RESOLVED FIXED 84821
fast/js/random-array-gc-stress times out in debug mode
https://bugs.webkit.org/show_bug.cgi?id=84821
Summary fast/js/random-array-gc-stress times out in debug mode
Mark Hahnenberg
Reported 2012-04-24 20:05:47 PDT
We should reduce the number of iterations so that it completes.
Attachments
Patch (1.71 KB, patch)
2012-04-24 20:07 PDT, Mark Hahnenberg
no flags
Patch (1.82 KB, patch)
2012-04-24 20:12 PDT, Mark Hahnenberg
no flags
Patch (1.21 KB, patch)
2012-04-25 16:39 PDT, Mark Hahnenberg
no flags
Patch (1.21 KB, patch)
2012-04-27 11:55 PDT, Mark Hahnenberg
no flags
Mark Hahnenberg
Comment 1 2012-04-24 20:07:06 PDT
Darin Adler
Comment 2 2012-04-24 20:09:31 PDT
Comment on attachment 138723 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=138723&action=review > LayoutTests/fast/js/script-tests/random-array-gc-stress.js:-12 > -function gc() > -{ > - if (this.GCController) > - GCController.collect(); > - else > - for (var i = 0; i < 10000; ++i) // Allocate a sufficient number of objects to force a GC. > - ({}); > -} Why are you removing this? Change log does not say.
Mark Hahnenberg
Comment 3 2012-04-24 20:12:51 PDT
Geoffrey Garen
Comment 4 2012-04-24 20:19:38 PDT
Comment on attachment 138724 [details] Patch Seems like this will fail to run due to no gc() function.
Darin Adler
Comment 5 2012-04-24 20:26:17 PDT
Comment on attachment 138724 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=138724&action=review > LayoutTests/ChangeLog:10 > + * fast/js/script-tests/random-array-gc-stress.js: Reduce number of iterations so > + that the test can complete without timing out. Also remove the gc() function because > + a gc() function is already provided in resources/js-test-pre.js. Geoff, here’s the explanation of why it’s good to remove the gc function.
WebKit Review Bot
Comment 6 2012-04-25 08:09:56 PDT
Comment on attachment 138724 [details] Patch Clearing flags on attachment: 138724 Committed r115208: <http://trac.webkit.org/changeset/115208>
WebKit Review Bot
Comment 7 2012-04-25 08:10:02 PDT
All reviewed patches have been landed. Closing bug.
Mark Hahnenberg
Comment 8 2012-04-25 11:09:23 PDT
Apparently the bots are too slow even for the new version, which passes locally on my machine. Will try to reduce the number of iterations even further.
Mark Hahnenberg
Comment 9 2012-04-25 16:39:34 PDT
Geoffrey Garen
Comment 10 2012-04-25 18:22:47 PDT
Comment on attachment 138897 [details] Patch r=me
WebKit Review Bot
Comment 11 2012-04-25 22:06:49 PDT
Comment on attachment 138897 [details] Patch Clearing flags on attachment: 138897 Committed r115287: <http://trac.webkit.org/changeset/115287>
WebKit Review Bot
Comment 12 2012-04-25 22:06:54 PDT
All reviewed patches have been landed. Closing bug.
Mark Hahnenberg
Comment 13 2012-04-26 14:28:22 PDT
Still timing out on some bots.
Mark Hahnenberg
Comment 14 2012-04-27 11:55:56 PDT
Darin Adler
Comment 15 2012-04-27 18:21:07 PDT
Comment on attachment 139243 [details] Patch OK. I am increasingly unhappy with this test if it takes so long to run.
WebKit Review Bot
Comment 16 2012-04-30 12:23:18 PDT
Comment on attachment 139243 [details] Patch Clearing flags on attachment: 139243 Committed r115658: <http://trac.webkit.org/changeset/115658>
WebKit Review Bot
Comment 17 2012-04-30 12:23:39 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.