154964 2016-03-03 03:29:42 -0800 Slow JSC stress tests times out in 32 bit debug mode 2016-05-19 05:02:21 -0700 1 1 1 Unclassified WebKit New Bugs Other Unspecified Unspecified RESOLVED FIXED P1 Critical --- 154858 1 ossy ossy ap commit-queue fpizlo ossy rniwa oldest_to_newest 1170177 0 ossy 2016-03-03 03:29:42 -0800 http://trac.webkit.org/changeset/197442 disabled FTL JIT testing on 32 bit bots, because there is no FTL JIT on any 32 bit platforms. But 25-35 tests started to fail on these bots: - https://build.webkit.org/builders/Apple%20El%20Capitan%2032-bit%20JSC%20%28BuildAndTest%29/builds/1658 - https://build.webkit.org/builders/Apple%20Yosemite%2032-bit%20JSC%20%28BuildAndTest%29 The problem is that these tests seems to be very slow in debug mode, but they didn't fail previously because jsc-stress-tests script duplicated the timeout in case of FTL JIT is enabled: if $enableFTL and ENV["JSCTEST_timeout"] # Currently, using the FTL is a performance regression particularly in real # (i.e. non-loopy) benchmarks. Account for this in the timeout. ENV["JSCTEST_timeout"] = (ENV["JSCTEST_timeout"].to_i * 2).to_s end It seems we should duplicate timeout in debug mode too. 1170178 1 ossy 2016-03-03 03:39:24 -0800 My first idea was to replace $enableFtl condition with ($enableFTL or !$ifJSCArgIsntProvidedAreWeReleaseBuild), but unfortunately run-javascriptcore-tests doesn't pass through the --debug option to run-jsc-stress-tests. I don't have time to investigate this bug, feel free to pick it up. But maybe the simplest fix is to set JSCTEST_timeout environment on these bots to a high enough number. 1170222 2 ap 2016-03-03 09:23:26 -0800 Will fix by rolling out r197442. There is no reason to keep such a regression. 1170228 3 ap 2016-03-03 09:26:55 -0800 *** This bug has been marked as a duplicate of bug 154967 *** 1170237 4 ossy 2016-03-03 09:31:30 -0800 Rolling out r197442 only hides this problem. The tests are still slow, relying on increased timeout set by --ftl-jit option is incorrect and should be fixed. 1171283 5 ap 2016-03-05 15:21:33 -0800 This bug used to be a duplicate, because it said that the tests failed on bots. That is fixed via bug 154967. You changed the title after the fact. 1171437 6 ossy 2016-03-06 11:34:27 -0800 (In reply to comment #5) > This bug used to be a duplicate, because it said that the tests failed on > bots. That is fixed via bug 154967. > > You changed the title after the fact. I agree, the bug mentioned in the original title of the bug was "solved" (papered over) by the rollout. But The description of this bug is clear and the bug is still valid and should be fixed in the near future. The original title was confusing, I just changed it to match with the desription. It's not reasonable to rely on running useless FTL JIT tests (+50% runtime) just to increase the JSCTEST_timeout on 32 bit debug JSC tester bots. We should increase the timeout in debug mode in this bug and then we can disable FTL JIT testing on all 32 bit bots in bug154858. 