Re: [PATCH 4/2] perf_counter: Fix swcounter context invariance

[Frederic Weisbecker <fweisbec@gmail.com>]

On Thu, Aug 13, 2009 at 10:22:58AM +0200, Peter Zijlstra wrote:
> On Thu, 2009-08-13 at 10:05 +0200, Frederic Weisbecker wrote:
> > On Thu, Aug 13, 2009 at 09:51:55AM +0200, Peter Zijlstra wrote:
> > > Not really related to this topic, but it needs posting anyway.
> > > 
> > > ---
> > > 
> > > Subject: perf_counter: Fix swcounter context invariance
> > > From: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > > Date: Fri Aug 07 13:29:13 CEST 2009
> > > 
> > > perf_swcounter_is_counting() uses a lock, which means we cannot use
> > > swcounters from NMI or when holding that particular lock, this is
> > > unintended.
> > > 
> > > The below removes the lock, this opens up race window, but not worse
> > > than the swcounters already experience due to RCU traversal of the
> > > context in perf_swcounter_ctx_event().
> > > 
> > > Cc: Paul Mackerras <paulus@samba.org>
> > > Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > 
> > 
> > 
> > As a side effect, it's possible this also fixes the hard lockups
> > while opening a lockdep tracepoint counter.
> 
> It will -- not a side-effect at all, intended consequence :-)


Great then! Important fix for .31 :-)