lkml.org 
[lkml]   [2005]   [Nov]   [26]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateFri, 25 Nov 2005 21:14:54 -0800
From"Paul E. McKenney" <>
SubjectRe: BUG: spinlock recursion on 2.6.14-mm2 when oprofiling
On Sat, Nov 26, 2005 at 11:39:55AM +0800, Wu Fengguang wrote:
> On Fri, Nov 25, 2005 at 03:28:29PM -0800, Paul E. McKenney wrote:
> > And here is an alternative patch that assumes that the answer to both
> > questions above is "no".  It is shorter, though mostly due to use of
> > the list_splice_init() and list_for_each_entry_safe() primitives.
> 
> Ok, I'll try it. But it may take time.
> 
> Currently I'm running oprofile on linux-2.6.15-rc2-mm1. It seems the bug is not
> quite reproducible. I'll report again if there are any new findings.

Yep, the bug is indeed non-deterministic.  For it to happen, softirq
must interrupt a "task_mortuary" critical section, and that softirq
must execute a task-struct RCU callback.

						Thanx, Paul
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-11-26 06:18    [from the cache]
©2003-2008