[patch 4/3] cpusets: memory_spread_user interleaves over all mems_allowed

[patch 4/3] cpusets: memory_spread_user interleaves over all mems_allowed

[David Rientjes]

Instead of using current's mems_allowed, which may differ from the
mems_allowed of the cpuset being updated, the newmask passed to
mpol_rebind_mm() is used as the interleave mask in the
memory_spread_user case.

Cc: Andi Kleen <ak@suse.de>
Cc: Paul Jackson <pj@sgi.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Signed-off-by: David Rientjes <rientjes@google.com>
---
 mm/mempolicy.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1740,7 +1740,7 @@ static void mpol_rebind_policy(struct mempolicy *pol,
 		break;
 	case MPOL_INTERLEAVE:
 		if (current_cpuset_is_spread_user())
-			tmp = cpuset_mems_allowed(current);
+			tmp = *newmask;
 		else
 			nodes_remap(tmp, pol->v.nodes, *mpolmask, *newmask);
 		pol->v.nodes = tmp;

Re: [patch 4/3] cpusets: memory_spread_user interleaves over all mems_allowed

[Paul Jackson]

David wrote:
> -			tmp = cpuset_mems_allowed(current);
> +			tmp = *newmask;

I see this as a nice little optimization, not a change in
what the code does.  That is, *newmask happens to already
hold cpuset_mems_allowed(current), so can be used for such.

Is that right?

If so, nice tweak - thanks.

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