[Webkit-unassigned] [Bug 173368] New: [GTK][WPE] Raise the timeout values for layout tests

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=173368

            Bug ID: 173368
           Summary: [GTK][WPE] Raise the timeout values for layout tests
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Tools / Tests
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: clopez at igalia.com
                CC: cgarcia at igalia.com, lforschler at apple.com,
                    zan at falconsigh.net

GTK+ debug test bots are constantly giving exit early errors (many times related to timeouts) and on the release test bot that also happens sometimes.
And the release test bot is running test with high priority to avoid random timeouts that would happen otherwise.

I have been checking the timeout values for all ports.

It turns out this is the current status quo:


 - Mac/iOS: They use a 30 second timeout (both for Release and Debug). A special case of 35 seconds is used for the case of the guard_malloc builds.
 - GTK+: It uses 6 seconds for release and 12 for debug. A special case of 10x seconds is used for the case of running under Valgrind.
 - WPE: It uses 6 seconds for both release and debug (There isn't currently any debug test bot for the WPE port)


I couldn't find any explanation for this difference other than r121497 <https://trac.webkit.org/r121497> where the Chromium port defined a 6/12 timeout for release/debug and they defined a default for all other ports of 35 seconds. This general 35 seconds was then lowered to 30 in r177471 <https://trac.webkit.org/r177471>.

I have been thinking between just using the default 30 seconds value like Mac/iOS ports are doing or to raise our values a bit.

Currently I'm leaning towards raising our values to have a 30 seconds for debug and 15 seconds for release. Debug builds are much slower, it makes sense to have higher timeouts.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170614/aeabdf6e/attachment-0001.html>