Firefox hang causes MacOS to crash
Categories
(Firefox :: General, defect)
Tracking
()
People
(Reporter: mozer624, Unassigned)
Details
User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:123.0) Gecko/20100101 Firefox/123.0
Steps to reproduce:
It's been happening to me a few times, but I don't know how to reproduce exactly. Each time is when I was watching a video on Rumble, and about ~30mins into the video firefox crashes, then my Mac crashes. When Firefox starts back up, before it restores any sessions, it will crash a few more times. Eventually I'm able to start a new session.
Comment 1•2 years ago
|
||
The Bugbug bot thinks this bug should belong to the 'Core::Widget: Cocoa' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.
Comment 2•2 years ago
|
||
Are the crashes listed in about:crashes?
| Reporter | ||
Comment 3•2 years ago
|
||
Yes, I see 3 from today which corresponds, and some from previous days. I just pressed Submit All. A bunch of old crashes says "failed" (to upload?), but it looks like all the recent ones were submitted. Here are the links:
https://crash-stats.mozilla.org/report/index/1c4611ff-00c3-41f9-a28c-ee5990240308
https://crash-stats.mozilla.org/report/index/bb82603e-e07c-4f62-82eb-6afd60240308
https://crash-stats.mozilla.org/report/index/7fff6eb6-db58-4e8e-838d-6ad3b0240308
https://crash-stats.mozilla.org/report/index/98c95630-5f05-402b-8f73-9be5d0240308
https://crash-stats.mozilla.org/report/index/c148afbd-ceeb-4fdf-a2a2-0fdc20240308
https://crash-stats.mozilla.org/report/index/085b081c-dd42-4b55-a92e-636b00240308
https://crash-stats.mozilla.org/report/index/991b5aa7-853a-482f-bb18-1f3d60240308
https://crash-stats.mozilla.org/report/index/ffff696a-8fbd-4431-a968-c077b0240305
Comment 4•2 years ago
|
||
All these crashes are completely different. Also, none of them appear to be startup crashes. It might be worth testing your RAM using Apple's utilities. Here is an article on how to do that: https://support.apple.com/en-us/102550
Once you've run the diagnostic, does the report show anything that could explain the crashes that you are seeing?
| Reporter | ||
Comment 5•2 years ago
|
||
I ran the diagnosis and it said there are no issues. ADP000
I reported this to Firefox because the crashes happened in the same pattern: Use Rumble, use PiP (maybe related?), crash loop 3 times, computer fine again.
Is there additional logging I can enable and run while using Firefox to capture this?
| Reporter | ||
Comment 6•2 years ago
|
||
I just had a crash loop happen again, but I wasn't Rumble or PiP, so it's probably unrelated.
Here's a crashed tab that just happened too: https://crash-stats.mozilla.org/report/index/b29b3195-80d9-4453-9133-574800240309
If that's also unrelated then we can close this.
Comment 7•2 years ago
|
||
(In reply to Michael Ozeryansky from comment #6)
I just had a crash loop happen again, but I wasn't Rumble or PiP, so it's probably unrelated.
Here's a crashed tab that just happened too: https://crash-stats.mozilla.org/report/index/b29b3195-80d9-4453-9133-574800240309
If that's also unrelated then we can close this.
This is yet another crash. I'm going to move this to a more general component in the hopes that we can better isolate this, since this isn't strictly widget related.
| Reporter | ||
Comment 8•2 years ago
|
||
This time Firefox crashed but my mac did not: https://crash-stats.mozilla.org/report/index/3818aafd-cafa-44c3-b783-06dd30240313
While using PiP. I can keep posting crash links on this thread if that could be useful, it's just Firefox has been crashing multiple times a day for me the past few weeks, and I can't imagine it's all for different reasons.
Comment 9•2 years ago
|
||
Moving this to Core::Javascript Engine in the hopes we might get some more insight into the cause of the crash.
Comment 10•2 years ago
|
||
There are two crashes here that touch the JS engine, but neither of them look like bugs in our code.
In this crash, we crash in JIT-compiled code with EXC_BAD_INSTRUCTION / EXC_I386_INVOP while trying to execute a perfectly good jump instruction. (The target of the jump instruction is close enough that it's present in the dump, so we can verify that we're not jumping to an inaccessible address.)
This crash claims to be crashing with EXC_BAD_ACCESS / EXC_I386_GPFLT on a sub r12, r13, which makes even less sense: there's nothing that could possibly go wrong while subtracting one register from another.
Neither of these instructions should ever crash on properly functioning hardware. This really looks like faulty hardware of some sort. Is it possible that watching a video for half an hour overheats your machine? Do you ever see crashes when running other similarly strenuous non-Firefox programs?
I'm going to move this back to a more general component, since there's nothing to be done here from a JS perspective.
Comment 11•2 years ago
|
||
See Comment #10 - would you be able to share more information?
| Reporter | ||
Comment 12•2 years ago
|
||
Since my last comment, I have upgraded my MacOS to the latest and have not had a crash since. I have watched a long video using PiP from Rumble (which was the original condition causing the most crashes) and had no issues.
Updated•2 years ago
|
Description
•