262 Commits

Author SHA1 Message Date
fbarchard@google.com
cf925c50bc Make Yvu vs Yuv use same code and structure but pass in a different version of the matrix
BUG=396
TESTED=ncval
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/34979004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1266 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-02-06 00:29:08 +00:00
fbarchard@google.com
1663996c52 Remove ifdef __SSE2__ and native client ifdef for r14 in register usage declarations.
BUG=395
TESTED=gcc build with nacl
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/34149004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1264 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-02-05 23:09:15 +00:00
fbarchard@google.com
3cb485533a NaCL port of YToARGB for AVX2
BUG=393
TESTED=d:\src\nacl_sdk\pepper_canary\tools\ncval.exe newlib/Release/nacltest_x86_32.nexe
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/39829004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1263 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-02-05 21:32:15 +00:00
fbarchard@google.com
baafc97d6b port YToARGB AVX2 to GCC
BUG=393
TESTED=untested
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/39819004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1262 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-02-05 20:17:27 +00:00
fbarchard@google.com
c4e032c543 change Y multiplier and bias to compensate for 257/256 which makes YToARGB exactly match float math.
Histogram Before
hist            -3      -2      -1      0       1       2       3
red             0       0       1809408 13140736        1827072 0       0
green           0       0       1679912 13471329        1625975 0       0
blue            168448  994816  1876480 10655488        1893376 1006336 182272
Histogram After
hist            -3      -2      -1      0       1       2       3
red             0       0       558848  15632128        586240  0       0
green           0       0       209907  16350588        216721  0       0
blue            14848   642816  1989376 11363328        2053120 695040  18688
BUG=394
TESTED=more stringent luma tests
R=brucedawson@google.com

Review URL: https://webrtc-codereview.appspot.com/38859004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1259 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-02-04 19:45:26 +00:00
fbarchard@google.com
29db9b0b89 C version of YToARGB with ubias removed to produce consistent luma ramp.
BUG=392
TESTED=TestGreyYUV
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/35869004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1251 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-01-26 23:07:46 +00:00
fbarchard@google.com
080a316492 port yuv chroma improvements to gcc. YUV to RGB is more accurate using a negative matrix. 2% slower but half as much error.
BUG=324
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/41629004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1249 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-01-26 04:35:51 +00:00
fbarchard@google.com
b089593610 xmm4 is unused - remove from NV21
BUG=324
TESTED=untested
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/40489004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1243 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-01-22 18:17:44 +00:00
fbarchard@google.com
e7873910df port YUV luma accuracy to posix
BUG=324
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/33049004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1236 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-01-21 00:36:30 +00:00
fbarchard@google.com
b2a6af1be6 Change rectangle low level functions to use more conventional row functions including 'any' variations. Previously the yuv function SetPlane stored 32 bit values. Now a more conventional memset() style function is used for YUV that stores bytes. On Haswell a rep stosb is used for YUV. Overall benefit of this CL is improved performance for 'any' width, and simpler row assembly instead of full image assembly. Previously ARGBRect used a low level function that supported a rectangle in assembly. Now it uses a row function, and relies on row coalesce to combine into a single low level call.
BUG=371
TESTED=untested
R=brucedawson@google.com, harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/35689004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1222 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-01-12 03:58:24 +00:00
fbarchard@google.com
966233e5eb Remove sub 16 from yuv conversions and change bias to include it.
BUG=388
TESTED=out\release\libyuv_unittest --gtest_catch_exceptions=0 --gtest_filter=*420ToARGB_Opt  | sortms
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/34609004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1216 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-31 01:07:02 +00:00
fbarchard@google.com
8723fc1109 Syntax fix for change 24 bit conversions to use single asm block instead of 2, but with memory counter
BUG=389, 378
TESTED=out\release\libyuv_unittest --gtest_catch_exceptions=0 --gtest_filter=*420ToRGB24_Opt | sortms
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/39399004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1215 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-30 22:26:10 +00:00
fbarchard@google.com
16338ba85f Change 24 bit conversions to use single asm block instead of 2, but with memory counter
BUG=389,378
TESTED=out\release\libyuv_unittest --gtest_catch_exceptions=0 --gtest_filter=*420ToRGB24_Opt | sortms
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/29359004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1214 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-30 21:35:35 +00:00
fbarchard@google.com
284d6bdf49 Port I422ToBGRA from Windows version that does 16 pixels at a time, for performance improvement.
BUG=386
TESTED=nacl build
R=brucedawson@google.com, harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/36549004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1207 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-16 23:56:04 +00:00
fbarchard@google.com
8b55212c83 Make vextop take the register selector parameter to access the upper portion of the avx registers.
BUG=269
TESTED=nacl
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/37399004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1205 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-16 00:30:51 +00:00
fbarchard@google.com
7892ea1fe1 Fix for ARGBToUV on AVX2
BUG=269
TESTED=local testing
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/33669004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1202 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-15 18:59:23 +00:00
fbarchard@google.com
f5f5d15dcd Fix register order for ARGBToUV_AVX2
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/29249004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1200 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-15 18:07:09 +00:00
fbarchard@google.com
11c3015712 Fix for I422ToARGB AVX2
BUG=269
TESTED=untested

