Re: [PATCH AUTOSEL 6.18-6.6] KVM: x86/xen: Prevent runstate times from becoming negative

[David Woodhouse]

On Wed, 2026-10-21 at 13:17 -0700, Sean Christopherson wrote:
> 
> 
> Two questions: do we care abou this specific flaw, and does anyone run Xen guests
> with KVM using an LTS kernel?  I agree with AI's assessment that this is low risk,
> but unless this will actually benefit someone, I don't see the point.

I concur. Nobody runs KVM and expects steal time and clocks not to be
entirely hosed once you start caring deeply about the details — even in
upstream. You want this stuff to work, you carefully curate the kernel
you're running.

Backporting this one to LTS kernels doesn't help anyone. The risk is
low, but the benefit is lower.

Those who care will already have it backported to 6.18 anyway.

Attachment: smime.p7s
Description: S/MIME cryptographic signature