Re: [PATCH v10 09/41] KVM: guest_memfd: Filter both shared and private when invalidating

Next message: Gregory Price: "Re: [PATCH] mm/memcontrol: remove unused memcg parameter in calculate_high_delay()"
Previous message: Chen, Yu C: "Re: [PATCH v6 5/9] x86/resctrl: Parse ACPI CMRC table"
In reply to: Sean Christopherson: "Re: [PATCH v10 09/41] KVM: guest_memfd: Filter both shared and private when invalidating"
Next in thread: Michael Roth: "Re: [PATCH v10 09/41] KVM: guest_memfd: Filter both shared and private when invalidating"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

From: Xiaoyao Li

Date: Tue Aug 25 2026 - 12:15:36 EST

On 8/20/2026 9:32 AM, Sean Christopherson wrote:

And vice versa, a PUNCH_HOLE on a SHARED gmem (if userspace is
using an INIT_SHARED gmem for the shared branch of a memslot) could invalidate the
PRIVATE mappings (of a different gmem instance).

I'm wondering now how this could happen.

the requirement for PUNCH_HOLE on gmem to trigger mapping invalidation is the gmem is bound with the memslot. But how can a memslot bound with two gmem instances?