diff options
author | Ronald S. Bultje <rsbultje@gmail.com> | 2010-09-14 13:36:26 +0000 |
---|---|---|
committer | Ronald S. Bultje <rsbultje@gmail.com> | 2010-09-14 13:36:26 +0000 |
commit | 1d16a1cf99488f16492b1bb48e023f4da8377e07 (patch) | |
tree | bbb29695e29d17f22118af56d4c511b286a08e11 /libavcodec/x86/Makefile | |
parent | 6ebf0bfc17253e29a226e888fb9b7b31574a2eed (diff) | |
download | ffmpeg-streaming-1d16a1cf99488f16492b1bb48e023f4da8377e07.zip ffmpeg-streaming-1d16a1cf99488f16492b1bb48e023f4da8377e07.tar.gz |
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
Diffstat (limited to 'libavcodec/x86/Makefile')
-rw-r--r-- | libavcodec/x86/Makefile | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile index 528d4f9..48e71ee 100644 --- a/libavcodec/x86/Makefile +++ b/libavcodec/x86/Makefile @@ -10,7 +10,7 @@ YASM-OBJS-$(CONFIG_FFT) += x86/fft_mmx.o \ MMX-OBJS-$(CONFIG_H264DSP) += x86/h264dsp_mmx.o YASM-OBJS-$(CONFIG_H264DSP) += x86/h264_deblock.o \ x86/h264_weight.o \ - x86/h264_idct_sse2.o \ + x86/h264_idct.o \ YASM-OBJS-$(CONFIG_H264PRED) += x86/h264_intrapred.o MMX-OBJS-$(CONFIG_H264PRED) += x86/h264_intrapred_init.o |