Created attachment 319954 [details] dmesg terminating with oops and a number of traces Description of problem: This is what I see in dmesg: BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: [<ffffffff81181fa7>] __list_add+0x16/0x87 PGD 1a195067 PUD 1a196067 PMD 1a13b067 PTE 0 Oops: 0000 [1] SMP Full dmesg output attached. No idea what "Tainted" means in this case. There are no external modules of any sort loaded or present on the machine. I have seen "Disabling IRQ #16" displayed all over the place on other occasssions. Version-Release number of selected component (if applicable): kernel-2.6.27-0.398.rc9.fc10.x86_64 How reproducible: I can actually dig out from logs other "NULL pointer dereference" messages. They go back to 2.6.27-0.211.rc1.git3.fc10 and through 2.6.27-0.226.rc1.git5.fc10. They all seem to be related to X but earlier ones point to drm_ati_pcigart_init+0x1be/0x210 though. Additional info: That particular one seems to be something new after an update to 2.6.27-0.398.rc9.fc10
Created attachment 319963 [details] additonal oopses and traces After running that kernel a bit longer I got also what is attached here; after "Disabling IRQ #16" line. So, yes, this seems to be repetable.
(In reply to comment #2) > Full dmesg output attached. No idea what "Tainted" means > in this case. There are no external modules of any sort loaded > or present on the machine. > G - all loaded modules are GPL M - there was an earlier machine check exception D - there was a previous oops
Got the same error with 2.6.27-0.408.rc9.git1.fc10.x86_64 too. Sometime after that a machine just died. Hard to tell if this was related as I got no information from that last incident.
Has this improved at all in recent rawhide?
> Has this improved at all in recent rawhide? Indeed it did improve. I do not see anymore this or "Disabling IRQ" messages. That was showing up in a seemingly random manner but I would think that it would still show up from time to time if still present.