summaryrefslogtreecommitdiff
path: root/libavcodec/i386
Commit message (Collapse)AuthorAgeFilesLines
* get_pixels_sse2, ~+12% performance compared to mmxBaptiste Coudurier2008-10-091-0/+35
| | | | Originally committed as revision 15591 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Add a bunch of const qualifiers to function arguments.Diego Biurrun2008-10-071-12/+17
| | | | | | Taken from the upstream libmpeg2 version of the file. Originally committed as revision 15585 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Sync cosmetics from upstream.Diego Biurrun2008-10-071-0/+3
| | | | Originally committed as revision 15576 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Declare ff_pw_53 and ff_pw_18 as assembler constants.Diego Pettenò2008-10-041-2/+2
| | | | | | Patch by Diego 'Flameeyes' Petten Originally committed as revision 15553 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove duplication of TRANSPOSE8 macro, also removing some redundant load/storesDavid Conrad2008-09-181-58/+2
| | | | Originally committed as revision 15355 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Make a separate macro for writing an entire block from registers to memoryDavid Conrad2008-09-181-9/+12
| | | | Originally committed as revision 15354 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Factorize SSE2_(Row|Column)_IDCT into one macroDavid Conrad2008-09-181-116/+19
| | | | Originally committed as revision 15353 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Cosmetics: remove excessive blank linesDavid Conrad2008-09-171-143/+0
| | | | Originally committed as revision 15351 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Ensure MMX/SSE2 VP3 IDCT selection isn't disabled when only Theora is enabledDavid Conrad2008-09-171-1/+1
| | | | Originally committed as revision 15350 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Cosmetics: reindentDavid Conrad2008-09-172-9/+9
| | | | Originally committed as revision 15349 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Rewrite SSE2 VP3 IDCT in inline asmDavid Conrad2008-09-171-283/+273
| | | | Originally committed as revision 15347 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Rewrite MMX VP3 IDCT in inline asmDavid Conrad2008-09-171-202/+185
| | | | Originally committed as revision 15346 to svn://svn.ffmpeg.org/ffmpeg/trunk
* MMX/SSE2 VP3 IDCT are bitexact now that the dequantization matrices are ↵David Conrad2008-09-171-3/+1
| | | | | | permutated correctly Originally committed as revision 15345 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Globally rename the header inclusion guard names.Stefano Sabatini2008-08-317-21/+21
| | | | | | | | | Consistently apply this rule: the guard name is obtained from the filename by stripping the leading "lib", converting '/' and '.' to '_' and uppercasing the resulting name. Guard names in the root directory have to be prefixed by "FFMPEG_". Originally committed as revision 15120 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use ff_vp3_idct_data in vp3dsp_mmx.c rather than duplicating itDavid Conrad2008-08-313-24/+2
| | | | Originally committed as revision 15118 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Declare ff_vp3_idct_data to be uint16_tDavid Conrad2008-08-311-1/+1
| | | | Originally committed as revision 15117 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Don't declare SSE vp3 idct data static, so it can be used in the mmx versionDavid Conrad2008-08-311-2/+2
| | | | Originally committed as revision 15116 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove unused dequantization code from SSE VP3 IDCTDavid Conrad2008-08-301-343/+0
| | | | Originally committed as revision 15054 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use ff_pw_8 in MMX/SSE VP3 IDCTDavid Conrad2008-08-302-16/+5
| | | | Originally committed as revision 15053 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Let ff_pw_8 be used as an SSE constantDavid Conrad2008-08-302-2/+2
| | | | Originally committed as revision 15052 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Add explicit (int) cast to i386 optimized MUL* macros.Vladimir Voroshilov2008-08-261-3/+3
| | | | | | | Wrong result is returned when 16-bit value is passed as value. Also fixes "Warning: using `%edx' instead of `%dx' due to `l' suffix". Originally committed as revision 14981 to svn://svn.ffmpeg.org/ffmpeg/trunk
* stricter constraints of asm() blocksAlexis Ballier2008-08-241-6/+6
| | | | | | | | | All these variables are used as left operands of a movd instruction, which does accept only memory or register operands while the "g" constraint also allows immediates. Use "rm" instead. Patch by Alexis Ballier %alexis P ballier A gmail P com% Originally committed as revision 14941 to svn://svn.ffmpeg.org/ffmpeg/trunk
* file which should have been added in r14749Loren Merritt2008-08-141-0/+92
| | | | Originally committed as revision 14751 to svn://svn.ffmpeg.org/ffmpeg/trunk
* missing prototypeLoren Merritt2008-08-141-0/+1
| | | | Originally committed as revision 14750 to svn://svn.ffmpeg.org/ffmpeg/trunk
* gcc chokes on the 7 registers needed for float_to_int16_interleave6 (even ↵Loren Merritt2008-08-141-43/+19
