Perma LeakSanitizer | leak at alloc, __rdl_alloc, alloc::alloc::alloc, _$LT$alloc..alloc..Global$u20$as$u20$core..alloc..Alloc$GT$::alloc
Categories
(Core :: WebRTC, defect, P5)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox-esr60 | --- | unaffected |
| firefox66 | --- | unaffected |
| firefox67 | --- | unaffected |
| firefox68 | --- | fixed |
People
(Reporter: intermittent-bug-filer, Assigned: chmanchester)
References
(Regression)
Details
(Keywords: regression, Whiteboard: [Tier2])
Attachments
(1 file)
#[markdown(off)]
Filed by: nbeleuzu [at] mozilla.com
https://treeherder.mozilla.org/logviewer.html#?job_id=239802877&repo=autoland
[task 2019-04-12T00:38:44.879Z] 00:38:44 INFO - TEST-UNEXPECTED-FAIL | LeakSanitizer | leak at alloc, __rdl_alloc, alloc::alloc::alloc, _$LT$alloc..alloc..Global$u20$as$u20$core..alloc..Alloc$GT$::alloc
[task 2019-04-12T00:38:44.880Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, NewEmptyScopeData
[task 2019-04-12T00:38:44.880Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
[task 2019-04-12T00:38:44.881Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at mozilla::dom::ChromeUtils::GenerateQI, mozilla::dom::ChromeUtils_Binding::generateQI, CallJSNative, js::InternalCallOrConstruct
[task 2019-04-12T00:38:44.882Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame mozilla::dom::ChromeUtils::GenerateQI matched a expected leak
[task 2019-04-12T00:38:44.882Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, js::TenuringTracer::moveElementsToTenured
[task 2019-04-12T00:38:44.882Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
[task 2019-04-12T00:38:44.882Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, CopyScopeData
[task 2019-04-12T00:38:44.883Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
[task 2019-04-12T00:38:44.883Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_realloc, js_pod_arena_realloc, maybe_pod_realloc, pod_realloc
[task 2019-04-12T00:38:44.883Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_realloc matched a expected leak
[task 2019-04-12T00:38:44.883Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, pod_malloc
[task 2019-04-12T00:38:44.884Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
[task 2019-04-12T00:38:44.885Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, mozilla::UniquePtr
[task 2019-04-12T00:38:44.885Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
[task 2019-04-12T00:38:44.886Z] 00:38:44 INFO - TEST-FAIL | LeakSanitizer | leak at js_arena_malloc, js_pod_arena_malloc, maybe_pod_malloc, js::TenuringTracer::moveSlotsToTenured
[task 2019-04-12T00:38:44.886Z] 00:38:44 INFO - INFO | LeakSanitizer | Frame js_arena_malloc matched a expected leak
|
||
Updated•5 years ago
|
|
|
||
Comment 1•5 years ago
•
|
||
I did some retriggers and seems the fail started from https://hg.mozilla.org/integration/autoland/rev/47660e04a4f9e098ddf459d05c99b3e68d351e13
:cmanchester , can you please take a look?
|
|
||
Updated•5 years ago
|
|
Assignee | |
Comment 2•5 years ago
|
||
Oddly the stacks here do look similar to bug 1532884, which we started seeing significantly before the rust update.
|
|
Assignee | |
Comment 3•5 years ago
|
||
Hi Eric. The latest rust update has made this intermittent lsan-detected leak fail permanently. Do you know of someone who may be able to help investigate this? Thank you.
|
||
Comment 4•5 years ago
|
||
My guess is that updating rust (comment 1) changed a signature in our whitelist. Andrew does that sound right?
|
||
Comment 5•5 years ago
|
||
Yeah. This file contains the whitelist for this directory:
https://searchfox.org/mozilla-central/source/testing/web-platform/meta/pointerevents/__dir__.ini
Two of the entries are std::sys::unix::alloc::$LT$impl$u20$core..alloc..GlobalAlloc$u20$for$u20$std..alloc..System$GT$::alloc and std::sys::unix::alloc::$LT$impl$u20$core..alloc..GlobalAlloc$u20$for$u20$std..alloc..System$GT$::realloc. Maybe alloc::alloc::alloc and alloc::alloc::realloc could get added to the list. Normally I'd say that whitelisting something that says "alloc" is bad, but I can't imagine C++ would call anything alloc::alloc::alloc...
|
|
||
Comment 6•5 years ago
|
||
There's two other directories that might need to be fixed up in a similar way: https://searchfox.org/mozilla-central/search?q=GlobalAlloc%24u20%24for%24u20%24std&path=
| Comment hidden (Intermittent Failures Robot) |
|
|
Assignee | |
Updated•5 years ago
|
|
|
Assignee | |
Comment 8•5 years ago
|
||
This is a smaller update than recommended by the wpt-update command, however
the entire suite passes with it on try.
Pushed by cmanchester@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/6dc813d4efdb Update wpt expectations for symbol name updated with rust update. r=mccr8
|
||
Comment 10•5 years ago
|
||
| bugherder | ||
| Comment hidden (Intermittent Failures Robot) |
|
||
Updated•5 years ago
|
|
||
Updated•2 years ago
|
Description
•