aboutsummaryrefslogtreecommitdiffhomepage
path: root/src
Commit message (Collapse)AuthorAge
* Revert "Revert "SkAdvancedTypefaceMetrics: factor out GlyphToUnicode""Gravatar Hal Canary2018-05-09
| | | | | | | | | This reverts commit 97c1108607584b6050a6880d6ce22846e4913a92. Change-Id: Ic3c6addc64ced39766bbee3e10b4d88faf61ba2f Reviewed-on: https://skia-review.googlesource.com/127021 Reviewed-by: Hal Canary <halcanary@google.com> Commit-Queue: Hal Canary <halcanary@google.com>
* Add fPreferExternalImagesOverES3 to GrContextOptionsGravatar Brian Osman2018-05-09
| | | | | | | | | | | | | This forces us to use the ES2 shading language when we have an ES3 context, but the driver claims to only support GL_OES_EGL_image_external (and not the _essl3 variant). Many of these devices will work correctly if we blindly enable both extensions, but there are some that won't. Bug: skia: Change-Id: Id632003a1905ea61b46166befd30905a57cead69 Reviewed-on: https://skia-review.googlesource.com/126681 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Osman <brianosman@google.com>
* Fix clipping of SkMatrixImageFilter inputGravatar Robert Phillips2018-05-09
| | | | | | | | Bug: skia:7931 Change-Id: I3e424dcd5c0b04542976f4f59a5836e7c3079d06 Reviewed-on: https://skia-review.googlesource.com/127020 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Remove SK_IGNORE_MASK_FILTER_GLYPH_FIX.Gravatar Ben Wagner2018-05-09
| | | | | | | | | | | This removes the SK_IGNORE_MASK_FILTER_GLYPH_FIX flag check and the code it guards. This flag existed for rebaselining and the rebaselines have been done. Change-Id: Ic2d6850051a7f4dcbf2e31c6174db1248ce9367a Reviewed-on: https://skia-review.googlesource.com/126925 Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
* Revert "SkAdvancedTypefaceMetrics: factor out GlyphToUnicode"Gravatar Greg Daniel2018-05-09
| | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 1c2bcd8b14e029a70e88b1e81acd29553cab0d1c. Reason for revert: breaking chrome roll Original change's description: > SkAdvancedTypefaceMetrics: factor out GlyphToUnicode > > Change-Id: Iedce8c1ea2c405d5ab64ccac353970d5cd2b9d63 > Reviewed-on: https://skia-review.googlesource.com/126507 > Commit-Queue: Hal Canary <halcanary@google.com> > Reviewed-by: Ben Wagner <bungeman@google.com> TBR=halcanary@google.com,bungeman@google.com,reed@google.com Change-Id: Ib1ff8484ffd09cdb88d461ac00745aa32c191124 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://skia-review.googlesource.com/127000 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* clean up some SkDebugf()Gravatar Mike Klein2018-05-09
| | | | | | | | | | | These annoy the crap out of me every time I run DM. Change-Id: I76edf1d5a5c1931a93d208ba1d8db5220987e62d Reviewed-on: https://skia-review.googlesource.com/126929 Commit-Queue: Mike Klein <mtklein@chromium.org> Commit-Queue: Florin Malita <fmalita@chromium.org> Auto-Submit: Mike Klein <mtklein@chromium.org> Reviewed-by: Florin Malita <fmalita@chromium.org>
* ccpr: Cleanup on-flush path statsGravatar Chris Dalton2018-05-09
| | | | | | | | | | | | Adds a wrapper struct to limit the number of free variables. Fixes a bug where we preallocated draw instances for clip paths. Counts conic weights and preallocates their buffer as well. Bug: skia: Change-Id: I72779c9017322a0dc64461c8cb927f975406b9af Reviewed-on: https://skia-review.googlesource.com/126844 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Chris Dalton <csmartdalton@google.com>
* Cleanup SK_IGNORE_BLUR_IMAGE_FILTER_FIX (since Chrome no longer uses it)Gravatar Robert Phillips2018-05-09
