searching for reviewer(mstange)
8e3201300324: Bug 1418722 - Mark nsDisplayPerspective items for removal when the inner transform frame gets removed from the frame tree. r=mstange a=gchang
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 28 Nov 2017 09:39:08 +1300 - rev 442618
Push 8274 by nerli@mozilla.com at 2017-11-30 14:47 +0000
Bug 1418722 - Mark nsDisplayPerspective items for removal when the inner transform frame gets removed from the frame tree. r=mstange a=gchang
3f82ccf7f33a: Bug 1360278 - Add preference to trigger context menu on mouse up for GTK+ and macOS, r=mstange,smaug a=gchang
Robin Grenet <robin.grenet@wanadoo.fr> - Thu, 16 Nov 2017 13:35:58 +0100 - rev 442443
Push 8211 by archaeopteryx@coole-files.de at 2017-11-23 12:27 +0000
Bug 1360278 - Add preference to trigger context menu on mouse up for GTK+ and macOS, r=mstange,smaug a=gchang MozReview-Commit-ID: Bg60bD8jIg6
93b43c6d7ab4: Bug 1415963 - Add a reftest. r=mstange a=gchang
Botond Ballo <botond@mozilla.com> - Mon, 13 Nov 2017 19:35:58 -0500 - rev 442394
Push 8189 by apavel@mozilla.com at 2017-11-22 13:36 +0000
Bug 1415963 - Add a reftest. r=mstange a=gchang MozReview-Commit-ID: kbVHp3lqCU
520fd735565e: Bug 1415963 - Take into account stroke (not just fill) when calculating the contribution of an SVG element to a clip rect for a clip path. r=mstange a=gchang
Botond Ballo <botond@mozilla.com> - Mon, 13 Nov 2017 19:35:51 -0500 - rev 442393
Push 8189 by apavel@mozilla.com at 2017-11-22 13:36 +0000
Bug 1415963 - Take into account stroke (not just fill) when calculating the contribution of an SVG element to a clip rect for a clip path. r=mstange a=gchang MozReview-Commit-ID: 7ZwwOH0ErRN
1c6cf35925b7: Add a reftest (bug 1409871, r=mstange)
Ryan Hunt <rhunt@eqrion.net> - Sun, 05 Nov 2017 10:05:14 -0600 - rev 442197
Push 8138 by ryanvm@gmail.com at 2017-11-11 14:08 +0000
Add a reftest (bug 1409871, r=mstange) MozReview-Commit-ID: 9UxLeFGZW8E
20d4af5881a1: Bug 1416306 - Make the function argument a const& instead of passing by value. r=mstange
Milan Sreckovic <milan@mozilla.com> - Fri, 10 Nov 2017 10:29:00 -0500 - rev 442195
Push 8138 by ryanvm@gmail.com at 2017-11-11 14:08 +0000
Bug 1416306 - Make the function argument a const& instead of passing by value. r=mstange
e7f0ad152631: Bug 1416067 - Avoid doing a bunch of useless event-regions work for frames with empty border boxes. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Thu, 09 Nov 2017 19:40:44 -0500 - rev 442040
Push 8134 by ryanvm@gmail.com at 2017-11-10 21:18 +0000
Bug 1416067 - Avoid doing a bunch of useless event-regions work for frames with empty border boxes. r=mstange MozReview-Commit-ID: 8gsIJojhQ6t
9d9e227c3a5e: Bug 1413073 - Part 3: Recompute ASR and clip-chains on wrap lists when merging. r=mstange
Matt Woodrow <mwoodrow@mozilla.com> - Fri, 10 Nov 2017 13:16:55 +1300 - rev 441951
Push 8134 by ryanvm@gmail.com at 2017-11-10 21:18 +0000
Bug 1413073 - Part 3: Recompute ASR and clip-chains on wrap lists when merging. r=mstange
d5a89d7d8163: Bug 1413073 - Part 2: Only force frames to have display items built if they intersect the dirty rectangle. r=mstange
Matt Woodrow <mwoodrow@mozilla.com> - Fri, 10 Nov 2017 13:16:45 +1300 - rev 441950
Push 8134 by ryanvm@gmail.com at 2017-11-10 21:18 +0000
Bug 1413073 - Part 2: Only force frames to have display items built if they intersect the dirty rectangle. r=mstange
