189063 2018-08-28 15:32:44 -0700 REGRESSION(r235408): GTK bots exiting early 2018-09-05 00:09:42 -0700 1 1 1 Unclassified WebKit WebKitGTK Other PC Linux RESOLVED FIXED P2 Normal --- 1 mcatanzaro simon.fraser aboya bugs-noreply commit-queue csaavedra keith_miller mcatanzaro msaboff simon.fraser oldest_to_newest 1454585 0 mcatanzaro 2018-08-28 15:32:44 -0700 The GTK bots are exiting early. I see many timeouts in certain animation and compositing tests, CSS3 tests, and editing tests: https://build.webkit.org/results/GTK%20Linux%2064-bit%20Release%20(Tests)/r235437%20(7940)/results.html The regression was introduced between r235407-r235410, inclusive. Most likely r235408 or r235409. 1454588 1 mcatanzaro 2018-08-28 15:35:31 -0700 There are also ten failing webkitperl tests, which is unusual as these never fail. These may or may not be related, I don't know: Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl .................................. Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl .................................. 1/13 # Failed test 'test262 test failed, ignore expectations (exit code: 1)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '1' # Failed test 'test262 test failed, ignore expectations (new failures: 2)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 142. # got: '0' # expected: '2' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'test262 test passed, ignore expectations (exit code: 0)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '0' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'test262 tests newly failed (exit code: 1)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '1' # Failed test 'test262 tests newly failed (new failures: 2)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 142. # got: '0' # expected: '2' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'test262 tests newly passed (exit code: 0)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '0' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'test262 tests fails, expected failure (exit code: 0)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '0' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'test262 tests fails, with unexpected error string (exit code: 1)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 136. # got: '255' # expected: '1' # Failed test 'test262 tests fails, with unexpected error string (new failures: 2)' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 142. # got: '0' # expected: '2' Experimental keys on scalar is now forbidden at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262/Runner.pm line 929. Compilation failed in require at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. BEGIN failed--compilation aborted at /home/buildbot/wpe/wpe-linux-64-release-tests/build/Tools/Scripts/test262-runner line 45. # Failed test 'expectations yaml file format' # at Tools/Scripts/webkitperl/test262_unittest/test262-runner-tests.pl line 157. # Looks like you failed 10 tests of 13. 