All of lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Yosry Ahmed" <yosry@kernel.org>
Cc: kvm@vger.kernel.org
Subject: Re: [PATCH v3 13/27] KVM: nSVM: Add a placeholder ASID for L2
Date: Tue, 13 Oct 2026 00:01:15 +0000	[thread overview]
Message-ID: <sashiko-outbox-168763@kernel.org> (raw)
In-Reply-To: <20261012233638.2546339-14-yosry@kernel.org>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] kvm: nSVM: missing vmcb_mark_dirty() when updating L2 ASID in SEV migration
--

commit c8f241845a6790f522df58195167e95ce8fea046
Author: Yosry Ahmed <yosry@kernel.org>

KVM: nSVM: Add a placeholder ASID for L2

This commit adds a placeholder ASID for L2 to prepare for future changes
where L1 and L2 will use different ASIDs. It also updates the SEV migration
path to copy the new ASID into asid02 and vmcb02 if nested is already
initialized on the destination vCPU.

> diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
> index 99577859e1aab..b2e6acfbb081f 100644
> --- a/arch/x86/kvm/svm/sev.c
> +++ b/arch/x86/kvm/svm/sev.c
[ ... ]
> @@ -2041,6 +2041,15 @@ static void sev_vcpu_migrate_asid(struct vcpu_svm *dst_svm,
>  	 */
>  	free_asid(dst_svm->asid);
>  	dst_svm->asid = asid;
> +
> +	/*
> +	 * If nested is already initialized on the destination vCPU, also update
> +	 * the nested ASID to match the new SEV ASID.
> +	 */
> +	if (dst_svm->nested.vmcb02.ptr) {
> +		dst_svm->nested.asid02 = asid;
> +		dst_svm->nested.vmcb02.ptr->control.asid = asid;

[Severity: High]
Does this code need a call to vmcb_mark_dirty(dst_svm->nested.vmcb02.ptr, VMCB_ASID)?

If userspace creates a non-SEV VM, runs it in nested SVM (allocating vmcb02
and caching it on a physical CPU), and then uses it as the destination for
an intra-host migration, the migration path updates the ASID here.

If the vCPU is scheduled back onto the same physical CPU, pre_svm_run() skips
vmcb_mark_all_dirty(). Since the VMCB_ASID clean bit might still be set from
the last exit from L2, could the CPU ignore the new ASID in memory and
incorrectly reuse the old non-SEV ASID from its internal VMCB cache?

> +	}
>  }
>  
>  static void sev_migrate_from(struct kvm *dst_kvm, struct kvm *src_kvm)

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261012233638.2546339-1-yosry@kernel.org?part=13

  reply	other threads:[~2026-10-13  0:01 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-12 23:36 [PATCH v3 00/27] KVM: nSVM: Optimize nSVM TLB flushes Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 01/27] KVM: nSVM: Flush the TLB after forcefully leaving nested Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 02/27] KVM: SVM: Document number of ASIDs CPUID setting Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 03/27] KVM: VMX: Generalize VPID allocation to be vendor-neutral Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 04/27] KVM: x86/mmu: Support specifying reserved TLB tags Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 05/27] KVM: SVM: Add helpers to set/clear ASID flush in VMCB Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 06/27] KVM: SVM: Fallback to flush everything if FLUSHBYASID is not available Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 07/27] KVM: SEV: Do ASID initialization at VMCB initialization Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 08/27] KVM: SEV: Expose sev_get_asid() outside of sev.c Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 09/27] KVM: SEV: Explicitly initialize the per vCPU ASID on SEV VM migration Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 10/27] KVM: SVM: Use a static ASID per vCPU Yosry Ahmed
2026-10-12 23:54   ` sashiko-bot
2026-10-12 23:55     ` Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 11/27] KVM: SVM: Only flush the fallback ASID when used by a different vCPU Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 12/27] KVM: nSVM: Drop svm->nested.initialized Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 13/27] KVM: nSVM: Add a placeholder ASID for L2 Yosry Ahmed
2026-10-13  0:01   ` sashiko-bot [this message]
2026-10-13 14:25     ` Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 14/27] KVM: x86: hyper-v: Rename kvm_hv_vcpu_purge_flush_tlb() Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 15/27] KVM: x86: hyper-v: Allow purging all TLB flush FIFOs Yosry Ahmed
2026-10-12 23:56   ` sashiko-bot
2026-10-13 14:42     ` Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 16/27] KVM: nSVM: Flush both L1 and L2 ASIDs on KVM_REQ_TLB_FLUSH Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 17/27] KVM: nSVM: Always switch VMCB before leaving guest mode Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 18/27] KVM: nSVM: Split nested_svm_transition_tlb_flush() into entry/exit fns Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 19/27] KVM: nSVM: Service local TLB flushes before nested transitions Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 20/27] KVM: x86: Add KVM_REQ_MMU_SYNC_ALL_ROOTS Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 21/27] KVM: nSVM: Handle nested TLB flush requests through TLB_CONTROL Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 22/27] KVM: nSVM: Flush the TLB if L1 changes L2's ASID in vmcb12 Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 23/27] KVM: nSVM: Do not reset TLB_CONTROL in vmcb02 on nested VM-Enter Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 24/27] KVM: nSVM: Flush L2's ASID when emulating INVLPGA Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 25/27] KVM: nSVM: Flush the ASID on nested transitions if shared by L1 and L2 Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 26/27] KVM: nSVM: Use different ASIDs for " Yosry Ahmed
2026-10-12 23:36 ` [PATCH v3 27/27] KVM: selftests: Add a test for nested TLB flushes Yosry Ahmed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=sashiko-outbox-168763@kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=kvm@vger.kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    --cc=yosry@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.