aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/core
Commit message (Collapse)AuthorAge
...
* use log2(scale) to compute mip levelGravatar reed2015-01-16
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/849333002
* Add Xfer Processor for GrCustomXfermodesGravatar egdaniel2015-01-16
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/852203003
* Fold alpha to the draw in savelayer-draw-restore patterns with non-opaque drawGravatar kkinnunen2015-01-16
| | | | | | | | | | | | | | | | Fold alpha of a save layer call to the subsequent draw call paint, even when the draw call paint color is already non-opaque. Comparing the difference of the unoptimized and the optimized call pattern with all (layer alpha, draw alpha) combinations, this produces off-by-one pixels ~50% of the time. Reduces layers of current desk_carsvg.skp (recorded with cross-process picture image filters allowed) from 122 to 115. BUG=skia:3119 Review URL: https://codereview.chromium.org/840483005
* Revert "choose better mip level"Gravatar reed2015-01-15
| | | | | | | | | | This reverts commit 09544fcf9611be6efc0c38ab9d9a3f82181341fb. BUG=skia: TBR= NOTREECHECKS=True Review URL: https://codereview.chromium.org/849173003
* Move XferEffects class to GrCustomXfermode fileGravatar egdaniel2015-01-15
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/844913003
* SkPictureShader should handle negative scaling gracefully.Gravatar fmalita2015-01-15
| | | | | | | | | Let's not choke on inverted tiles. BUG=chromium:447707 R=robertphillips@google.com,reed@google.com Review URL: https://codereview.chromium.org/852213002
* Why are we afraid of saveLayer() bounds?Gravatar mtklein2015-01-15
| | | | | | | | | They're only hints. We can ignore them in this optimization. My only hesitation is that this pattern doesn't seem to happen in our .skps. BUG=skia: Review URL: https://codereview.chromium.org/849243002
* choose better mip levelGravatar reed2015-01-15
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/845793005
* Use device-space stroke width for SkDraw::drawRect() quick-rejectGravatar fmalita2015-01-15
| | | | | | | | | | The stroke width needs to be CTM-adjusted when applied to device space rects. BUG=skia:3313 R=reed@google.com Review URL: https://codereview.chromium.org/801353008
* separate out the scaline-choosing to simplify chooseProcsGravatar reed2015-01-14
| | | | | | | BUG=skia: TBR= Review URL: https://codereview.chromium.org/796063007
* Do some minor pre cleanup work before converting all xfermodes to XPs.Gravatar egdaniel2015-01-14
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/853543003
* rename blitrow::proc and add (uncalled) hook for colorproc16Gravatar reed2015-01-13
| | | | | | BUG=skia:3302 Review URL: https://codereview.chromium.org/847443003
* Simplify SkInstCntGravatar mtklein2015-01-13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This code requires fewer macros to use it (just one), has less code in macro definitions, and has simpler synchronization code (just atomic ints, no SkOnce, no SkMutex, etc.) A minor downside, we lose indentation and reverse-ordering in the final report: Leaked SkRefCntBase: 7 Leaked SkFontMgr: 1 Leaked SkWeakRefCnt: 1 Leaked SkTypeface: 1 Leaked SkFlattenable: 3 Leaked SkXfermode: 3 Leaked SkPathRef: 1 Leaked SkPixelRef: 1 Leaked SkMallocPixelRef: 1 becomes Leaked SkXfermode: 3 Leaked SkMallocPixelRef: 1 Leaked SkPixelRef: 1 Leaked SkPathRef: 1 Leaked SkFlattenable: 3 Leaked SkTypeface: 1 Leaked SkWeakRefCnt: 1 Leaked SkFontMgr: 1 Leaked SkRefCntBase: 7 This is motivated by wanting to land https://codereview.chromium.org/806473006/, which makes sure all static use of SkOnce are in global scope. The current implementation of SkInstCnt uses them in function scope, which isn't safe. BUG=skia: No public API changes. TBR=reed@google.com Review URL: https://codereview.chromium.org/841263004
* Revert of Fix GPU clipped-AA vs. non-AA drawRect discrepancy (patchset #2 ↵Gravatar robertphillips2015-01-12