1454589 2 mcatanzaro 2018-08-28 15:35:56 -0700 (In reply to Michael Catanzaro from comment #1) > There are also ten failing webkitperl tests, which is unusual as these never > fail. These may or may not be related, I don't know: Well, they started failing in the same range of commits. 1454596 3 simon.fraser 2018-08-28 15:46:53 -0700 You're sure it wasn't https://trac.webkit.org/changeset/235399/webkit ? 1454600 4 simon.fraser 2018-08-28 15:52:56 -0700 The perl tests started failing here: https://trac.webkit.org/changeset/235359/webkit 1454612 5 mcatanzaro 2018-08-28 16:10:39 -0700 (In reply to Simon Fraser (smfr) from comment #3) > You're sure it wasn't https://trac.webkit.org/changeset/235399/webkit ? Hm, that commit is mighty suspicious, but we do have successful test runs at r235402 and r235406, on two different bots. So I don't think so, not unless old versions of the test runner are somehow being used by mistake. (In reply to Simon Fraser (smfr) from comment #4) > The perl tests started failing here: > https://trac.webkit.org/changeset/235359/webkit Ooops, I misread the test results page. So that is surely unrelated. (Do you want to report the bug for it, or shall I?) 1454682 6 mcatanzaro 2018-08-28 20:03:43 -0700 (In reply to Michael Catanzaro from comment #5) > (In reply to Simon Fraser (smfr) from comment #3) > > You're sure it wasn't https://trac.webkit.org/changeset/235399/webkit ? > > Hm, that commit is mighty suspicious, but we do have successful test runs at > r235402 and r235406, on two different bots. So I don't think so, not unless > old versions of the test runner are somehow being used by mistake. I've tested r235406 locally and verified it's good. Tested r235408 and verified it's bad. The only change in r235407 is guarded by PLATFORM(COCOA). So we can pin this on r235408. The tests are printing on stderr: FAIL: TestControllerRunLoop timed out. Looking at the commit, I don't immediately notice anything problematic, so I'll need to attempt to debug why some particular test is stalling. I did spot one unrelated bug in the commit, in Tools/DumpRenderTree/win/DumpRenderTree.cpp: // DumpRenderTree does not support checking for abandonded documents. if (!strcmp("#CHECK FOR ABANDONED DOCUMENTS", command)) { I noticed that "abandonded" was misspelled. Then I noticed that the command here is different than everywhere else. Should probably be: // DumpRenderTree does not support checking for world leaks. if (!strcmp("#CHECK FOR WORLD LEAKS", command)) { 1454771 7 mcatanzaro 2018-08-29 08:02:16 -0700 *** Bug 189079 has been marked as a duplicate of this bug. *** 1454897 8 simon.fraser 2018-08-29 13:00:25 -0700 Happy to help out if we can be on irc at the same time. 1455815 9 mcatanzaro 2018-08-31 12:53:31 -0700 (In reply to Simon Fraser (smfr) from comment #8) > Happy to help out if we can be on irc at the same time. Thanks... it's been a busy week and next week will be too, so probably not