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: <20210323084515.1346540-1-vkuznets@redhat.com>
In-Reply-To: <20210323084515.1346540-1-vkuznets@redhat.com>
From:   Haiwei Li <lihaiwei.kernel@gmail.com>
Date:   Thu, 25 Mar 2021 15:51:52 +0800
Message-ID: <CAB5KdObQ7t4aXFsYioNdVfNt6B+ChJLB5dKsWxAtoXMYpgSoBA@mail.gmail.com>
Subject: Re: [PATCH] KVM: x86/vPMU: Forbid writing to MSR_F15H_PERF MSRs when
 guest doesn't have X86_FEATURE_PERFCTR_CORE
To:     Vitaly Kuznetsov <vkuznets@redhat.com>
Cc:     kvm list <kvm@vger.kernel.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Sean Christopherson <seanjc@google.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Wei Huang <wei.huang2@amd.com>, Joerg Roedel <joro@8bytes.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Mar 23, 2021 at 4:48 PM Vitaly Kuznetsov <vkuznets@redhat.com> wrote:
>
> MSR_F15H_PERF_CTL0-5, MSR_F15H_PERF_CTR0-5 MSRs are only available when
> X86_FEATURE_PERFCTR_CORE CPUID bit was exposed to the guest. KVM, however,
> allows these MSRs unconditionally because kvm_pmu_is_valid_msr() ->
> amd_msr_idx_to_pmc() check always passes and because kvm_pmu_set_msr() ->
> amd_pmu_set_msr() doesn't fail.
>
> In case of a counter (CTRn), no big harm is done as we only increase
> internal PMC's value but in case of an eventsel (CTLn), we go deep into
> perf internals with a non-existing counter.
>
> Note, kvm_get_msr_common() just returns '0' when these MSRs don't exist
> and this also seems to contradict architectural behavior which is #GP
> (I did check one old Opteron host) but changing this status quo is a bit
> scarier.

When msr doesn't exist, kvm_get_msr_common() returns KVM_MSR_RET_INVALID
in `default:` and kvm_complete_insn_gp() will inject #GP to guest.

Also i have wrote a kvm-unit-test, tested both on amd EPYC and intel
CascadeLake. A #GP error was printed.
Just like:

Unhandled exception 13 #GP at ip 0000000000400420
error_code=0000      rflags=00010006      cs=00000008
rax=0000000000000000 rcx=0000000000000620 rdx=00000000006164a0
rbx=0000000000009500
rbp=0000000000517490 rsi=0000000000616ae0 rdi=0000000000000001
 r8=0000000000000001  r9=00000000000003f8 r10=000000000000000d
r11=0000000000000000
r12=0000000000000000 r13=0000000000000000 r14=0000000000000000
r15=0000000000000000
cr0=0000000080000011 cr2=0000000000000000 cr3=000000000040b000
cr4=0000000000000020
cr8=0000000000000000
STACK: @400420 400338