On Wed, Oct 18, 2017 at 3:55 PM, Andrew Jones wrote:
> On Wed, Oct 18, 2017 at 03:18:43PM +0200, Christoffer Dall wrote:
>> On Wed, Oct 18, 2017 at 02:13:05PM +0200, Andrew Jones wrote:
>> > On Sat, Oct 14, 2017 at 09:13:35PM +0200, Christoffer Dall wrote:
>> > > On Fri, Sep
On Wed, Oct 18, 2017 at 03:18:43PM +0200, Christoffer Dall wrote:
> On Wed, Oct 18, 2017 at 02:13:05PM +0200, Andrew Jones wrote:
> > On Sat, Oct 14, 2017 at 09:13:35PM +0200, Christoffer Dall wrote:
> > > On Fri, Sep 29, 2017 at 01:30:41PM +0200, Andrew Jones wrote:
> > > > When the vPMU is in
On Wed, Oct 18, 2017 at 02:13:05PM +0200, Andrew Jones wrote:
> On Sat, Oct 14, 2017 at 09:13:35PM +0200, Christoffer Dall wrote:
> > On Fri, Sep 29, 2017 at 01:30:41PM +0200, Andrew Jones wrote:
> > > When the vPMU is in use if a VCPU's perf event overflow handler
> > > were to fire after the
On Sat, Oct 14, 2017 at 09:13:35PM +0200, Christoffer Dall wrote:
> On Fri, Sep 29, 2017 at 01:30:41PM +0200, Andrew Jones wrote:
> > When the vPMU is in use if a VCPU's perf event overflow handler
> > were to fire after the VCPU started waiting, then the wake up
> > done by the kvm_vcpu_kick()
On Fri, Sep 29, 2017 at 01:30:41PM +0200, Andrew Jones wrote:
> When the vPMU is in use if a VCPU's perf event overflow handler
> were to fire after the VCPU started waiting, then the wake up
> done by the kvm_vcpu_kick() call in the handler would do nothing,
> as no "pmu overflow" state is
On Thu, Oct 05, 2017 at 10:37:32AM +0100, Marc Zyngier wrote:
> On 29/09/17 12:30, Andrew Jones wrote:
> > When the vPMU is in use if a VCPU's perf event overflow handler
> > were to fire after the VCPU started waiting, then the wake up
> > done by the kvm_vcpu_kick() call in the handler would do
On 29/09/17 12:30, Andrew Jones wrote:
> When the vPMU is in use if a VCPU's perf event overflow handler
> were to fire after the VCPU started waiting, then the wake up
> done by the kvm_vcpu_kick() call in the handler would do nothing,
> as no "pmu overflow" state is checked in
When the vPMU is in use if a VCPU's perf event overflow handler
were to fire after the VCPU started waiting, then the wake up
done by the kvm_vcpu_kick() call in the handler would do nothing,
as no "pmu overflow" state is checked in kvm_arch_vcpu_runnable().
Fix this by checking the IRQ_PENDING