Re: [PATCH v2] kvm: nVMX: Set nested_run_pending in vmx_set_nested_state after checks complete

From: Aaron Lewis
Date: Wed May 08 2019 - 13:54:34 EST


From: Sean Christopherson <sean.j.christopherson@xxxxxxxxx>
Date: Wed, May 8, 2019 at 7:20 AM
To: Paolo Bonzini
Cc: <linux-kernel@xxxxxxxxxxxxxxx>, <kvm@xxxxxxxxxxxxxxx>, Peter
Shier, Aaron Lewis

> On Wed, May 08, 2019 at 02:16:39PM +0200, Paolo Bonzini wrote:
> > From: Aaron Lewis <aaronlewis@xxxxxxxxxx>
>
> If this is actually attributed to Aaron it needs his SOB.
>
> > nested_run_pending=1 implies we have successfully entered guest mode.
> > Move setting from external state in vmx_set_nested_state() until after
> > all other checks are complete.
>
> It'd be helpful to at least mention the flag is consumed by
> nested_vmx_enter_non_root_mode().
>
> > Based on a patch by Aaron Lewis.
> >
> > Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
> > ---
>
> For the code itself:
>
> Reviewed-by: Sean Christopherson <sean.j.christopherson@xxxxxxxxx>
>
> > arch/x86/kvm/vmx/nested.c | 10 ++++++----
> > 1 file changed, 6 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
> > index cec77f30f61c..e58caff92694 100644
> > --- a/arch/x86/kvm/vmx/nested.c
> > +++ b/arch/x86/kvm/vmx/nested.c
> > @@ -5420,9 +5420,6 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
> > if (!(kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE))
> > return 0;
> >
> > - vmx->nested.nested_run_pending =
> > - !!(kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING);
> > -
> > if (nested_cpu_has_shadow_vmcs(vmcs12) &&
> > vmcs12->vmcs_link_pointer != -1ull) {
> > struct vmcs12 *shadow_vmcs12 = get_shadow_vmcs12(vcpu);
> > @@ -5446,9 +5443,14 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
> > return -EINVAL;
> >
> > vmx->nested.dirty_vmcs12 = true;
> > + vmx->nested.nested_run_pending =
> > + !!(kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING);
> > +
> > ret = nested_vmx_enter_non_root_mode(vcpu, false);
> > - if (ret)
> > + if (ret) {
> > + vmx->nested.nested_run_pending = 0;
> > return -EINVAL;
> > + }
> >
> > return 0;
> > }
> > --
> > 1.8.3.1
> >

nested_run_pending is also checked in
nested_vmx_check_vmentry_postreqs
(https://elixir.bootlin.com/linux/v5.1/source/arch/x86/kvm/vmx/nested.c#L2709)
so I think the setting needs to be moved to just prior to that call
with Paolo's rollback along with another for if the prereqs and
postreqs fail. I put a patch together below:

------------------------------------

nested_run_pending=1 implies we have successfully entered guest mode.
Move setting from external state in vmx_set_nested_state() until after
all other checks are complete.

Signed-off-by: Aaron Lewis <aaronlewis@xxxxxxxxxx>
Reviewed-by: Peter Shier <pshier@xxxxxxxxxx>
---
arch/x86/kvm/vmx/nested.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c
index 6401eb7ef19c..cf1f810223d2 100644
--- a/arch/x86/kvm/vmx/nested.c
+++ b/arch/x86/kvm/vmx/nested.c
@@ -5460,9 +5460,6 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
if (!(kvm_state->flags & KVM_STATE_NESTED_GUEST_MODE))
return 0;

- vmx->nested.nested_run_pending =
- !!(kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING);
-
if (nested_cpu_has_shadow_vmcs(vmcs12) &&
vmcs12->vmcs_link_pointer != -1ull) {
struct vmcs12 *shadow_vmcs12 = get_shadow_vmcs12(vcpu);
@@ -5480,14 +5477,21 @@ static int vmx_set_nested_state(struct kvm_vcpu *vcpu,
return -EINVAL;
}

+ vmx->nested.nested_run_pending =
+ !!(kvm_state->flags & KVM_STATE_NESTED_RUN_PENDING);
+
if (nested_vmx_check_vmentry_prereqs(vcpu, vmcs12) ||
- nested_vmx_check_vmentry_postreqs(vcpu, vmcs12, &exit_qual))
+ nested_vmx_check_vmentry_postreqs(vcpu, vmcs12, &exit_qual)) {
+ vmx->nested.nested_run_pending = 0;
return -EINVAL;
+ }

vmx->nested.dirty_vmcs12 = true;
ret = nested_vmx_enter_non_root_mode(vcpu, false);
- if (ret)
+ if (ret) {
+ vmx->nested.nested_run_pending = 0;
return -EINVAL;
+ }

return 0;
}