Intermittent toolkit/components/telemetry/tests/unit/test_UtilityScalars.js | application crashed [@ mozilla::ipc::UtilityProcessImpl::Init(int, char**)]
Categories
(Core :: Security: Process Sandboxing, defect)
Tracking
()
Tracking | Status | |
---|---|---|
thunderbird_esr102 | --- | unaffected |
firefox-esr102 | --- | unaffected |
firefox107 | --- | unaffected |
firefox108 | --- | wontfix |
firefox109 | --- | fixed |
People
(Reporter: intermittent-bug-filer, Assigned: mkmelin)
References
(Regression)
Details
(Keywords: crash, intermittent-failure, regression)
Crash Data
Attachments
(2 files)
Filed by: mkmelin [at] iki.fi
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=396113291&repo=comm-central
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/Wz-ruXBFT1aWv2lb1hbzvQ/runs/0/artifacts/public/logs/live_backing.log
Comment hidden (Intermittent Failures Robot) |
Assignee | ||
Comment 2•1 year ago
|
||
Mozilla crash reason: MOZ_CRASH(Unable to preload module) - https://searchfox.org/mozilla-central/rev/3d01e045479a6e5237958bd9aa8eb64306e1f48a/ipc/glue/UtilityProcessImpl.cpp#30. Looks like this is from bug 1797301 then - https://hg.mozilla.org/mozilla-central/rev/f8406f212dc308033f6454e07be7aea0aaa3646b
Any ideas?
Comment 3•1 year ago
|
||
Set release status flags based on info from the regressing bug 1797301
Comment hidden (Intermittent Failures Robot) |
Updated•1 year ago
|
Comment 5•1 year ago
|
||
No ?
Comment 6•1 year ago
|
||
macosx1015-64-qr: 1
opt: 1
The code is not used on macOS, I dont see how it could exists there.
Comment 7•1 year ago
|
||
This is all failing only on comm-central
Assignee | ||
Comment 8•1 year ago
|
||
Yes, looks like basically perma fail on windows. see the X1 failures e.g. at https://treeherder.mozilla.org/jobs?repo=comm-central&revision=6159998bfa02d443a2ebf3e4104c4cd63a26f77f
Assignee | ||
Comment 9•1 year ago
|
||
And here is one for debug: https://treeherder.mozilla.org/logviewer?job_id=397151001&repo=comm-central&lineNumber=3920
What library load could be failing?
Comment 10•1 year ago
|
||
ipc/glue/UtilityAudioDecoderParent.cpp:76
Comment 11•1 year ago
|
||
I dont know where that is, there is nothing on searchfox: https://searchfox.org/comm-central/source/ipc/glue/UtilityAudioDecoderParent.cpp
Comment 12•1 year ago
|
||
The best I get is https://searchfox.org/comm-central/rev/cc5837b209537d99febe8d64eef7d9a76807fe62/mozilla/ipc/glue/UtilityAudioDecoderParent.cpp#69-76 so line 76 is the end of the function, not sure which one it is. Are those libs even available on comm-central ?
Assignee | ||
Comment 13•1 year ago
|
||
Thanks, looks like we need to package those like https://searchfox.org/mozilla-central/rev/83b86005c6913c2062419efb8aabdf2e683aa47f/browser/installer/package-manifest.in#150
For the library loads in UtilityAudioDecoderParent.cpp, that ifdef probably should be conditional on MOZ_FFVPX
Comment 14•1 year ago
|
||
(In reply to Magnus Melin [:mkmelin] from comment #13)
Thanks, looks like we need to package those like https://searchfox.org/mozilla-central/rev/83b86005c6913c2062419efb8aabdf2e683aa47f/browser/installer/package-manifest.in#150
For the library loads in UtilityAudioDecoderParent.cpp, that ifdef probably should be conditional on MOZ_FFVPX
Can you give a try to if defined(XP_WIN) && defined(MOZ_FFPVX)
then ? If it fixes, I can review that quickly. I'm focused on some quite urgent issue with debug symbols on Snap, so spinning up a comm-central build etc would likely take me too much time to confirm
Assignee | ||
Comment 15•1 year ago
|
||
Yep, I'll see if I can get it to work
Assignee | ||
Comment 16•1 year ago
|
||
Assignee | ||
Comment 17•1 year ago
|
||
Comment 18•1 year ago
|
||
Re-componenting based on regressing bug.
Comment 19•1 year ago
|
||
Pushed by mkmelin@iki.fi: https://hg.mozilla.org/integration/autoland/rev/41bebff09700 mozavcodec and mozavutil should be ifdef MOZ_FFPVX. r=gerard-majax DONTBUILD
Comment 20•1 year ago
|
||
Backed out changeset 41bebff09700 (Bug 1800033) for causing failures in browser_utility_audioDecodeCrash.js CLOSED TREE
Log: https://treeherder.mozilla.org/logviewer?job_id=397439019&repo=autoland&lineNumber=10035
Backout: https://hg.mozilla.org/integration/autoland/rev/17a7f82366276081b22f16184e69a2cb99f747db
Comment 21•1 year ago
|
||
From the error it looks like decoding is happening in the content process: TEST-UNEXPECTED-FAIL | ipc/glue/test/browser/browser_utility_audioDecodeCrash.js | playback https://example.com/browser/ipc/glue/test/browser/small-shot.mp3 was from decoder 'ffmpeg audio decoder', expected 'Utility Generic' - false == true - got false, expected true (operator ==)
This code would verify if it's on Generic or WMF process: https://searchfox.org/mozilla-central/rev/05f486c76f84830041e51b7c68e3e697dc768454/ipc/glue/test/browser/head.js#89-91, but being in the content process is symptom of something else.
Comment 22•1 year ago
|
||
Ok, it's a simple typo in the MOZ_FFPVX
usage that should have been MOZ_FFVPX
.
Comment 23•1 year ago
|
||
with the current patch: https://treeherder.mozilla.org/jobs?repo=try&revision=1262d40b631d88b781001f7ac41453f9b1d44434
with the correct define: https://treeherder.mozilla.org/jobs?repo=try&revision=32d40588f385ee15dda8116c36eda72c0d031e71
Assignee | ||
Comment 24•1 year ago
|
||
Hmm, the correct one doesn't look happy either.
Comment 25•1 year ago
|
||
(In reply to Magnus Melin [:mkmelin] from comment #24)
Hmm, the correct one doesn't look happy either.
It's all green, I dont know what you expect for successs? ASAN failures are expected, it is supposed to be skipped, but my try ran it anyway
Comment 26•1 year ago
|
||
Pushed by mkmelin@iki.fi: https://hg.mozilla.org/integration/autoland/rev/6c75b4aa4b74 mozavcodec and mozavutil should be ifdef MOZ_FFPVX. r=gerard-majax
Comment 27•1 year ago
|
||
bugherder |
Comment 28•1 year ago
|
||
Pushed by mkmelin@iki.fi: https://hg.mozilla.org/comm-central/rev/db00b84f4fee package mozavutil+mozavcodec for mail. r=aleca
Comment 29•1 year ago
|
||
The patch landed in nightly and beta is affected.
:mkmelin, is this bug important enough to require an uplift?
- If yes, please nominate the patch for beta approval.
- If no, please set
status-firefox108
towontfix
.
For more information, please visit auto_nag documentation.
Assignee | ||
Updated•1 year ago
|
Assignee | ||
Updated•1 year ago
|
Comment hidden (Intermittent Failures Robot) |
Description
•