| | | | | | | Change-Id: I3eebd6233d6f6a8b43ce0d865ffafd85d58f0c16 Reviewed-on: https://skia-review.googlesource.com/126926 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* When validating transfer fns, allow D > 1Gravatar Brian Osman2018-05-09
| | | | | | | | | | | | | | This matches behavior in skcms. Previously, we could get a profile with a large D value, skcms would decide it was okay, but then we'd be unable to create an SkColorSpace from it. Bug: chromium:841110 Change-Id: I2418a0b770172bff62a14d2d4d0fc7ba70f33e0b Reviewed-on: https://skia-review.googlesource.com/126920 Commit-Queue: Brian Osman <brianosman@google.com> Commit-Queue: Mike Klein <mtklein@google.com> Reviewed-by: Mike Klein <mtklein@google.com> Auto-Submit: Brian Osman <brianosman@google.com>
* SkAdvancedTypefaceMetrics: factor out GlyphToUnicodeGravatar Hal Canary2018-05-09
| | | | | | | Change-Id: Iedce8c1ea2c405d5ab64ccac353970d5cd2b9d63 Reviewed-on: https://skia-review.googlesource.com/126507 Commit-Queue: Hal Canary <halcanary@google.com> Reviewed-by: Ben Wagner <bungeman@google.com>
* fonts: Fix a trace name in SkRemoteGlyphCacheGravatar Khushal2018-05-09
| | | | | | | | | TBR=herb@google.com Change-Id: I5df4295a94d9af93e29ebf64b06f5259dc54ce90 Reviewed-on: https://skia-review.googlesource.com/126842 Reviewed-by: Khusal Sagar <khushalsagar@chromium.org> Commit-Queue: Khusal Sagar <khushalsagar@chromium.org>
* fonts: Fix subpixel position mismatch in font remotingGravatar Khushal2018-05-08
| | | | | | | | | | R=herb@google.com Bug: skia:7786 Change-Id: Icb52be92d193eb17ac20734bfa7cb00370c6ea21 Reviewed-on: https://skia-review.googlesource.com/126664 Commit-Queue: Khusal Sagar <khushalsagar@chromium.org> Reviewed-by: Herb Derby <herb@google.com>
* Treat SkFILEStream's fOriginalOffset as the startGravatar Leon Scroggins III2018-05-08
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Bug: b/78866720 ::rewind() rewinds to fOriginalOffset ::seek(position) seeks to position + fOriginalOffset ::move(offset) will not move < fOriginalOffset ::getPosition() returns position relative to fOriginalOffset ::getLength() returns full size minus fOriginalOffset ::duplicate() and ::fork() pass on fOriginalOffset Android may create an SkFILEStream using a file descriptor whose offset is at the beginning of the data that Android cares about. Treat all positions in SkFILEStream as relative to that original offset. This allows AnimatedImageDrawable to read directly from the SkFILEStream, rather than using an SkFrontBufferedStream and forcing SkGifCodec to cache data for later use. This fixes a TODO that was introduced in https://skia-review.googlesource.com/c/skia/+/9498 and takes it a step further. In that CL, bungeman@ and I discussed the change and decided to "leave this alone for now to avoid changing behavior". Doing a code search today, the only two callers want the new behavior. Change-Id: I9211394d5b730adf528fac0df0af7a664b1295be Reviewed-on: https://skia-review.googlesource.com/126511 Reviewed-by: Ben Wagner <bungeman@google.com> Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Leon Scroggins <scroggo@google.com>
* Only enable noperspective with ES3 shading languageGravatar Brian Osman2018-05-08
| | | | | | | | | | | | | | This will never happen in real usage - the extension requires ES3 support. However, we're likely to be testing/running with ES3 and ES2 shading language to work around driver bugs on other devices. Across many devices, this was the only issue I found when forcing ES2 shading language, so it seemed worth fixing. Change-Id: I8b75c6935827c1a24c712cec4b36515581aea353 Reviewed-on: https://skia-review.googlesource.com/126640 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Brian Osman <brianosman@google.com>
* SkBlurMask to blur all the things.Gravatar Ben Wagner2018-05-08
| | | | | | | | | | | | | | | SkBlurMask used to handle only A8 as input, and was recently extended to ARGB as input. This adds LCD16 and A1 as inputs as well, allowing SkScalerContext to no longer force all requests with a mask filter to A8. This also fixes the shadows for LCD16 and A1 glyphs, which at small sigma were quite obviously wrong since they were generated from an A8 version. Change-Id: I151cdbf5556d18c2938da990ecee5bcd8ee31be8 Reviewed-on: https://skia-review.googlesource.com/126615 Reviewed-by: Herb Derby <herb@google.com> Reviewed-by: Jim Van Verth <jvanverth@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
* Use device and key descriptorsGravatar Herb Derby2018-05-08
| | | | | | | Change-Id: I2ac13303376f3d1464dd1e259637374a9c5ef237 Reviewed-on: https://skia-review.googlesource.com/125823 Commit-Queue: Herb Derby <herb@google.com> Reviewed-by: Khushal Sagar <khushalsagar@google.com>
* Add a drawPaths method to GrCCPathProcessorGravatar Chris Dalton2018-05-08
| | | | | | | | Bug: skia: Change-Id: I5d3762203cbe1d823ea3f3967240bcab76651f2c Reviewed-on: https://skia-review.googlesource.com/126662 Reviewed-by: Brian Salomon <bsalomon@google.com> Commit-Queue: Chris Dalton <csmartdalton@google.com>
* Fix a bug in SkGpuBlurUtils::GaussianBlur (take 3)Gravatar Robert Phillips2018-05-08
| | | | | | | | | | | | | This fixes the remaining imageblurclampmode bug and a preexisting un-reported error in the imageblurrepeatmode GM. This relies on https://chromium-review.googlesource.com/c/chromium/src/+/1044113 (Add a flag to disable a blur imagefilter bug fix in Skia) landing in Chrome first. Bug: skia:7765 Change-Id: I66e27384541002829b2bd4b42d9cbda837aa7c3d Reviewed-on: https://skia-review.googlesource.com/125961 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-on: https://skia-review.googlesource.com/126220
* Avoid rerunning path renderer chain when path known to require swGravatar Brian Salomon2018-05-08
| | | | | | | Change-Id: If8bdeadf95543917a6e00570b5806a3ee23aa335 Reviewed-on: https://skia-review.googlesource.com/126508 Commit-Queue: Robert Phillips <robertphillips@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com>
* text-blob: Fix OOM issues during blob deserialization.Gravatar Khushal2018-05-07
| | | | | | | | | | | | Limit the number of glyphs allowed during blob deserialization. Allocating for an arbitrary number can lead to OOM issues. Bug: 840096 Change-Id: I1673fd312438b99ff76dadcfdc9f9427acdca13b R=bsalomon@chromium.org Reviewed-on: https://skia-review.googlesource.com/126603 Commit-Queue: Khusal Sagar <khushalsagar@chromium.org> Reviewed-by: Florin Malita <fmalita@chromium.org>
* Visit the image proxy in GrLatticeOpGravatar Brian Salomon2018-05-07
| | | | | | | | | Bug: b/77917978 Change-Id: Ib6b40ce891f576f4a1c7d624a2409f3bba36b344 Reviewed-on: https://skia-review.googlesource.com/126620 Commit-Queue: Brian Salomon <bsalomon@google.com> Auto-Submit: Brian Salomon <bsalomon@google.com> Reviewed-by: Greg Daniel <egdaniel@google.com>
* Add null proxy test to SkGpuDevice::drawProducerLatticeGravatar Brian Salomon2018-05-07
| | | | | | | | | | | | Bug: b/77917978 TBR=brianosman@google.com Change-Id: Ib7ecb00dd0beac2444ca95106c2560c2b2940465 Reviewed-on: https://skia-review.googlesource.com/126509 Reviewed-by: Brian Salomon <bsalomon@google.com> Reviewed-by: Brian Osman <brianosman@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* SkAdvancedTypefaceMetrics: simplify moreGravatar Hal Canary2018-05-07
