On Mon, Jun 28, 2010 at 03:18, Milton Miller [off-list ref] wrote:

On Fri Jun 25 around 14:01:51 EST 2010 Kyle Moffett wrote:

quoted

I've got a new P2020 (32bit mpc85xx family) board I'm working on a
port for that includes 2 NOR flashes (128MB each) and a removable
SO-RDIMM of 2GB or 4GB. =C2=A0Unfortunately when I configure both flashe=

s

quoted

in the device-tree off my elbc, Linux is completely unable to access
the second one because it attempts to ioremap() the entire virtual
address space of both FLASH chips.

Even with only one flash chip enabled, there's a bit of a noticeable
performance degradation because the mapping consumes almost all of my
available vmalloc space and forces bounce-buffering for all my
HIGHMEM.

It looks like the "of-flash" driver currently requires that the whole
chip be mapped in the kernel at once. =C2=A0I would much rather have a 5=

0%

quoted

performance penalty on flash accesses (which are already very slow)
and regain most of the vmap space.

So the question is, is there a way to convince the MTD layer to
iomap() only what it needs to access to do reads and writes? =C2=A0If no=

t,

quoted

what changes would need to be made to MTD and/or "of-flash" to create
such functionality?

I believe the MTD layer would be happy, but it is beyond the scope of
the physmap_of driver.

Well, I believe that the physmap_of driver is the correct place for
it; that driver is used by a large number of embedded PPC systems and
at present it is completely unable to handle some of the larger flash
memories currently being produced (256MByte+), and there's a decent
system-wide performance penalty (due to lack of ioremap space) even
for the larger flash memories which do function.

Cheers,
Kyle Moffett