Messages in this thread |  | | | Date | Wed, 16 Apr 2008 15:07:25 -0400 (EDT) | | From | Steven Rostedt <> | | Subject | Re: [BUG/PATCH] x86 mmiotrace: dynamically disable non-boot CPUs |
On Wed, 16 Apr 2008, Ingo Molnar wrote:
>
> so lets fix those preemptability bugs. They show that the
> cpu-up/cpu-down ops are called from atomic context - it should normally
> be straightforward to sort out - there's no particular reason why the
> ->open()/->close() methods of an ftrace plugin should run in atomic
> context. Steve, any ideas where the atomicity might come from?
>
They shouldn't be called in an atomic section. The only thing I do to
protect them is call mutex_lock/unlock. Those should allow preemption to
take place.
-- Steve
|  |