lkml.org 
[lkml]   [2014]   [Apr]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 1/2] smp: Non busy-waiting IPI queue
On Wed, Apr 02, 2014 at 08:30:23PM +0200, Frederic Weisbecker wrote:
> 2014-04-02 20:05 GMT+02:00 Paul E. McKenney <paulmck@linux.vnet.ibm.com>:
> > On Wed, Apr 02, 2014 at 06:26:05PM +0200, Frederic Weisbecker wrote:
> >> diff --git a/kernel/smp.c b/kernel/smp.c
> >> index 06d574e..bfe7b36 100644
> >> --- a/kernel/smp.c
> >> +++ b/kernel/smp.c
> >> @@ -265,6 +265,50 @@ int smp_call_function_single_async(int cpu, struct call_single_data *csd)
> >> }
> >> EXPORT_SYMBOL_GPL(smp_call_function_single_async);
> >>
> >> +void generic_smp_queue_function_single_interrupt(void *info)
> >> +{
> >> + struct queue_single_data *qsd = info;
> >> +
> >> + WARN_ON_ONCE(xchg(&qsd->pending, 0) != 1);
> >
> > I am probably missing something here, but shouldn't this function copy
> > *qsd to a local on-stack variable before doing the above xchg()? What
> > prevents the following from happening?
> >
> > o CPU 0 does smp_queue_function_single(), which sets ->pending
> > and fills in ->func and ->data.
> >
> > o CPU 1 takes IPI, invoking generic_smp_queue_function_single_interrupt().
> >
> > o CPU 1 does xchg(), so that ->pending is now zero.
> >
> > o An attempt to reuse the queue_single_data sees ->pending equal
> > to zero, so the ->func and ->data is overwritten.
> >
> > o CPU 1 calls the new ->func with the new ->data (or any of the other
> > two possible unexpected outcomes), which might not be helpful to
> > the kernel's actuarial statistics.
> >
> > So what am I missing?
>
> Ah, I forgot to precise that the function must remain the same for all
> calls on a single qsd. And the data is always the qsd so this one can
> only stay stable. So that shouldn't be a problem.

I did indeed miss that particular constraint. ;-)

> But you're right. The fact that we pass the function as an argument of
> smp_queue_function_single() suggests that we can pass a different
> function across various calls on a same qsd. So that's confusing.
> Perhaps changing smp_queue_function_single() such that it only takes
> the qsd as an argument would make that clearer? Then it's up to the
> caller to initialize the qsd with the constant function. I could
> define smp_queue_function_init() for that purpose. Or
> DEFINE_QUEUE_FUNCTION_DATA() for static initializers.
>
> How does that sound?

Sounds good!

Thanx, Paul



\
 
 \ /
  Last update: 2014-04-02 21:41    [W:0.036 / U:0.216 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site