From: Vitaly Kuznetsov <vkuznets@redhat.com>
To: kvm@vger.kernel.org, Paolo Bonzini <pbonzini@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>,
Wanpeng Li <wanpengli@tencent.com>,
Jim Mattson <jmattson@google.com>,
Michael Kelley <mikelley@microsoft.com>,
Yuan Yao <yuan.yao@linux.intel.com>,
Maxim Levitsky <mlevitsk@redhat.com>,
linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v6 15/33] KVM: nVMX: Support PERF_GLOBAL_CTRL with enlightened VMCS
Date: Tue, 30 Aug 2022 15:37:19 +0200 [thread overview]
Message-ID: <20220830133737.1539624-16-vkuznets@redhat.com> (raw)
In-Reply-To: <20220830133737.1539624-1-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>
---
arch/x86/kvm/hyperv.c | 2 +-
arch/x86/kvm/vmx/evmcs.c | 30 +++++++++++++++++++++++++++---
arch/x86/kvm/vmx/evmcs.h | 9 +++------
arch/x86/kvm/vmx/vmx.c | 2 +-
4 files changed, 32 insertions(+), 11 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 b64e29f1359f..5fc4834be1f6 100644
--- a/arch/x86/kvm/vmx/evmcs.c
+++ b/arch/x86/kvm/vmx/evmcs.c
@@ -412,10 +412,28 @@ static u32 evmcs_get_unsupported_ctls(enum evmcs_ctrl_type ctrl_type)
return evmcs_unsupported_ctrls[ctrl_type][evmcs_rev];
}
-void nested_evmcs_filter_control_msr(u32 msr_index, u64 *pdata)
+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, u32 msr_index, u64 *pdata)
{
u32 ctl_low = (u32)*pdata;
u32 ctl_high = (u32)(*pdata >> 32);
+ u32 unsupported_ctrls;
/*
* Hyper-V 2016 and 2019 try using these features even when eVMCS
@@ -424,11 +442,17 @@ void nested_evmcs_filter_control_msr(u32 msr_index, u64 *pdata)
switch (msr_index) {
case MSR_IA32_VMX_EXIT_CTLS:
case MSR_IA32_VMX_TRUE_EXIT_CTLS:
- ctl_high &= ~evmcs_get_unsupported_ctls(EVMCS_EXIT_CTRLS);
+ unsupported_ctrls = evmcs_get_unsupported_ctls(EVMCS_EXIT_CTRLS);
+ if (!evmcs_has_perf_global_ctrl(vcpu))
+ unsupported_ctrls |= VM_EXIT_LOAD_IA32_PERF_GLOBAL_CTRL;
+ ctl_high &= ~unsupported_ctrls;
break;
case MSR_IA32_VMX_ENTRY_CTLS:
case MSR_IA32_VMX_TRUE_ENTRY_CTLS:
- ctl_high &= ~evmcs_get_unsupported_ctls(EVMCS_ENTRY_CTRLS);
+ unsupported_ctrls = evmcs_get_unsupported_ctls(EVMCS_ENTRY_CTRLS);
+ if (!evmcs_has_perf_global_ctrl(vcpu))
+ unsupported_ctrls |= VM_ENTRY_LOAD_IA32_PERF_GLOBAL_CTRL;
+ ctl_high &= ~unsupported_ctrls;
break;
case MSR_IA32_VMX_PROCBASED_CTLS2:
ctl_high &= ~evmcs_get_unsupported_ctls(EVMCS_2NDEXEC);
diff --git a/arch/x86/kvm/vmx/evmcs.h b/arch/x86/kvm/vmx/evmcs.h
index f886a8ff0342..d0b2861325a0 100644
--- a/arch/x86/kvm/vmx/evmcs.h
+++ b/arch/x86/kvm/vmx/evmcs.h
@@ -42,8 +42,6 @@ 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,
@@ -61,9 +59,8 @@ 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)
struct evmcs_field {
@@ -243,7 +240,7 @@ bool nested_enlightened_vmentry(struct kvm_vcpu *vcpu, u64 *evmcs_gpa);
uint16_t nested_get_evmcs_version(struct kvm_vcpu *vcpu);
int nested_enable_evmcs(struct kvm_vcpu *vcpu,
uint16_t *vmcs_version);
-void nested_evmcs_filter_control_msr(u32 msr_index, u64 *pdata);
+void nested_evmcs_filter_control_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
int nested_evmcs_check_controls(struct vmcs12 *vmcs12);
#endif /* __KVM_X86_VMX_EVMCS_H */
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index d4ed802947d7..f91f09db6e65 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -1931,7 +1931,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
* features out.
