public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: kvm@vger.kernel.org
Subject: [RFC PATCH v6 16/36] KVM: nVMX: Support PERF_GLOBAL_CTRL with enlightened VMCS
Date: Wed, 24 Aug 2022 03:01:18 +0000	[thread overview]
Message-ID: <20220824030138.3524159-17-seanjc@google.com> (raw)
In-Reply-To: <20220824030138.3524159-1-seanjc@google.com>

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Enlightened VMCS v1 got updated and now includes the required fields
for loading PERF_GLOBAL_CTRL upon VMENTER/VMEXIT features. For KVM on
Hyper-V enablement, KVM can just observe VMX control MSRs and use the
features (with or without eVMCS) when possible.

Hyper-V on KVM is messier as Windows 11 guests fail to boot if the
controls are advertised and a new PV feature flag, CPUID.0x4000000A.EBX
BIT(0), is not set.  Honor the Hyper-V CPUID feature flag to play nice
with Windows guests.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/hyperv.c    |  2 +-
 arch/x86/kvm/vmx/evmcs.c | 32 ++++++++++++++++++++++++++++++++
 arch/x86/kvm/vmx/evmcs.h |  7 ++-----
 3 files changed, 35 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index a7478b61088b..0adf4a437e85 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -2546,7 +2546,7 @@ int kvm_get_hv_cpuid(struct kvm_vcpu *vcpu, struct kvm_cpuid2 *cpuid,
 		case HYPERV_CPUID_NESTED_FEATURES:
 			ent->eax = evmcs_ver;
 			ent->eax |= HV_X64_NESTED_MSR_BITMAP;
-
+			ent->ebx |= HV_X64_NESTED_EVMCS1_PERF_GLOBAL_CTRL;
 			break;
 
 		case HYPERV_CPUID_SYNDBG_VENDOR_AND_MAX_FUNCTIONS:
diff --git a/arch/x86/kvm/vmx/evmcs.c b/arch/x86/kvm/vmx/evmcs.c
index bd1dcc077c85..38ec41939cab 100644
--- a/arch/x86/kvm/vmx/evmcs.c
+++ b/arch/x86/kvm/vmx/evmcs.c
@@ -442,6 +442,23 @@ u64 nested_evmcs_get_unsupported_ctrls(struct vcpu_vmx *vmx, u32 msr_index)
 	return 0;
 }
 
+static bool evmcs_has_perf_global_ctrl(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
+
+	/*
+	 * PERF_GLOBAL_CTRL has a quirk where some Windows guests may fail to
+	 * boot if a PV CPUID feature flag is not also set.  Treat the fields
+	 * as unsupported if the flag is not set in guest CPUID.  This should
+	 * be called only for guest accesses, and all guest accesses should be
+	 * gated on Hyper-V being enabled and initialized.
+	 */
+	if (WARN_ON_ONCE(!hv_vcpu))
+		return false;
+
+	return hv_vcpu->cpuid_cache.nested_ebx & HV_X64_NESTED_EVMCS1_PERF_GLOBAL_CTRL;
+}
+
 void nested_evmcs_filter_control_msr(struct kvm_vcpu *vcpu,
 				     struct msr_data *msr_info)
 {
@@ -455,6 +472,21 @@ void nested_evmcs_filter_control_msr(struct kvm_vcpu *vcpu,
 		return;
 
 	unsupported_ctrls = nested_evmcs_get_unsupported_ctrls(vmx, msr_info->index);
+	switch (msr_info->index) {
+	case MSR_IA32_VMX_EXIT_CTLS:
+	case MSR_IA32_VMX_TRUE_EXIT_CTLS:
+		if (!evmcs_has_perf_global_ctrl(vcpu))
+			unsupported_ctrls |= VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL;
+		break;
+	case MSR_IA32_VMX_ENTRY_CTLS:
+	case MSR_IA32_VMX_TRUE_ENTRY_CTLS:
+		if (!evmcs_has_perf_global_ctrl(vcpu))
+			unsupported_ctrls |= VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL;
+		break;
+	default:
+		break;
+	}
+
 	if (msr_info->index == MSR_IA32_VMX_VMFUNC)
 		msr_info->data &= ~unsupported_ctrls;
 	else
diff --git a/arch/x86/kvm/vmx/evmcs.h b/arch/x86/kvm/vmx/evmcs.h
index e2b3aeee57ac..35b326386c50 100644
--- a/arch/x86/kvm/vmx/evmcs.h
+++ b/arch/x86/kvm/vmx/evmcs.h
@@ -43,8 +43,6 @@ DECLARE_STATIC_KEY_FALSE(enable_evmcs);
  *	PLE_GAP                         = 0x00004020,
  *	PLE_WINDOW                      = 0x00004022,
  *	VMX_PREEMPTION_TIMER_VALUE      = 0x0000482E,
- *      GUEST_IA32_PERF_GLOBAL_CTRL     = 0x00002808,
- *      HOST_IA32_PERF_GLOBAL_CTRL      = 0x00002c04,
  *
  * Currently unsupported in KVM:
  *	GUEST_IA32_RTIT_CTL		= 0x00002814,
@@ -62,9 +60,8 @@ DECLARE_STATIC_KEY_FALSE(enable_evmcs);
 	 SECONDARY_EXEC_TSC_SCALING |					\
 	 SECONDARY_EXEC_PAUSE_LOOP_EXITING)
 #define EVMCS1_UNSUPPORTED_VMEXIT_CTRL					\
-	(VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL |				\
-	 VM_EXIT_SAVE_VMX_PREEMPTION_TIMER)
-#define EVMCS1_UNSUPPORTED_VMENTRY_CTRL (VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL)
+	(VM_EXIT_SAVE_VMX_PREEMPTION_TIMER)
+#define EVMCS1_UNSUPPORTED_VMENTRY_CTRL (0)
 #define EVMCS1_UNSUPPORTED_VMFUNC (VMX_VMFUNC_EPTP_SWITCHING)
 
