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;
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
From:   Andy Lutomirski <luto@amacapital.net>
Mime-Version: 1.0 (1.0)
Subject: Re: FSGSBASE causing panic on 5.9-rc1
Date:   Thu, 20 Aug 2020 13:36:46 -0700
Message-Id: <5756198D-C8BD-4290-BFCA-04424EB230A6@amacapital.net>
References: <f6694ee6-d672-1cf9-deaf-4d600bc4e9eb@amd.com>
Cc:     Dave Hansen <dave.hansen@intel.com>,
        Jim Mattson <jmattson@google.com>,
        Andy Lutomirski <luto@kernel.org>,
        Sean Christopherson <sean.j.christopherson@intel.com>,
        Joerg Roedel <joro@8bytes.org>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        X86 ML <x86@kernel.org>,
        "Chang S. Bae" <chang.seok.bae@intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Sasha Levin <sashal@kernel.org>,
        Borislav Petkov <bp@alien8.de>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@kernel.org>
In-Reply-To: <f6694ee6-d672-1cf9-deaf-4d600bc4e9eb@amd.com>
To:     Tom Lendacky <thomas.lendacky@amd.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On Aug 20, 2020, at 1:15 PM, Tom Lendacky <thomas.lendacky@amd.com> wrote:=

>=20
> =EF=BB=BFOn 8/20/20 3:07 PM, Dave Hansen wrote:
>> On 8/20/20 12:05 PM, Tom Lendacky wrote:
>>>> I added a quick hack to save TSC_AUX to a new variable in the SVM
>>>> struct and then restore it right after VMEXIT (just after where GS is
>>>> restored in svm_vcpu_enter_exit()) and my guest is no longer crashing.
>>>=20
>>> Sorry, I mean my host is no longer crashing.
>> Just to make sure I've got this:
>> 1. Older CPUs didn't have X86_FEATURE_RDPID
>> 2. FSGSBASE patches started using RDPID in the NMI entry path when
>>    supported *AND* FSGSBASE was enabled
>> 3. There was a latent SVM bug which did not restore the RDPID data
>>    before NMIs were reenabled after VMEXIT
>> 4. If an NMI comes in the window between VMEXIT and the
>>    wrmsr(TSC_AUX)... boom
>=20
> Right, which means that the setting of TSC_AUX to the guest value needs to=
 be moved, too.
>=20

Depending on how much of a perf hit this is, we could also skip using RDPID i=
n the paranoid path on SVM-capable CPUs.=