Re: [PATCH 0/3] x86/mtrr, pat: make PAT independent from MTRR

[Kani, Toshi]

On Tue, 2019-08-13 at 00:49 -0700, Isaku Yamahata wrote:
> On Fri, Aug 09, 2019 at 07:51:17PM +0000,
> "Kani, Toshi" <toshi.kani@xxxxxxx> wrote:
> 
> > On Fri, 2019-08-09 at 09:06 +0200, Borislav Petkov wrote:
> > > On Thu, Aug 08, 2019 at 08:54:17PM -0700, Isaku Yamahata wrote:
> > > > Make PAT(Page Attribute Table) independent from
> > > > MTRR(Memory Type Range Register).
> > > > Some environments (mainly virtual ones) support only PAT, but not MTRR
> > > > because PAT replaces MTRR.
> > > > It's tricky and no gain to support both MTRR and PAT except compatibility.
> > > > So some VM technologies don't support MTRR, but only PAT.
> > 
> > I do not think it is technically correct on bare metal.  AFAIK, MTRR is
> > still the only way to setup cache attribute in real-mode, which BIOS SMI
> > handler relies on in SMM.
> 
> Then you're claiming if it's baremetal, both MTRR and PAT should be
> enabled/disabled at the same time?

No, I did not say that. My point:
- Your statement of MTTR being useless is not correct. It's still used.
The OS should leave the MTTR hand-off state.

I agree with you in general that PAT and MTTR init should be
independent. However, as Boris said, please verify the impact of your
change. As I mentioned in the Xen's example, hypervisor may have non-
default PAT hand-off setting.

Thanks,
-Toshi