summaryrefslogtreecommitdiff
path: root/test/vp9_quantize_test.cc
AgeCommit message (Collapse)Author
2022-10-17vp9 quantize: rewrite ssse3 in intrinsicsJohann
Change-Id: I3177251a5935453a23a23c39ea5f6fd41254775e
2022-09-24quantize: add untested functionJohann
vp9_quantize_fp_sse2 was only tested in non-hbd configuration. Missed when fixing this for vpx_quantize_b_sse2. Change-Id: Ide346e5727d74281c774f605c90d280050e0bf62
2022-09-18quantize: test lowbd in highbd buildsJohann
Change-Id: I7af273e979415a8b8cafb7494728d2736862f4a5
2022-08-20[NEON] Added vpx_highbd_quantize_b* functionsKonstantinos Margaritis
Total gain for 12-bit encoding: * ~4.8% for best profile * ~6.2% for rt profile Change-Id: I61e646ab7aedf06a25db1365d6d1cf7b05101c21
2022-08-16Add vp9_highbd_quantize_fp_32x32_neon().Scott LaVarnway
Up to 2.6x faster than vp9_highbd_quantize_fp_32x32_c() for full calculations. Bug: b/237714063 Change-Id: Icfeff2ad4dcd57d0ceb47fe04789710807b9cbad
2022-08-15VPX: Add vp9_highbd_quantize_fp_neon().Scott LaVarnway
Up to 4.1x faster than vp9_highbd_quantize_fp_c() for full calculations. ~1.3% overall encoder improvement for the test clip used. Bug: b/237714063 Change-Id: I8c6466bdbcf1c398b1d8b03cab4165c1d8556b0c
2022-08-03VPX: Add vp9_highbd_quantize_fp_32x32_avx2().Scott LaVarnway
~4x faster than vp9_highbd_quantize_fp_32x32_c() for full calculations. Bug: b/237714063 Change-Id: Iff2182b8e7b1ac79811e33080d1f6cac6679382d
2022-08-03VPX: Add vp9_highbd_quantize_fp_avx2().Scott LaVarnway
Up to 5.37x faster than vp9_highbd_quantize_fp_c() for full calculations. ~1.6% overall encoder improvement for the test clip used. Bug: b/237714063 Change-Id: I584fd1f60a3e02f1ded092de98970725fc66c5b8
2022-08-01VPX: Add vp9_quantize_fp_32x32_avx2().Scott LaVarnway
Up to 1.80x faster than vp9_quantize_fp_32x32_ssse3() for full calculations. Bug: b/237714063 Change-Id: Ic4ae4724fce7ac85c7a089535b16a999e02f0a10
2022-07-27VPX: vp9_quantize_fp_avx2() cleanup.Scott LaVarnway
No change in performance. Bug: b/237714063 Change-Id: I8ea42759cc4dc57be6a29c23784997cb90ad4090
2022-07-26VPX: Add vpx_highbd_quantize_b_32x32_avx2().Scott LaVarnway
Up to 11.78x faster than vpx_quantize_b_32x32_sse2() for full calculations. ~1.7% overall encoder improvement for the test clip used. Bug: b/237714063 Change-Id: Ib759056db94d3487239cb2748ffef1184a89ae18
2022-07-25VPX: Add vpx_highbd_quantize_b_avx2().Scott LaVarnway
Up to 3.61x faster than vpx_highbd_quantize_b_sse2() for full calculations. ~2.3% overall encoder improvement for the test clip used. Bug: b/237714063 Change-Id: I23f88d2a7f96aaa4103778372f4f552207f73cee
2022-07-19VPX: Add vpx_quantize_b_32x32_avx2().Scott LaVarnway
Up to 1.36x faster than vpx_quantize_b_32x32_avx() for full calculations. Up to 1.29x faster for VP9_HIGHBITDEPTH builds. Bug: b/237714063 Change-Id: I97aa6a18d4dc2f3187b76800f91bbba7be447ef1
2022-07-11VPX: Add vpx_quantize_b_avx2().Scott LaVarnway
Up to 1.58x faster than vpx_quantize_b_avx() depending on the size. Bug: b/237714063 Change-Id: I595a6bb32ebee63f69f27b5a15322fdeae1bf70e
2022-07-07VPX: Add quantize speed test for ref vs opt.Scott LaVarnway
