[PATCH] kvm: clear guest TSC on reset

["Fernando Luis Vázquez Cao" <fernando_b1@lab.ntt.co.jp>]

VCPU TSC is not cleared by a warm reset (*), which leaves many Linux
guests vulnerable to the overflow in cyc2ns_offset fixed by upstream
commit 9993bc635d01a6ee7f6b833b4ee65ce7c06350b1 ("sched/x86: Fix overflow
in cyc2ns_offset").

To put it in a nutshell, if a Linux guest without the patch above applied
has been up more than 208 days and attempts a warm reset chances are that
the newly booted kernel will panic or hang.

(*) Intel Xeon E5 processors show the same broken behavior due to
    the errata "TSC is Not Affected by Warm Reset" (Intel® Xeon®
    Processor E5 Family Specification Update - August 2013): "The
    TSC (Time Stamp Counter MSR 10H) should be cleared on
    reset. Due to this erratum the TSC is not affected by warm
    reset."

Cc: stable@vger.kernel.org
Cc: Will Auld <will.auld@intel.com>
Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
---

diff -urNp linux-3.13-rc2-orig/arch/x86/kvm/x86.c linux-3.13-rc2/arch/x86/kvm/x86.c
--- linux-3.13-rc2-orig/arch/x86/kvm/x86.c	2013-11-30 05:57:14.000000000 +0900
+++ linux-3.13-rc2/arch/x86/kvm/x86.c	2013-12-03 14:51:53.747600839 +0900
@@ -6716,18 +6716,24 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu
 	return r;
 }
 
-int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
+static void kvm_tsc_reset(struct kvm_vcpu *vcpu)
 {
-	int r;
 	struct msr_data msr;
 
-	r = vcpu_load(vcpu);
-	if (r)
-		return r;
 	msr.data = 0x0;
 	msr.index = MSR_IA32_TSC;
 	msr.host_initiated = true;
 	kvm_write_tsc(vcpu, &msr);
+}
+
+int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu)
+{
+	int r;
+
+	r = vcpu_load(vcpu);
+	if (r)
+		return r;
+	kvm_tsc_reset(vcpu);
 	vcpu_put(vcpu);
 
 	return r;
@@ -6770,6 +6776,10 @@ void kvm_vcpu_reset(struct kvm_vcpu *vcp
 
 	kvm_pmu_reset(vcpu);
 
+	kvm_tsc_reset(vcpu);
+	if (guest_cpuid_has_tsc_adjust(vcpu))
+		vcpu->arch.ia32_tsc_adjust_msr = 0x0;
+
 	memset(vcpu->arch.regs, 0, sizeof(vcpu->arch.regs));
 	vcpu->arch.regs_avail = ~0;
 	vcpu->arch.regs_dirty = ~0;