Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20211011194615.2955791-1-vipinsh@google.com> <YWSdTpkzNt3nppBc@google.com>
 <CALMp9eRzPXg2WS6-Yy6U90+B8wXm=zhVSkmAym4Y924m7FM-7g@mail.gmail.com> <YWhgxjAwHhy0POut@google.com>
In-Reply-To: <YWhgxjAwHhy0POut@google.com>
From:   Jim Mattson <jmattson@google.com>
Date:   Thu, 14 Oct 2021 10:05:35 -0700
Message-ID: <CALMp9eQ4y+YO7THjfpHzJPmoODkUqoPUURaBvL+OdGjZhAMuTA@mail.gmail.com>
Subject: Re: [PATCH] KVM: VMX: Add a wrapper for reading INVPCID/INVEPT/INVVPID
 type
To:     Sean Christopherson <seanjc@google.com>
Cc:     Vipin Sharma <vipinsh@google.com>, pbonzini@redhat.com,
        dmatlack@google.com, kvm@vger.kernel.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Oct 14, 2021 at 9:54 AM Sean Christopherson <seanjc@google.com> wrote:
>
> On Mon, Oct 11, 2021, Jim Mattson wrote:
> > On Mon, Oct 11, 2021 at 1:23 PM Sean Christopherson <seanjc@google.com> wrote:
> > >
> > > On Mon, Oct 11, 2021, Vipin Sharma wrote:
> > > > -     if (type > 3) {
> > > > +     if (type > INVPCID_TYPE_MAX) {
> > >
> > > Hrm, I don't love this because it's not auto-updating in the unlikely chance that
> > > a new type is added.  I definitely don't like open coding '3' either.  What about
> > > going with a verbose option of
> > >
> > >         if (type != INVPCID_TYPE_INDIV_ADDR &&
> > >             type != INVPCID_TYPE_SINGLE_CTXT &&
> > >             type != INVPCID_TYPE_ALL_INCL_GLOBAL &&
> > >             type != INVPCID_TYPE_ALL_NON_GLOBAL) {
> > >                 kvm_inject_gp(vcpu, 0);
> > >                 return 1;
> > >         }
> >
> > Better, perhaps, to introduce a new function, valid_invpcid_type(),
> > and squirrel away the ugliness there?
>
> Oh, yeah, definitely.  I missed that SVM's invpcid_interception() has the same
> open-coded check.
>
> Alternatively, could we handle the invalid type in the main switch statement?  I
> don't see anything in the SDM or APM that architecturally _requires_ the type be
> checked before reading the INVPCID descriptor.  Hardware may operate that way,
> but that's uArch specific behavior unless there's explicit documentation.

Right. INVVPID and INVEPT are explicitly documented to check the type
first, but INVPCID is not.