sparc64: Adjust vmalloc region size based upon available virtual address bits.
authorDavid S. Miller <davem@davemloft.net>
Sat, 27 Sep 2014 18:05:21 +0000 (11:05 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 30 Oct 2014 16:38:27 +0000 (09:38 -0700)
commitf2450cb9e4e54ffadee804a91fc50b272088db2f
treed9e2a70e74b0cad13511d4484fc5b68a210e0c92
parente0b18223c64d6e11248553fae8a2343d4fe66a42
sparc64: Adjust vmalloc region size based upon available virtual address bits.

[ Upstream commit bb4e6e85daa52a9f6210fa06a5ec6269598a202b ]

In order to accomodate embedded per-cpu allocation with large numbers
of cpus and numa nodes, we have to use as much virtual address space
as possible for the vmalloc region.  Otherwise we can get things like:

PERCPU: max_distance=0x380001c10000 too large for vmalloc space 0xff00000000

So, once we select a value for PAGE_OFFSET, derive the size of the
vmalloc region based upon that.

Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Bob Picco <bob.picco@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/sparc/include/asm/page_64.h
arch/sparc/include/asm/pgtable_64.h
arch/sparc/kernel/ktlb.S
arch/sparc/mm/init_64.c