Subject: Re: sh4 pmap bug? (Re: port-dreamcast/34243)
To: None <port-sh3@NetBSD.org>
From: Valeriy E. Ushakov <firstname.lastname@example.org>
Date: 08/24/2006 04:24:55
On Thu, Aug 24, 2006 at 09:02:58 +0900, Izumi Tsutsui wrote:
> email@example.com wrote:
> > > Maybe we should check virtual cache index in pmap_pv_enter() and
> > > allow multiple mappings if VA pages being mapped have the same
> > > virtual indexes like mips pmap?
> > My reading of Schimmel's book sections 2.2.5 and 3.3.6 is that for
> > that to work the cache must use write-allocate so that if we write via
> > one alias, we cause the line to be loaded into the cache so that
> > references via the other alias pick up the change.
> AFAIK, SH4 use virtual indexed cache with physical tags,
> so you should refer the section 5.2.6, not 3.3.6.
> I.e. no alias could happen if all mapped pages have the same
> virtual indexes (colors).
Right, but 5.2.6 says:
"... The case of a single process attaching the same shared
memory region two or more times within its own address space is
the same as described in the previous chapters (see Section
firstname.lastname@example.org | Zu Grunde kommen
http://snark.ptc.spbu.ru/~uwe/ | Ist zu Grunde gehen