Received-SPF: pass (google.com: domain of linux-kernel+bounces-137112-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240404121327.3107131-1-pbonzini@redhat.com> <20240404121327.3107131-8-pbonzini@redhat.com>
 <43d1ade0461868016165e964e2bc97f280aee9d4.camel@intel.com> <ZhSYEVCHqSOpVKMh@google.com>
In-Reply-To: <ZhSYEVCHqSOpVKMh@google.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Tue, 9 Apr 2024 16:43:59 +0200
Message-ID: <CABgObfZAJ50Z30VzFLdSrQFOEaPxpyFWuvVr1iGogjhs2_+bGA@mail.gmail.com>
Subject: Re: [PATCH v5 07/17] KVM: x86: add fields to struct kvm_arch for CoCo features
To: Sean Christopherson <seanjc@google.com>
Cc: Rick P Edgecombe <rick.p.edgecombe@intel.com>, "kvm@vger.kernel.org" <kvm@vger.kernel.org>, 
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, 
	"michael.roth@amd.com" <michael.roth@amd.com>, Isaku Yamahata <isaku.yamahata@intel.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Apr 9, 2024 at 3:21=E2=80=AFAM Sean Christopherson <seanjc@google.c=
om> wrote:
> > I'm a little late to this conversation, so hopefully not just complicat=
ing
> > things. But why not deduce has_private_mem and has_protected_state from=
 the
> > vm_type during runtime? Like if kvm.arch.vm_type was instead a bit mask=
 with
> > the bit position of the KVM_X86_*_VM set, kvm_arch_has_private_mem() co=
uld
> > bitwise-and with a compile time mask of vm_types that have primate memo=
ry.
> > This also prevents it from ever transitioning through non-nonsensical s=
tates
> > like vm_type =3D=3D KVM_X86_TDX_VM, but !has_private_memory, so would b=
e a little
> > more robust.
>
> LOL, time is a circle, or something like that.  Paolo actually did this i=
n v2[*],
> and I objected, vociferously.

To be fair, Rick is asking for something much less hideous - just set

 kvm->arch.vm_type =3D (1 << vm_type);

and then define kvm_has_*(kvm) as !!(kvm->arch.vm_type & SOME_BIT_MASK).

And indeed it makes sense as an alternative. It also feels a little
bit more restrictive and the benefit is small, so I think I'm going to
go with this version.

Paolo