|Summary:||Reshuffle some code in WebKitDriver._read_block in preparation for reading stderr/stdout separately|
|Product:||WebKit||Reporter:||Eric Seidel (no email) <eric>|
|Component:||New Bugs||Assignee:||Eric Seidel (no email) <eric>|
|Severity:||Normal||CC:||abarth, rniwa, webkit.review.bot|
|Version:||528+ (Nightly build)|
|Bug Depends on:||67557|
Description Eric Seidel (no email) 2011-09-02 15:13:03 PDT
Reshuffle some code in WebKitDriver._read_block in preparation for reading stderr/stdout separately
Comment 2 Eric Seidel (no email) 2011-09-02 15:19:23 PDT
Committed r94453: <http://trac.webkit.org/changeset/94453>
Comment 3 Ryosuke Niwa 2011-09-02 23:56:23 PDT
Comment 4 Ryosuke Niwa 2011-09-03 00:02:11 PDT
I have confirmed that this patch is causing the failures. Since these failures are preventing the rest of tests to run, I'm going to rollout the patch for now unless either one of you can fix it in the next 10 minutes or so.
Comment 5 Ryosuke Niwa 2011-09-03 00:29:04 PDT
Rolled the patch out in http://trac.webkit.org/changeset/94481.
Comment 6 WebKit Review Bot 2011-09-03 00:32:13 PDT
Comment on attachment 106205 [details] Patch Rejecting attachment 106205 [details] from commit-queue. Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 2 Last 500 characters of output: Scripts/svn-apply', u'--reviewer', u'Adam Barth', u'--force']" exit_code: 1 Parsed 2 diffs from patch file(s). patching file Tools/ChangeLog Hunk #1 succeeded at 1 with fuzz 3. patching file Tools/Scripts/webkitpy/layout_tests/port/webkit.py Hunk #1 FAILED at 509. 1 out of 1 hunk FAILED -- saving rejects to file Tools/Scripts/webkitpy/layout_tests/port/webkit.py.rej Failed to run "[u'/mnt/git/webkit-commit-queue/Tools/Scripts/svn-apply', u'--reviewer', u'Adam Barth', u'--force']" exit_code: 1 Full output: http://queues.webkit.org/results/9587229
Comment 7 Ryosuke Niwa 2011-09-03 00:44:49 PDT
Also rolled out the unreviewed followup in http://trac.webkit.org/changeset/94483.
Comment 8 Eric Seidel (no email) 2011-09-03 01:17:41 PDT
Comment 9 Eric Seidel (no email) 2011-09-09 11:43:11 PDT
I guess I'll try this again. :)
Comment 10 Eric Seidel (no email) 2011-09-12 12:59:51 PDT
I found the bug in the original patch, fixed. I also realize now that this is subtly changing behavior. Before we were not correctly respecting the deadline. This should make our timeouts much more accurate, as we're re-computing when the timeout is every time we read a line, instead of computing up front that we should wait 30 seconds, and then making every read wait 30 seconds. The old code would wait forever for a test which slowly produced output.