Bug: b/237714063 Change-Id: I4304ba8d976fed3613e28442983b04a9cfc15b79
2022-05-18vp9[loongarch]: Optimize vpx_quantize_b/b_32x32yuanhecai
1. vpx_quantize_b_lsx 2. vpx_quantize_b_32x32_lsx Bug: webm:1755 Change-Id: I476c8677a2c2aed7248e088e62c3777c9bed2adb
2022-03-31Revert "quantize: replace highbd versions"James Zern
This reverts commit 2200039d33c49a9f7a5c438656df143755b022c4. This causes failures with VP9/EndToEndTestLarge.EndtoEndPSNRTest/*; it seems the assembly does not match the C code. Bug: webm:1586 Change-Id: I4c63beebf88d4c12789d681b0d38014510b147fe
2022-03-31Revert "quantize: remove highbd version"James Zern
This reverts commit 89cfe3835c47dabf77d38edb3af190155984fa9a. This is a prerequisite for reverting 2200039d33c49a9f7a5c438656df143755b022c4 which causes high bitdepth test failures Bug: webm:1586 Change-Id: I28f3b98f3339f3573b1492b88bf733dade133fc0
2022-03-31quantize: remove highbd versionJohann
The only difference between the code is the clamp. For 8 bit it is purely an optimization. The values outside this range will still saturate. Change-Id: I2a770b140690d99e151b00957789bd72f7a11e13
2022-03-31quantize: replace highbd versionsJohann
The optimized quantize functions were already built to handle highbd values. The only difference is the clamping. All highbd functions expand to 32bits when running in highbd mode. Removes vpx_highbd_quantize_32x32_sse2 as it is slower than the C version in the worst case. Bug: webm:1586 Change-Id: I49bf8a6a2041f78450bf43a4f655c67656b0f8d9
2022-03-30remove skip_block from quantizeJohann
Whether a block is skipped is handled by mi->skip. x->skip_block is kept exclusively to verify that the quantize functions are not called for skip blocks. Finishes the cleanup in 13eed991f Bug: libvpx:1612 Change-Id: I1598c3b682d3c5e6c57a15fa4cb5df2c65b3a58a
2020-07-27NULL -> nullptr in CPP filesJerome Jiang
This should clean up clangtidy warnings Change-Id: Ifb5a986121b2d0bd71b9ad39a79dd46c63bdb998
2020-06-18update googletest to v1.10.0James Zern
this moves the framework to c++11 and changes *_TEST_CASE* to _TEST_SUITE BUG=webm:1695 Change-Id: I07f2c20850312a9c7e381b38353d2f9f45889cb1
2019-09-30namespace ARCH_* definesJames Zern
this prevents redefinition warnings if a toolchain sets one BUG=b/117240165 Change-Id: Ib5d8c303cd05b4dbcc8d42c71ecfcba8f6d7b90c
2019-08-22test/vp9_quantize_test: fix int sanitizer warningHien Ho
implicit conversion from type 'int' of value 42126 (32-bit, signed) to type 'tran_low_t' (aka 'short') changed the value to -23410 (16-bit, signed) BUG=webm:1615 Change-Id: I339c640fce81e9f2dd73ef9c9bee084b6a5638dc
2019-08-01Fix saturation issue in vp9_quantize_fp_neonJerome Jiang
Change-Id: I7850a5c5aea3633e50e9a2efc8116b9e16383a8f
2018-12-07test/*: use std::*tupleJames Zern
since: 77fa51003 Replace deprecated scoped_ptr with unique_ptr c++11 has been required so <tuple> is safe to use Change-Id: I873cb953104b361a8503b5839a3372ce2b99e73c
2018-12-03quantize neon: fix hbd buildsJohann
BUG=webm:1448 Change-Id: I2140fb9b6ce92716d2d9509f3031244088a62127
2018-11-30quantize 32x32: saturate dqcoeff on x86Johann
