Age | Commit message (Collapse) | Author | |
---|---|---|---|
2013-11-14 | Deprecate obsolete XvMC hardware decoding support | Diego Biurrun | |
XvMC has long ago been superseded by newer acceleration APIs, such as VDPAU, and few downstreams still support it. Furthermore XvMC is not implemented within the hwaccel framework, but requires its own specific code in the MPEG-1/2 decoder, which is a maintenance burden. | |||
2013-11-08 | dsputil: Split off H.263 bits into their own H263DSPContext | Diego Biurrun | |
2013-11-05 | x86: rv40dsp: Use PAVGB instruction macro where appropriate | Diego Biurrun | |
2013-11-04 | x86: hpeldsp: Use PAVGB instruction macro where necessary | Mikulas Patocka | |
Signed-off-by: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz> Signed-off-by: Diego Biurrun <diego@biurrun.de> | |||
2013-11-02 | x86: vp8dsp: Split loopfilter code into a separate file | Diego Biurrun | |
2013-10-07 | x86: h264_idct: Update comments to match 8/10-bit depth optimization split | Diego Biurrun | |
2013-10-07 | x86inc: Utilize the shadow space on 64-bit Windows | Henrik Gramner | |
Store XMM6 and XMM7 in the shadow space in functions that clobbers them. This way we don't have to adjust the stack pointer as often, reducing the number of instructions as well as code size. Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com> | |||
2013-10-07 | x86: fdct: Employ more specific ifdefs | Diego Biurrun | |
This avoids building mmxext and sse2 code when disabled by configure. | |||
2013-10-05 | x86: dsputil: Separate ff_add_hfyu_median_prediction_cmov from dsputil_mmx | Diego Biurrun | |
The function does not depend on MMX and compilation without MMX enabled fails if the function is compiled conditional on MMX availability. | |||
2013-10-05 | x86: fdct: Initialize optimized fdct implementations in the standard way | Diego Biurrun | |
2013-10-05 | x86: xviddct: Employ more specific ifdefs | Diego Biurrun | |
This avoids building mmxext and sse2 code when disabled by configure. | |||
2013-10-04 | x86: fdct: Only build fdct code if encoders have been enabled | Diego Biurrun | |
fdct is only initialized if encoders are enabled. | |||
2013-09-16 | x86: Add an xmm clobbering wrapper for avcodec_encode_video2 | Martin Storsjö | |
This is required since 187105ff8 when we started trying to wrap this function as well. Signed-off-by: Martin Storsjö <martin@martin.st> | |||
2013-09-15 | mathops/x86: work around inline asm miscompilation with GCC 4.8.1 | Hendrik Leppkes | |
The volatile is not required here, and prevents a miscompilation with GCC 4.8.1 when building on x86 with --cpu=i686 Signed-off-by: Michael Niedermayer <michaelni@gmx.at> Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com> | |||
2013-08-29 | x86: avcodec: Consistently structure CPU extension initialization | Diego Biurrun | |
2013-08-29 | x86: avcodec: Use convenience macros to check for CPU flags | Diego Biurrun | |
2013-08-29 | x86: rv40dsp: Move inline assembly optimizations out of YASM init section | Diego Biurrun | |
2013-08-29 | dsputil: x86: Hide arch-specific initialization details | Diego Biurrun | |
Also give consistent names to init functions. | |||
2013-08-23 | vp56: Mark VP6-only optimizations as such. | Diego Biurrun | |
Most of our VP56 optimizations are VP6-only and will stay that way. So avoid compiling them for VP5-only builds. | |||
2013-08-21 | x86: Split DCT and FFT initialization into separate files | Diego Biurrun | |
2013-08-21 | x86: h264_idct: Remove incorrect comment | Diego Biurrun | |
