Re: [PATCH 1/1] KVM: Correct deliver_bitmask in kvm_get_intr_delivery_bitmask

[Sheng Yang <sheng@linux.intel.com>]

On Monday 02 March 2009 15:12:52 Gleb Natapov wrote:
> On Fri, Feb 27, 2009 at 12:58:55PM +0800, Sheng Yang wrote:
> > 1. deliver_bitmask is a bitmap rather than a unsigned long intereger.
> > 2. Lowest priority target bitmap wrong calculated by mistake.
> > 3. Prevent potential NULL reference.
> >
> > Signed-off-by: Sheng Yang <sheng@linux.intel.com>
> > ---
> >  virt/kvm/irq_comm.c |    6 +++++-
> >  1 files changed, 5 insertions(+), 1 deletions(-)
> >
> > diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c
> > index d165e05..8e1869d 100644
> > --- a/virt/kvm/irq_comm.c
> > +++ b/virt/kvm/irq_comm.c
> > @@ -54,8 +54,12 @@ void kvm_get_intr_delivery_bitmask(struct kvm_ioapic
> > *ioapic, deliver_bitmask);
> >  	switch (entry->fields.delivery_mode) {
> >  	case IOAPIC_LOWEST_PRIORITY:
> > +		/* Select one in deliver_bitmask */
> >  		vcpu = kvm_get_lowest_prio_vcpu(ioapic->kvm,
> >  				entry->fields.vector, deliver_bitmask);
> > +		bitmap_zero(deliver_bitmask, KVM_MAX_VCPUS);
> > +		if (!vcpu)
> > +			return;
> >  		__set_bit(vcpu->vcpu_id, deliver_bitmask);
> >  		break;
>
> The only caller of kvm_ioapic_get_delivery_bitmask() is
> kvm_get_intr_delivery_bitmask() why not add delivery_mode parameter to
> kvm_ioapic_get_delivery_bitmask() and get rid of
> kvm_get_intr_delivery_bitmask()?

Oh yeah, that's right. I think it's OK to merge 
kvm_ioapic_get_delivery_bitmask() into kvm_get_intr_delivery_bitmask(). If no 
objection, I would send a patch to merge them.

Thanks.

-- 
regards
Yang, Sheng
>
> >  	case IOAPIC_FIXED:
> > @@ -65,7 +69,7 @@ void kvm_get_intr_delivery_bitmask(struct kvm_ioapic
> > *ioapic, if (printk_ratelimit())
> >  			printk(KERN_INFO "kvm: unsupported delivery mode %d\n",
> >  				entry->fields.delivery_mode);
> > -		*deliver_bitmask = 0;
> > +		bitmap_zero(deliver_bitmask, KVM_MAX_VCPUS);
> >  	}
> >  }
> >
> > --
> > 1.5.4.5
>
> --
> 			Gleb.