Commit Graph

46 Commits

Author SHA1 Message Date
Reimar Döffinger
b1c32fb5e5 Use "d" suffix for general-purpose registers used with movd.
This increases compatibilty with nasm and is also more consistent,
e.g. with h264_intrapred.asm and h264_chromamc.asm that already
do it that way.

Originally committed as revision 25042 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-05 10:10:16 +00:00
Ronald S. Bultje
3611c45ab7 Mark xmm registers as clobbered in simple loopfilter. Should fix the last
two VP8-related fate failures on Win64.

Originally committed as revision 24908 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-08-24 16:52:27 +00:00
Ronald S. Bultje
684d608bde Fix segfaults in VP8 SIMD code on Win64 (and FATE/win64 failures).
Originally committed as revision 24871 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-08-23 02:41:22 +00:00
Jason Garrett-Glaser
827d43bb9d VP8: move zeroing of luma DC block into the WHT
Lets us do the zeroing in asm instead of C.
Also makes it consistent with the way the regular iDCT code does it.

Originally committed as revision 24668 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-08-02 20:18:09 +00:00
Ronald S. Bultje
6341838f3c Use word-writing instead of dword-writing (with two cached but otherwise
unchanged bytes) in the horizontal simple loopfilter. This makes the filter
quite a bit faster in itself (~30 cycles less on Core1), probably mostly
because we don't need a complex 4x4 transpose, but only a simple byte
interleave. Also allows using pextrw on SSE4, which speeds up even more
(e.g. 25% faster on Core i7).

Originally committed as revision 24638 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-31 23:13:15 +00:00
Ronald S. Bultje
ab4d031889 Use pmaddubsw for the mbedge_filter (>=ssse3), 6-10 cycles faster.
Originally committed as revision 24514 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 21:18:19 +00:00
Jason Garrett-Glaser
e25dee602f VP8: Much faster SSE2 MC
5-10% faster or more on Phenom, Athlon 64, and some others.
Helps some on pre-SSSE3 Intel chips as well, but not as much.

Originally committed as revision 24513 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 19:34:00 +00:00
Ronald S. Bultje
48adb7e7a4 Enable no-loop memory/register saving for ssse3/sse4 also.
Originally committed as revision 24511 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 14:07:57 +00:00
Ronald S. Bultje
2a180c69ea Save a register (or regsize of stackspace for x86-32) for the no-loop
mbedge loopfilter functions, by re-using space that holds a variable
that we no longer need.

Originally committed as revision 24510 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 14:00:15 +00:00
Ronald S. Bultje
bcd4aa6498 Use nested ifs instead of &&, which appears to not work with %ifidn (i.e. this
construct was always enabled, even for <ssse3 versions).

Originally committed as revision 24509 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 13:56:51 +00:00
Ronald S. Bultje
2208053bd3 Split pextrw macro-spaghetti into several opt-specific macros, this will make
future new optimizations (imagine a sse5) much easier. Also fix a bug where
we used the direction (%2) rather than optimization (%1) to enable this, which
means it wasn't ever actually used...

Originally committed as revision 24507 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-26 13:50:59 +00:00
Ronald S. Bultje
6de5b7c6b8 Fix obvious bug in assignment. Somehow, the test vectors don't test this...
Originally committed as revision 24489 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-25 02:42:40 +00:00
Ronald S. Bultje
e3f7bf774c Fix SPLATB_REG mess. Used to be a if/elseif/elseif/elseif spaghetti, so this
splits it into small optimization-specific macros which are selected for each
DSP function. The advantage of this approach is that the sse4 functions now
use the ssse3 codepath also without needing an explicit sse4 codepath.

Originally committed as revision 24487 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-24 19:33:05 +00:00
Jason Garrett-Glaser
3ae079a3c8 VP8: optimize DC-only chroma case in the same way as luma.
Add MMX idct_dc_add4uv function for this case.
~40% faster chroma idct.

Originally committed as revision 24455 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-23 06:02:52 +00:00
Jason Garrett-Glaser
51c9156438 VP8 asm: cosmetics (spacing)
Originally committed as revision 24453 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-23 03:02:56 +00:00
Jason Garrett-Glaser
8a467b2d44 VP8: 30% faster idct_mb
Take shortcuts based on statistically common situations.
Add 4-at-a-time idct_dc function (mmx and sse2) since rows of 4 DC-only DCT
blocks are common.
TODO: tie this more directly into the MB mode, since the DC-level transform is
only used for non-splitmv blocks?

Originally committed as revision 24452 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-23 02:58:27 +00:00
Jason Garrett-Glaser
c25c776708 VP8: clear DCT blocks in iDCT instead of using clear_blocks.
~0.3% faster overall.

Originally committed as revision 24448 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-23 00:07:16 +00:00
Ronald S. Bultje
dc5eec8085 Use pextrw for SSE4 mbedge filter result writing, speedup 5-10cycles on
CPUs supporting it.

Originally committed as revision 24437 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-22 19:59:34 +00:00
Ronald S. Bultje
003243c3c2 Fix and enable horizontal >=SSE2 mbedge loopfilter.
Originally committed as revision 24409 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-22 01:35:26 +00:00
Jason Garrett-Glaser
8731dbd890 Eliminate one instruction in VP8 dc_add_sse4
Originally committed as revision 24405 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-21 22:41:37 +00:00
Jason Garrett-Glaser
7dd224a42d Various VP8 x86 deblocking speedups
SSSE3 versions, improve SSE2 versions a bit.
SSE2/SSSE3 mbedge h functions are currently broken, so explicitly disable them.

Originally committed as revision 24403 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-21 22:11:03 +00:00
Jason Garrett-Glaser
b8b231b5dc Make mmx VP8 WHT faster
Avoid pextrw, since it's slow on many older CPUs.
Now it doesn't require mmxext either.

Originally committed as revision 24397 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-21 20:51:01 +00:00
Ronald S. Bultje
e9e456d850 VP8 MBedge loopfilter MMX/MMX2/SSE2 functions for both luma (width=16)
and chroma (width=8).

Originally committed as revision 24378 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-20 22:58:56 +00:00
Ronald S. Bultje
268821e76e Chroma (width=8) inner loopfilter MMX/MMX2/SSE2 for VP8 decoder.
Originally committed as revision 24377 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-20 22:04:18 +00:00
Ronald S. Bultje
c60ed66dbe Revert r24339 (it causes fate failures on x86-64) - I'll figure out what's
wrong with it tomorrow or so, then re-submit.

Originally committed as revision 24341 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-19 23:57:09 +00:00
Ronald S. Bultje
1878f685c0 Implement chroma (width=8) inner loopfilter MMX/MMX2/SSE2 functions.
Originally committed as revision 24339 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-19 21:53:28 +00:00
Ronald S. Bultje
fb9bdf048c Be more efficient with registers or stack memory. Saves 8/16 bytes stack
for x86-32, or 2 MM registers on x86-64.

Originally committed as revision 24338 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-19 21:45:36 +00:00
Ronald S. Bultje
3facfc99da Change function prototypes for width=8 inner and mbedge loopfilter functions
so that it does both U and V planes at the same time. This will have speed
advantages when using SSE2 (or higher) optimizations, since we can do both
the U and V rows together in a single xmm register.

This also renames filter16 to filter16y and filter8 to filter8uv so that it's
more obvious what each function is used for.

Originally committed as revision 24337 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-19 21:18:04 +00:00
Ronald S. Bultje
819b2dd2b1 Attempt to fix x86-64 testsuite on fate.
Originally committed as revision 24275 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 21:35:30 +00:00
Ronald S. Bultje
6f323f1251 Remove duplicate define.
Originally committed as revision 24272 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 19:54:47 +00:00
Ronald S. Bultje
889b2c26ee Revert 24270, it contained some stuff that shouldn't have been in there.
Originally committed as revision 24271 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 19:54:25 +00:00
Ronald S. Bultje
2356a7834b Remove duplicate define.
Originally committed as revision 24270 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 19:42:32 +00:00
Ronald S. Bultje
ede1b9665a Give x86 r%d registers names, this will simplify implementation of the chroma
inner loopfilter, and it also allows us to save one register on x86-64/sse2.

Originally committed as revision 24269 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 19:38:10 +00:00
Ronald S. Bultje
526e831a46 Change return statement, the REP_RET is a mistake since the else case (x86-64,
sse2) doesn't actually loop, so REP_RET isn't necessary.

Originally committed as revision 24268 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-16 18:29:14 +00:00
Ronald S. Bultje
a711eb4829 VP8 H/V inner loopfilter MMX/MMXEXT/SSE2 optimizations.
Originally committed as revision 24250 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-15 23:02:34 +00:00
Ronald S. Bultje
f2a30bd840 Simple H/V loopfilter for VP8 in MMX, MMX2 and SSE2 (yay for yasm macros).
Originally committed as revision 24029 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-03 19:26:30 +00:00
Jason Garrett-Glaser
b06855f18a SSSE3 versions of vp8 width4 bilinear MC functions
Originally committed as revision 24013 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-03 00:48:12 +00:00
Jason Garrett-Glaser
dcc602d802 SSSE3 versions of width4 VP8 6-tap MC functions
Also make some small changes to saturation order of 4-tap SSSE3 MC to fix a
non-bitexactness bug.

Patch mostly by Eli Friedman <eli.friedman AT gmail DOT com>.

Originally committed as revision 23965 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-07-02 05:27:41 +00:00
Jason Garrett-Glaser
82a8d0f114 Use add instead of lshift in mmxext vp8 idct
Originally committed as revision 23891 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-29 17:23:17 +00:00
Ronald S. Bultje
565344e7e4 Remove unused macros (duplicates from the now-LGPL x86util.asm).
Originally committed as revision 23890 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-29 17:04:29 +00:00
Ronald S. Bultje
2dd2f71692 MMX idct_add for VP8.
Originally committed as revision 23886 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-29 14:43:11 +00:00
Jason Garrett-Glaser
004cda8e79 Add mmxext version of VP8 DC Hadamard transform
Originally committed as revision 23878 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-29 01:41:59 +00:00
Jason Garrett-Glaser
a912da761d Fix VP8 bilinear mc on x86_64
Originally committed as revision 23872 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-28 22:13:14 +00:00
Jason Garrett-Glaser
0fecad09fe Add x86 asm functions for VP8 put_pixels
Originally committed as revision 23858 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-28 19:14:40 +00:00
Jason Garrett-Glaser
a173aa8940 Add MMX, SSE2, SSSE3 asm for VP8 bilinear MC
Originally committed as revision 23857 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-06-28 18:56:24 +00:00
Jason Garrett-Glaser
0178d14fe5 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 02:01:45 +00:00