On Mon, Nov 23, 2009 at 1:16 PM, Albert Herranz [off-list ref] w=
rote:

Grant Likely wrote:

quoted

On Sun, Nov 22, 2009 at 3:01 PM, Albert Herranz <albert_herranz@yahoo.es=

wrote:

quoted

quoted

The Nintendo Wii has two discontiguous RAM memory areas called
MEM1 and MEM2.
MEM1 starts at 0x00000000 and contains 24MB of 1T-SRAM.
MEM2 starts at 0x10000000 and contains 64MB of DDR2 RAM.
Between both memory address ranges there is an address space
where memory-mapped I/O registers are found.

Currently, Linux 32-bit PowerPC does not support RAM in
discontiguous memory address spaces. Thus, in order to use
both RAM areas, we declare as RAM the range from the start of
MEM1 to the end of useable MEM2 and exclude the needed parts
with /memreserve/ statements, at the expense of wasting a bit
of memory.

As a side effect, we need to allow ioremapping RAM areas
because the I/O address space sits within the memreserve'd part
of the declared RAM region.
Note that this is not safe if the region ioremapped is covered
by an existing BAT mapping used to map RAM, so this is
specifically banned here.

Signed-off-by: Albert Herranz <redacted>
---
=A0arch/powerpc/mm/pgtable_32.c | =A0 19 ++++++++++++++++---
=A01 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.=

c

quoted

quoted

index cb96cb2..ba00cb1 100644

--- a/arch/powerpc/mm/pgtable_32.c
+++ b/arch/powerpc/mm/pgtable_32.c

@@ -191,9 +191,22 @@ __ioremap_caller(phys_addr_t addr, unsigned long s=

ize, unsigned long flags,

quoted

quoted

=A0 =A0 =A0 =A0 * Don't allow anybody to remap normal RAM that we're us=

ing.

quoted

quoted

=A0 =A0 =A0 =A0 * mem_init() sets high_memory so only do the check afte=

r that.

quoted

quoted

=A0 =A0 =A0 =A0 */
- =A0 =A0 =A0 if (mem_init_done && (p < virt_to_phys(high_memory))) {
- =A0 =A0 =A0 =A0 =A0 =A0 =A0 printk("__ioremap(): phys addr 0x%llx is =

RAM lr %p\n",

quoted

quoted

- =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0(unsigned long long)p, __b=

uiltin_return_address(0));

quoted

quoted

+ =A0 =A0 =A0 if (mem_init_done && (p < virt_to_phys(high_memory))
+#ifdef CONFIG_WII
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 /*
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* On some systems, though, we may want=

 to remap an area

quoted

quoted

+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* declared as normal RAM that we have =

memreserve'd at the

quoted

quoted

+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* device tree. See wii.dts.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* But we can't do that safely if we ar=

e using BATs to map

quoted

quoted

+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0* part of that area.
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0*/
+ =A0 =A0 =A0 =A0 =A0 && !__map_without_bats
+#endif
+ =A0 =A0 =A0 =A0 =A0 ) {
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 printk(KERN_WARNING
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0"__ioremap(): phys addr 0x=

%llx is RAM lr %p\n",

quoted

quoted

+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0(unsigned long long)p,
+ =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0__builtin_return_addre=

ss(0));

quoted

This could adversely affect multiplatform kernels. =A0I'd rather get the
RAM problem fixed and not hack up common code to work around the hack.

g.

Would it be acceptable to create a global var __allow_ioremap_normal_ram =

that by default would have a value of 0 and would be set _only_ for those p=
latforms needing it?

I'm not the best one to answer this since I don't dig into the mm code
very often.  Ben?  Kumar?  Becky?  Thoughts?

The other solutions I see is:
- add support for discontiguous memory to powerpc 32-bits (which is not s=

omething that I can look into now)

I of course like this option.  :-)

- don't use the precious second 64MB area (which is a waste)

Not exactly nice, but it might be wise to do this now so that
discussion about how to fix it best won't block getting the bulk of
support into mainline.

g.

--=20
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.