| | | | | | | | | | | | | | | | | | | | | | | | | id:20001 of https://codereview.chromium.org/839883003/) Reason for revert: This CL introduces rendering conflicts with hairlines (i.e., the hairlines get overwritten). These conflicts are particularly visible on the following GMs (for the Ubuntu and Android gpu configs): coloremoji & complexclip2_rrect_bw Original issue's description: > Fix GPU clipped-AA vs. non-AA drawRect discrepancy > > In the clip stack we were manually rounding out non-AA clip rects but leaving the hardening of non-AA drawRects up to the GPU. In some border cases the GPU can truncate rather than round out resulting in visual discrepancies. > > BUG=423834 > > Committed: https://skia.googlesource.com/skia/+/933a03fecb65c83f81cf65d5cf9870c69aa379ff TBR=bsalomon@google.com,jvanverth@google.com NOTREECHECKS=true NOTRY=true BUG=423834 Review URL: https://codereview.chromium.org/847033002
* Fix GPU clipped-AA vs. non-AA drawRect discrepancyGravatar robertphillips2015-01-12
| | | | | | | | In the clip stack we were manually rounding out non-AA clip rects but leaving the hardening of non-AA drawRects up to the GPU. In some border cases the GPU can truncate rather than round out resulting in visual discrepancies. BUG=423834 Review URL: https://codereview.chromium.org/839883003
* Fix overlap test for 64-bit pointers in SkMatrixGravatar qiankun.miao2015-01-12
| | | | | | BUG=skia:3297 Review URL: https://codereview.chromium.org/803493008
* Fix up all the easy virtual ... SK_OVERRIDE cases.Gravatar mtklein2015-01-09
| | | | | | | | | | | | This fixes every case where virtual and SK_OVERRIDE were on the same line, which should be the bulk of cases. We'll have to manually clean up the rest over time unless I level up in regexes. for f in (find . -type f); perl -p -i -e 's/virtual (.*)SK_OVERRIDE/\1SK_OVERRIDE/g' $f; end BUG=skia: Review URL: https://codereview.chromium.org/806653007
* Remove SkTileGrid (except for TileGridInfo).Gravatar mtklein2015-01-09
| | | | | | | | TBR=reed@google.com BUG=skia:3085 Review URL: https://codereview.chromium.org/845623002
* Revert of Cache blur mask for rects which can not break into nine-patch ↵Gravatar qiankun.miao2015-01-07
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (patchset #10 id:200001 of https://codereview.chromium.org/729463002/) Reason for revert: revert it due to a memory leak. ==8017==ERROR: LeakSanitizer: detected memory leaks Direct leak of 25992 byte(s) in 2 object(s) allocated from: #0 0x7feb53030e0b in __interceptor_malloc (/home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/gm+0x24fe0b) #1 0x7feb54d54f76 in sk_malloc_flags(unsigned long, unsigned int) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/ports/SkMemory_malloc.cpp:54:15 #2 0x7feb54d54d4a in sk_malloc_throw(unsigned long) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/ports/SkMemory_malloc.cpp:40:12 #3 0x7feb539f5a77 in SkMask::AllocImage(unsigned long) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkMask.cpp:37:22 #4 0x7feb53fe5c34 in (anonymous namespace)::copy_cacheddata_to_mask(SkCachedData*, SkMask*) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkMaskCache.cpp:25:20 #5 0x7feb53fea064 in SkMaskCache::FindAndCopy(float, SkBlurStyle, SkBlurQuality, SkRect const*, int, SkMask*) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkMaskCache.cpp:224:26 #6 0x7feb539f957e in SkMaskFilter::filterPath(SkPath const&, SkMatrix const&, SkRasterClip const&, SkBlitter*, SkPaint::Style) const /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkMaskFilter.cpp:270:14 #7 0x7feb5392e920 in SkDraw::drawPath(SkPath const&, SkPaint const&, SkMatrix const*, bool, bool, SkBlitter*) const /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkDraw.cpp:1117:13 #8 0x7feb53694afc in SkDraw::drawPath(SkPath const&, SkPaint const&, SkMatrix const*, bool) const /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../include/core/SkDraw.h:54:9 #9 0x7feb5368d799 in SkBitmapDevice::drawPath(SkDraw const&, SkPath const&, SkPaint const&, SkMatrix const*, bool) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkBitmapDevice.cpp:216:5 #10 0x7feb5386aa57 in SkCanvas::onDrawPath(SkPath const&, SkPaint const&) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkCanvas.cpp:1908:9 #11 0x7feb5386386b in SkCanvas::drawPath(SkPath const&, SkPaint const&) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../src/core/SkCanvas.cpp:1703:5 #12 0x7feb53109572 in Blur2RectsNonNinePatchGM::onDraw(SkCanvas*) /home/default/storage/skia-repo/buildbot/skiabot-linux-xsan-000/build/slave/Test-Ubuntu13_10-GCE-NoGPU-x86_64-Debug-ASAN/build/skia/out/Debug/../../gm/blurs.cpp:166:9 Original issue's description: > Cache blur mask for rects which can not break into nine-patch > > With this CL performance improves: > blurrectsnonninepatch 42.4us -> 20.5us 0.48x > > BUG=431021,skia:3118 > > Committed: https://skia.googlesource.com/skia/+/5a5b4e900ee69540fc35952882a323c63d6d0db9 TBR=reed@google.com,humper@google.com,reed@chromium.org NOTREECHECKS=true NOTRY=true BUG=431021,skia:3118 Review URL: https://codereview.chromium.org/844673002
* Cache blur mask for rects which can not break into nine-patchGravatar qiankun.miao2015-01-07
| | | | | | | | | With this CL performance improves: blurrectsnonninepatch 42.4us -> 20.5us 0.48x BUG=431021,skia:3118 Review URL: https://codereview.chromium.org/729463002
* It is dangerous to ignore SkRect::intersect's return valueGravatar robertphillips2015-01-07
| | | | | | Committed: https://skia.googlesource.com/skia/+/152f524fd325b7776b01f84afbfe2fa071648a05 Review URL: https://codereview.chromium.org/833943002
* Remove distance field generation and storage from SkGlyphCache.Gravatar jvanverth2015-01-07
| | | | | | BUG=skia:3103 Review URL: https://codereview.chromium.org/834103004
* Revert of It is dangerous to ignore SkRect::intersect's return value ↵Gravatar robertphillips2015-01-07
| | | | | | | | | | | | | | | | | | (patchset #6 id:100001 of https://codereview.chromium.org/833943002/) Reason for revert: Still more Chromium clean up to do Original issue's description: > It is dangerous to ignore SkRect::intersect's return value > > Committed: https://skia.googlesource.com/skia/+/152f524fd325b7776b01f84afbfe2fa071648a05 TBR=reed@google.com NOTREECHECKS=true NOTRY=true Review URL: https://codereview.chromium.org/825983005
* Remove macros that make it look like it's a good idea to not be able to flatten.Gravatar mtklein2015-01-07
| | | | | | | | | | | | | | | | | There are only a handful of SkFlattenables that are not flattenable. That there are any seems highly illogical. To make this look less like a normal thing, this removes both macros that marked SkFlattenables as non-flattenable (in slightly different ways). The handful of SkFlattenables in our codebase that can't be flattened now assert violently that they can't be flattened. They're internal or part of animator... places where we'll never actually flatten them. TestLooper and DummyRasterizer were so trivial that I just made them flattenable. BUG=skia: Review URL: https://codereview.chromium.org/841753002
* It is dangerous to ignore SkRect::intersect's return valueGravatar robertphillips2015-01-07
| | | | Review URL: https://codereview.chromium.org/833943002
* Verify size_t overflowGravatar sugoi2015-01-07
| | | | | | | | In 32 bits, it's possible that multiplying 2 32b values might overflow a size_t, which could be 32b unsigned in that context, so I added a check for size_t overflow. BUG=445831 Review URL: https://codereview.chromium.org/836733005
* Rename SkDrawPictureCallback to SkPicture::AbortCallbackGravatar robertphillips2015-01-07
| | | | | | Committed: https://skia.googlesource.com/skia/+/7ef197255deb4e2fa64c03c7130d56ddf164e83c Review URL: https://codereview.chromium.org/829983003
* use conics for ovals and roundrectsGravatar reed2015-01-06
| | | | | | | | | This reverts commit e3b61eb7997d12c5f9b844dd3af1847492d9f411. BUG=skia: TBR= Review URL: https://codereview.chromium.org/834303002
* Revert of Rename SkDrawPictureCallback to SkPicture::AbortCallback (patchset ↵Gravatar reed2015-01-06
| | | | | | | | | | | | | | | | | | #2 id:20001 of https://codereview.chromium.org/829983003/) Reason for revert: speculative revert to unblock DEPS roll Original issue's description: > Rename SkDrawPictureCallback to SkPicture::AbortCallback > > Committed: https://skia.googlesource.com/skia/+/7ef197255deb4e2fa64c03c7130d56ddf164e83c TBR=reed@google.com,robertphillips@google.com NOTREECHECKS=true NOTRY=true Review URL: https://codereview.chromium.org/824013004
* Comment out unused names in src/core/SkGlyphCache.hGravatar tomhudson2015-01-06
| | | | | | | | | | | | Android violates our encapsulation and depends on this file; to keep their compiles clean they want us to hide the names of unused parameters. Follow-up to https://codereview.chromium.org/726923002/. R=djsollen@google.com Review URL: https://codereview.chromium.org/839583005
* Rename SkDrawPictureCallback to SkPicture::AbortCallbackGravatar robertphillips2015-01-06
| | | | Review URL: https://codereview.chromium.org/829983003
* fixes for conicsGravatar reed2015-01-06
| | | | | | | | | | - use std tolerance in edgebuilder, since the path has not been scaled-up to its super-sample size (that happens in the builder methods. - off-by-1 fix for pathops when using the output of the conicquadder BUG=skia: Review URL: https://codereview.chromium.org/837023002
* Revert of use conics for ovals and roundrects (patchset #1 id:1 of ↵Gravatar reed2015-01-06
| | | | | | | | | | | | | | | | | | | | | | | | https://codereview.chromium.org/817283005/) Reason for revert: need fix in edgebuilder+aa Original issue's description: > use conics for ovals and roundrects > > adapted from https://codereview.chromium.org/831313002/ > > BUG=skia: > TBR= > > Committed: https://skia.googlesource.com/skia/+/2bf1c4d58b84c73c6421edcb25e3afb4971da200 TBR=reed@chromium.org NOTREECHECKS=true NOTRY=true BUG=skia: Review URL: https://codereview.chromium.org/840573002
* use conics for ovals and roundrectsGravatar reed2015-01-05
| | | | | | | | | adapted from https://codereview.chromium.org/831313002/ BUG=skia: TBR= Review URL: https://codereview.chromium.org/817283005
* Remove SkPath::asRectGravatar robertphillips2015-01-05
| | | | | | | | | | AFAICT the asRect entry point is not needed. Greg: GPU Reed: API Cary: Path Review URL: https://codereview.chromium.org/833193002
* Fix thin rect stroking.Gravatar fmalita2015-01-05
| | | | | | | | | | | When both edges of a thin (sub-unit) stroke fall within the same pixel, we compute the wrong coverage and also blit the same line twice. To avoid these issues, tweak the inner/outer hulls to ensure one of the edges is always pixel-aligned. BUG=skia:3278 R=reed@google.com Review URL: https://codereview.chromium.org/819253003
* Cleanup isRect variantsGravatar robertphillips2015-01-05
| | | | Review URL: https://codereview.chromium.org/832083002
* Print out failing PathRef test on ChromeOSGravatar robertphillips2015-01-05
| | | | | | | | This is intended to provide additional information for a flaky assertion failure on ChromeOS Daisy. BUG=skia:3282 Review URL: https://codereview.chromium.org/821683002
* Revert of Revert of move remaining virtual draw methods to onDraw (patchset ↵Gravatar reed2015-01-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | #1 id:1 of https://codereview.chromium.org/835913002/) Reason for revert: test to see if it wasn't this CL that broke chromeos Original issue's description: > Revert of move remaining virtual draw methods to onDraw (patchset #5 id:80001 of https://codereview.chromium.org/817723005/) > > Reason for revert: > did this cause chromeos heap corruption on skp? > > Original issue's description: > > move remaining virtual draw methods to onDraw > > > > BUG=skia: > > > > Committed: https://skia.googlesource.com/skia/+/2e0f7d53bba5438c9bf11ee5ccae2c301e348419 > > TBR=robertphillips@google.com,djsollen@google.com,bsalomon@google.com,mtklein@google.com > NOTREECHECKS=true > NOTRY=true > BUG=skia: > > Committed: https://skia.googlesource.com/skia/+/e87e383351e1547b21d96d02bcd0c3cbbe0837f1 TBR=robertphillips@google.com,djsollen@google.com,bsalomon@google.com,mtklein@google.com,reed@chromium.org NOTREECHECKS=true NOTRY=true BUG=skia: Review URL: https://codereview.chromium.org/831253002
* Clean up dead clear() code in SkRecord.Gravatar mtklein2015-01-05
| | | | | | | | | | | | | This removes the SkRecords::Clear struct and everything that refers to it. Notice there is nothing actually creating a Clear, which means this is all dead code. Now that all ops obey the clip, I don't think we need the weird inflate-empty-to-epsilon hack for BBH queries. BUG=skia: Review URL: https://codereview.chromium.org/835813002
* Revert of move remaining virtual draw methods to onDraw (patchset #5 ↵Gravatar reed2015-01-05
| | | | | | | | | | | | | | | | | | | | | id:80001 of https://codereview.chromium.org/817723005/) Reason for revert: did this cause chromeos heap corruption on skp? Original issue's description: > move remaining virtual draw methods to onDraw > > BUG=skia: > > Committed: https://skia.googlesource.com/skia/+/2e0f7d53bba5438c9bf11ee5ccae2c301e348419 TBR=robertphillips@google.com,djsollen@google.com,bsalomon@google.com,mtklein@google.com NOTREECHECKS=true NOTRY=true BUG=skia: Review URL: https://codereview.chromium.org/835913002
* move remaining virtual draw methods to onDrawGravatar reed2015-01-04
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/817723005
* remove unused drawDataGravatar reed2015-01-02
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/830083002
* Fix computation of bound in FillBounds::updateSaveBoundsGravatar robertphillips2014-12-30
| | | | | | | | intersect doesn't change the bounds when the two bounds do not intersect. This is definitely not the intended behavior. With the SKPs captured on 12/23/14, Chrome began passing Skia drawPicture ops that did not intersect the current clip - which revealed this bug. Review URL: https://codereview.chromium.org/817483004
* Update stroke path to use rect returned from isRect (to fix trailing moveTo bug)Gravatar robertphillips2014-12-29
| | | | | | | | | | This basically recreates what was done in: https://codereview.chromium.org/16950021/ (add rect-output parameter to isRect, allowing us to return the correct bounds even if a rectagular path has a trailing moveTo) with the addition of GM representation BUG=skia:247770 Review URL: https://codereview.chromium.org/834503002
* Fix bound returned by SkPath::isRect when the path contains a trailing moveToGravatar robertphillips2014-12-29
| | | | | | | | | | | | | | | | | | Oddly enough this was fixed in: https://codereview.chromium.org/16950021/ (add rect-output parameter to isRect, allowing us to return the correct bounds even if a rectagular path has a trailing moveTo) but was reverted here: https://skia.googlesource.com/skia/+/8fd160350ca5f57fbb1b2e03383c5778414a9b48 since it appeared to be crashing Chrome's trybots. I think it just fell through the cracks after that. If this sticks I will land a follow on patch for the stroke issue reported in the original bug (crbug.com/247770). BUG=247770,445368 Review URL: https://codereview.chromium.org/834483002
* Add device space skshader GM to test kDevice_GrCoordSetGravatar bsalomon2014-12-29
| | | | Review URL: https://codereview.chromium.org/816003002
* fix images in multi-thread by doing shallow-copiesGravatar reed2014-12-25
| | | | | | | BUG=skia: TBR=mtklein@google.com Review URL: https://codereview.chromium.org/791473006
* add private flag for sprite-as-bitmapGravatar reed2014-12-22
| | | | | | BUG=skia:3272 Review URL: https://codereview.chromium.org/806853005
* add testing flag to ignore saveLayer boundsGravatar reed2014-12-22
| | | | | | BUG=skia: Review URL: https://codereview.chromium.org/819123003
generated by cgit v1.2.3 (git 2.39.1) at 2024-10-24 09:59:20 +0000