anytime soon unfortunately. > I did spot one unrelated bug in the commit, in > Tools/DumpRenderTree/win/DumpRenderTree.cpp: > > // DumpRenderTree does not support checking for abandonded documents. > if (!strcmp("#CHECK FOR ABANDONED DOCUMENTS", command)) { > > I noticed that "abandonded" was misspelled. Then I noticed that the command > here is different than everywhere else. Should probably be: > > // DumpRenderTree does not support checking for world leaks. > if (!strcmp("#CHECK FOR WORLD LEAKS", command)) { Don't forget about this. ^ 1455819 10 simon.fraser 2018-08-31 13:01:39 -0700 (In reply to Michael Catanzaro from comment #9) > (In reply to Simon Fraser (smfr) from comment #8) > > Happy to help out if we can be on irc at the same time. > > Thanks... it's been a busy week and next week will be too, so probably not > anytime soon unfortunately. > > > I did spot one unrelated bug in the commit, in > > Tools/DumpRenderTree/win/DumpRenderTree.cpp: > > > > // DumpRenderTree does not support checking for abandonded documents. > > if (!strcmp("#CHECK FOR ABANDONED DOCUMENTS", command)) { > > > > I noticed that "abandonded" was misspelled. Then I noticed that the command > > here is different than everywhere else. Should probably be: > > > > // DumpRenderTree does not support checking for world leaks. > > if (!strcmp("#CHECK FOR WORLD LEAKS", command)) { > > Don't forget about this. ^ Fixed in r235408. 1456299 11 csaavedra 2018-09-04 06:30:03 -0700 I debugged this a bit today, have some findings that might shed some light. - This only happens if you run a series of tests sharing the DumpRendererTree. If you use --run-singly, the tests pass. So it probably has to be with the remainder state of DumpRendererTree. - I noticed that there's a 20_s in the AsyncTask that is added for the world leaks in the offending commit. I'm not familiar with this but I'm guessing that this is the timeout for the task. The timeout for the tests is 5 s, AFAIR, so it's probable that the world leaks task is timing out. I increased the tests timeout when running them locally to 20s, and then the tests pass. Similarly, if I change the 20_s to 5_s, the tests also pass. - But what I don't understand is why this task is being called at all? Isn't it so that --world-leaks needs to be called explicitly for the check to run? What I see is that this is done unconditionally in willDestroyWebView(). Is this intended? 1456301 12 csaavedra 2018-09-04 06:47:33 -0700 (In reply to Claudio Saavedra from comment #11) > What I see is that this is done unconditionally in willDestroyWebView(). Is > this intended? Apologies in advance if this is gross, but just not doing this unconditional call also allows the tests to pass without timing out. 1456313 13 simon.fraser 2018-09-04 08:41:12 -0700 (In reply to Claudio Saavedra from comment #12) > (In reply to Claudio Saavedra from comment #11) > > > What I see is that this is done unconditionally in willDestroyWebView(). Is > > this intended? > > Apologies in advance if this is gross, but just not doing this unconditional > call also allows the tests to pass without timing out. Oh, that call in DestroyWebView should not be unconditional! 