[Webkit-unassigned] [Bug 192855] New: [GTK][WPE] Unify TestController::platformRunUntil() and honor condition flag
bugzilla-daemon at webkit.org
bugzilla-daemon at webkit.org
Wed Dec 19 08:26:02 PST 2018
https://bugs.webkit.org/show_bug.cgi?id=192855
Bug ID: 192855
Summary: [GTK][WPE] Unify TestController::platformRunUntil()
and honor condition flag
Product: WebKit
Version: Other
Hardware: Unspecified
OS: Unspecified
Status: NEW
Severity: Normal
Priority: P2
Component: Tools / Tests
Assignee: webkit-unassigned at lists.webkit.org
Reporter: aperez at igalia.com
CC: lforschler at apple.com
Currently the TestController::platformRunUntil(bool&, WTF::Seconds) function
is different for the GTK+ and WPE ports, which does not really make much sense,
and also it's not honoring the condition flag at all. Ignoring the flag means
that the following can happen:
1. Some process is started in background which can finish very fast.
2. The background process finishes and sets the flag to “false”, and calls
TestController::notifyDone() to stop the event loop.
3. TestController::platformRunUntil() gets called, and the event loop
gets started even when the condition was already met.
4. The event loop is never stopped because ::notifyDone() is not called
(the early completion already happened), making the test cast timeout.
I have experienced the above scenario about 99% of the time while working
on the content extensions support for the GTK+ and WPE ports, but it could
happen for other kinds of tests as well. I have a strong suspicion that this
may be the cause for the flakyness of some of our tests.
--
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/20181219/14255a67/attachment.html>
More information about the webkit-unassigned
mailing list