mbox series

[v3,0/2] KVM: x86/mmu: Skip mmu_notifier changes when possible

Message ID 20210128060515.1732758-1-stevensd@google.com (mailing list archive)
Headers show
Series KVM: x86/mmu: Skip mmu_notifier changes when possible | expand

Message

David Stevens Jan. 28, 2021, 6:05 a.m. UTC 
These patches reduce how often mmu_notifier updates block guest page
faults. The primary benefit of this is the reduction in the likelihood
of extreme latency when handling a page fault due to another thread
having been preempted while modifying host virtual addresses.

v2 -> v3:
 - Added patch to skip check for MMIO page faults
 - Style changes

David Stevens (1):
  KVM: x86/mmu: Consider the hva in mmu_notifier retry

Sean Christopherson (1):
  KVM: x86/mmu: Skip mmu_notifier check when handling MMIO page fault

 arch/powerpc/kvm/book3s_64_mmu_hv.c    |  2 +-
 arch/powerpc/kvm/book3s_64_mmu_radix.c |  2 +-
 arch/x86/kvm/mmu/mmu.c                 | 16 ++++++++------
 arch/x86/kvm/mmu/paging_tmpl.h         |  7 ++++---
 include/linux/kvm_host.h               | 25 +++++++++++++++++++++-
 virt/kvm/kvm_main.c                    | 29 ++++++++++++++++++++++----
 6 files changed, 65 insertions(+), 16 deletions(-)

Comments

Paolo Bonzini Jan. 28, 2021, 12:48 p.m. UTC | #1 
On 28/01/21 07:05, David Stevens wrote:
> These patches reduce how often mmu_notifier updates block guest page
> faults. The primary benefit of this is the reduction in the likelihood
> of extreme latency when handling a page fault due to another thread
> having been preempted while modifying host virtual addresses.
> 
> v2 -> v3:
>   - Added patch to skip check for MMIO page faults
>   - Style changes
> 
> David Stevens (1):
>    KVM: x86/mmu: Consider the hva in mmu_notifier retry
> 
> Sean Christopherson (1):
>    KVM: x86/mmu: Skip mmu_notifier check when handling MMIO page fault
> 
>   arch/powerpc/kvm/book3s_64_mmu_hv.c    |  2 +-
>   arch/powerpc/kvm/book3s_64_mmu_radix.c |  2 +-
>   arch/x86/kvm/mmu/mmu.c                 | 16 ++++++++------
>   arch/x86/kvm/mmu/paging_tmpl.h         |  7 ++++---
>   include/linux/kvm_host.h               | 25 +++++++++++++++++++++-
>   virt/kvm/kvm_main.c                    | 29 ++++++++++++++++++++++----
>   6 files changed, 65 insertions(+), 16 deletions(-)
> 

Queued, thanks.

Paolo
David Stevens Feb. 5, 2021, 5:58 a.m. UTC | #2 
These patches might be responsible for some instability in one of our
stress tests. I'll send an update once I figure out what's going on.

Thanks,
David

On Thu, Jan 28, 2021 at 9:48 PM Paolo Bonzini <pbonzini@redhat.com> wrote:
>
> On 28/01/21 07:05, David Stevens wrote:
> > These patches reduce how often mmu_notifier updates block guest page
> > faults. The primary benefit of this is the reduction in the likelihood
> > of extreme latency when handling a page fault due to another thread
> > having been preempted while modifying host virtual addresses.
> >
> > v2 -> v3:
> >   - Added patch to skip check for MMIO page faults
> >   - Style changes
> >
> > David Stevens (1):
> >    KVM: x86/mmu: Consider the hva in mmu_notifier retry
> >
> > Sean Christopherson (1):
> >    KVM: x86/mmu: Skip mmu_notifier check when handling MMIO page fault
> >
> >   arch/powerpc/kvm/book3s_64_mmu_hv.c    |  2 +-
> >   arch/powerpc/kvm/book3s_64_mmu_radix.c |  2 +-
> >   arch/x86/kvm/mmu/mmu.c                 | 16 ++++++++------
> >   arch/x86/kvm/mmu/paging_tmpl.h         |  7 ++++---
> >   include/linux/kvm_host.h               | 25 +++++++++++++++++++++-
> >   virt/kvm/kvm_main.c                    | 29 ++++++++++++++++++++++----
> >   6 files changed, 65 insertions(+), 16 deletions(-)
> >
>
> Queued, thanks.
>
> Paolo
>