 /* TODO: explicitly define these */
-- 
2.37.1.595.g718a3a8f04-goog


  parent reply	other threads:[~2022-08-24  3:04 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-24  3:01 [RFC PATCH v6 00/36] KVM: x86: eVMCS rework Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 01/36] x86/hyperv: Fix 'struct hv_enlightened_vmcs' definition Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 02/36] x86/hyperv: Update " Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 03/36] KVM: x86: Zero out entire Hyper-V CPUID cache before processing entries Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 04/36] KVM: x86: Check for existing Hyper-V vCPU in kvm_hv_vcpu_init() Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 05/36] KVM: x86: Report error when setting CPUID if Hyper-V allocation fails Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 06/36] KVM: nVMX: Treat eVMCS as enabled for guest iff Hyper-V is also enabled Sean Christopherson
2022-08-25 10:21   ` Vitaly Kuznetsov
2022-08-25 14:48     ` Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 07/36] KVM: nVMX: Refactor unsupported eVMCS controls logic to use 2-d array Sean Christopherson
2022-08-25 10:24   ` Vitaly Kuznetsov
2022-08-24  3:01 ` [RFC PATCH v6 08/36] KVM: nVMX: Use CC() macro to handle eVMCS unsupported controls checks Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 09/36] KVM: nVMX: Enforce unsupported eVMCS in VMX MSRs for host accesses Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 10/36] KVM: VMX: Define VMCS-to-EVMCS conversion for the new fields Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 11/36] KVM: nVMX: Support several new fields in eVMCSv1 Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 12/36] KVM: x86: hyper-v: Cache HYPERV_CPUID_NESTED_FEATURES CPUID leaf Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 13/36] KVM: selftests: Add ENCLS_EXITING_BITMAP{,HIGH} VMCS fields Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 14/36] KVM: selftests: Switch to updated eVMCSv1 definition Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 15/36] KVM: nVMX: WARN once and fail VM-Enter if eVMCS sees VMFUNC[63:32] != 0 Sean Christopherson
2022-08-24  3:01 ` Sean Christopherson [this message]
2022-08-24  3:01 ` [RFC PATCH v6 17/36] KVM: nVMX: Support TSC scaling with enlightened VMCS Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 18/36] KVM: selftests: Enable TSC scaling in evmcs selftest Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 19/36] KVM: VMX: Get rid of eVMCS specific VMX controls sanitization Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 20/36] KVM: nVMX: Don't propagate vmcs12's PERF_GLOBAL_CTRL settings to vmcs02 Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 21/36] KVM: nVMX: Always emulate PERF_GLOBAL_CTRL VM-Entry/VM-Exit controls Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 22/36] KVM: VMX: Check VM_ENTRY_IA32E_MODE in setup_vmcs_config() Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 23/36] KVM: VMX: Check CPU_BASED_{INTR,NMI}_WINDOW_EXITING " Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 24/36] KVM: VMX: Tweak the special handling of SECONDARY_EXEC_ENCLS_EXITING " Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 25/36] KVM: VMX: Don't toggle VM_ENTRY_IA32E_MODE for 32-bit kernels/KVM Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 26/36] KVM: VMX: Extend VMX controls macro shenanigans Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 27/36] KVM: VMX: Move CPU_BASED_CR8_{LOAD,STORE}_EXITING filtering out of setup_vmcs_config() Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 28/36] KVM: VMX: Add missing VMEXIT controls to vmcs_config Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 29/36] KVM: VMX: Add missing CPU based VM execution " Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 30/36] KVM: VMX: Adjust CR3/INVPLG interception for EPT=y at runtime, not setup Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 31/36] KVM: x86: VMX: Replace some Intel model numbers with mnemonics Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 32/36] KVM: VMX: Move LOAD_IA32_PERF_GLOBAL_CTRL errata handling out of setup_vmcs_config() Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 33/36] KVM: nVMX: Always set required-1 bits of pinbased_ctls to PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 34/36] KVM: nVMX: Use sanitized allowed-1 bits for VMX control MSRs Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 35/36] KVM: VMX: Cache MSR_IA32_VMX_MISC in vmcs_config Sean Christopherson
2022-08-24  3:01 ` [RFC PATCH v6 36/36] KVM: nVMX: Use cached host MSR_IA32_VMX_MISC value for setting up nested MSR Sean Christopherson
2022-08-25 18:08 ` [RFC PATCH v6 00/36] KVM: x86: eVMCS rework Vitaly Kuznetsov
2022-08-25 18:29   ` Sean Christopherson
2022-08-26 17:19     ` Vitaly Kuznetsov
2022-08-27 14:03       ` Vitaly Kuznetsov
2022-08-29 15:54         ` Sean Christopherson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220824030138.3524159-17-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=vkuznets@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox