Commit Graph

246 Commits

Author SHA1 Message Date
Måns Rullgård
55fde95e3b rename cropTbl -> ff_cropTbl
Originally committed as revision 6992 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-11-12 20:08:09 +00:00
Panagiotis Issaris
c2212338e8 Remove the H264Context parameter from decode_init_vlc() as it is not being used.
Originally committed as revision 6955 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-11-09 16:40:33 +00:00
Michael Niedermayer
65342b2b4a merge clip() into the alpha/beta/tc0 tables (10% faster filter_mb_fast() on P3)
Originally committed as revision 6863 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-11-01 19:34:30 +00:00
Reimar Döffinger
d55f46e5a8 Reenable AMD64 optimizations for cabac accidentially disabled in r6852
Originally committed as revision 6853 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-11-01 17:18:49 +00:00
Diego Biurrun
419b878494 Add ARCH_X86_32 as a new define for 32 bit x86 architectures and change
the semantics of ARCH_X86 to mean both 32 and 64 bits.

Originally committed as revision 6852 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-11-01 17:01:33 +00:00
Diego Biurrun
d5cd50ed73 Fix compilation with PIC enabled, BRANCHLESS_GET_CABAC is defined under
!PIC but gets used without a check for !PIC.

Originally committed as revision 6834 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-29 13:14:37 +00:00
Michael Niedermayer
e08f580644 decode_significance_8x8_x86()
8% faster decode_cabac_residual() (8x8 case only) on P3

Originally committed as revision 6750 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-20 21:34:02 +00:00
Michael Niedermayer
f624f029f0 slightly faster decode_cabac_mb_mvd()
Originally committed as revision 6735 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-19 14:54:34 +00:00
Michael Niedermayer
aa1da263aa 1 variable less in decode_cabac_residual() no speed change (but less vars means less things gcc can load and store redundantly)
Originally committed as revision 6734 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-19 10:33:56 +00:00
Michael Niedermayer
ebd624b662 optimize sign decoding code in decode_residual()
x86 is 4% faster on P3
C sign stuff + x86 code for everything else is also faster then before (sorry forgot to test pure C)
... and if i replace the second occurance of the sign decoding in decode_residual by the asm too then everything gets slower iam starting to think that it might be best to write the whole function in asm, playing this avoid random deoptimizations game with gcc is not fun at all

Originally committed as revision 6732 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-19 01:19:03 +00:00
Michael Niedermayer
30865e2aa1 copy cabac context onto the stack for the c code in decode_cabac_residual() (15% faster on P3 but still slower then the current asm)
Originally committed as revision 6725 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-17 23:59:21 +00:00
Michael Niedermayer
eb73bf723d x86 asm version of the decode significance loop (not 8x8) of decode_residual() 5% faster decode_residual() on P3
Originally committed as revision 6724 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-17 22:18:29 +00:00
Michael Niedermayer
b566bd65e2 simplify escape decoding
Originally committed as revision 6685 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-13 16:19:15 +00:00
Michael Niedermayer
de1013e1ca 10l int vs uint8_t
Originally committed as revision 6675 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-12 18:41:29 +00:00
Michael Niedermayer
851ded8918 prevent "mb level" get_cabac() calls from being inlined (3% faster decode_mb_cabac() on P3)
Originally committed as revision 6674 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-12 14:49:19 +00:00
Michael Niedermayer
1f6049f515 int->uint8_t for 2 tables
Originally committed as revision 6673 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-12 12:27:02 +00:00
Diego Biurrun
c26abfa541 Rename ABS macro to FFABS.
Originally committed as revision 6666 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-11 23:17:58 +00:00
Loren Merritt
a957c27b03 more prefetches, 2% faster h264
Originally committed as revision 6664 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-11 21:34:15 +00:00
Michael Niedermayer
d61c4e731e make state transition tables global as they are constant and the code is slightly faster that way
Originally committed as revision 6655 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-11 14:44:17 +00:00
Michael Niedermayer
0fa352c7e6 make lps_range a global table its constant anyway (saves 1 addition for accessing it)
Originally committed as revision 6653 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-11 13:21:42 +00:00
Michael Niedermayer
7822e1c1ff reverse remainder of the failed attempt to optimize *state=c->mps_state[s]
Originally committed as revision 6609 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-09 21:14:16 +00:00
Michael Niedermayer
1c2a417f6a move outcommented START/STOP_TIMER to a hopefully better place for benchmarking ...
Originally committed as revision 6605 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-09 18:20:00 +00:00
Michael Niedermayer
99ce10873d 5% faster get_cabac()
Originally committed as revision 6586 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-08 11:24:37 +00:00
Diego Biurrun
e16c407a81 output typo fixes
Originally committed as revision 6585 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-08 10:11:58 +00:00
Michael Niedermayer
5659b509c7 refill cabac variables in 16bit steps, 3% faster get_cabac()
Originally committed as revision 6578 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-07 15:44:14 +00:00
Diego Biurrun
b78e7197a8 Change license headers to say 'FFmpeg' instead of 'this program/this library'
and fix GPL/LGPL version mismatches.

