Closed Bug 1553694 Opened 5 years ago Closed 5 years ago

Intermittent browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PCompositorBridge::Msg_WillClose 1 more times than expected before handling user events -

Categories

(Firefox :: General, defect, P5)

defect

Tracking

()

RESOLVED FIXED
Firefox 69
Tracking Status
firefox-esr60 --- unaffected
firefox67 --- unaffected
firefox68 --- unaffected
firefox69 --- fixed

People

(Reporter: intermittent-bug-filer, Assigned: florian)

References

(Regression)

Details

(Keywords: intermittent-failure, regression)

Attachments

(1 file)

Filed by: ncsoregi [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer.html#?job_id=247856554&repo=autoland
Full log: https://queue.taskcluster.net/v1/task/ZBjckuNHSfmiEzSGOFxPsQ/runs/0/artifacts/public/logs/live_backing.log


23:26:26 INFO - PCompositorBridge::Msg_MakeSnapshot - at most 1 times
23:26:26 INFO - TEST-PASS | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PLayerTransaction::Msg_GetTextureFactoryIdentifier happened as many times as expected before handling user events -
23:26:26 INFO - TEST-PASS | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PCompositorBridge::Msg_Initialize happened as many times as expected before handling user events -
23:26:26 INFO - Buffered messages finished
23:26:26 INFO - TEST-UNEXPECTED-FAIL | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PCompositorBridge::Msg_WillClose 1 more times than expected before handling user events -
23:26:26 INFO - Stack trace:
23:26:26 INFO - chrome://mochikit/content/browser-test.js:test_ok:1314
23:26:26 INFO - chrome://mochitests/content/browser/browser/base/content/test/performance/browser_startup_syncIPC.js:null:241
23:26:26 INFO - chrome://mochikit/content/browser-test.js:Tester_execTest/<:1116
23:26:26 INFO - chrome://mochikit/content/browser-test.js:Tester_execTest:1144
23:26:26 INFO - chrome://mochikit/content/browser-test.js:nextTest/<:1005
23:26:26 INFO - chrome://mochikit/content/tests/SimpleTest/SimpleTest.js:SimpleTest.waitForFocus/waitForFocusInner/focusedOrLoaded/<:803
23:26:26 INFO - TEST-PASS | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PCompositorBridge::Msg_MakeSnapshot allowed 1 more times before handling user events -
23:26:26 INFO - whitelisted sync IPC before becoming idle:
23:26:26 INFO - PAPZInputBridge::Msg_ProcessUnhandledEvent - at most 1 times
23:26:26 INFO - PAPZInputBridge::Msg_ReceiveMouseInputEvent - at most 1 times
23:26:26 INFO - TEST-PASS | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PAPZInputBridge::Msg_ProcessUnhandledEvent happened as many times as expected before becoming idle -
23:26:26 INFO - TEST-PASS | browser/base/content/test/performance/browser_startup_syncIPC.js | sync IPC PAPZInputBridge::Msg_ReceiveMouseInputEvent happened as many times as expected before becoming idle -
23:26:26 INFO - Not taking screenshot here: see the one that was previously logged
23:26:26 INFO - TEST-UNEXPECTED-FAIL | browser/base/content/test/performance/browser_startup_syncIPC.js | Unexpected sync IPC behavior during startup; open the profile_startup_syncIPC.json artifact in the Firefox Profiler to see what happened -
23:26:26 INFO - Stack trace:
23:26:26 INFO - chrome://mochikit/content/browser-test.js:test_ok:1314
23:26:26 INFO - chrome://mochitests/content/browser/browser/base/content/test/performance/browser_startup_syncIPC.js:null:261
23:26:26 INFO - chrome://mochikit/content/browser-test.js:Tester_execTest/<:1116
23:26:26 INFO - chrome://mochikit/content/browser-test.js:Tester_execTest:1144
23:26:26 INFO - chrome://mochikit/content/browser-test.js:nextTest/<:1005
23:26:26 INFO - chrome://mochikit/content/tests/SimpleTest/SimpleTest.js:SimpleTest.waitForFocus/waitForFocusInner/focusedOrLoaded/<:803
23:26:26 INFO - Leaving test bound

Looking at the profiles, we still have only 2 PCompositorBridge::Msg_WillClose sync IPC, and they were already whitelisted 'before first paint'. Having them moved to 'before handling user events' is actually an improvement. From the profile, it seems this is triggered when we are done with the sanitytest window, so the time when it happens probably isn't deterministic. I think we can just whitelist and move on.

Thanks for taking this Florian. I'm not sure why this commit triggered any change at all on windows - the test is only run on osx (https://hg.mozilla.org/integration/autoland/rev/dfabf5135cf1b219f4ea809d4b066ffe5d8a3cef#l1.7) and the one line that changed doesn't have any other test coverage AFAICT.

Assignee: nobody → florian
Status: NEW → ASSIGNED
Flags: needinfo?(bgrinstead)
Pushed by fqueze@mozilla.com:
https://hg.mozilla.org/integration/autoland/rev/0ca9a03071b7
accept a second PCompositorBridge::Msg_WillClose sync IPC in the 'before handling user events' phase, r=mconley.
Status: ASSIGNED → RESOLVED
Closed: 5 years ago
Resolution: --- → FIXED
Target Milestone: --- → Firefox 69
Has Regression Range: --- → yes
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: