aboutsummaryrefslogtreecommitdiff
path: root/libavcodec/i386
AgeCommit message (Expand)Author
2007-12-21Fix typo in macro name: WARPER8_16_SQ --> WRAPPER8_16_SQ.Diego Biurrun
2007-12-16move FLAC mmx dsp to its own fileAurelien Jacobs
2007-12-15Add 'l' suffix where it is necessary because type can not always beReimar Döffinger
2007-12-12add required include to make this file self-containedAurelien Jacobs
2007-12-11typo/clarificationDiego Biurrun
2007-12-02Misc spelling fixes, prefer American over British English.Diego Biurrun
2007-12-01spellingVitor Sessak
2007-11-27remove some unused ff_p* vars from dsputilAurelien Jacobs
2007-11-27remove useless #ifdef around extern declarationAurelien Jacobs
2007-11-27cosmetics: indentationAurelien Jacobs
2007-11-27convert some #ifdef CONFIG_ to if(ENABLE_Aurelien Jacobs
2007-11-27build vc1dsp_mmx.c in its own compilation unitAurelien Jacobs
2007-11-27use ff_ prefix for extern varsAurelien Jacobs
2007-11-27make ff_p* vars extern so that they can be used in various *_mmx.c filesAurelien Jacobs
2007-11-25Typo fix. Previous version had some picture error building up until next keyf...Christophe Gisquet
2007-11-24Strip debug stuff from vc1dsp_mmx.c, patch by Christophe GISQUET %hristophe P...Christophe Gisquet
2007-11-21add VC-1 MMX DSP functions, under MIT license.Christophe Gisquet
2007-11-12tring to workaround gcc 2.95 bug which causes random failuresMichael Niedermayer
2007-10-17Explain why there are no multiple inclusion guards in these header files.Diego Biurrun
2007-10-17Remove wrong multiple inclusion guards.Aurelien Jacobs
2007-10-17Add FFMPEG_ prefix to all multiple inclusion guards.Diego Biurrun
2007-10-17Add missing multiple inclusion guards.Diego Biurrun
2007-10-11Fix intended order of operations for 4 assert() checks.Shane
2007-09-29sse2 version of compute_autocorr().Loren Merritt
2007-08-30Replace complicated and currently broken manual alignment code byReimar Döffinger
2007-08-27typoMichael Niedermayer
2007-08-26CONFIG_7REGS has been renamed to HAVE_7REGSRamiro Polla
2007-08-26workaround gcc bug, untested as my gcc is not complainingMichael Niedermayer
2007-08-26optimize the first vertical lifting step, this also prevents anotherMichael Niedermayer
2007-08-26optimize 1st horizontal lifting stepMichael Niedermayer
2007-08-26typoMichael Niedermayer
2007-08-26get rid of totally senseless "m" + read in register we have enoughMichael Niedermayer
2007-08-26simplify senselessly complex addressingMichael Niedermayer
2007-08-26cosmeticsMichael Niedermayer
2007-08-26avoid overflow in the 3rd lifting step, this now needs mmx2 at minimumMichael Niedermayer
2007-08-25avoid an overflow in the 1 horizontal lifting stepMichael Niedermayer
2007-08-25prevent one overflow in the first vertical lifting stepMichael Niedermayer
2007-08-25update mmx code to latest snow changesMichael Niedermayer
2007-08-25use 16bit IDWT (a SIMD implementation of it should be >2x faster then withMichael Niedermayer
2007-08-21Change rounding of the horizontal DWT to match the vertical one.Michael Niedermayer
2007-08-21remove code which become unused by the previous changesMichael Niedermayer
2007-08-21Simplify and optimize the 4th vertical lifting step of the SSE2 code (untested)Michael Niedermayer
2007-08-21Simplify and optimize the 4th vertical lifting step of the MMX codeMichael Niedermayer
2007-08-20Simplify and speedup code, reduce needed headroom by 2 bits in the 3rdMichael Niedermayer
2007-08-20simplify, speedup and reduce needed headroom by 2 bits in the 3rdMichael Niedermayer
2007-08-20replace <<1 by add for SSE2 (untested)Michael Niedermayer
2007-08-20replace <<1 by addMichael Niedermayer
2007-08-20slightly change horizontal lift3 so it needs 1 bit less headroomMichael Niedermayer
2007-08-20remove idiotc double subtraction from the sse2 code (untested, no sse2 here)Michael Niedermayer
2007-08-20fixme noteMichael Niedermayer
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-15 16:34:10 +0000