Commit Graph

324 Commits

Author SHA1 Message Date
James Almer
3b06208a57 x86/float_dsp: remove duplicated code from vector_dmul_scalar
Use the xm# and ym# aliases as they remain in sync with m# after a SWAP.
No actual changes to the assembly.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-19 14:21:51 +02:00
James Almer
76ed71a72b x86: move horizontal add macros to x86util
Also port relevant AVX2/XOP optimizations from x264 with permission
to relicense to LGPL from the corresponding authors

Signed-off-by: James Almer <jamrial@gmail.com>
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-17 14:15:09 +02:00
James Almer
11b36b1ee0 x86/float_dsp: unroll loop in vector_fmac_scalar
~6% faster SSE2 performance. AVX/FMA3 are unaffected.

Signed-off-by: James Almer <jamrial@gmail.com>
Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-16 18:36:52 +02:00
James Almer
3b808900af x86/float_dsp: use SWAP in vector_fmac_scalar Win64
The mova is unnecessary

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-16 15:46:21 +02:00
James Almer
2d9821a208 x86/cpu: check for OS support before enabling AVX2
AV_CPU_FLAG_AVX is enabled at this point only if there's OS support.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-25 17:56:43 +01:00
Matt Oliver
8236747511 Automatically change MANGLE() into named inline asm operands when direct symbol reference in inline asm are not supported.
This is part of the patch-set for intel C inline asm on windows support

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-18 23:39:30 +01:00
James Almer
7d7487e85c x86/float_dsp: add ff_vector_{fmul_add, fmac_scalar}_fma3
~7% faster than AVX

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-13 04:34:05 +01:00
Michael Niedermayer
4159f702a7 avutil/timer: Fix units for x86 after c708b54033
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-09 15:22:02 +01:00
James Almer
3f3d748cab x86: Move XOP emulation to x86util
We need the emulation to support the cases where the first
argument is the same as the fourth. To achieve this a fifth
argument working as a temporary may be needed.
Emulation that doesn't obey the original instruction semantics
can't be in x86inc.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-24 08:30:19 +01:00
Michael Niedermayer
bd8d73ea8b Merge remote-tracking branch 'qatar/master'
* qatar/master:
  x86: add detection for Bit Manipulation Instruction sets

Conflicts:
	libavutil/x86/cpu.c

See: 0bc3de19ff
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-23 22:52:58 +01:00
Michael Niedermayer
d9574069c1 Merge commit '1b932eb1508f550fac9e911923a0383efda53aa3'
* commit '1b932eb1508f550fac9e911923a0383efda53aa3':
  x86: add detection for FMA3 instruction set

Conflicts:
	configure
	libavutil/cpu.h
	libavutil/x86/cpu.c

See: a2af8eddab
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-23 22:43:08 +01:00
James Almer
d59fcdaff3 x86: add detection for Bit Manipulation Instruction sets
Based on x264 code

Signed-off-by: James Almer <jamrial@gmail.com>
2014-02-23 15:29:36 +01:00
James Almer
1b932eb150 x86: add detection for FMA3 instruction set
Based on x264 code

Signed-off-by: James Almer <jamrial@gmail.com>
2014-02-23 15:29:36 +01:00
James Almer
10b0161d78 x86: add missing XOP checks and macros
Signed-off-by: James Almer <jamrial@gmail.com>
2014-02-23 15:29:36 +01:00
James Almer
0bc3de19ff x86: add detection for Bit Manipulation Instruction sets
Based on x264 code

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-22 17:26:00 +01:00
James Almer
a2af8eddab x86: add detection for FMA3 instruction set
Based on x264 code

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-22 17:25:52 +01:00
Christophe Gisquet
996697e266 x86: float dsp: unroll SSE versions
vector_fmul and vector_fmac_scalar are guaranteed that they can process in
batch of 16 elements, but their SSE versions only does 8 at a time.

Therefore, unroll them a bit.
299 to 261c for 256 elements in vector_fmac_scalar on Arrandale/Win64.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
2014-02-20 14:18:05 +01:00
Christophe Gisquet
133b34207c x86: float dsp: unroll SSE versions
vector_fmul and vector_fmac_scalar are guaranteed that they can process in
batch of 16 elements, but their SSE versions only does 8 at a time.

