@@ -3619,6 +3619,7 @@ static void kvm_accel_instance_init(Object *obj)
s->xen_version = 0;
s->xen_gnttab_max_frames = 64;
s->xen_evtchn_max_pirq = 256;
+ s->pmu_cap_disabled = false;
}
/**
@@ -120,6 +120,7 @@ struct KVMState
uint32_t xen_caps;
uint16_t xen_gnttab_max_frames;
uint16_t xen_evtchn_max_pirq;
+ bool pmu_cap_disabled;
};
void kvm_memory_listener_register(KVMState *s, KVMMemoryListener *kml,
@@ -188,6 +188,7 @@ DEF("accel", HAS_ARG, QEMU_OPTION_accel,
" dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n"
" eager-split-size=n (KVM Eager Page Split chunk size, default 0, disabled. ARM only)\n"
" notify-vmexit=run|internal-error|disable,notify-window=n (enable notify VM exit and set notify window, x86 only)\n"
+ " pmu-cap-disabled=true|false (disable KVM_CAP_PMU_CAPABILITY, x86 only, default false)\n"
" thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
SRST
``-accel name[,prop=value[,...]]``
@@ -269,6 +270,12 @@ SRST
open up for a specified of time (i.e. notify-window).
Default: notify-vmexit=run,notify-window=0.
+ ``pmu-cap-disabled=true|false``
+ When the KVM accelerator is used, it controls whether to disable the
+ KVM_CAP_PMU_CAPABILITY via KVM_PMU_CAP_DISABLE. When disabled, the
+ PMU virtualization is disabled at the KVM module side. This is for
+ x86 host only.
+
ERST
DEF("smp", HAS_ARG, QEMU_OPTION_smp,
@@ -138,6 +138,7 @@ static bool has_msr_ucode_rev;
static bool has_msr_vmx_procbased_ctls2;
static bool has_msr_perf_capabs;
static bool has_msr_pkrs;
+static bool has_pmu_cap;
static uint32_t has_architectural_pmu_version;
static uint32_t num_architectural_pmu_gp_counters;
@@ -2713,6 +2714,23 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
}
}
+ has_pmu_cap = kvm_check_extension(s, KVM_CAP_PMU_CAPABILITY);
+
+ if (s->pmu_cap_disabled) {
+ if (has_pmu_cap) {
+ ret = kvm_vm_enable_cap(s, KVM_CAP_PMU_CAPABILITY, 0,
+ KVM_PMU_CAP_DISABLE);
+ if (ret < 0) {
+ s->pmu_cap_disabled = false;
+ error_report("kvm: Failed to disable pmu cap: %s",
+ strerror(-ret));
+ }
+ } else {
+ s->pmu_cap_disabled = false;
+ error_report("kvm: KVM_CAP_PMU_CAPABILITY is not supported");
+ }
+ }
+
return 0;
}
@@ -5772,6 +5790,28 @@ static void kvm_arch_set_xen_evtchn_max_pirq(Object *obj, Visitor *v,
s->xen_evtchn_max_pirq = value;
}
+static void kvm_set_pmu_cap_disabled(Object *obj, Visitor *v,
+ const char *name, void *opaque,
+ Error **errp)
+{
+ KVMState *s = KVM_STATE(obj);
+ bool pmu_cap_disabled;
+ Error *error = NULL;
+
+ if (s->fd != -1) {
+ error_setg(errp, "Cannot set properties after the accelerator has been initialized");
+ return;
+ }
+
+ visit_type_bool(v, name, &pmu_cap_disabled, &error);
+ if (error) {
+ error_propagate(errp, error);
+ return;
+ }
+
+ s->pmu_cap_disabled = pmu_cap_disabled;
+}
+
void kvm_arch_accel_class_init(ObjectClass *oc)
{
object_class_property_add_enum(oc, "notify-vmexit", "NotifyVMexitOption",
@@ -5811,6 +5851,12 @@ void kvm_arch_accel_class_init(ObjectClass *oc)
NULL, NULL);
object_class_property_set_description(oc, "xen-evtchn-max-pirq",
"Maximum number of Xen PIRQs");
+
+ object_class_property_add(oc, "pmu-cap-disabled", "bool",
+ NULL, kvm_set_pmu_cap_disabled,