| | | | | | | Change-Id: I80c28c50aa96c28181d2fef13daad4a80ec91a12 Reviewed-on: https://skia-review.googlesource.com/120160 Reviewed-by: Ben Wagner <bungeman@google.com> Commit-Queue: Hal Canary <halcanary@google.com>
* make SkColorSpace::MakeICC strict as skcmsGravatar Mike Klein2018-05-07
| | | | | | | | | | | | | | | We do this somewhat trivially by calling skcms_Parse() and failing if it fails. This will help keep us sane while we work to remove SkColorSpace::MakeICC(). Change-Id: Icb1bc1a088083e181f46465dbb3a530f0e004193 Reviewed-on: https://skia-review.googlesource.com/126510 Commit-Queue: Mike Klein <mtklein@chromium.org> Commit-Queue: Brian Osman <brianosman@google.com> Auto-Submit: Mike Klein <mtklein@chromium.org> Reviewed-by: Brian Osman <brianosman@google.com>
* Handle failure to create SkColorSpaceXform in SkConvertPixelsGravatar Brian Osman2018-05-07
| | | | | | | | | | | | For now, this will just fallback to the pipeline code. Eventually, we may need to support more color spaces in skcms, (or put in a simpler fallback that just does type conversion and ignores color space)? Bug: chromium:838115 Change-Id: Id71eec6a72202683957c884620ee3fe90df6c776 Reviewed-on: https://skia-review.googlesource.com/126201 Commit-Queue: Mike Klein <mtklein@google.com> Reviewed-by: Mike Klein <mtklein@google.com>
* Make sure the ScalerContextRec is denseGravatar Herb Derby2018-05-07
| | | | | | | | | | | Make sure that the Rec structure is dense because the SkDescriptor code takes checksums of this structure. Change-Id: Id384ed49e17f4613647da538c810f09a066efd90 Reviewed-on: https://skia-review.googlesource.com/126503 Commit-Queue: Herb Derby <herb@google.com> Reviewed-by: Mike Klein <mtklein@google.com>
* Make GPU lattice/nine patch not bleed across cells.Gravatar Brian Salomon2018-05-07
| | | | | | | | | | | | | Consolidate code for handling various image/bitmap and lattice/ninepatch flavors. Makes refTextureForParams virtual on GrTextureProducer. Previously both subclasses had non-virtual flavors of this. Bug: b/77917978 Change-Id: I14787faef33c4617ef359039e81453d683f33ff1 Reviewed-on: https://skia-review.googlesource.com/125520 Reviewed-by: Brian Osman <brianosman@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Update canvas2D read/write pixels fast path to handle BGRA input/outputGravatar Brian Salomon2018-05-07
| | | | | | | Change-Id: Ib77b45dc0ea2b00c728b080d6d847f97b1471bff Reviewed-on: https://skia-review.googlesource.com/126482 Commit-Queue: Brian Salomon <bsalomon@google.com> Reviewed-by: Robert Phillips <robertphillips@google.com>
* Remove android specific code - v2Gravatar Herb Derby2018-05-07
| | | | | | | Change-Id: I2010b168f1a1f3bc48b86a075b0398eaf46ea328 Reviewed-on: https://skia-review.googlesource.com/126480 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Herb Derby <herb@google.com>
* Reland "Avoid drawing BW clipped color glyphs many times."Gravatar Ben Wagner2018-05-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 8a95244f626c2f7d4b66c8ca18266c2b0640bebb. Reason for revert: Didn't actually break anything. Original change's description: > Revert "Avoid drawing BW clipped color glyphs many times." > > This reverts commit bba38ee80e4140908ac0c351dc18a2f1224e5f0a. > > Reason for revert: possibly breaking chrome text > > Original change's description: > > Avoid drawing BW clipped color glyphs many times. > > > > Prior to this change if a color glyph is drawn through the raster device > > with an aliased clip the glyph is drawn once for each rectangular region > > in the aliased clip. In addition, even when the glyph was not a color > > glyph in this situation, the mask was checked for being a color glyph > > once for each rectangular region of the aliased clip. This change hoists > > the test for the color format out of the loop to ensure that the mask > > format is checked once and the mask is drawn once. > > > > This issue was discovered by rotating the coloremoji_blendmodes gm. > > > > Change-Id: I18b6b546356780e0b00948fff7b65783219f5c92 > > Reviewed-on: https://skia-review.googlesource.com/125868 > > Reviewed-by: Mike Reed <reed@google.com> > > Commit-Queue: Ben Wagner <bungeman@google.com> > > TBR=mtklein@google.com,bungeman@google.com,reed@google.com > > # Not skipping CQ checks because original CL landed > 1 day ago. > > Change-Id: Ife6d7255f9cb88b6f9d7a3c5800e88cc065e475b > Reviewed-on: https://skia-review.googlesource.com/126401 > Reviewed-by: Greg Daniel <egdaniel@google.com> > Commit-Queue: Greg Daniel <egdaniel@google.com> TBR=egdaniel@google.com,mtklein@google.com,bungeman@google.com,reed@google.com Change-Id: I7fc4c1f901bcee482f6b118b192caee14c112e43 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://skia-review.googlesource.com/126540 Reviewed-by: Ben Wagner <bungeman@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
* Respect SkBitmapDeviceFilteredSurfaceProps in SkThreadedBMPDeviceGravatar Yuqian Li2018-05-07
| | | | | | | | | | | | Bug: skia:7909 Change-Id: I13ccfbdc883f764f60701383b90537d3175a8126 Reviewed-on: https://skia-review.googlesource.com/126100 Commit-Queue: Ben Wagner <bungeman@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com> Reviewed-by: Ben Wagner <bungeman@google.com> Auto-Submit: Yuqian Li <liyuqian@google.com> Reviewed-on: https://skia-review.googlesource.com/126463 Reviewed-by: Greg Daniel <egdaniel@google.com>
* track bounds of top-level control opsGravatar Mike Klein2018-05-07
| | | | | | | | | | | | | | | | We don't need an explicit save-restore block to determine the bounds of top-level control operations... the implicit save-restore that all picutres have should logically work the same way. The commented test failed before this and passes now. Bug: skia:7735 Change-Id: Ibd31a3a9b0b48042ab3869a6bb57bc8d8bb78c09 Reviewed-on: https://skia-review.googlesource.com/126460 Commit-Queue: Mike Klein <mtklein@chromium.org> Commit-Queue: Brian Osman <brianosman@google.com> Auto-Submit: Mike Klein <mtklein@chromium.org> Reviewed-by: Brian Osman <brianosman@google.com>
* SkRecorder: templates are better than macrosGravatar Hal Canary2018-05-07
| | | | | | | Change-Id: I594da617a0fc8970a01c2165e41cdd803173ffc5 Reviewed-on: https://skia-review.googlesource.com/126420 Reviewed-by: Mike Klein <mtklein@google.com> Commit-Queue: Hal Canary <halcanary@google.com>
* Revert "Avoid drawing BW clipped color glyphs many times."Gravatar Greg Daniel2018-05-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit bba38ee80e4140908ac0c351dc18a2f1224e5f0a. Reason for revert: possibly breaking chrome text Original change's description: > Avoid drawing BW clipped color glyphs many times. > > Prior to this change if a color glyph is drawn through the raster device > with an aliased clip the glyph is drawn once for each rectangular region > in the aliased clip. In addition, even when the glyph was not a color > glyph in this situation, the mask was checked for being a color glyph > once for each rectangular region of the aliased clip. This change hoists > the test for the color format out of the loop to ensure that the mask > format is checked once and the mask is drawn once. > > This issue was discovered by rotating the coloremoji_blendmodes gm. > > Change-Id: I18b6b546356780e0b00948fff7b65783219f5c92 > Reviewed-on: https://skia-review.googlesource.com/125868 > Reviewed-by: Mike Reed <reed@google.com> > Commit-Queue: Ben Wagner <bungeman@google.com> TBR=mtklein@google.com,bungeman@google.com,reed@google.com # Not skipping CQ checks because original CL landed > 1 day ago. Change-Id: Ife6d7255f9cb88b6f9d7a3c5800e88cc065e475b Reviewed-on: https://skia-review.googlesource.com/126401 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Revert "Remove android specific code"Gravatar Greg Daniel2018-05-07
| | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 5570ea0c2549722f4069add1875743e6c44fc50b. Reason for revert: Possibly breaking chrome Original change's description: > Remove android specific code > > Change-Id: Ib9b1f936d1aee920ba7064a9eccffaca6696812f > Reviewed-on: https://skia-review.googlesource.com/125341 > Commit-Queue: Mike Reed <reed@google.com> > Reviewed-by: Mike Reed <reed@google.com> > Auto-Submit: Herb Derby <herb@google.com> TBR=herb@google.com,reed@google.com # Not skipping CQ checks because original CL landed > 1 day ago. Change-Id: I95578d720fc4d5924296245522f880dc15624be9 Reviewed-on: https://skia-review.googlesource.com/126360 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Revert "Respect SkBitmapDeviceFilteredSurfaceProps in SkThreadedBMPDevice"Gravatar Greg Daniel2018-05-06
| | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit b9a1e688fd9c3e64c50d9159481e72a812f978c1. Reason for revert: breaking msan Original change's description: > Respect SkBitmapDeviceFilteredSurfaceProps in SkThreadedBMPDevice > > Bug: skia:7909 > Change-Id: I3e480a57f16eec97706296137c8c1061555b9701 > Reviewed-on: https://skia-review.googlesource.com/126100 > Commit-Queue: Ben Wagner <bungeman@google.com> > Commit-Queue: Yuqian Li <liyuqian@google.com> > Reviewed-by: Ben Wagner <bungeman@google.com> > Auto-Submit: Yuqian Li <liyuqian@google.com> TBR=mtklein@google.com,bungeman@google.com,liyuqian@google.com # Not skipping CQ checks because original CL landed > 1 day ago. Bug: skia:7909 Change-Id: Iee605220721a8b9f74b652106d22b4516a87900a Reviewed-on: https://skia-review.googlesource.com/126320 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Revert "Restore old SkBitmapDeviceFilteredSurfaceProps"Gravatar Greg Daniel2018-05-06
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 3d5b83bf95333a0b31b08fb4a889fa7697c87c03. Reason for revert: Need to revert to revert previous change Original change's description: > Restore old SkBitmapDeviceFilteredSurfaceProps > > The old seemingly unnecessary assignment is needed, otherwise > skbug.com/7909 won't be fixed, and we will also introduce some > unexpected gold changes. > > Ben: did you remember why you originally have that additional > assignment? Probably related with SkTLazy? > > Bug: skia:7909 > TBR: bungeman@google.com > Change-Id: I20293b76bb448733d3ab4a90d40282b1420000d0 > Reviewed-on: https://skia-review.googlesource.com/126204 > Reviewed-by: Yuqian Li <liyuqian@google.com> > Commit-Queue: Yuqian Li <liyuqian@google.com> TBR=bungeman@google.com,liyuqian@google.com # Not skipping CQ checks because original CL landed > 1 day ago. Bug: skia:7909 Change-Id: I27841ba2040ad026e67402e3b11a2b0ba5a55cb9 Reviewed-on: https://skia-review.googlesource.com/126321 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Greg Daniel <egdaniel@google.com>
* Restore old SkBitmapDeviceFilteredSurfacePropsGravatar Yuqian Li2018-05-04
| | | | | | | | | | | | | | | | The old seemingly unnecessary assignment is needed, otherwise skbug.com/7909 won't be fixed, and we will also introduce some unexpected gold changes. Ben: did you remember why you originally have that additional assignment? Probably related with SkTLazy? Bug: skia:7909 TBR: bungeman@google.com Change-Id: I20293b76bb448733d3ab4a90d40282b1420000d0 Reviewed-on: https://skia-review.googlesource.com/126204 Reviewed-by: Yuqian Li <liyuqian@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com>