Review URL: https://webrtc-codereview.appspot.com/32809004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1199 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-13 19:22:09 +00:00
fbarchard@google.com
ada2a3eb12 Fix for ARGBToY on AVX
BUG=269
TESTED=local build on osx
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/29229005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1198 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-13 01:44:33 +00:00
fbarchard@google.com
b0abc62c21 Fix for UYVYToI422 AVX2 version
BUG=269
TESTED=untested
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/32329004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1197 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-13 00:15:11 +00:00
fbarchard@google.com
a9734a4492 ARGBMirror for AVX had wrong loop counting. This fixes it to match windows, and reenables the function.
BUG=269
TESTED=try bots
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/33639004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1196 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-12 22:43:55 +00:00
fbarchard@google.com
08daa3e22b Disable AVX2 code that fails on GCC unittests until issues can be resolved.
BUG=269
TESTED=sde-external-7.8.0-2014-10-02-mac/sde -ast -hsw -- out/Release/libyuv_unittest

Review URL: https://webrtc-codereview.appspot.com/29219004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1195 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-12 19:30:15 +00:00
fbarchard@google.com
233a931cb6 Port ARGBToUV to AVX2.
BUG=269
TESTED=ncval
R=brucedawson@google.com, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/35449004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1194 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-10 22:48:58 +00:00
fbarchard@google.com
e0bb4c26e2 Interpolate Row ported to AVX2 GCC/NaCL.
BUG=269
TESTED=nacl build
R=brucedawson@google.com, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/25329004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1193 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-09 22:21:53 +00:00
fbarchard@google.com
044938f485 convert ARGB to UV for SSSE3 use single asm block.
BUG=378
TESTED=nacl build
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/28179004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1191 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-06 19:23:12 +00:00
fbarchard@google.com
540e8af80c remove add 16 from ARGBToYJ and add rounding, for consistency with Windows version. row.h header macros sorted alphabetically.
BUG=269
TESTED=untested
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/32579005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1185 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-02 22:37:47 +00:00
fbarchard@google.com
b036cf700b ARGBToYRow_AVX2 and ARGBToYJRow_AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=brucedawson@google.com, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/30299004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1184 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-02 22:00:08 +00:00
fbarchard@google.com
d0bfd10147 I422ToRGBARow_AVX2 ported to GCC.
BUG=269
TESTED=nacl build
R=brucedawson@google.com

Review URL: https://webrtc-codereview.appspot.com/32259004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1183 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-02 18:59:33 +00:00
fbarchard@google.com
702e237d5f I422ToABGR_AVX2 port from Visual C to GCC/NaCL.
BUG=269
TESTED=builds with nacl compiler.
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/33449004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1182 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-02 03:54:08 +00:00
fbarchard@google.com
0c472f9d42 gcc port of I422ToARGB_AVX2 from Visual C. Uses Macros for read of I422 and conversion from YUV to RGB. Shares constants from I422ToBGRA structure.
BUG=269
TESTED=nacl builds.
R=brucedawson@google.com, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/27279004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1181 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-12-02 00:22:56 +00:00
fbarchard@google.com
ef14972df0 MergeUV AVX2 use vextractf128 to store results to avoid shuffling.
BUG=none
TESTED=intel sde on unittests
R=brucedawson@google.com

