coffee/libyuv
1
0
Fork 0
mirror of https://chromium.googlesource.com/libyuv/libyuv synced 2025-12-07 17:26:49 +08:00
Code Issues Packages Projects Releases Wiki Activity
104 Commits 3 Branches 0 Tags BSD-3-Clause 60 MiB
Commit Graph

84 Commits

Author SHA1 Message Date
fbarchard@google.com
e7ae4e1062 nv12 stores ordered for better core2 performance. 221->211ms 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/330001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@108 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-15 20:01:43 +00:00
fbarchard@google.com
373cdbdc58 reorder stores for FastConvertYUVToABGRRow_SSSE3 and FastConvertYUVToBGRARow_SSSE3. ReverseRow_SSE2. cpu detect allow environment variable override set LIBYUV_DISABLE_SSSE3=1 set LIBYUV_DISABLE_SSE2=1. Reorder stores in rotate for core2 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/317010

git-svn-id: http://libyuv.googlecode.com/svn/trunk@107 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-14 21:10:07 +00:00
fbarchard@google.com
8b9759c4a7 I400ToARGB use 8.8 fixed point to avoid a shift. gcc generate constants to avoid fpic performance stall 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/322013

git-svn-id: http://libyuv.googlecode.com/svn/trunk@106 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-14 04:17:39 +00:00
fbarchard@google.com
3fe369661a FastConvertYUVToARGBRow_SSSE3 use 2 pack and then 2 stores, which works better on Core2 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/323007

git-svn-id: http://libyuv.googlecode.com/svn/trunk@103 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-13 02:49:22 +00:00
fbarchard@google.com
19a248ab63 fix for underscore on gcc 32 bit for linux 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/325001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@102 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-12 20:21:52 +00:00
fbarchard@google.com
fe5ff7ed54 clang on linux prefers no args for rep movsl/stosl. ifdef for C that removes namespace. extern c allows c++ to be used to compile, including with namespace, but C program can link libyuv 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/320004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@101 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-10 07:45:58 +00:00
fbarchard@google.com
1d61cd91bb mingw behaves like apple as far as labels go 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/322002

git-svn-id: http://libyuv.googlecode.com/svn/trunk@99 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-08 19:57:48 +00:00
frkoenig@google.com
e0bf6f0934 Fix build for OSX. 
Clang defines __APPLE__, not OSX, use that
for platform specific code.

Removed spurious volatile that was giving a warning
when compiling for OSX.
Review URL: http://webrtc-codereview.appspot.com/305005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@98 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-07 01:54:13 +00:00
fbarchard@google.com
e5ffa14f31 CopyRow_X86 for gcc 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/300007

git-svn-id: http://libyuv.googlecode.com/svn/trunk@97 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-05 22:45:32 +00:00
fbarchard@google.com
c2b74366ec CopyRows_X86 had parameters wrong 
TESTED=videotool save to file
 NEEDS_QA=false
Review URL: http://webrtc-codereview.appspot.com/304006

git-svn-id: http://libyuv.googlecode.com/svn/trunk@96 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-02 23:19:27 +00:00
fbarchard@google.com
ca26f860b9 ARGBRect use SetRows32 to do full image. 744 ms -> 288 ms on Linux, 688 ms -> 277 ms on Mac. Started ARGBCopy function. 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/290006

git-svn-id: http://libyuv.googlecode.com/svn/trunk@95 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-12-01 23:29:14 +00:00
fbarchard@google.com
120b8d7ee7 ARGBRect optimized for visual c that will do rep stosd. Fix x/y calculation. 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/292008

git-svn-id: http://libyuv.googlecode.com/svn/trunk@94 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-28 19:28:41 +00:00
fbarchard@google.com
3e8a577bb3 add ARGBRect, fix for 444 support in ConvertToI420 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/289003

git-svn-id: http://libyuv.googlecode.com/svn/trunk@93 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-23 18:58:38 +00:00
fbarchard@google.com
10f5556a9f I444 fourcc and ConvertToI420 support 
BUG=none
TEST=untested
Review URL: http://webrtc-codereview.appspot.com/290001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@92 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-23 00:01:24 +00:00
fbarchard@google.com
1c5136d069 use IS_ALIGNED for width and stride to avoid mod that generates 6 instructions 
BUG=none
TEST=disassemble to confirm smaller/simplier alignment checks
Review URL: http://webrtc-codereview.appspot.com/287001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@91 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-22 18:15:17 +00:00
fbarchard@google.com
3c00cf04f3 inline cpuid for better branch prediction 
BUG=none
TEST=disassembly shows 6 instructions (due to auto init)
Review URL: http://webrtc-codereview.appspot.com/285001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@90 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-22 00:40:39 +00:00
fbarchard@google.com
12d048335d I420Mirror rewrite, I420Copy movdqa, I444To420, I420To422 
Review URL: http://webrtc-codereview.appspot.com/267025

git-svn-id: http://libyuv.googlecode.com/svn/trunk@89 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-21 23:54:38 +00:00
fbarchard@google.com
1616c315e0 I420ToYUY2 SSE2 version. 5.5x faster 
BUG=none
TEST=LmiVideoFrameTest.ConvertToYUY2Buffer
Review URL: http://webrtc-codereview.appspot.com/283001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@88 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-20 18:34:15 +00:00
frkoenig@google.com
5e56498d4b Fix FastConvertYUVToABGRRow_SSSE3 for posix. 
Extraneous lines in assembly code.
Review URL: http://webrtc-codereview.appspot.com/279013

git-svn-id: http://libyuv.googlecode.com/svn/trunk@87 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-19 01:43:44 +00:00
frkoenig@google.com
93d003f874 YUVToARGB, YUVToBGRA,YUVToABGR optimized. 
Review URL: http://webrtc-codereview.appspot.com/267022

git-svn-id: http://libyuv.googlecode.com/svn/trunk@86 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-19 01:39:57 +00:00
fbarchard@google.com
bc8f28eb52 remove row table, make C use math that mimics SIMD for exactness. Also 2x faster than old code which mimiced old SIMD via tables. 9000 ms instead of 20000 ms 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/267020

git-svn-id: http://libyuv.googlecode.com/svn/trunk@85 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-18 21:03:54 +00:00
fbarchard@google.com
15c3d45cbe Change WIN32 to _M_IX86 and use YUV_DISABLE_ASM consistently 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/277008

git-svn-id: http://libyuv.googlecode.com/svn/trunk@84 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-17 22:13:17 +00:00
fbarchard@google.com
1c2d8be1e1 port yuv to rgb to mac 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/269017

git-svn-id: http://libyuv.googlecode.com/svn/trunk@83 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-17 21:57:54 +00:00
fbarchard@google.com
8219d27a19 i422 to i420 sse2 
BUG=none
TEST=media_unittest
Review URL: http://webrtc-codereview.appspot.com/269016

git-svn-id: http://libyuv.googlecode.com/svn/trunk@82 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-16 19:06:31 +00:00
fbarchard@google.com
c59000bf0f osx 64 bit does not want _ in variable names 
BUG=none
TEST=build with m64
Review URL: http://webrtc-codereview.appspot.com/268011

git-svn-id: http://libyuv.googlecode.com/svn/trunk@81 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-15 22:27:45 +00:00
fbarchard@google.com
228bdc24e4 port yuv to rgb ssse3 to gcc 
BUG=none
TEST=media_unittest
Review URL: http://webrtc-codereview.appspot.com/269015

git-svn-id: http://libyuv.googlecode.com/svn/trunk@80 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-15 21:58:26 +00:00
fbarchard@google.com
4cf70bd6db compare SumSquareError_SSE2 ported to gcc 
BUG=none
TEST=media_unittest
Review URL: http://webrtc-codereview.appspot.com/279005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@79 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-15 18:44:31 +00:00
fbarchard@google.com
2430e04e0a fix internal compiler error on mac by removing const from vectors 
BUG=none
TEST=compile on mac
Review URL: http://webrtc-codereview.appspot.com/266013

git-svn-id: http://libyuv.googlecode.com/svn/trunk@78 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-11 21:57:06 +00:00
fbarchard@google.com
eaedc1d727 remove mmx functions 
BUG=none
TEST=builds
Review URL: http://webrtc-codereview.appspot.com/269010

git-svn-id: http://libyuv.googlecode.com/svn/trunk@77 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-11 18:41:47 +00:00
frkoenig@google.com
c82af4a59c Order includes. 
Use the following order.
local
system
libyuv
Review URL: http://webrtc-codereview.appspot.com/270007

git-svn-id: http://libyuv.googlecode.com/svn/trunk@76 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-11 00:54:34 +00:00
fbarchard@google.com
d4ff5b42e6 document ConvertToI420 parameters 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/267008

git-svn-id: http://libyuv.googlecode.com/svn/trunk@75 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-11 00:06:01 +00:00
mikhal@webrtc.org
1b06484b4a libyuv: clean up: 
1. Removing libyuv:: when not needed
2. Test clean-up 
Review URL: http://webrtc-codereview.appspot.com/279002

git-svn-id: http://libyuv.googlecode.com/svn/trunk@74 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 23:58:32 +00:00
fbarchard@google.com
f7a5048f54 align asm new line to column 48 
BUG=none
TEST=builds
Review URL: http://webrtc-codereview.appspot.com/268008

git-svn-id: http://libyuv.googlecode.com/svn/trunk@73 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 22:41:20 +00:00
frkoenig@google.com
2cb934c624 PSNR and SSIM plane comparision functions. 
Code sourced from libvpx.
  http://www.webmproject.org/code/#libvpx_the_vp8_codec_sdk
Review URL: http://webrtc-codereview.appspot.com/267004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@72 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 20:52:24 +00:00
fbarchard@google.com
6334808d9d remove _ from simd in row 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/270006

