Re: [PATCH] make might_sleep() display the oopsing process

[Arjan van de Ven <arjan@infradead.org>]

On Sat, 30 Aug 2008 10:45:20 -0400
Joe Korty <joe.korty@ccur.com> wrote:

> On Fri, Aug 29, 2008 at 07:20:03PM -0400, Andrew Morton wrote:
> > On Thu, 28 Aug 2008 11:36:28 +0200
> > Ingo Molnar <mingo@elte.hu> wrote:
> > 
> > > > Expand might_sleep's printk to indicate the oopsing process.
> > >
> > > good idea
> > 
> > Why?  I don't recall ever having been interested in this
> > information.
> > 
> > (I shouldn't have to ask questions like this.  Please spend a little
> > more time in preparing the patch descriptions).
> 
> I wrote and used the patch once, as I wanted to know
> which program in a lengthy multiple-program test suite
> was triggering the problem.  That made it easier coming
> up with a shortened test that would replicate the problem.
> 
> However I only needed it that one time so this patch can
> be considered to be exceptionally optional :)

WARN() and WARN_ON() already print the oopsing process.
so if might_sleep() just calls one of those it already prints it.
If it doesn't call one of these.. maybe it should?


-- 
If you want to reach me at my work email, use arjan@linux.intel.com
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org