Open Bug 1746524 Opened 2 years ago Updated 4 months ago

Enable browser.tabs.documentchannel.parent-controlled pref

Categories

(Core :: DOM: Navigation, task, P1)

task

Tracking

()

People

(Reporter: annyG, Assigned: farre, NeedInfo)

References

(Depends on 3 open bugs, Blocks 1 open bug, Regressed 1 open bug)

Details

(Keywords: perf-alert)

Attachments

(1 file, 1 obsolete file)

Unfortunately, work in bug 1721217 expanded because of a few test failures that came up recently. To make this work more manageable, It makes more sense to land patches in that bug without this pref enabled, and instead track the required work for doing that here.

Blocks: 1743636
No longer blocks: 1743636
Depends on: 1743636
Blocks: 1647550
Type: defect → task
Priority: -- → P1

Anny, should this P1 bug be assigned to you? Or will someone else finish the work to enable the browser.tabs.documentchannel.parent-controlled pref?

Severity: -- → N/A
Flags: needinfo?(agakhokidze)
Flags: needinfo?(agakhokidze)
Depends on: 1742886
Depends on: 1748327
Depends on: 1748364
See Also: → 1725716

Just pasting a comment from here, when we enable browser.tabs.documentchannel.parent-controlled and SHIP by default, we will not need to check for isNavigating here anymore.

This tiny task is probably blocked on removal of support for non-SHIP and non-Fission configurations.

Assignee: nobody → continuation
Blocks: 1743636
No longer depends on: 1743636
Depends on: 1765652
Depends on: 1766305
No longer blocks: 1743636
Depends on: 1743636
Depends on: 1776080

I still want to wait for bug 1766305 to merge to m-c, but things seem reasonably green so far on Linux, and it doesn't look like it is breaking anything on Android.

One dodgy thing remaining is that download tests need to be written in a special way, but I'm not sure there's any way around it. I guess I'll write an email to Firefox-dev about that.

Pushed by amccreight@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/5dda3f9ee3d5
Enable browser.tabs.documentchannel.parent-controlled pref. r=nika

Backed out for causing multiple failures.

Backout link
Push with failures

Link to bc failure log
Link to X5 failure log
Failure line :
TEST-UNEXPECTED-FAIL | browser/components/extensions/test/browser/browser_ext_tabs_discard_reversed.js
TEST-UNEXPECTED-FAIL | xpcshell-remote.ini:toolkit/components/extensions/test/xpcshell/test_ext_userScripts_register.js

Flags: needinfo?(continuation)
Flags: needinfo?(continuation)

I managed to reproduce the X5 failure on my try push. It is an intermittent failure. It seems to fail about half of the time across 9 tries on inbound and 3 out of 8 times on try. I found evidence of a few similar looking intermittent failures in bugzilla, but if it occurs without parent controlled navigation, the failure rate is much lower, because it happened 0 times in a dozen retries.

The BC failure happened 3 out of 3 times on try. It seems odd that it is opt only. Maybe there's a race condition.

Depends on: 1780488

I've fixed the XPCShell failure in bug 1780488. I haven't managed to reproduce the BC failure (browser_ext_tabs_discard_reversed.js), though it does look like an existing issue. This test has already been disabled on OSX, Windows 10 Fission debug, and Linux 64 debug, so I think it is a very flaky test. Maybe I just got unlucky with my landing, though the prior analysis of this test does sound like something this patch could cause to fail more.

With my patch, I tried retriggering the configuration that hit the BC failure in comment 7, and didn't hit the issue after 12 retries. I also re-enabled the test on Linux debug (the debug failure rate was 4 or so times higher before it was disabled) and retriggered it on try 10 times and did not hit the issue.

I think I'm going to try relanding it, after confirming there are no new regressions with my patch. If browser_ext_tabs_discard_reversed.js starts to fail too frequently I'll try talking to the WebExtensions people about disabling the title checking part of the test, as it sounds like it is somewhat racy.

I also tried running browser_ext_tabs_discard_reversed.js locally on OSX debug with --verify and my patch, four or five times, and was unable to reproduce any failure.

The try run looks okay-ish so I'll try relanding this.

