2015-12-18 09:42:17

by Takuya Yoshikawa

Subject: [PATCH] KVM: x86: MMU: Use clear_page() instead of init_shadow_page_table()

Not just in order to clean up the code, but to make it faster by using
enhanced instructions: the initialization became 20-30% faster on our
testing machine.

Signed-off-by: Takuya Yoshikawa <[email protected]>
---
arch/x86/kvm/mmu.c | 10 +---------
1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index a1a3d19..7f5a82b 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -2041,14 +2041,6 @@ static void mmu_sync_children(struct kvm_vcpu *vcpu,
}
}

-static void init_shadow_page_table(struct kvm_mmu_page *sp)
-{
- int i;
-
- for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
- sp->spt[i] = 0ull;
-}
-
static void __clear_sp_write_flooding_count(struct kvm_mmu_page *sp)
{
sp->write_flooding_count = 0;
@@ -2128,7 +2120,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
account_shadowed(vcpu->kvm, sp);
}
sp->mmu_valid_gen = vcpu->kvm->arch.mmu_valid_gen;
- init_shadow_page_table(sp);
+ clear_page(sp->spt);
trace_kvm_mmu_get_page(sp, true);
return sp;
}
--
2.1.0

2015-12-18 18:08:36

by Paolo Bonzini

Subject: Re: [PATCH] KVM: x86: MMU: Use clear_page() instead of init_shadow_page_table()

On 18/12/2015 10:54, Takuya Yoshikawa wrote:
> Not just in order to clean up the code, but to make it faster by using
> enhanced instructions: the initialization became 20-30% faster on our
> testing machine.
>
> Signed-off-by: Takuya Yoshikawa <[email protected]>

Applied locally, but I'm not sure when it will appear on kernel.org...
it depends on the amount of testing I can do next week before the
holidays. Anyway, it's not lost. :)

Paolo

> ---
> arch/x86/kvm/mmu.c | 10 +---------
> 1 file changed, 1 insertion(+), 9 deletions(-)
>
> diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
> index a1a3d19..7f5a82b 100644
> --- a/arch/x86/kvm/mmu.c
> +++ b/arch/x86/kvm/mmu.c
> @@ -2041,14 +2041,6 @@ static void mmu_sync_children(struct kvm_vcpu *vcpu,
> }
> }
>
> -static void init_shadow_page_table(struct kvm_mmu_page *sp)
> -{
> - int i;
> -
> - for (i = 0; i < PT64_ENT_PER_PAGE; ++i)
> - sp->spt[i] = 0ull;
> -}
> -
> static void __clear_sp_write_flooding_count(struct kvm_mmu_page *sp)
> {
> sp->write_flooding_count = 0;
> @@ -2128,7 +2120,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu,
> account_shadowed(vcpu->kvm, sp);
> }
> sp->mmu_valid_gen = vcpu->kvm->arch.mmu_valid_gen;
> - init_shadow_page_table(sp);
> + clear_page(sp->spt);
> trace_kvm_mmu_get_page(sp, true);
> return sp;
> }
>