virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
Code Issues Releases Activity
60198742ff
FFmpeg/libavcodec/x86/Makefile

92 lines
5.4 KiB
Makefile
Raw Normal View History

dsputil: Make dsputil selectable Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-12 17:28:11 +03:00
OBJS += x86/constants.o \
x86/fmtconvert_init.o \
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_AAC_DECODER) += x86/sbrdsp_init.o
OBJS-$(CONFIG_AC3DSP) += x86/ac3dsp_init.o
x86: cavs: call ff_cavsdsp_init_x86() under if (ARCH_X86)
2012-10-07 18:23:54 +03:00
OBJS-$(CONFIG_CAVS_DECODER) += x86/cavsdsp.o
x86: Always compile files with functions that are called unconditionally
2012-08-28 18:47:39 +03:00
OBJS-$(CONFIG_DNXHD_ENCODER) += x86/dnxhdenc.o
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_FFT) += x86/fft_init.o
h264chroma: x86: Fix building with yasm disabled Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-06 17:34:39 +03:00
OBJS-$(CONFIG_H264CHROMA) += x86/h264chroma_init.o
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_H264DSP) += x86/h264dsp_init.o
OBJS-$(CONFIG_H264PRED) += x86/h264_intrapred_init.o
dsputil: Separate h264 qpel The sh4 optimizations are removed, because the code is 100% identical to the C code, so it is unlikely to provide any real practical benefit. Signed-off-by: Diego Biurrun <diego@biurrun.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-24 07:24:53 +03:00
OBJS-$(CONFIG_H264QPEL) += x86/h264_qpel.o
x86: hpeldsp: Only compile MMX hpeldsp code if MMX is enabled
2013-05-06 02:01:05 +03:00
OBJS-$(CONFIG_HPELDSP) += x86/hpeldsp_init.o
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_LPC) += x86/lpc.o
Simplify arch-specific object file lists Originally committed as revision 22570 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-16 23:23:03 +02:00
OBJS-$(CONFIG_MLP_DECODER) += x86/mlpdsp.o
mpegaudiosp: More consistent names for ppc/x86 optimization files
2013-04-27 02:03:37 +03:00
OBJS-$(CONFIG_MPEGAUDIODSP) += x86/mpegaudiodsp.o
x86: Always compile files with functions that are called unconditionally
2012-08-28 18:47:39 +03:00
OBJS-$(CONFIG_MPEGVIDEO) += x86/mpegvideo.o
OBJS-$(CONFIG_MPEGVIDEOENC) += x86/mpegvideoenc.o
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_PNG_DECODER) += x86/pngdsp_init.o
OBJS-$(CONFIG_PRORES_DECODER) += x86/proresdsp_init.o
OBJS-$(CONFIG_RV30_DECODER) += x86/rv34dsp_init.o
OBJS-$(CONFIG_RV40_DECODER) += x86/rv34dsp_init.o \
x86/rv40dsp_init.o
Simplify arch-specific object file lists Originally committed as revision 22570 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-16 23:23:03 +02:00
OBJS-$(CONFIG_TRUEHD_DECODER) += x86/mlpdsp.o
x86: vc1: call ff_vc1dsp_init_x86() under if (ARCH_X86)
2012-10-07 18:41:10 +03:00
OBJS-$(CONFIG_VC1_DECODER) += x86/vc1dsp_init.o
lavc: introduce VideoDSPContext Move some functions from dsputil. The idea is that videodsp contains functions that are useful for a large and varied set of video decoders. Currently, it contains emulated_edge_mc() and prefetch(). Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2012-12-15 20:46:02 +03:00
OBJS-$(CONFIG_VIDEODSP) += x86/videodsp_init.o
Move vorbis_inverse_coupling from dsputil to vorbisdspcontext. Conveniently (together with Justin's earlier patches), this makes our vorbis decoder entirely independent of dsputil.
2013-01-20 09:21:10 +03:00
OBJS-$(CONFIG_VORBIS_DECODER) += x86/vorbisdsp_init.o
vp3: move idct and loop filter pointers to new vp3dsp context This moves all VP3-specific function pointers from dsputil to a new vp3dsp context. There is no reason to ever use the VP3 IDCT where an MPEG2 IDCT is expected or vice versa. Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-07-17 18:47:43 +03:00
OBJS-$(CONFIG_VP3DSP) += x86/vp3dsp_init.o
x86: call most of the x86 dsp init functions under if (ARCH_X86) Rename the called dsp init functions to *_init_x86.
2012-10-05 20:54:10 +03:00
OBJS-$(CONFIG_VP5_DECODER) += x86/vp56dsp_init.o
OBJS-$(CONFIG_VP6_DECODER) += x86/vp56dsp_init.o
OBJS-$(CONFIG_VP8_DECODER) += x86/vp8dsp_init.o
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
OBJS-$(CONFIG_XMM_CLOBBER_TEST) += x86/w64xmmtest.o
Move arch-specific makefile parts into $arch/Makefile Originally committed as revision 22569 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-16 23:22:59 +02:00
x86: dsputil: Split inline assembly from init code Also remove some pointless comments.
2013-05-08 02:45:39 +03:00
MMX-OBJS-$(CONFIG_DSPUTIL) += x86/dsputil_init.o \
x86/dsputil_mmx.o \
x86: avcodec: Drop silly "_mmx" suffixes from filenames
2012-08-08 03:12:17 +03:00
x86/fdct.o \
x86: Move duplicated put_pixels{8|16}_mmx functions into their own file
2013-04-23 18:10:59 +03:00
x86/fpel_mmx.o \
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
x86/idct_mmx_xvid.o \
x86/idct_sse2_xvid.o \
x86: dsputil hpeldsp: Move shared template functions into separate object
2013-05-07 01:42:22 +03:00
x86/rnd_mmx.o \
x86: avcodec: Drop silly "_mmx" suffixes from filenames
2012-08-08 03:12:17 +03:00
x86/simple_idct.o \
dct32: port SSE 32-point DCT to YASM
2011-05-17 19:26:01 +03:00
x86: dsputil: Only compile motion_est code when encoders are enabled
2012-09-09 22:19:50 +03:00
MMX-OBJS-$(CONFIG_ENCODERS) += x86/dsputilenc_mmx.o \
x86/motion_est.o
x86: hpeldsp: Move avg_pixels8_x2_mmx() out of hpeldsp_rnd_template.c The function is only instantiated once, so there is no point in keeping it in a template file.
2013-04-23 19:36:25 +03:00
MMX-OBJS-$(CONFIG_HPELDSP) += x86/fpel_mmx.o \
x86: dsputil hpeldsp: Move shared template functions into separate object
2013-05-07 01:42:22 +03:00
x86/hpeldsp_mmx.o \
x86/rnd_mmx.o
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
MMX-OBJS-$(CONFIG_VC1_DECODER) += x86/vc1dsp_mmx.o
build: cosmetics: Place unconditional before conditional OBJS lines Signed-off-by: Martin Storsjö <martin@martin.st>
2013-05-30 00:15:17 +03:00
YASM-OBJS += x86/deinterlace.o \
x86/fmtconvert.o \
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
YASM-OBJS-$(CONFIG_AAC_DECODER) += x86/sbrdsp.o
YASM-OBJS-$(CONFIG_AC3DSP) += x86/ac3dsp.o
x86: avcodec: Drop silly "_sse" suffixes from filenames
2012-08-08 02:49:46 +03:00
YASM-OBJS-$(CONFIG_DCT) += x86/dct32.o
dsputil: Make dsputil selectable Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-12 17:28:11 +03:00
YASM-OBJS-$(CONFIG_DSPUTIL) += x86/dsputil.o \
x86: hpeldsp: Move half-pel assembly from dsputil to hpeldsp Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-11 01:37:59 +03:00
x86/fpel.o \
