On 2016/1/15 21:28, Andrew Jones wrote:
On Fri, Jan 15, 2016 at 02:27:55PM +0800, Shannon Zhao wrote:
From: Shannon Zhao <shannon.z...@linaro.org>
To configure the virtual PMUv3 overflow interrupt number, we use the
vcpu kvm_device ioctl, encapsulating the KVM_ARM_VCPU_PMU_V3_IRQ
attribute within the KVM_ARM_VCPU_PMU_V3_CTRL group.
After configuring the PMUv3, call the vcpu iotcl with attribute
^ ioctl
KVM_ARM_VCPU_PMU_V3_INIT to initialize the PMUv3.
Signed-off-by: Shannon Zhao <shannon.z...@linaro.org>
---
CC: Peter Maydell <peter.mayd...@linaro.org>
---
Documentation/virtual/kvm/devices/vcpu.txt | 24 ++++++
arch/arm/include/asm/kvm_host.h | 15 ++++
arch/arm/kvm/arm.c | 3 +
arch/arm64/include/asm/kvm_host.h | 6 ++
arch/arm64/include/uapi/asm/kvm.h | 5 ++
arch/arm64/kvm/guest.c | 51 ++++++++++++
include/kvm/arm_pmu.h | 23 ++++++
virt/kvm/arm/pmu.c | 128 +++++++++++++++++++++++++++++
8 files changed, 255 insertions(+)
diff --git a/Documentation/virtual/kvm/devices/vcpu.txt
b/Documentation/virtual/kvm/devices/vcpu.txt
index 3cc59c5..60cbac8 100644
--- a/Documentation/virtual/kvm/devices/vcpu.txt
+++ b/Documentation/virtual/kvm/devices/vcpu.txt
@@ -6,3 +6,27 @@ KVM_GET_DEVICE_ATTR, and KVM_HAS_DEVICE_ATTR. The interface
uses the same struct
kvm_device_attr as other devices, but targets VCPU-wide settings and controls.
The groups and attributes per virtual cpu, if any, are architecture specific.
+
+1. GROUP: KVM_ARM_VCPU_PMU_V3_CTRL
+Architectures: ARM64
+
+1.1. ATTRIBUTE: KVM_ARM_VCPU_PMU_V3_IRQ
+Parameters: in kvm_device_attr.addr the address for PMU overflow interrupt
+Returns: -EBUSY: The PMU overflow interrupt is already set
+ -ENODEV: Not set the corresponding vcpu feature bit or getting the PMU
+ overflow interrupt number while it's not set
I think we should have two separate error numbers. One for "PMUv3 not
supported" and another for the overflow interrupt not set when
attempting to get it (ENXIO?). In any case, please rephrase.
+ -EINVAL: Invalid PMU overflow interrupt number supplied
+
+A value describing the PMUv3 (Performance Monitor Unit v3) overflow interrupt
+number for this vcpu. This interrupt could be a PPI or SPI, but the interrupt
+type must be same for each vcpu. As a PPI, the interrupt number is same for all
+vcpus, while as a SPI it must be different for each vcpu.
/as a/as an/
+
+1.2 ATTRIBUTE: KVM_ARM_VCPU_PMU_V3_INIT
+Parameters: no additional parameter in kvm_device_attr.addr
+Returns: -ENODEV: No PMUv3 supported
"PMUv3 not supported"
+ -ENXIO: PMUv3 not properly configured as required prior to calling
this
+ attribute
+ -EBUSY: PMUv3 already initialized
+
+Request the initialization of the PMUv3.
diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h
index f9f2779..6dd0992 100644
--- a/arch/arm/include/asm/kvm_host.h
+++ b/arch/arm/include/asm/kvm_host.h
@@ -242,5 +242,20 @@ static inline void kvm_arm_init_debug(void) {}
static inline void kvm_arm_setup_debug(struct kvm_vcpu *vcpu) {}
static inline void kvm_arm_clear_debug(struct kvm_vcpu *vcpu) {}
static inline void kvm_arm_reset_debug_ptr(struct kvm_vcpu *vcpu) {}
+static inline int kvm_arm_vcpu_arch_set_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
+static inline int kvm_arm_vcpu_arch_get_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
+static inline int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
#endif /* __ARM_KVM_HOST_H__ */
diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
index 34d7395..dc8644f 100644
--- a/arch/arm/kvm/arm.c
+++ b/arch/arm/kvm/arm.c
@@ -833,6 +833,7 @@ static int kvm_arm_vcpu_set_attr(struct kvm_vcpu *vcpu,
switch (attr->group) {
default:
+ ret = kvm_arm_vcpu_arch_set_attr(vcpu, attr);
break;
}
@@ -846,6 +847,7 @@ static int kvm_arm_vcpu_get_attr(struct kvm_vcpu *vcpu,
switch (attr->group) {
default:
+ ret = kvm_arm_vcpu_arch_get_attr(vcpu, attr);
break;
}
@@ -859,6 +861,7 @@ static int kvm_arm_vcpu_has_attr(struct kvm_vcpu *vcpu,
switch (attr->group) {
default:
+ ret = kvm_arm_vcpu_arch_has_attr(vcpu, attr);
break;
}
diff --git a/arch/arm64/include/asm/kvm_host.h
b/arch/arm64/include/asm/kvm_host.h
index cb220b7..a855a30 100644
--- a/arch/arm64/include/asm/kvm_host.h
+++ b/arch/arm64/include/asm/kvm_host.h
@@ -359,5 +359,11 @@ void kvm_arm_init_debug(void);
void kvm_arm_setup_debug(struct kvm_vcpu *vcpu);
void kvm_arm_clear_debug(struct kvm_vcpu *vcpu);
void kvm_arm_reset_debug_ptr(struct kvm_vcpu *vcpu);
+int kvm_arm_vcpu_arch_set_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
+int kvm_arm_vcpu_arch_get_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
+int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
#endif /* __ARM64_KVM_HOST_H__ */
diff --git a/arch/arm64/include/uapi/asm/kvm.h
b/arch/arm64/include/uapi/asm/kvm.h
index 6aedbe3..dc69037 100644
--- a/arch/arm64/include/uapi/asm/kvm.h
+++ b/arch/arm64/include/uapi/asm/kvm.h
@@ -205,6 +205,11 @@ struct kvm_arch_memory_slot {
#define KVM_DEV_ARM_VGIC_GRP_CTRL 4
#define KVM_DEV_ARM_VGIC_CTRL_INIT 0
+/* Device Control API on vcpu fd */
+#define KVM_ARM_VCPU_PMU_V3_CTRL 0
+#define KVM_ARM_VCPU_PMU_V3_IRQ 0
+#define KVM_ARM_VCPU_PMU_V3_INIT 1
too many spaces
This just wants to be consistent with the existing ones.
+
/* KVM_IRQ_LINE irq field index values */
#define KVM_ARM_IRQ_TYPE_SHIFT 24
#define KVM_ARM_IRQ_TYPE_MASK 0xff
diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
index fcb7788..dbe45c3 100644
--- a/arch/arm64/kvm/guest.c
+++ b/arch/arm64/kvm/guest.c
@@ -380,3 +380,54 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu
*vcpu,
}
return 0;
}
+
+int kvm_arm_vcpu_arch_set_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ int ret;
+
+ switch (attr->group) {
+ case KVM_ARM_VCPU_PMU_V3_CTRL:
+ ret = kvm_arm_pmu_v3_set_attr(vcpu, attr);
+ break;
+ default:
+ ret = -ENXIO;
+ break;
+ }
+
+ return ret;
+}
+
+int kvm_arm_vcpu_arch_get_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ int ret;
+
+ switch (attr->group) {
+ case KVM_ARM_VCPU_PMU_V3_CTRL:
+ ret = kvm_arm_pmu_v3_get_attr(vcpu, attr);
+ break;
+ default:
+ ret = -ENXIO;
+ break;
+ }
+
+ return ret;
+}
+
+int kvm_arm_vcpu_arch_has_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ int ret;
+
+ switch (attr->group) {
+ case KVM_ARM_VCPU_PMU_V3_CTRL:
+ ret = kvm_arm_pmu_v3_has_attr(vcpu, attr);
+ break;
+ default:
+ ret = -ENXIO;
+ break;
+ }
+
+ return ret;
+}
diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
index c35b11d..5351109 100644
--- a/include/kvm/arm_pmu.h
+++ b/include/kvm/arm_pmu.h
@@ -36,6 +36,7 @@ struct kvm_pmu {
};
#define kvm_arm_pmu_v3_ready(v) ((v)->arch.pmu.ready)
+#define kvm_arm_pmu_irq_initialized(v) ((v)->arch.pmu.irq_num >= VGIC_NR_SGIS)
u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx);
u64 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu);
void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu);
@@ -49,11 +50,18 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val);
void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
u64 select_idx);
int kvm_arm_support_pmu_v3(void);
+int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
+int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
+int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr);
#else
struct kvm_pmu {
};
#define kvm_arm_pmu_v3_ready(v) (false)
+#define kvm_arm_pmu_irq_initialized(v) (false)
should tab the false over
static inline u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu,
u64 select_idx)
{
@@ -74,6 +82,21 @@ static inline void kvm_pmu_handle_pmcr(struct kvm_vcpu
*vcpu, u64 val) {}
static inline void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu,
u64 data, u64 select_idx) {}
static inline int kvm_arm_support_pmu_v3(void) { return 0; }
+static inline int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
+static inline int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
+static inline int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr)
+{
+ return -ENXIO;
+}
#endif
#endif
diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c
index cb373d4..99a6e8a 100644
--- a/virt/kvm/arm/pmu.c
+++ b/virt/kvm/arm/pmu.c
@@ -19,6 +19,7 @@
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <linux/perf_event.h>
+#include <linux/uaccess.h>
#include <asm/kvm_emulate.h>
#include <kvm/arm_pmu.h>
#include <kvm/arm_vgic.h>
@@ -383,3 +384,130 @@ int kvm_arm_support_pmu_v3(void)
*/
return perf_num_counters();
}
+
+static int kvm_arm_pmu_v3_init(struct kvm_vcpu *vcpu)
+{
+ if (kvm_arm_pmu_v3_ready(vcpu))
+ return -EBUSY;
+
+ if (kvm_arm_support_pmu_v3 <= 0)
+ return -ENODEV;
+
+ if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features) ||
+ !kvm_arm_pmu_irq_initialized(vcpu))
+ return -ENXIO;
Would seem more logical for these checks to be in the following order,
if supported
if enabled (has-feature-bit)
if ready
sure, will change.
+
+ kvm_pmu_vcpu_reset(vcpu);
+ vcpu->arch.pmu.ready = true;
+
+ return 0;
+}
+
+static int kvm_arm_pmu_irq_access(struct kvm_vcpu *vcpu,
+ struct kvm_device_attr *attr,
+ int *irq, bool is_set)
+{
+ if (!is_set) {
+ if (!kvm_arm_pmu_irq_initialized(vcpu))
+ return -ENODEV;
+
+ *irq = vcpu->arch.pmu.irq_num;
+ } else {
+ if (kvm_arm_pmu_irq_initialized(vcpu))
+ return -EBUSY;
I didn't look at earlier reviews about this one, but I'm curious why we
can't change an irq for a vcpu?
It intends to avoid changing the irq during the vm running. This will
call wrong interrupt inject since guest already gets the interrupt
number from DTS.
Can't we change it, even if it requires
a pmu reset? Actually, if we did do a reset and a ready=true here, then
could we completely drop KVM_ARM_VCPU_PMU_V3_INIT?
This is suggested by Peter and I agree with him now because it's better
that userspace explictly tell KVM that "I configure the device
compeletly, you could initialize it now".
+
+ kvm_debug("Set kvm ARM PMU irq: %d\n", *irq);
+ vcpu->arch.pmu.irq_num = *irq;
+ }
+
+ return 0;
+}
+
+static bool irq_is_invalid(struct kvm *kvm, int irq, bool is_ppi)
personally I prefer positive checks, i.e. irq_is_valid
+{
+ int i;
+ struct kvm_vcpu *vcpu;
+
+ kvm_for_each_vcpu(i, vcpu, kvm) {
+ if (!kvm_arm_pmu_irq_initialized(vcpu))
+ continue;
+
+ if (is_ppi) {
+ if (vcpu->arch.pmu.irq_num != irq)
+ return true;
+ } else {
+ if (vcpu->arch.pmu.irq_num == irq)
+ return true;
+ }
+ }
+
+ return false;
+}
+
+
+int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr
*attr)
+{
+ switch (attr->attr) {
+ case KVM_ARM_VCPU_PMU_V3_IRQ: {
+ int __user *uaddr = (int __user *)(long)attr->addr;
+ int reg;
+
+ if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+ return -ENODEV;
+
+ if (get_user(reg, uaddr))
+ return -EFAULT;
+
+ /*
+ * The PMU overflow interrupt could be a PPI or SPI, but for one
+ * VM the interrupt type must be same for each vcpu. As a PPI,
+ * the interrupt number is same for all vcpus, while as a SPI it
+ * must be different for each vcpu.
+ */
+ if (reg < VGIC_NR_SGIS || reg >= vcpu->kvm->arch.vgic.nr_irqs ||
+ irq_is_invalid(vcpu->kvm, reg, reg < VGIC_NR_PRIVATE_IRQS))
+ return -EINVAL;
+
+ return kvm_arm_pmu_irq_access(vcpu, attr, ®, true);
+ }
+ case KVM_ARM_VCPU_PMU_V3_INIT:
+ return kvm_arm_pmu_v3_init(vcpu);
+ }
+
+ return -ENXIO;
+}
+
+int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr
*attr)
+{
+ int ret;
+
+ switch (attr->attr) {
+ case KVM_ARM_VCPU_PMU_V3_IRQ: {
+ int __user *uaddr = (int __user *)(long)attr->addr;
+ int reg = -1;
+
+ if (!test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+ return -ENODEV;
+
+ ret = kvm_arm_pmu_irq_access(vcpu, attr, ®, false);
+ if (ret)
+ return ret;
+ return put_user(reg, uaddr);
+ }
+ }
+
+ return -ENXIO;
+}
+
+int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr
*attr)
+{
+ switch (attr->attr) {
+ case KVM_ARM_VCPU_PMU_V3_IRQ:
+ case KVM_ARM_VCPU_PMU_V3_INIT:
+ if (kvm_arm_support_pmu_v3() > 0 &&
+ test_bit(KVM_ARM_VCPU_PMU_V3, vcpu->arch.features))
+ return 0;
+ }
+
+ return -ENXIO;
+}
--
2.0.4
Thanks,
drew
--
Shannon
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
