Re: [RFC PATCH V3 13/16] x86/sev: Add Check of #HV event in path

[Tianyu Lan]

On 3/9/2023 12:18 AM, Gupta, Pankaj wrote:
> On 3/1/2023 12:11 PM, Gupta, Pankaj wrote:
> > On 1/22/2023 3:46 AM, Tianyu Lan wrote:

> > > diff --git a/arch/x86/kernel/sev.c b/arch/x86/kernel/sev.c
> > > index a8862a2eff67..fe5e5e41433d 100644
> > > --- a/arch/x86/kernel/sev.c
> > > +++ b/arch/x86/kernel/sev.c
> > > @@ -179,6 +179,45 @@ void noinstr __sev_es_ist_enter(struct pt_regs 
> > > *regs)
> > >       this_cpu_write(cpu_tss_rw.x86_tss.ist[IST_INDEX_VC], new_ist);
> > >   }
> > > +static void do_exc_hv(struct pt_regs *regs)
> > > +{
> > > +    /* Handle #HV exception. */
> > > +}
> > > +
> > > +void check_hv_pending(struct pt_regs *regs)
> > > +{
> > > +    if (!cc_platform_has(CC_ATTR_GUEST_SEV_SNP))
> > > +        return;
> > > +
> > > +    if ((regs->flags & X86_EFLAGS_IF) == 0)
> > > +        return;
> >
> > Will this return and prevent guest from executing NMI's
> > while irqs are disabled?
>
> I think we need to handle NMI's even when irqs are disabled.

Yes, nice catch!

> As we reset "no_further_signal" in hv_raw_handle_exception()
> and return from check_hv_pending() when irqs are disabled, this
> can result in loss/delay of NMI event?

Will fix this in the next version.

Thanks.