Guest writes to MSR 0x2f8 are considered valid (no #GP) but trigger a
WARN_ON in kvm_valid_mtrr and do not get handled in any meaningful
way:
29 static bool msr_mtrr_valid(unsigned msr)
30 {
31 switch (msr) {
32 case 0x200 ... 0x200 + 2 * KVM_NR_VAR_MTRR - 1:
33 case MSR_MTRRfix64K_00000:
...
45 case MSR_IA32_CR_PAT:
46 return true;
47 case 0x2f8:
48 return true;
49 }
50 return false;
51 }
63 bool kvm_mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data)
64 {
65 int i;
66 u64 mask;
67
68 if (!msr_mtrr_valid(msr))
69 return false;
...
87 /* variable MTRRs */
88 WARN_ON(!(msr >= 0x200 && msr < 0x200 + 2 * KVM_NR_VAR_MTRR));
0x2f8 is not a valid MSR on any known Intel CPUs and I could not find
any reference to it in AMD's SDM either. KVM support for 0x2f8 traces
back to this commit:
http://marc.info/?l=git-commits-head&m=121661643317231&w=2
Avi, do you remember why you added this MSR? Did ESX use it?
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