dsputil: Make dsputil selectable Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-12 17:28:11 +03:00
x86/mpeg4qpel.o \
x86/qpel.o
x86: Drop silly "_yasm" suffixes from filenames
2012-08-08 01:35:43 +03:00
YASM-OBJS-$(CONFIG_ENCODERS) += x86/dsputilenc.o
x86: avcodec: Drop silly "_mmx" suffixes from filenames
2012-08-08 03:12:17 +03:00
YASM-OBJS-$(CONFIG_FFT) += x86/fft.o
x86: dsputil: Fix h263 loop filter link error in some configurations This was caused by unconditionally referencing a conditionally compiled table. Now the code is also compiled conditionally. Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-02-18 18:50:06 +03:00
YASM-OBJS-$(CONFIG_H263_DECODER) += x86/h263_loopfilter.o
YASM-OBJS-$(CONFIG_H263_ENCODER) += x86/h263_loopfilter.o
h264chroma: x86: Fix building with yasm disabled Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-06 17:34:39 +03:00
YASM-OBJS-$(CONFIG_H264CHROMA) += x86/h264_chromamc.o \
build: conditionally compile x86 H.264 chroma optimizations
2011-12-14 00:54:52 +03:00
x86/h264_chromamc_10bit.o
Port latest x264 deblock asm (before they moved to using NV12 as internal format), LGPL'ed with permission from Jason and Loren. This includes mmx2 code, so remove inline asm from h264dsp_mmx.c accordingly. Originally committed as revision 25031 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-03 19:52:46 +03:00
YASM-OBJS-$(CONFIG_H264DSP) += x86/h264_deblock.o \
Port x86 10-bit H.264 deblock asm from x264
2011-05-10 18:55:12 +03:00
x86/h264_deblock_10bit.o \
Rename h264_idct_sse2.asm to h264_idct.asm; move inline IDCT asm from h264dsp_mmx.c to h264_idct.asm (as yasm code). Because the loops are now coded in asm instead of C, this is (depending on the function) up to 50% faster for cases where gcc didn't do a great job at looping. Since h264_idct_add8() is now faster than the manual loop setup in h264.c, in-asm idct calling can now be enabled for chroma as well (see r16207). For MMX, this is 5% faster. For SSE2 (which isn't done for chroma if h264.c does the looping), this makes it up to 50% faster. Speed gain overall is ~0.5-1.0%. Originally committed as revision 25119 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-14 16:36:26 +03:00
x86/h264_idct.o \
Add IDCT functions for 10-bit H.264. Ports the majority of IDCT functions for 10-bit H.264. Parts are inspired from 8-bit IDCT code in Libav; other parts ported from x264 with relicensing permission from author. Signed-off-by: Ronald S. Bultje <rbultje@google.com>
2011-05-24 22:14:38 +03:00
x86/h264_idct_10bit.o \
x86/h264_weight.o \
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
x86/h264_weight_10bit.o
Add x86 assembly for some 10-bit H.264 intra predict functions. Parts are inspired from the 8-bit H.264 predict code in Libav. Other parts ported from x264 with relicensing permission from author. Signed-off-by: Diego Biurrun <diego@biurrun.de>
2011-06-06 02:20:05 +03:00
YASM-OBJS-$(CONFIG_H264PRED) += x86/h264_intrapred.o \
x86/h264_intrapred_10bit.o
x86: h264: Convert 8-bit QPEL inline assembly to YASM Signed-off-by: Diego Biurrun <diego@biurrun.de>
2012-10-13 18:04:50 +03:00
YASM-OBJS-$(CONFIG_H264QPEL) += x86/h264_qpel_8bit.o \
dsputil: Make dsputil selectable Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-12 17:28:11 +03:00
