From: sashiko-bot@kernel.org
To: "Sean Christopherson" <seanjc@google.com>
Cc: kvm@vger.kernel.org
Subject: Re: [PATCH v4 06/11] KVM: x86/xen: Punt singleshot timer hcalls to userspace if Xen vCPU ID isn't set
Date: Tue, 30 Jun 2026 23:08:52 +0000 [thread overview]
Message-ID: <20260630230852.DCE9B1F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260630225619.511632-7-seanjc@google.com>
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] KVM_XEN_VCPU_SET_ATTR unconditionally rejects the uninitialized state (XEN_VCPU_ID_INVALID) returned by KVM_XEN_VCPU_GET_ATTR, breaking live migration.
--
commit a0e30df39328c44b4229534da902c6e3bed5d228
Author: Sean Christopherson <seanjc@google.com>
KVM: x86/xen: Punt singleshot timer hcalls to userspace if Xen vCPU ID isn't set
This commit explicitly invalidates the Xen vCPU ID at creation instead of
trying to default it to the vCPU index. It also forwards singleshot timer
hypercalls to userspace if the VMM hasn't explicitly set the Xen vCPU ID.
Link: https://lore.kernel.org/all/20260612233017.1F9771F000E9@smtp.kernel.org
> diff --git a/arch/x86/kvm/xen.c b/arch/x86/kvm/xen.c
> --- a/arch/x86/kvm/xen.c
> +++ b/arch/x86/kvm/xen.c
> @@ -1103,6 +1104,8 @@ int kvm_xen_vcpu_set_attr(struct kvm_vcpu *vcpu, struct kvm_xen_vcpu_attr *data)
> break;
>
> case KVM_XEN_VCPU_ATTR_TYPE_VCPU_ID:
> + BUILD_BUG_ON(XEN_VCPU_ID_INVALID < KVM_MAX_VCPUS);
> +
> if (data->u.vcpu_id >= KVM_MAX_VCPUS)
> r = -EINVAL;
> else {
[Severity: High]
Will this break live migration by making GET and SET operations asymmetric?
Since kvm_xen_init_vcpu() now initializes the vCPU ID to XEN_VCPU_ID_INVALID:
kvm_xen_init_vcpu()
vcpu->arch.xen.vcpu_id = XEN_VCPU_ID_INVALID;
During live migration, if the VMM hasn't explicitly configured the ID, the
source VMM will read this initial state via kvm_xen_vcpu_get_attr(), receiving
the U32_MAX value.
When the destination VMM attempts to restore this exact state using
kvm_xen_vcpu_set_attr(), it appears it will be unconditionally rejected here
because U32_MAX is >= KVM_MAX_VCPUS, returning -EINVAL.
Should kvm_xen_vcpu_set_attr() be updated to explicitly allow restoring
XEN_VCPU_ID_INVALID to preserve standard save and restore loops?
--
Sashiko AI review · https://sashiko.dev/#/patchset/20260630225619.511632-1-seanjc@google.com?part=6
next prev parent reply other threads:[~2026-06-30 23:08 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-30 22:56 [PATCH v4 00/11] KVM: x86/hyperv: Fix racy usage of vcpu->arch.hyperv Sean Christopherson
2026-06-30 22:56 ` [PATCH v4 01/11] KVM: x86/hyperv: Get target FIFO in hv_tlb_flush_enqueue(), not caller Sean Christopherson
2026-06-30 22:56 ` [PATCH v4 02/11] KVM: x86/hyperv: Check for NULL vCPU Hyper-V object in kvm_hv_get_tlb_flush_fifo() Sean Christopherson
2026-07-01 15:02 ` Philippe Mathieu-Daudé
2026-06-30 22:56 ` [PATCH v4 03/11] KVM: x86/hyperv: Ensure vCPU's Hyper-V object is initialized on cross-vCPU accesses Sean Christopherson
2026-06-30 22:56 ` [PATCH v4 04/11] KVM: x86/xen: Always route non-singleshot-timer vCPU hypercalls to userspace Sean Christopherson
2026-06-30 22:56 ` [PATCH v4 05/11] KVM: x86/xen: Consolidate checks on Xen vCPU ID for singleshot timer hypercalls Sean Christopherson
2026-07-01 15:03 ` Philippe Mathieu-Daudé
2026-06-30 22:56 ` [PATCH v4 06/11] KVM: x86/xen: Punt singleshot timer hcalls to userspace if Xen vCPU ID isn't set Sean Christopherson
2026-06-30 23:08 ` sashiko-bot [this message]
2026-06-30 22:56 ` [PATCH v4 07/11] KVM: Initialize a vCPU's index to '-1' while it's being created Sean Christopherson
2026-07-01 15:06 ` Philippe Mathieu-Daudé
2026-06-30 22:56 ` [PATCH v4 08/11] KVM: Move nVMX's lockdep logic for vcpu->mutex to a common helper Sean Christopherson
2026-07-01 15:04 ` Philippe Mathieu-Daudé
2026-06-30 22:56 ` [PATCH v4 09/11] KVM: x86: Treat a vCPU as unreachable if its index is invalid Sean Christopherson
2026-07-01 15:07 ` Philippe Mathieu-Daudé
2026-06-30 22:56 ` [PATCH v4 10/11] KVM: x86/hyperv: Assert vCPU's mutex is held in to_hv_vcpu() Sean Christopherson
2026-06-30 22:56 ` [PATCH v4 11/11] KVM: x86/hyperv: Use {READ,WRITE}_ONCE for cross-task synic->active accesses Sean Christopherson
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=20260630230852.DCE9B1F000E9@smtp.kernel.org \
--to=sashiko-bot@kernel.org \
--cc=kvm@vger.kernel.org \
--cc=sashiko-reviews@lists.linux.dev \
--cc=seanjc@google.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox