[PATCH v2 1/2] KVM: X86: Processor States following Reset or INIT

[PATCH v2 1/2] KVM: X86: Processor States following Reset or INIT

[Wanpeng Li]

From: Wanpeng Li <wanpeng.li@hotmail.com>

- XCR0 is zeroed by RESET but not INIT
- XSS is zeroed by both RESET and INIT
- BNDCFGU, BND0-BND3, BNDCFGS are zeroed by both RESET and INIT

This patch does this according to SDM.

Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Jim Mattson <jmattson@google.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
---
v1 -> v2:
 * XCR0 is not zeroed by INIT
 * XSS, BNDCFGU, BND0-BND3, BNDCFGS are zeroed by both RESET and INIT

 arch/x86/kvm/vmx.c |  2 ++
 arch/x86/kvm/x86.c | 11 +++++++++++
 2 files changed, 13 insertions(+)

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 244e366..ab0f16e 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -5581,6 +5581,8 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 	vmcs_write32(GUEST_ACTIVITY_STATE, GUEST_ACTIVITY_ACTIVE);
 	vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0);
 	vmcs_writel(GUEST_PENDING_DBG_EXCEPTIONS, 0);
+	if (kvm_mpx_supported())
+		vmcs_write64(GUEST_BNDCFGS, 0);
 
 	setup_msrs(vmx);
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index b0d2915..f353825 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7804,18 +7804,29 @@ void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event)
 	kvm_async_pf_hash_reset(vcpu);
 	vcpu->arch.apf.halted = false;
 
+	if (kvm_mpx_supported()) {
+		memset(get_xsave_addr(&vcpu->arch.guest_fpu.state.xsave,
+			XFEATURE_MASK_BNDREGS), 0, sizeof(struct mpx_bndreg_state));
+		memset(get_xsave_addr(&vcpu->arch.guest_fpu.state.xsave,
+			XFEATURE_MASK_BNDCSR), 0, sizeof(u64));
+	}
+
 	if (!init_event) {
 		kvm_pmu_reset(vcpu);
 		vcpu->arch.smbase = 0x30000;
 
 		vcpu->arch.msr_platform_info = MSR_PLATFORM_INFO_CPUID_FAULT;
 		vcpu->arch.msr_misc_features_enables = 0;
+
+		vcpu->arch.xcr0 = XFEATURE_MASK_FP;
 	}
 
 	memset(vcpu->arch.regs, 0, sizeof(vcpu->arch.regs));
 	vcpu->arch.regs_avail = ~0;
 	vcpu->arch.regs_dirty = ~0;
 
+	vcpu->arch.ia32_xss = 0;
+
 	kvm_x86_ops->vcpu_reset(vcpu, init_event);
 }
 
-- 
2.7.4

[PATCH v2 2/2] KVM: VMX: Don't expose unrestricted_guest is enabled if ept is disabled

[Wanpeng Li]

From: Wanpeng Li <wanpeng.li@hotmail.com>

SDM mentioned: 

 "If either the “unrestricted guest” VM-execution control or the “mode-based 
  execute control for EPT” VM- execution control is 1, the “enable EPT” 
  VM-execution control must also be 1."

However, we can still observe unrestricted_guest is Y after inserting the kvm-intel.ko 
w/ ept=N. It depends on later starts a guest in order that the function 
vmx_compute_secondary_exec_control() can be executed, then both the module parameter 
and exec control fields will be amended.

This patch fixes it by amending module parameter immediately during vmcs data setup.

Reviewed-by: Jim Mattson <jmattson@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Jim Mattson <jmattson@google.com>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
---
v1 -> v2: 
 * reduce to just "enable_ept = 0"

 arch/x86/kvm/vmx.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index ab0f16e..dedff59 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -6730,16 +6730,13 @@ static __init int hardware_setup(void)
 	if (!cpu_has_vmx_ept() ||
 	    !cpu_has_vmx_ept_4levels() ||
 	    !cpu_has_vmx_ept_mt_wb() ||
-	    !cpu_has_vmx_invept_global()) {
+	    !cpu_has_vmx_invept_global())
 		enable_ept = 0;
-		enable_unrestricted_guest = 0;
-		enable_ept_ad_bits = 0;
-	}
 
 	if (!cpu_has_vmx_ept_ad_bits() || !enable_ept)
 		enable_ept_ad_bits = 0;
 
-	if (!cpu_has_vmx_unrestricted_guest())
+	if (!cpu_has_vmx_unrestricted_guest() || !enable_ept)
 		enable_unrestricted_guest = 0;
 
 	if (!cpu_has_vmx_flexpriority())
-- 
2.7.4

Re: [PATCH v2 2/2] KVM: VMX: Don't expose unrestricted_guest is enabled if ept is disabled

[Jim Mattson]

Looks great.

