summaryrefslogtreecommitdiffstats
path: root/libavcodec/i386
Commit message (Expand)AuthorAgeFilesLines
* Explain why there are no multiple inclusion guards in these header files.Diego Biurrun2007-10-173-0/+12
* Remove wrong multiple inclusion guards.Aurelien Jacobs2007-10-173-15/+0
* Add FFMPEG_ prefix to all multiple inclusion guards.Diego Biurrun2007-10-174-12/+12
* Add missing multiple inclusion guards.Diego Biurrun2007-10-173-1/+13
* Fix intended order of operations for 4 assert() checks.Shane2007-10-111-4/+4
* sse2 version of compute_autocorr().Loren Merritt2007-09-291-0/+123
* Replace complicated and currently broken manual alignment code byReimar Döffinger2007-08-301-3/+1
* typoMichael Niedermayer2007-08-271-1/+1
* CONFIG_7REGS has been renamed to HAVE_7REGSRamiro Polla2007-08-262-4/+4
* workaround gcc bug, untested as my gcc is not complainingMichael Niedermayer2007-08-262-0/+6
* optimize the first vertical lifting step, this also prevents anotherMichael Niedermayer2007-08-261-25/+25
* optimize 1st horizontal lifting stepMichael Niedermayer2007-08-261-34/+22
* typoMichael Niedermayer2007-08-261-1/+1
* get rid of totally senseless "m" + read in register we have enoughMichael Niedermayer2007-08-261-72/+52
* simplify senselessly complex addressingMichael Niedermayer2007-08-261-27/+28
* cosmeticsMichael Niedermayer2007-08-261-36/+27
* avoid overflow in the 3rd lifting step, this now needs mmx2 at minimumMichael Niedermayer2007-08-262-30/+77
* avoid an overflow in the 1 horizontal lifting stepMichael Niedermayer2007-08-251-12/+12
* prevent one overflow in the first vertical lifting stepMichael Niedermayer2007-08-251-6/+6
* update mmx code to latest snow changesMichael Niedermayer2007-08-252-238/+210
* use 16bit IDWT (a SIMD implementation of it should be >2x faster then withMichael Niedermayer2007-08-251-0/+2
* Change rounding of the horizontal DWT to match the vertical one.Michael Niedermayer2007-08-211-27/+17
* remove code which become unused by the previous changesMichael Niedermayer2007-08-211-9/+0
* Simplify and optimize the 4th vertical lifting step of the SSE2 code (untested)Michael Niedermayer2007-08-211-2/+1
* Simplify and optimize the 4th vertical lifting step of the MMX codeMichael Niedermayer2007-08-211-2/+1
* Simplify and speedup code, reduce needed headroom by 2 bits in the 3rdMichael Niedermayer2007-08-201-5/+4
* simplify, speedup and reduce needed headroom by 2 bits in the 3rdMichael Niedermayer2007-08-201-5/+4
* replace <<1 by add for SSE2 (untested)Michael Niedermayer2007-08-201-2/+2
* replace <<1 by addMichael Niedermayer2007-08-201-2/+2
* slightly change horizontal lift3 so it needs 1 bit less headroomMichael Niedermayer2007-08-201-6/+6
* remove idiotc double subtraction from the sse2 code (untested, no sse2 here)Michael Niedermayer2007-08-201-18/+15
* fixme noteMichael Niedermayer2007-08-201-1/+1
* and of course the unneeded double subtractions were blindly put in theMichael Niedermayer2007-08-201-18/+15
* help some gcc version to optimize out those functionsAurelien Jacobs2007-07-241-0/+4
* cosmetics: indentationAurelien Jacobs2007-07-101-2/+2
* Avoid linking with h263.c functions when the relevant codecsAurelien Jacobs2007-07-101-0/+3
* make arguments to ssd_int8_vs_int16() constMåns Rullgård2007-07-081-1/+1
* typosDiego Biurrun2007-07-052-2/+2
* license header consistency cosmeticsDiego Biurrun2007-07-051-1/+1
* Group all copyright and author notices together.Diego Biurrun2007-07-056-66/+56
* add a comment to indicate which #endif belong to which #defineGuillaume Poirier2007-06-171-1/+1
* add multiple inclusion guards to headersMåns Rullgård2007-06-171-0/+4
* correct type for {put,avg}_h264_chroma_mc2_mmx2Måns Rullgård2007-06-161-2/+2
* misc typo fixesDiego Biurrun2007-06-122-3/+3
* misc spelling fixesDiego Biurrun2007-06-125-14/+15
* work around issues with the old version of Gnu Assembler shipped onGraham Booker2007-06-081-8/+8
* Preprocessor #defines starting with __ are reserved for the system.Diego Biurrun2007-06-071-1/+1
* Use FFmpeg-specific CPU feature definitions.Ronald S. Bultje2007-06-071-17/+17
* rename attribute_unused to av_unused and moves its declaration to common.hCarl Eugen Hoyos2007-05-301-5/+5
* Remove incorrect comment; MMX2 is preferred over 3DNow! on AthlonZuxy Meng2007-05-201-1/+0
OpenPOWER on IntegriCloud