08e3bd4e82de: Bug 1413073 - Part 1: Add new crashtests for complex ASR cases. r=mstange
Matt Woodrow <mwoodrow@mozilla.com> - Fri, 10 Nov 2017 13:16:36 +1300 - rev 441949
Push 8134 by ryanvm@gmail.com at 2017-11-10 21:18 +0000
Bug 1413073 - Part 1: Add new crashtests for complex ASR cases. r=mstange
6ca404d46fc2: Bug 1411627 - Add reftests. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 07 Nov 2017 10:16:54 -0500 - rev 441415
Push 8130 by ryanvm@gmail.com at 2017-11-09 00:28 +0000
Bug 1411627 - Add reftests. r=mstange MozReview-Commit-ID: apbFCe2uzF
857a0e87e490: Bug 1411627 - Send the applied offset for sticky frames to WR. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 07 Nov 2017 10:16:48 -0500 - rev 441414
Push 8130 by ryanvm@gmail.com at 2017-11-09 00:28 +0000
Bug 1411627 - Send the applied offset for sticky frames to WR. r=mstange There are cases where we do a main-thread paint at a scroll position where sticky offsets have been applied in order to keep sticky items visually unmoving. From that paint, it's possible to do an async-scroll in the direction that reduces the sticky offset. In order for WR to handle this case properly we need to tell WR how much of a sticky offset was already applied on the main thread. MozReview-Commit-ID: 79DsfPpsIfA
a165d1dcc904: Bug 1411627 - Stop sending sticky clips to WR for scrollframes with no ASR. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 07 Nov 2017 10:16:05 -0500 - rev 441413
Push 8130 by ryanvm@gmail.com at 2017-11-09 00:28 +0000
Bug 1411627 - Stop sending sticky clips to WR for scrollframes with no ASR. r=mstange If the scrollframe has no ASR and is not async scrollable, WR will never need to adjust the position of the sticky item. So we don't actually need to send this data to WR. In theory this would be harmless but the later patches in this bug actually expose another coordinate space mismatch between the gecko and WR sticky-position code, so this patch is needed to avoid the latent problem from manifesting to the user. MozReview-Commit-ID: J0LMcU1FudA
6f7cae36d92d: Bug 1413693 - Ensure data copied from discarded front buffer isn't also painted. r=mstange
Jamie Nicol <jnicol@mozilla.com> - Thu, 02 Nov 2017 11:46:08 +0000 - rev 440702
Push 8120 by ryanvm@gmail.com at 2017-11-04 17:45 +0000
Bug 1413693 - Ensure data copied from discarded front buffer isn't also painted. r=mstange Bug 1092294 introduced a regression in to the code to copy from the discarded front buffers to the new backbuffers in SingleTiledContentClient. The aim was to ensure that if locking the the frontbuffers failed, meaning the region could not be copied, that it would be painted instead. However due to incorrect logic the region would both be copied and painted in cases where there was no onWhite buffers. To fix this we take both locks (frontLock, and frontOnWhiteLock if required) up front, so that we either copy both buffers or neither. MozReview-Commit-ID: 3iepOuweruk
7ae1ff4001aa: Bug 1386511 - Part 1: Revise the format of the GC profiler markers. r=jonco, r=mstange
Paul Bone <pbone@mozilla.com> - Tue, 31 Oct 2017 14:32:49 +1100 - rev 440684
Push 8120 by ryanvm@gmail.com at 2017-11-04 17:45 +0000
Bug 1386511 - Part 1: Revise the format of the GC profiler markers. r=jonco, r=mstange This change also updates the documentation to correct the semantics of GC slice numbers.
b17661987940: Bug 1412272 - Have ComputePartialPrerenderArea() handle the cases where aDirtyRect is larger than aPrerenderSize. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 01 Nov 2017 15:27:16 -0400 - rev 440530
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1412272 - Have ComputePartialPrerenderArea() handle the cases where aDirtyRect is larger than aPrerenderSize. r=mstange Rendering at least aDirtyRect is more important than staying under aPrerenderSize, so that's what we do. MozReview-Commit-ID: 8Ze1biaNzqX
ce7f71c2f402: Bug 1382534 - Add an async-scrolling reftest. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 04 Oct 2017 16:39:47 -0400 - rev 440525
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Add an async-scrolling reftest. r=mstange MozReview-Commit-ID: 2MuOY0tf7Jr
29a923b7d812: Bug 1382534 - Add a crashtest for the ASR-related assertion failure. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 04 Oct 2017 16:27:15 -0400 - rev 440524
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Add a crashtest for the ASR-related assertion failure. r=mstange MozReview-Commit-ID: J8hm3FmQpaE
296d10ae6f34: Bug 1382534 - Try to give nsDisplayMask items proper scroll metadata. r=mstange
Botond Ballo <botond@mozilla.com> - Fri, 20 Oct 2017 18:16:50 -0400 - rev 440523
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Try to give nsDisplayMask items proper scroll metadata. r=mstange MozReview-Commit-ID: L6kYkzC1F8S
5a6831152e15: Bug 1382534 - Use GetBounds() in nsDisplayMask::ComputeVisibility(). r=mstange
Botond Ballo <botond@mozilla.com> - Mon, 16 Oct 2017 17:18:03 -0400 - rev 440522
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Use GetBounds() in nsDisplayMask::ComputeVisibility(). r=mstange The code previously used GetClippedBoundsWithRespectToASR() (changed in bug 1298218), but this violated the requirements of nsDisplayList:: ComputeVisibilityForSublist(). MozReview-Commit-ID: F9UVMvVKLAp
486bc3b21ec6: Bug 1382534 - Relax the requirement of having a clip with respect to an ASR for nsDisplayMask items. r=mstange
Botond Ballo <botond@mozilla.com> - Mon, 25 Sep 2017 19:15:23 -0400 - rev 440521
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Relax the requirement of having a clip with respect to an ASR for nsDisplayMask items. r=mstange We allow an nsDisplayMask item to satisfy the requirement even it does not itself have a clip, as long as its children have finite bounds with respect to the ASR. MozReview-Commit-ID: 8zKE0bguY38
a5cc27bbb7b4: Bug 1382534 - Add nsCSSClipPathInstance::GetBoundingRectForBasicShapeClip(). r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 04 Oct 2017 16:46:44 -0400 - rev 440520
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Add nsCSSClipPathInstance::GetBoundingRectForBasicShapeClip(). r=mstange MozReview-Commit-ID: 85mhNcj5V2J
441b39c10487: Bug 1382534 - Make layout/reftests/svg/clipPath-polygon-elementFromPoint-01.svg a bit more permissive. r=mstange
Botond Ballo <botond@mozilla.com> - Mon, 25 Sep 2017 16:19:03 -0400 - rev 440517
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Make layout/reftests/svg/clipPath-polygon-elementFromPoint-01.svg a bit more permissive. r=mstange The test was asserting that a point on the boundary of an SVG path is inside the path, but for a point that's right on the boundary, either answer (inside or outside) is reasonable. MozReview-Commit-ID: 8srcoPFmDq
8148797eab63: Bug 1382534 - Add a UnionMaybeRects() method to gfx/2d/Rect.h. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 23 Aug 2017 17:45:15 -0400 - rev 440515
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Add a UnionMaybeRects() method to gfx/2d/Rect.h. r=mstange MozReview-Commit-ID: 3G4Hlj54Thy
4923de331172: Bug 1382534 - Move ComputeOffsetToUserSpace() from nsDisplayList.cpp to nsLayoutUtils. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 23 Aug 2017 17:26:25 -0400 - rev 440514
