aboutsummaryrefslogtreecommitdiffhomepage
path: root/src
Commit message (Collapse)AuthorAge
* Revert "reduce SkDisplacementImageFilter size"Gravatar Mike Reed2017-06-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit ed68a92ca65762f0d0c4060668ceeb9c92506cf9. Reason for revert: breaks layouttests (on BGRA?) Original change's description: > reduce SkDisplacementImageFilter size > > saves ~8K on mac laptop > > - remove templates > - check for 0 0r 0xFF in alpha > > Before: (using modified bench w/o caching) > > 23/23 MB 2 134µs 151µs 146µs 156µs 6% █▇▆▆▆▆▄▁▁▁ 8888 displacement_full_large > 23/23 MB 2 253µs 254µs 255µs 260µs 1% █▄▄▅▂▁▂▁▁▁ 8888 displacement_alpha_large > 24/24 MB 1 410µs 410µs 415µs 445µs 3% █▃▁▂▁▁▁▁▁▁ 8888 displacement_zero_large > 24/24 MB 151 840ns 841ns 853ns 938ns 4% █▁▁▁▁▁▁▂▁▁ 8888 displacement_full_small > 24/24 MB 180 832ns 835ns 836ns 851ns 1% █▂▁▂▂▁▂▂▂▂ 8888 displacement_alpha_small > 24/24 MB 9 60µs 60.9µs 69.9µs 101µs 19% █▄▄▄▁▁▁▁▁▁ 8888 displacement_zero_small > > After: > > 23/23 MB 3 47.4µs 48µs 48.2µs 51.4µs 2% █▂▂▃▃▁▁▁▂▁ 8888 displacement_full_large > 23/23 MB 2 140µs 141µs 145µs 166µs 7% ▂▁▁▁▁▁▁▁▇█ 8888 displacement_alpha_large > 24/24 MB 2 189µs 196µs 197µs 225µs 5% █▃▂▂▂▂▃▂▁▁ 8888 displacement_zero_large > 24/24 MB 134 588ns 594ns 597ns 616ns 1% ▂▁▁█▂▆▂▂▃▄ 8888 displacement_full_small > 24/24 MB 168 590ns 592ns 592ns 599ns 0% ▃▂▁▁▁▁▁█▂▂ 8888 displacement_alpha_small > 24/24 MB 9 8.39µs 8.41µs 8.42µs 8.53µs 0% █▄▃▂▁▁▃▂▁▂ 8888 displacement_zero_small > > Bug: skia: > Change-Id: Ia3b12dc8420b32b65633bb1cf76a15241e420eac > Reviewed-on: https://skia-review.googlesource.com/20181 > Reviewed-by: Mike Klein <mtklein@google.com> > Commit-Queue: Mike Reed <reed@google.com> TBR=mtklein@google.com,robertphillips@google.com,fmalita@chromium.org,reed@google.com Change-Id: I811c5f25a328c75527ce5caa0d9b8e123d535583 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/20182 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Mike Reed <reed@google.com>
* reduce SkDisplacementImageFilter sizeGravatar Mike Reed2017-06-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | saves ~8K on mac laptop - remove templates - check for 0 0r 0xFF in alpha Before: (using modified bench w/o caching) 23/23 MB 2 134µs 151µs 146µs 156µs 6% █▇▆▆▆▆▄▁▁▁ 8888 displacement_full_large 23/23 MB 2 253µs 254µs 255µs 260µs 1% █▄▄▅▂▁▂▁▁▁ 8888 displacement_alpha_large 24/24 MB 1 410µs 410µs 415µs 445µs 3% █▃▁▂▁▁▁▁▁▁ 8888 displacement_zero_large 24/24 MB 151 840ns 841ns 853ns 938ns 4% █▁▁▁▁▁▁▂▁▁ 8888 displacement_full_small 24/24 MB 180 832ns 835ns 836ns 851ns 1% █▂▁▂▂▁▂▂▂▂ 8888 displacement_alpha_small 24/24 MB 9 60µs 60.9µs 69.9µs 101µs 19% █▄▄▄▁▁▁▁▁▁ 8888 displacement_zero_small After: 23/23 MB 3 47.4µs 48µs 48.2µs 51.4µs 2% █▂▂▃▃▁▁▁▂▁ 8888 displacement_full_large 23/23 MB 2 140µs 141µs 145µs 166µs 7% ▂▁▁▁▁▁▁▁▇█ 8888 displacement_alpha_large 24/24 MB 2 189µs 196µs 197µs 225µs 5% █▃▂▂▂▂▃▂▁▁ 8888 displacement_zero_large 24/24 MB 134 588ns 594ns 597ns 616ns 1% ▂▁▁█▂▆▂▂▃▄ 8888 displacement_full_small 24/24 MB 168 590ns 592ns 592ns 599ns 0% ▃▂▁▁▁▁▁█▂▂ 8888 displacement_alpha_small 24/24 MB 9 8.39µs 8.41µs 8.42µs 8.53µs 0% █▄▃▂▁▁▃▂▁▂ 8888 displacement_zero_small Bug: skia: Change-Id: Ia3b12dc8420b32b65633bb1cf76a15241e420eac Reviewed-on: https://skia-review.googlesource.com/20181 Reviewed-by: Mike Klein <mtklein@google.com> Commit-Queue: Mike Reed <reed@google.com>
* Add a verbs iterator to SkPathPrivGravatar Chris Dalton2017-06-17
| | | | | | | | Bug: skia: Change-Id: If846eeac2722b67a80266f01ede7ef03ef027ac8 Reviewed-on: https://skia-review.googlesource.com/20027 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Chris Dalton <csmartdalton@google.com>
* Enable constexpr for VC++ 2017Gravatar Bruce Dawson2017-06-17
| | | | | | | | | | | | | | | VC++ 2017 fixes the internal compiler error which prevented using constexpr for some BlendFormula arrays, so tweak the #ifdefs to enable constexpr for 2017. This gets rid of two constructors and slightly shrinks chrome_child.dll BUG=chromium:341941 Change-Id: Ic4079b2da393ee89dd38da5bbb2d7c8ede8020e8 Reviewed-on: https://skia-review.googlesource.com/20160 Commit-Queue: Mike Klein <mtklein@chromium.org> Reviewed-by: Mike Klein <mtklein@chromium.org>
* support g8 and 4444 in SkRPBlitterGravatar Mike Klein2017-06-17
| | | | | | | | | BUG=skia:6766 Change-Id: I1bf8720cac24092352fd99d8d509252a10270b54 Reviewed-on: https://skia-review.googlesource.com/20020 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Mike Klein <mtklein@chromium.org>
* Trim gEntries down from 24K to 3K.Gravatar Mike Klein2017-06-17
| | | | | | | | | Today gCount maxes out at 66, so 128 seems like plenty of room for now. Change-Id: Id29d7a0232047d07a5ea7c43e4f2d874f055165a Reviewed-on: https://skia-review.googlesource.com/19740 Commit-Queue: Mike Klein <mtklein@chromium.org> Reviewed-by: Mike Reed <reed@google.com>
* refactor lighting imagefilter to save codesizeGravatar Mike Reed2017-06-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Goal was to just remove template args and add virtuals where needed. Ended up having to also move a couple of classes/helpers earlier in the file, but nothing beyond that. Size savings (on Mac laptop) : 67896 Before: 8/8 MB 1 30.6ms 30.7ms 30.7ms 30.9ms 0% █▁▅▁▄▃▄▁▃▃ 8888 lightingspotlitspecular_large 8/8 MB 1 34.2ms 37.6ms 38.5ms 45.9ms 9% ▂▂▃▅▁▅█▂▅▂ 8888 lightingspotlitspecular_small 8/8 MB 1 31.3ms 35ms 34.6ms 38.4ms 7% ▁▄▅▆▆▁▇▃▂█ 8888 lightingdistantlitspecular_large 8/8 MB 1 29.9ms 33.5ms 34.3ms 39.2ms 8% ▁▅█▃▄▇▄▂▄▆ 8888 lightingdistantlitspecular_small 8/8 MB 1 30.4ms 34.1ms 34.5ms 40.7ms 11% ▂▄█▃▁▅▂▅▁▇ 8888 lightingpointlitspecular_large 8/8 MB 1 29.8ms 36.2ms 34.8ms 41.4ms 12% ▂▃▅▁▅▇▂▁▆█ 8888 lightingpointlitspecular_small 8/8 MB 1 16.3ms 19.5ms 20.5ms 26.6ms 17% ▁▆▃▃█▃▂▅▁▆ 8888 lightingspotlitdiffuse_large 8/8 MB 1 17.3ms 19.3ms 19.4ms 23.3ms 10% ▄▁▂▁▆▅▁▃▂█ 8888 lightingspotlitdiffuse_small 8/8 MB 1 12.7ms 14.9ms 17ms 27.9ms 30% ▂▁█▁▂▄▂▂▂▆ 8888 lightingdistantlitdiffuse_large 8/8 MB 1 12.4ms 14.5ms 15.8ms 24.7ms 23% ▁▂▅▂▁▃█▂▃▂ 8888 lightingdistantlitdiffuse_small 8/8 MB 1 13.6ms 14.9ms 16.6ms 22.5ms 22% ▁█▅▁▆▂▁▂▁█ 8888 lightingpointlitdiffuse_large 8/8 MB 1 13.8ms 17ms 16.5ms 19.2ms 11% ▆▆▁▁▆▅▅█▃▂ 8888 lightingpointlitdiffuse_small After: 8/8 MB 1 23.5ms 23.6ms 23.8ms 25.1ms 2% ▃▁█▁▁▂▁▂▁▁ 8888 lightingspotlitspecular_large 8/8 MB 1 23.5ms 23.6ms 23.9ms 24.9ms 2% █▅▂▁▁▁▁▁▆▃ 8888 lightingspotlitspecular_small 8/8 MB 1 21.6ms 21.8ms 21.9ms 22.3ms 1% █▆▃▁▄▄▃▂▄▂ 8888 lightingdistantlitspecular_large 8/8 MB 1 21.6ms 21.7ms 21.9ms 22.7ms 2% █▅▂▂▂▁▂▂▂▃ 8888 lightingdistantlitspecular_small 8/8 MB 1 22.3ms 22.9ms 22.8ms 23.2ms 1% ▇▆▁▄▆▄▃█▄▆ 8888 lightingpointlitspecular_large 8/8 MB 1 22.1ms 22.2ms 22.5ms 23.6ms 3% ██▁▁▂▂▁▁▂▂ 8888 lightingpointlitspecular_small 8/8 MB 1 12.8ms 13ms 13ms 13.3ms 2% ▃▇█▇▁▅▄▃▂▄ 8888 lightingspotlitdiffuse_large 8/8 MB 1 12.8ms 13ms 13ms 13.2ms 1% ▄▃██▄▁▃▅▂▇ 8888 lightingspotlitdiffuse_small 8/8 MB 1 10.6ms 10.7ms 10.8ms 11.1ms 2% ▂▄██▂▁▃▃▂▂ 8888 lightingdistantlitdiffuse_large 8/8 MB 1 10.6ms 10.7ms 10.8ms 11.3ms 2% ▂▂▇█▃▁▂▅▁▂ 8888 lightingdistantlitdiffuse_small 8/8 MB 1 10.9ms 11.1ms 11.2ms 11.6ms 2% ▁▂▇▁█▃▂▂▇▁ 8888 lightingpointlitdiffuse_large 8/8 MB 1 10.9ms 11.1ms 11.1ms 11.5ms 2% ▄▃█▅▃▃▁▃▂▁ 8888 lightingpointlitdiffuse_small Bug: skia: Change-Id: I7542a5ca1f209a732630f646b4ceb4fb08150ce4 Reviewed-on: https://skia-review.googlesource.com/20155 Commit-Queue: Mike Reed <reed@google.com> Reviewed-by: Florin Malita <fmalita@chromium.org>
* Fix logic error when drawing path as nested rectsGravatar Brian Salomon2017-06-16
| | | | | | | | Bug: chromium:732350 Change-Id: I42770e9fa8c201780f16ce8df58b208e08aef640 Reviewed-on: https://skia-review.googlesource.com/20158 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* ICC: SkICC::WriteToICC Description Tag is function of inputGravatar Hal Canary2017-06-16
| | | | | | | | BUG=skia:6720 Change-Id: I038079a6e15f884eb77b84d9c7c75f6b7fbedd37 Reviewed-on: https://skia-review.googlesource.com/20152 Commit-Queue: Hal Canary <halcanary@google.com> Reviewed-by: Matt Sarett <msarett@google.com>
