Re: [PATCH v4 0/4] KVM: arm64: Fix underallocation of storage for SVE state

[Mark Brown]

On Mon, Jul 08, 2024 at 04:30:30PM +0100, Dave Martin wrote:

> One thing confuses me:

> The host could never use over-max VLs except in non-preemptible kernel
> code, since code doing that would be non-migratable to other physical
> CPUs.  This is done to probe SVE only, and the extra bits in the vector
> registers are never used at all.

> Can't pKVM just hide the non symmetrically supported VLs using ZCR_EL2,
> just as regular KVM does for the guest?

> (I may be making bad assumptions about pKVM's relationship with the host
> kernel.)

That'd be another way to do it, constrain the VLs the host can set - I
assume there's something about how pKVM or hVHE does things that makes
problems for that.

Attachment: signature.asc
Description: PGP signature