linux-kernel - Re: [PATCH 0/2] KVM: kvm-coco-queue: Support protected TSC

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <c6cea2d0-80b5-4d05-84a7-0dc25c219d1d@redhat.com>
Date: Wed, 12 Mar 2025 13:22:24 +0100
From: Paolo Bonzini <pbonzini@...hat.com>
To: Isaku Yamahata <isaku.yamahata@...el.com>, kvm@...r.kernel.org
Cc: Sean Christopherson <seanjc@...gle.com>, chao.gao@...el.com,
 rick.p.edgecombe@...el.com, yan.y.zhao@...el.com,
 linux-kernel@...r.kernel.org, isaku.yamahata@...il.com,
 Nikunj A Dadhania <nikunj@....com>, Marcelo Tosatti <mtosatti@...hat.com>
Subject: Re: [PATCH 0/2] KVM: kvm-coco-queue: Support protected TSC

On 10/12/24 09:55, Isaku Yamahata wrote:
> The current x86 KVM implementation conflicts with protected TSC because the
> VMM can't change the TSC offset/multiplier.  Disable or ignore the KVM
> logic to change/adjust the TSC offset/multiplier somehow.
> 
> Because KVM emulates the TSC timer or the TSC deadline timer with the TSC
> offset/multiplier, the TSC timer interrupts are injected to the guest at the
> wrong time if the KVM TSC offset is different from what the TDX module
> determined.
> 
> Originally the issue was found by cyclic test of rt-test [1] as the latency in
> TDX case is worse than VMX value + TDX SEAMCALL overhead.  It turned out that
> the KVM TSC offset is different from what the TDX module determines.
> 
> The solution is to keep the KVM TSC offset/multiplier the same as the value of
> the TDX module somehow. [...] Ignore (or don't call related functions) the
> request to change the TSC offset/multiplier.
> 
> [...]  With this patch series, SEV-SNP secure TSC can be supported.

Thanks, I've squashed these changes (apart from setting
vcpu->arch.guest_tsc_protected) into the corresponding patches in
kvm-coco-queue.  Just one small change is needed in patch 2, to
which I will reply.

For SEV-SNP, all that's necessary on top should be

diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index d92e97baea0f..beddeed90ff0 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -2481,6 +2481,9 @@ static int snp_launch_update_vmsa(struct kvm *kvm, struct kvm_sev_cmd *argp)
  		}
  
  		svm->vcpu.arch.guest_state_protected = true;
+		if (snp_secure_tsc_enabled(kvm))
+			svm->vcpu.arch.guest_tsc_protected = true;
+
  		/*
  		 * SEV-ES (and thus SNP) guest mandates LBR Virtualization to
  		 * be _always_ ON. Enable it only after setting

For the sake of testing, I applied the latest SEV-SNP host patches
from https://github.com/AMDESE/linux-kvm/commits/sectsc-host-latest
to kvm-coco-queue as well, plus the above hunk; Nikunj can integrate
it in the next revision of
https://lore.kernel.org/kvm/20250310064347.13986-1-nikunj@amd.com/T/.

Paolo