| Summary: | [ITP] Crash under ResourceLoadStatisticsMemoryStore::removeDataRecords() | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Chris Dumez <cdumez> | ||||
| Component: | WebKit2 | Assignee: | Chris Dumez <cdumez> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | beidson, bfulgham, commit-queue, ddkilzer, rniwa, webkit-bug-importer, wilander, youennf | ||||
| Priority: | P2 | Keywords: | InRadar | ||||
| Version: | WebKit Nightly Build | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Attachments: |
|
||||||
|
Description
Chris Dumez
2018-07-19 14:47:20 PDT
Created attachment 345384 [details]
Patch
Looks good to me. I assume we weren't able to add a test case that caused the crash? (In reply to John Wilander from comment #3) > Looks good to me. I assume we weren't able to add a test case that caused > the crash? I assume this code path is exercised on the bots. However, this is racy and to experience the crash, the store would need to get destroyed on the background thread *while* the WebPageProxy operation is going on on the main thread. Comment on attachment 345384 [details]
Patch
r=me
Comment on attachment 345384 [details] Patch Clearing flags on attachment: 345384 Committed r234020: <https://trac.webkit.org/changeset/234020> All reviewed patches have been landed. Closing bug. |