| | | | | | inside HAVE_7REGS), so write it in yasm Originally committed as revision 14749 to svn://svn.ffmpeg.org/ffmpeg/trunk
* gcc chokes on xmm constraints, so pessimize int32_to_float_fmul_scalar_sse a ↵Loren Merritt2008-08-141-2/+2
| | | | | | little Originally committed as revision 14748 to svn://svn.ffmpeg.org/ffmpeg/trunk
* special case 6 channel version of float_to_int16_interleaveLoren Merritt2008-08-131-5/+51
| | | | | | 5% faster ac3 Originally committed as revision 14744 to svn://svn.ffmpeg.org/ffmpeg/trunk
* simd int->floatLoren Merritt2008-08-131-0/+46
| | | | | | 20% faster ac3 if downmixing, 15% if not Originally committed as revision 14743 to svn://svn.ffmpeg.org/ffmpeg/trunk
* simd downmixLoren Merritt2008-08-131-0/+100
| | | | | | 13% faster ac3 if downmixing Originally committed as revision 14742 to svn://svn.ffmpeg.org/ffmpeg/trunk
* cosmetics in dsp initLoren Merritt2008-08-121-6/+4
| | | | Originally committed as revision 14704 to svn://svn.ffmpeg.org/ffmpeg/trunk
* remove mdct tmp bufferLoren Merritt2008-08-122-4/+2
| | | | Originally committed as revision 14702 to svn://svn.ffmpeg.org/ffmpeg/trunk
* optimize imdct_half:Loren Merritt2008-08-122-283/+191
| | | | | | | | remove tmp buffer. skip fft reinterleave pass, leaving data in a format more convenient for simd. merge post-rotate with post-reorder. Originally committed as revision 14700 to svn://svn.ffmpeg.org/ffmpeg/trunk
* split-radix FFTLoren Merritt2008-08-124-309/+523
| | | | | | c is 1.9x faster than previous c (on various x86 cpus), sse is 1.6x faster than previous sse. Originally committed as revision 14698 to svn://svn.ffmpeg.org/ffmpeg/trunk
* import yasm macros from x264Loren Merritt2008-08-111-0/+546
| | | | Originally committed as revision 14697 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark add_png_paeth_prediction_* functions which are only used within this fileUoti Urpala2008-08-021-1/+1
| | | | | | as static. patch by Uoti Urpala, uoti.urpala pp1.inet fi Originally committed as revision 14509 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix h264_loop_filter_strength_mmx2() so it works with PAFF.Michael Niedermayer2008-07-191-1/+11
| | | | | | | | | fixed at least: CVFI1_Sony_D.jsv CVFI1_SVA_C.264 MR6_BT_B.h264 Originally committed as revision 14310 to svn://svn.ffmpeg.org/ffmpeg/trunk
* float_to_int16_interleave: change src to an array of pointers instead of ↵Loren Merritt2008-07-161-10/+10
| | | | | | | | assuming it's contiguous. this has no immediate effect, but will allow it to be used in more codecs. Originally committed as revision 14252 to svn://svn.ffmpeg.org/ffmpeg/trunk
* 10l, float_to_int16_interleave_sse/3dnow wrote the wrong samplesLoren Merritt2008-07-151-2/+2
| | | | Originally committed as revision 14236 to svn://svn.ffmpeg.org/ffmpeg/trunk
* exploit mdct symmetryLoren Merritt2008-07-133-22/+161
| | | | | | 2% faster vorbis on conroe, k8. 7% on celeron. Originally committed as revision 14207 to svn://svn.ffmpeg.org/ffmpeg/trunk
* simplify vorbis windowingLoren Merritt2008-07-131-1/+120
| | | | Originally committed as revision 14205 to svn://svn.ffmpeg.org/ffmpeg/trunk
* SSE2 optimizations for Monkey's Audio decoder vector functionsKostya Shishkov2008-07-111-0/+78
| | | | Originally committed as revision 14161 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Make the function prototype visible to comply with C99 inline.Alexander Strange2008-07-091-0/+1
| | | | | | Fixes building with gcc -std=gnu99. Originally committed as revision 14140 to svn://svn.ffmpeg.org/ffmpeg/trunk
* float_to_int16_sse2()Michael Niedermayer2008-07-091-0/+20
| | | | | | 20% faster than sse Originally committed as revision 14138 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Make LOAD4/STORE4 macros more generic.Victor Pollex2008-07-082-18/+18
| | | | | | | | Patch by Victor Pollex victor pollex web de Original thread: [PATCH] mmx implementation of vc-1 inverse transformations Date: 06/21/2008 03:37 PM Originally committed as revision 14108 to svn://svn.ffmpeg.org/ffmpeg/trunk
* 1 c-asm loop less and 1x unroll of float_to_int16_sse()Michael Niedermayer2008-07-071-12/+19
| | | | | | 25% faster Originally committed as revision 14104 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix x86-64Michael Niedermayer2008-07-071-2/+2
| | | | Originally committed as revision 14103 to svn://svn.ffmpeg.org/ffmpeg/trunk
* dont use C-asm loops and unroll once float_to_int16_3dnow()Michael Niedermayer2008-07-071-12/+19
| | | | | | 30% faster Originally committed as revision 14102 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Realign newlines.Alexander Strange2008-06-281-62/+62
| | | | Originally committed as revision 14023 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Use MANGLE() instead of memory operands to read globals.Alexander Strange2008-06-281-10/+10
| | | | | | (fixes out of registers with apple gcc 4.2) Originally committed as revision 14022 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix add_bytes_mmx and add_bytes_l2_mmx for w < 16Reimar Döffinger2008-06-221-2/+6
| | | | Originally committed as revision 13877 to svn://svn.ffmpeg.org/ffmpeg/trunk
View Lorry Controller Status