Therefore, unroll them a bit.
299 to 261c for 256 elements in vector_fmac_scalar on Arrandale/Win64.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-15 18:54:21 +01:00
James Almer
23a8c63452 x86inc: Extend FMA_INSTR functionality
Support the cases where the first and last operand of
the XOP instruction are the same.

Also add vpmacsdql emulation.

Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-13 22:14:24 +01:00
James Almer
6c12b1de06 x86: add missing XOP checks and macros
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-11 03:46:52 +01:00
Loren Merritt
b7d0d10a1d x86inc: Speed up assembling with Yasm
Work around Yasm's inefficiency with handling large numbers of variables
in the global scope.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2014-01-26 18:40:08 +01:00
Loren Merritt
4d55fe7204 x86inc: speed up compilation with yasm
Work around yasm's inefficiency with handling large numbers of variables
in the global scope.
2014-01-18 01:19:16 +01:00
Michael Niedermayer
c3814ab654 rename new lls code to lls2 to avoid conflict with the old which has a different ABI
also remove failed attempt at a compatibility layer, the code simply cannot work

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-11-17 16:41:08 +01:00
Michael Niedermayer
bbe66ef912 avutil: rename lls to lls2
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-11-17 16:30:23 +01:00
Michael Niedermayer
a665704402 Merge commit '4d6ee0725553a43ba88d6f8327ebcf8f1c5ae8d4'
* commit '4d6ee0725553a43ba88d6f8327ebcf8f1c5ae8d4':
  libavutil: x86: Add AVX2 capable CPU detection.

Conflicts:
	libavutil/cpu.c
	libavutil/cpu.h
	libavutil/x86/cpu.c

See: 865b70bc5d
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-26 02:36:36 +02:00
Kieran Kunhya
865b70bc5d Add AVX2 capable CPU detection. Patch based on x264's AVX2 detection
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-26 02:34:22 +02:00
Kieran Kunhya
4d6ee07255 libavutil: x86: Add AVX2 capable CPU detection.
Patch based on x264's AVX2 detection

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-25 19:36:55 +01:00
Michael Niedermayer
f9bef2bec9 Merge remote-tracking branch 'qatar/master'
* qatar/master:
  x86: more AVX2 framework

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-14 16:13:57 +02:00
Michael Niedermayer
e3e0e3d0c9 Merge commit 'c6908d6b4b377a04a5d055ba874bdbcf06c80497'
* commit 'c6908d6b4b377a04a5d055ba874bdbcf06c80497':
  x86inc: FMA3/4 Support

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-14 16:06:22 +02:00
Michael Niedermayer
9ac124c889 Merge commit '206895708ea2b464755d340e44501daf9a07c310'
* commit '206895708ea2b464755d340e44501daf9a07c310':
  x86inc: Remove our FMA4 support

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-14 15:54:23 +02:00
Michael Niedermayer
12e4493f9c Merge commit 'c108ba0175d4fc3a3253a8b0f782fbfb96ba5098'
* commit 'c108ba0175d4fc3a3253a8b0f782fbfb96ba5098':
  x86inc: Use VEX-encoded instructions in AVX functions

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-14 15:48:34 +02:00
Jason Garrett-Glaser
a3fabc6cb3 x86: more AVX2 framework
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-14 12:41:56 +01:00
Jason Garrett-Glaser
c6908d6b4b x86inc: FMA3/4 Support
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-14 12:41:54 +01:00
Derek Buitenhuis
206895708e x86inc: Remove our FMA4 support
This is so we can sync to x264's version of FMA4 support.

This partialy reverts commit 79687079a9.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-14 12:39:29 +01:00
Henrik Gramner
c108ba0175 x86inc: Use VEX-encoded instructions in AVX functions
Automatically use VEX-encoding in AVX/AVX2/XOP/FMA3/FMA4
functions for all instructions that exists in a VEX-encoded
version.

This change makes it easier to extend existing code to use AVX2.

