[MODERATED] Re: [PATCH v3 06/10] [PATCH v3 6/9] Linux Patch #6

[Borislav Petkov <bp@suse.de>]

On Mon, Apr 23, 2018 at 10:45:10PM +0200, speck for Jiri Kosina wrote:
> Don't we really want it to appear in /proc/cpuinfo though?
> 
> At the end of the day, it really is about toggling an internal CPU feature 
> behavior, so I'd say it belongs there.
> 
> We even have PTI there, and this is way more actually really internal to 
> the CPU.

Well, the thing is that other CPUs have store bypassing too but they
don't necessarily have the disabling bit.

So if anything, we should use the X86_BUG_CPU_SPEC_STORE_BYPASS
bit so that it appears in the bugs: line.

I mean, one of the two: X86_FEATURE_STBUF_BYPASS or
X86_BUG_CPU_SPEC_STORE_BYPASS is redundant AFAICT so we can stick with
the X86_BUG one.

Also, we have the state in /sys/devices/system/cpu/vulnerabilities/ too,
in case something needs it.

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
--