Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull second batch of kvm updates from Paolo Bonzini:
 "Four changes:

   - x86: work around two nasty cases where a benign exception occurs
     while another is being delivered.  The endless stream of exceptions
     causes an infinite loop in the processor, which not even NMIs or
     SMIs can interrupt; in the virt case, there is no possibility to
     exit to the host either.

   - x86: support for Skylake per-guest TSC rate.  Long supported by
     AMD, the patches mostly move things from there to common
     arch/x86/kvm/ code.

   - generic: remove local_irq_save/restore from the guest entry and
     exit paths when context tracking is enabled.  The patches are a few
     months old, but we discussed them again at kernel summit.  Andy
     will pick up from here and, in 4.5, try to remove it from the user
     entry/exit paths.

   - PPC: Two bug fixes, see merge commit 370289756becc for details"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (21 commits)
  KVM: x86: rename update_db_bp_intercept to update_bp_intercept
  KVM: svm: unconditionally intercept #DB
  KVM: x86: work around infinite loop in microcode when #AC is delivered
  context_tracking: avoid irq_save/irq_restore on guest entry and exit
  context_tracking: remove duplicate enabled check
  KVM: VMX: Dump TSC multiplier in dump_vmcs()
  KVM: VMX: Use a scaled host TSC for guest readings of MSR_IA32_TSC
  KVM: VMX: Setup TSC scaling ratio when a vcpu is loaded
  KVM: VMX: Enable and initialize VMX TSC scaling
  KVM: x86: Use the correct vcpu's TSC rate to compute time scale
  KVM: x86: Move TSC scaling logic out of call-back read_l1_tsc()
  KVM: x86: Move TSC scaling logic out of call-back adjust_tsc_offset()
  KVM: x86: Replace call-back compute_tsc_offset() with a common function
  KVM: x86: Replace call-back set_tsc_khz() with a common function
  KVM: x86: Add a common TSC scaling function
  KVM: x86: Add a common TSC scaling ratio field in kvm_vcpu_arch
  KVM: x86: Collect information for setting TSC scaling ratio
  KVM: x86: declare a few variables as __read_mostly
  KVM: x86: merge handle_mmio_page_fault and handle_mmio_page_fault_common
  KVM: PPC: Book3S HV: Don't dynamically split core when already split
  ...

1 files changed, 137 insertions, 22 deletions

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4a6eff166fc6..00462bd63129 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -93,10 +93,10 @@ static void update_cr8_intercept(struct kvm_vcpu *vcpu);
 static void process_nmi(struct kvm_vcpu *vcpu);
 static void __kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags);
 
-struct kvm_x86_ops *kvm_x86_ops;
+struct kvm_x86_ops *kvm_x86_ops __read_mostly;
 EXPORT_SYMBOL_GPL(kvm_x86_ops);
 
-static bool ignore_msrs = 0;
+static bool __read_mostly ignore_msrs = 0;
 module_param(ignore_msrs, bool, S_IRUGO | S_IWUSR);
 
 unsigned int min_timer_period_us = 500;
@@ -105,20 +105,25 @@ module_param(min_timer_period_us, uint, S_IRUGO | S_IWUSR);
 static bool __read_mostly kvmclock_periodic_sync = true;
 module_param(kvmclock_periodic_sync, bool, S_IRUGO);
 
-bool kvm_has_tsc_control;
+bool __read_mostly kvm_has_tsc_control;
 EXPORT_SYMBOL_GPL(kvm_has_tsc_control);
-u32  kvm_max_guest_tsc_khz;
+u32  __read_mostly kvm_max_guest_tsc_khz;
 EXPORT_SYMBOL_GPL(kvm_max_guest_tsc_khz);
+u8   __read_mostly kvm_tsc_scaling_ratio_frac_bits;
+EXPORT_SYMBOL_GPL(kvm_tsc_scaling_ratio_frac_bits);
+u64  __read_mostly kvm_max_tsc_scaling_ratio;
+EXPORT_SYMBOL_GPL(kvm_max_tsc_scaling_ratio);
+static u64 __read_mostly kvm_default_tsc_scaling_ratio;
 
 /* tsc tolerance in parts per million - default to 1/2 of the NTP threshold */
-static u32 tsc_tolerance_ppm = 250;
+static u32 __read_mostly tsc_tolerance_ppm = 250;
 module_param(tsc_tolerance_ppm, uint, S_IRUGO | S_IWUSR);
 
 /* lapic timer advance (tscdeadline mode only) in nanoseconds */
-unsigned int lapic_timer_advance_ns = 0;
+unsigned int __read_mostly lapic_timer_advance_ns = 0;
 module_param(lapic_timer_advance_ns, uint, S_IRUGO | S_IWUSR);
 
-static bool backwards_tsc_observed = false;
+static bool __read_mostly backwards_tsc_observed = false;
 
 #define KVM_NR_SHARED_MSRS 16
 
@@ -1249,14 +1254,53 @@ static u32 adjust_tsc_khz(u32 khz, s32 ppm)
 	return v;
 }
 
-static void kvm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 this_tsc_khz)
+static int set_tsc_khz(struct kvm_vcpu *vcpu, u32 user_tsc_khz, bool scale)
+{
+	u64 ratio;
+
+	/* Guest TSC same frequency as host TSC? */
+	if (!scale) {
+		vcpu->arch.tsc_scaling_ratio = kvm_default_tsc_scaling_ratio;
+		return 0;
+	}
+
+	/* TSC scaling supported? */
+	if (!kvm_has_tsc_control) {
+		if (user_tsc_khz > tsc_khz) {
+			vcpu->arch.tsc_catchup = 1;
+			vcpu->arch.tsc_always_catchup = 1;
+			return 0;
+		} else {
+			WARN(1, "user requested TSC rate below hardware speed\n");
+			return -1;
+		}
+	}
+
+	/* TSC scaling required  - calculate ratio */
+	ratio = mul_u64_u32_div(1ULL << kvm_tsc_scaling_ratio_frac_bits,
+				user_tsc_khz, tsc_khz);
+
+	if (ratio == 0 || ratio >= kvm_max_tsc_scaling_ratio) {
+		WARN_ONCE(1, "Invalid TSC scaling ratio - virtual-tsc-khz=%u\n",
+			  user_tsc_khz);
+		return -1;
+	}
+
+	vcpu->arch.tsc_scaling_ratio = ratio;
+	return 0;
+}
+
+static int kvm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 this_tsc_khz)
 {
 	u32 thresh_lo, thresh_hi;
 	int use_scaling = 0;
 
 	/* tsc_khz can be zero if TSC calibration fails */
-	if (this_tsc_khz == 0)
-		return;
+	if (this_tsc_khz == 0) {
+		/* set tsc_scaling_ratio to a safe value */
+		vcpu->arch.tsc_scaling_ratio = kvm_default_tsc_scaling_ratio;
+		return -1;
+	}
 
 	/* Compute a scale to convert nanoseconds in TSC cycles */
 	kvm_get_time_scale(this_tsc_khz, NSEC_PER_SEC / 1000,
@@ -1276,7 +1320,7 @@ static void kvm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 this_tsc_khz)
 		pr_debug("kvm: requested TSC rate %u falls outside tolerance [%u,%u]\n", this_tsc_khz, thresh_lo, thresh_hi);
 		use_scaling = 1;
 	}
-	kvm_x86_ops->set_tsc_khz(vcpu, this_tsc_khz, use_scaling);
+	return set_tsc_khz(vcpu, this_tsc_khz, use_scaling);
 }
 
 static u64 compute_guest_tsc(struct kvm_vcpu *vcpu, s64 kernel_ns)
@@ -1322,6 +1366,48 @@ static void update_ia32_tsc_adjust_msr(struct kvm_vcpu *vcpu, s64 offset)
 	vcpu->arch.ia32_tsc_adjust_msr += offset - curr_offset;
 }
 
+/*
+ * Multiply tsc by a fixed point number represented by ratio.
+ *
+ * The most significant 64-N bits (mult) of ratio represent the
+ * integral part of the fixed point number; the remaining N bits
+ * (frac) represent the fractional part, ie. ratio represents a fixed
+ * point number (mult + frac * 2^(-N)).
+ *
+ * N equals to kvm_tsc_scaling_ratio_frac_bits.
+ */
+static inline u64 __scale_tsc(u64 ratio, u64 tsc)
+{
+	return mul_u64_u64_shr(tsc, ratio, kvm_tsc_scaling_ratio_frac_bits);
+}
+
+u64 kvm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc)
+{
+	u64 _tsc = tsc;
+	u64 ratio = vcpu->arch.tsc_scaling_ratio;
+
+	if (ratio != kvm_default_tsc_scaling_ratio)
+		_tsc = __scale_tsc(ratio, tsc);
+
+	return _tsc;
+}
+EXPORT_SYMBOL_GPL(kvm_scale_tsc);
+
+static u64 kvm_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc)
+{
+	u64 tsc;
+
+	tsc = kvm_scale_tsc(vcpu, rdtsc());
+
+	return target_tsc - tsc;
+}
+
+u64 kvm_read_l1_tsc(struct kvm_vcpu *vcpu, u64 host_tsc)
+{
+	return kvm_x86_ops->read_l1_tsc(vcpu, kvm_scale_tsc(vcpu, host_tsc));
+}
+EXPORT_SYMBOL_GPL(kvm_read_l1_tsc);
+
 void kvm_write_tsc(struct kvm_vcpu *vcpu, struct msr_data *msr)
 {
 	struct kvm *kvm = vcpu->kvm;
@@ -1333,7 +1419,7 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, struct msr_data *msr)
 	u64 data = msr->data;
 
 	raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags);
-	offset = kvm_x86_ops->compute_tsc_offset(vcpu, data);
+	offset = kvm_compute_tsc_offset(vcpu, data);
 	ns = get_kernel_ns();
 	elapsed = ns - kvm->arch.last_tsc_nsec;
 
@@ -1390,7 +1476,7 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, struct msr_data *msr)
 		} else {
 			u64 delta = nsec_to_cycles(vcpu, elapsed);
 			data += delta;
-			offset = kvm_x86_ops->compute_tsc_offset(vcpu, data);
+			offset = kvm_compute_tsc_offset(vcpu, data);
 			pr_debug("kvm: adjusted tsc offset by %llu\n", delta);
 		}
 		matched = true;
@@ -1447,6 +1533,20 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, struct msr_data *msr)
 
 EXPORT_SYMBOL_GPL(kvm_write_tsc);
 
+static inline void adjust_tsc_offset_guest(struct kvm_vcpu *vcpu,
+					   s64 adjustment)
+{
+	kvm_x86_ops->adjust_tsc_offset_guest(vcpu, adjustment);
+}
+
+static inline void adjust_tsc_offset_host(struct kvm_vcpu *vcpu, s64 adjustment)
+{
+	if (vcpu->arch.tsc_scaling_ratio != kvm_default_tsc_scaling_ratio)
+		WARN_ON(adjustment < 0);
+	adjustment = kvm_scale_tsc(vcpu, (u64) adjustment);
+	kvm_x86_ops->adjust_tsc_offset_guest(vcpu, adjustment);
+}
+
 #ifdef CONFIG_X86_64
 
 static cycle_t read_tsc(void)
@@ -1608,7 +1708,7 @@ static void kvm_gen_update_masterclock(struct kvm *kvm)
 
 static int kvm_guest_time_update(struct kvm_vcpu *v)
 {
-	unsigned long flags, this_tsc_khz;
+	unsigned long flags, this_tsc_khz, tgt_tsc_khz;
 	struct kvm_vcpu_arch *vcpu = &v->arch;
 	struct kvm_arch *ka = &v->kvm->arch;
 	s64 kernel_ns;
@@ -1645,7 +1745,7 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
 		kernel_ns = get_kernel_ns();
 	}
 
-	tsc_timestamp = kvm_x86_ops->read_l1_tsc(v, host_tsc);
+	tsc_timestamp = kvm_read_l1_tsc(v, host_tsc);
 
 	/*
 	 * We may have to catch up the TSC to match elapsed wall clock
@@ -1671,7 +1771,9 @@ static int kvm_guest_time_update(struct kvm_vcpu *v)
 		return 0;
 
 	if (unlikely(vcpu->hw_tsc_khz != this_tsc_khz)) {
-		kvm_get_time_scale(NSEC_PER_SEC / 1000, this_tsc_khz,
+		tgt_tsc_khz = kvm_has_tsc_control ?
+			vcpu->virtual_tsc_khz : this_tsc_khz;
+		kvm_get_time_scale(NSEC_PER_SEC / 1000, tgt_tsc_khz,
 				   &vcpu->hv_clock.tsc_shift,
 				   &vcpu->hv_clock.tsc_to_system_mul);
 		vcpu->hw_tsc_khz = this_tsc_khz;
@@ -2617,7 +2719,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 		if (tsc_delta < 0)
 			mark_tsc_unstable("KVM discovered backwards TSC");
 		if (check_tsc_unstable()) {
-			u64 offset = kvm_x86_ops->compute_tsc_offset(vcpu,
+			u64 offset = kvm_compute_tsc_offset(vcpu,
 						vcpu->arch.last_guest_tsc);
 			kvm_x86_ops->write_tsc_offset(vcpu, offset);
 			vcpu->arch.tsc_catchup = 1;
@@ -3319,9 +3421,9 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 		if (user_tsc_khz == 0)
 			user_tsc_khz = tsc_khz;
 
-		kvm_set_tsc_khz(vcpu, user_tsc_khz);
+		if (!kvm_set_tsc_khz(vcpu, user_tsc_khz))
+			r = 0;
 
-		r = 0;
 		goto out;
 	}
 	case KVM_GET_TSC_KHZ: {
@@ -6452,8 +6554,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 	if (hw_breakpoint_active())
 		hw_breakpoint_restore();
 
-	vcpu->arch.last_guest_tsc = kvm_x86_ops->read_l1_tsc(vcpu,
-							   rdtsc());
+	vcpu->arch.last_guest_tsc = kvm_read_l1_tsc(vcpu, rdtsc());
 
 	vcpu->mode = OUTSIDE_GUEST_MODE;
 	smp_wmb();
@@ -7015,7 +7116,7 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 	 */
 	kvm_set_rflags(vcpu, rflags);
 
-	kvm_x86_ops->update_db_bp_intercept(vcpu);
+	kvm_x86_ops->update_bp_intercept(vcpu);
 
 	r = 0;
 
@@ -7364,6 +7465,20 @@ int kvm_arch_hardware_setup(void)
 	if (r != 0)
 		return r;
 
+	if (kvm_has_tsc_control) {
+		/*
+		 * Make sure the user can only configure tsc_khz values that
+		 * fit into a signed integer.
+		 * A min value is not calculated needed because it will always
+		 * be 1 on all machines.
+		 */
+		u64 max = min(0x7fffffffULL,
+			      __scale_tsc(kvm_max_tsc_scaling_ratio, tsc_khz));
+		kvm_max_guest_tsc_khz = max;
+
+		kvm_default_tsc_scaling_ratio = 1ULL << kvm_tsc_scaling_ratio_frac_bits;
+	}
+
 	kvm_init_msr_list();
 	return 0;
 }