mirror of https://git.ffmpeg.org/ffmpeg.git
1d16a1cf99
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 |
||
|---|---|---|
| .. | ||
| Makefile | ||
| cavsdsp_mmx.c | ||
| dct32_sse.c | ||
| deinterlace.asm | ||
| dnxhd_mmx.c | ||
| dsputil_mmx.c | ||
| dsputil_mmx.h | ||
| dsputil_mmx_avg_template.c | ||
| dsputil_mmx_qns_template.c | ||
| dsputil_mmx_rnd_template.c | ||
| dsputil_yasm.asm | ||
| dsputilenc_mmx.c | ||
| fdct_mmx.c | ||
| fft.c | ||
| fft.h | ||
| fft_3dn.c | ||
| fft_3dn2.c | ||
| fft_mmx.asm | ||
| fft_sse.c | ||
| h264_chromamc.asm | ||
| h264_deblock.asm | ||
| h264_i386.h | ||
| h264_idct.asm | ||
| h264_intrapred.asm | ||
| h264_intrapred_init.c | ||
| h264_qpel_mmx.c | ||
| h264_weight.asm | ||
| h264dsp_mmx.c | ||
| idct_mmx.c | ||
| idct_mmx_xvid.c | ||
| idct_sse2_xvid.c | ||
| idct_xvid.h | ||
| lpc_mmx.c | ||
| mathops.h | ||
| mlpdsp.c | ||
| motion_est_mmx.c | ||
| mpegaudiodec_mmx.c | ||
| mpegvideo_mmx.c | ||
| mpegvideo_mmx_template.c | ||
| simple_idct_mmx.c | ||
| snowdsp_mmx.c | ||
| vc1dsp_mmx.c | ||
| vc1dsp_yasm.asm | ||
| vp3dsp.asm | ||
| vp8dsp-init.c | ||
| vp8dsp.asm | ||
| vp56_arith.h | ||
| vp56dsp.asm | ||
| vp56dsp_init.c | ||
| x86inc.asm | ||
| x86util.asm | ||