If the web inspector is used during a QtLauncher session, when the program is closed it always results in a segfault: zsh: segmentation fault ./bin/QtLauncher
Created attachment 35099 [details] patch to fix segfault
Is this possible to test?
Would it be possible to test this by creating InspectorController and watching the signal when it gets closed?
Comment on attachment 35099 [details] patch to fix segfault I'm no Qt expert, but this looks sane to me.
Comment on attachment 35099 [details] patch to fix segfault Hum... removing cq+ now that Qt folks are commenting here too. :) A qt committer can set it again if they're comfortable with this, or if the bug goes stale I'll set it later this week.
Comment on attachment 35099 [details] patch to fix segfault Rejecting patch 35099 from commit-queue. This patch will require manual commit. ['WebKitTools/Scripts/run-webkit-tests', '--no-launch-safari', '--quiet', '--exit-after-n-failures=1'] failed with exit code 1
Comment on attachment 35099 [details] patch to fix segfault Sorry, you were yet another victim of bug 28845! It's a bad regression which is causing flakey media tests. :( media/controls-after-reload.html -> crashed
Comment on attachment 35099 [details] patch to fix segfault Rejecting patch 35099 from commit-queue. This patch will require manual commit. WebKitTools/Scripts/build-webkit failed with exit code 1
Comment on attachment 35099 [details] patch to fix segfault I think the build was broken for other reasons and the bots had not yet rolled red.
Comment on attachment 35099 [details] patch to fix segfault Clearing flags on attachment: 35099 Committed r48039: <http://trac.webkit.org/changeset/48039>
All reviewed patches have been landed. Closing bug.
(In reply to comment #7) > (From update of attachment 35099 [details]) > Sorry, you were yet another victim of bug 28845! It's a bad regression which > is causing flakey media tests. :( > media/controls-after-reload.html -> crashed Actually, I was wrong, the stack trace looks different. I've filed bug 29037.