Re: Yet another KPTI regression with 4.14.x series in a VM

[Borislav Petkov]

On Sat, Jan 13, 2018 at 01:50:55PM +0100, Thomas Gleixner wrote:
>  2:0 Ignored
>    3 PWT
>    4 PCD

Btw, those last two are "(implies PCD=PWT=0)" according to
http://www.sandpile.org/x86/crx.htm with PCID. I was wondering recently
what happens with those bits when PCID is enabled and CR3[11:0] is the
ASID. I couldn't find it in the SDM.

And it kinda makes sense, PCD=1b is probably only for testing some
hardware crud and not really sensible for normal production.

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.