31 Commits

Author SHA1 Message Date
Frank Barchard
11cbf8f976 Add LIBYUV_BIT_EXACT macro to force C to match SIMD
- C code use ARM path, so NEON and C match
- C used on Intel platforms, disabling AVX.

Bug: libyuv:908, b/202888439
Change-Id: Ie035a150a60d3cf4ee7c849a96819d43640cf020
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/3223507
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: richard winterton <rrwinterton@gmail.com>
2021-10-14 20:37:39 +00:00
Frank Barchard
49ebc996aa Make 2 step transitive tests measure 2 step time.
Add tests of all macros used by libyuv public headers

When a 1 step conversion is added, a 2 step test can compare
the old 2 step method to the 1 step.  A 1 step unittest is
also added which compares C to SIMD.  Making the 2 step
conversions measure performance of the 2 steps allows the
old 2 step performance to be compared to 1 step.

All macros used in public headers are added to an ifdef test.
Showing them in a unittest allows some diagnostics when
a test is failing.

Bug: libyuv:901
Change-Id: I7ffa6ed0cb3b506fa1b7fd4b7b1b729658c3c266
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2857916
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Mirko Bonadei <mbonadei@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2021-04-30 18:14:57 +00:00
Frank Barchard
5e05f26a2b Switch win32 to row_gcc for clangcl.
Bug: libyuv:900, libyuv:848, b/178283356, b/185922513
Change-Id: I7697953753391c555a778198db36412c853fb29e
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2844962
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
2021-04-22 19:32:32 +00:00
Frank Barchard
5e83cac0d5 Disable win32 SIMD
Bug: libyuv:900, libyuv:848, b/178283356, b/185922513
Change-Id: Iee7d9970c7991856c8f51158cd12ec72ee9c57eb
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2844779
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
2021-04-21 21:37:44 +00:00
Martin Storsjö
9b772abf97 Restore the file mode for source files
This was changed in 21be9122aadf7824efe3fc19b2a09ff253a688e1.

Change-Id: I6c04dc92f673557e10c231bd090ec8aa88b6bee4
Reviewed-on: https://chromium-review.googlesource.com/1146183
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2018-08-06 18:53:32 +00:00
lixia zhang
21be9122aa libyuv:loongson optimize compare/row/scale/rotate files with mmi.
Currently, libyuv supports MIPS SIMD Arch(MSA),
but libyuv does not supports MultiMedia Instruction(MMI)(such as loongson3a platform).

In order to improve performance of libyuv on loongson3a platform,
this provides optimize 98 functions with mmi.

BUG=libyuv:804

Change-Id: I8947626009efad769b3103a867363ece25d79629
Reviewed-on: https://chromium-review.googlesource.com/1122064
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2018-07-20 22:53:04 +00:00
Frank Barchard
7e389884a1 Switch to C99 types
Append _t to all sized types.
uint64 becomes uint64_t etc

Bug: libyuv:774
Test: try bots build on all platforms
Change-Id: Ide273d7f8012313d6610415d514a956d6f3a8cac
Reviewed-on: https://chromium-review.googlesource.com/879922
Reviewed-by: Miguel Casas <mcasas@chromium.org>
2018-01-23 19:16:05 +00:00
Frank Barchard
a2142148e9 Remove x64 native_client macros.
Bug: libyuv:702
Test: try bots pass
Change-Id: I76d74b5f02fe9843418108b84742e2f714d1ab0a
Reviewed-on: https://chromium-review.googlesource.com/855656
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
2018-01-10 01:27:22 +00:00
Frank Barchard
3b81288ece Remove Mips DSPR2 code
Bug: libyuv:765
Test: build for mips still passes
Change-Id: I99105ad3951d2210c0793e3b9241c178442fdc37
Reviewed-on: https://chromium-review.googlesource.com/826404
Reviewed-by: Weiyong Yao <braveyao@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
2017-12-14 18:22:16 +00:00
Frank Barchard
735ace2ed3 Re-enable x86 assembly without requiring -msse2
clang does not require -msse2 or -msse for inline, except
the "x" parameter.  So change this to "m" for 32 bit.  64 bit
requires sse2 so use "x" for 64 bit.

gcc requires -msse for xmm registers in clobber list.
Reduce compiler requirement from -msse2 to -msse for enabling
assembly.