1191523 7 278495 ossy 2016-05-10 08:38:16 -0700 Created attachment 278495 Patch 1193819 8 278495 mcatanzaro 2016-05-17 08:36:32 -0700 Comment on attachment 278495 Patch (Please commit when you're around to watch the bots.) 1194723 9 278495 commit-queue 2016-05-19 05:02:16 -0700 Comment on attachment 278495 Patch Clearing flags on attachment: 278495 Committed r201158: <http://trac.webkit.org/changeset/201158> 1194724 10 commit-queue 2016-05-19 05:02:21 -0700 All reviewed patches have been landed. Closing bug. 278495 2016-05-10 08:38:16 -0700 2016-05-19 05:02:16 -0700 Patch bug-154964-20160510083903.patch text/plain 2112 ossy U3VidmVyc2lvbiBSZXZpc2lvbjogMjAwNjIyCmRpZmYgLS1naXQgYS9Ub29scy9DaGFuZ2VMb2cg Yi9Ub29scy9DaGFuZ2VMb2cKaW5kZXggYjI2NzA3M2E0ZWNiODQwZTFhZDgzOTk3OWQzMjFiMmMz YWI5MGU3ZC4uNTgyYTY5YzFkYjc4YmU0ZmYzNTM0MjJjNzE0ZTNiZjVhNzY1MWVmMiAxMDA2NDQK LS0tIGEvVG9vbHMvQ2hhbmdlTG9nCisrKyBiL1Rvb2xzL0NoYW5nZUxvZwpAQCAtMSwzICsxLDE0 IEBACisyMDE2LTA1LTEwICBDc2FiYSBPc3p0cm9nb27DoWMgIDxvc3N5QHdlYmtpdC5vcmc+CisK KyAgICAgICAgU2xvdyBKU0Mgc3RyZXNzIHRlc3RzIHRpbWVzIG91dCBpbiAzMiBiaXQgZGVidWcg bW9kZQorICAgICAgICBodHRwczovL2J1Z3Mud2Via2l0Lm9yZy9zaG93X2J1Zy5jZ2k/aWQ9MTU0 OTY0CisKKyAgICAgICAgUmV2aWV3ZWQgYnkgTk9CT0RZIChPT1BTISkuCisKKyAgICAgICAgKiBT Y3JpcHRzL3J1bi1qYXZhc2NyaXB0Y29yZS10ZXN0czoKKyAgICAgICAgKHJ1bkpTQ1N0cmVzc1Rl c3RzKTogUGFzcyB0aHJvdWdoIHRoZSAtLWRlYnVnIG9wdGlvbi4KKyAgICAgICAgKiBTY3JpcHRz L3J1bi1qc2Mtc3RyZXNzLXRlc3RzOiBJbmNyZWFzZSB0aGUgdGltZW91dCBmb3IgZGVidWcgYnVp bGRzLgorCiAyMDE2LTA1LTEwICBNaWNoYWVsIENhdGFuemFybyAgPG1jYXRhbnphcm9AaWdhbGlh LmNvbT4KIAogICAgICAgICBbTGludXhdIFJlbW92ZSBzZWNjb21wIGZpbHRlcnMgc3VwcG9ydApk aWZmIC0tZ2l0IGEvVG9vbHMvU2NyaXB0cy9ydW4tamF2YXNjcmlwdGNvcmUtdGVzdHMgYi9Ub29s cy9TY3JpcHRzL3J1bi1qYXZhc2NyaXB0Y29yZS10ZXN0cwppbmRleCBiMDcwZGJhMjNhOTMyODQ2 MGUxZmNlZjMxZmFlYjJmYzA0OWI0MzM5Li43NDhjOTE1MDc5MDE2ZGY0NzM5ZTkxMmYwNGU5NzQx MzVkMDBkMjg1IDEwMDc1NQotLS0gYS9Ub29scy9TY3JpcHRzL3J1bi1qYXZhc2NyaXB0Y29yZS10 ZXN0cworKysgYi9Ub29scy9TY3JpcHRzL3J1bi1qYXZhc2NyaXB0Y29yZS10ZXN0cwpAQCAtMjU3 LDYgKzI1NywxMCBAQCBzdWIgcnVuSlNDU3RyZXNzVGVzdHMKICAgICAgICAgc2hpZnQgQGpzY1N0 cmVzc0RyaXZlckNtZDsgIyBSZW1vdmUgL3Vzci9iaW4vZW52CiAgICAgfQogCisgICAgaWYgKGNv bmZpZ3VyYXRpb24oKSBlcSAiRGVidWciKSB7CisgICAgICAgIHB1c2goQGpzY1N0cmVzc0RyaXZl ckNtZCwgIi0tZGVidWciKTsKKyAgICB9CisKICAgICBpZiAoJGVudlZhcnMgbmUgIiIpIHsKICAg ICAgICAgICAgIHB1c2goQGpzY1N0cmVzc0RyaXZlckNtZCwgIi0tZW52LXZhcnMiKTsKICAgICAg ICAgICAgIHB1c2goQGpzY1N0cmVzc0RyaXZlckNtZCwgJGVudlZhcnMpOwpkaWZmIC0tZ2l0IGEv VG9vbHMvU2NyaXB0cy9ydW4tanNjLXN0cmVzcy10ZXN0cyBiL1Rvb2xzL1NjcmlwdHMvcnVuLWpz Yy1zdHJlc3MtdGVzdHMKaW5kZXggOTU2NjNlY2M2NmE4OWFlMjA4N2YzNDE5OTgxYTUzNjdkNGFl NGFmOC4uOTVkMzkzNzM2ZTY1YWE2MjZiMDA4YWVjMmE2ZmQ0MjNkZTlmZmYxYiAxMDA3NTUKLS0t IGEvVG9vbHMvU2NyaXB0cy9ydW4tanNjLXN0cmVzcy10ZXN0cworKysgYi9Ub29scy9TY3JpcHRz L3J1bi1qc2Mtc3RyZXNzLXRlc3RzCkBAIC0xOTQ2LDkgKzE5NDYsMTAgQEAgaWYgISRudW1DaGls ZFByb2Nlc3NlcwogICAgIGVuZAogZW5kCiAKLWlmICRlbmFibGVGVEwgYW5kIEVOVlsiSlNDVEVT VF90aW1lb3V0Il0KK2lmICRlbmFibGVGVEwgYW5kIEVOVlsiSlNDVEVTVF90aW1lb3V0Il0gb3Ig IWlmSlNDQXJnSXNudFByb3ZpZGVkQXJlV2VSZWxlYXNlQnVpbGQKICAgICAjIEN1cnJlbnRseSwg dXNpbmcgdGhlIEZUTCBpcyBhIHBlcmZvcm1hbmNlIHJlZ3Jlc3Npb24gcGFydGljdWxhcmx5IGlu IHJlYWwKICAgICAjIChpLmUuIG5vbi1sb29weSkgYmVuY2htYXJrcy4gQWNjb3VudCBmb3IgdGhp cyBpbiB0aGUgdGltZW91dC4KKyAgICAjIEluY3JlYXNlIHRoZSB0aW1lb3V0IGZvciBkZWJ1ZyBi dWlsZHMgdG9vLiAoLS1kZWJ1ZyBjb21tYW5kIGxpbmUgb3B0aW9uKQogICAgIEVOVlsiSlNDVEVT VF90aW1lb3V0Il0gPSAoRU5WWyJKU0NURVNUX3RpbWVvdXQiXS50b19pICogMikudG9fcwogZW5k CiAK