lkml.org 
[lkml]   [2014]   [Sep]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] x86: Quark: Flush TLB via CR3 not CR4.PGE in setup_arch()

* Henrique de Moraes Holschuh <hmh@hmh.eng.br> wrote:

> On Thu, 25 Sep 2014, Ingo Molnar wrote:
> > > If we're adding a comment though the first thing the comment ought to say is
> > > what the code does for everybody else - stuff CR3 and flush the TLB, then it
> > > should comment on the exception for Quark.
> > >
> > > /*
> > > * Locate the page directory and flush the TLB.
> > > *
> > > * On Quark CPUs we still have the PGE bit set so
> > > * __flush_tlb_all() is not yet doing what it says - but
> > > * accidentally we have a cr3 flush here which is what is
> > > * needed - so there's no need to add a Quark quirk here.
> > > */
> > >
> > > ?
> >
> > Yeah, fair enough. You can even put the latter in parentheses, to
> > signal that it's all a rare case.
>
> I'd have mentioned "erratum" there, otherwise people won't understand why
> the PGE bit being set would be a problem in the first place.
>
> Something like:
>
> /*
> * Locate the page directory and flush the TLB.
> *
> * On Quark X1000 CPUs we still have the PGE bit incorrectly set
> * due to a processor erratum, so __flush_tlb_all() is not yet
> * doing what it says. Fortunately we have a cr3 flush here,
> * which is what is needed in this processor to flush TLBs, so
> * there's no need to add a Quark X1000 quirk here.
> */

Yeah.

I'd also add the fact that it's an unintended erratum to the
Quark quirk section of early_init_intel() as well.

Thanks,

Ingo


\
 
 \ /
  Last update: 2014-09-25 21:01    [W:0.043 / U:1.764 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site