linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
* [RFC] mbind: Restrict nodes to the currently allowed cpuset
@ 2007-01-04 19:16 Christoph Lameter
  2007-01-04 19:20 ` Paul Jackson
  2007-01-10  0:23 ` Andi Kleen
  0 siblings, 2 replies; 3+ messages in thread
From: Christoph Lameter @ 2007-01-04 19:16 UTC (permalink / raw)
  To: ak; +Cc: linux-mm, pj

Currently one can specify an arbitrary node mask to mbind that includes nodes
not allowed. If that is done with an interleave policy then we will go around
all the nodes. Those outside of the currently allowed cpuset will be redirected
to the border nodes. Interleave will then create imbalances at the borders
of the cpuset.

This patch restricts the nodes to the currently allowed cpuset.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

----

I still wonder if this is the right approach. Could mbind be used to set 
up policies that are larger than the existing cpuset? Or could mbind be 
used to set up a policy and then the cpuset would change?



Index: linux-2.6.19-mm1/mm/mempolicy.c
===================================================================
--- linux-2.6.19-mm1.orig/mm/mempolicy.c	2006-12-11 19:00:38.224610647 -0800
+++ linux-2.6.19-mm1/mm/mempolicy.c	2006-12-13 11:13:10.175294067 -0800
@@ -882,6 +882,7 @@ asmlinkage long sys_mbind(unsigned long 
 	int err;
 
 	err = get_nodes(&nodes, nmask, maxnode);
+	nodes_and(nodes, nodes, current->mems_allowed);
 	if (err)
 		return err;
 	return do_mbind(start, len, mode, &nodes, flags);

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFC] mbind: Restrict nodes to the currently allowed cpuset
  2007-01-04 19:16 [RFC] mbind: Restrict nodes to the currently allowed cpuset Christoph Lameter
@ 2007-01-04 19:20 ` Paul Jackson
  2007-01-10  0:23 ` Andi Kleen
  1 sibling, 0 replies; 3+ messages in thread
From: Paul Jackson @ 2007-01-04 19:20 UTC (permalink / raw)
  To: Christoph Lameter; +Cc: ak, linux-mm

Christoph wrote:
> Could mbind be used to set 
> up policies that are larger than the existing cpuset? Or could mbind be 
> used to set up a policy and then the cpuset would change?

My intention (hopefully the code matches this) is that mbind nodes are
constrained to fit in the cpuset.  If you ask to mbind more nodes, those
outside the cpuset are masked off.  If you later change the cpuset, then
we mask more nodes off to continue to fit in the cpuset.  If this gets us
down to an empty mbind list, then you get to use whatever memory nodes are
in your new cpuset.

-- 
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj@sgi.com> 1.925.600.0401

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [RFC] mbind: Restrict nodes to the currently allowed cpuset
  2007-01-04 19:16 [RFC] mbind: Restrict nodes to the currently allowed cpuset Christoph Lameter
  2007-01-04 19:20 ` Paul Jackson
@ 2007-01-10  0:23 ` Andi Kleen
  1 sibling, 0 replies; 3+ messages in thread
From: Andi Kleen @ 2007-01-10  0:23 UTC (permalink / raw)
  To: Christoph Lameter; +Cc: linux-mm, pj

On Thursday 04 January 2007 20:16, Christoph Lameter wrote:
> Currently one can specify an arbitrary node mask to mbind that includes nodes
> not allowed. If that is done with an interleave policy then we will go around
> all the nodes. Those outside of the currently allowed cpuset will be redirected
> to the border nodes. Interleave will then create imbalances at the borders
> of the cpuset.
> 
> This patch restricts the nodes to the currently allowed cpuset.

Fine by me.

-Andi

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2007-01-10  0:23 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-01-04 19:16 [RFC] mbind: Restrict nodes to the currently allowed cpuset Christoph Lameter
2007-01-04 19:20 ` Paul Jackson
2007-01-10  0:23 ` Andi Kleen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox