Issue metadata
Sign in to add a comment
|
Unexpected box-shadow afterglow
Reported by
matthieu...@gmail.com,
Mar 30 2018
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36 Steps to reproduce the problem: 1. Focus and leave inputs many times 2. Afterglow randomly appear What is the expected behavior? box-shadow should not leave an afterglow when the property is set to none What went wrong? A removed box-shadow leave an afterglow (sometimes), browser repaint is not consistent Did this work before? No Does this work in other browsers? Yes Chrome version: 65.0.3325.181 Channel: stable OS Version: 10.0 Flash Version:
,
Mar 30 2018
I'm not seeing any problem in M-66. Can test team please check M-65 to see if this can be reproduced. Will remove release block once we understand the situation.
,
Mar 30 2018
I can confirm it's still an issue on canary (67) It can be related to zoom level, I have a base zoom level of 125% on Windows
,
Mar 30 2018
In theory I tested with that, although via a 125% windows DPI setting. Is your Display Properties zoom set to 100%? Just trying to nail down what we should test with. Invalidation issues with zoom are not unusual right now.
,
Apr 1 2018
,
Apr 2 2018
Unable to reproduce the issue on chrome reported version 65.0.3325.181 using windows-10 with steps mentioned below: 1) Launched chrome reported version and opened codepen: https://codepen.io/frlinw/pen/dmmVLm?editors=1100 given in comment#1 2) Focused and left the input field many times, didn't observed any break in glow after leaving input filed for manytimes @Reporter: Please find the attached screen cast for your reference and let us know if we missed anything in reproducing the issue, provide your feedback on it which help in further triaging it.
,
Apr 2 2018
,
Apr 2 2018
Thank you for providing more feedback. Adding the requester to the cc list. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Apr 2 2018
OK, we'll try some more to reproduce it.
,
Apr 2 2018
,
Apr 9 2018
OK, I can reproduce this on Windows Canary. The really odd thing is that the blur that should not be there blinks with the cursor, despite not showing in DevTools as being repainted. The only explanation I can think of for this is that a paint chunk for the blur is being associated with the caret and re-drawn as the caret redraws. But if I start with --disable-slimming-paint-v175 I get the same result. Over to someone who might have a better theory of what's causing this.
,
Apr 9 2018
I have ever seen in another bug (bug 816854) that some corrupted painting blinks with the cursor. At least that bug was not a paint invalidation issue. The phenomenon seems related to some inconsistent rasterization result of some paint operations affected by other paint operations (e.g. the caret). For bug 816854, the problem is that Skia sometimes modifies pixels that are far away from a filter effect when Skia rasterizes the filter. The solution was to add explicit bounds for the filter. Not sure if this bug's situation is similar. Will investigate more.
,
Apr 10 2018
@schenney, the command-line flag for disabling SPV175 is: --disable-blink-features=SlimmingPaintV175 (adding this here because I keep having to look it up too)
,
Apr 12 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/2d526f29f2a8d8df3cab30b7ae407106574ce547 commit 2d526f29f2a8d8df3cab30b7ae407106574ce547 Author: Xianzhu Wang <wangxianzhu@chromium.org> Date: Thu Apr 12 05:10:50 2018 [PE] Let visual rect cover sub-pixel visual effect outsets If an overflowing visual effect has sub-pixel geometry and is painted with anti-aliasing along pixel-snapped border box, the pixel-snapping may cause the anti-aliased edge overflow the calculated visual rect. Let LayoutBox::VisualRectOutsetForRasterEffects() return 1 if there are any sub-pixel visual effect outsets to ensure the visual rect covers all pixels. Bug: 827516 Change-Id: I48f8350fb2124d002333a4cc76486c63ca55f8c9 Reviewed-on: https://chromium-review.googlesource.com/1003492 Reviewed-by: Chris Harrelson <chrishtr@chromium.org> Reviewed-by: Florin Malita <fmalita@chromium.org> Cr-Commit-Position: refs/heads/master@{#550014} [add] https://crrev.com/2d526f29f2a8d8df3cab30b7ae407106574ce547/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move-expected.html [add] https://crrev.com/2d526f29f2a8d8df3cab30b7ae407106574ce547/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move.html [modify] https://crrev.com/2d526f29f2a8d8df3cab30b7ae407106574ce547/third_party/blink/renderer/core/layout/layout_box.cc [modify] https://crrev.com/2d526f29f2a8d8df3cab30b7ae407106574ce547/third_party/blink/renderer/core/layout/layout_box.h [modify] https://crrev.com/2d526f29f2a8d8df3cab30b7ae407106574ce547/third_party/blink/renderer/core/layout/overflow_model.h
