On Fri, 2019-03-08 at 08:54 +0100, Paolo Bonzini wrote: > On 08/03/19 07:10, Xiaoyao Li wrote: > > > so that non-virtualizable features are hidden and > > > > > > if (boot_cpu_has(X86_FEATURE_SPLIT_LOCK_DETECT)) > > > data |= CORE_CAP_SPLIT_LOCK_DETECT; > > > > > > so that userspace gets "for free" the FMS list that will be added > > > later to the kernel. > > > > I think it's redundant. Because there is no case that > > rdmsrl_safe(MSR_IA32_CORE_CAPABILITY, &data) shows no split lock detection > > while > > boot_cpu_has(X86_FEATURE_SPLIT_LOCK_DETECT). > > There will be when the kernel will add a list of FMS values that have > split lock detection but lack the core capabilities MSR. Or at least > that is what Fenghua said in the cover letter. > > Paolo Got it. Thanks for your explanation.
