Re: [Xen-devel] [PATCH v5 3/6] xen/PMU: Initialization code for Xen PMU

[Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>]

On 07/02/2015 12:21 PM, David Vrabel wrote:
> On 02/07/15 15:53, Boris Ostrovsky wrote:
> > Map shared data structure that will hold CPU registers, VPMU context,
> > V/PCPU IDs of the CPU interrupted by PMU interrupt. Hypervisor fills
> > this information in its handler and passes it to the guest for further
> > processing.
> >
> > Set up PMU VIRQ.
> >
> > Now that perf infrastructure will assume that PMU is available on a PV
> > guest we need to be careful and make sure that accesses via RDPMC
> > instruction don't cause fatal traps by the hypervisor. Provide a nop
> > RDPMC handler.
> >
> > For the same reason avoid issuing a warning on a write to APIC's LVTPC.
> >
> > Both of these will be made functional in later patches.
> [...]
> > +               rc = bind_virq_to_irqhandler(VIRQ_XENPMU, cpu,
> > +                                            xen_pmu_irq_handler,
> > +                                            IRQF_PERCPU|IRQF_NOBALANCING,
> > +                                            pmu_name, NULL);
> If you bind a VIRQ as IRQF_PERCPU it can only be safely unbound on the
> CPU it is bound to (because the percpu handler does not take the
> desc->lock).
This is a true per-cpu VIRQ and we only unbind it on cpu_die() path so I 
think IRQF_PERCPU should be safe, no? (it behaves similar to VIRQ_TIMER 
in how/when it is unbound).
-boris

> Otherwise, Reviewed-by: David Vrabel <david.vrabel@xxxxxxxxxx>
>
> David

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel