sched/numa: Update migrate_improves/degrades_locality()

Update the migrate_improves/degrades_locality() functions with
knowledge of pseudo-interleaving.

Do not consider moving tasks around within the set of group's active
nodes as improving or degrading locality. Instead, leave the load
balancer free to balance the load between a numa_group's active nodes.

Also, switch from the group/task_weight functions to the group/task_fault
functions. The "weight" functions involve a division, but both calls use
the same divisor, so there's no point in doing that from these functions.

On a 4 node (x10 core) system, performance of SPECjbb2005 seems
unaffected, though the number of migrations with 2 8-warehouse wide
instances seems to have almost halved, due to the scheduler running
each instance on a single node.

Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: mgorman@suse.de
Cc: chegu_vinod@hp.com
Link: http://lkml.kernel.org/r/20140515130306.61aae7db@cuia.bos.redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b899613f2bc636dd925b6643a09764de369ecabf..503f750c2d2535be0cd7504112501998b7f028a8 100644 (file)
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5123,6 +5123,7 @@ task_hot(struct task_struct *p, u64 now)
 /* Returns true if the destination node has incurred more faults */
 static bool migrate_improves_locality(struct task_struct *p, struct lb_env *env)
 {
+       struct numa_group *numa_group = rcu_dereference(p->numa_group);
        int src_nid, dst_nid;
 
        if (!sched_feat(NUMA_FAVOUR_HIGHER) || !p->numa_faults_memory ||
@@ -5136,21 +5137,29 @@ static bool migrate_improves_locality(struct task_struct *p, struct lb_env *env)
        if (src_nid == dst_nid)
                return false;
 
-       /* Always encourage migration to the preferred node. */
-       if (dst_nid == p->numa_preferred_nid)
-               return true;
+       if (numa_group) {
+               /* Task is already in the group's interleave set. */
+               if (node_isset(src_nid, numa_group->active_nodes))
+                       return false;
+
+               /* Task is moving into the group's interleave set. */
+               if (node_isset(dst_nid, numa_group->active_nodes))
+                       return true;
 
-       /* If both task and group weight improve, this move is a winner. */
-       if (task_weight(p, dst_nid) > task_weight(p, src_nid) &&
-           group_weight(p, dst_nid) > group_weight(p, src_nid))
+               return group_faults(p, dst_nid) > group_faults(p, src_nid);
+       }
+
+       /* Encourage migration to the preferred node. */
+       if (dst_nid == p->numa_preferred_nid)
                return true;
 
-       return false;
+       return task_faults(p, dst_nid) > task_faults(p, src_nid);
 }
 
 
 static bool migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 {
+       struct numa_group *numa_group = rcu_dereference(p->numa_group);
        int src_nid, dst_nid;
 
        if (!sched_feat(NUMA) || !sched_feat(NUMA_RESIST_LOWER))
@@ -5165,16 +5174,23 @@ static bool migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
        if (src_nid == dst_nid)
                return false;
 
+       if (numa_group) {
+               /* Task is moving within/into the group's interleave set. */
+               if (node_isset(dst_nid, numa_group->active_nodes))
+                       return false;
+
+               /* Task is moving out of the group's interleave set. */
+               if (node_isset(src_nid, numa_group->active_nodes))
+                       return true;
+
+               return group_faults(p, dst_nid) < group_faults(p, src_nid);
+       }
+
        /* Migrating away from the preferred node is always bad. */
        if (src_nid == p->numa_preferred_nid)
                return true;
 
-       /* If either task or group weight get worse, don't do it. */
-       if (task_weight(p, dst_nid) < task_weight(p, src_nid) ||
-           group_weight(p, dst_nid) < group_weight(p, src_nid))
-               return true;
-
-       return false;
+       return task_faults(p, dst_nid) < task_faults(p, src_nid);
 }
 
 #else