* Revert "Revert "Remove Gr*Gpu includes from header files.""Gravatar Greg Daniel2017-06-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 56af45b03c542ef0bc198e42687bf685aafaae81. Reason for revert: Relanding now that pre CL has relanded Original change's description: > Revert "Remove Gr*Gpu includes from header files." > > This reverts commit 2bd381bffd36cdcffacf606d952547ce66fed7e9. > > Reason for revert: Need to revert change this landed on top of > > Original change's description: > > Remove Gr*Gpu includes from header files. > > > > Some general cleanup to try to keep use of the backend gpu class in cpps. > > > > Bug: skia: > > Change-Id: I1a28e0e5c2e55562b3fff56c284c3405b5b6d8bc > > Reviewed-on: https://skia-review.googlesource.com/20057 > > Reviewed-by: Robert Phillips <robertphillips@google.com> > > Commit-Queue: Greg Daniel <egdaniel@google.com> > > TBR=egdaniel@google.com,robertphillips@google.com > > Change-Id: I5b27d50654a17502c056d59b2d4f9609376d189e > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: skia: > Reviewed-on: https://skia-review.googlesource.com/20064 > Reviewed-by: Greg Daniel <egdaniel@google.com> > Commit-Queue: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,robertphillips@google.com Change-Id: Id1656a7708223bc4dcc8eb4444ef41ada61be5d0 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/20148 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Remove GrRectOpFactory::MakeAAFillWithDevRectGravatar Brian Salomon2017-06-16
| | | | | | | | | This was only used in a rare case. Change-Id: I69aaca40ce6529a492311fa0893a85adcc08aa4d Reviewed-on: https://skia-review.googlesource.com/20143 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Rename GrDrawOp::xpRequiresDstTexture to finalize and change return type to ↵Gravatar Brian Salomon2017-06-16
| | | | | | | | | | an enum Bug: skia: Change-Id: I0f9d9d3ef1dd20821f171f7d6237491921fbbd97 Reviewed-on: https://skia-review.googlesource.com/20142 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Greg Daniel <egdaniel@google.com>
* Revert "Revert "Add API for flushing surfaces with gpu semaphores""Gravatar Greg Daniel2017-06-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 7292231905c34ed290ba479338f26b56ae2a7792. This change relands the original plus the follow on change: https://skia-review.googlesource.com/20059. Additionally it adds a blacklist for the mac intel bots which don't see to respect the added fences on the GPU. Original change's description: > Revert "Add API for flushing surfaces with gpu semaphores" > > This reverts commit 66366c697853e906d961ae691e2bc5209cdcfa62. > > Reason for revert: Failing test on mac bots > > Original change's description: > > Add API for flushing surfaces with gpu semaphores > > > > BUG=skia: > > > > Change-Id: Ia4bfef784cd5f2516ceccafce958be18a86f91d1 > > Reviewed-on: https://skia-review.googlesource.com/11488 > > Commit-Queue: Greg Daniel <egdaniel@google.com> > > Reviewed-by: Brian Salomon <bsalomon@google.com> > > Reviewed-by: Forrest Reiling <freiling@google.com> > > TBR=egdaniel@google.com,jvanverth@google.com,bsalomon@google.com,brianosman@google.com,freiling@google.com > > Change-Id: I75633a2732d2d48b1926f9ad818a9f1a9196d211 > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: skia: > Reviewed-on: https://skia-review.googlesource.com/20063 > Commit-Queue: Greg Daniel <egdaniel@google.com> > Reviewed-by: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,jvanverth@google.com,bsalomon@google.com,brianosman@google.com,freiling@google.com Change-Id: I4dc6c0e1deb0398eeb165a34f0a26af7a58259f1 Reviewed-on: https://skia-review.googlesource.com/20141 Commit-Queue: Greg Daniel <egdaniel@google.com> Reviewed-by: Greg Daniel <egdaniel@google.com>
* Hide GrGpuResourceRef from prying eyesGravatar Robert Phillips2017-06-16
| | | | | | | Change-Id: I5535b8cc8e82ed4b6fe847b96b417531b4bae348 Reviewed-on: https://skia-review.googlesource.com/19967 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Add GrTextureProxy.h to GrAHardwareBufferImageGenerator.cpp (for Android roll)Gravatar Robert Phillips2017-06-16
| | | | | | | | | TBR=rmistry@google.com Change-Id: Id08f821447dc6b1229667a07b7845f87399b1cce Reviewed-on: https://skia-review.googlesource.com/20140 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Make GrAAConvexPathRenderer use vertex colorGravatar Brian Salomon2017-06-15
| | | | | | | Change-Id: Iea2902a73fad9493d69b24e8bc88cdafd6aea9fc Reviewed-on: https://skia-review.googlesource.com/20060 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Revert "Add API for flushing surfaces with gpu semaphores"Gravatar Greg Daniel2017-06-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 66366c697853e906d961ae691e2bc5209cdcfa62. Reason for revert: Failing test on mac bots Original change's description: > Add API for flushing surfaces with gpu semaphores > > BUG=skia: > > Change-Id: Ia4bfef784cd5f2516ceccafce958be18a86f91d1 > Reviewed-on: https://skia-review.googlesource.com/11488 > Commit-Queue: Greg Daniel <egdaniel@google.com> > Reviewed-by: Brian Salomon <bsalomon@google.com> > Reviewed-by: Forrest Reiling <freiling@google.com> TBR=egdaniel@google.com,jvanverth@google.com,bsalomon@google.com,brianosman@google.com,freiling@google.com Change-Id: I75633a2732d2d48b1926f9ad818a9f1a9196d211 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/20063 Commit-Queue: Greg Daniel <egdaniel@google.com> Reviewed-by: Greg Daniel <egdaniel@google.com>
* Revert "Remove Gr*Gpu includes from header files."Gravatar Greg Daniel2017-06-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 2bd381bffd36cdcffacf606d952547ce66fed7e9. Reason for revert: Need to revert change this landed on top of Original change's description: > Remove Gr*Gpu includes from header files. > > Some general cleanup to try to keep use of the backend gpu class in cpps. > > Bug: skia: > Change-Id: I1a28e0e5c2e55562b3fff56c284c3405b5b6d8bc > Reviewed-on: https://skia-review.googlesource.com/20057 > Reviewed-by: Robert Phillips <robertphillips@google.com> > Commit-Queue: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,robertphillips@google.com Change-Id: I5b27d50654a17502c056d59b2d4f9609376d189e No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/20064 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Improve computeMatrices singular matrix handling.Gravatar Ben Wagner2017-06-15
| | | | | | | | | | | | | | | The existing singular matrix detection in computeMatrices is sufficient but not necessary. Since compute matrices is already doing a QR decomposition, use that to determine singularity instead. This is both faster and more accurate than the previous method for the common case. BUG=https://bugzilla.mozilla.org/show_bug.cgi?id=1305085 Change-Id: Iccef8368527b45e4eb565eddbebbbcf41ca66a2c Reviewed-on: https://skia-review.googlesource.com/20054 Reviewed-by: Lee Salzman <lsalzman@mozilla.com> Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
* Remove Gr*Gpu includes from header files.Gravatar Greg Daniel2017-06-15
| | | | | | | | | | Some general cleanup to try to keep use of the backend gpu class in cpps. Bug: skia: Change-Id: I1a28e0e5c2e55562b3fff56c284c3405b5b6d8bc Reviewed-on: https://skia-review.googlesource.com/20057 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Migrate AAConvexPathOp off of GrLegacyMeshDrawOpGravatar Brian Salomon2017-06-15
| | | | | | | Change-Id: If6df4292ec5c78a7f2fdde0a4540f7327e69e294 Reviewed-on: https://skia-review.googlesource.com/20058 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Revert "explore always using SkRasterPipeline for color space transforms"Gravatar Mike Klein2017-06-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 58564425e5acb5911cc3719f9c1e39190cc829b8. Reason for revert: iOS Original change's description: > explore always using SkRasterPipeline for color space transforms > > On my trashcan, bench times change: mostly slowdowns, some speedups. > On the other hand, this cuts about 270K of code out of Skia. > > Change-Id: Ib1069792508ced361c11ea809b0b8187f5e28a5c > Reviewed-on: https://skia-review.googlesource.com/19744 > Reviewed-by: Matt Sarett <msarett@google.com> > Commit-Queue: Mike Klein <mtklein@chromium.org> TBR=mtklein@chromium.org,msarett@google.com Change-Id: I26469e174f0281597ac5720ec790686ef3fbb4e0 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://skia-review.googlesource.com/20036 Reviewed-by: Mike Klein <mtklein@google.com> Commit-Queue: Mike Klein <mtklein@google.com>
* Add API for flushing surfaces with gpu semaphoresGravatar Greg Daniel2017-06-15
| | | | | | | | | | BUG=skia: Change-Id: Ia4bfef784cd5f2516ceccafce958be18a86f91d1 Reviewed-on: https://skia-review.googlesource.com/11488 Commit-Queue: Greg Daniel <egdaniel@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com> Reviewed-by: Forrest Reiling <freiling@google.com>
* Readd GrResourceProvider.h to GrAHardwareBufferImageGenerator.cpp (for ↵Gravatar Robert Phillips2017-06-15
| | | | | | | | | | Android roll) TBR=rmistry@google.com Change-Id: If6aa6a024bcc87846c3b24b0d004ffe7177895a8 Reviewed-on: https://skia-review.googlesource.com/20052 Reviewed-by: Robert Phillips <robertphillips@google.com>
* Add processor info dumping to non-legacy mesh draw opsGravatar Brian Salomon2017-06-15
| | | | | | | Change-Id: I2ee77f0971a1b627905ac547bc0511042c40ac38 Reviewed-on: https://skia-review.googlesource.com/19816 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com>
* Revert "implemented mustImplementGSInvocationsWithLoop workaround in sksl"Gravatar Greg Daniel2017-06-15
| | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit d5d323f45777c4766ddf4c74eedff36ee5d9bc56. Reason for revert: Looks like this might be breaking vulkan Original change's description: > implemented mustImplementGSInvocationsWithLoop workaround in sksl > > Bug: skia: > Change-Id: I8434be89537a8baf9e9c5c7a643d28c03108f4ea > Reviewed-on: https://skia-review.googlesource.com/18154 > Reviewed-by: Chris Dalton <csmartdalton@google.com> > Commit-Queue: Ethan Nicholas <ethannicholas@google.com> TBR=csmartdalton@google.com,ethannicholas@google.com Change-Id: I606aad071d3c084eb7eb7ec6c1c1a3cf10479cec No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/20045 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Remove GrGpu.h from GrResourceProvider.hGravatar Robert Phillips2017-06-15
| | | | | | | | | Noticed this while cleaning up GrResourceProvider.h's usage. It seems like a powerful header to just be splashing around. Change-Id: I686fea61354a7e3c1c759627ffe4a560f7945f83 Reviewed-on: https://skia-review.googlesource.com/20040 Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-by: Greg Daniel <egdaniel@google.com>
* implemented mustImplementGSInvocationsWithLoop workaround in skslGravatar Ethan Nicholas2017-06-15
| | | | | | | | Bug: skia: Change-Id: I8434be89537a8baf9e9c5c7a643d28c03108f4ea Reviewed-on: https://skia-review.googlesource.com/18154 Reviewed-by: Chris Dalton <csmartdalton@google.com> Commit-Queue: Ethan Nicholas <ethannicholas@google.com>
* Clean up GrResourceProvider usageGravatar Robert Phillips2017-06-15
| | | | | | | | | | | | | The only substantive changes are the removal of GrProxy instantiation in: SkGpuBlurUtils::GaussianBlur GrSimpleTextureEffect::Make* Change-Id: I10970609693bd6ff5b3a3c21b41d82642bb277bc Reviewed-on: https://skia-review.googlesource.com/19965 Reviewed-by: Greg Daniel <egdaniel@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Revert "Revert "Converts remaining rect ops from GrLegacyMeshDrawOp to ↵Gravatar Brian Salomon2017-06-15
| | | | | | | | | | | | GrMeshDrawOp subclasses."" This reverts commit a0485d94529905e76320b7aa941a0d94b5578ac2. Bug: skia: Change-Id: If50b2d1af285a6c3c943373bb7258b56e1af28d6 Reviewed-on: https://skia-review.googlesource.com/19961 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* explore always using SkRasterPipeline for color space transformsGravatar Mike Klein2017-06-15
| | | | | | | | | | On my trashcan, bench times change: mostly slowdowns, some speedups. On the other hand, this cuts about 270K of code out of Skia. Change-Id: Ib1069792508ced361c11ea809b0b8187f5e28a5c Reviewed-on: https://skia-review.googlesource.com/19744 Reviewed-by: Matt Sarett <msarett@google.com> Commit-Queue: Mike Klein <mtklein@chromium.org>
* delete lowp plusGravatar Mike Klein2017-06-15
| | | | | | | | | | | | | | | | | | | | | | | I have figured out how to implement lowp clamp_1/clamp_a, and implementing clamp_1 would make lowp plus active. But... the way we have factored blend modes requires us to be able to lerp between the dst and possibly-out-of-range src values. This is not possible in lowp. If we try to multiply with values in [0x8001,0xffff], we'll just get garbage. We'll clamp them back in range, but sadly clamped garbage is still garbage. So the simplest thing to do is keep plus blends in floats. This CL doesn't even change that... we'd use floats before and after it. It just removes the lowp plus stage code that is both dead and buggy. As far as I can tell, no other drawing is currently gated by lowp missing clamp_1 or clamp_a. Change-Id: I55b73c840614f1bff9cd610dff90ca5e2b5c73e5 Reviewed-on: https://skia-review.googlesource.com/19909 Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Mike Klein <mtklein@google.com>
* Revert "Converts remaining rect ops from GrLegacyMeshDrawOp to GrMeshDrawOp ↵Gravatar Brian Salomon2017-06-14
| | | | | | | | | | | | | | | | | | | subclasses." This reverts commit 1ec03f33cf493352174c748662d4a3cca29f78fd. Revert "Fix logic reversal in NonAAFillRectOp test factory" This reverts commit 89c1c2552ec5b9ad8949988f7c9532a298b55987. Reason: Unexpected GM changes. Bug: skia: Change-Id: I9edf5f0e4a54b5cad86bd438a505aaaef38563de Reviewed-on: https://skia-review.googlesource.com/19960 Reviewed-by: Brian Salomon <bsalomon@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* SkPDF: Glyph-by-glyph bounds-reject.Gravatar Hal Canary2017-06-14
| | | | | | | | BUG=chromium:682644 Change-Id: I9d20dd81886d4de4e74ba043349d456230c89cb1 Reviewed-on: https://skia-review.googlesource.com/19706 Commit-Queue: Hal Canary <halcanary@google.com> Reviewed-by: Ben Wagner <bungeman@google.com>
* Fix logic reversal in NonAAFillRectOp test factoryGravatar Brian Salomon2017-06-14
| | | | | | | Change-Id: I3c8514ca58377f90adeb2b6191bff18c08f1d6e4 Reviewed-on: https://skia-review.googlesource.com/19920 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Prevent onFlushCB created opLists from being grouped with the normal ones (new)Gravatar Robert Phillips2017-06-14
| | | | | | | Change-Id: I8ee04bb40a51f5b432b4c896a0c3e50559977463 Reviewed-on: https://skia-review.googlesource.com/19902 Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-by: Chris Dalton <csmartdalton@google.com>
* Converts remaining rect ops from GrLegacyMeshDrawOp to GrMeshDrawOp subclasses.Gravatar Brian Salomon2017-06-14
| | | | | | | | | | | Consolidates op factory functions to a rewritten GrRectOpFactory. Removes GrRenderTargetContext::drawNonAAFilledRect() in favor of creating and adding ops directly by the callers. Change-Id: I57e5fc739bf4e92b4a4710c739e6d22cce82a479 Reviewed-on: https://skia-review.googlesource.com/17711 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com>
* Drop the ref on the GrOpList's target in makeClosed (take 2)Gravatar Robert Phillips2017-06-14
| | | | | | | | | | | Bug: 729233 TBR=bsalomon@google.com Change-Id: I5c9a0cb793c7c6564ad355a4a63b29fdc12f6cd7 Reviewed-on: https://skia-review.googlesource.com/19860 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Fix failing dm with GrPrimitiveType::kLinesAdjacencyGravatar Chris Dalton2017-06-14
| | | | | | | | | Bug: skia: Change-Id: I9a73ec49b2bf10700ca95c2073b84cfb606d2fec Reviewed-on: https://skia-review.googlesource.com/19722 Commit-Queue: Chris Dalton <csmartdalton@google.com> Reviewed-by: Ravi Mistry <rmistry@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com>
* Revert "Revert "Go back to using dual source blending for lcd src-over even ↵Gravatar Greg Daniel2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | with non-opaque color"" This reverts commit 7d6fe0b9964d139de64ca88c46d87eba41c7f84e. Reason for revert: Relanding with fix Original change's description: > Revert "Go back to using dual source blending for lcd src-over even with non-opaque color" > > This reverts commit b54bdef86eb5cf63b94588afaa9197f49374a5f5. > > Reason for revert: breaking some bots > Original change's description: > > Go back to using dual source blending for lcd src-over even with non-opaque color > > > > This is change is currently still safe since earlier in Skia we are still requiring > > the dst to be opaque. The change is a workaround to spots where trying to read the > > dst to do in shader blending is failing for some reason. This also should give back > > a little performance since doing dual source blending should be better than shader > > blends. > > > > Bug: chromium:732341 > > Change-Id: I795f8a520f87f3fbf5d63a9509fbd9f394ea2b29 > > Reviewed-on: https://skia-review.googlesource.com/19703 > > Reviewed-by: Brian Salomon <bsalomon@google.com> > > Commit-Queue: Greg Daniel <egdaniel@google.com> > > TBR=egdaniel@google.com,bsalomon@google.com > > Change-Id: Ibb9bc1ef4ec5967dabcd62c81f62c0989c14fbb8 > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: chromium:732341 > Reviewed-on: https://skia-review.googlesource.com/19815 > Reviewed-by: Greg Daniel <egdaniel@google.com> > Commit-Queue: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,bsalomon@google.com Bug: chromium:732341 Change-Id: I7481755a9aa64364371d8149af4458fc2c15c8aa Reviewed-on: https://skia-review.googlesource.com/19840 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Revert "Drop the ref on the GrOpList's target in makeClosed"Gravatar Robert Phillips2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit dcd499caed823f23bc70c07df7804a6dc1306606. Reason for revert: memory issue Original change's description: > Drop the ref on the GrOpList's target in makeClosed > > Bug: 729233 > Change-Id: Ifb66b745e604d7f7c22c2907bcffa91d2086236d > Reviewed-on: https://skia-review.googlesource.com/19495 > Commit-Queue: Robert Phillips <robertphillips@google.com> > Reviewed-by: Brian Salomon <bsalomon@google.com> TBR=bsalomon@google.com,robertphillips@google.com Change-Id: I4269c696c57a93cdd056e7c3a887c35b8765ac6d No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 729233 Reviewed-on: https://skia-review.googlesource.com/19793 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Force AAType to MSAA if the render target has MSAA and the API doesn't ↵Gravatar Brian Salomon2017-06-14
| | | | | | | | | support disabling it. Change-Id: I88c29b8117fa82c6f41166b9333537a06bb841cc Reviewed-on: https://skia-review.googlesource.com/19818 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Drop the ref on the GrOpList's target in makeClosedGravatar Robert Phillips2017-06-14
| | | | | | | | Bug: 729233 Change-Id: Ifb66b745e604d7f7c22c2907bcffa91d2086236d Reviewed-on: https://skia-review.googlesource.com/19495 Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-by: Brian Salomon <bsalomon@google.com>
* Revert "Go back to using dual source blending for lcd src-over even with ↵Gravatar Greg Daniel2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | non-opaque color" This reverts commit b54bdef86eb5cf63b94588afaa9197f49374a5f5. Reason for revert: breaking some bots Original change's description: > Go back to using dual source blending for lcd src-over even with non-opaque color > > This is change is currently still safe since earlier in Skia we are still requiring > the dst to be opaque. The change is a workaround to spots where trying to read the > dst to do in shader blending is failing for some reason. This also should give back > a little performance since doing dual source blending should be better than shader > blends. > > Bug: chromium:732341 > Change-Id: I795f8a520f87f3fbf5d63a9509fbd9f394ea2b29 > Reviewed-on: https://skia-review.googlesource.com/19703 > Reviewed-by: Brian Salomon <bsalomon@google.com> > Commit-Queue: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,bsalomon@google.com Change-Id: Ibb9bc1ef4ec5967dabcd62c81f62c0989c14fbb8 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: chromium:732341 Reviewed-on: https://skia-review.googlesource.com/19815 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Go back to using dual source blending for lcd src-over even with non-opaque ↵Gravatar Greg Daniel2017-06-14
| | | | | | | | | | | | | | | | color This is change is currently still safe since earlier in Skia we are still requiring the dst to be opaque. The change is a workaround to spots where trying to read the dst to do in shader blending is failing for some reason. This also should give back a little performance since doing dual source blending should be better than shader blends. Bug: chromium:732341 Change-Id: I795f8a520f87f3fbf5d63a9509fbd9f394ea2b29 Reviewed-on: https://skia-review.googlesource.com/19703 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Revert "Revert "Do not return Index8 from ↵Gravatar Matt Sarett2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SkAndroidCodec::computeOutputColorType"" This reverts commit 81c83a7db4e524b19d33bf7c8a9b537b9d606c93. Reason for revert: <INSERT REASONING HERE> Original change's description: > Revert "Do not return Index8 from SkAndroidCodec::computeOutputColorType" > > This reverts commit b6f4767294261dca3beef6f280c4bac69df3f930. > > Reason for revert: This breaks CTS tests in Android. Doh. > > Original change's description: > > Do not return Index8 from SkAndroidCodec::computeOutputColorType > > > > Given that this is the only known use of Index8 color type, > > this is essentially an experimental delete. > > > > Bug: skia:6620 > > Change-Id: Ib363d237e0217f6e7f461a62e54d32892c428095 > > Reviewed-on: https://skia-review.googlesource.com/10586 > > Reviewed-by: Leon Scroggins <scroggo@google.com> > > Commit-Queue: Matt Sarett <msarett@google.com> > > TBR=msarett@google.com,scroggo@google.com,reed@google.com > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: skia:6620 > > Change-Id: I2b44c695b8b95659520e9532901f636f56e01e2a > Reviewed-on: https://skia-review.googlesource.com/19084 > Reviewed-by: Matt Sarett <msarett@google.com> > Commit-Queue: Matt Sarett <msarett@google.com> TBR=msarett@google.com,scroggo@google.com,reviews@skia.org,reed@google.com # Not skipping CQ checks because original CL landed > 1 day ago. Bug: skia:6620 Change-Id: I7e4e3a5ec068102244ad3a0259aa6aded4f12f36 Reviewed-on: https://skia-review.googlesource.com/19802 Reviewed-by: Matt Sarett <msarett@google.com> Commit-Queue: Matt Sarett <msarett@google.com>
* use #ifdef instead of #if for SK_VULKANGravatar Mike Reed2017-06-14
| | | | | | | | | | Try to fix android roll Bug: skia: Change-Id: Iae32027b299d8a975d3caf70f0f8910541ca972d Reviewed-on: https://skia-review.googlesource.com/19800 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Mike Reed <reed@google.com>
* Revert "Revert "Update skia to use ifdefs for Vulkan code instead of dummy ↵Gravatar Robert Phillips2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | header"" This reverts commit fad9e3f54112ea8c8bb6bb72384f47b9759578f5. Reason for revert: Can't find the error message anymore (?!?) Let's try again shall we Original change's description: > Revert "Update skia to use ifdefs for Vulkan code instead of dummy header" > > This reverts commit c0f8e426c59eec6c720b8e1329dcb966cf1b6800. > > Reason for revert: Experiment to see if this will unblock the Android roll > > Original change's description: > > Update skia to use ifdefs for Vulkan code instead of dummy header > > > > Bug: skia:6721 > > Change-Id: I80a4c9f2acc09c174497f625c50ed12a8bb76505 > > Reviewed-on: https://skia-review.googlesource.com/19547 > > Reviewed-by: Mike Klein <mtklein@google.com> > > Commit-Queue: Greg Daniel <egdaniel@google.com> > > TBR=egdaniel@google.com,mtklein@google.com,bsalomon@google.com > > Change-Id: Ib51c1672570f2071a17b6fbde692a5174b0358ce > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: skia:6721 > Reviewed-on: https://skia-review.googlesource.com/19724 > Reviewed-by: Robert Phillips <robertphillips@google.com> > Commit-Queue: Robert Phillips <robertphillips@google.com> TBR=egdaniel@google.com,mtklein@google.com,bsalomon@google.com,robertphillips@google.com Change-Id: Iecef7ddcfe31d82938336120a4193525ac6693be No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia:6721 Reviewed-on: https://skia-review.googlesource.com/19782 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Revert "Revert "use pipeline for non-opt xfermodes""Gravatar Mike Reed2017-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 4f8c695736c8ae4fed2190a8e1301a4f4a979898. Reason for revert: suppression re-landed in chrome Original change's description: > Revert "use pipeline for non-opt xfermodes" > > This reverts commit e93cf97175d348ce1400762bdd8d9acabdd29766. > > Reason for revert: unblock Chrome roll > > Original change's description: > > use pipeline for non-opt xfermodes > > > > Produces slightly different results for Hue, Saturation, Color, Luminosity, > > seemingly around the aa edging. > > > > Bug: skia: > > Change-Id: I6364818c9788863e5fad6d14cad4797d073dbea3 > > Reviewed-on: https://skia-review.googlesource.com/19554 > > Reviewed-by: Mike Klein <mtklein@google.com> > > Commit-Queue: Mike Reed <reed@google.com> > > TBR=mtklein@google.com,fmalita@chromium.org,reed@google.com > > Change-Id: Ib85a51753d21ce778fa5cfedd6b7d1b5b2b87096 > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: skia: > Reviewed-on: https://skia-review.googlesource.com/19745 > Reviewed-by: Mike Klein <mtklein@chromium.org> > Commit-Queue: Mike Klein <mtklein@chromium.org> TBR=mtklein@chromium.org,mtklein@google.com,fmalita@chromium.org,reed@google.com Change-Id: I1201b5abae1514edf9143927064ce060095592f4 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia: Reviewed-on: https://skia-review.googlesource.com/19760 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Mike Reed <reed@google.com>
* Add GrTexture.h include to GrAHardwareBufferImageGenerator.cppGravatar Robert Phillips2017-06-13
| | | | | | | TBR=bsalomon@google.com Change-Id: I0cee1dc3f6cdfeaf86d1fbe2eb033c78ff8fe686 Reviewed-on: https://skia-review.googlesource.com/19725 Reviewed-by: Robert Phillips <robertphillips@google.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-06 16:50:50 +0000