44199 2010-08-18 13:21:11 -0700 perf tests are flaky 2011-02-09 22:31:17 -0800 1 1 1 Unclassified WebKit Tools / Tests 528+ (Nightly build) PC OS X 10.5 RESOLVED FIXED P2 Normal --- 1 ojan webkit-unassigned abarth eric mrobinson rniwa tony oldest_to_newest 266335 0 ojan 2010-08-18 13:21:11 -0700 Some of the new perf tests seem to be flaky. I'll skip them on non-chromium platforms and mark them as PASS TEXT for chromium. That way, we can get flakiness data and make them non-flaky without turning bots red in in interim. 268834 1 65331 ojan 2010-08-24 15:04:53 -0700 Created attachment 65331 Patch 268843 2 65331 tony 2010-08-24 15:18:18 -0700 Comment on attachment 65331 Patch Ok for testing flakiness. If it works, we'll need to do this properly with layoutTestController, otherwise we can just revert this patch. 268847 3 ojan 2010-08-24 15:22:25 -0700 Committed r65940: <http://trac.webkit.org/changeset/65940> 347392 4 mrobinson 2011-02-08 08:06:09 -0800 Any updates on this? 348816 5 ojan 2011-02-09 22:30:31 -0800 It did not fix the problem. I don't remember if it improved things though. Some of the tests are not flaky at all, but some of them are super flaky. See http://test-results.appspot.com/dashboards/flakiness_dashboard.html#master=ChromiumWebkit&tests=%5Eperf for the results of the flaky tests on the chromium bots. The ones that are not flaky don't show up there. Also, some of the ones listed there look like they crashed when all the tests in that run crashed (i.e. are not flaky). It's been vaguely on my TODO list to try and understand the cause of the flakiness, but I haven't gotten around to it. 348817 6 ojan 2011-02-09 22:31:17 -0800 I suppose we could take the non-flaky ones off the Skipped lists. 65331 2010-08-24 15:04:53 -0700 2010-08-24 15:18:17 -0700 Patch bug-44199-20100824150452.patch text/plain 2763 ojan ZGlmZiAtLWdpdCBhL0xheW91dFRlc3RzL0NoYW5nZUxvZyBiL0xheW91dFRlc3RzL0NoYW5nZUxv ZwppbmRleCAyYWI5OTg0N2IxOWE3OTZkYjE3MjQ3MGI4YTIxMTg1ZjE4YjE4OWFjLi43YzBmMWUx YWRlMmE1Y2EwZjAzYWExZGI3ZGRhZjNkOGVmZmQ2YjEwIDEwMDY0NAotLS0gYS9MYXlvdXRUZXN0 cy9DaGFuZ2VMb2cKKysrIGIvTGF5b3V0VGVzdHMvQ2hhbmdlTG9nCkBAIC0xLDMgKzEsMTkgQEAK KzIwMTAtMDgtMjQgIE9qYW4gVmFmYWkgIDxvamFuQGNocm9taXVtLm9yZz4KKworICAgICAgICBS ZXZpZXdlZCBieSBOT0JPRFkgKE9PUFMhKS4KKworICAgICAgICBwZXJmIHRlc3RzIGFyZSBmbGFr eQorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9NDQxOTkK KworICAgICAgICBTcGVjdWxhdGl2ZSBmaXggZm9yIHBlcmYgdGVzdCBmbGFraW5lc3MuIFVzZSBj aHJvbWl1bS5JbnRlcnZhbAorICAgICAgICB0byBnZXQgYSBtaWNyb3NlY29uZCBncmFudWxhcml0 eSB0aW1lci4gSWYgdGhpcyBkb2VzIGVuZCB1cAorICAgICAgICByZWR1Y2luZyBmbGFraW5lc3Ms IHRoZW4gd2UgY2FuIGV4cG9zZSBzb21ldGhpbmcgc2ltaWxhciBmb3IKKyAgICAgICAgSlNDIGFu ZC9vciBsYXlvdXRUZXN0Q29udHJvbGxlci4KKworICAgICAgICAqIHJlc291cmNlcy9tYWduaXR1 ZGUtcGVyZi5qczoKKyAgICAgICAgKE1hZ25pdHVkZS5fcnVuSXRlcmF0aW9uKToKKyAgICAgICAg KE1hZ25pdHVkZSk6CisKIDIwMTAtMDgtMjQgIFNpbW9uIEZyYXNlciAgPHNpbW9uLmZyYXNlckBh cHBsZS5jb20+CiAKICAgICAgICAgUmV2aWV3ZWQgYnkgU2FtIFdlaW5pZy4KZGlmZiAtLWdpdCBh L0xheW91dFRlc3RzL3Jlc291cmNlcy9tYWduaXR1ZGUtcGVyZi5qcyBiL0xheW91dFRlc3RzL3Jl c291cmNlcy9tYWduaXR1ZGUtcGVyZi5qcwppbmRleCBhOGM3MjczOGY2M2E0Y2VhY2JiYTNjNWQx ODAwZTI0NjNlZmRlYTk2Li5jYmQxYjAyNzVhMzQ4ZTE3MWYwZmQ5M2IzNGNiMzU3NTEwYWUzZDBj IDEwMDY0NAotLS0gYS9MYXlvdXRUZXN0cy9yZXNvdXJjZXMvbWFnbml0dWRlLXBlcmYuanMKKysr IGIvTGF5b3V0VGVzdHMvcmVzb3VyY2VzL21hZ25pdHVkZS1wZXJmLmpzCkBAIC0yMDMsMTQgKzIw MywzMCBAQCBNYWduaXR1ZGUuX3J1bkl0ZXJhdGlvbiA9IGZ1bmN0aW9uKHNldHVwLCB0ZXN0LCBt YWduaXR1ZGUsIG1pbGxpc2Vjb25kcywgcnVuc1BlcgogICAgIE1hZ25pdHVkZS5fZGVidWcoJ3J1 biBpdGVyYXRpb24uIG1hZ25pdHVkZSAnICsgbWFnbml0dWRlICsgIiBtaWxsaXNlY29uZHMgIiAr IG1pbGxpc2Vjb25kcyArICIgcnVuc1Blckl0ZXJhdGlvbiAiICsgcnVuc1Blckl0ZXJhdGlvbik7 CiAgICAgc2V0dXAobWFnbml0dWRlKTsKIAotICAgIHZhciBzdGFydCA9IERhdGUubm93KCk7CiAg ICAgdmFyIGl0ZXJhdGlvbnMgPSAwOwotICAgIHdoaWxlIChEYXRlLm5vdygpIC0gc3RhcnQgPCBt aWxsaXNlY29uZHMpIHsKLSAgICAgICAgLy8gTG9vcCBydW5zUGVySXRlcmF0aW9uIHRpbWVzIHRv IHJlZHVjZSBlcnJvcnMgZHVlIHRvIHRoZSBvdmVyaGVhZCBhbmQgZ3JhbnVsYXJpdHkgb2YgdGhl IERhdGUgb2JqZWN0LgotICAgICAgICBmb3IgKHZhciBpID0gMDsgaSA8IHJ1bnNQZXJJdGVyYXRp b247IGkrKykgewotICAgICAgICAgICAgdGVzdChtYWduaXR1ZGUpOwotICAgICAgICB9Ci0gICAg ICAgIGl0ZXJhdGlvbnMrKzsKKyAgICBpZiAod2luZG93LmNocm9taXVtKSB7CisgICAgICAvLyBG SVhNRTogSWYgdXNpbmcgbWljcm9zZWNvbmRzIHR1cm5zIG91dCB0byBiZSBsZXNzIGZsYWt5LCBl eHBvc2UgbWljcm9zZWNvbmRzCisgICAgICAvLyBmcm9tIEpTQyBvciBsYXlvdXRUZXN0Q29udHJv bGxlciBhbmQgdXNlIHRoZW0uIE90aGVyd2lzZSwgZ2V0IHJpZCBvZiB0aGlzIGJsb2NrLgorICAg ICAgdmFyIG1pY3Jvc2Vjb25kcyA9IG1pbGxpc2Vjb25kcyAqIDEwMDA7CisgICAgICB2YXIgaW50 ZXJ2YWwgPSBuZXcgY2hyb21pdW0uSW50ZXJ2YWwoKTsKKyAgICAgIGludGVydmFsLnN0YXJ0KCk7 CisgICAgICB3aGlsZSAoaW50ZXJ2YWwubWljcm9zZWNvbmRzKCkgPCBtaWNyb3NlY29uZHMpIHsK KyAgICAgICAgICAvLyBMb29wIHJ1bnNQZXJJdGVyYXRpb24gdGltZXMgdG8gcmVkdWNlIGVycm9y cyBkdWUgdG8gdGhlIG92ZXJoZWFkIGFuZCBncmFudWxhcml0eSBvZiB0aGUgRGF0ZSBvYmplY3Qu CisgICAgICAgICAgZm9yICh2YXIgaSA9IDA7IGkgPCBydW5zUGVySXRlcmF0aW9uOyBpKyspIHsK KyAgICAgICAgICAgICAgdGVzdChtYWduaXR1ZGUpOworICAgICAgICAgIH0KKyAgICAgICAgICBp dGVyYXRpb25zKys7CisgICAgICB9CisgICAgICBpbnRlcnZhbC5zdG9wKCk7CisgICAgfSBlbHNl IHsKKyAgICAgIHZhciBzdGFydCA9IERhdGUubm93KCk7CisgICAgICB3aGlsZSAoRGF0ZS5ub3co KSAtIHN0YXJ0IDwgbWlsbGlzZWNvbmRzKSB7CisgICAgICAgICAgLy8gTG9vcCBydW5zUGVySXRl cmF0aW9uIHRpbWVzIHRvIHJlZHVjZSBlcnJvcnMgZHVlIHRvIHRoZSBvdmVyaGVhZCBhbmQgZ3Jh bnVsYXJpdHkgb2YgdGhlIERhdGUgb2JqZWN0LgorICAgICAgICAgIGZvciAodmFyIGkgPSAwOyBp IDwgcnVuc1Blckl0ZXJhdGlvbjsgaSsrKSB7CisgICAgICAgICAgICAgIHRlc3QobWFnbml0dWRl KTsKKyAgICAgICAgICB9CisgICAgICAgICAgaXRlcmF0aW9ucysrOworICAgICAgfQogICAgIH0K ICAgICByZXR1cm4gaXRlcmF0aW9uczsKIH0K