[PATCH v3 RESEND 08/17] ARM: LPAE: use phys_addr_t in free_memmap()

On Mon, Sep 24, 2012 at 05:55:55PM +0100, Russell King - ARM Linux wrote:

On Mon, Sep 24, 2012 at 12:41:33PM -0400, Cyril Chemparathy wrote:

quoted

It appears to be not that hard actually... Or maybe I'm totally missing  
your point.  Could it be that you last looked at this prior to the  
nobootmem compatibility stuff being added in?

Quite possible - we converted over to memblock here:

commit 2778f62056ada442414392d7ccd41188bb631619
Author: Russell King [off-list ref]
Date:   Fri Jul 9 16:27:52 2010 +0100

It looks like nobootmem.c was created here:

commit 0932587328d9bd5b500a640fbaff3290c8d4cabf
Author: Yinghai Lu [off-list ref]
Date:   Thu Feb 24 14:43:05 2011 +0100

So, when I was porting ARM to memblock, we had:

commit 08677214e318297f228237be0042aac754f48f1d
Author: Yinghai Lu [off-list ref]
Date:   Wed Feb 10 01:20:20 2010 -0800

which causes a load of bootmem functions to panic() when bootmem is
disabled.

quoted

The following patch appears to work just fine on a faked sparsemem system:

Great.  Don't forget to also check flatmem too, just to be sure.

There is still a problem since nobootmem.c doesn't handle 64-bit
physical addresses, so calling free_bootmem() still truncates the
values. Simply replacing bootmem.c with nobootmem.c is not enough, we
need to also replace bootmem calls with the corresponding memblock ones.

-- 
Catalin