1456469 14 348851 simon.fraser 2018-09-04 15:03:39 -0700 Created attachment 348851 Patch 1456522 15 348851 commit-queue 2018-09-04 16:45:13 -0700 Comment on attachment 348851 Patch Clearing flags on attachment: 348851 Committed r235646: <https://trac.webkit.org/changeset/235646> 1456523 16 commit-queue 2018-09-04 16:45:15 -0700 All reviewed patches have been landed. Closing bug. 1456656 17 csaavedra 2018-09-05 00:09:42 -0700 Glad to have helped! 348851 2018-09-04 15:03:39 -0700 2018-09-04 16:45:13 -0700 Patch bug-189063-20180904150339.patch text/plain 3822 simon.fraser U3VidmVyc2lvbiBSZXZpc2lvbjogMjM1NjMyCmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggNzhjYjIxMjg1YjE1NWVlMDUwNzVhNjBlY2Q4NDgyNDNi MTA0MjEyOC4uOWQwNTQ4NTU5MGQ4Nzg3Y2U0MjU1OTI3MDQ4OGY1MTNlZjFkOGUyOSAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDI2 IEBACisyMDE4LTA5LTA0ICBTaW1vbiBGcmFzZXIgIDxzaW1vbi5mcmFzZXJAYXBwbGUuY29tPgor CisgICAgICAgIFJFR1JFU1NJT04ocjIzNTQwOCk6IEdUSyBib3RzIGV4aXRpbmcgZWFybHkKKyAg ICAgICAgaHR0cHM6Ly9idWdzLndlYmtpdC5vcmcvc2hvd19idWcuY2dpP2lkPTE4OTA2MworCisg ICAgICAgIFJldmlld2VkIGJ5IE5PQk9EWSAoT09QUyEpLgorICAgICAgICAKKyAgICAgICAgV2Vi S2l0VGVzdFJ1bm5lciB3YXMgcnVubmluZyB0aGUgd29ybGQgbGVhayBjaGVja3MgZXZlbiB3aGVu IHJ1biB3aXRob3V0IC0td29ybGQtbGVha3MsCisgICAgICAgIGNhdXNpbmcgR1RLIGJvdCB0aW1l b3V0cy4gU28gZ3VhcmQgdXBkYXRlTGl2ZURvY3VtZW50c0FmdGVyVGVzdCgpLCBjaGVja0Zvcldv cmxkTGVha3MoKQorICAgICAgICBhbmQgZmluZEFuZER1bXBXb3JsZExlYWtzKCkgd2l0aCBtX2No ZWNrRm9yV29ybGRMZWFrcyBjaGVja3MsIGFuZCBpbgorICAgICAgICBUZXN0Q29udHJvbGxlcjo6 aGFuZGxlQ29udHJvbENvbW1hbmQoKSBwcmludCBhIG1lc3NhZ2UgaWYgdGhlIGNvbnRyb2wgc2Ny aXB0IHNlbmRzIHRoZQorICAgICAgICAiI0NIRUNLIEZPUiBXT1JMRCBMRUFLUyIgY29tbWFuZCBp ZiBXVFIgd2FzIG5vdCBydW4gd2l0aCAtLXdvcmxkLWxlYWtzLgorICAgICAgICAKKyAgICAgICAg SSB0ZXN0ZWQgdGhhdCBydW5uaW5nIHdpdGggLS13b3JsZC1sZWFrcyBzdGlsbCB3b3Jrcy4KKwor ICAgICAgICAqIFdlYktpdFRlc3RSdW5uZXIvVGVzdENvbnRyb2xsZXIuY3BwOgorICAgICAgICAo V1RSOjpUZXN0Q29udHJvbGxlcjo6cmVzZXRTdGF0ZVRvQ29uc2lzdGVudFZhbHVlcyk6CisgICAg ICAgIChXVFI6OlRlc3RDb250cm9sbGVyOjp1cGRhdGVMaXZlRG9jdW1lbnRzQWZ0ZXJUZXN0KToK KyAgICAgICAgKFdUUjo6VGVzdENvbnRyb2xsZXI6OmNoZWNrRm9yV29ybGRMZWFrcyk6CisgICAg ICAgIChXVFI6OlRlc3RDb250cm9sbGVyOjpmaW5kQW5kRHVtcFdvcmxkTGVha3MpOgorICAgICAg ICAoV1RSOjpUZXN0Q29udHJvbGxlcjo6aGFuZGxlQ29udHJvbENvbW1hbmQpOgorICAgICAgICAo V1RSOjpUZXN0Q29udHJvbGxlcjo6cnVuKToKKwogMjAxOC0wOS0wNCAgQ2hyaXMgRHVtZXogIDxj ZHVtZXpAYXBwbGUuY29tPgogCiAgICAgICAgIEFkZCBwcm9jZXNzIHBvb2wgY29uZmlndXJhdGlv biBmbGFnIHRvIHR1cm4gb24gYXV0b21hdGljIHByb2Nlc3MgcHJlLXdhcm1pbmcKZGlmZiAtLWdp dCBhL1Rvb2xzL1dlYktpdFRlc3RSdW5uZXIvVGVzdENvbnRyb2xsZXIuY3BwIGIvVG9vbHMvV2Vi S2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKaW5kZXggN2NkYWFmMGE4NjlhYmFjMmJh NjZkNTdmYmZlZWEyZDczMjc5MWE0Zi4uY2ZmOGQ2YTQ4ZDA2MzRhNTc3NTMzOWQ4NjFiZTEyZmNi