Originally committed as revision 6577 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-07 15:30:46 +00:00
Loren Merritt
e6e77eb6fa simplify write_back_motion for skip mbs
Originally committed as revision 6548 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-04 07:08:26 +00:00
Loren Merritt
2833fc4646 approximate qpel functions: sacrifice some quality for some decoding speed. enabled on B-frames with -lavdopts fast.
Originally committed as revision 6412 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-10-01 21:25:17 +00:00
Måns Rullgård
191e8ca752 fix some signedness warnings
Originally committed as revision 6355 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-09-27 19:47:39 +00:00
Loren Merritt
776fe466b2 slightly faster deblock
Originally committed as revision 6121 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-08-29 03:07:58 +00:00
Loren Merritt
3e20143ee7 mmx implementation of deblocking strength decision.
2-3% faster h264.

Originally committed as revision 6113 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-08-28 09:33:01 +00:00
Loren Merritt
ae08a5631e 10l in r5768 (broke mbaff)
Originally committed as revision 5831 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-07-27 20:37:21 +00:00
Loren Merritt
717b173366 slightly faster h264 fill_caches()
Originally committed as revision 5768 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-07-16 18:55:18 +00:00
Måns Rullgård
af9e7d1810 --enable/disable parsers. Warning: some combinations are broken.
Originally committed as revision 5691 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-07-09 12:02:15 +00:00
Michael Niedermayer
784036e404 simplify intra prediction mode decoding
Originally committed as revision 5612 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-07-04 17:06:43 +00:00
Måns Rullgård
91c58c944f fill_mbaff_ref_list is missing a return statement, its return value
is ignored, and it can't possibly fail, so make it void

Originally committed as revision 5566 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-07-01 17:33:48 +00:00
Loren Merritt
c4da83fb0f h264: accept rbsp_trailing longer than 1 byte
Originally committed as revision 5467 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-06-12 09:30:17 +00:00
Michael Niedermayer
b3aaa7bea6 typo me thinks ...
Originally committed as revision 5421 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-29 16:39:40 +00:00
Loren Merritt
5d18eaad13 h264: MBAFF interlaced decoding
Originally committed as revision 5419 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-28 22:44:14 +00:00
Loren Merritt
67a8208695 h264: faster fill_rectangle()
Originally committed as revision 5418 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-28 22:28:08 +00:00
Michael Niedermayer
2d661e073e cosmetics (initalize variables where they are used)
Originally committed as revision 5356 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-10 11:05:03 +00:00
Michael Niedermayer
11b681687d decoders should (IMHO) not duplicate frames, the application can do this if it wants
Originally committed as revision 5355 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-10 10:57:27 +00:00
Loren Merritt
09786209e6 reinit quant matrices if pps is overwritten by another of the same pps_id.
Originally committed as revision 5346 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-05-05 18:08:01 +00:00
Loren Merritt
89ddb72aa1 warn on interlaced streams
Originally committed as revision 5337 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-04-30 19:31:45 +00:00
Loren Merritt
2848ce84d2 don't force asserts in release builds. 2% faster h264.
Originally committed as revision 5332 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-04-29 00:43:15 +00:00
Loren Merritt
2ea392529a 10l in 1.199 (write_back_motion with B-pyramid + spatial)
Originally committed as revision 5215 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-03-25 09:46:43 +00:00
Loren Merritt
513fbd8e5a prefetch pixels for future motion compensation. 2-5% faster h264.
Originally committed as revision 5203 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-03-23 20:16:36 +00:00
Loren Merritt
53b19144e7 1.5x faster write_back_motion, 1-3% overall
Originally committed as revision 5198 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-03-22 12:41:02 +00:00
Michael Niedermayer
4cb8574dc8 fixing coverity warnig CID: 253 (reading over the end of the aspect_ratio array for illegal aspect ratios)
Originally committed as revision 5163 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-03-13 23:46:47 +00:00
David S. Miller
bf4665eeec Alignment fixes, caused problems on SPARC.
patch by "David S. Miller" davem <<at>> davemloft <<dot>> net

Originally committed as revision 5130 to svn://svn.ffmpeg.org/ffmpeg/trunk
2006-03-09 08:05:45 +00:00