Messages in this thread | | | Subject | Re: [PATCH v2 1/8] x86/fpu/xstate: Define new macros for supervisor and user xstates | From | Yu-cheng Yu <> | Date | Thu, 20 Feb 2020 12:23:00 -0800 |
| |
On Thu, 2020-02-20 at 12:47 +0100, Borislav Petkov wrote: > On Tue, Jan 21, 2020 at 12:18:36PM -0800, Yu-cheng Yu wrote: > > From: Fenghua Yu <fenghua.yu@intel.com> > > [...] > > +/* All currently supported supervisor features */ > > +#define SUPPORTED_XFEATURES_MASK_SUPERVISOR (0) > > + > > +/* > > + * Unsupported supervisor features. When a supervisor feature in this mask is > > + * supported in the future, move it to the supported supervisor feature mask. > > + */ > > +#define UNSUPPORTED_XFEATURES_MASK_SUPERVISOR (XFEATURE_MASK_PT) > > + > > +/* All supervisor states including supported and unsupported states. */ > > +#define ALL_XFEATURES_MASK_SUPERVISOR (SUPPORTED_XFEATURES_MASK_SUPERVISOR | \ > > + UNSUPPORTED_XFEATURES_MASK_SUPERVISOR) > > So frankly having the namespace prepended in those macros makes it more > readable to me: you know that those masks all belong together if you had > this: > > XFEATURE_MASK_SUPERVISOR > XFEATURE_MASK_SUPERVISOR_SUPPORTED > XFEATURE_MASK_SUPERVISOR_UNSUPPORTED > XFEATURE_MASK_SUPERVISOR_ALL > XFEATURE_MASK_USER_SUPPORTED > > Now they all begin with different words: "ALL", "UNSUPPORTED", > "SUPPORTED", ... and makes you go and look up the mask to make sure it > is the correct type of mask used. > > Even more so if the single feature masks also start with > "XFEATURE_MASK_" so it is only logical to have them all start with > XFEATURE_MASK_ IMO.
I will make the changes.
Yu-cheng
| |