Date:   Thu, 9 Dec 2021 18:22:20 +0000
From:   Sean Christopherson <seanjc@google.com>
To:     Jim Mattson <jmattson@google.com>
Cc:     Paolo Bonzini <pbonzini@redhat.com>, Marc Orr <marcorr@google.com>,
        vkuznets@redhat.com, wanpengli@tencent.com, joro@8bytes.org,
        tglx@linutronix.de, mingo@redhat.com, bp@alien8.de,
        dave.hansen@linux.intel.com, x86@kernel.org, hpa@zytor.com,
        thomas.lendacky@amd.com, mlevitsk@redhat.com, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] KVM: x86: Always set kvm_run->if_flag
Message-ID: <YbJJXKFevTV/L3in@google.com>
References: <20211209155257.128747-1-marcorr@google.com>
 <5f8c31b4-6223-a965-0e91-15b4ffc0335e@redhat.com>
 <CALMp9eThf3UtvoLFjajkrXtvOEWQvc8_=Xf6-m6fHXkOhET+GA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CALMp9eThf3UtvoLFjajkrXtvOEWQvc8_=Xf6-m6fHXkOhET+GA@mail.gmail.com>
Precedence: bulk

On Thu, Dec 09, 2021, Jim Mattson wrote:
> On Thu, Dec 9, 2021 at 9:48 AM Paolo Bonzini <pbonzini@redhat.com> wrote:
> >
> > On 12/9/21 16:52, Marc Orr wrote:
> > > The kvm_run struct's if_flag is a part of the userspace/kernel API. The
> > > SEV-ES patches failed to set this flag because it's no longer needed by
> > > QEMU (according to the comment in the source code). However, other
> > > hypervisors may make use of this flag. Therefore, set the flag for
> > > guests with encrypted registers (i.e., with guest_state_protected set).
> > >
> > > Fixes: f1c6366e3043 ("KVM: SVM: Add required changes to support intercepts under SEV-ES")
> > > Signed-off-by: Marc Orr<marcorr@google.com>
> >
> > Applied, though I wonder if it is really needed by those other VMMs
> > (which? gVisor is the only one that comes to mind that is interested in
> > userspace APIC).
> 
> Vanadium appears to have one use of it.
> 
> > It shouldn't be necessary for in-kernel APIC (where userspace can inject
> > interrupts at any time), and ready_for_interrupt_injection is superior
> > for userspace APIC.
> 
> LOL. Here's that one use...
> 
> if (vcpu_run_state_->request_interrupt_window &&
> vcpu_run_state_->ready_for_interrupt_injection &&
> vcpu_run_state_->if_flag) {
> ...
> }
> 
> So, maybe this is much ado about nothing?

I assume the issue is that SEV-ES always squishes if_flag, so that above statement
can never evaluate true.