,
Apr 12 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf commit 2195c17ababe2e9f24c9e9fe7d5da2a9915349cf Author: Xianzhu Wang <wangxianzhu@chromium.org> Date: Thu Apr 12 16:13:23 2018 Revert "[PE] Let visual rect cover sub-pixel visual effect outsets" This reverts commit 2d526f29f2a8d8df3cab30b7ae407106574ce547. Reason for revert: crbug.com/832020 Bug: 832020 Original change's description: > [PE] Let visual rect cover sub-pixel visual effect outsets > > If an overflowing visual effect has sub-pixel geometry and is painted > with anti-aliasing along pixel-snapped border box, the pixel-snapping > may cause the anti-aliased edge overflow the calculated visual rect. > > Let LayoutBox::VisualRectOutsetForRasterEffects() return 1 if there are > any sub-pixel visual effect outsets to ensure the visual rect covers > all pixels. > > Bug: 827516 > Change-Id: I48f8350fb2124d002333a4cc76486c63ca55f8c9 > Reviewed-on: https://chromium-review.googlesource.com/1003492 > Reviewed-by: Chris Harrelson <chrishtr@chromium.org> > Reviewed-by: Florin Malita <fmalita@chromium.org> > Cr-Commit-Position: refs/heads/master@{#550014} TBR=wangxianzhu@chromium.org,fmalita@chromium.org,chrishtr@chromium.org Change-Id: Iea3cfeb8e661dac43f27f85ab1eb2e7f00da8b73 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 827516 Reviewed-on: https://chromium-review.googlesource.com/1010582 Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#550227} [delete] https://crrev.com/210c995247166fa2cdd38ab71e1c5cfc38871894/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move-expected.html [delete] https://crrev.com/210c995247166fa2cdd38ab71e1c5cfc38871894/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move.html [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/layout_box.cc [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/layout_box.h [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/overflow_model.h
,
Apr 12 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/19327a652bb7c8806ec437ba577f280c5962e12e commit 19327a652bb7c8806ec437ba577f280c5962e12e Author: Xianzhu Wang <wangxianzhu@chromium.org> Date: Thu Apr 12 20:32:43 2018 Reland "[PE] Let visual rect cover sub-pixel visual effect outsets" This is a reland of 2d526f29f2a8d8df3cab30b7ae407106574ce547 Was reverted because it caused crbug.com/832020 which is a crash in the following case: <div style="width: 0; height: 0; box-shadow: 0 5.5px"></div> The div has zero size and visual effect outset on one side only, so the overflow is still zero and doesn't create OverflowModel. Original change's description: > [PE] Let visual rect cover sub-pixel visual effect outsets > > If an overflowing visual effect has sub-pixel geometry and is painted > with anti-aliasing along pixel-snapped border box, the pixel-snapping > may cause the anti-aliased edge overflow the calculated visual rect. > > Let LayoutBox::VisualRectOutsetForRasterEffects() return 1 if there are > any sub-pixel visual effect outsets to ensure the visual rect covers > all pixels. > > Bug: 827516 > Change-Id: I48f8350fb2124d002333a4cc76486c63ca55f8c9 > Reviewed-on: https://chromium-review.googlesource.com/1003492 > Reviewed-by: Chris Harrelson <chrishtr@chromium.org> > Reviewed-by: Florin Malita <fmalita@chromium.org> > Cr-Commit-Position: refs/heads/master@{#550014} Bug: 827516 , 832020 Change-Id: I4e7831d7d91cc541127670ac102ec0292ae50387 Reviewed-on: https://chromium-review.googlesource.com/1010583 Reviewed-by: Chris Harrelson <chrishtr@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#550334} [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move-expected.html [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move.html [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-zero-size-box-crash.html [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/layout_box.cc [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/layout_box.h [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/overflow_model.h
,
Apr 13 2018
,
Apr 13 2018
[Auto-generated comment by a script] We noticed that this issue is targeted for M-67; it appears the fix may have landed after branch point, meaning a merge might be required. Please confirm if a merge is required here - if so add Merge-Request-67 label, otherwise remove Merge-TBD label. Thanks.
,
Apr 13 2018
M67 branch #3396 was branched at chromium revision 550428, so CLs listed here already in M67. Pls double check and remove Merge-TBD label. Thank you.
