Re: cdrecord deadlocks linux 2.6.8.1 (problem in setscheduler)

From: Chris Wright (chrisw@private)
Date: Tue Oct 19 2004 - 14:09:47 PDT


* James Morris (jmorris@private) wrote:
> On Tue, 19 Oct 2004, Stephen Smalley wrote:
> 
> > On Tue, 2004-10-19 at 16:09, Stephen Smalley wrote:
> > > NMI Watchdog detected LOCKUP on CPU1, eip c030f99c, registers:
> > 
> > Sorry, that one wasn't very useful.  Re-tried, got a nicer one.
> > Not sure why the audit_log_end_irq calls don't show up in the call
> > trace.
> 
> Yes, it's the runqueue lock.  One simple possibility would be to convert
> the audit code over to use the keventd workqueue, and use
> schedule_delayed_work() to kick the audit logging via a timer in this
> case.

Yeah, guess that would work, but it's not that nice a solution ;-/



This archive was generated by hypermail 2.1.3 : Tue Oct 19 2004 - 14:10:15 PDT