Re: [RFC PATCH 08/13] KVM: nSVM: Flush both L1 and L2 ASIDs on KVM_REQ_TLB_FLUSH

[Yosry Ahmed <yosry.ahmed@linux.dev>]

On Fri, Feb 28, 2025 at 08:58:04PM -0500, Maxim Levitsky wrote:
> On Wed, 2025-02-05 at 18:23 +0000, Yosry Ahmed wrote:
> > KVM_REQ_TLB_FLUSH is used to flush all TLB entries for all contexts
> > (e.g. in kvm_flush_remote_tlbs()). Flush both L1 and L2 ASIDs in
> > svm_flush_tlb_all() to handle it appropriately.
> > 
> > This is currently not required as nested transitions do unconditional
> > TLB flushes, but this is a step toward eliminating that.
> > 
> > Signed-off-by: Yosry Ahmed <yosry.ahmed@linux.dev>
> > ---
> >  arch/x86/kvm/svm/nested.c | 1 -
> >  arch/x86/kvm/svm/svm.c    | 4 +++-
> >  2 files changed, 3 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
> > index 0e9b0592c1f83..0735177b95a1d 100644
> > --- a/arch/x86/kvm/svm/nested.c
> > +++ b/arch/x86/kvm/svm/nested.c
> > @@ -491,7 +491,6 @@ static void nested_svm_entry_tlb_flush(struct kvm_vcpu *vcpu)
> >  	 * TODO: optimize unconditional TLB flush/MMU sync.  A partial list of
> >  	 * things to fix before this can be conditional:
> >  	 *
> > -	 *  - Flush TLBs for both L1 and L2 remote TLB flush
> >  	 *  - Honor L1's request to flush an ASID on nested VMRUN
> >  	 *  - Sync nested NPT MMU on VMRUN that flushes L2's ASID[*]
> >  	 *  - Don't crush a pending TLB flush in vmcb02 on nested VMRUN
> > diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
> > index 9e29f87d3bd93..8342c7eadbba8 100644
> > --- a/arch/x86/kvm/svm/svm.c
> > +++ b/arch/x86/kvm/svm/svm.c
> > @@ -4044,7 +4044,9 @@ static void svm_flush_tlb_all(struct kvm_vcpu *vcpu)
> >  	if (WARN_ON_ONCE(svm_hv_is_enlightened_tlb_enabled(vcpu)))
> >  		hv_flush_remote_tlbs(vcpu->kvm);
> >  
> > -	svm_flush_tlb_asid(vcpu, svm->current_vmcb);
> > +	svm_flush_tlb_asid(vcpu, &svm->vmcb01);
> > +	if (svm->nested.initialized)
> > +		svm_flush_tlb_asid(vcpu, &svm->nested.vmcb02);
> >  }
> 
> This makes sense.
> 
> Note that this doesn't really flush the ASID used, but rather ensures
> that we will flush it on next entry via that vmcb. (because of new asid,
> that will be picked, or because we set tlb_ctl in that vmcb)

Right, what I mean by 'flush' here is to setup the flush. For SVM all
flushes are done on VM-enter anyway.

> 
> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>

Thanks!