ARM DEN0022D 5.19 "SYSTEM_SUSPEND" describes a PSCI call that may be
used to request a system be suspended. This is optional for PSCI v1.0
and to date KVM has elected to not implement the call. However, a
VMM/operator may wish to provide their guests with the ability to
suspend/resume, necessitating this PSCI call.
Implement support for SYSTEM_SUSPEND according to the prescribed
behavior in the specification. Add a new system event exit type,
KVM_SYSTEM_EVENT_SUSPEND, to notify userspace when a VM has requested a
system suspend. Make KVM_MP_STATE_HALTED a valid state on arm64.
Userspace can set this to request an in-kernel emulation of the suspend.
Signed-off-by: Oliver Upton <oupton@xxxxxxxxxx>
---
Documentation/virt/kvm/api.rst | 6 ++++
arch/arm64/include/asm/kvm_host.h | 3 ++
arch/arm64/kvm/arm.c | 8 +++++
arch/arm64/kvm/psci.c | 60 +++++++++++++++++++++++++++++++
include/uapi/linux/kvm.h | 2 ++
5 files changed, 79 insertions(+)
diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
index a6729c8cf063..361a57061b8f 100644
--- a/Documentation/virt/kvm/api.rst
+++ b/Documentation/virt/kvm/api.rst
@@ -5656,6 +5656,7 @@ should put the acknowledged interrupt vector into the 'epr' field.
#define KVM_SYSTEM_EVENT_SHUTDOWN 1
#define KVM_SYSTEM_EVENT_RESET 2
#define KVM_SYSTEM_EVENT_CRASH 3
+ #define KVM_SYSTEM_EVENT_SUSPEND 4
__u32 type;
__u64 flags;
} system_event;
@@ -5680,6 +5681,11 @@ Valid values for 'type' are:
has requested a crash condition maintenance. Userspace can choose
to ignore the request, or to gather VM memory core dump and/or
reset/shutdown of the VM.
+ - KVM_SYSTEM_EVENT_SUSPEND -- the guest has requested that the VM
+ suspends. Userspace is not obliged to honor this, and may call KVM_RUN
+ again. Doing so will cause the guest to resume at its requested entry
+ point. For ARM64, userspace can request in-kernel suspend emulation
+ by setting the vCPU's MP state to KVM_MP_STATE_HALTED.
::
diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
index 1beda1189a15..441eb6fa7adc 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -137,6 +137,9 @@ struct kvm_arch {
/* Memory Tagging Extension enabled for the guest */
bool mte_enabled;
+
+ /* PSCI SYSTEM_SUSPEND call enabled for the guest */
+ bool suspend_enabled;
};
struct kvm_vcpu_fault_info {
diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c
index f1a375648e25..d875d3bcf3c5 100644
--- a/arch/arm64/kvm/arm.c
+++ b/arch/arm64/kvm/arm.c
@@ -101,6 +101,10 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
}
mutex_unlock(&kvm->lock);
break;
+ case KVM_CAP_ARM_SYSTEM_SUSPEND:
+ r = 0;
+ kvm->arch.suspend_enabled = true;
+ break;
default:
r = -EINVAL;
break;
@@ -215,6 +219,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
case KVM_CAP_SET_GUEST_DEBUG:
case KVM_CAP_VCPU_ATTRIBUTES:
case KVM_CAP_PTP_KVM:
+ case KVM_CAP_ARM_SYSTEM_SUSPEND:
r = 1;
break;
case KVM_CAP_SET_GUEST_DEBUG2:
@@ -470,6 +475,9 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
int ret = 0;
switch (mp_state->mp_state) {
+ case KVM_MP_STATE_HALTED:
+ kvm_make_request(KVM_REQ_SUSPEND, vcpu);
+ fallthrough;
case KVM_MP_STATE_RUNNABLE:
vcpu->arch.power_off = false;
break;
diff --git a/arch/arm64/kvm/psci.c b/arch/arm64/kvm/psci.c
index d453666ddb83..cf869f1f8615 100644
--- a/arch/arm64/kvm/psci.c
+++ b/arch/arm64/kvm/psci.c
@@ -203,6 +203,46 @@ static void kvm_psci_system_reset(struct kvm_vcpu *vcpu)
kvm_prepare_system_event(vcpu, KVM_SYSTEM_EVENT_RESET);
}
+static int kvm_psci_system_suspend(struct kvm_vcpu *vcpu)
+{
+ unsigned long entry_addr, context_id;
+ struct kvm *kvm = vcpu->kvm;
+ unsigned long psci_ret = 0;
+ struct kvm_vcpu *tmp;
+ int ret = 0;
+ int i;
+
+ /*
+ * The SYSTEM_SUSPEND PSCI call requires that all vCPUs (except the
+ * calling vCPU) be in an OFF state, as determined by the
+ * implementation.
