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;
Subject: Re: [PATCH] kvm: x86: rewrite kvm_spec_ctrl_valid_bits
To:     Sean Christopherson <sean.j.christopherson@intel.com>
Cc:     Maxim Levitsky <mlevitsk@redhat.com>, kvm@vger.kernel.org,
        "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@kernel.org>,
        linux-kernel@vger.kernel.org, Joerg Roedel <joro@8bytes.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Wanpeng Li <wanpengli@tencent.com>,
        "H. Peter Anvin" <hpa@zytor.com>, Borislav Petkov <bp@alien8.de>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Ingo Molnar <mingo@redhat.com>,
        Jim Mattson <jmattson@google.com>
References: <20200702174455.282252-1-mlevitsk@redhat.com>
 <20200702181606.GF3575@linux.intel.com>
 <3793ae0da76fe00036ed0205b5ad8f1653f58ef2.camel@redhat.com>
 <20200707061105.GH5208@linux.intel.com>
 <7c1d9bbe-5f59-5b86-01e9-43c929b24218@redhat.com>
 <20200707081444.GA7417@linux.intel.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <e5da32da-6cb2-85b1-a12b-da796843d2bb@redhat.com>
Date:   Tue, 7 Jul 2020 10:17:14 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.9.0
MIME-Version: 1.0
In-Reply-To: <20200707081444.GA7417@linux.intel.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 07/07/20 10:14, Sean Christopherson wrote:
>>> One oddity with this whole thing is that by passing through the MSR, KVM is
>>> allowing the guest to write bits it doesn't know about, which is definitely
>>> not normal.  It also means the guest could write bits that the host VMM
>>> can't.
>> That's true.  However, the main purpose of the kvm_spec_ctrl_valid_bits
>> check is to ensure that host-initiated writes are valid; this way, you
>> don't get a #GP on the next vmentry's WRMSR to MSR_IA32_SPEC_CTRL.
>> Checking the guest CPUID bit is not even necessary.
> Right, what I'm saying is that rather than try and decipher specs to
> determine what bits are supported, just throw the value at hardware and
> go from there.  That's effectively what we end up doing for the guest writes
> anyways.

Yes, it would prevent the #GP.

> Actually, the current behavior will break migration if there are ever legal
> bits that KVM doesn't recognize, e.g. guest writes a value that KVM doesn't
> allow and then migration fails when the destination tries to stuff the value
> into KVM.

Yes, unfortunately migration would also be broken if the target (and the
guest CPUID) is an older CPU.  But that's not something we can fix
without trapping all writes which would be unacceptably slow.

Paolo