> It seems like the highmem additions are causing conceptual
> changes (some definately good) in the ARCHITECTURE INDEPENDANT code.
> While it makes some sense to alter the architecture independant code
> in linux/mm, it also kind of seems like the highmem stuff should be in
> arch/i386.
generally there were no changes in the 'core' kernel that were purely
highmem-motivated. All of the changes were basically cleanups which make
conceptual sense to other architectures as well. Eg. the VM cleanup wrt.
'struct page *' simplified the code by separating virtual and physical
addresses, bootmem allows more efficient use of RAM and a much more
complex RAM setup, the zone-allocator enables NUMA (and other types of
assymetric memory configurations) and uses less RAM. Maybe the only thing
that is unique to highmem is 'kmap(page)/kunmap(page)', but i think the
current interface is low-impact enough.
-- mingo
-
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/