+ *
+ * See ARM DEN0022D, 5.19 "SYSTEM_SUSPEND" for more details.
+ */
+ mutex_lock(&kvm->lock);
+ kvm_for_each_vcpu(i, tmp, kvm) {
+ if (tmp != vcpu && !tmp->arch.power_off) {
+ psci_ret = PSCI_RET_DENIED;
+ ret = 1;
+ goto out;
+ }
+ }
+
+ entry_addr = smccc_get_arg1(vcpu);
+ context_id = smccc_get_arg2(vcpu);
+
+ kvm_psci_vcpu_request_reset(vcpu, entry_addr, context_id,
+ kvm_vcpu_is_be(vcpu));
+
+ memset(&vcpu->run->system_event, 0, sizeof(vcpu->run->system_event));
+ vcpu->run->system_event.type = KVM_SYSTEM_EVENT_SUSPEND;
+ vcpu->run->exit_reason = KVM_EXIT_SYSTEM_EVENT;
+out:
+ mutex_unlock(&kvm->lock);
+ smccc_set_retval(vcpu, psci_ret, 0, 0, 0);
+ return ret;
+}
+
static void kvm_psci_narrow_to_32bit(struct kvm_vcpu *vcpu)
{
int i;
@@ -223,6 +263,14 @@ static unsigned long kvm_psci_check_allowed_function(struct kvm_vcpu *vcpu, u32
if ((fn & PSCI_0_2_64BIT) && vcpu_mode_is_32bit(vcpu))
return PSCI_RET_NOT_SUPPORTED;
+ switch (fn) {
+ case PSCI_1_0_FN_SYSTEM_SUSPEND:
+ case PSCI_1_0_FN64_SYSTEM_SUSPEND:
+ if (!vcpu->kvm->arch.suspend_enabled)
+ return PSCI_RET_NOT_SUPPORTED;
+ break;
+ }
+
return 0;
}
@@ -316,6 +364,10 @@ static int kvm_psci_1_0_call(struct kvm_vcpu *vcpu)
unsigned long val;
int ret = 1;
+ val = kvm_psci_check_allowed_function(vcpu, psci_fn);
+ if (val)
+ goto out;
+
switch(psci_fn) {
case PSCI_0_2_FN_PSCI_VERSION:
val = KVM_ARM_PSCI_1_0;
@@ -339,6 +391,8 @@ static int kvm_psci_1_0_call(struct kvm_vcpu *vcpu)
case PSCI_0_2_FN_SYSTEM_OFF:
case PSCI_0_2_FN_SYSTEM_RESET:
case PSCI_1_0_FN_PSCI_FEATURES:
+ case PSCI_1_0_FN_SYSTEM_SUSPEND:
+ case PSCI_1_0_FN64_SYSTEM_SUSPEND:
case ARM_SMCCC_VERSION_FUNC_ID:
val = 0;
break;
@@ -347,10 +401,16 @@ static int kvm_psci_1_0_call(struct kvm_vcpu *vcpu)
break;
}
break;
+ case PSCI_1_0_FN_SYSTEM_SUSPEND:
+ kvm_psci_narrow_to_32bit(vcpu);
+ fallthrough;
+ case PSCI_1_0_FN64_SYSTEM_SUSPEND:
+ return kvm_psci_system_suspend(vcpu);
default:
return kvm_psci_0_2_call(vcpu);
}
+out:
smccc_set_retval(vcpu, val, 0, 0, 0);
return ret;
}
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index a067410ebea5..052b0e717b08 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -433,6 +433,7 @@ struct kvm_run {
#define KVM_SYSTEM_EVENT_SHUTDOWN 1
#define KVM_SYSTEM_EVENT_RESET 2
#define KVM_SYSTEM_EVENT_CRASH 3
+#define KVM_SYSTEM_EVENT_SUSPEND 4
__u32 type;
__u64 flags;
} system_event;
@@ -1112,6 +1113,7 @@ struct kvm_ppc_resize_hpt {
#define KVM_CAP_BINARY_STATS_FD 203
#define KVM_CAP_EXIT_ON_EMULATION_FAILURE 204
#define KVM_CAP_ARM_MTE 205
+#define KVM_CAP_ARM_SYSTEM_SUSPEND 206
#ifdef KVM_CAP_IRQ_ROUTING
--
2.33.0.685.g46640cef36-goog
