I think the option #2 is good idea, since it affects as little.
Ingo suggests me (and I've agreed once) that "shift"-ing the kernel
space mapping so that
__va(PHISICAL_MEMORY_START) == PAGE_OFFSET
and changing the assumption of __pa(PAGE_OFFSET) == 0, and I've tried
to change them in SH part.
However, I've found that it results kind of ugly hack for SH. SH has
MMU architecture similar to MIPS. Kernel space resides at 0x80000000
(upto 0x9FFFFFFF), and it maps to physical memory 0x00000000, with
_no_ TLB lookup. And the physical memory usually start at 0x08000000
or 0x0c000000 (depends on the board).
I'd like to provide a patch for the option #2, could you please take a
look at it. Is it OK for you, Ingo?
Later,
-- Niibe Yutaka- 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/