2013-07-18 | Consistently use "cpu_flags" as variable/parameter name for CPU flags | Diego Biurrun | |
2013-07-17 | fmtconvert: Explicitly use int32_t instead of int | Christophe Gisquet | |
Signed-off-by: Martin Storsjö <martin@martin.st> | |||
2013-07-12 | mlpdsp: x86: Respect cpuflags | Luca Barbato | |
2013-07-04 | cabac: x86 version of get_cabac_bypass | Jason Garrett-Glaser | |
Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | |||
2013-05-30 | build: cosmetics: Place unconditional before conditional OBJS lines | Diego Biurrun | |
Signed-off-by: Martin Storsjö <martin@martin.st> | |||
2013-05-16 | mpegvideo: Remove commented-out PARANOID debug cruft | Diego Biurrun | |
2013-05-13 | x86: dsputil: Rename dsputil_mmx.h --> dsputil_x86.h | Diego Biurrun | |
The header is not (anymore) MMX-specific. | |||
2013-05-13 | x86: dsputil: Split inline assembly from init code | Diego Biurrun | |
Also remove some pointless comments. | |||
2013-05-13 | x86: dsputil: Refactor pixels16 wrapper functions with a macro | Diego Biurrun | |
2013-05-12 | configure: Rename cmov processor capability to i686 | Diego Biurrun | |
The goal is to make the capapility slightly more general and have it cover the availability of the nopl instruction in addition to cmov. | |||
2013-05-10 | x86: sbrdsp: implement SSE2 qmf_pre_shuffle | Christophe Gisquet | |
From 253 to 51 cycles on Arrandale and Win64. 44 cycles on SandyBridge. Signed-off-by: Anton Khirnov <anton@khirnov.net> | |||
2013-05-08 | x86: dsputil: Remove unused argument from QPEL_OP macro | Diego Biurrun | |
2013-05-08 | x86: dsputil: Move TRANSPOSE4 macro to the only place it is used | Diego Biurrun | |
2013-05-08 | x86: dsputil: Move constant declarations into separate header | Diego Biurrun | |
2013-05-08 | x86: dsputil: Group all assembly constants together in constants.c | Diego Biurrun | |
2013-05-08 | x86: dsputil: Move ff_pd assembly constants to the only place they are used | Diego Biurrun | |
2013-05-07 | x86: dsputil: Remove unused ff_pb_3F constant | Diego Biurrun | |
2013-05-07 | x86: dsputil: Remove unused MOVQ_BONE macro | Diego Biurrun | |
2013-05-07 | x86: dsputil: Move rv40-specific functions where they belong | Diego Biurrun | |
2013-05-07 | x86: dsputil hpeldsp: Move shared template functions into separate object | Diego Biurrun | |
2013-05-07 | x86: rnd_template: Eliminate pointless OP_AVG macro indirection | Diego Biurrun | |
2013-05-06 | x86: hpeldsp: Move avg_pixels8_x2_mmx() out of hpeldsp_rnd_template.c | Diego Biurrun | |
The function is only instantiated once, so there is no point in keeping it in a template file. | |||
2013-05-06 | x86: hpeldsp: Only compile MMX hpeldsp code if MMX is enabled | Diego Biurrun | |
2013-05-06 | x86: More specific ifdefs for dsputil/hpeldsp init functions | Diego Biurrun | |
2013-05-04 | avcodec: Add av_cold attributes to init functions missing them | Diego Biurrun | |
2013-05-03 | silly typo fixes | Diego Biurrun | |
2013-05-03 | x86: sbrdsp: Implement SSE2 qmf_deint_bfly | Christophe Gisquet | |
Sandybridge: 47 cycles Having a loop counter is a 7 cycle gain. Unrolling is another 7 cycle gain. Working in reverse scan is another 6 cycles. Signed-off-by: Diego Biurrun <diego@biurrun.de> | |||
2013-05-02 | x86: dsputil: Move cavs and vc1-specific functions where they belong | Diego Biurrun | |
2013-05-02 | x86: dsputil: Move avg_pixels16_mmx() out of rnd_template.c | Diego Biurrun | |
The function does not do any rounding, so there is no point in keeping it in a round template file. |