Messages in this thread | | | Date | Fri, 28 Aug 2015 07:31:47 -0700 | From | Dave Hansen <> | Subject | Re: [PATCH 09/11] x86, fpu: correct and check XSAVE xstate size calculations |
| |
On 08/27/2015 09:54 PM, Ingo Molnar wrote: > > * Dave Hansen <dave@sr71.net> wrote: > >> +static int xfeature_is_supervisor(int xfeature_nr) >> +{ >> + /* >> + * We currently do not suport supervisor states, but if >> + * we did, we could find out like this. >> + * >> + * SDM says: If state component i is a user state component, >> + * ECX[0] return 0; if state component i is a supervisor >> + * state component, ECX[0] returns 1. >> + u32 eax, ebx, ecx, edx; >> + cpuid_count(XSTATE_CPUID, xfeature_nr, &eax, &ebx, &ecx, &edx); >> + return !!(ecx & 1); >> + */ >> + return 0; >> +} > > So if this CPUID is documented to work, why not use it to sanity check things? > > I.e. do something like: > > u32 eax, ebx, ecx, edx; > > cpuid_count(XSTATE_CPUID, xfeature_nr, &eax, &ebx, &ecx, &edx); > > /* Linux doesn't support supervisor states (yet): */ > WARN_ON_ONCE(ecx & 1); > > return 0; > > That would give us a gentle way to double check our assumptions here.
Actually, the newest state that you will see in the wild is for Processor Trace, and it _is_ a supervisor state. However, we don't use it in Linux for our Processor Trace support, and Andi says we probably never will.
So we probably shouldn't warn on it.
| |