Andrew Morton wrote:

..
huge_pte_alloc() is a sleeping function.

Well we know what's asleep around here, and it's not huge_pte_alloc.
 

When you plug that one, I'd appreciate it if you could find a way
of not taking mapping->page_lock inside mm->page_table_lock.  Those
locks have "no relationship" at present (I think), and it'd be nice
to keep it that way.

Whatever.
--
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/