git-svn-id: http://libyuv.googlecode.com/svn/trunk@71 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 20:48:35 +00:00
fbarchard@google.com
75988ad3a8 remove align16 macro 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/266012

git-svn-id: http://libyuv.googlecode.com/svn/trunk@70 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 20:28:35 +00:00
fbarchard@google.com
706fc412ad move code from table to common 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/269009

git-svn-id: http://libyuv.googlecode.com/svn/trunk@69 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 20:13:57 +00:00
fbarchard@google.com
124b76e5c9 add a comparison function 
BUG=none
TEST=none
Review URL: http://webrtc-codereview.appspot.com/270005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@68 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 20:13:42 +00:00
mikhal@webrtc.org
b744e3734a libyuv: Updating general functions: 
1. Deleting Crop and CropPad functions
2. Moving I420Mirror to planar functions
3. deleting general.h and general.cc
Review URL: http://webrtc-codereview.appspot.com/269008

git-svn-id: http://libyuv.googlecode.com/svn/trunk@67 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 18:44:37 +00:00
fbarchard@google.com
d93d4486eb row functions for windows use ssse3 for yuv to rgb. mac use sse3 for rgb to yuv 
Review URL: http://webrtc-codereview.appspot.com/267007

git-svn-id: http://libyuv.googlecode.com/svn/trunk@66 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-10 18:26:20 +00:00
fbarchard@google.com
82ba1b77ae bayer off by one on width 
BUG=none
TEST=media_unittest
Review URL: http://webrtc-codereview.appspot.com/276003

git-svn-id: http://libyuv.googlecode.com/svn/trunk@65 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-08 19:35:52 +00:00
frkoenig@google.com
86e9487d0d Fix for InitCpuFlags for android. 
Review URL: http://webrtc-codereview.appspot.com/277001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@64 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-08 00:12:33 +00:00
frkoenig@google.com
e518542da5 ifdef platform specific code. 
Not all posix platforms have SSSE3 functionality.
Review URL: http://webrtc-codereview.appspot.com/276001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@62 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-07 23:07:57 +00:00
fbarchard@google.com
b61497636a ARGB to and from I420 ported to x64 
BUG=none
TEST=media_unittests
Review URL: http://webrtc-codereview.appspot.com/266003

git-svn-id: http://libyuv.googlecode.com/svn/trunk@61 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-07 21:58:52 +00:00
frkoenig@google.com
755de365c4 Change add + shift syntax for arm assembly. 
Short syntax is not supported by all compilers.
Review URL: http://webrtc-codereview.appspot.com/265001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@60 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-03 23:33:01 +00:00
frkoenig@google.com
f7e74a1a6f Move neon rotate code from straight assembly to inline. 
Allow assemblers with a slightly different syntax to use
the optimized neon routines.

Removed extra constraints on the calling of the optimized
routines.  All neon routines can load unaligned and handle
odd widths.

Align allocated buffers in rotate_test.cc

Add neon rotate file to gyp file for arm targets.
Review URL: http://webrtc-codereview.appspot.com/253007

git-svn-id: http://libyuv.googlecode.com/svn/trunk@59 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-03 22:41:59 +00:00
frkoenig@google.com
7aa6f06e70 Avoid using Q4-Q7 registers for NEON. 
Was not paying attention to which registers were for scratch.
Avoid the need to preserve registers by not using registers
in the Q4-Q7 range.

Fix ScaleDown2Int_NEON by changing how rounding was applied.

ScaleDownRow4 changed to process 4 output pixels per loop.

No need to push/pop registers for UV Transpose, removed
functions.

Fix for CPU Flag for scale_test.cc to turn on/off optimizations
for timing.
Review URL: http://webrtc-codereview.appspot.com/259002

git-svn-id: http://libyuv.googlecode.com/svn/trunk@58 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-03 00:54:50 +00:00
fbarchard@google.com
799796b272 avoid xmm7 in formatconversion 
BUG=none
TEST=build/run on linux
Review URL: http://webrtc-codereview.appspot.com/255010

git-svn-id: http://libyuv.googlecode.com/svn/trunk@57 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-02 23:06:50 +00:00
fbarchard@google.com
d14936a0af scale avoid xmm6 and xmm7 usage 
BUG=none
TEST=scale test in talk
Review URL: http://webrtc-codereview.appspot.com/256009

git-svn-id: http://libyuv.googlecode.com/svn/trunk@56 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-02 23:06:40 +00:00
fbarchard@google.com
4562b62caa Reduced usage of xmm6 and xmm7 which need to be saved on win64. 
BUG=none
TEST=tested with talk unittests
Review URL: http://webrtc-codereview.appspot.com/261003

git-svn-id: http://libyuv.googlecode.com/svn/trunk@55 16f28f9a-4ce2-e073-06de-1de4eb20be90
 2011-11-02 23:06:27 +00:00