summaryrefslogtreecommitdiffstats
path: root/libavcodec/i386
Commit message (Expand)AuthorAgeFilesLines
* added define for builtins use - inverse fix by Romain DolbeauFabrice Bellard2003-01-071-0/+12
* PIC / ebx fixMichael Niedermayer2003-01-072-8/+16
* more gcc bug workaroundsMichael Niedermayer2003-01-072-7/+15
* gcc 3.2.2 -O3 bug workaround (older gcc are very likely affected too but didn...Michael Niedermayer2003-01-071-3/+3
* 1000l (push & esp) using mangle now ...Michael Niedermayer2003-01-061-82/+59
* qpel in b frames bugfixesMichael Niedermayer2003-01-052-149/+152
* slowdown / gcc 2.95.* bug workaround (this should be reversed as soon as gcc ...Michael Niedermayer2003-01-051-9/+34
* qpel in mmx2/3dnowMichael Niedermayer2003-01-053-2/+1181
* warnings patch by (bubu <bubu at bubu dot net>)bubu2003-01-011-1/+0
* qpel encodingMichael Niedermayer2002-12-272-2/+178
* 100l (sad8x8_x2 had params in wrong order)Michael Niedermayer2002-12-031-1/+2
* via c3 fix patch by (Francisco Javier Cabello Torres <fjcabello at visual-too...Francisco Javier Cabello Torres2002-11-261-1/+2
* via c3 detection patch by (Francisco Javier Cabello Torres <fjcabello at visu...Francisco Javier Cabello Torres2002-11-251-0/+15
* huffyuvMichael Niedermayer2002-11-141-0/+48
* * using DSPContext - so each codec could use its local (sub)set of CPU extensionZdenek Kabelac2002-11-111-5/+6
* * using DSPContext - so each codec could use its local (sub)set of CPU extensionZdenek Kabelac2002-11-111-2/+2
* * using DSPContext - so each codec could use its local (sub)set of CPU extensionZdenek Kabelac2002-11-111-121/+120
* cleanupMichael Niedermayer2002-11-051-8/+8
* new generic FFT/MDCT code for audio codecsFabrice Bellard2002-10-281-0/+128
* idct_permutation_type variable, so the permutation type can quickly be identi...Michael Niedermayer2002-10-251-17/+3
* idct permutation cleanup, idct can be selected per context nowMichael Niedermayer2002-09-294-151/+241
* pix_sum16_mmx()Michael Niedermayer2002-09-201-0/+39
* fdct_mmx -> ff_fdct_mmx (renamed to avoid namespace conflict with xvid)Arpi2002-09-193-3/+3
* put/avg_pixels16Michael Niedermayer2002-09-113-61/+325
* fixing encoding at high QPs (was broken after the dct accuracy increase due t...Michael Niedermayer2002-09-021-5/+5
* dct cleanupMichael Niedermayer2002-09-012-4/+4
* be less verbose patch by (Lennert Buytenhek <buytenh at math dot leidenuniv d...Lennert Buytenhek2002-09-011-8/+8
* bitexact hack for the simple mmx idctMichael Niedermayer2002-08-311-0/+26
* slow but accurate integer dct from IJG (should be ok with the LGPL as the old...Michael Niedermayer2002-08-293-7/+11
* warning patch by (Dominik Mierzejewski <dominik at rangers dot eu dot org>)Dominik Mierzejewski2002-08-281-2/+2
* dont trash mm0, its still used in the next iteration of the loop (fixes stran...Michael Niedermayer2002-08-201-2/+2
* mpeg4 mpeg quantizer encodingMichael Niedermayer2002-08-061-1/+2
* pix_abs mmx2 isnt bit-exact ...Michael Niedermayer2002-07-211-0/+8
* * reimplemented remaing avg_ pixel functionsZdenek Kabelac2002-06-031-359/+12
* * implemented remaing avg_ pixel functions (these are not used offen)Zdenek Kabelac2002-06-031-3/+156
* * removed MANGLE from macros for setting constantsZdenek Kabelac2002-05-303-189/+112
* * using unrolled loops seems to help to MMX CeleronsZdenek Kabelac2002-05-291-32/+19
* * reenabled original xy2 put routine - rounding error is really bad withZdenek Kabelac2002-05-291-1/+33
* * code with new PAVGB for MMX only CPU splited into separate fileZdenek Kabelac2002-05-292-218/+171
* * new mmx code - based upon http://aggregate.org/MAGICZdenek Kabelac2002-05-291-2/+103
* * cosmetic minor changeZdenek Kabelac2002-05-291-1/+1
* * optimized remaing avg_pixels_xy2Zdenek Kabelac2002-05-291-34/+26
* * optimized avg_* functions (except xy2)Zdenek Kabelac2002-05-281-153/+139
* * more PIC friendly and faster codeZdenek Kabelac2002-05-271-96/+63
* * removed TESTCPU_MAIN - not needed for testingZdenek Kabelac2002-05-271-3/+0
* * undefine local defines when they are no longer neededZdenek Kabelac2002-05-271-1/+7
* * avoids double inclusion of this headerZdenek Kabelac2002-05-271-0/+4
* license/copyright changeFabrice Bellard2002-05-258-85/+83
* * baling 8 seems to have the same speedZdenek Kabelac2002-05-231-1/+1
* * always baling even though it seems to produce sometime worseZdenek Kabelac2002-05-231-2/+0
OpenPOWER on IntegriCloud