Subject: Re: PXA250 Xscale and page tables
To: Martin Husemann <martin@duskware.de>
From: Allen Briggs <briggs@wasabisystems.com>
List: tech-kern
Date: 02/01/2003 14:51:11
On Sat, Feb 01, 2003 at 08:43:16PM +0100, Martin Husemann wrote:
> On Sat, Feb 01, 2003 at 10:13:55AM -0700, Colin Cook wrote:
> > What I want to know is why was this in there in the first place. Turning off
> > the cache is a very bad thing to do.
> I think it's there to avoid illegal cache aliases, PVF_NC means
> mapping is non-cacheable, there are multiple mappings (for this VA).

Precisely.  See the comments above pmap_vac_me_harder() earlier in the
file.

> But at first glance it's not clear to me if the test for this bit is
> inverted.

The test there isn't inverted, but it's still possible that something's
awry if all user pages end up that way.  Simply removing that line is
certainly not the correct solution, but if the code's ending up there
when it shouldn't, there is something wrong.

-allen

-- 
 Allen Briggs                     briggs@wasabisystems.com
 http://www.wasabisystems.com/    Quality NetBSD CDs, Sales, Support, Service
NetBSD development for Alpha, ARM, M68K, MIPS, PowerPC, SuperH, XScale, etc...