Bug: libyuv:754, libyuv:757
Test: CC=clang CXX=clang++ CFLAGS="-m32" CXXFLAGS="-m32 -mno-sse -O2" make -f linux.mk
Change-Id: I86df72cfee80b7d349561c1fd7c97ad360767255
Reviewed-on: https://chromium-review.googlesource.com/759303
Reviewed-by: richard winterton <rrwinterton@gmail.com>
Reviewed-by: Frank Barchard <fbarchard@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-11-09 00:51:06 +00:00
Frank Barchard
d997ac287d Revert "Enable SSE2 code without -msse"
This reverts commit 01e994d74e4e3937ee1a3efdc048320a1e51f818.

Change-Id: Ie76710d0f4e641e071889c5125fd3be23cdcdb59
Reviewed-on: https://chromium-review.googlesource.com/758499
Reviewed-by: Frank Barchard <fbarchard@google.com>
2017-11-08 19:33:09 +00:00
Frank Barchard
01e994d74e Enable SSE2 code without -msse
Bug: libyuv:754
Test: CC=clang CXX=clang++ CFLAGS="-m32" CXXFLAGS="-m32 -mno-sse -O2" make -f linux.mk
Change-Id: I74bf8d032013694e65ea7637bc38d3253db53ff2
Reviewed-on: https://chromium-review.googlesource.com/758043
Reviewed-by: Frank Barchard <fbarchard@google.com>
2017-11-08 02:54:41 +00:00
Frank Barchard
8fa02df3c0 mingw fix ifdefs to use gcc source
mingw gcc sets the macro _M_IX86 which is normally only set
by Visual C and clangcl which are Visual C style source code
style for assembly, but gcc is not Visual C compatible.
Add _MSC_VER to most ifdefs to detect that its really Visual C
or clangcl and not mingw gcc so the gcc source code will be used.

Bug: libyuv:744
Test: CXXFLAGS=-m32 CXX=~/prebuilts/gcc/linux-x86/host/x86_64-w64-mingw32-4.8/bin/x86_64-w64-mingw32-g++ make -f linux.mk
Change-Id: I3431aa486eb769b145faa8d5eb75ed639f9d6f5e
Reviewed-on: https://chromium-review.googlesource.com/722319
Reviewed-by: Cheng Wang <wangcheng@google.com>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-10-17 17:36:35 +00:00
Frank Barchard
54f2094a5e Rename mips source files to dspr2.
Add MSA detect to unittest.
Change macro to disable DSPR2 code to LIBYUV_DISABLE_DSPR2

BUG=libyuv:634
TEST=try bots

Change-Id: I9e0aa2452204fc529bb6f9e6fd93c4e1c379bba6
Reviewed-on: https://chromium-review.googlesource.com/433463
Reviewed-by: Henrik Kjellander <kjellander@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@google.com>
2017-01-27 23:11:43 +00:00
Manojkumar Bhosale
73a6f100a9 Add MSA optimized rotate functions (used 16x16 transpose)
R=fbarchard@google.com
BUG=libyuv:634

Performance Gain (vs C vectorized)
TransposeWx16_MSA        - ~6.0x
TransposeWx16_Any_MSA    - ~4.7x
TransposeUVWx16_MSA      - ~6.3x
TransposeUVWx16_Any_MSA  - ~5.4x

Performance Gain (vs C non-vectorized)
TransposeWx16_MSA        - ~6.0x
TransposeWx16_Any_MSA    - ~4.8x
TransposeUVWx16_MSA      - ~6.3x
TransposeUVWx16_Any_MSA  - ~5.4x

Review-Url: https://codereview.chromium.org/2617703002 .
2017-01-13 15:50:02 +05:30
Manojkumar Bhosale
6fa5e4eb78 Add MSA optimized TransposeWx8_MSA and TransposeUVWx8_MSA functions
R=fbarchard@google.com
BUG=libyuv:634

Performance Gain (vs C vectorized)
TransposeWx8_MSA          - ~2.7x
TransposeWx8_Any_MSA      - ~2.1x
TransposeUVWx8_MSA        - ~2.5x
TransposeUVWx8_Any_MSA    - ~2.7x

