Bug 229668

Summary: [GPU Process](REGRESSION): A detached canvas leaks all the images it draws
Product: WebKit Reporter: Maximilian Böhm <mb>
Component: CanvasAssignee: Said Abou-Hallawa <sabouhallawa>
Status: RESOLVED FIXED    
Severity: Normal CC: cdumez, changseok, dino, eric.carlson, esprehn+autocc, ews-watchlist, gyuyoung.kim, kangil.han, kondapallykalyan, sabouhallawa, simon.fraser, webkit-bug-importer, youennf
Priority: P2 Keywords: InRadar
Version: Safari Technology Preview   
Hardware: iPhone / iPad   
OS: Unspecified   
See Also: https://bugs.webkit.org/show_bug.cgi?id=226813
https://bugs.webkit.org/show_bug.cgi?id=212594
https://bugs.webkit.org/show_bug.cgi?id=229790
Attachments:
Description Flags
Memory usage
none
Patch
none
Patch
none
Patch
none
simpler repro test case none

Maximilian Böhm
Reported 2021-08-30 07:07:55 PDT
Created attachment 436766 [details] Memory usage Basically we are having a video (getUserMedia) and mirror it into a canvas. As long as the canvas is not appended to the DOM, we have a memory leak. After 40 seconds Safari reloads the page... Funny, we are having 18446744073,98 GB of RAM! (See screenshot in attachment) Reproduction: https://jsfiddle.net/mkn2tb3h/ Device: iPad 8 iOS: 15.0 (19A5337a) Kind regards, Maximilian Böhm
Attachments
Memory usage (218.22 KB, image/png)
2021-08-30 07:07 PDT, Maximilian Böhm 		no flags
Details
Patch (4.32 KB, patch)
2021-08-31 11:05 PDT, Said Abou-Hallawa 		no flags
DetailsFormatted DiffDiff
Patch (2.08 KB, patch)
2021-08-31 13:19 PDT, Said Abou-Hallawa 		no flags
DetailsFormatted DiffDiff
Patch (3.46 KB, patch)
2021-08-31 14:11 PDT, Said Abou-Hallawa 		no flags
DetailsFormatted DiffDiff
simpler repro test case (1.45 KB, text/html)
2021-09-01 00:25 PDT, Said Abou-Hallawa 		no flags
Details
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.
Maximilian Böhm
Comment 1 2021-08-30 07:15:30 PDT
GPU Process: Canvas Rendering has to be enabled. If it is disabled, the error does not occur.
Radar WebKit Bug Importer
Comment 2 2021-08-30 09:45:35 PDT
<rdar://problem/82532484>
Said Abou-Hallawa
Comment 3 2021-08-31 00:37:29 PDT
This bug is similar to bug 226813.
Said Abou-Hallawa
Comment 4 2021-08-31 01:23:25 PDT
The reason of this bug is the if-statement in this loop in Document::prepareCanvasesForDisplayIfNeeded(): for (auto& canvas : canvases) { // However, if they are not in the document body, then they won't // be composited and thus don't need preparation. Unfortunately they // can't tell at the time they were added to the list, since they // could be inserted or removed from the document body afterwards. if (!canvas->isInTreeScope()) continue; canvas->prepareForDisplay(); } So if the canvas was not added to the DOM, prepareForDisplay() will not be called. This causes the images of the video to be accumulated in the WebProcess side and hence they are never released in the GPUProcess as well.
Said Abou-Hallawa
Comment 5 2021-08-31 01:43:31 PDT
This if-statement was added in r262498.
Said Abou-Hallawa
Comment 6 2021-08-31 11:05:55 PDT
Created attachment 436916 [details] Patch
Said Abou-Hallawa
Comment 7 2021-08-31 13:19:33 PDT
Created attachment 436932 [details] Patch
Simon Fraser (smfr)
Comment 8 2021-08-31 13:33:35 PDT
Comment on attachment 436932 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=436932&action=review > Source/WebCore/dom/Document.cpp:8942 > + if (canvas->isInTreeScope() || (canvas->renderingContext() && canvas->renderingContext()->is2d())) > + canvas->prepareForDisplay(); We instead always call preprareForDisplay() but inside the implementation of that function do the isInTreeScope() check when appropriate.
Said Abou-Hallawa
Comment 9 2021-08-31 14:11:01 PDT
Created attachment 436941 [details] Patch
Simon Fraser (smfr)
Comment 10 2021-08-31 15:58:46 PDT
Comment on attachment 436941 [details] Patch Can we write a test that detect the memory growth?
Said Abou-Hallawa
Comment 11 2021-09-01 00:24:30 PDT
(In reply to Simon Fraser (smfr) from comment #10) > Comment on attachment 436941 [details] > Patch > > Can we write a test that detect the memory growth? There is no easy way to check the WebProcess memory status before and after drawing images to the canvas. There should be an Internals API to facilitate this check. I think this can be done in a separate patch. I am going to attach a simpler repro test case which can be added as a layout test once the Internals API is there.
Said Abou-Hallawa
Comment 12 2021-09-01 00:25:26 PDT
Created attachment 436998 [details] simpler repro test case
EWS
Comment 13 2021-09-01 00:55:46 PDT
Committed r281839 (241172@main): <https://commits.webkit.org/241172@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 436941 [details].
Note You need to log in before you can comment on or make changes to this bug.