Re: [PATCH v2 00/10] KVM: Consolidate and optimize MMU notifiers

2021-04-12 Thread Marc Zyngier
On Fri, 02 Apr 2021 13:17:45 +0100, Paolo Bonzini wrote: > > On 02/04/21 02:56, Sean Christopherson wrote: > > The end goal of this series is to optimize the MMU notifiers to take > > mmu_lock if and only if the notification is relevant to KVM, i.e. the hva > > range overlaps a memslot. Large V

Re: [PATCH v2 00/10] KVM: Consolidate and optimize MMU notifiers

2021-04-02 Thread Paolo Bonzini
On 02/04/21 02:56, Sean Christopherson wrote: The end goal of this series is to optimize the MMU notifiers to take mmu_lock if and only if the notification is relevant to KVM, i.e. the hva range overlaps a memslot. Large VMs (hundreds of vCPUs) are very sensitive to mmu_lock being taken for wri

[PATCH v2 00/10] KVM: Consolidate and optimize MMU notifiers

2021-04-01 Thread Sean Christopherson
The end goal of this series is to optimize the MMU notifiers to take mmu_lock if and only if the notification is relevant to KVM, i.e. the hva range overlaps a memslot. Large VMs (hundreds of vCPUs) are very sensitive to mmu_lock being taken for write at inopportune times, and such VMs also tend