This slows down low bitdepth builds but is necessary to obtain correct values. BUG=webm:1448 Change-Id: I4ca9145f576089bb8496fcfeedeb556dc8fe6574
2018-11-28quantize 32x32: fix dqcoeffJohann
Calculate the high bits of dqcoeff and store them appropriately in high bit depth builds. Low bit depth builds still do not pass. C truncates the results after division. X86 only supports packing with saturation at this step. BUG=webm:1448 Change-Id: Ic80def575136c7ca37edf18d21e26925b475da98
2018-11-28quantize: fix x86 hbd buildsJohann
Calculate the high bits of dqcoeff in high bit depth builds and store them appropriately. BUG=webm:1448 Change-Id: I61a2f8bfcf2e30765f10a94073c4d58321d2fa24
2018-06-18include msvc.h for snprintf support in benchmarksLuc Trudeau
include vpx_ports/msvc.h to avoid issues with snprintf issues with MSVC. Change-Id: Ida09cff8ee3b84e09fd61de131f84b32c113fa1a
2018-06-11VSX Version of vp9_quantize_fp_32x32Luc Trudeau
Low bit depth version only. Passes the VP9QuantizeTest test suite. VP9QuantizeTest Speed Test (POWER8 Model 2.1) 32x32 C time = 93.1 ms (±0.4 ms), VSX time = 6.5 ms (±0.2 ms) [14.4x] Change-Id: I7f1fd0fc987af86baf2b74147a25aee811289112
2018-06-11VSX Version of vp9_quantize_fpLuc Trudeau
Low bit depth version only. Passes the VP9QuantizeTest test suite. VP9QuantizeTest Speed Test (POWER8 Model 2.1) 4x4 C time = 86.3 ms (±0.7 ms), VSX time = 18.2 ms (±0.0 ms) [ 4.7x] 8x8 C time = 57.7 ms (±0.3 ms), VSX time = 7.6 ms (±0.0 ms) [ 7.6x] 16x16 C time = 50.7 ms (±0.1 ms), VSX time = 4.9 ms (±0.0 ms) [10.3x] Change-Id: Ic09bc786c57cc89bba14624064216b52996075eb
2018-06-04test,cosmetics: fix func/member naming, decl orderJames Zern
functions: upper camelcase members: lowercase with trailing '_' decl order: functions (overrides marked virtual), members after: 656e8ac61 VSX version of vpx_post_proc_down_and_across_mb_row 766d875b9 VSX version of vpx_mbpost_proc_ip 35e98a70b VSX version of vpx_mbpost_proc_down b2898a9ad Bench Class For More Robust Speed Tests Change-Id: Ib257bd607c5c1248d30e619ec9e8a47cc629825b
2018-05-29Bench Class For More Robust Speed TestsLuc Trudeau
To make speed testing more robust, the AbstractBench runs the desired code multiple times and report the median run time with mean absolute deviation around the median. To use the AbstractBench, simply add it as a parent to your test class, and implement the run() method (with the code you want to benchmark). Sample output for VP9QuantizeTest [ BENCH ] Bypass calculations 4x4 165.8 ms ( ±1.0 ms ) [ BENCH ] Full calculations 4x4 165.8 ms ( ±0.9 ms ) [ BENCH ] Bypass calculations 8x8 129.7 ms ( ±0.9 ms ) [ BENCH ] Full calculations 8x8 130.3 ms ( ±1.4 ms ) [ BENCH ] Bypass calculations 16x16 110.3 ms ( ±1.4 ms ) [ BENCH ] Full calculations 16x16 110.1 ms ( ±0.9 ms ) Change-Id: I1dd649754cb8c4c621eee2728198ea6a555f38b3
2018-05-14VSX version of vpx_quantize_b_32x32_vsxLuc Trudeau
Low bit depth version only. Passes the VP9QuantizeTest. VP9QuantizeTest Speed Test (POWER8 Model 2.1) Full calculations: C time = 1456 ms, VSX time = 80 ms (18x) Change-Id: I1b1d6d03b1aeff63640efbdeb222cab857ddd95e
2018-05-09VSX version of vpx_quantize_b_vsxLuc Trudeau
Low bit depth version only. Passes the VP9QuantizeTest. Change-Id: I6546f872864bd404a7e353348b0554aab1de5bf0
2018-03-28test: use testing::*tuple instead of std::tr1James Zern
googletest imports tuple into testing to allow for compatibility across c++ versions where tuple may be in std::tr1 or std. fixes deprecation warnings under visual studio 2017 Change-Id: Id78b372d5478b12d8c8f63fd3f2166fec25aa8be
2018-01-18vp9_quantize_fp_avx2()Scott LaVarnway
Started from vp9_quantize_fp_sse2 and tweaked to use avx2. Change-Id: Ic2da50cc9d73896c7ef2f3cd3db5b1c5d7795b8b
2017-12-26Add quantize_fp_32x32_nz_c()Scott LaVarnway
This c version uses the shortcuts found in the vp9_quantize_fp_32x32_ssse3 function. Change-Id: I2e983adb00064e070b7f2b1ac088cc58cf778137
2017-12-21Add vp9_quantize_fp_nz_c() -- 2Scott LaVarnway
This c version uses the shortcuts found in the x86 vp9_quantize_fp functions. The test was updated to use the correct quant/round range. Change-Id: Ie5871f710d9eb39047d8d9f48b907c0633e1f830
2017-12-21Revert "Add vp9_quantize_fp_nz_c()"James Zern
This reverts commit 86842855d30d6ca6befdcf5108003e027d90daa9. SSSE3/VP9QuantizeTest.EOBCheck/1 fails on Mac and the build breaks under visual studio due to a #if within another macro. Change-Id: I475095a04aafcc714fade2b24e4df7b682be2cd1
2017-12-19Add vp9_quantize_fp_nz_c()Scott LaVarnway
This c version uses the shortcuts found in the x86 vp9_quantize_fp functions. The test was updated to use the correct quant/round range. Change-Id: I5d19f8af2fddda8e50910249eafb740acb29415b
2017-09-12Revert "Revert "quantize avx: copy 32x32 implementation""Johann
This reverts commit 8c42237bb200253931c49e2c530838f3a877dd65. Because ssse3 code is used for the reference, the qcoeff and dqcoeff reference buffers must be aligned. Original change's description: > quantize avx: copy 32x32 implementation > > Ensure avx and ssse3 stay in sync by testing them against each other. > > Change-Id: I699f3b48785c83260825402d7826231f475f697c Change-Id: Ieeef11b9406964194028b0d81d84bcb63296ae06
2017-08-25Merge "Revert "quantize avx: copy 32x32 implementation""Marco Paniconi
2017-08-25Revert "quantize avx: copy 32x32 implementation"Marco Paniconi
This reverts commit f60d1dcd3de46f72bafc5eeef481bd1a4e203301. Reason for revert: <INSERT REASONING HERE> Failures in AVX/VP9QuantizeTest in nightly tests. Original change's description: > quantize avx: copy 32x32 implementation > > Ensure avx and ssse3 stay in sync by testing them against each other. > > Change-Id: I699f3b48785c83260825402d7826231f475f697c TBR=slavarnway@google.com,johannkoenig@google.com,builds@webmproject.org Change-Id: Ibd38636212269328317dd0721be9d25452113d1c No-Presubmit: true No-Tree-Checks: true No-Try: true
2017-08-24Merge "quantize avx: copy 32x32 implementation"Johann Koenig
2017-08-24Merge "quantize test: skip block was removed"Johann Koenig
2017-08-24quantize avx: copy 32x32 implementationJohann
Ensure avx and ssse3 stay in sync by testing them against each other. Change-Id: I699f3b48785c83260825402d7826231f475f697c
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-14 19:20:14 +0000