On Mon, Jan 31, 2011 at 03:31:15PM -0800, Dave Hansen wrote:

wait_split_huge_page() is really only used in a few spots at the moment.
I was trying to use it in fs/proc/task_mmu.c, but simply including
huge_mm.h gets this:

fs/proc/task_mmu.c: In function a??smaps_pte_rangea??:
fs/proc/task_mmu.c:392: error: dereferencing pointer to incomplete type

I like what you're doing eheh ;)

I think it's due to the __anon_vma dereference below.  #including rmap.h
makes it go away, but I don't think it's really the correct thing to do
here.  Directly including rmap.h in huge_mm.h ends up with some really
interesting header dependencies and does not work either.

Any ideas?  Should we move the existing huge_mm.h stuff to a private
header and have a more public one that also brings in rmap.h?

Solution:

+#include <linux/rmap.h>

And avoid including an explicit #include huge_mm.h which is never
needed (not even huge_memory.c includes huge_mm.h, rmap.h is all you
need)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>