Pushed by amccreight@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/44255b7d9b1c
Enable browser.tabs.documentchannel.parent-controlled pref. r=nika
Status: NEW → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → 105 Branch

Backed out for causing xpcshell failures on test_ext_userScripts_register.js (requested by dev)

Backout link

Push with failures

Failure log

Flags: needinfo?(continuation)
Status: RESOLVED → REOPENED
Resolution: FIXED → ---

The test only seems to fail with Fission disabled (including on Android), so I'm going to try disabling it under those conditions in 1763197

Depends on: 1763197
Flags: needinfo?(continuation)
Pushed by amccreight@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/24d6d6816f34
Enable browser.tabs.documentchannel.parent-controlled pref. r=nika
Status: REOPENED → RESOLVED
Closed: 2 years ago2 years ago
Resolution: --- → FIXED
Regressions: 1785290
Status: RESOLVED → REOPENED
Flags: needinfo?(continuation)
Resolution: FIXED → ---
Target Milestone: 105 Branch → ---

Thanks.

Flags: needinfo?(continuation)

(In reply to Norisz Fay [:noriszfay] from comment #14)

Backed out for causing xpcshell failures on test_ext_userScripts_register.js (requested by dev)

Backout link

Push with failures

Failure log

== Change summary for alert #35188 (as of Mon, 22 Aug 2022 08:30:10 GMT) ==

Improvements:

Ratio Test Platform Options Absolute values (old vs new)
10% bing FirstVisualChange android-hw-a51-11-0-aarch64-shippable-qr warm webrender 374.08 -> 336.42

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=35188

Depends on: 1814293

I did a fresh Linux debug try push to see if any new tests have failed. I filed a bug 1814293 for one that seems to be permafail. Nothing else seems to be permafail, though I am suspicious of the unexpected pass in /navigation-timing/secure-connection-start-non-zero.https.html that happened 2 out of 4 times on my try push, after only happening twice in the last 30 days otherwise on Treeherder.

I'm also suspicious of it because of the "navigation timing" in the path sounds like the sort of thing that a change in navigation could affect.

Depends on: 1814311

I did another Linux debug try run. WPT2, dt8, M7(http3) have a high failure rate, but they are having a very high failure rate on trunk, so I don't think that's a regression. M-a11y-checks dt6 has a high failure rate, but we don't seem to run that at all on m-c or autoland so I'll assume that's not a regression either. The other failures didn't show up on another couple of retriggers.

Yet another Linux debug try run, and it looks like the high failure rates have gone. There are some other failures, have the failing tests moved to other chunks. Doesn't look like it at first glance.

Does this mean that we should try to land again after soft freeze is over?

Flags: needinfo?(continuation)

(In reply to Andreas Farre [:farre] from comment #25)

Does this mean that we should try to land again after soft freeze is over?

The issue that got it backed out initially was a bunch of performance regressions (bug 1785290). I investigated that for awhile but didn't solve the issue. It was a bit odd because the tests that regressed were initiating the loads in the parent (loads were basically started like you were pasting the URL in the URL bar and hitting enter).

Flags: needinfo?(continuation)

The perf problems seems to still be present. The revisions used where base and new. I've tried looking at profiles a bit, but I'm not really seeing anything.

When I looked at it, I think the "new" kind of looked like the old one, shifted over by about 5ms. Nothing was present in the profiler for that little early bit that was taking 5ms more time. I don't know if the profiler wasn't looking at the right thread or the sample rate wasn't high enough or what.

Assignee: continuation → afarre

This patch also does some cosmetic changes for when parent controlled
loads should be used, but more importantly introduces the pref
'browser.tabs.documentchannel.speculative-load' with the default value
false.

It turns out that speculatively loading in the parent when a parent
controlled load isn't possible is the cause for performance
regressions.

The perf results look good, I'm trying to land this!

Pushed by afarre@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/e210c2c7d93a
Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug
Attachment #9361455 - Attachment description: Bug 1746524 - Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug! → Bug 1746524 - Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug
Pushed by afarre@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/149e69327bbb
Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug

Backed out for causing gv-junit failures on bypassHTTPSOnlyError

[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.298 D/GeckoViewProgress( 3452): StateTracker onStateChange: isTopLevel=true, flags=0xc0010, status=NS_OK loadType=8388609
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 D/GeckoSession( 3452): handleMessage GeckoView:PageStop uri=null
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): Error
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): java.lang.AssertionError: onLoadRequest should be called specified number of times
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): Expected: <1>
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452):      but: was <0>
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.Assert.assertThat(Assert.java:964)
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector$1.call(ErrorCollector.java:77)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector.checkSucceeds(ErrorCollector.java:90)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector.checkThat(ErrorCollector.java:75)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.checkThat(GeckoSessionTestRule.java:929)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.assertMatchesCount(GeckoSessionTestRule.java:965)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.-$$Nest$massertMatchesCount(GeckoSessionTestRule.java)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$CallbackDelegates.clearAndAssert(GeckoSessionTestRule.java:724)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.afterWait(GeckoSessionTestRule.java:1875)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.waitForJS(GeckoSessionTestRule.java:2646)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.BaseSessionTest.waitForJS(BaseSessionTest.kt:246)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.NavigationDelegateTest.bypassHTTPSOnlyError(NavigationDelegateTest.kt:676)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.lang.reflect.Method.invoke(Native Method)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:59)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:56)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.Verifier$1.evaluate(Verifier.java:35)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2.lambda$evaluate$0(GeckoSessionTestRule.java:1509)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2.$r8$lambda$eGWMAiU-In-hdc_ISYHPMUhHPKA(GeckoSessionTestRule.java)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2$$ExternalSyntheticLambda1.run(D8$$SyntheticClass)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:428)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.util.concurrent.FutureTask.run(FutureTask.java:237)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.app.Instrumentation$SyncRunnable.run(Instrumentation.java:1950)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Handler.handleCallback(Handler.java:751)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Handler.dispatchMessage(Handler.java:95)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Looper.loop(Looper.java:154)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.app.ActivityThread.main(ActivityThread.java:6077)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.lang.reflect.Method.invoke(Native Method)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:866)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:756)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.301 D/GeckoViewContentDelegateParent( 3452): receiveMessage: DispatcherMessage
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.303 D/GeckoViewStartup( 3452): onEvent GeckoView:SetDefaultPrefs
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.305 D/GeckoViewContent( 3452): handleEvent: DOMWindowClose
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.306 D/GeckoViewConsole( 3452): onEvent GeckoView:WebExtension:List null
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.309 E/GeckoSessionTestRule( 3452): test_end c5ee677f-bc83-49bd-9e28-2d35f3d0f059 bypassHTTPSOnlyError(org.mozilla.geckoview.test.NavigationDelegateTest)
[task 2023-11-07T09:35:28.059Z] 09:35:28  WARNING -  TEST-UNEXPECTED-FAIL | org.mozilla.geckoview.test.NavigationDelegateTest#bypassHTTPSOnlyError | java.lang.AssertionError: onLoadRequest should be called specified number of times
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  TEST-INFO took 1144ms
Attachment #9283129 - Attachment is obsolete: true
Pushed by afarre@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/f3e41764fab0
Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug

(In reply to Cosmin Sabou [:CosminS] from comment #36)

Backed out for causing gv-junit failures on NavigationDelegateTest.

Push with failures: https://treeherder.mozilla.org/jobs?repo=autoland&selectedTaskRun=KKCjeO_PSe6Ms1HM550YaQ.0&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel&revision=f3e41764fab0e413f58f1b42e270801107cc08ea

Failure log: https://treeherder.mozilla.org/logviewer?job_id=435564829&repo=autoland

Backout link: https://hg.mozilla.org/integration/autoland/rev/eca77d1db747f2995c74d3dc6c48256b109501c7

== Change summary for alert #40202 (as of Thu, 09 Nov 2023 19:57:26 GMT) ==

Regressions:

Ratio Test Platform Options Absolute values (old vs new)
5% cpstartup content-process-startup linux1804-64-shippable-qr e10s fission stylo webrender 76.96 -> 80.50

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=40202

(In reply to Andrej (:aglavic) from comment #37)

(In reply to Cosmin Sabou [:CosminS] from comment #36)

Backed out for causing gv-junit failures on NavigationDelegateTest.

Push with failures: https://treeherder.mozilla.org/jobs?repo=autoland&selectedTaskRun=KKCjeO_PSe6Ms1HM550YaQ.0&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel&revision=f3e41764fab0e413f58f1b42e270801107cc08ea

Failure log: https://treeherder.mozilla.org/logviewer?job_id=435564829&repo=autoland

Backout link: https://hg.mozilla.org/integration/autoland/rev/eca77d1db747f2995c74d3dc6c48256b109501c7

== Change summary for alert #40202 (as of Thu, 09 Nov 2023 19:57:26 GMT) ==

Regressions:

Ratio Test Platform Options Absolute values (old vs new)
5% cpstartup content-process-startup linux1804-64-shippable-qr e10s fission stylo webrender 76.96 -> 80.50

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=40202

So this patch got backed out, but the degraded perf is still there. Are you sure this patch is causing it?

Flags: needinfo?(afarre) → needinfo?(aglavic)

Looking at a perf run of this test in isolation it even looks better.

