable easily either way)

Originally committed as revision 22052 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 14:02:39 +00:00
Michael Niedermayer c2186cbddc unroll tiny and trivial loop. Same speed but clearer.
Originally committed as revision 22051 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 12:51:32 +00:00
Benoit Fouet a5517dcc9a asf: indent.
Patch from Anton Khirnov wyskas gmail com

Originally committed as revision 22050 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 09:10:37 +00:00
Benoit Fouet 9b54f4328e asf: add more entries to metadata conv table.
Patch from Anton Khirnov wyskas gmail com

Originally committed as revision 22049 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 09:09:39 +00:00
Michael Niedermayer e1c88a2138 Cut the size of mvd_table by yet another factor of 2.
The code read/write code itself was 1 cycle faster, overall its
likely more due to cache effects

Originally committed as revision 22048 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 04:11:33 +00:00
Michael Niedermayer d43c192236 Keep mvd_table values of only 2 mb rows.
Originally committed as revision 22047 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 02:42:25 +00:00
Stefano Sabatini b6a4185367 10l: add prefix "TAG:" to the metadata tags key showed for each stream.
This is consistent with the metadata displaying in show_format() and
with the documentation.

Originally committed as revision 22046 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 00:16:43 +00:00
Stefano Sabatini 912dd63ecb Cosmetics: replace "filename" to "arg" for the name of the argument of
opt_input_file(). More consistent and more clear, as "filename" can be
easily confused with the global "input_filename".

Originally committed as revision 22045 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-25 00:05:01 +00:00
Alex Converse c4a90caae2 aac: Keep decode_band_types() from eating all padding at the end of a buffer.
Due to a shortcoming in the AAC specification, if an all zero buffer is
fed to section data decoding it will never terminate. That means without
a buffer exhaustion check decode_band_types() will consume all input
buffer padding. Worse if a get_bits() implementation that returns zeros
when padding is exhausted is used, the function will never terminate.

The fixes that by added a buffer exhaustion check in the sectioning
decoding loop.

Originally committed as revision 22044 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 23:56:52 +00:00
Stefano Sabatini 3314799368 Extend doxy for the src parameter of av_fifo_generic_write().
@patchby Tomas Härdin |tomas dot hardin at codemill dot se|

Originally committed as revision 22043 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 23:52:52 +00:00
Michael Niedermayer 09ef1ace4b 3rd and hopefully last 100l fix.
Originally committed as revision 22041 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 22:13:20 +00:00
Michael Niedermayer 3393bff10d Fix doxy and assert().
Originally committed as revision 22040 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 22:05:51 +00:00
Michael Niedermayer 377fc38d5d Try to fix 100l compilation failure on some systems.
Originally committed as revision 22039 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 21:55:55 +00:00
Michael Niedermayer b5bd070029 Change mvd_cache & mvd_table to 8bit, this is overall a bit faster
for high resolution videos.
about 20cycles faster per MB for cathederal.

Originally committed as revision 22038 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 20:43:06 +00:00
Michael Niedermayer 8ef4e65e94 AV_COPY16() & AV_ZERO16()
Originally committed as revision 22037 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 20:38:31 +00:00
Michael Niedermayer f3ea29bfa3 Extend fill_rectangle() support for 16bit
Originally committed as revision 22036 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 20:37:58 +00:00
Michael Niedermayer 81b5e4ee92 Calculate mvd without abs()
same speed (ask gcc why, i dont know)

Originally committed as revision 22035 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:50:02 +00:00
Anton Khirnov fce4448948 Read ASF metadata as proper UTF-16 and spit it out as proper UTF-8 in our
metadata system.

Patch by Anton Khirnov <wyskas gmail com>.

Originally committed as revision 22034 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:19:54 +00:00
Anton Khirnov a106d4112d Don't add WM prefixes to all written ASF tags.
Patch by Anton Khirnov <wyskas gmail com>.

Originally committed as revision 22033 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:17:32 +00:00
Michael Niedermayer 855a1ba5e8 switch back to (amvd>2)+(amvd>32), its 5 cpu cycles faster now.
Originally committed as revision 22032 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:16:48 +00:00
Anton Khirnov 531d8fa308 Eliminate put_str16().
Patch by Anton Khirnov <wyskas gmail com>.

Originally committed as revision 22031 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:09:49 +00:00
Anton Khirnov ae529ddb32 Add PUT_UTF16() macro.
Patch by Anton Khirnov <wyskas gmail com>.

Originally committed as revision 22030 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:08:30 +00:00
Michael Niedermayer 01b35be14a Factorize common code from the top of decode_cabac_mb_mvd()
10-15 cpu cycles faster.

Originally committed as revision 22029 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-02-24 18:06:02 +00:00