+ NULL, NULL);
+ object_class_property_set_description(oc, "pmu-cap-disabled",
+ "Disable KVM_CAP_PMU_CAPABILITY");
}
void kvm_set_max_apic_id(uint32_t max_apic_id)
The "perf stat" at the VM side still works even we set "-cpu host,-pmu" in the QEMU command line. That is, neither "-cpu host,-pmu" nor "-cpu EPYC" could disable the pmu virtualization in an AMD environment. We still see below at VM kernel side ... [ 0.510611] Performance Events: Fam17h+ core perfctr, AMD PMU driver. ... although we expect something like below. [ 0.596381] Performance Events: PMU not available due to virtualization, using software events only. [ 0.600972] NMI watchdog: Perf NMI watchdog permanently disabled This is because the AMD pmu does not rely on cpuid to decide if the pmu virtualization is supported. We introduce a new property 'pmu-cap-disabled' for KVM accel to set KVM_PMU_CAP_DISABLE if KVM_CAP_PMU_CAPABILITY is supported. Only x86 host is supported because currently KVM uses KVM_CAP_PMU_CAPABILITY only for x86. Cc: Joe Jin <joe.jin@oracle.com> Cc: Like Xu <likexu@tencent.com> Cc: Denis V. Lunev <den@virtuozzo.com> Signed-off-by: Dongli Zhang <dongli.zhang@oracle.com> --- This is to resurrect the patch to disable PMU. I split the patchset and send the patch to disable PMU. Changed since v1: [PATCH 1/3] kvm: introduce a helper before creating the 1st vcpu https://lore.kernel.org/all/20221119122901.2469-2-dongli.zhang@oracle.com/ [PATCH 2/3] i386: kvm: disable KVM_CAP_PMU_CAPABILITY if "pmu" is disabled https://lore.kernel.org/all/20221119122901.2469-3-dongli.zhang@oracle.com/ - In version 1 we did not introduce the new property. We ioctl KVM_PMU_CAP_DISABLE only before the creation of the 1st vcpu. We had introduced a helpfer function to do this job before creating the 1st KVM vcpu in v1. Changed since v2: https://lore.kernel.org/all/20230621013821.6874-2-dongli.zhang@oracle.com/ Nothing. I split the patchset and send this as a single patch. As a summary: - Greg Kurz and Liang Yan suggested introduce the machine property to disable the PMU (e.g., with the concern of live migration, or vCPU prop theoretically be different for each vCPU). - Denis V. Lunev and Like Xu preferred the method in v1 patch: to re-use cpu->enable_pmu. Would you please suggest if we may go via v1 (re-use cpu->enable_pmu) or v2 (to introduce new machine prop) 1. The v1 is to re-use cpu->enable_pmu. It disables KVM_PMU_CAP_DISABLE when creating the 1st vCPU. We may use the vCPU id or (current_cpu == first_cpu) to check when it is the 1st vCPU creation. The benefit is that the QEMU user (e.g., libvirt will not require much change). 2. The v2 is to introduce the new machine property as in this patch. The benefit: the 'pmu' is to configure cpuid, while KVM_PMU_CAP_DISABLE is a different KVM feature. They are orthogonal features. Perhaps there is another option to sum both v1 and v2 together ... Perhaps the maintainer can help make decision on that :) Thank you very much! accel/kvm/kvm-all.c | 1 + include/sysemu/kvm_int.h | 1 + qemu-options.hx | 7 ++++++ target/i386/kvm/kvm.c | 46 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 55 insertions(+)