Sun, 10 Feb 2008 01:45:42 +0000 |
lorenm |
asm argument that might be in memory needs a size
libavcodec
|
Wed, 06 Feb 2008 12:32:31 +0000 |
lorenm |
sse2 h264 motion compensation. not new code, just separate out the cases that didn't need ssse3.
libavcodec
|
Wed, 06 Feb 2008 04:44:21 +0000 |
lorenm |
put loop counter in a register if possible. makes some of the qpel functions 3% faster.
libavcodec
|
Wed, 06 Feb 2008 04:14:07 +0000 |
lorenm |
fix aliasing warnings. simpler too.
libavcodec
|
Tue, 05 Feb 2008 11:22:55 +0000 |
lorenm |
ssse3 h264 motion compensation.
libavcodec
|
Tue, 05 Feb 2008 03:58:13 +0000 |
lorenm |
add qpel rounder once during hv rather than twice during hv and whatever it's averaged with
libavcodec
|
Tue, 05 Feb 2008 01:16:48 +0000 |
lorenm |
clean up an ugliness introduced in r11826. this syntax will require fewer changes when adding future sse2 code.
libavcodec
|
Sun, 03 Feb 2008 16:19:37 +0000 |
lorenm |
use mmx2/3dnow avg functions in avg_qpel*_mc00
libavcodec
|
Sun, 03 Feb 2008 07:05:11 +0000 |
lorenm |
ff_h264_idct8_add_sse2.
libavcodec
|
Sun, 03 Feb 2008 03:21:47 +0000 |
lorenm |
remove some movq in ff_h264_idct8_add_mmx. 225->217 cycles on core2.
libavcodec
|
Fri, 11 Jan 2008 08:29:58 +0000 |
diego |
Factorize some duplicated code from CAVS and H.264 into a common file.
libavcodec
|
Fri, 21 Dec 2007 23:11:22 +0000 |
gpoirier |
add MMX version for put_no_rnd_h264_chroma_mc8_c, used in VC-1 decoding.
libavcodec
|
Tue, 27 Nov 2007 22:36:15 +0000 |
aurel |
use ff_ prefix for extern vars
libavcodec
|
Tue, 27 Nov 2007 22:23:34 +0000 |
aurel |
make ff_p* vars extern so that they can be used in various *_mmx.c files
libavcodec
|
Thu, 11 Oct 2007 22:18:38 +0000 |
cehoyos |
Fix intended order of operations for 4 assert() checks.
libavcodec
|
Tue, 08 May 2007 17:55:56 +0000 |
lorenm |
cosmetics: remove duplicate transpose macro
libavcodec
|
Thu, 19 Apr 2007 16:12:06 +0000 |
reimar |
Fix parts missed in clip -> av_clip rename
libavcodec
|
Fri, 16 Feb 2007 21:22:53 +0000 |
michael |
add a few asserts to ensure alignment
libavcodec
|
Fri, 16 Feb 2007 21:21:07 +0000 |
michael |
prevent h.264 MC related functions from being inlined (yes this is much faster the code just doesnt fit in the code cache otherwise)
libavcodec
|
Fri, 03 Nov 2006 16:54:05 +0000 |
michael |
ensure alignment (no speed change)
libavcodec
|
Fri, 03 Nov 2006 16:02:18 +0000 |
michael |
merging mov & and (no speedchange)
libavcodec
|
Fri, 03 Nov 2006 15:40:57 +0000 |
michael |
2 instructions less (same speed)
libavcodec
|
Fri, 03 Nov 2006 15:17:36 +0000 |
michael |
comment about failed optimization
libavcodec
|
Fri, 03 Nov 2006 14:28:30 +0000 |
michael |
move luma tc0 related init into asm
libavcodec
|
Fri, 03 Nov 2006 12:07:53 +0000 |
michael |
2 instructions less in h264_loop_filter_luma_mmx2()
libavcodec
|
Fri, 03 Nov 2006 11:07:35 +0000 |
michael |
preempt possible overflow
libavcodec
|
Fri, 03 Nov 2006 09:59:15 +0000 |
michael |
1 instruction less
libavcodec
|
Fri, 03 Nov 2006 02:03:56 +0000 |
michael |
optimize H264_DEBLOCK_P0_Q0
libavcodec
|
Sat, 07 Oct 2006 15:30:46 +0000 |
diego |
Change license headers to say 'FFmpeg' instead of 'this program/this library'
libavcodec
|
Mon, 28 Aug 2006 09:33:01 +0000 |
lorenm |
mmx implementation of deblocking strength decision.
libavcodec
|
Sun, 02 Jul 2006 22:01:31 +0000 |
mru |
remove redundant macro definitions
libavcodec
|
Wed, 22 Mar 2006 22:05:00 +0000 |
lorenm |
added mmx implementation of h264_chroma_mc2
libavcodec
|
Tue, 07 Mar 2006 22:45:56 +0000 |
lorenm |
h264_idct8_add_mmx
libavcodec
|
Tue, 07 Mar 2006 22:33:32 +0000 |
lorenm |
h264_idct_add only needs mmx1
libavcodec
|
Sun, 05 Mar 2006 19:02:35 +0000 |
lorenm |
gcc2.95 workaround
libavcodec
|
Sat, 04 Mar 2006 19:56:01 +0000 |
lorenm |
remove some useless instructions
libavcodec
|
Thu, 02 Mar 2006 08:21:08 +0000 |
lorenm |
4% faster h264_qpel_mc
libavcodec
|
Fri, 10 Feb 2006 06:55:25 +0000 |
lorenm |
h264: special case dc-only idct. ~1% faster overall
libavcodec
|
Thu, 09 Feb 2006 02:43:23 +0000 |
lorenm |
10l in 1.12
libavcodec
|
Tue, 07 Feb 2006 16:10:48 +0000 |
lorenm |
10l (x86_64)
libavcodec
|
Tue, 07 Feb 2006 10:52:25 +0000 |
lorenm |
18% faster put_h264_qpel16_mc[13]2_mmx2
libavcodec
|
Tue, 07 Feb 2006 07:35:03 +0000 |
lorenm |
11% faster put_h264_qpel16_v_lowpass_mmx2
libavcodec
|
Tue, 07 Feb 2006 06:25:14 +0000 |
lorenm |
15% faster put_h264_qpel16_hv_lowpass_mmx2
libavcodec
|
Thu, 12 Jan 2006 22:43:26 +0000 |
diego |
Update licensing information: The FSF changed postal address.
libavcodec
|
Mon, 09 Jan 2006 03:38:37 +0000 |
lorenm |
tweak h264_biweight
libavcodec
|
Fri, 30 Dec 2005 23:47:41 +0000 |
lorenm |
fix some potential arithmetic overflows in pred_direct_motion() and
libavcodec
|
Thu, 22 Dec 2005 01:10:11 +0000 |
diego |
COSMETICS: tabs --> spaces, some prettyprinting
libavcodec
|
Sat, 17 Dec 2005 18:14:38 +0000 |
diego |
COSMETICS: Remove all trailing whitespace.
libavcodec
|
Thu, 27 Oct 2005 06:45:29 +0000 |
lorenm |
faster h264_chroma_mc8_mmx, added h264_chroma_mc4_mmx.
libavcodec
|
Sun, 09 Oct 2005 23:38:52 +0000 |
lorenm |
h264 mmx weighted prediction. up to 3% overall speedup.
libavcodec
|
Sat, 03 Sep 2005 09:17:30 +0000 |
michael |
simplify (d&a) and (d&~a) calculation, hint by skal
libavcodec
|
Thu, 02 Jun 2005 20:45:35 +0000 |
lorenm |
sort H.264 mmx dsp functions into their own file
libavcodec
|