Re: [PATCH v9 04/12] x86/mm: get INVLPGB count max from CPUID

[Brendan Jackman]

Oh, sorry

On Fri, 7 Feb 2025 at 16:10, Brendan Jackman <jackmanb@xxxxxxxxxx> wrote:
>
> On Thu, 6 Feb 2025 at 05:45, Rik van Riel <riel@xxxxxxxxxxx> wrote:
> > diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
> > index 17b6590748c0..f9b832e971c5 100644
> > --- a/arch/x86/include/asm/cpufeatures.h
> > +++ b/arch/x86/include/asm/cpufeatures.h
> > @@ -338,6 +338,7 @@
> >  #define X86_FEATURE_CLZERO             (13*32+ 0) /* "clzero" CLZERO instruction */
> >  #define X86_FEATURE_IRPERF             (13*32+ 1) /* "irperf" Instructions Retired Count */
> >  #define X86_FEATURE_XSAVEERPTR         (13*32+ 2) /* "xsaveerptr" Always save/restore FP error pointers */
> > +#define X86_FEATURE_INVLPGB            (13*32+ 3) /* INVLPGB and TLBSYNC instruction supported. */
>
> Why no "invlpgb" here? Seems like having this flag visible in cpuinfo
> would be worthwhile.
>
> If there's a reason to hide it maybe add a comment to explain the
> reason? Sorry if this is a stupid question - I also can't see an
> obvious rationale for why existing flags do or don't get a name at
> runtime.

Oh, found it:

https://lore.kernel.org/lkml/20250102120450.GNZ3aA4oVPnoJYRVUL@fat_crate.local/

Sorry for the noise, please ignore.