* Remove GR_TESSELLATOR_LEGACY_INVERSION_HANDLING (unused).Gravatar Stephen White2018-05-04
| | | | | | | | | | No user-visible change. Bug: skia: Change-Id: Ib665e79c96f8fbbc4891595a95359ab085039950 Reviewed-on: https://skia-review.googlesource.com/126200 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Stephen White <senorblanco@chromium.org>
* SkSinglyLinkedList: use unique_ptr to manage ownershipGravatar Hal Canary2018-05-04
| | | | | | | Change-Id: I7b0fe6bd601b9ae0e9042ff28aa5b4dec9a9639b Reviewed-on: https://skia-review.googlesource.com/126180 Reviewed-by: Ben Wagner <bungeman@google.com> Commit-Queue: Hal Canary <halcanary@google.com>
* Respect SkBitmapDeviceFilteredSurfaceProps in SkThreadedBMPDeviceGravatar Yuqian Li2018-05-04
| | | | | | | | | | Bug: skia:7909 Change-Id: I3e480a57f16eec97706296137c8c1061555b9701 Reviewed-on: https://skia-review.googlesource.com/126100 Commit-Queue: Ben Wagner <bungeman@google.com> Commit-Queue: Yuqian Li <liyuqian@google.com> Reviewed-by: Ben Wagner <bungeman@google.com> Auto-Submit: Yuqian Li <liyuqian@google.com>
* Revert "Fix a bug in SkGpuBlurUtils::GaussianBlur (take 2)"Gravatar Robert Phillips2018-05-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 72de74504f09f5de17af281b363fbc114ece3e73. Reason for revert: We are missing some tests - Chrome images look different Original change's description: > Fix a bug in SkGpuBlurUtils::GaussianBlur (take 2) > > This fixes the remaining imageblurclampmode bug and a preexisting un-reported error in the imageblurrepeatmode GM. > > This relies on https://chromium-review.googlesource.com/c/chromium/src/+/1044113 (Add a flag to disable a blur imagefilter bug fix in Skia) landing in Chrome first. > > Bug: skia:7765 > Change-Id: I568d18c17bb95c9e270e4cfd304c8dc134a48a70 > Reviewed-on: https://skia-review.googlesource.com/125961 > Reviewed-by: Greg Daniel <egdaniel@google.com> > Commit-Queue: Robert Phillips <robertphillips@google.com> TBR=egdaniel@google.com,robertphillips@google.com Change-Id: Ia56b3cf3009a95b7baaa9af62e97be390d0868fa No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: skia:7765 Reviewed-on: https://skia-review.googlesource.com/126160 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Remove unused SK_DISABLE_ARC_TO_LINE_TO_CHECK build flagGravatar Brian Salomon2018-05-04
| | | | | | | Change-Id: Ibcdb8c53a95a135a8686b3fe22df5a9d32e939c7 Reviewed-on: https://skia-review.googlesource.com/126060 Reviewed-by: Robert Phillips <robertphillips@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Remove could not invert matrix SkDebugfsGravatar Brian Salomon2018-05-04
| | | | | | | Change-Id: I547513b816bdbf58e7628ae3d50c2597b173537f Reviewed-on: https://skia-review.googlesource.com/126080 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Brian Salomon <bsalomon@google.com>
* Don't use getDeviceClipBounds() to bound pic ops.Gravatar Mike Klein2018-05-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The values returned by SkCanvas::getDeviceClipBounds() are in the right space, but have extra constraints on them that are not desirable for bounding the logical bounds of draw operations: - they are integral - they are non-negative We've been intersecting the bounds of each operation with these bounds, which means we're mixing these bogus constraints into the bounds of each recorded operation. This percolates up to the SkPicutre cull rect too. The most egregious way to see the problem is to record a draw op entirely in negative space... it'll come back with empty