> and if you had to be precise, the code should do:
>
> if (cpu_feature_enabled(X86_FEATURE_FRED)) {
> if (cpu_feature_enabled(X86_FEATURE_WRMSRNS))
> wrmsrns(MSR_IA32_FRED_RSP0, (unsigned
> long)task_stack_page(task) + THREAD_SIZE);
> else
> wrmsr(MSR_IA32_FRED_RSP0, (unsigned
> long)task_stack_page(task) + THREAD_SIZE);
> }
This is exactly what tglx wanted to avoid.
And I love the idea "baseline", especially we have a ton of CPU features.
>
> > Another patch set should replace WRMSR with WRMSRNS, with SERIALIZE
> > added when needed.
>
> I sense someone wants to optimize MSR writes ... :-)
:-)
