55 Commits

Author SHA1 Message Date
Frank Barchard
60f433fbd9 Revert "ComputeHammingDistance reduce SIMD loop to 1 call when possible."
This reverts commit ec75df5894845b8d6b1341885a78db1de83decd8.

Reason for revert: <INSERT REASONING HERE>

Original change's description:
> ComputeHammingDistance reduce SIMD loop to 1 call when possible.
> 
> 32 bit x86 has high overhead due to -fpic.  So this reduces the
> number of calls by 1.
> 
> TBR=kjellander@chromium.org
> Bug: libyuv:701
> Test: BenchmarkHammingDistance
> Change-Id: I7f557ef047920db65eab362a5f93abbd274ca051
> Reviewed-on: https://chromium-review.googlesource.com/701755
> Reviewed-by: Frank Barchard <fbarchard@google.com>
> Reviewed-by: Cheng Wang <wangcheng@google.com>

TBR=rrwinterton@gmail.com,fbarchard@google.com,wangcheng@google.com

Change-Id: Ia61e8558a8f083c14be5f51e0e141550b6f2b5c1
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: libyuv:701
Reviewed-on: https://chromium-review.googlesource.com/707823
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-10-10 01:16:15 +00:00
Frank Barchard
ec75df5894 ComputeHammingDistance reduce SIMD loop to 1 call when possible.
32 bit x86 has high overhead due to -fpic.  So this reduces the
number of calls by 1.

TBR=kjellander@chromium.org
Bug: libyuv:701
Test: BenchmarkHammingDistance
Change-Id: I7f557ef047920db65eab362a5f93abbd274ca051
Reviewed-on: https://chromium-review.googlesource.com/701755
Reviewed-by: Frank Barchard <fbarchard@google.com>
Reviewed-by: Cheng Wang <wangcheng@google.com>
2017-10-09 22:51:23 +00:00
Frank Barchard
1734712a6f Fix odd length HammingDistance
If length of HammingDistance was not a multiple of 4,
the result was incorrect.  The old tests did not catch this
so a new test is done to count 1s.

Bug: libyuv:740
Test: LibYUVCompareTest.TestHammingDistance
Change-Id: I93db5437821c597f1f162ac263d4a594bb83231f
Reviewed-on: https://chromium-review.googlesource.com/699614
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Reviewed-by: Cheng Wang <wangcheng@google.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-10-04 22:21:36 +00:00
Frank Barchard
fecd741794 Port HammingDistance to SSSE3
Bug: libyuv:701
Test: BenchmarkHammingDistance_Opt
Change-Id: Ibdd5d382677ebef4f82a62e0d5c3b88614a3b6e4
Reviewed-on: https://chromium-review.googlesource.com/696290
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: Cheng Wang <wangcheng@google.com>
2017-10-03 19:11:05 +00:00
Frank Barchard
bde789b176 Hamming Distance SSE2 and AVX2 optimized
Bug: None
Test: None
Change-Id: Id52663f9c957aac3172fba92d888ad1b041d5cf0
Reviewed-on: https://chromium-review.googlesource.com/692981
Reviewed-by: Cheng Wang <wangcheng@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-10-02 22:32:54 +00:00
Frank Barchard
db25485ee2 Move compare functions into a unittest class
BUG=None
TEST=LibYUVCompareTest.*
R=jkellander@chromium.org

Change-Id: I3131ca73020f855ead08255d09aa7a846bf0d556
Reviewed-on: https://chromium-review.googlesource.com/540064
Commit-Queue: Frank Barchard <fbarchard@google.com>
Reviewed-by: Henrik Kjellander <kjellander@chromium.org>
2017-06-19 19:39:10 +00:00
Frank Barchard
77f6916da2 use __popcnt for visual c HammingDistance_X86
BUG=libyuv:701
TEST=HammingDistance unittest performance is comparable to x64
R=wangcheng@google.com

Change-Id: I8abe861e086e0162ba4c7ba6f1ef7d1c006cd9d4
Reviewed-on: https://chromium-review.googlesource.com/505454
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-05-12 22:59:00 +00:00
Frank Barchard
e0615c0e69 Optimize Hamming Distance C code to do 64 bits at a time.
BUG=libyuv:701
TEST=LibYUVBaseTest.BenchmarkHammingDistance_C
R=wangcheng@google.com

Change-Id: I243003b098bea8ef3809298bbec349ed52a43d8c
Reviewed-on: https://chromium-review.googlesource.com/499487
Reviewed-by: Cheng Wang <wangcheng@google.com>
2017-05-12 17:53:52 +00:00
Frank Barchard
2136e349da Hamming code difference of 2 memory blocks
BUG=libyuv:701
TEST=built and disassembled for aarch64
R=kjellander@chromium.org

Change-Id: I7712b1c7934e5dfb55fda1fa7c8405c32d6964ce
Reviewed-on: https://chromium-review.googlesource.com/495327
Reviewed-by: Henrik Kjellander <kjellander@chromium.org>
Reviewed-by: Cheng Wang <wangcheng@google.com>
2017-05-08 21:37:51 +00:00
Frank Barchard
e62309f259 clang-format libyuv
BUG=libyuv:654
R=kjellander@chromium.org

Review URL: https://codereview.chromium.org/2469353005 .
2016-11-07 17:37:23 -08:00
Frank Barchard
198bce3959 Cast for clang-cl 64 bit build warnings in unittests
R=kjellander@chromium.org
BUG=libyuv:649

Review URL: https://codereview.chromium.org/2414763002 .
2016-10-12 13:09:57 -07:00
Frank Barchard
ade85fb55c remove row.h from unittests
add SIMD_ALIGNED to unittest header.

BUG=libyuv:594
TESTED=local build passes with row.h removed from tests.
R=harryjin@google.com

Review URL: https://codereview.chromium.org/2001373002 .
2016-05-27 10:57:49 -07:00
Frank Barchard
6020d2aa64 fix off by one in unitest BenchmarkARGBDetect_Unaligned where array is initialized beyond end by 1.
TBR=harryjin@google.com
BUG=libyuv:595
TESTED=local unittest passes on try bots

Review URL: https://codereview.chromium.org/2012603002 .
2016-05-24 15:33:27 -07:00
Frank Barchard
c629cb3afe add command line cpu info to allow android neon test
in order to compare C and Neon code, a new command line flag is added.
historically environment variables controlled cpu features, but on
android apk it is easier to pass a command line option to disable cpu
optimizations.

R=harryjin@google.com
BUG=libyuv:516

Review URL: https://codereview.chromium.org/1407193009 .
2015-11-03 17:01:48 -08:00
Frank Barchard
26db4de2ae break up unittests into categories
R=harryjin@google.com
BUG=none

Review URL: https://codereview.chromium.org/1399523004 .
2015-10-13 16:01:07 -07:00
Frank Barchard
16f12b58cc Replace random with fastrand
random / rand is slow and impacts performance testing.
Although its only called to clear a frame once, a typical profile shows
it high in the overall profile, when doing 1000 frames for a benchmark.

95.10%  libyuv_unittest  libyuv_unittest      [.] YUY2ToARGBRow_SSSE3
 2.01%  libyuv_unittest  libc-2.19.so         [.] __random_r
 1.13%  libyuv_unittest  libc-2.19.so         [.] __random

Replace random is a faster version for unittests.

set LIBYUV_WIDTH=1280
set LIBYUV_HEIGHT=720
set LIBYUV_REPEAT=999
set LIBYUV_FLAGS=-1
out\release\libyuv_unittest --gtest_filter=*YUY2ToARGB*  | findms

Was
libyuvTest.YUY2ToARGB_Opt (497 ms)

Now
libyuvTest.YUY2ToARGB_Opt (454 ms)

R=harryjin@google.com
BUG=none

Review URL: https://codereview.chromium.org/1361813002 .
2015-09-22 15:47:36 -07:00
fbarchard@google.com
f16f33d4ce All cpu flags to be set so that instead of comparing C code, compare assembler to assembler, for benchmarking purposes.
BUG=none
TESTED=libyuv_unittest.exe
R=bcornell@google.com

Review URL: https://webrtc-codereview.appspot.com/50499004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1346 16f28f9a-4ce2-e073-06de-1de4eb20be90
2015-03-26 18:22:29 +00:00
fbarchard@google.com
a6025e8b6b ARGBDetect do 2 pixels at a time for improved performance.
BUG=375
TESTED=libyuvTest.BenchmarkARGBDetect_Opt
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/26049004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1155 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-05 23:23:17 +00:00
fbarchard@google.com
b661b3ee0d Detect Endian of ARGB image.
BUG=375
TESTED=libyuv builds, but no test app for it yet
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/32389004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1154 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-11-05 18:46:06 +00:00
fbarchard@google.com
9c4c82181b Remove alignment constraint for SSE2. Allows the optimized function to be used with unaligned memory, improving performance in that use case. Hurts performance on core2 and prior where memory was faster with movdqa instruction.
BUG=365
TESTED=psnr, ssim and djb2 unittests pass.
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/22859004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@1100 16f28f9a-4ce2-e073-06de-1de4eb20be90
2014-09-30 18:53:34 +00:00
fbarchard@google.com
ae9a1388a7 Use malloc for row buffers in rotate
BUG=296
TESTED=rotate_test
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/6329004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@922 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-12-26 21:41:11 +00:00
fbarchard@google.com
0014ce0056 test odd width and fix for unaligned used on odd width conversion.
BUG=283
TESTED=try bots
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/4729004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@883 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-12-02 20:07:29 +00:00
fbarchard@google.com
095f33d870 Coalesce rows by changing width/height and dropping into code instead of recursing. Improve coalesce by setting stride to 0 so it can be used even on odd width images. Reduce unittests to improve time to run emulators.
BUG=277
TEST=unittests all build and pass
R=ryanpetrie@google.com

Review URL: https://webrtc-codereview.appspot.com/2589004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@819 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-10-21 19:29:10 +00:00
fbarchard@google.com
88ce3c0caa Change unittests to pass when size is 1 tall.
BUG=202
TEST=set LIBYUV_HEIGHT=1 & out\release\libyuv_unittest
R=ryanpetrie@google.com

Review URL: https://webrtc-codereview.appspot.com/2391004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@811 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-10-14 19:36:26 +00:00
fbarchard@google.com
c9f0d966ed FixedDiv port to gcc
BUG=250
TEST=*Div*
R=ryanpetrie@google.com

Review URL: https://webrtc-codereview.appspot.com/1682005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@733 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-06-18 18:58:33 +00:00
fbarchard@google.com
8ffe78abd2 Scale down by 4 used 3rd pixel
BUG=232
TEST=convert.exe -f 0 faces_640x480_P420.yuv face2_160x120_P420.yuv
R=changjun.yang@intel.com

Review URL: https://webrtc-codereview.appspot.com/1579005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@709 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-05-30 07:14:14 +00:00
fbarchard@google.com
0066be9716 test hash of quick brown fox for known hash value
BUG=none
TESTED=out\release\libyuv_unittest --gtest_filter=libyuvTest.Djb2_Test
R=johannkoenig@google.com

Review URL: https://webrtc-codereview.appspot.com/1527004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@699 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-05-22 02:10:47 +00:00
fbarchard@google.com
9f24b14e05 Compare typo fix
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/1390005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@691 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-05-09 01:43:54 +00:00
fbarchard@google.com
b06d3f5344 Add more tests for hash
BUG=none
TEST=out\release\libyuv_unittest --gtest_filter=*Djb2*
Review URL: https://webrtc-codereview.appspot.com/1448004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@690 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-05-09 00:34:05 +00:00
fbarchard@google.com
6a34ee200e Test DJB2 hash with half the buffer same.
BUG=none
TEST=libyuvTest.BenchmakDjb2_Test
Review URL: https://webrtc-codereview.appspot.com/1446004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@689 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-05-08 23:05:08 +00:00
fbarchard@google.com
4127a2637d ARGBInterpolate odd width support and inverted odd width test. ARGBToNV12/21 odd height fix. Compare test tolerate small height with warning.
BUG=202
TEST=libyuvTest.ARGBInterpolate85_Any_Invert
Review URL: https://webrtc-codereview.appspot.com/1325004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@663 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-04-15 10:43:33 +00:00
fbarchard@google.com
4e0d7cc2c6 Y coefficients for J420 need to be scaled by 255/219 to full range.
BUG=159
TESTED=out\release\libyuv_unittest --gtest_filter=*J*
Review URL: https://webrtc-codereview.appspot.com/1264004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@624 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-03-27 07:35:03 +00:00
fbarchard@google.com
04de744106 fix for bayer unittest, transpose and rotate, and Ssim tolerance.
BUG=198
TESTED=*Bayer*
Review URL: https://webrtc-codereview.appspot.com/1194004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@604 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-03-13 01:52:21 +00:00
fbarchard@google.com
cde587092f Replace two spaces with one after .
BUG=none
TEST=lint
Review URL: https://webrtc-codereview.appspot.com/1063010