,
Apr 13 2018
Yes, it's in branch #3396.
,
Apr 16 2018
The NextAction date has arrived: 2018-04-16
,
Apr 17 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf commit 2195c17ababe2e9f24c9e9fe7d5da2a9915349cf Author: Xianzhu Wang <wangxianzhu@chromium.org> Date: Thu Apr 12 16:13:23 2018 Revert "[PE] Let visual rect cover sub-pixel visual effect outsets" This reverts commit 2d526f29f2a8d8df3cab30b7ae407106574ce547. Reason for revert: crbug.com/832020 Bug: 832020 Original change's description: > [PE] Let visual rect cover sub-pixel visual effect outsets > > If an overflowing visual effect has sub-pixel geometry and is painted > with anti-aliasing along pixel-snapped border box, the pixel-snapping > may cause the anti-aliased edge overflow the calculated visual rect. > > Let LayoutBox::VisualRectOutsetForRasterEffects() return 1 if there are > any sub-pixel visual effect outsets to ensure the visual rect covers > all pixels. > > Bug: 827516 > Change-Id: I48f8350fb2124d002333a4cc76486c63ca55f8c9 > Reviewed-on: https://chromium-review.googlesource.com/1003492 > Reviewed-by: Chris Harrelson <chrishtr@chromium.org> > Reviewed-by: Florin Malita <fmalita@chromium.org> > Cr-Commit-Position: refs/heads/master@{#550014} TBR=wangxianzhu@chromium.org,fmalita@chromium.org,chrishtr@chromium.org Change-Id: Iea3cfeb8e661dac43f27f85ab1eb2e7f00da8b73 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: 827516 Reviewed-on: https://chromium-review.googlesource.com/1010582 Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#550227} [delete] https://crrev.com/210c995247166fa2cdd38ab71e1c5cfc38871894/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move-expected.html [delete] https://crrev.com/210c995247166fa2cdd38ab71e1c5cfc38871894/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move.html [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/layout_box.cc [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/layout_box.h [modify] https://crrev.com/2195c17ababe2e9f24c9e9fe7d5da2a9915349cf/third_party/blink/renderer/core/layout/overflow_model.h
,
Apr 17 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/19327a652bb7c8806ec437ba577f280c5962e12e commit 19327a652bb7c8806ec437ba577f280c5962e12e Author: Xianzhu Wang <wangxianzhu@chromium.org> Date: Thu Apr 12 20:32:43 2018 Reland "[PE] Let visual rect cover sub-pixel visual effect outsets" This is a reland of 2d526f29f2a8d8df3cab30b7ae407106574ce547 Was reverted because it caused crbug.com/832020 which is a crash in the following case: <div style="width: 0; height: 0; box-shadow: 0 5.5px"></div> The div has zero size and visual effect outset on one side only, so the overflow is still zero and doesn't create OverflowModel. Original change's description: > [PE] Let visual rect cover sub-pixel visual effect outsets > > If an overflowing visual effect has sub-pixel geometry and is painted > with anti-aliasing along pixel-snapped border box, the pixel-snapping > may cause the anti-aliased edge overflow the calculated visual rect. > > Let LayoutBox::VisualRectOutsetForRasterEffects() return 1 if there are > any sub-pixel visual effect outsets to ensure the visual rect covers > all pixels. > > Bug: 827516 > Change-Id: I48f8350fb2124d002333a4cc76486c63ca55f8c9 > Reviewed-on: https://chromium-review.googlesource.com/1003492 > Reviewed-by: Chris Harrelson <chrishtr@chromium.org> > Reviewed-by: Florin Malita <fmalita@chromium.org> > Cr-Commit-Position: refs/heads/master@{#550014} Bug: 827516 , 832020 Change-Id: I4e7831d7d91cc541127670ac102ec0292ae50387 Reviewed-on: https://chromium-review.googlesource.com/1010583 Reviewed-by: Chris Harrelson <chrishtr@chromium.org> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org> Cr-Commit-Position: refs/heads/master@{#550334} [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move-expected.html [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-move.html [add] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/WebKit/LayoutTests/paint/invalidation/subpixel-shadow-zero-size-box-crash.html [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/layout_box.cc [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/layout_box.h [modify] https://crrev.com/19327a652bb7c8806ec437ba577f280c5962e12e/third_party/blink/renderer/core/layout/overflow_model.h |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by matthieu...@gmail.com
, Mar 30 2018