Re: [PATCH 1/2] KVM: x86/mmu: Avoid multiple hash lookups in kvm_get_mmu_page()

2020-06-23 Thread Jon Cargille
LGTM. Reviewed-By: Jon Cargille On Tue, Jun 23, 2020 at 12:40 PM Sean Christopherson wrote: > > Refactor for_each_valid_sp() to take the list of shadow pages instead of > retrieving it from a gfn to avoid doing the gfn->list hash and lookup > multiple times during kvm_get_mmu

Re: [PATCH 2/2] KVM: x86/mmu: Optimize MMU page cache lookup for fully direct MMUs

2020-06-23 Thread Jon Cargille
LGTM. Reviewed-By: Jon Cargille On Tue, Jun 23, 2020 at 12:40 PM Sean Christopherson wrote: > > Skip the unsync checks and the write flooding clearing for fully direct > MMUs, which are guaranteed to not have unsync'd or indirect pages (write > flooding detection only ap

[PATCH] kvm: x86 mmu: avoid mmu_page_hash lookup for direct_map-only VM

2020-05-08 Thread Jon Cargille
cking if the SP already exists. This optimization has proven significant in batch large SP shattering where the hash lookup accounted for 95% of the overhead. Signed-off-by: Peter Feiner Signed-off-by: Jon Cargille Reviewed-by: Jim Mattson --- arch/x86/include/asm/kvm_host.h | 13

[PATCH] kvm: add halt-polling cpu usage stats

2020-05-08 Thread Jon Cargille
low at ~500 years, which seems reasonably large. Signed-off-by: David Matlack Signed-off-by: Jon Cargille Reviewed-by: Jim Mattson --- arch/x86/include/asm/kvm_host.h | 2 ++ arch/x86/kvm/x86.c | 2 ++ virt/kvm/kvm_main.c | 20 +--- 3 files changed,