| Summary: | [Gstreamer] test http/tests/security/webaudio-render-remote-audio-allowed-crossorigin.html fails | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Arcady Goldmints-Orlov <crzwdjk> | ||||||
| Component: | Tools / Tests | Assignee: | Arcady Goldmints-Orlov <crzwdjk> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Minor | CC: | eocanha, eric.carlson, ews-watchlist, glenn, jer.noble, philipj, pnormand, sergio, webkit-bug-importer | ||||||
| Priority: | P2 | Keywords: | InRadar | ||||||
| Version: | WebKit Nightly Build | ||||||||
| Hardware: | PC | ||||||||
| OS: | Linux | ||||||||
| Attachments: |
|
||||||||
|
Description
Arcady Goldmints-Orlov
2021-12-01 16:59:43 PST
Created attachment 445637 [details]
Patch
I'm not sure adding a new timer is the solution here, it might just make the test flaky :( In a 40 cores machine loaded with "stress -c 400" (10 times more than what the machine can process, showing a load of 395 when running top) the loop that samples the frequency data every 30 ms usually doesn't need extra iterations to get non-silence values. I ran the test (modified to end one second later on purpose) several times and I think only once or twice I got the need of one or two extra iterations at most. 100 ms would allow 3 extra iterations of the frequency sampler. I think that's enough, even on a loaded machine. If I was a reviewer (I'm not), I would accept the patch as is and increase the timeout if the test ever fails (flakes) again. Comment on attachment 445637 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=445637&action=review > LayoutTests/http/tests/security/webaudio-render-remote-audio-allowed-crossorigin.html:46 > + }, 100); ok then please fix indentation here and you'll get the r+ you deserve Created attachment 445749 [details]
Patch
Committed r286486 (244825@main): <https://commits.webkit.org/244825@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 445749 [details]. |