On Tue, 27 Mar 2012, Lindsay Sprinter wrote:
On Mon, 26 Mar 2012, Anthony Shipman wrote:
I've been getting occasional kernel errors over the past few weeks. Should I worry?
WARNING: at kernel/sched.c:5914 thread_return+0x232/0x79d() (Tainted: P W ---------------- ) Pid: 11507, comm: firefox Tainted: P W ---------------- 2.6.32-220.2.1.el6.x86_64 #1 Call Trace: [<ffffffff81069997>] ? warn_slowpath_common+0x87/0xc0 [<ffffffff810699ea>] ? warn_slowpath_null+0x1a/0x20 [<ffffffff814eccc5>] ? thread_return+0x232/0x79d [<ffffffff810958e3>] ? __hrtimer_start_range_ns+0x1a3/0x460 [<ffffffff81051dc6>] ? enqueue_task+0x66/0x80 [<ffffffff810a2239>] ? futex_wait_queue_me+0xb9/0xf0 [<ffffffff810a3308>] ? futex_wait+0x1f8/0x380 [<ffffffff810705b7>] ? current_fs_time+0x27/0x30 [<ffffffff81094b70>] ? hrtimer_wakeup+0x0/0x30 [<ffffffff81095bd4>] ? hrtimer_start_range_ns+0x14/0x20 [<ffffffff810a4a51>] ? do_futex+0x121/0xb00 [<ffffffff81090a10>] ? autoremove_wake_function+0x0/0x40 [<ffffffff81218def>] ? selinux_file_permission+0xbf/0x150 [<ffffffff810a54ab>] ? sys_futex+0x7b/0x170 [<ffffffff8100b0f2>] ? system_call_fastpath+0x16/0x1b
WARNING: at kernel/sched.c:5914 thread_return+0x232/0x79d() Pid: 6679, comm: Xorg Tainted: P W ---------------- 2.6.32-220.2.1.el6.x86_64 #1 Call Trace: [<ffffffff81069997>] ? warn_slowpath_common+0x87/0xc0 [<ffffffff810699ea>] ? warn_slowpath_null+0x1a/0x20 [<ffffffff814eccc5>] ? thread_return+0x232/0x79d [<ffffffff810958e3>] ? __hrtimer_start_range_ns+0x1a3/0x460 [<ffffffff81094dc1>] ? lock_hrtimer_base+0x31/0x60 [<ffffffff814ee498>] ? schedule_hrtimeout_range+0xc8/0x160 [<ffffffff81094b70>] ? hrtimer_wakeup+0x0/0x30 [<ffffffff81095bd4>] ? hrtimer_start_range_ns+0x14/0x20 [<ffffffff8118b079>] ? poll_schedule_timeout+0x39/0x60 [<ffffffff8118b6e8>] ? do_select+0x578/0x6b0 [<ffffffff8104d80d>] ? check_preempt_curr+0x6d/0x90 [<ffffffff8118b820>] ? __pollwait+0x0/0xf0
I cannot directly help you with the worried query except that whenever the kernel starts putting out errors one shouid get worried. Now the traces say tainted, this usually means you have a closed src driver loaded, in this case possibly a driver for a graphics card. This will be the place to start. Closed src drivers cannot directly talk to the kernel and as there programming is unknown this is a long time src of problems.
Whereas every single oops I've experienced on a tainted machine has been a real software or hardware problem related to the area in which the oops was reported. I think this noise about tainting is just that, unfortunately — noise. (I get remarkably little robustness out of any kernels these days on dell laptops with various graphics chips that used to work reasonably flawlessly. I don't have high expectations of today's software unfortunately - X and the in-kernel drivers appears to be going rapidly downhill in quality and stability). But if you temporarily disable the thing that's tainting the kernel (hopefully it's not essential to a barely working machine!), somehow reproduce the problem, get the same stack trace and report it on bugzilla, at least you'll make the devs happy. But in this case, the stack trace appears to be remarkably generic, so good luck! Turn off selinux perhaps? -- Tim Connors