1

I've been using Linux Mint MATE edition for a while and suffering the odd freeze. Today, it switched to a terminal-style output and gave me some actual info before it died. I took a photo, see below (open it in a new tab to zoom in and read the text).

enter image description here

The only "lead" i can see is a line that says

BUG: scheduling while atomic: swapper/0/0/0x10010000

but, for all i know i'm only seeing the bottom of the stack trace here and the actual problem is much further up.

Questions: a) can anyone see if there's anything in the photo that gives a clue as to what's going wrong? b) is the stuff in this trace liable to be in a log file somewhere so i can have a proper look at it?

  • The first step is memtest86. It's available as a stand-alone tool (bootable iso cd), or in most Linux CD/DVD. Boot from your CD and look into the installation menu. When started, memtest86 will perform memory stress test: let it run for a couple of hours. – AndrewQ Sep 20 '13 at 13:35
  • I doubt it is a bad memory error, but if you have time memtest can't hurt. – goldilocks Sep 20 '13 at 13:51
1

This pretty much means what it says:

BUG: scheduling while atomic

The kernel is multi-threaded, and the kernel API has various mechanisms for locking access to resources in a way similar to userspace APIs such as pthreads. There are rules surrounding their use, and if you break the rules you can create problems. I.e., someone did exactly that, probably in a module.

I don't think this would necessarily bring the kernel or the system down -- the error obviously was caught -- however, it would bring the module down and that could bring the system down.

You need to file an official bug report. With some software you should start "downstream", i.e., with the distro, but although distros do patch the kernel a bit, it is less likely this is their fault, so you might as well start "upstream" (the kernel developers) with the link. You have to create an account, which is free and easy. Then you want to try and determine if the bug has been reported before by searching the site (don't use the search bar on the main page, click through to the search page). From my perspective, there is not much more than your photograph to go on, so I searched for "BUG: scheduling while atomic: swapper/0/0/0x10010000" and got this list of hits. Maybe something rings a bell in there for you, maybe not. You have more details so should be able to refine the search.

If you think you have found a duplicate, read it and add a comment reporting your experience. You should try and include a copy of the log in the photo, but preferably not the log itself -- almost certainly also went to /var/log/syslog or some such, so:

grep 'BUG: scheduling while atomic: swapper/0/0/0x10010000' /var/log/*

If you have had the problem before or similar problems it may be in there more than once, each incident will look similar. Copy/paste one or two into a text file and attach it in the bug report (either an existing one, or create a new one and explain what happened).

They'll get back to you in an indeterminate period of time (but it probably won't be too long).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.