Re: [PATCH 4/4] VMX: x86: Only reset MMU when necessary

[Avi Kivity <avi@redhat.com>]

On 05/12/2010 09:33 AM, Sheng Yang wrote:
> Only modifying some bits of CR0/CR4 needs paging mode switch.
>
> Add update_rsvd_bits_mask() to address EFER.NX bit updating for reserved bits.
>
>
> @@ -2335,6 +2335,19 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level)
>   	}
>   }
>
> +void update_rsvd_bits_mask(struct kvm_vcpu *vcpu)
> +{
> +	if (!is_paging(vcpu))
> +		return;
> +	if (is_long_mode(vcpu))
> +		reset_rsvds_bits_mask(vcpu, PT64_ROOT_LEVEL);
> +	else if (is_pae(vcpu))
> +		reset_rsvds_bits_mask(vcpu, PT32E_ROOT_LEVEL);
> +	else
> +		reset_rsvds_bits_mask(vcpu, PT32_ROOT_LEVEL);
> +}
> +EXPORT_SYMBOL_GPL(update_rsvd_bits_mask);
>    

Needs a kvm_ prefix if made a global symbol.  But isn't nx switching 
rare enough so we can reload the mmu completely?

> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index b59fc67..971a295 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -416,6 +416,10 @@ out:
>
>   static int __kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
>   {
> +	unsigned long old_cr0 = kvm_read_cr0(vcpu);
> +	unsigned long update_bits = X86_CR0_PG | X86_CR0_PE |
> +				    X86_CR0_CD | X86_CR0_NW;
>    

PE doesn't affect paging, CD, NW don't either?

What about WP?

> +
>   	cr0 |= X86_CR0_ET;
>
>   #ifdef CONFIG_X86_64
> @@ -449,7 +453,8 @@ static int __kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0)
>
>   	kvm_x86_ops->set_cr0(vcpu, cr0);
>
> -	kvm_mmu_reset_context(vcpu);
> +	if ((cr0 ^ old_cr0)&  update_bits)
> +		kvm_mmu_reset_context(vcpu);
>   	return 0;
>   }
>
>
> @@ -692,6 +698,8 @@ static u32 emulated_msrs[] = {
>
>   static int set_efer(struct kvm_vcpu *vcpu, u64 efer)
>   {
> +	u64 old_efer = vcpu->arch.efer;
> +
>   	if (efer&  efer_reserved_bits)
>   		return 1;
>
> @@ -722,6 +730,9 @@ static int set_efer(struct kvm_vcpu *vcpu, u64 efer)
>
>   	vcpu->arch.mmu.base_role.nxe = (efer&  EFER_NX)&&  !tdp_enabled;
>
> +	if ((efer ^ old_efer)&  EFER_NX)
> +		update_rsvd_bits_mask(vcpu);
> +
>   	return 0;
>   }
>    

I think it's fine to reset the entire mmu context here, most guests 
won't toggle nx all the time.  But it needs to be in patch 3, otherwise 
we have a regression between 3 and 4.

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.