On Wed, Dec 05, 2012 at 09:11:40PM -0200, Marcelo Tosatti wrote:
> > >> +static int handle_apic_write(struct kvm_vcpu *vcpu)
> > >> +{
> > >> + unsigned long exit_qualification = vmcs_readl(EXIT_QUALIFICATION);
> > >> + u32 offset = exit_qualification & 0xfff;
> > >> +
> > >> + /* APIC-write VM exit is trap-like and thus no need to adjust IP */
> > >> + return kvm_apic_write_nodecode(vcpu, offset) == 0;
> > >> +}
> > >
> > > Point of return value == 0?
> > if kvm_apic_write_nodecode() handle successfully, it will return zero. Then it will return 1 for this vmexit handle.
> > What's wrong?
>
> If kvm_apic_write_nodecode fails (return 0 for the vmexit handle),
> there is an exit to userspace. Why is that necessary?
kvm_apic_write_nodecode() returns apic_reg_write() value.
apic_reg_write() returns 1 when #GP should be injected if used from
MSR, if used from mmio apic_reg_write() is ignored since errors cannot
be reported to a cpu. Since kvm_apic_write_nodecode() is used for mmio
only make it void and ignore apic_reg_write() return value.
--
Gleb.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
