[PATCH] undo more numa maxnode confusions

[Paul Jackson <pj@sgi.com>]

Argh ... undo Andi's patch that undid Brent's patch that undid Andi's API ...

It was Linus' bk tree, not Andrew's *-mm tree, that needed the
"--maxnode" put back in.  Andrew's tree already has the "--maxnode"
that it should have in the "get_nodes()" routine.

If you apply Andi's patch to Andrew's *-mm tree, then a second nearby
similar looking routine "get_zonemask()" gets the second "--maxnode",
and she doesn't build anymore, failing with:

  mm/mempolicy.c: In function `get_zonemask':
  mm/mempolicy.c:419: error: `maxnode' undeclared (first use in this function)

This applies on top of 2.6.9-rc1-mm5.

Index: 2.6.9-rc1-mm5/mm/mempolicy.c
===================================================================
--- 2.6.9-rc1-mm5.orig/mm/mempolicy.c	2004-09-13 04:23:57.000000000 -0700
+++ 2.6.9-rc1-mm5/mm/mempolicy.c	2004-09-13 05:00:15.000000000 -0700
@@ -416,7 +416,6 @@ static void get_zonemask(struct mempolic
 {
 	int i;
 
-	--maxnode;
 	bitmap_zero(nodes, MAX_NUMNODES);
 	switch (p->policy) {
 	case MPOL_BIND:


-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj@sgi.com> 1.650.933.1373