Review URL: https://webrtc-codereview.appspot.com/33369004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1178 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-22 03:33:33 +00:00
fbarchard@google.com
147f7b70f5 Quick fix for build gcc - remove unused argument kARGBShuffleMirror from ARGBMirror SSE2.
BUG=none
TESTED=untested
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/30209004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1177 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-22 01:12:19 +00:00
fbarchard@google.com
ef67597b48 ARGBMirror use SSE2 pshufd instruction instead of SSSE3 pshufb.
BUG=269
TESTED=local benchmark for ARGBMirror
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/32509004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1176 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-21 19:25:14 +00:00
fbarchard@google.com
91f240c5db Move sub before branch for loops.
Remove CopyRow_x86
Add CopyRow_Any versions for AVX, SSE2 and Neon.
BUG=269
TESTED=local build
R=harryjin@google.com, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/26209004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1175 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-20 21:14:27 +00:00
fbarchard@google.com
813bf9f97d Change lea macros from memaccess to memlea to fix nacl 64 bit build errors.
BUG=381
TESTED=local nacl build and validate
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/32129004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1174 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-19 23:02:04 +00:00
fbarchard@google.com
db7a7f61ff Port ARGBMirror AVX2 code to gcc/NaCL.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/24329004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1173 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-19 20:03:37 +00:00
fbarchard@google.com
55db4ec23b port lea removal for mirror to gcc
BUG=none
TESTED=none
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/27209004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1169 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-17 20:06:40 +00:00
fbarchard@google.com
5822505e0a Remove extra unaligned loop from alphablender. Both aligned and unaligned loops were the same, so remove the extra.
BUG=none
TESTED=try bots.
R=brucedawson@google.com, harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/29059004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1166 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-17 18:33:07 +00:00
fbarchard@google.com
35508d0979 Mirror_AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/32079004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1164 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-13 23:11:10 +00:00
fbarchard@google.com
91000425a3 ARGBUnattenuate_AVX2 ported to GCC. Minor cleanup of constants to use broadcast to make 16 byte constant instead of 32 byte.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/30999004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1163 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-13 17:57:33 +00:00
fbarchard@google.com
f8c334473b ARGBAttenuate_AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/29049004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1162 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-12 18:38:06 +00:00
fbarchard@google.com
a843cafbe4 ARGBMultiply_AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/27139005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1160 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-11 20:33:33 +00:00
fbarchard@google.com
0387df5186 ARGBSubtract_AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/27129004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1159 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-11 19:12:38 +00:00
fbarchard@google.com
9e9e26d60a ARGBAdd ported AVX2 ported to GCC.
BUG=269
TESTED=try bots
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/32449004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1158 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-11 19:01:29 +00:00
fbarchard@google.com
10d9c0d0a7 MergeUV for AVX2 ported to gcc. Add missing vzeroupper to all avx2 functions.
BUG=none
TESTED=ncval for nacl
R=brucedawson@google.com, harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/25059005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1157 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-10 19:19:12 +00:00
fbarchard@google.com
d1885bcc73 SplitUVRow_AVX2 ported to GCC/NaCL.
BUG=269
TESTED=validator for nacl.
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/28979004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1156 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-06 01:39:26 +00:00
fbarchard@google.com
bb3a4b41e9 vextractf128 requuires a constant argument for which dqword to extract, so add a new macro.
BUG=none
TESTED=local build on clang for osx
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/30869004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1153 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-04 21:05:55 +00:00
fbarchard@google.com
3f87404769 Port YUY2ToUV, YUY2ToUV422, UYVYToUV and UYVYToUV422 to AVX2 on GCC/Nacl.
BUG=269
TESTED=ncval
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/26029004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1152 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-04 18:24:10 +00:00
fbarchard@google.com
067892c5a1 Port YUY2ToYRow_AVX2 and UYVYToYRow_AVX2 to gcc/NaCL from Windows AVX code.
BUG=269
TESTED=ncval
R=brucedawson@google.com, harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/25039004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1151 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-03 18:30:17 +00:00
fbarchard@google.com
260e3b2273 now that libyuv requires newer nacl compiler, bundles can be assumed and bundle align macro can be removed. no impact on code gen.
BUG=none
TESTED=validator still passes
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/30019004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1150 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-10-30 20:02:03 +00:00