summaryrefslogtreecommitdiffstats
path: root/libavutil/x86
Commit message (Expand)AuthorAgeFilesLines
* x86inc: fully concatenate tokens to fix macro expansion for nasmJanne Grunau2012-12-131-2/+2
* x86inc: fix stack alignment on win64Ronald S. Bultje2012-12-121-4/+6
* x86inc: support stack mem allocation and re-alignment in PROLOGUERonald S. Bultje2012-12-121-36/+149
* x86: float_dsp: fix loading of the len parameter on x86-32Justin Ruggles2012-12-071-1/+4
* x86: float_dsp: fix compilation of ff_vector_dmul_scalar_avx() on x86-32Justin Ruggles2012-12-061-1/+1
* x86: af_volume: add SSE2/SSSE3/AVX-optimized s32 volume scalingJustin Ruggles2012-12-051-0/+1
* float_dsp: add vector_dmul_scalar() to multiply a vector of doublesJustin Ruggles2012-12-053-0/+65
* x86: cpu: Drop unused HAVE_RWEFLAGS conditionDiego Biurrun2012-11-281-1/+1
* x86: float_dsp: add SSE version of vector_fmul_scalar()Justin Ruggles2012-11-262-0/+33
* x86: SPLATD: port to cpuflagsDiego Biurrun2012-11-181-16/+6
* x86: mmx2 ---> mmxext in asm constructsDiego Biurrun2012-11-141-1/+1
* build: Drop AVX assembly ifdefsDiego Biurrun2012-11-111-4/+0
* x86inc: Set program_name outside of x86inc.asmDiego Biurrun2012-11-112-1/+2
* x86: PALIGNR: port to cpuflagsDiego Biurrun2012-11-091-8/+8
* x86: PABSW: port to cpuflagsDiego Biurrun2012-11-051-12/+10
* x86: Refactor PSWAPD fallback implementations and port to cpuflagsDiego Biurrun2012-11-021-0/+12
* x86: PMINUB: port to cpuflagsDiego Biurrun2012-11-021-5/+5
* x86util: Add cpuflags_mmxext alias for cpuflags_mmx2Diego Biurrun2012-11-021-0/+2
* x86inc: Only define program_name if the macro is unsetDiego Biurrun2012-11-021-1/+3
* x86: Fix assembly with NASMDave Yeo2012-10-311-1/+1
* x86: MMX2 ---> MMXEXT in macro namesDiego Biurrun2012-10-311-3/+3
* x86: include x86inc.asm in x86util.asmDiego Biurrun2012-10-313-2/+3
* Remove INIT_AVX from x86inc.asm.Ronald S. Bultje2012-10-291-8/+0
* x86: get_cpu_flags: add necessary ifdefs around function bodyDiego Biurrun2012-10-041-0/+5
* x86: Drop CPU detection intrinsicsDiego Biurrun2012-10-041-36/+0
* x86: Add YASM implementations of cpuid and xgetbv from x264Diego Biurrun2012-10-044-2/+113
* x86: cpu: Break out test for cpuid capabilities into separate functionDiego Biurrun2012-10-041-11/+21
* x86: ff_get_cpu_flags_x86(): Avoid a pointless variable indirectionDiego Biurrun2012-10-041-4/+1
* x86: Replace checks for CPU extensions and flags by convenience macrosDiego Biurrun2012-09-081-4/+3
* x86: float_dsp: fix ff_vector_fmac_scalar_avx() on Win64Justin Ruggles2012-09-071-3/+3
* x86: Add convenience macros to check for CPU extensions and flagsDiego Biurrun2012-09-041-0/+57
* x86: Split inline and external assembly #ifdefsDiego Biurrun2012-08-311-2/+2
* x86: cosmetics: Comment some #endifs for better readabilityDiego Biurrun2012-08-301-2/+2
* vf_hqdn3d: x86 asmLoren Merritt2012-08-261-0/+1
* lavr: x86: optimized 6-channel s16 to fltp conversionJustin Ruggles2012-08-231-0/+12
* x86: remove FASTDIV inline asmMans Rullgard2012-08-221-37/+0
* Add more missing includes after removing the implicit common.hMartin Storsjö2012-08-161-0/+1
* Add some more missing includes after removing the implicit common.hMartin Storsjö2012-08-151-0/+1
* x86: move MANGLE() and related macros to libavutil/x86/asm.hMans Rullgard2012-08-091-0/+12
* x86: rename libavutil/x86_cpu.h to libavutil/x86/asm.hMans Rullgard2012-08-092-1/+99
* x86: fix build with nasm 2.08Mans Rullgard2012-08-071-11/+7
* x86: use nop cpu directives only if supportedMans Rullgard2012-08-071-3/+9
* x86: fix rNmp macros with nasmMans Rullgard2012-08-071-2/+2
* x86: add colons after labelsMans Rullgard2012-08-071-2/+2
* x86: build: replace mmx2 by mmxextDiego Biurrun2012-08-031-2/+2
* x86: Use consistent 3dnowext function and macro name suffixesDiego Biurrun2012-08-031-1/+1
* x86inc: clip num_args to 7 on x86-32.Loren Merritt2012-07-281-0/+3
* x86inc: sync to latest version from x264.Ronald S. Bultje2012-07-281-92/+124
* x86: add support for fmaddps fma4 instruction with abstraction to avx/sseJustin Ruggles2012-07-271-5/+11
* x86inc: automatically insert vzeroupper for YMM functions.Ronald S. Bultje2012-07-262-13/+12
OpenPOWER on IntegriCloud