35166 2010-02-19 10:39:44 -0800 [V8] Worker crashes due to r54972 2010-02-19 11:42:07 -0800 1 1 1 Unclassified WebKit WebCore Misc. 528+ (Nightly build) All All RESOLVED FIXED P2 Normal --- 1 japhet japhet dglazkov oldest_to_newest 191969 0 japhet 2010-02-19 10:39:44 -0800 I didn't properly test the worker-related changes in the final version of r54972, and I screwed up unwrapping global objects when we're in a WorkerContext. Patch momentarily. 191971 1 49092 japhet 2010-02-19 10:51:28 -0800 Created attachment 49092 patch 191977 2 49092 dglazkov 2010-02-19 10:53:14 -0800 Comment on attachment 49092 patch ok. 191993 3 japhet 2010-02-19 11:26:29 -0800 http://trac.webkit.org/changeset/55023 192001 4 ap 2010-02-19 11:33:33 -0800 Is this covered by existing regression tests? 192002 5 japhet 2010-02-19 11:42:07 -0800 (In reply to comment #4) > Is this covered by existing regression tests? Yes, but unfortunately, most (all?) of the tests that cover v8-related worker code are run under a full build of chromium, so problems didn't show up in any buildbots until chromium tried to merge passed r54972. I thought I had run those tests locally prior to committing r54972, but apparently I hadn't. Mea culpa. 49092 2010-02-19 10:51:28 -0800 2010-02-19 10:53:14 -0800 patch patch.txt text/plain 1281 japhet SW5kZXg6IFdlYkNvcmUvQ2hhbmdlTG9nCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KLS0tIFdlYkNvcmUvQ2hhbmdlTG9n CShyZXZpc2lvbiA1NTAyMCkKKysrIFdlYkNvcmUvQ2hhbmdlTG9nCSh3b3JraW5nIGNvcHkpCkBA IC0xLDMgKzEsMTQgQEAKKzIwMTAtMDItMTkgIE5hdGUgQ2hhcGluICA8amFwaGV0QGNocm9taXVt Lm9yZz4KKworICAgICAgICBSZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBb VjhdIEZpeCBXb3JrZXIgY3Jhc2ggcmVncmVzc2lvbiBpbiByNTQ5NzIKKworICAgICAgICBodHRw czovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MzUxNjYKKworICAgICAgICAqIGJp bmRpbmdzL3Y4L1Y4RE9NV3JhcHBlci5jcHA6CisgICAgICAgIChXZWJDb3JlOjpWOERPTVdyYXBw ZXI6Omluc3RhbnRpYXRlVjhPYmplY3QpOiBQcm9wZXJseSB1bndyYXAgZ2xvYmFsIG9iamVjdHMg d2hlbiBpbnNpZGUgYSBXb3JrZXJDb250ZXh0LgorCiAyMDEwLTAyLTE5ICBTdGV2ZSBCbG9jayAg PHN0ZXZlYmxvY2tAZ29vZ2xlLmNvbT4KIAogICAgICAgICBSZXZpZXdlZCBieSBEYXZpZCBMZXZp bi4KSW5kZXg6IFdlYkNvcmUvYmluZGluZ3MvdjgvVjhET01XcmFwcGVyLmNwcAo9PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 Ci0tLSBXZWJDb3JlL2JpbmRpbmdzL3Y4L1Y4RE9NV3JhcHBlci5jcHAJKHJldmlzaW9uIDU1MDIw KQorKysgV2ViQ29yZS9iaW5kaW5ncy92OC9WOERPTVdyYXBwZXIuY3BwCSh3b3JraW5nIGNvcHkp CkBAIC0yOTUsNyArMjk1LDcgQEAKICAgICAgICAgICAgIGlmIChnbG9iYWxPYmplY3RQcm90b3R5 cGVJc0RPTVdpbmRvdyhnbG9iYWxQcm90b3R5cGUpKQogICAgICAgICAgICAgICAgIHByb3h5ID0g VjhQcm94eTo6cmV0cmlldmUoVjhET01XaW5kb3c6OnRvTmF0aXZlKGdsb2JhbFByb3RvdHlwZSkt PmZyYW1lKCkpOwogICAgICAgICAgICAgZWxzZQotICAgICAgICAgICAgICAgIHdvcmtlckNvbnRl eHQgPSBWOFdvcmtlckNvbnRleHQ6OnRvTmF0aXZlKGdsb2JhbFByb3RvdHlwZSk7CisgICAgICAg ICAgICAgICAgd29ya2VyQ29udGV4dCA9IFY4V29ya2VyQ29udGV4dDo6dG9OYXRpdmUobG9va3Vw RE9NV3JhcHBlcihWOFdvcmtlckNvbnRleHQ6OkdldFRlbXBsYXRlKCksIGNvbnRleHQtPkdsb2Jh bCgpKSk7CiAgICAgICAgIH0KICAgICB9CiAK