Re: [PATCH v3] KVM: arm64: nv: Keep the shadow S2 MMUs at fixed addresses

On Fri, 07 Aug 2026 02:13:17 +0100,
Karl Mehltretter <[email protected]> wrote:
> 
> On Thu, Aug 06, 2026 at 10:54:03PM +0100, Wei-Lin Chang wrote:
> > 
> > Hey sorry for nitpicking once more, I actually like v2 more, as this
> > version splits the error handling to two places. Marc's implementation
> > [1] is also good, but that requires a twist in kvm_arch_flush_shadow_all().
> > So I'd personally go with v2 (+my suggestions for it :)), but also fine if
> > others have other opinions.
> > 
> > I also think the comment in kvm_vcpu_init_nested() is worth adjusting
> > slightly, too.
> > 
> 
> No worries, thanks for taking another look.
> 
> I was following the guidance in include/linux/cleanup.h:
> 
>   the expectation is that usage of "goto" and cleanup helpers is never
>   mixed in the same function.
> 
> But maybe keeping goto is better. I'll wait a bit for others
> to comment.

The version I proposed in response to your first version had no goto,
and trivial error handling. Can we *please* stop the bike shedding and
settle on the simplest possible fix? Or *any* fix at all?

> 
> For the comment, I propose replacing the current text with:
> 
> 	/*
> 	 * If allocating an MMU object fails, keep the VNCR page alive so
> 	 * that userspace can retry vCPU initialisation without affecting
> 	 * the VM.
> 	 */
> 
> Is that the adjustment you had in mind?

I don't think we need to change any comment. If anything, it should be
deleted altogether. This is a security fix, not a literary exercise.

	M.

-- 
Jazz isn't dead. It just smells funny.