On Tue, Sep 24, 2019 at 09:11:50AM -0700, Sean Christopherson wrote:
> With respect to more SGX feature flags, the original changelog even
> stated "with more expected in the not-too-distant future".
That means nothing, you know that, right? :)
There's a big difference between expectation and it actually happening
and besides, the longterm plan with all those feature words which are
scattered, is to propagate them to proper ->x86_capability[] words once
the number of feature bits used is gradually growing.
Also,...
> I'm not arguing that this isn't ugly, just want to make it clear that
> we're not wantonly throwing junk into the kernel. I'm all for a dedicated
> SGX word, it makes our lives easier.
... you didn't do the first-8-bits-need-to-match-the-CPUID-leaf for KVM
thing then, you're doing now. Which would make word 8 half-hard-coded
and the other half Linux-defined.
Which makes a separate leaf look much better now. :)
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