x86/h264_qpel_10bit.o \
x86: hpeldsp: Move half-pel assembly from dsputil to hpeldsp Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-11 01:37:59 +03:00
x86/fpel.o \
dsputil: Make dsputil selectable Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-12 17:28:11 +03:00
x86/qpel.o
x86: hpeldsp: Move half-pel assembly from dsputil to hpeldsp Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-11 01:37:59 +03:00
YASM-OBJS-$(CONFIG_HPELDSP) += x86/fpel.o \
x86/hpeldsp.o
x86: avcodec: Drop silly "_sse" suffixes from filenames
2012-08-08 02:49:46 +03:00
YASM-OBJS-$(CONFIG_MPEGAUDIODSP) += x86/imdct36.o
build: prettyprinting cosmetics
2012-02-03 01:55:57 +03:00
YASM-OBJS-$(CONFIG_PNG_DECODER) += x86/pngdsp.o
YASM-OBJS-$(CONFIG_PRORES_DECODER) += x86/proresdsp.o
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
YASM-OBJS-$(CONFIG_RV30_DECODER) += x86/rv34dsp.o
rv40: x86 SIMD for biweight Provide MMX, SSE2 and SSSE3 versions, with a fast-path when the weights are multiples of 512 (which is often the case when the values round up nicely). *_TIMER report for the 16x16 and 8x8 cases: C: 9015 decicycles in 16, 524257 runs, 31 skips 2656 decicycles in 8, 524271 runs, 17 skips MMX: 4156 decicycles in 16, 262090 runs, 54 skips 1206 decicycles in 8, 262131 runs, 13 skips MMX on fast-path: 2760 decicycles in 16, 524222 runs, 66 skips 995 decicycles in 8, 524252 runs, 36 skips SSE2: 2163 decicycles in 16, 262131 runs, 13 skips 832 decicycles in 8, 262137 runs, 7 skips SSE2 with fast path: 1783 decicycles in 16, 524276 runs, 12 skips 711 decicycles in 8, 524283 runs, 5 skips SSSE3: 2117 decicycles in 16, 262136 runs, 8 skips 814 decicycles in 8, 262143 runs, 1 skips SSSE3 with fast path: 1315 decicycles in 16, 524285 runs, 3 skips 578 decicycles in 8, 524286 runs, 2 skips This means around a 4% speedup for some sequences. Signed-off-by: Diego Biurrun <diego@biurrun.de>
2012-01-12 02:11:15 +03:00
YASM-OBJS-$(CONFIG_RV40_DECODER) += x86/rv34dsp.o \
x86/rv40dsp.o
x86: Drop silly "_yasm" suffixes from filenames
2012-08-08 01:35:43 +03:00
YASM-OBJS-$(CONFIG_VC1_DECODER) += x86/vc1dsp.o
lavc: introduce VideoDSPContext Move some functions from dsputil. The idea is that videodsp contains functions that are useful for a large and varied set of video decoders. Currently, it contains emulated_edge_mc() and prefetch(). Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2012-12-15 20:46:02 +03:00
YASM-OBJS-$(CONFIG_VIDEODSP) += x86/videodsp.o
vorbisdsp: convert x86 simd functions from inline asm to yasm.
2013-01-16 04:00:41 +03:00
YASM-OBJS-$(CONFIG_VORBIS_DECODER) += x86/vorbisdsp.o
build: add CONFIG_VP3DSP, reduce repetition in OBJS lists Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-07-16 14:50:14 +03:00
YASM-OBJS-$(CONFIG_VP3DSP) += x86/vp3dsp.o
YASM-OBJS-$(CONFIG_VP6_DECODER) += x86/vp56dsp.o
First shot at VP8 optimizations: - MMXEXT, SSE2 and SSSE3 MC functions - MMX and SSE4 IDCT dc_add functions Patch by Jason Garrett-Glaser <darkshikari gmail com> and myself. Originally committed as revision 23815 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-27 05:01:45 +03:00
YASM-OBJS-$(CONFIG_VP8_DECODER) += x86/vp8dsp.o
Reference in New Issue Copy Permalink