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Move ComputeOffsetToUserSpace() from nsDisplayList.cpp to nsLayoutUtils. r=mstange MozReview-Commit-ID: 8w9lDViJuAj
98355059c8d1: Bug 1382534 - Layerize active masks even with HWA disabled. r=mstange
Botond Ballo <botond@mozilla.com> - Wed, 09 Aug 2017 16:25:19 -0400 - rev 440513
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1382534 - Layerize active masks even with HWA disabled. r=mstange MozReview-Commit-ID: 6lvOE3fGFAy
e12eb0aa5165: Bug 1406217 - Fix computation of scrollframe's offset to the reference frame. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 31 Oct 2017 09:04:04 -0400 - rev 440468
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1406217 - Fix computation of scrollframe's offset to the reference frame. r=mstange We need to get the offset across documents boundaries, since the scrollframe might be inside a subdocument. In such cases we were previously computing a (incorrect) zero offset for the scrollframe, which was throwing off the margin for the sticky items inside the scrollframe. MozReview-Commit-ID: AWeQ9ay2cmp
5e19d7125bbb: Bug 1413629. Don't require a PresContext to apply a clip. r=mstange
Jeff Muizelaar <jmuizelaar@mozilla.com> - Wed, 01 Nov 2017 14:32:21 -0400 - rev 440431
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1413629. Don't require a PresContext to apply a clip. r=mstange The only thing we use the PresContext for is the app units. I'd rather not worry about keeping around a PresContext for the webrender display item grouping.
f9632977cebf: Bug 1413073 - Don't check the dirty rect when deciding if we need to store OutOfFlowDisplayData for a frame. r=mstange
Matt Woodrow <mwoodrow@mozilla.com> - Wed, 01 Nov 2017 16:58:53 +1300 - rev 440286
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1413073 - Don't check the dirty rect when deciding if we need to store OutOfFlowDisplayData for a frame. r=mstange
1713aff08afb: Add profiler markers for painting on the paint thread. (bug 1412951, r=mstange)
Ryan Hunt <rhunt@eqrion.net> - Mon, 30 Oct 2017 14:48:16 -0400 - rev 440109
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Add profiler markers for painting on the paint thread. (bug 1412951, r=mstange) The following commits added the necessary tracking to ensure we always end a profiler marker when we start one, and this just plugs them in. MozReview-Commit-ID: C5VXpnlH2QA
41566b43f974: Bug 1412648 - Initialize some uninitialized fields in tools/profiler/. r=mstange
Tristan Bourvon <tbourvon@mozilla.com> - Tue, 31 Oct 2017 11:20:12 +0100 - rev 440090
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1412648 - Initialize some uninitialized fields in tools/profiler/. r=mstange
fff30c59e02e: Bug 1412645 - Initialize some uninitialized fields in widget/cocoa/. r=mstange
Tristan Bourvon <tbourvon@mozilla.com> - Tue, 31 Oct 2017 11:25:04 +0100 - rev 440089
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1412645 - Initialize some uninitialized fields in widget/cocoa/. r=mstange
f4df790d1477: Bug 1412130: Fix window levels on macOS to ensure that popups hide when the application is deactivated. r=mstange
Stephen A Pohl <spohl.mozilla.bugs@gmail.com> - Mon, 30 Oct 2017 21:33:59 -0400 - rev 440005
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1412130: Fix window levels on macOS to ensure that popups hide when the application is deactivated. r=mstange
ff9f8970cfe1: Bug 1408481 - send gfxMacFont font smoothing setting to WebRender. r=mstange
Lee Salzman <lsalzman@mozilla.com> - Sun, 29 Oct 2017 20:21:10 -0400 - rev 439731
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1408481 - send gfxMacFont font smoothing setting to WebRender. r=mstange MozReview-Commit-ID: 9BwLVrfjnsP
908908392903: Bug 1411879 - Use kPMDataFormatXMLCompressed parameter into PMPageFormatCreateDataRepresentation(). r=mstange
Mantaroh Yoshinaga <mantaroh@gmail.com> - Thu, 26 Oct 2017 17:36:16 +0900 - rev 439662
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1411879 - Use kPMDataFormatXMLCompressed parameter into PMPageFormatCreateDataRepresentation(). r=mstange We can use kPMDataFormatXMLCompressed parameter when storing the page format data. As result, this preference data will be 20 times smaller. MozReview-Commit-ID: HMQzhodQyA
8cb78523022e: Bug 1401609 - Add UpdateWebRenderCanvasData() to handle CanvasRenderer re-creation r=jrmuizel,mstange
sotaro <sotaro.ikeda.g@gmail.com> - Sat, 28 Oct 2017 19:07:27 +0900 - rev 439619
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1401609 - Add UpdateWebRenderCanvasData() to handle CanvasRenderer re-creation r=jrmuizel,mstange
1570e72037e2: Bug 1403690 Part 3: macOS change nsLookAndFeel::NativeGetColor to use cached colors. r=mstange
Brad Werth <bwerth@mozilla.com> - Mon, 02 Oct 2017 17:13:48 -0700 - rev 439537
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1403690 Part 3: macOS change nsLookAndFeel::NativeGetColor to use cached colors. r=mstange MozReview-Commit-ID: FHm6d2Vz2sJ
aa6f720f86f0: Bug 1411249 - Handle yet more clipping cases. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Fri, 27 Oct 2017 13:22:16 -0400 - rev 439492
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1411249 - Handle yet more clipping cases. r=mstange This extends the fix in bug 1373802 to account for extra levels of display item nesting. If those extra intermediate display items don't push any clips we still want to pick up the ClipAndScroll from the enclosing ancestor that has it. MozReview-Commit-ID: AmxRz4fBKnX
d76b2163ba1a: Bug 1373802 - Handle more clipping cases. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Fri, 27 Oct 2017 11:41:37 -0400 - rev 439491
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1373802 - Handle more clipping cases. r=mstange We already support cases where we have scrolling clips applied to fixed items. However if we had to build nested clips inside those items, then those nested clips would not properly inherit from the scrolling clips. This patch addresses that case. MozReview-Commit-ID: CWp1x0EsyaP
7b501bcc9d3f: Bug 1410777 - Force a full transaction for WR when the window overlay changes. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Thu, 26 Oct 2017 14:17:19 -0400 - rev 439484
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1410777 - Force a full transaction for WR when the window overlay changes. r=mstange The window buttons are drawn as part of the AddWindowOverlayWebRenderCommands function which is invoked in the full-transaction codepath. It should be possible to have the empty transaction codepath simply update the image (without building a full WR display list) and do a recomposite. That would be more performant but it requires some plumbing to build and ship across a IpcResourceUpdateQueue on empty transactions. MozReview-Commit-ID: 2Mrb0wELD6E
15b3f352292e: Bug 1410777 - Remove unused function and dead code left over from the layers-full WR days. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Thu, 26 Oct 2017 14:17:18 -0400 - rev 439483
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1410777 - Remove unused function and dead code left over from the layers-full WR days. r=mstange This is functionally unrelated to the bug but I noticed it while fiddling with the code, and the lines affected are kind of intertwined with the next patch so I'm just doing the code removal as part of this bug. MozReview-Commit-ID: CwmluhyCdbR
732521699363: Bug 1409508 - Handle another clip edge case. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Wed, 25 Oct 2017 15:05:20 -0400 - rev 439043
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409508 - Handle another clip edge case. r=mstange This edge case happens when: - we have a display item A nested inside another display item B - B has pushed an out-of-band clip, and - A's clip chain doesn't connect directly to the end of B's clip chain but instead connects somewhere farther up the clip chain See comments in the patch for more details. MozReview-Commit-ID: 4mCCaVUQuvH
