i915 potential deadlock
I attempted to send a potential deadlock issue from my work email (acoffin@sfcompute.com), and my personal email (alex@cyberialabs.net) and neither seemed to show up on the mailing list so I'm just going to try to report it here instead. I really apologize if I am making a mistake somehow, or spamming, but I really can't figure out how this is supposed to work and I am suspicious that my emails are just being blackholed by a spam filter.
Edit: Apparently, it just hadn't been manually approved yet (sorry) as the mailing list has been experiencing a lot of spam recently. As I was asked to post a bug report here, I will leave this bug report up.
I was attempting to investigate an issue that I believe may be related to iwlwifi, but managed to stumble across this lockdep report first when I was attempting to debug the kernel in general. This is from Linux Kernel version v6.1.112 (technically the Debian patched version, but I don't think that is relevant to this issue).
I haven't had a chance to read this code yet, but I thought that I would report it so that someone more familiar with this code base could have a chance to investigate before I do (if I have time). As far as I can see there weren't any changes to i915 after 6.1.112 that were relevant (I only see "drm/i915/gem: fix bitwise and logical AND mixup") so unless a backport is missing I believe that this issue would also affect mainline version, but as I don't know the steps to reproduce this issue I can't easily test the mainline version.
I am including 3 text files.
- A log from
journalctl -k
- The same log from before but with the relevant portion highlighted
- An annotated version of the prior relevant portion (passed through
./scripts/decode_stacktrace.sh -r 6.1.112).
Checklist before filing a bug
Have you tried a recent kernel?
I would consider the latest second latest LTS release recent, but considering I don't know how to trigger this bug I and I'm not exactly interesting in using mainline Linux on my work laptop either this bug report has to be sufficient, or this issue should be closed.
Are you running an upstream kernel?
As said above, nope. I am running with the Debian patches (Debian version 6.1.112-1 of the kernel to be precise after recompiling with lockdep enabled) although I didn't see any that patches that touch i915.
Is i915 or xe the right component?
I would say so from the logs.
Does the bug report exist already?
Not that I could find.
Are you willing to follow-up with the bug?
As I don't really know how to reproduce the issue I doubt I could do much follow up, but I may be able to follow up with a bit more work depending upon the request.
Are you filing a security bug?
Not that I am aware of.