> -Original Message-
> From: Huang, Kai
> Sent: Friday, May 23, 2025 3:18 AM
> To: Reshetova, Elena ; Hansen, Dave
>
> Cc: Raynor, Scott ; [email protected];
> [email protected]; Scarlata, Vincent R ;
> [email protected]; [email protected]; Annapurve, Vishal
> ; [email protected]; Mallick, Asit K
> ; Aktas, Erdem ; Cai,
> Chong ; [email protected]; linux-
> [email protected]; [email protected]
> Subject: Re: [PATCH v6 2/5] x86/cpufeatures: Add
> X86_FEATURE_SGX_EUPDATESVN feature flag
>
> On Thu, 2025-05-22 at 12:21 +0300, Elena Reshetova wrote:
> > --- a/tools/arch/x86/include/asm/cpufeatures.h
> > +++ b/tools/arch/x86/include/asm/cpufeatures.h
> > @@ -481,6 +481,7 @@
> > #define X86_FEATURE_AMD_HTR_CORES (21*32+ 6) /* Heterogeneous
> Core Topology */
> > #define X86_FEATURE_AMD_WORKLOAD_CLASS (21*32+ 7) /*
> Workload Classification */
> > #define X86_FEATURE_PREFER_YMM (21*32+ 8) /* Avoid ZMM
> registers due to downclocking */
> > +#define X86_FEATURE_SGX_EUPDATESVN (21*32+11) /* Support for
> ENCLS[EUPDATESVN] instruction */
>
> [Sorry for not mentioning in the previous version.]
>
> Nit:
>
> I am not sure we need to change tool headers.
>
> Per commit
>
> f6d9883f8e68 ("tools/include: Sync x86 headers with the kernel sources")
>
> .. and tools/include/uapi/README:
>
> ...
>
> What we are doing now is a third option:
>
>- A software-enforced copy-on-write mechanism of kernel headers to
> tooling, driven by non-fatal warnings on the tooling side build when
> kernel headers get modified:
>
> Warning: Kernel ABI header differences:
> diff -u tools/include/uapi/drm/i915_drm.h include/uapi/drm/i915_drm.h
> diff -u tools/include/uapi/linux/fs.h include/uapi/linux/fs.h
> diff -u tools/include/uapi/linux/kvm.h include/uapi/linux/kvm.h
> ...
>
> The tooling policy is to always pick up the kernel side headers as-is,
> and integate them into the tooling build. The warnings above serve as a
> notification to tooling maintainers that there's changes on the kernel
> side.
>
> We've been using this for many years now, and it might seem hacky, but
> works surprisingly well.
>
> .. I interpret the updating to tools headers is not mandatory (unless building
> tools fails w/o the new feature bit definition which I believe isn't the case
> of
> SGX_UPDATESVN). The tools maintainers will eventually do the sync.
>
> But on the other hand, modifying tools headers in this patch also reduces
> tools
> maintainer's effort in the future.
>
> That being said, I am unclear with the rule here. Perhaps Dave/Ingo can help
> to
> clarify.
Thank you Kai! I am also not sure what is the rule since I have checked before
and different patches to x86/cpufeatures.c do it differently (some do the
updates to tools and some don't).
I also would like to hear suggestions from Dave and Ingo on this.
Best Regards,
Elena.
