Re: [Xen-devel] [PATCH 03/12] evtchn delivery on HVM

[Stefano Stabellini <stefano.stabellini@eu.citrix.com>]

On Mon, 14 Jun 2010, Konrad Rzeszutek Wilk wrote:
> > +static int __cpuinit xen_hvm_cpu_notify(struct notifier_block *self,
> > +				    unsigned long action, void *hcpu)
> > +{
> > +	int cpu = (long)hcpu;
> > +	switch (action) {
> > +	case CPU_UP_PREPARE:
> > +		per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu];
> > +		break;
> 
> Is there no need to do anything if the CPUs are brought down?
> 

nope


> > +/* the callback vector mechanism is a newer alternative way of receiving
> > + * event channel notifications from Xen: we can receive vector callbacks
> > + * on any vcpus and we don't need any PCI or IO APIC support */
> 
> I am not an expert in English, but I think you need to revise this
> comment.


/* vector callbacks are better than PCI interrupts to receive event
 * channel notifications because we can receive vector callbacks on any
 * vcpu and we don't need PCI support or APIC interactions */

is it good enough?


> > +void xen_callback_vector(void)
> > +{
> > +	int rc;
> > +	uint64_t callback_via;
> > +	if (xen_have_vector_callback) {
> > +		callback_via = HVM_CALLBACK_VECTOR(XEN_HVM_EVTCHN_CALLBACK);
> > +		rc = xen_set_callback_via(callback_via);
> > +		if (rc) {
> > +			printk(KERN_ERR "request for callback vector failed\n");
> 
> Perhaps mention which type? Say 'Request for Xen HVM callback vector
> failed.\n' ?
> 

good idea, I'll do that