ZTA3ODQ0NCAxMDA2NDQKLS0tIGEvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxl ci5jcHAKKysrIGIvVG9vbHMvV2ViS2l0VGVzdFJ1bm5lci9UZXN0Q29udHJvbGxlci5jcHAKQEAg LTkzMyw3ICs5MzMsNyBAQCBib29sIFRlc3RDb250cm9sbGVyOjpyZXNldFN0YXRlVG9Db25zaXN0 ZW50VmFsdWVzKGNvbnN0IFRlc3RPcHRpb25zJiBvcHRpb25zLCBSZQogICAgIGlmICghbV9kb25l UmVzZXR0aW5nKQogICAgICAgICByZXR1cm4gZmFsc2U7CiAgICAgCi0gICAgaWYgKHJlc2V0U3Rh Z2UgPT0gUmVzZXRTdGFnZTo6QWZ0ZXJUZXN0ICYmIG1fY2hlY2tGb3JXb3JsZExlYWtzKQorICAg IGlmIChyZXNldFN0YWdlID09IFJlc2V0U3RhZ2U6OkFmdGVyVGVzdCkKICAgICAgICAgdXBkYXRl TGl2ZURvY3VtZW50c0FmdGVyVGVzdCgpOwogCiAgICAgcmV0dXJuIG1fZG9uZVJlc2V0dGluZzsK QEAgLTk0MSw2ICs5NDEsOSBAQCBib29sIFRlc3RDb250cm9sbGVyOjpyZXNldFN0YXRlVG9Db25z aXN0ZW50VmFsdWVzKGNvbnN0IFRlc3RPcHRpb25zJiBvcHRpb25zLCBSZQogCiB2b2lkIFRlc3RD b250cm9sbGVyOjp1cGRhdGVMaXZlRG9jdW1lbnRzQWZ0ZXJUZXN0KCkKIHsKKyAgICBpZiAoIW1f Y2hlY2tGb3JXb3JsZExlYWtzKQorICAgICAgICByZXR1cm47CisKICAgICBBc3luY1Rhc2soW10o KSB7CiAgICAgICAgIC8vIEFmdGVyIGVhY2ggdGVzdCwgd2UgdXBkYXRlIHRoZSBsaXN0IG9mIGxp dmUgZG9jdW1lbnRzIHNvIHRoYXQgd2UgY2FuIGRldGVjdCB3aGVuIGFuIGFiYW5kb25lZCBkb2N1 bWVudCBmaXJzdCBzaG93ZWQgdXAuCiAgICAgICAgIFdLUmV0YWluUHRyPFdLU3RyaW5nUmVmPiBt ZXNzYWdlTmFtZSA9IGFkb3B0V0soV0tTdHJpbmdDcmVhdGVXaXRoVVRGOENTdHJpbmcoIkdldExp dmVEb2N1bWVudHMiKSk7CkBAIC05NTAsNyArOTUzLDcgQEAgdm9pZCBUZXN0Q29udHJvbGxlcjo6 dXBkYXRlTGl2ZURvY3VtZW50c0FmdGVyVGVzdCgpCiAKIHZvaWQgVGVzdENvbnRyb2xsZXI6OmNo ZWNrRm9yV29ybGRMZWFrcygpCiB7Ci0gICAgaWYgKCFUZXN0Q29udHJvbGxlcjo6c2luZ2xldG9u KCkubWFpbldlYlZpZXcoKSkKKyAgICBpZiAoIW1fY2hlY2tGb3JXb3JsZExlYWtzIHx8ICFUZXN0 Q29udHJvbGxlcjo6c2luZ2xldG9uKCkubWFpbldlYlZpZXcoKSkKICAgICAgICAgcmV0dXJuOwog CiAgICAgQXN5bmNUYXNrKFtdKCkgewpAQCAtOTYyLDYgKzk2NSw5IEBAIHZvaWQgVGVzdENvbnRy b2xsZXI6OmNoZWNrRm9yV29ybGRMZWFrcygpCiAKIHZvaWQgVGVzdENvbnRyb2xsZXI6OmZpbmRB bmREdW1wV29ybGRMZWFrcygpCiB7CisgICAgaWYgKCFtX2NoZWNrRm9yV29ybGRMZWFrcykKKyAg ICAgICAgcmV0dXJuOworCiAgICAgY2hlY2tGb3JXb3JsZExlYWtzKCk7CiAKICAgICBTdHJpbmdC dWlsZGVyIGJ1aWxkZXI7CkBAIC0xMzkwLDYgKzEzOTYsMTAgQEAgYm9vbCBUZXN0Q29udHJvbGxl cjo6d2FpdEZvckNvbXBsZXRpb24oY29uc3QgV1RGOjpGdW5jdGlvbjx2b2lkICgpPiYgZnVuY3Rp b24sIFcKIGJvb2wgVGVzdENvbnRyb2xsZXI6OmhhbmRsZUNvbnRyb2xDb21tYW5kKGNvbnN0IGNo YXIqIGNvbW1hbmQpCiB7CiAgICAgaWYgKCFzdHJjbXAoIiNDSEVDSyBGT1IgV09STEQgTEVBS1Mi LCBjb21tYW5kKSkgeworICAgICAgICBpZiAoIW1fY2hlY2tGb3JXb3JsZExlYWtzKSB7CisgICAg ICAgICAgICBXVEZMb2dBbHdheXMoIldlYktpdFRlc3RSdW5uZXIgYXNrZWQgdG8gY2hlY2sgZm9y IHdvcmxkIGxlYWtzLCBidXQgd2FzIG5vdCBydW4gd2l0aCAtLXdvcmxkLWxlYWtzIik7CisgICAg ICAgICAgICByZXR1cm4gdHJ1ZTsKKyAgICAgICAgfQogICAgICAgICBmaW5kQW5kRHVtcFdvcmxk TGVha3MoKTsKICAgICAgICAgcmV0dXJuIHRydWU7CiAgICAgfQpAQCAtMTQyNCw3ICsxNDM0LDgg QEAgdm9pZCBUZXN0Q29udHJvbGxlcjo6cnVuKCkKICAgICAgICAgICAgIGlmICghcnVuVGVzdCht X3BhdGhzW2ldLmNfc3RyKCkpKQogICAgICAgICAgICAgICAgIGJyZWFrOwogICAgICAgICB9Ci0g ICAgICAgIGZpbmRBbmREdW1wV29ybGRMZWFrcygpOworICAgICAgICBpZiAobV9jaGVja0Zvcldv cmxkTGVha3MpCisgICAgICAgICAgICBmaW5kQW5kRHVtcFdvcmxkTGVha3MoKTsKICAgICB9CiB9 CiAK