Messages in this thread | | | Subject | Re: [PATCH] mm: Speed up mremap on large regions | From | Paolo Bonzini <> | Date | Fri, 12 Oct 2018 09:24:32 +0200 |
| |
On 12/10/2018 05:21, Jann Horn wrote: > I don't know how this interacts with shadow paging implementations.
Shadow paging simply uses MMU notifiers and that does not assume that PTE invalidation is atomic. The invalidate_range_start and invalidate_range_end calls are not affected by Joel's patch, so it's ok for KVM and also for other users of MMU notifiers.
Paolo
| |