Performance Gain (vs C non-vectorized)
TransposeWx8_MSA          - ~4.6x
TransposeWx8_Any_MSA      - ~2.9x
TransposeUVWx8_MSA        - ~4.4x
TransposeUVWx8_Any_MSA    - ~3.7x

Review URL: https://codereview.chromium.org/2553403002 .
2016-12-15 10:06:01 +05:30
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
bcd823805c remove guard nolints from all headers
Remove NOLINT from guards

TEST=git cl lint
BUG=libyuv:634

Review URL: https://codereview.chromium.org/2374653002 .
2016-09-26 18:02:09 -07:00
Frank Barchard
837aa1e2af disable assembly in header for msan=1
GYP_DEFINES="target_arch=x64 msan=1" ./gyp_libyuv
ninja -j7 -C out/Release

R=impjdi@google.com
BUG=libyuv:575

Review URL: https://codereview.chromium.org/1805683003 .
2016-03-15 18:45:38 -07:00
Frank Barchard
0d880e5bc0 rename MIPS_DSPR2 to DSPR2 for consistency
When attempting to normalize function names to end in Row_SIMD it was made
harder with MIPS_DSPR2 naming convention.
Other CPUs do not include the vendor.  This should be named consistently.

Removed the DISABLE_MIPS in favour of DISABLE_ASM for consistency with other
processors.

TBR=harryjin@google.com
BUG=libyuv:562

Review URL: https://codereview.chromium.org/1677633002 .
2016-02-05 14:49:54 -08:00
Frank Barchard
216e93b4e8 Fix MIPS DSPR2 build failure.
Fixing the failure:
 'TransposeWx8_Fast_MIPS_DSPR2' was not declared in this scope

BUG=none
R=fbarchard@chromium.org

Review URL: https://codereview.chromium.org/1527243002 .
2015-12-16 10:37:42 -08:00
Frank Barchard
94d4269936 clang use scalewin
R=harryjin@google.com
TBR=harryjin@google.com
BUG=libyuv:469

Review URL: https://webrtc-codereview.appspot.com/51329004.
2015-08-18 14:50:27 -07:00
Frank Barchard
58f0020137 use visual c 32 bit code for clangcl
R=harryjin@google.com
BUG=libyuv:483

Review URL: https://webrtc-codereview.appspot.com/54819004.
2015-08-11 10:10:45 -07:00
Frank Barchard
9425c4b01a rotate nv12 any width
BUG=libyuv:464
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/55709004.
2015-08-07 23:48:38 -07:00
Frank Barchard
e40384b6d9 remove 32 bit gcc version of UV transpose
TBR=harryjin@google.com
BUG=libyuv:481

Review URL: https://webrtc-codereview.appspot.com/52249004.
2015-08-03 18:03:55 -07:00
Frank Barchard
f14c433916 rotate macros used for source
R=brucedawson@chromium.org, harryjin@google.com
BUG=libyuv:481

Review URL: https://webrtc-codereview.appspot.com/52239004.
2015-08-03 16:12:18 -07:00
Frank Barchard
cb54e8b69a rename rotate macros and functions to match
BUG=libyuv:477
R=harryjin@google.com

Review URL: https://webrtc-codereview.appspot.com/52199004.
2015-07-27 17:00:41 -07:00
Frank Barchard
6a1d01220a disable x86 consistently
R=harryjin@google.com, jzern@chromium.org
BUG=libyuv:476

Review URL: https://webrtc-codereview.appspot.com/55699004.
2015-07-27 12:49:54 -07:00
Frank Barchard
2fa4f5a3ea Adds files and functions for rotate any, but does not hook them up to the caller.
rotate any

R=harryjin@google.com
BUG=libyuv:464

Review URL: https://webrtc-codereview.appspot.com/53769004.
2015-07-27 10:32:08 -07:00
Frank Barchard
892807d860 move asm out of rotate into win/gcc and header
R=harryjin@google.com
BUG=libyuv:468

Review URL: https://webrtc-codereview.appspot.com/51319004.
2015-07-22 11:22:55 -07:00
Frank Barchard
673fe7a684 create rotate_row header
R=tpsiaki@google.com, tpsiaki
BUG=none
TESTED=local build still works.

Review URL: https://webrtc-codereview.appspot.com/50329004.
2015-07-09 14:40:35 -07:00