On Mon, Oct 9, 2017 at 4:00 AM, Wanpeng Li <kernellwp@gmail.com> wrote:
> From: Wanpeng Li <wanpeng.li@hotmail.com>
>
> SDM mentioned:
>
>  "If either the “unrestricted guest” VM-execution control or the “mode-based
>   execute control for EPT” VM- execution control is 1, the “enable EPT”
>   VM-execution control must also be 1."
>
> However, we can still observe unrestricted_guest is Y after inserting the kvm-intel.ko
> w/ ept=N. It depends on later starts a guest in order that the function
> vmx_compute_secondary_exec_control() can be executed, then both the module parameter
> and exec control fields will be amended.
>
> This patch fixes it by amending module parameter immediately during vmcs data setup.
>
> Reviewed-by: Jim Mattson <jmattson@google.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Jim Mattson <jmattson@google.com>
> Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
> ---
> v1 -> v2:
>  * reduce to just "enable_ept = 0"
>
>  arch/x86/kvm/vmx.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index ab0f16e..dedff59 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -6730,16 +6730,13 @@ static __init int hardware_setup(void)
>         if (!cpu_has_vmx_ept() ||
>             !cpu_has_vmx_ept_4levels() ||
>             !cpu_has_vmx_ept_mt_wb() ||
> -           !cpu_has_vmx_invept_global()) {
> +           !cpu_has_vmx_invept_global())
>                 enable_ept = 0;
> -               enable_unrestricted_guest = 0;
> -               enable_ept_ad_bits = 0;
> -       }
>
>         if (!cpu_has_vmx_ept_ad_bits() || !enable_ept)
>                 enable_ept_ad_bits = 0;
>
> -       if (!cpu_has_vmx_unrestricted_guest())
> +       if (!cpu_has_vmx_unrestricted_guest() || !enable_ept)
>                 enable_unrestricted_guest = 0;
>
>         if (!cpu_has_vmx_flexpriority())
> --
> 2.7.4
>

Re: [PATCH v2 2/2] KVM: VMX: Don't expose unrestricted_guest is enabled if ept is disabled

[Paolo Bonzini]

On 09/10/2017 13:00, Wanpeng Li wrote:
> From: Wanpeng Li <wanpeng.li@hotmail.com>
> 
> SDM mentioned: 
> 
>  "If either the “unrestricted guest” VM-execution control or the “mode-based 
>   execute control for EPT” VM- execution control is 1, the “enable EPT” 
>   VM-execution control must also be 1."
> 
> However, we can still observe unrestricted_guest is Y after inserting the kvm-intel.ko 
> w/ ept=N. It depends on later starts a guest in order that the function 
> vmx_compute_secondary_exec_control() can be executed, then both the module parameter 
> and exec control fields will be amended.
> 
> This patch fixes it by amending module parameter immediately during vmcs data setup.
> 
> Reviewed-by: Jim Mattson <jmattson@google.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Jim Mattson <jmattson@google.com>
> Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
> ---
> v1 -> v2: 
>  * reduce to just "enable_ept = 0"
> 
>  arch/x86/kvm/vmx.c | 7 ++-----
>  1 file changed, 2 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index ab0f16e..dedff59 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -6730,16 +6730,13 @@ static __init int hardware_setup(void)
>  	if (!cpu_has_vmx_ept() ||
>  	    !cpu_has_vmx_ept_4levels() ||
>  	    !cpu_has_vmx_ept_mt_wb() ||
> -	    !cpu_has_vmx_invept_global()) {
> +	    !cpu_has_vmx_invept_global())
>  		enable_ept = 0;
> -		enable_unrestricted_guest = 0;
> -		enable_ept_ad_bits = 0;
> -	}
>  
>  	if (!cpu_has_vmx_ept_ad_bits() || !enable_ept)
>  		enable_ept_ad_bits = 0;
>  
> -	if (!cpu_has_vmx_unrestricted_guest())
> +	if (!cpu_has_vmx_unrestricted_guest() || !enable_ept)
>  		enable_unrestricted_guest = 0;
>  
>  	if (!cpu_has_vmx_flexpriority())
> 

Queued both, thanks.

Paolo