lkml.org 
[lkml]   [2009]   [Nov]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRE: [regression bisect -next] BUG: using smp_processor_id() in preemptible [00000000] code: rmmod
From
Date
On Thu, 2009-11-26 at 12:09 -0500, Leyendecker, Robert wrote:
>
> I have same oops (different cause-vector). Details here
> http://lkml.org/lkml/2009/11/23/548

Which is an -rt kernel and very likely totally unrelated.

> Can anyone tell me if they are related and/or recommend any
> work-arounds or troubleshooting steps?
>
> Could this be related to scheduler trying to reschedule a task/thread
> on other proc

No, the warning says you try to get the current cpu number when the code
in question is in no way tied to a particular cpu, therefore the cpu
number obtained is totally useless.

> and possibly a false alarm?

Most likely not.

Look at your particular source-tree and try to fine the
smp_processor_id() call that triggers this, then look what it needs the
cpu number for and fix the code so that it makes sense ;-)

> Nov 23 17:24:15 localhost kernel: BUG: using smp_processor_id() in preemptible [00000000] code: smash/3062
> Nov 23 17:24:15 localhost kernel: caller is __schedule+0x93/0x7d4
> Nov 23 17:24:15 localhost kernel: Pid: 3062, comm: smash Not tainted 2.6.31.6-rt19 #1
> Nov 23 17:24:15 localhost kernel: Call Trace:
> Nov 23 17:24:15 localhost kernel: [<c06dc3df>] ? printk+0xf/0x18
> Nov 23 17:24:15 localhost kernel: [<c05499a6>] debug_smp_processor_id+0xa6/0xbc
> Nov 23 17:24:15 localhost kernel: [<c06dc70e>] __schedule+0x93/0x7d4
> Nov 23 17:24:15 localhost kernel: [<c046621f>] ? audit_syscall_exit+0xfa/0x10f
> Nov 23 17:24:15 localhost kernel: [<c040a955>] ? syscall_trace_leave+0xc8/0xef
> Nov 23 17:24:15 localhost kernel: [<c040338a>] work_resched+0x5/0x19
> Nov 23 17:24:15 localhost kernel: __ratelimit: 6 callbacks suppressed




\
 
 \ /
  Last update: 2009-11-26 18:25    [W:0.068 / U:0.132 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site