wrapper/FFmpeg
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
FFmpeg/libavutil/x86
History
Ganesh Ajjanagadde 5989add4ab lavu/x86/lls: add fma3 optimizations for update_lls 
This improves accuracy (very slightly) and speed for processors having
fma3.

Sample benchmark (fate flac-16-lpc-cholesky, Haswell):
old:
5993610 decicycles in ff_lpc_calc_coefs,      64 runs,      0 skips
5951528 decicycles in ff_lpc_calc_coefs,     128 runs,      0 skips

new:
5252410 decicycles in ff_lpc_calc_coefs,      64 runs,      0 skips
5232869 decicycles in ff_lpc_calc_coefs,     128 runs,      0 skips

Tested with FATE and --disable-fma3, also examined contents of
lavu/lls-test.

Reviewed-by: James Almer <jamrial@gmail.com>
Reviewed-by: Henrik Gramner <henrik@gramner.com>
Signed-off-by: Ganesh Ajjanagadde <gajjanagadde@gmail.com>
2016-01-15 16:46:13 -05:00
..
asm.h avutil/x86/asm: rename REG_SP to REG_sp 2015-08-22 02:56:53 +02:00
bswap.h avutil/x86/bswap: Remove warning about bswap intrinsics with msvc. 2015-11-23 23:03:32 +11:00
cpu.c lavu: add AESNI CPU flag 2015-10-28 04:23:14 -05:00
cpu.h lavu: add AESNI CPU flag 2015-10-28 04:23:14 -05:00
cpuid.asm Merge commit '79793f8337' 2014-07-01 15:43:40 +02:00
emms.asm
emms.h
float_dsp.asm x86/float_dsp: zero extend offset from ff_scalarproduct_float_sse 2016-01-08 16:14:32 -03:00
float_dsp_init.c x86: check for AV_CPU_FLAG_AVXSLOW where useful 2015-06-01 00:15:35 +02:00
intmath.h x86/intmath: add missing early clobber to output operands 2016-01-15 13:32:58 -03:00
intreadwrite.h
lls.asm lavu/x86/lls: add fma3 optimizations for update_lls 2016-01-15 16:46:13 -05:00
lls_init.c lavu/x86/lls: add fma3 optimizations for update_lls 2016-01-15 16:46:13 -05:00
Makefile avutil/x86/Makefile: fix conditional x86/emms.o build 2015-04-09 01:12:51 +02:00
pixelutils.asm x86inc: Drop SECTION_TEXT macro 2015-08-04 20:13:09 +02:00
pixelutils.h avutil: add pixelutils API 2014-08-05 21:05:52 +02:00
pixelutils_init.c pixelutils: Comment on (lack of) sad_8x8_sse2 2015-03-04 21:58:53 +01:00
timer.h Merge commit 'd1a6cb195f' 2015-07-09 12:28:09 +02:00
w64xmmtest.h Reinstate proper FFmpeg license for all files. 2013-08-30 15:47:38 +00:00
x86inc.asm lavu: add AESNI CPU flag 2015-10-28 04:23:14 -05:00
x86util.asm x86: port PSIGNW to cpuflags 2015-09-11 23:27:03 -03:00