Also add support for AVX emulation of a few instructions that
were missing before.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-14 12:36:11 +01:00
Michael Niedermayer
31d0d35560 Merge remote-tracking branch 'qatar/master'
* qatar/master:
  x86inc: Remove .rodata kludges

Conflicts:
	libavutil/x86/x86inc.asm

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-09 14:29:42 +02:00
Henrik Gramner
ad7d7d4f6a x86inc: Remove .rodata kludges
The Mach-O bug was fixed in yasm 0.8.0 and we don't
support versions that old anymore.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-09 07:44:30 -04:00
Michael Niedermayer
19c3890819 Merge commit '3e2fa991db7ef172579422accd61624d52777e5a'
* commit '3e2fa991db7ef172579422accd61624d52777e5a':
  x86inc: remove misaligned cpu flag

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 12:02:21 +02:00
Michael Niedermayer
31d9aa6b2e Merge commit '71155665414b551ad350622d5abed20e58371fbf'
* commit '71155665414b551ad350622d5abed20e58371fbf':
  x86inc: various minor backports from x264

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:57:39 +02:00
Michael Niedermayer
3f965ab95d Merge commit '47f9d7ce5493e119e09d1227d017414feaaf8d97'
* commit '47f9d7ce5493e119e09d1227d017414feaaf8d97':
  x86inc: Check for __OUTPUT_FORMAT__ having a value of "x64"

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:37:22 +02:00
Michael Niedermayer
1f17619fe4 Merge commit 'bbe4a6db44f0b55b424a5cc9d3e89cd88e250450'
* commit 'bbe4a6db44f0b55b424a5cc9d3e89cd88e250450':
  x86inc: Utilize the shadow space on 64-bit Windows

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:23:00 +02:00
Michael Niedermayer
17d9c7c208 Merge commit '3fb78e99a04d0ed8db834d813d933eb86c37142a'
* commit '3fb78e99a04d0ed8db834d813d933eb86c37142a':
  x86inc: create xm# and ym#, analagous to m#

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:15:17 +02:00
Michael Niedermayer
3352fdb292 Merge commit '49ebe3f9fe02174ae7e14548001fd146ed375cc2'
* commit '49ebe3f9fe02174ae7e14548001fd146ed375cc2':
  x86inc: fix some corner cases of SWAP

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:07:03 +02:00
Michael Niedermayer
006c0fcfea Merge commit '63f0d623100bdb0c6081456127f4b6713e83d3db'
* commit '63f0d623100bdb0c6081456127f4b6713e83d3db':
  x86inc: Use SSE instead of SSE2 for copying data

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 11:01:40 +02:00
Michael Niedermayer
faafffaf82 Merge commit 'ad76e6e7e193b98e7335156422d35467816f9ef1'
* commit 'ad76e6e7e193b98e7335156422d35467816f9ef1':
  x86inc: Set ELF hidden visibility for global constants

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 10:52:51 +02:00
Michael Niedermayer
c1488fab3d Merge commit '25cb0c1a1e66edacc1667acf6818f524c0997f10'
* commit '25cb0c1a1e66edacc1667acf6818f524c0997f10':
  x86inc: activate REP_RET automatically

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 10:27:30 +02:00
Henrik Gramner
3e2fa991db x86inc: remove misaligned cpu flag
Prevents a crash if the misaligned exception mask bit is
cleared for some reason.

Misaligned SSE functions are only used on AMD Phenom CPUs
and the benefit is miniscule. They also require modifying
the MXCSR control register and by removing those functions
we can get rid of that complexity altogether.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-07 06:27:38 -04:00
Jason Garrett-Glaser
7115566541 x86inc: various minor backports from x264
Small backports that sneaked into other asm commits in x264.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-07 06:27:22 -04:00
Derek Buitenhuis
47f9d7ce54 x86inc: Check for __OUTPUT_FORMAT__ having a value of "x64"
This is also a valid value for WIN64.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-07 06:27:08 -04:00
Henrik Gramner
bbe4a6db44 x86inc: Utilize the shadow space on 64-bit Windows
Store XMM6 and XMM7 in the shadow space in functions that
clobbers them. This way we don't have to adjust the stack
pointer as often, reducing the number of instructions as
well as code size.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-10-07 06:25:35 -04:00