*/
if (!msr_info->host_initiated && guest_cpuid_has_evmcs(vcpu))
- nested_evmcs_filter_control_msr(msr_info->index,
+ nested_evmcs_filter_control_msr(vcpu, msr_info->index,
&msr_info->data);
break;
case MSR_IA32_RTIT_CTL:
--
2.37.2
next prev parent reply other threads:[~2022-08-30 13:40 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-30 13:37 [PATCH v6 00/33] KVM: VMX: Support updated eVMCSv1 revision + use vmcs_config for L1 VMX MSRs Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 01/33] x86/hyperv: Fix 'struct hv_enlightened_vmcs' definition Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 02/33] x86/hyperv: Update " Vitaly Kuznetsov
2022-09-05 17:06 ` Wei Liu
2022-08-30 13:37 ` [PATCH v6 03/33] KVM: x86: Zero out entire Hyper-V CPUID cache before processing entries Vitaly Kuznetsov
2022-09-05 17:07 ` Wei Liu
2022-08-30 13:37 ` [PATCH v6 04/33] KVM: x86: Check for existing Hyper-V vCPU in kvm_hv_vcpu_init() Vitaly Kuznetsov
2022-09-05 17:08 ` Wei Liu
2022-08-30 13:37 ` [PATCH v6 05/33] KVM: x86: Report error when setting CPUID if Hyper-V allocation fails Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 06/33] KVM: nVMX: Treat eVMCS as enabled for guest iff Hyper-V is also enabled Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 07/33] KVM: nVMX: Refactor unsupported eVMCS controls logic to use 2-d array Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 08/33] KVM: nVMX: Use CC() macro to handle eVMCS unsupported controls checks Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 09/33] KVM: VMX: Define VMCS-to-EVMCS conversion for the new fields Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 10/33] KVM: nVMX: Support several new fields in eVMCSv1 Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 11/33] KVM: x86: hyper-v: Cache HYPERV_CPUID_NESTED_FEATURES CPUID leaf Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 12/33] KVM: selftests: Add ENCLS_EXITING_BITMAP{,HIGH} VMCS fields Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 13/33] KVM: selftests: Switch to updated eVMCSv1 definition Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 14/33] KVM: nVMX: WARN once and fail VM-Enter if eVMCS sees VMFUNC[63:32] != 0 Vitaly Kuznetsov
2022-08-30 13:37 ` Vitaly Kuznetsov [this message]
2022-08-30 13:37 ` [PATCH v6 16/33] KVM: VMX: Get rid of eVMCS specific VMX controls sanitization Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 17/33] KVM: nVMX: Don't propagate vmcs12's PERF_GLOBAL_CTRL settings to vmcs02 Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 18/33] KVM: nVMX: Always emulate PERF_GLOBAL_CTRL VM-Entry/VM-Exit controls Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 19/33] KVM: VMX: Check VM_ENTRY_IA32E_MODE in setup_vmcs_config() Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 20/33] KVM: VMX: Check CPU_BASED_{INTR,NMI}_WINDOW_EXITING " Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 21/33] KVM: VMX: Tweak the special handling of SECONDARY_EXEC_ENCLS_EXITING " Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 22/33] KVM: VMX: Don't toggle VM_ENTRY_IA32E_MODE for 32-bit kernels/KVM Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 23/33] KVM: VMX: Extend VMX controls macro shenanigans Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 24/33] KVM: VMX: Move CPU_BASED_CR8_{LOAD,STORE}_EXITING filtering out of setup_vmcs_config() Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 25/33] KVM: VMX: Add missing VMEXIT controls to vmcs_config Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 26/33] KVM: VMX: Add missing CPU based VM execution " Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 27/33] KVM: VMX: Adjust CR3/INVPLG interception for EPT=y at runtime, not setup Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 28/33] KVM: x86: VMX: Replace some Intel model numbers with mnemonics Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 29/33] KVM: VMX: Move LOAD_IA32_PERF_GLOBAL_CTRL errata handling out of setup_vmcs_config() Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 30/33] KVM: nVMX: Always set required-1 bits of pinbased_ctls to PIN_BASED_ALWAYSON_WITHOUT_TRUE_MSR Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 31/33] KVM: nVMX: Use sanitized allowed-1 bits for VMX control MSRs Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 32/33] KVM: VMX: Cache MSR_IA32_VMX_MISC in vmcs_config Vitaly Kuznetsov
2022-08-30 13:37 ` [PATCH v6 33/33] KVM: nVMX: Use cached host MSR_IA32_VMX_MISC value for setting up nested MSR Vitaly Kuznetsov
2022-09-17 1:05 ` [PATCH v6 00/33] KVM: VMX: Support updated eVMCSv1 revision + use vmcs_config for L1 VMX MSRs 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=20220830133737.1539624-16-vkuznets@redhat.com \
--to=vkuznets@redhat.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=linux-hyperv@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mikelley@microsoft.com \
--cc=mlevitsk@redhat.com \
--cc=pbonzini@redhat.com \
--cc=seanjc@google.com \
--cc=wanpengli@tencent.com \
--cc=yuan.yao@linux.intel.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
all inboxes | Powered by JetHome®