Closed Bug 1694106 Opened 3 years ago Closed 7 months ago

Crash in [@ arena_dalloc | core::ptr::drop_in_place<T> | webrender::render_backend::RenderBackend::process_transaction]

Categories

(Core :: Graphics: WebRender, defect)

Product:
Core
Component:
Graphics: WebRender
Platform:
Unspecified
Windows
Type:
defect

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: gsvelto, Unassigned)

Details

(Keywords: crash)

Crash Data

Crash report: https://crash-stats.mozilla.org/report/index/cfbf6021-c03e-4d52-9de0-a0bc40210221

Reason: EXCEPTION_ACCESS_VIOLATION_READ

Top 10 frames of crashing thread:

0 mozglue.dll arena_dalloc memory/build/mozjemalloc.cpp:3359
1 xul.dll core::ptr::drop_in_place<webrender::scene::BuiltScene> ../cb75ad5db02783e8b0222fee363c5f63f7e2cf5b/library/core/src/ptr/mod.rs:179
2 xul.dll webrender::render_backend::RenderBackend::process_transaction gfx/wr/webrender/src/render_backend.rs:946
3 xul.dll webrender::render_backend::RenderBackend::process_api_msg gfx/wr/webrender/src/render_backend.rs:1219
4  @0x7ffd 
5 xul.dll std::sys_common::backtrace::__rust_begin_short_backtrace<closure-5, tuple<>> ../cb75ad5db02783e8b0222fee363c5f63f7e2cf5b/library/std/src/sys_common/backtrace.rs:125
6 xul.dll core::ops::function::FnOnce::call_once<closure-0, tuple<>> ../cb75ad5db02783e8b0222fee363c5f63f7e2cf5b/library/core/src/ops/function.rs:227
7 xul.dll std::sys::windows::thread::{{impl}}::new::thread_start ../cb75ad5db02783e8b0222fee363c5f63f7e2cf5b//library/std/src/sys/windows/thread.rs:56
8 kernel32.dll BaseThreadInitThunk 
9 mozglue.dll patched_BaseThreadInitThunk mozglue/dllservices/WindowsDllBlocklist.cpp:591

It's unclear if this is a real problem or just people with bad hardware. The stacks are relatively consistent between different crashes but not great in general. Some crashes appear to be accessing freed objects but not all of them. Filing this to track the volume in case it picks up.

Severity: -- → S3

I found a bunch of other signatures with stacks similar to this one, like for example:

And there's many more with lower crash counts. Does these look to you like they came from people with broken hardware? I remember stumbling upon something similar not too long ago and you mentioned that these were just common errors coming up often with users having flaky machines.

Flags: needinfo?(emilio)

These do look quite similar to bug 1406996 indeed... It might be worth to chkimg the binaries and such to see if there are bitflips on the dump...

A bunch of these crashes do look "impossible" (as in, null pointers that should never be null, etc)

Flags: needinfo?(emilio)

Closing because no crashes reported for 12 weeks.

Status: NEW → RESOLVED
Closed: 7 months ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.