On 03/22/2010 11:13 AM, Sheng Yang wrote:
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 06108f3..f971b9b 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1804,9 +1804,15 @@ static u64 construct_eptp(unsigned long root_hpa) { u64 eptp; - /* TODO write the value reading from MSR */ - eptp = VMX_EPT_DEFAULT_MT | - VMX_EPT_DEFAULT_GAW<< VMX_EPT_GAW_EPTP_SHIFT; + if (cpu_has_vmx_eptp_writeback()) + eptp = VMX_EPT_MT_WRBACK | + VMX_EPT_DEFAULT_GAW<< VMX_EPT_GAW_EPTP_SHIFT;I prefer to ensure WB is supported and used as default. Otherwise it would be a big trouble for memory subsystem(to use UC for all memory). Both WB and UC EPT memory types are ensured to be support in hardware. And you can remove VMX_EPT_DEFAULT_MT as well.
I agree, hopefully we never ever see a cpu that doesn't support EPT WB. -- error compiling committee.c: too many arguments to function -- 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
