Re: [PATCH 4.19 72/98] x86/CPU/AMD: Clear RDRAND CPUID bit on AMD family 15h/16h

[Pavel Machek]

Hi!

> > > There is no way to reinitialize RDRAND from the kernel otherwise we would
> > > have exactly done that. If you know how to do that please tell.
> > 
> > Would they? AMD is not exactly doing good job with communication
> 
> Yes they would. Stop making up weird conspiracy theories.

> > here. If BIOS can do it, kernel can do it, too...
> 
> May I recommend to read up on SMM and BIOS being able to lock down access
> to certain facilities?
> 
> > or do you have information saying otherwise?
> 
> Yes. It was clearly stated by Tom that it can only be done in the
> BIOS.

Do you have a link for that? Because I don't think I seen that one.

> > > Also disabling it for every BIOS is the only way which can be done because
> > > there is no way to know whether the BIOS is fixed or not at cold boot
> > > time. And it has to be known there because applications cache the
> > 
> > I'm pretty sure DMI-based whitelist would help here. It should be
> > reasonably to fill it with the common machines at least.
> 
> Send patches to that effect.

Why should it be my job? AMD screwed this up, they should fix it
properly. And you should insist on proper fix.

> > Plus, where is the CVE, and does AMD do anything to make BIOS vendors
> > fix them?
> 
> May I redirect you to: https://www.amd.com/en/corporate/contact

That will certainly make communication easier, right.

								Pavel
-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

Attachment: signature.asc
Description: Digital signature