Skip to contentSkip to conversation

i386: simplify Hyper-V enlightenments enablement

PATCH New Vitaly Kuznetsov 1 participant 3 messages Dec 17, 2020

lore
Share & tools
lore
#0 lore raw

v2 cover i386: simplify Hyper-V enlightenments enablement

Changes since v1:
- Moved X86MachineClass related setup to x86_cpu_pre_plug [Eduardo], introduced kvm_hv_evmcs_available() to support the change.
- Droped PATCHes0-4 from v1 as x86_cpu_hyperv_realize() is not needed for the purpose of this series. I'll be sending them out as part of "KVM: expand Hyper-V features early" series after 5.11-rc1 Linux release.

Original description:

This series is a part of the previously sent "[PATCH RFC v3 00/23] i386:
KVM: expand Hyper-V features early":
https://lists.gnu.org/archive/html/qemu-devel/2020-10/msg02443.html

We're not ready to merge the full patch set yet because the required KVM capability is only queued for 5.11. We can, however, extract the part providing 'hyperv=on' option to x86 machine types which is valuable on its own.

Vitaly Kuznetsov (2):
i386: introduce kvm_hv_evmcs_available()
i386: provide simple 'hyperv=on' option to x86 machine types

 docs/hyperv.txt        |  8 ++++++++
 hw/i386/x86.c          | 42 ++++++++++++++++++++++++++++++++++++++++++
 include/hw/i386/x86.h  |  7 +++++++
 target/i386/kvm-stub.c |  5 +++++
 target/i386/kvm.c      |  8 ++++++++
 target/i386/kvm_i386.h |  1 +
 6 files changed, 71 insertions(+)
Signature · 1 line
2.29.2
#1 lore raw

PATCH v2 1/2 i386: introduce kvm_hv_evmcs_available()

Enlightened VMCS feature is hardware specific, it is only supported on Intel CPUs. Introduce a simple kvm_hv_evmcs_available() helper, it will be used to filter out 'hv_evmcs' when 'hyperv=on' option is added to X86MachineClass.

---
 target/i386/kvm-stub.c | 5 +++++
 target/i386/kvm.c      | 8 ++++++++
 target/i386/kvm_i386.h | 1 +
 3 files changed, 14 insertions(+)
target/i386/kvm-stub.c +5 -0
index 92f49121b8fa..0a163ae207c5 100644
@@ -39,3 +39,8 @@ bool kvm_hv_vpindex_settable(void)
 {
     return false;
 }
+
+bool kvm_hv_evmcs_available(void)
+{
+    return false;
+}
target/i386/kvm.c +8 -0
index bcfa4b03e077..96395d499729 100644
@@ -95,6 +95,7 @@ static bool has_msr_hv_crash;
 static bool has_msr_hv_reset;
 static bool has_msr_hv_vpindex;
 static bool hv_vpindex_settable;
+static bool hv_evmcs_available;
 static bool has_msr_hv_runtime;
 static bool has_msr_hv_synic;
 static bool has_msr_hv_stimer;
@@ -192,6 +193,11 @@ bool kvm_hv_vpindex_settable(void)
     return hv_vpindex_settable;
 }
 
+bool kvm_hv_evmcs_available(void)
+{
+    return hv_evmcs_available;
+}
+
 static int kvm_get_tsc(CPUState *cs)
 {
     X86CPU *cpu = X86_CPU(cs);
@@ -2124,6 +2130,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
     has_pit_state2 = kvm_check_extension(s, KVM_CAP_PIT_STATE2);
 
     hv_vpindex_settable = kvm_check_extension(s, KVM_CAP_HYPERV_VP_INDEX);
+    hv_evmcs_available =
+        kvm_check_extension(s, KVM_CAP_HYPERV_ENLIGHTENED_VMCS);
 
     has_exception_payload = kvm_check_extension(s, KVM_CAP_EXCEPTION_PAYLOAD);
     if (has_exception_payload) {
target/i386/kvm_i386.h +1 -0
index dc725083891c..08968cfb33f1 100644
@@ -47,6 +47,7 @@ bool kvm_has_x2apic_api(void);
 bool kvm_has_waitpkg(void);
 
 bool kvm_hv_vpindex_settable(void);
+bool kvm_hv_evmcs_available(void);
 
 uint64_t kvm_swizzle_msi_ext_dest_id(uint64_t address);
 
Signature · 1 line
2.29.2
#2 lore raw

PATCH v2 2/2 i386: provide simple 'hyperv=on' option to x86 machine types

In reply to Vitaly Kuznetsov

Enabling Hyper-V emulation for a Windows VM is a tiring experience as it requires listing all currently supported enlightenments ("hv_*" CPU features) explicitly. We do have a 'hv_passthrough' mode enabling everything but it can't be used in production as it prevents migration.

Introduce a simple 'hyperv=on' option for all x86 machine types enabling all currently supported Hyper-V enlightenments. Later, when new enlightenments get implemented, we will be adding them to newer machine types only (by disabling them for legacy machine types) thus preserving migration.

---
 docs/hyperv.txt       |  8 ++++++++
 hw/i386/x86.c         | 42 ++++++++++++++++++++++++++++++++++++++++++
 include/hw/i386/x86.h |  7 +++++++
 3 files changed, 57 insertions(+)
docs/hyperv.txt +8 -0
index 5df00da54fc4..1a76a07f8417 100644
@@ -29,6 +29,14 @@ When any set of the Hyper-V enlightenments is enabled, QEMU changes hypervisor
 identification (CPUID 0x40000000..0x4000000A) to Hyper-V. KVM identification
 and features are kept in leaves 0x40000100..0x40000101.
 
+Hyper-V enlightenments can be enabled in bulk by specifying 'hyperv=on' to an
+x86 machine type:
+
+  qemu-system-x86_64 -machine q35,accel=kvm,kernel-irqchip=split,hyperv=on ...
+
+Note, new enlightenments are only added to the latest (in-develompent) machine
+type, older machine types keep the list of the supported features intact to
+safeguard migration.
 
 3. Existing enlightenments
 ===========================
hw/i386/x86.c +42 -0

66 lines not on this page · raw message

include/hw/i386/x86.h +7 -0
index 56080bd1fb14..c6e9c8500357 100644
@@ -38,6 +38,9 @@ struct X86MachineClass {
     bool save_tsc_khz;
     /* Enables contiguous-apic-ID mode */
     bool compat_apic_id_mode;
+
+    /* Hyper-V features enabled with 'hyperv=on' */
+    uint64_t default_hyperv_features;
 };
 
 struct X86MachineState {
@@ -72,10 +75,14 @@ struct X86MachineState {
      * will be translated to MSI messages in the address space.
      */
     AddressSpace *ioapic_as;
+
+    /* Hyper-V emulation */
+    bool hyperv_enabled;
 };
 
 #define X86_MACHINE_SMM              "smm"
 #define X86_MACHINE_ACPI             "acpi"
+#define X86_MACHINE_HYPERV           "hyperv"
 
 #define TYPE_X86_MACHINE   MACHINE_TYPE_NAME("x86")
 OBJECT_DECLARE_TYPE(X86MachineState, X86MachineClass, X86_MACHINE)
Signature · 1 line
2.29.2