184c3c90cf23: Bug 1410527 - Update how we tell WR about position:sticky elements. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Wed, 25 Oct 2017 15:21:32 -0400 - rev 439042
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1410527 - Update how we tell WR about position:sticky elements. r=mstange My original understanding of the API was flawed, and so while I had position:sticky working in some cases it didn't work properly in a lot of other cases. This patch corrects the usage of the API to match what WR is expecting and fixes a lot of test cases. MozReview-Commit-ID: AdMux19Fk9U
8990a530c1ea: Bug 1405146 - Don't make the decision to build nsDisplayWrapList based on the number of child items, since this can vary when doing partial display list builds. r=mstange
Matt Woodrow <mwoodrow@mozilla.com> - Wed, 25 Oct 2017 14:17:25 +1300 - rev 438836
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1405146 - Don't make the decision to build nsDisplayWrapList based on the number of child items, since this can vary when doing partial display list builds. r=mstange
e036a5afa854: Bug 1409446 - Remove old code that is now unused. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:46:00 -0400 - rev 438767
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Remove old code that is now unused. r=mstange MozReview-Commit-ID: KEtcbIGTrn3
53894a33da8c: Bug 1409446 - Switch over to the new code. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:46:00 -0400 - rev 438766
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Switch over to the new code. r=mstange One of the sticky-pos tests was only passing because of two wrongs that cancelled each other out in the old code. Specifically, instead of defining a nested clip with the sticky clip as an ancestor, the clip was being defined with the root ASR as an ancestor. Both resulted in the nested clip not scrolling with the actual scrolling scrollframe and so the test was passing. The new code changes things so that the nested clip is defined with the actual scrolling scrollframe as the ancestor, causing the reftest to fail. Fixing the clip ancestry is not hard but it reveals other problems so so I'm deferring that to a follow-up bug. MozReview-Commit-ID: DldAKi1AP4l
4d8523e88330: Bug 1409446 - Treat sticky clips as out-of-band clips. r=ethlin,mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:46:00 -0400 - rev 438765
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Treat sticky clips as out-of-band clips. r=ethlin,mstange MozReview-Commit-ID: C0KEuSrCPEQ
debf2683f01f: Bug 1409446 - Handle nested display item scenarios. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:46:00 -0400 - rev 438764
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Handle nested display item scenarios. r=mstange This handles some cases where a nested display item's clip chain implicitly extends from the wrapper item's clip chain. MozReview-Commit-ID: DmghxOWi81K
4a1434cdcae3: Bug 1409446 - Work around the dual-ancestor case that WR doesn't handle yet. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:45:59 -0400 - rev 438763
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Work around the dual-ancestor case that WR doesn't handle yet. r=mstange Bug 1409442 is tracking a change that will allow scroll layers to have multiple ancestors. Without that, there are cases we cannot properly handle, and so we need to ignore a clip in those scenarios. This patch makes sure we do that instead of crashing. MozReview-Commit-ID: 7AU4uyzT6if
40c84480b4d4: Bug 1409446 - Deal with scenario of two interchangeable DisplayItemClipChain objects causing a cache miss. r=mstange
Kartikaya Gupta <kgupta@mozilla.com> - Tue, 24 Oct 2017 15:45:59 -0400 - rev 438762
Push 8114 by jlorenzo@mozilla.com at 2017-11-02 16:33 +0000
Bug 1409446 - Deal with scenario of two interchangeable DisplayItemClipChain objects causing a cache miss. r=mstange MozReview-Commit-ID: 4FQvOmMUUKH