Re: MM with fragmented memory

Werner Almesberger (almesber@lrc.di.epfl.ch)
Thu, 22 Oct 1998 11:57:38 +0200 (MET DST)


Jakub Jelinek wrote:
> BTW: Are those frags at fixed locations or do their addresses depend on
> how somebody stuffed SIMMs into it?

They seem to be fixed on Psions, although I'd like to keep this part generic,
because (1) you never know if there isn't some odd Psion model with a
different layout, and (2) we could then re-use things on the Geofox and
other 7110-based systems.

> Another question: how large VA space do you have and how large PA space the
> machine has?

The CPU has VA = PA = 4 GB, but RAM is confined to the upper 1 GB.

> If e.g. max PA range is 2GB, but VA is 4GB, then you could
> happily use 1:(1+off) mapping (or even 1:1), and map the kernel with a
> virtual mapping to some other VA (so that kernel's 2MB (or so) would have
> two mappings).

Hmm, that's a very interesting idea ! There's one issue with multiple
mappings, though: the cache works with virtual addresses, so if we
access the same page via two different addresses, our cache becomes
inconsistent. Can such a thing ever happen to kernel text, data, bss,
and friends ?

> If the above is not possible, then this should work just fine, although a
> tiny bit slower. That's what sparc32/sun4m does in most cases, so the little
> voice can stay back in your head, such scheme is tested and works.

Excellent. Things look much brighter now :)

Thanks a lot !

- Werner

-- 
  _________________________________________________________________________
 / Werner Almesberger, DI-ICA,EPFL,CH   werner.almesberger@lrc.di.epfl.ch /
/_IN_R_131__Tel_+41_21_693_6621__Fax_+41_21_693_6610_____________________/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/