Re: [PATCH v1 06/23] KVM: VMX: Defer enabling FRED MSRs save/load until after set CPUID

[Chao Gao <chao.gao@intel.com>]

On Wed, Nov 08, 2023 at 10:29:46AM -0800, Xin Li wrote:
>Clear FRED VM entry/exit controls when initializing a vCPU, and set
>these controls only if FRED is enumerated after set CPUID.
>
>FRED VM entry/exit controls need to be set to establish context
>sufficient to support FRED event delivery immediately after VM entry
>and exit.  However it is not required to save/load FRED MSRs for
>a non-FRED guest, which aren't supposed to access FRED MSRs.
>
>A non-FRED guest should get #GP upon accessing FRED MSRs, otherwise
>it corrupts host FRED MSRs.
>
>Tested-by: Shan Kang <shan.kang@intel.com>
>Signed-off-by: Xin Li <xin3.li@intel.com>
>---
> arch/x86/kvm/vmx/vmx.c | 34 +++++++++++++++++++++++++++++++++-
> 1 file changed, 33 insertions(+), 1 deletion(-)
>
>diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
>index 9186f41974ab..5d4786812664 100644
>--- a/arch/x86/kvm/vmx/vmx.c
>+++ b/arch/x86/kvm/vmx/vmx.c
>@@ -4423,6 +4423,9 @@ static u32 vmx_vmentry_ctrl(void)
> 	if (cpu_has_perf_global_ctrl_bug())
> 		vmentry_ctrl &= ~VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL;
> 
>+	/* Whether to load guest FRED MSRs is deferred until after set CPUID */
>+	vmentry_ctrl &= ~VM_ENTRY_LOAD_IA32_FRED;
>+
> 	return vmentry_ctrl;
> }
> 
>@@ -4458,7 +4461,13 @@ static u32 vmx_vmexit_ctrl(void)
> 
> static u64 vmx_secondary_vmexit_ctrl(void)
> {
>-	return vmcs_config.secondary_vmexit_ctrl;
>+	u64 secondary_vmexit_ctrl = vmcs_config.secondary_vmexit_ctrl;
>+
>+	/* Whether to save/load FRED MSRs is deferred until after set CPUID */
>+	secondary_vmexit_ctrl &= ~(SECONDARY_VM_EXIT_SAVE_IA32_FRED |
>+				   SECONDARY_VM_EXIT_LOAD_IA32_FRED);
>+
>+	return secondary_vmexit_ctrl;
> }
> 
> static void vmx_refresh_apicv_exec_ctrl(struct kvm_vcpu *vcpu)
>@@ -7785,10 +7794,33 @@ static void update_intel_pt_cfg(struct kvm_vcpu *vcpu)
> 		vmx->pt_desc.ctl_bitmask &= ~(0xfULL << (32 + i * 4));
> }
> 
>+static void vmx_vcpu_config_fred_after_set_cpuid(struct kvm_vcpu *vcpu)
>+{
>+	struct vcpu_vmx *vmx = to_vmx(vcpu);
>+
>+	if (!cpu_feature_enabled(X86_FEATURE_FRED) ||
>+	    !guest_cpuid_has(vcpu, X86_FEATURE_FRED))
>+		return;
>+
>+	/* Enable loading guest FRED MSRs from VMCS */
>+	vm_entry_controls_setbit(vmx, VM_ENTRY_LOAD_IA32_FRED);
>+
>+	/*
>+	 * Enable saving guest FRED MSRs into VMCS and loading host FRED MSRs
>+	 * from VMCS.
>+	 */
>+	vm_exit_controls_setbit(vmx, VM_EXIT_ACTIVATE_SECONDARY_CONTROLS);
>+	secondary_vm_exit_controls_setbit(vmx,
>+					  SECONDARY_VM_EXIT_SAVE_IA32_FRED |
>+					  SECONDARY_VM_EXIT_LOAD_IA32_FRED);

all above vmcs controls need to be cleared if guest doesn't enumerate FRED, see

https://lore.kernel.org/all/ZJYzPn7ipYfO0fLZ@google.com/

Clearing VM_EXIT_ACTIVATE_SECONDARY_CONTROLS may be problematic when
new bits are added to secondary vmcs controls. Why not keep
VM_EXIT_ACTIVATE_SECONDARY_CONTROLS always on if it is supported? or you see
any perf impact?

>+}
>+
> static void vmx_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
> {
> 	struct vcpu_vmx *vmx = to_vmx(vcpu);
> 
>+	vmx_vcpu_config_fred_after_set_cpuid(vcpu);
>+
> 	/*
> 	 * XSAVES is effectively enabled if and only if XSAVE is also exposed
> 	 * to the guest.  XSAVES depends on CR4.OSXSAVE, and CR4.OSXSAVE can be
>-- 
>2.42.0
>
>