git-svn-id: http://libyuv.googlecode.com/svn/trunk@553 16f28f9a-4ce2-e073-06de-1de4eb20be90
2013-01-28 00:02:35 +00:00
fbarchard@google.com
d985cf1f62 clamp maxy for first row of bilinear
BUG=153
TEST=none
Review URL: https://webrtc-codereview.appspot.com/934014

git-svn-id: http://libyuv.googlecode.com/svn/trunk@490 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-11-14 08:15:55 +00:00
fbarchard@google.com
3e46444727 Mirror source to continuous destination with Neon
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/937020

git-svn-id: http://libyuv.googlecode.com/svn/trunk@488 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-11-14 02:03:49 +00:00
fbarchard@google.com
55d5b48e4d bilinear scale use fast 2 pass unless image is beyond row buffer size.
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/965020

git-svn-id: http://libyuv.googlecode.com/svn/trunk@484 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-11-13 17:31:08 +00:00
fbarchard@google.com
bd4fcb9877 Switch to 128x72 for default testing size when not benchmarking.
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/935004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@442 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-24 19:52:14 +00:00
fbarchard@google.com
3bb60b37fd Use width variable for all tests
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/934004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@441 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-24 18:31:55 +00:00
fbarchard@google.com
042acf0458 Reduce allowed error on simple conversions
BUG=none
TEST=unittests
Review URL: https://webrtc-codereview.appspot.com/917006

git-svn-id: http://libyuv.googlecode.com/svn/trunk@431 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-20 06:37:21 +00:00
fbarchard@google.com
0a6b7a6eb7 test rgb565 by converting to ARGB and then doing MSE
BUG=114
TEST=libyuvTest.I420ToRGB565_Any
Review URL: https://webrtc-codereview.appspot.com/919004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@428 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-19 20:10:47 +00:00
fbarchard@google.com
c7277d08e8 Add convert_from_argb.h for all conversion functions from ARGB to something else.
BUG=none
TEST=convert_test
Review URL: https://webrtc-codereview.appspot.com/857014

git-svn-id: http://libyuv.googlecode.com/svn/trunk@408 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-11 23:10:53 +00:00
fbarchard@google.com
e91bdaca36 Move HalfRow to row_win and port to row_neon
BUG=118
TEST=libyuvTest.I420ToI422_OptVsC (247 ms)
Review URL: https://webrtc-codereview.appspot.com/855012

git-svn-id: http://libyuv.googlecode.com/svn/trunk@400 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-10-09 21:09:33 +00:00
fbarchard@google.com
7781d943b0 Disable V210 unittest that fails valgrind, and some warning fixes for android.
BUG=91
TEST=try
Review URL: https://webrtc-codereview.appspot.com/813004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@362 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-09-18 22:58:34 +00:00
fbarchard@google.com
b0c9797589 Update Copyright notice to follow new chromium conventions.
BUG=63
TEST=none
Review URL: https://webrtc-codereview.appspot.com/730004

git-svn-id: http://libyuv.googlecode.com/svn/trunk@314 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-08-08 19:04:24 +00:00
fbarchard@google.com
81b804e35c ARGBQuantize to do a posterizing effect. Added random resolution unittest.
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/654005

git-svn-id: http://libyuv.googlecode.com/svn/trunk@289 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-06-20 02:15:01 +00:00
fbarchard@google.com
810cd91079 ARGBUnattenuateRow_SSE2 use reciprocal table and pmul
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/497001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@244 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-04-20 20:15:27 +00:00
fbarchard@google.com
4d3bd8343b unaligned sse4 djb2 cleanup
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/486001

git-svn-id: http://libyuv.googlecode.com/svn/trunk@238 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-04-06 00:27:09 +00:00
fbarchard@google.com
ddf9051bd9 DJB2 hash with SSE4 pmulld
BUG=none
TEST=none
Review URL: https://webrtc-codereview.appspot.com/484002

git-svn-id: http://libyuv.googlecode.com/svn/trunk@237 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-04-05 20:44:03 +00:00
fbarchard@google.com
2b9c210803 lint fix test
BUG=none
TEST=gcl lint lintfix2
Review URL: https://webrtc-codereview.appspot.com/458003

git-svn-id: http://libyuv.googlecode.com/svn/trunk@220 16f28f9a-4ce2-e073-06de-1de4eb20be90
2012-03-22 22:36:44 +00:00