KVM: X86: Implement userspace interface to set virtual_tsc_khz

This patch implements two new vm-ioctls to get and set the
virtual_tsc_khz if the machine supports tsc-scaling. Setting
the tsc-frequency is only possible before userspace creates
any vcpu.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Avi Kivity <avi@redhat.com>

2 files changed, 55 insertions, 0 deletions

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 8c4549bef4ed..a98873762433 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -64,6 +64,8 @@ MODULE_LICENSE("GPL");
 #define DEBUGCTL_RESERVED_BITS (~(0x3fULL))
 
 #define TSC_RATIO_RSVD          0xffffff0000000000ULL
+#define TSC_RATIO_MIN           0x0000000000000001ULL
+#define TSC_RATIO_MAX           0x000000ffffffffffULL
 
 static bool erratum_383_found __read_mostly;
 
@@ -189,6 +191,7 @@ static int nested_svm_intercept(struct vcpu_svm *svm);
 static int nested_svm_vmexit(struct vcpu_svm *svm);
 static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr,
                                       bool has_error_code, u32 error_code);
+static u64 __scale_tsc(u64 ratio, u64 tsc);
 
 enum {
         VMCB_INTERCEPTS, /* Intercept vectors, TSC offset,
@@ -798,6 +801,23 @@ static __init int svm_hardware_setup(void)
         if (boot_cpu_has(X86_FEATURE_FXSR_OPT))
                 kvm_enable_efer_bits(EFER_FFXSR);
 
+        if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) {
+                u64 max;
+
+                kvm_has_tsc_control = true;
+
+                /*
+                 * 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 and a value of 0 is used to disable
+                 * tsc-scaling for the vcpu.
+                 */
+                max = min(0x7fffffffULL, __scale_tsc(tsc_khz, TSC_RATIO_MAX));
+
+                kvm_max_guest_tsc_khz = max;
+        }
+
         if (nested) {
                 printk(KERN_INFO "kvm: Nested Virtualization enabled\n");
                 kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 579ce34e7904..1d5a7f418795 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -87,6 +87,11 @@ EXPORT_SYMBOL_GPL(kvm_x86_ops);
 int ignore_msrs = 0;
 module_param_named(ignore_msrs, ignore_msrs, bool, S_IRUGO | S_IWUSR);
 
+bool kvm_has_tsc_control;
+EXPORT_SYMBOL_GPL(kvm_has_tsc_control);
+u32  kvm_max_guest_tsc_khz;
+EXPORT_SYMBOL_GPL(kvm_max_guest_tsc_khz);
+
 #define KVM_NR_SHARED_MSRS 16
 
 struct kvm_shared_msrs_global {
@@ -1986,6 +1991,7 @@ int kvm_dev_ioctl_check_extension(long ext)
         case KVM_CAP_X86_ROBUST_SINGLESTEP:
         case KVM_CAP_XSAVE:
         case KVM_CAP_ASYNC_PF:
+        case KVM_CAP_GET_TSC_KHZ:
                 r = 1;
                 break;
         case KVM_CAP_COALESCED_MMIO:
@@ -2012,6 +2018,9 @@ int kvm_dev_ioctl_check_extension(long ext)
         case KVM_CAP_XCRS:
                 r = cpu_has_xsave;
                 break;
+        case KVM_CAP_TSC_CONTROL:
+                r = kvm_has_tsc_control;
+                break;
         default:
                 r = 0;
                 break;
@@ -3045,6 +3054,32 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
                 r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, u.xcrs);
                 break;
         }
+        case KVM_SET_TSC_KHZ: {
+                u32 user_tsc_khz;
+
+                r = -EINVAL;
+                if (!kvm_has_tsc_control)
+                        break;
+
+                user_tsc_khz = (u32)arg;
+
+                if (user_tsc_khz >= kvm_max_guest_tsc_khz)
+                        goto out;
+
+                kvm_x86_ops->set_tsc_khz(vcpu, user_tsc_khz);
+
+                r = 0;
+                goto out;
+        }
+        case KVM_GET_TSC_KHZ: {
+                r = -EIO;
+                if (check_tsc_unstable())
+                        goto out;
+
+                r = vcpu_tsc_khz(vcpu);
+
+                goto out;
+        }
         default:
                 r = -EINVAL;
         }