Re: [PATCH] Only call EvtchnFlush on valid Cpus

[Paul Durrant <xadimgnik@xxxxxxxxx>]

On 03/07/2023 17:46, Paul Durrant wrote:
> On 20/06/2023 15:33, Owen Smith wrote:
> > The Evtchn processor array is created using 
> > KeQueryMaximumProcessorCountEx, which
> > can include processors that do not get initialized.
> > Skip cleanup and flushing uninitialized event channels
> >
> > Signed-off-by: Owen Smith <owen.smith@xxxxxxxxx>
> Acked-by: Paul Durrant <paul@xxxxxxx>
Actually, taking another look I think we should actually fix 
EvtchnTwoLevelIsProcessorEnabled() to check SystemProcessorVcpuId() 
rather than blindly returning TRUE (the FIFO variant already checks) and 
then we can use EvtchnIsProcessorEnabled()...
> > ---
> >   src/xenbus/evtchn.c | 6 ++++++
> >   1 file changed, 6 insertions(+)
> >
> > diff --git a/src/xenbus/evtchn.c b/src/xenbus/evtchn.c
> > index 2e63edb..ec4ac28 100644
> > --- a/src/xenbus/evtchn.c
> > +++ b/src/xenbus/evtchn.c
> > @@ -1752,10 +1752,16 @@ EvtchnRelease(
> >       for (Cpu = 0; Cpu < Context->ProcessorCount; Cpu++) {
> >           PXENBUS_EVTCHN_PROCESSOR Processor;
> > +        unsigned int             vcpu_id;
> > +        NTSTATUS                 status;
> >           ASSERT(Context->Processor != NULL);
> >           Processor = &Context->Processor[Cpu];
> > +        status = SystemProcessorVcpuId(Cpu, &vcpu_id);
... here instead. I'll send a follow-up patch.

  Paul

> > +        if (!NT_SUCCESS(status))
> > +            continue;
> > +
> >           EvtchnFlush(Context, Cpu);
> >           (VOID) KeRemoveQueueDpc(&Processor->Dpc);