Some code was initializing some xmm registers in one asm block and using them
in the following block, assuming they wouldn't be changed in between blocks.
Originally committed as revision 25568 to svn://svn.ffmpeg.org/ffmpeg/trunk
prediction (plus some with different rounding for svq3/rv40). Speedup (for
SSSE3) about ~6-fold, 3.6% faster overall with cathedral sample.
Originally committed as revision 25361 to svn://svn.ffmpeg.org/ffmpeg/trunk
Fixes compilation with clang's builtin assembler
Patch by İsmail Dönmez, ismail at namtrac dot org
Originally committed as revision 25331 to svn://svn.ffmpeg.org/ffmpeg/trunk
inline asm works for gcc-3.x also (hopefully). Should fix gcc-3.x FATE
breakage after r25254.
Originally committed as revision 25262 to svn://svn.ffmpeg.org/ffmpeg/trunk
increase to e.g. vc1, snow and mpeg decoding.
Patch by Eli Friedman <eli dot friedman gmail com>.
Originally committed as revision 25259 to svn://svn.ffmpeg.org/ffmpeg/trunk
from memory locations/offsets depending on b_idx plus constants, rather than
having gcc do this. This saves several lea calls and together saves about
10 cycles in h264_loop_filter_strength_mmx2().
Originally committed as revision 25256 to svn://svn.ffmpeg.org/ffmpeg/trunk
a pxor, or remove the instruction alltogether. Altogether, this saves 1
instruction.
Originally committed as revision 25255 to svn://svn.ffmpeg.org/ffmpeg/trunk
This has no measurable speed effect because the surrounding code doesn't
take advantage of this yet.
Originally committed as revision 25254 to svn://svn.ffmpeg.org/ffmpeg/trunk
of the d_idx variable and therefore allows for future optimizations. No speed
difference by this commit itself.
Originally committed as revision 25253 to svn://svn.ffmpeg.org/ffmpeg/trunk
inlining various constants within the loop code. 20 cycles faster on
cathedral sample.
Originally committed as revision 25252 to svn://svn.ffmpeg.org/ffmpeg/trunk
inlines scan8[] and removes loop setup. 15% faster, 0.4% overall.
See "[PATCH] unroll loop in h264_idct_add8_sse2()" thread on ML.
Originally committed as revision 25172 to svn://svn.ffmpeg.org/ffmpeg/trunk
code directly also and remove loop setup. 20% faster in function, 0.8% overall.
See "[PATCH] unroll loop in h264_idct_add8_sse2()" thread on ML.
Originally committed as revision 25171 to svn://svn.ffmpeg.org/ffmpeg/trunk
which will hopefully solve the Win64/FATE failures caused by these functions.
Originally committed as revision 25137 to svn://svn.ffmpeg.org/ffmpeg/trunk
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
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
format), LGPL'ed with permission from Jason and Loren. This includes mmx2
code, so remove inline asm from h264dsp_mmx.c accordingly.
Originally committed as revision 25031 to svn://svn.ffmpeg.org/ffmpeg/trunk
biweight code to sse2/ssse3; add sse2 weight code; and use that same code to
create mmx2 functions also, so that the inline asm in h264dsp_mmx.c can be
removed. OK'ed by Jason on IRC.
Originally committed as revision 25019 to svn://svn.ffmpeg.org/ffmpeg/trunk
still #included in dsputil_mmx.c and is part of DSPContext, and h264dsp_mmx.c,
which represents H264DSPContext and is now compiled on its own.
Originally committed as revision 25018 to svn://svn.ffmpeg.org/ffmpeg/trunk
into its own file, it doesn't belong in h264dsp_mmx.c (much less so in
dsputil_mmx.c).
Originally committed as revision 24990 to svn://svn.ffmpeg.org/ffmpeg/trunk
The stride can be negative and must be sign extended before being
used in pointer arithmetic.
Originally committed as revision 24926 to svn://svn.ffmpeg.org/ffmpeg/trunk
This is to avoid split asm sections that attempt to preserve some
registers between sections.
Originally committed as revision 24869 to svn://svn.ffmpeg.org/ffmpeg/trunk
Grab from the bitstream in 16-bit chunks instead of 8-bit chunks.
TODO: grab in 32-bit chunks on 64-bit systems.
Originally committed as revision 24783 to svn://svn.ffmpeg.org/ffmpeg/trunk
Many H.264 derivatives, like RV40 and VP8, use the H.264 prediction functions
but not the weight/loopfilter functions.
This should reduce the size of builds with one of these derivatives but without
H.264 decoding itself.
Originally committed as revision 24741 to svn://svn.ffmpeg.org/ffmpeg/trunk
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
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
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
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
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
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
This is a lot more reliable to get cmov rather than trying to trick gcc into
generating it, useful since it's 2% faster overall.
Patch by Eli Friedman <eli.friedman at gmail>
Originally committed as revision 24471 to svn://svn.ffmpeg.org/ffmpeg/trunk
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
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
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
Should fix compilation with icc and should help prevent any future duplicates
Originally committed as revision 24380 to svn://svn.ffmpeg.org/ffmpeg/trunk
regular MMX code. Examples of this are the Core1 CPU. Instead, set a new flag,
FF_MM_SSE2/3SLOW, which can be checked for particular SSE2/3 functions that
have been checked specifically on such CPUs and are actually faster than
their MMX counterparts.
In addition, use this flag to enable particular VP8 and LPC SSE2 functions
that are faster than their MMX counterparts.
Based on a patch by Loren Merritt <lorenm AT u washington edu>.
Originally committed as revision 24340 to svn://svn.ffmpeg.org/ffmpeg/trunk
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
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
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
Whose idea was it to have a CPU that didn't SIGILL on an invalid instruction?
Originally committed as revision 23927 to svn://svn.ffmpeg.org/ffmpeg/trunk
- 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
Modify the asm accordingly.
GLOBAL is now no longoer necessary for PIC-compliant loads.
Originally committed as revision 23739 to svn://svn.ffmpeg.org/ffmpeg/trunk