(In reply to Cosmin Sabou [:CosminS] from comment #36)

Backed out for causing gv-junit failures on NavigationDelegateTest.

Push with failures: https://treeherder.mozilla.org/jobs?repo=autoland&selectedTaskRun=KKCjeO_PSe6Ms1HM550YaQ.0&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel&revision=f3e41764fab0e413f58f1b42e270801107cc08ea

Failure log: https://treeherder.mozilla.org/logviewer?job_id=435564829&repo=autoland

Backout link: https://hg.mozilla.org/integration/autoland/rev/eca77d1db747f2995c74d3dc6c48256b109501c7

Looking at this push, it looks like that failure is bug 1732521.

So I think that this patch should be landable as is. Is it OK if I try again?

Flags: needinfo?(csabou)
Regressions: 1864640

Regarding 1864640. This has already been backed out. See bug 1746524 comment 33.

Pushed by afarre@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/f768135b85f2
Enable browser.tabs.documentchannel.parent-controlled pref. r=smaug
Status: REOPENED → RESOLVED
Closed: 2 years ago5 months ago
Resolution: --- → FIXED
Target Milestone: --- → 121 Branch

(In reply to Natalia Csoregi [:nataliaCs] from comment #34)

Backed out for causing gv-junit failures on bypassHTTPSOnlyError

[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.298 D/GeckoViewProgress( 3452): StateTracker onStateChange: isTopLevel=true, flags=0xc0010, status=NS_OK loadType=8388609
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 D/GeckoSession( 3452): handleMessage GeckoView:PageStop uri=null
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): Error
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): java.lang.AssertionError: onLoadRequest should be called specified number of times
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): Expected: <1>
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452):      but: was <0>
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.Assert.assertThat(Assert.java:964)
[task 2023-11-07T09:35:28.054Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector$1.call(ErrorCollector.java:77)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector.checkSucceeds(ErrorCollector.java:90)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.ErrorCollector.checkThat(ErrorCollector.java:75)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.checkThat(GeckoSessionTestRule.java:929)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.assertMatchesCount(GeckoSessionTestRule.java:965)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.-$$Nest$massertMatchesCount(GeckoSessionTestRule.java)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$CallbackDelegates.clearAndAssert(GeckoSessionTestRule.java:724)
[task 2023-11-07T09:35:28.055Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.afterWait(GeckoSessionTestRule.java:1875)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule.waitForJS(GeckoSessionTestRule.java:2646)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.BaseSessionTest.waitForJS(BaseSessionTest.kt:246)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.NavigationDelegateTest.bypassHTTPSOnlyError(NavigationDelegateTest.kt:676)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.lang.reflect.Method.invoke(Native Method)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.runners.model.FrameworkMethod$1.runReflectiveCall(FrameworkMethod.java:59)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:56)
[task 2023-11-07T09:35:28.056Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.junit.rules.Verifier$1.evaluate(Verifier.java:35)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2.lambda$evaluate$0(GeckoSessionTestRule.java:1509)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2.$r8$lambda$eGWMAiU-In-hdc_ISYHPMUhHPKA(GeckoSessionTestRule.java)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at org.mozilla.geckoview.test.rule.GeckoSessionTestRule$2$$ExternalSyntheticLambda1.run(D8$$SyntheticClass)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:428)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.util.concurrent.FutureTask.run(FutureTask.java:237)
[task 2023-11-07T09:35:28.057Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.app.Instrumentation$SyncRunnable.run(Instrumentation.java:1950)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Handler.handleCallback(Handler.java:751)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Handler.dispatchMessage(Handler.java:95)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.os.Looper.loop(Looper.java:154)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at android.app.ActivityThread.main(ActivityThread.java:6077)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at java.lang.reflect.Method.invoke(Native Method)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:866)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.300 E/GeckoSessionTestRule( 3452): 	at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:756)
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.301 D/GeckoViewContentDelegateParent( 3452): receiveMessage: DispatcherMessage
[task 2023-11-07T09:35:28.058Z] 09:35:28     INFO -  11-07 09:35:27.303 D/GeckoViewStartup( 3452): onEvent GeckoView:SetDefaultPrefs
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.305 D/GeckoViewContent( 3452): handleEvent: DOMWindowClose
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.306 D/GeckoViewConsole( 3452): onEvent GeckoView:WebExtension:List null
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  11-07 09:35:27.309 E/GeckoSessionTestRule( 3452): test_end c5ee677f-bc83-49bd-9e28-2d35f3d0f059 bypassHTTPSOnlyError(org.mozilla.geckoview.test.NavigationDelegateTest)
[task 2023-11-07T09:35:28.059Z] 09:35:28  WARNING -  TEST-UNEXPECTED-FAIL | org.mozilla.geckoview.test.NavigationDelegateTest#bypassHTTPSOnlyError | java.lang.AssertionError: onLoadRequest should be called specified number of times
[task 2023-11-07T09:35:28.059Z] 09:35:28     INFO -  TEST-INFO took 1144ms

== Change summary for alert #40218 (as of Fri, 10 Nov 2023 16:12:42 GMT) ==

Improvements:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
23% fandom fcp windows10-64-shippable-qr bytecode-cached fission warm webrender 27.73 -> 21.33 Before/After
21% wikipedia fcp android-hw-a51-11-0-aarch64-shippable-qr warm webrender 86.90 -> 68.27
8% wikipedia PerceptualSpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 203.18 -> 186.02
8% wikipedia ContentfulSpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 203.91 -> 186.92
8% wikipedia SpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 234.56 -> 216.13
5% wikipedia loadtime android-hw-a51-11-0-aarch64-shippable-qr warm webrender 330.58 -> 312.97

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=40218

Flags: needinfo?(csabou)

I don't understand these numbers, if you're indicating that you're planning a backout please talk to me first.

Flags: needinfo?(afinder)
Regressions: 1865831

(In reply to Andreas Farre [:farre] from comment #45)

I don't understand these numbers, if you're indicating that you're planning a backout please talk to me first.

Hi Andreas! The results posted in Comment #44 are for revision 5d6699b34edce04ffd8886be86de9d604d88a89a which backed out the changes introduced by 149e69327bbb1bc316787380a6a3e54f3e9195e0, as visible in the following graph. Please let me know via a needinfo if you have other questions or concerns. Thanks!

Flags: needinfo?(afinder)
Flags: needinfo?(aglavic)
Regressions: 1866107
Regressions: 1866419
Regressions: 1866868

I think we'll back this out for now.

Status: RESOLVED → REOPENED
Flags: needinfo?(afarre)
Resolution: FIXED → ---
Target Milestone: 121 Branch → ---
Status: REOPENED → NEW

(In reply to Narcis Beleuzu [:NarcisB] from comment #48)

Backed out for causing perf regressions
Backout link: https://hg.mozilla.org/integration/autoland/rev/e7df49cb67e7b69c365956231cacc4ad6c9e2ef8

== Change summary for alert #40602 (as of Fri, 08 Dec 2023 13:32:12 GMT) ==

Regressions:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
17% fandom FirstVisualChange linux1804-64-shippable-qr bytecode-cached fission warm webrender 92.43 -> 107.77 Before/After
16% fandom ContentfulSpeedIndex linux1804-64-shippable-qr bytecode-cached fission warm webrender 95.06 -> 110.40 Before/After

Improvements:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
18% wikipedia fcp linux1804-64-shippable-qr bytecode-cached fission warm webrender 43.41 -> 35.72 Before/After
14% youtube FirstVisualChange android-hw-a51-11-0-aarch64-shippable-qr warm webrender 173.33 -> 149.87
5% youtube PerceptualSpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 520.04 -> 494.47
5% youtube fcp android-hw-a51-11-0-aarch64-shippable-qr warm webrender 499.77 -> 475.43
4% youtube ContentfulSpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 564.13 -> 539.74
4% youtube SpeedIndex android-hw-a51-11-0-aarch64-shippable-qr warm webrender 674.86 -> 650.11

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=40602

(In reply to Narcis Beleuzu [:NarcisB] from comment #49)

Backed out of beta for 121.0b6
Backout link: https://hg.mozilla.org/releases/mozilla-beta/rev/c1852b5f0fd3cae3ab6e22effca4d9f2689aeeeb

== Change summary for alert #40497 (as of Sat, 02 Dec 2023 09:38:49 GMT) ==

Regressions:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
3% facebook ContentfulSpeedIndex linux1804-64-shippable-qr cold fission webrender 1,131.07 -> 1,163.03 Before/After

Improvements:

Ratio Test Platform Options Absolute values (old vs new) Performance Profiles
19% wikipedia fcp linux1804-64-shippable-qr bytecode-cached fission warm webrender 43.13 -> 35.02 Before/After
16% wikipedia fcp linux1804-64-shippable-qr fission warm webrender 47.33 -> 39.67 Before/After

For up to date results, see: https://treeherder.mozilla.org/perfherder/alerts?id=40497

You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: