1 files changed, 172 insertions, 17 deletions

diff --git a/arch/riscv/kvm/vcpu_timer.c b/arch/riscv/kvm/vcpu_timer.c
index 5c4c37ff2d48..85a7262115e1 100644
--- a/arch/riscv/kvm/vcpu_timer.c
+++ b/arch/riscv/kvm/vcpu_timer.c
@@ -11,8 +11,8 @@
 #include <linux/kvm_host.h>
 #include <linux/uaccess.h>
 #include <clocksource/timer-riscv.h>
-#include <asm/csr.h>
 #include <asm/delay.h>
+#include <asm/kvm_nacl.h>
 #include <asm/kvm_vcpu_timer.h>
 
 static u64 kvm_riscv_current_cycles(struct kvm_guest_timer *gt)
@@ -69,7 +69,18 @@ static int kvm_riscv_vcpu_timer_cancel(struct kvm_vcpu_timer *t)
 	return 0;
 }
 
-int kvm_riscv_vcpu_timer_next_event(struct kvm_vcpu *vcpu, u64 ncycles)
+static int kvm_riscv_vcpu_update_vstimecmp(struct kvm_vcpu *vcpu, u64 ncycles)
+{
+#if defined(CONFIG_32BIT)
+	ncsr_write(CSR_VSTIMECMP, ncycles & 0xFFFFFFFF);
+	ncsr_write(CSR_VSTIMECMPH, ncycles >> 32);
+#else
+	ncsr_write(CSR_VSTIMECMP, ncycles);
+#endif
+	return 0;
+}
+
+static int kvm_riscv_vcpu_update_hrtimer(struct kvm_vcpu *vcpu, u64 ncycles)
 {
 	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
 	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
@@ -88,6 +99,63 @@ int kvm_riscv_vcpu_timer_next_event(struct kvm_vcpu *vcpu, u64 ncycles)
 	return 0;
 }
 
+int kvm_riscv_vcpu_timer_next_event(struct kvm_vcpu *vcpu, u64 ncycles)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	return t->timer_next_event(vcpu, ncycles);
+}
+
+static enum hrtimer_restart kvm_riscv_vcpu_vstimer_expired(struct hrtimer *h)
+{
+	u64 delta_ns;
+	struct kvm_vcpu_timer *t = container_of(h, struct kvm_vcpu_timer, hrt);
+	struct kvm_vcpu *vcpu = container_of(t, struct kvm_vcpu, arch.timer);
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+
+	if (kvm_riscv_current_cycles(gt) < t->next_cycles) {
+		delta_ns = kvm_riscv_delta_cycles2ns(t->next_cycles, gt, t);
+		hrtimer_forward_now(&t->hrt, ktime_set(0, delta_ns));
+		return HRTIMER_RESTART;
+	}
+
+	t->next_set = false;
+	kvm_vcpu_kick(vcpu);
+
+	return HRTIMER_NORESTART;
+}
+
+bool kvm_riscv_vcpu_timer_pending(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+
+	if (!kvm_riscv_delta_cycles2ns(t->next_cycles, gt, t) ||
+	    kvm_riscv_vcpu_has_interrupts(vcpu, 1UL << IRQ_VS_TIMER))
+		return true;
+	else
+		return false;
+}
+
+static void kvm_riscv_vcpu_timer_blocking(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
+	u64 delta_ns;
+
+	if (!t->init_done)
+		return;
+
+	delta_ns = kvm_riscv_delta_cycles2ns(t->next_cycles, gt, t);
+	hrtimer_start(&t->hrt, ktime_set(0, delta_ns), HRTIMER_MODE_REL);
+	t->next_set = true;
+}
+
+static void kvm_riscv_vcpu_timer_unblocking(struct kvm_vcpu *vcpu)
+{
+	kvm_riscv_vcpu_timer_cancel(&vcpu->arch.timer);
+}
+
 int kvm_riscv_vcpu_get_reg_timer(struct kvm_vcpu *vcpu,
 				 const struct kvm_one_reg *reg)
 {
@@ -102,7 +170,7 @@ int kvm_riscv_vcpu_get_reg_timer(struct kvm_vcpu *vcpu,
 	if (KVM_REG_SIZE(reg->id) != sizeof(u64))
 		return -EINVAL;
 	if (reg_num >= sizeof(struct kvm_riscv_timer) / sizeof(u64))
-		return -EINVAL;
+		return -ENOENT;
 
 	switch (reg_num) {
 	case KVM_REG_RISCV_TIMER_REG(frequency):
@@ -119,7 +187,7 @@ int kvm_riscv_vcpu_get_reg_timer(struct kvm_vcpu *vcpu,
 					  KVM_RISCV_TIMER_STATE_OFF;
 		break;
 	default:
-		return -EINVAL;
+		return -ENOENT;
 	}
 
 	if (copy_to_user(uaddr, &reg_val, KVM_REG_SIZE(reg->id)))
@@ -143,14 +211,15 @@ int kvm_riscv_vcpu_set_reg_timer(struct kvm_vcpu *vcpu,
 	if (KVM_REG_SIZE(reg->id) != sizeof(u64))
 		return -EINVAL;
 	if (reg_num >= sizeof(struct kvm_riscv_timer) / sizeof(u64))
-		return -EINVAL;
+		return -ENOENT;
 
 	if (copy_from_user(&reg_val, uaddr, KVM_REG_SIZE(reg->id)))
 		return -EFAULT;
 
 	switch (reg_num) {
 	case KVM_REG_RISCV_TIMER_REG(frequency):
-		ret = -EOPNOTSUPP;
+		if (reg_val != riscv_timebase)
+			return -EINVAL;
 		break;
 	case KVM_REG_RISCV_TIMER_REG(time):
 		gt->time_delta = reg_val - get_cycles64();
@@ -165,7 +234,7 @@ int kvm_riscv_vcpu_set_reg_timer(struct kvm_vcpu *vcpu,
 			ret = kvm_riscv_vcpu_timer_cancel(t);
 		break;
 	default:
-		ret = -EINVAL;
+		ret = -ENOENT;
 		break;
 	}
 
@@ -179,11 +248,22 @@ int kvm_riscv_vcpu_timer_init(struct kvm_vcpu *vcpu)
 	if (t->init_done)
 		return -EINVAL;
 
-	hrtimer_init(&t->hrt, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
-	t->hrt.function = kvm_riscv_vcpu_hrtimer_expired;
 	t->init_done = true;
 	t->next_set = false;
 
+	/* Enable sstc for every vcpu if available in hardware */
+	if (riscv_isa_extension_available(NULL, SSTC)) {
+		t->sstc_enabled = true;
+		hrtimer_setup(&t->hrt, kvm_riscv_vcpu_vstimer_expired, CLOCK_MONOTONIC,
+			      HRTIMER_MODE_REL);
+		t->timer_next_event = kvm_riscv_vcpu_update_vstimecmp;
+	} else {
+		t->sstc_enabled = false;
+		hrtimer_setup(&t->hrt, kvm_riscv_vcpu_hrtimer_expired, CLOCK_MONOTONIC,
+			      HRTIMER_MODE_REL);
+		t->timer_next_event = kvm_riscv_vcpu_update_hrtimer;
+	}
+
 	return 0;
 }
 
@@ -199,27 +279,102 @@ int kvm_riscv_vcpu_timer_deinit(struct kvm_vcpu *vcpu)
 
 int kvm_riscv_vcpu_timer_reset(struct kvm_vcpu *vcpu)
 {
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	t->next_cycles = -1ULL;
 	return kvm_riscv_vcpu_timer_cancel(&vcpu->arch.timer);
 }
 
-void kvm_riscv_vcpu_timer_restore(struct kvm_vcpu *vcpu)
+static void kvm_riscv_vcpu_update_timedelta(struct kvm_vcpu *vcpu)
 {
 	struct kvm_guest_timer *gt = &vcpu->kvm->arch.timer;
 
-#ifdef CONFIG_64BIT
-	csr_write(CSR_HTIMEDELTA, gt->time_delta);
+#if defined(CONFIG_32BIT)
+	ncsr_write(CSR_HTIMEDELTA, (u32)(gt->time_delta));
+	ncsr_write(CSR_HTIMEDELTAH, (u32)(gt->time_delta >> 32));
+#else
+	ncsr_write(CSR_HTIMEDELTA, gt->time_delta);
+#endif
+}
+
+void kvm_riscv_vcpu_timer_restore(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	kvm_riscv_vcpu_update_timedelta(vcpu);
+
+	if (!t->sstc_enabled)
+		return;
+
+#if defined(CONFIG_32BIT)
+	ncsr_write(CSR_VSTIMECMP, (u32)t->next_cycles);
+	ncsr_write(CSR_VSTIMECMPH, (u32)(t->next_cycles >> 32));
+#else
+	ncsr_write(CSR_VSTIMECMP, t->next_cycles);
+#endif
+
+	/* timer should be enabled for the remaining operations */
+	if (unlikely(!t->init_done))
+		return;
+
+	kvm_riscv_vcpu_timer_unblocking(vcpu);
+}
+
+void kvm_riscv_vcpu_timer_sync(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	if (!t->sstc_enabled)
+		return;
+
+#if defined(CONFIG_32BIT)
+	t->next_cycles = ncsr_read(CSR_VSTIMECMP);
+	t->next_cycles |= (u64)ncsr_read(CSR_VSTIMECMPH) << 32;
 #else
-	csr_write(CSR_HTIMEDELTA, (u32)(gt->time_delta));
-	csr_write(CSR_HTIMEDELTAH, (u32)(gt->time_delta >> 32));
+	t->next_cycles = ncsr_read(CSR_VSTIMECMP);
 #endif
 }
 
-int kvm_riscv_guest_timer_init(struct kvm *kvm)
+void kvm_riscv_vcpu_timer_save(struct kvm_vcpu *vcpu)
+{
+	struct kvm_vcpu_timer *t = &vcpu->arch.timer;
+
+	if (!t->sstc_enabled)
+		return;
+
+	/*
+	 * The vstimecmp CSRs are saved by kvm_riscv_vcpu_timer_sync()
+	 * upon every VM exit so no need to save here.
+	 *
+	 * If VS-timer expires when no VCPU running on a host CPU then
+	 * WFI executed by such host CPU will be effective NOP resulting
+	 * in no power savings. This is because as-per RISC-V Privileged
+	 * specificaiton: "WFI is also required to resume execution for
+	 * locally enabled interrupts pending at any privilege level,
+	 * regardless of the global interrupt enable at each privilege
+	 * level."
+	 *
+	 * To address the above issue, vstimecmp CSR must be set to -1UL
+	 * over here when VCPU is scheduled-out or exits to user space.
+	 */
+
+	csr_write(CSR_VSTIMECMP, -1UL);
+#if defined(CONFIG_32BIT)
+	csr_write(CSR_VSTIMECMPH, -1UL);
+#endif
+
+	/* timer should be enabled for the remaining operations */
+	if (unlikely(!t->init_done))
+		return;
+
+	if (kvm_vcpu_is_blocking(vcpu))
+		kvm_riscv_vcpu_timer_blocking(vcpu);
+}
+
+void kvm_riscv_guest_timer_init(struct kvm *kvm)
 {
 	struct kvm_guest_timer *gt = &kvm->arch.timer;
 
 	riscv_cs_get_mult_shift(&gt->nsec_mult, &gt->nsec_shift);
 	gt->time_delta = -get_cycles64();
-
-	return 0;
 }