logical bounds rather than its correct (negative-space) bounds. I've added a test for this, and another test I also think should be passing but left making it so as a follow up. I've had to disable a couple tests asserting clips affect the bounds. :/ A possible follow-up might go back to using the clips to tighten the bounds of the ops, just so long as we take the original user bounds and map them with the CTM through to device space ourselves, rather than relying on the recording canvas' clip stack. I think this means we'd need to maintain our own stack of device-space float SkRect clip bounds while calculating these op bounds. Bug: skia:7735 Change-Id: I6bf15f6b2a9ba4329a4eeae7f9d57aa8729ec1bb Reviewed-on: https://skia-review.googlesource.com/126002 Commit-Queue: Mike Klein <mtklein@chromium.org> Commit-Queue: Brian Osman <brianosman@google.com> Auto-Submit: Mike Klein <mtklein@chromium.org> Reviewed-by: Brian Osman <brianosman@google.com>
* Remove android specific codeGravatar Herb Derby2018-05-04
| | | | | | | | Change-Id: Ib9b1f936d1aee920ba7064a9eccffaca6696812f Reviewed-on: https://skia-review.googlesource.com/125341 Commit-Queue: Mike Reed <reed@google.com> Reviewed-by: Mike Reed <reed@google.com> Auto-Submit: Herb Derby <herb@google.com>
* Avoid drawing BW clipped color glyphs many times.Gravatar Ben Wagner2018-05-04
| | | | | | | | | | | | | | | | | Prior to this change if a color glyph is drawn through the raster device with an aliased clip the glyph is drawn once for each rectangular region in the aliased clip. In addition, even when the glyph was not a color glyph in this situation, the mask was checked for being a color glyph once for each rectangular region of the aliased clip. This change hoists the test for the color format out of the loop to ensure that the mask format is checked once and the mask is drawn once. This issue was discovered by rotating the coloremoji_blendmodes gm. Change-Id: I18b6b546356780e0b00948fff7b65783219f5c92 Reviewed-on: https://skia-review.googlesource.com/125868 Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
* Make degenerate kernel in fill_in_2D_gaussian_kernel sum to 1Gravatar Robert Phillips2018-05-04
| | | | | | | | Bug: 836113 Change-Id: I1b2ed1d67648163571df6c497365d2a293b3b589 Reviewed-on: https://skia-review.googlesource.com/126020 Reviewed-by: Herb Derby <herb@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Fix a bug in SkGpuBlurUtils::GaussianBlur (take 2)Gravatar Robert Phillips2018-05-04
| | | | | | | | | | | | This fixes the remaining imageblurclampmode bug and a preexisting un-reported error in the imageblurrepeatmode GM. This relies on https://chromium-review.googlesource.com/c/chromium/src/+/1044113 (Add a flag to disable a blur imagefilter bug fix in Skia) landing in Chrome first. Bug: skia:7765 Change-Id: I568d18c17bb95c9e270e4cfd304c8dc134a48a70 Reviewed-on: https://skia-review.googlesource.com/125961 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Robert Phillips <robertphillips@google.com>
* Make the src rects of truths in SkLatticeIter be integral.Gravatar Brian Salomon2018-05-04
| | | | | | | | | | | | | | | This will make the non-bleed gpu implementation more sane as we will know that the input src rects are at integers and if offset by +/-0.5 we will be at pixel centers. No change in behavior, since src rects really were integral all along. Bug: b/77917978 Change-Id: I7193ff0b38de5d10debc24a0121e643573c47cda Reviewed-on: https://skia-review.googlesource.com/125740 Commit-Queue: Leon Scroggins <scroggo@google.com> Reviewed-by: Leon Scroggins <scroggo@google.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-08 03:41:28 +0000