Closed Bug 1760382 Opened 4 years ago Closed 3 years ago

Intermittent freezes when compiling Firefox

Categories

(Core :: Widget: Cocoa, defect, P3)

ARM64
macOS
defect

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: mstange, Unassigned)

References

Details

Crash Data

Attachments

(2 files)

Steps to reproduce:

  1. Compile Firefox on an Apple Silicon Mac.
  2. During compilation, switch to a running instance of Firefox and interact with it.

Expected results:
Firefox should always be somewhat responsive, even with a build running in the background.

Actual results:
Sometimes Firefox is frozen for multiple seconds.

I saw this on my M1 Max which is running macOS 12.2.1. Nazim reported the same issue a few days ago.
I don't remember this happening in the past.

The symptoms sound like my normal experience for quite a while with Firefox on my 2019 macbook pro with known thermal issues.

Timothy, do you also time being spent in _OSSpinLockLockYield in activity monitor when it happens to you?

Flags: needinfo?(tnikkel)
See Also: → 1670885

(In reply to Timothy Nikkel (:tnikkel) from comment #2)

The symptoms sound like my normal experience for quite a while with Firefox on my 2019 macbook pro with known thermal issues.

In my experience, my M1 Max is cool all the time. Whatever this is, I don't think this is thermal throttling. I'll try to reproduce as well.

See Also: → 1757324
Depends on: 1670885
See Also: 1670885
Attached file Sample of Firefox.txt

Not sure if this is useful or not, didn't seem to resolve symbols in Firefox, maybe because I'm running beta and not nightly?

Flags: needinfo?(tnikkel) → needinfo?(jmuizelaar)

There's only one thread that seems stuck in _OSSpinLockLockYield so it's hard to say if it's the same thing.

Flags: needinfo?(jmuizelaar)

The severity field is not set for this bug.
:spohl, could you have a look please?

For more information, please visit auto_nag documentation.

Flags: needinfo?(spohl.mozilla.bugs)
Severity: -- → S3
Flags: needinfo?(spohl.mozilla.bugs)
Priority: -- → P3

I'm able to reproduce this as well.

I captured a profile of it: https://share.firefox.dev/3KdUoMT Only the parent process seems frozen (not capturing any sample). I unfroze it to capture the profile with a ctrl-z on the terminal where I was building.

I just experienced this today and used a "kill -SIGABRT" on Firefox to crash it, generating a crash report which I tagged with this Bug number. Hopefully it will get connected.

Crash Signature: syscall_thread_switch | _OSSpinLockLockYield
Crash Signature: syscall_thread_switch | _OSSpinLockLockYield → [@ syscall_thread_switch | _OSSpinLockLockYield ]
See Also: 1670885

Closing because no crashes reported for 12 weeks.

Status: NEW → RESOLVED
Closed: 3 years ago
Resolution: --- → WORKSFORME

We'd only hit that signature if a dev abort their process during a hang.

Status: RESOLVED → REOPENED
Resolution: WORKSFORME → ---

Closing because no crashes reported for 12 weeks.

Status: REOPENED → RESOLVED
Closed: 3 years ago3 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: