Subject: Re: pmap_syncicache question
To: Matt Thomas <>
From: Ramesh BN <>
List: port-powerpc
Date: 09/23/2003 13:01:04

Thanks for the reply. Actually i was looking
at the function in 1.5.x and 1.4.x where the 
MMU is not disabled.Is there a 1:1 translation 
maintained in Netbsd 1.4.x and 1.5.x for 
managed physical memory for every mmu context ?



--- Matt Thomas <> wrote:
> At 12:02 PM 9/23/2003, Ramesh BN wrote:
> >Hi,
> >
> >pmap_syncicache() take in a "physical" address and
> >tries
> >to flush the data cache and invalidate the
> instruction
> >cache using the instructions "dcbst" and "icbi".
> >
> >The description of "dcbst" says it needs a EA, and
> >translation and protection will be checked. How is
> >this supposed to work if we use a "pa" instead of
> >a "va"
> instruction and data relocation are disabled during
> the synchronization.  (previously it relied on the
> fact that BATs are used as a very small TLB so all
> physical memory could be accessed by swapping BAT
> entries).
> -- 
> Matt Thomas                     email:
> 3am Software Foundry              www:
> Cupertino, CA              disclaimer: I avow all
> knowledge of this message.

Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software