sched: Add SD_PREFER_LOCAL

And turn it on for NUMA and MC domains. This improves
locality in balancing decisions by keeping up to
capacity amount of tasks local before looking for idle
CPUs. (and twice the capacity if SD_POWERSAVINGS_BALANCE
is set.)

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

include/linux/sched.h

@@ -805,7 +805,7 @@ enum cpu_idle_type {
 #define SD_BALANCE_FORK		0x0008	/* Balance on fork, clone */
 #define SD_BALANCE_WAKE		0x0010  /* Balance on wakeup */
 #define SD_WAKE_AFFINE		0x0020	/* Wake task to waking CPU */
-
+#define SD_PREFER_LOCAL		0x0040  /* Prefer to keep tasks local to this domain */
 #define SD_SHARE_CPUPOWER	0x0080	/* Domain members share cpu power */
 #define SD_POWERSAVINGS_BALANCE	0x0100	/* Balance for power savings */
 #define SD_SHARE_PKG_RESOURCES	0x0200	/* Domain members share cpu pkg resources */

include/linux/topology.h

@@ -129,6 +129,7 @@ int arch_update_cpu_topology(void);
 				| 1*SD_BALANCE_FORK			\
 				| 1*SD_BALANCE_WAKE			\
 				| 1*SD_WAKE_AFFINE			\
+				| 1*SD_PREFER_LOCAL			\
 				| 0*SD_SHARE_CPUPOWER			\
 				| 1*SD_SHARE_PKG_RESOURCES		\
 				| 0*SD_SERIALIZE			\
@@ -161,6 +162,7 @@ int arch_update_cpu_topology(void);
 				| 1*SD_BALANCE_FORK			\
 				| 1*SD_BALANCE_WAKE			\
 				| 1*SD_WAKE_AFFINE			\
+				| 1*SD_PREFER_LOCAL			\
 				| 0*SD_SHARE_CPUPOWER			\
 				| 0*SD_SHARE_PKG_RESOURCES		\
 				| 0*SD_SERIALIZE			\

kernel/sched_fair.c

@@ -1360,7 +1360,7 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int flags)
 		 * If power savings logic is enabled for a domain, see if we
 		 * are not overloaded, if so, don't balance wider.
 		 */
-		if (tmp->flags & SD_POWERSAVINGS_BALANCE) {
+		if (tmp->flags & (SD_POWERSAVINGS_BALANCE|SD_PREFER_LOCAL)) {
 			unsigned long power = 0;
 			unsigned long nr_running = 0;
 			unsigned long capacity;
@@ -1373,7 +1373,10 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int flags)
 
 			capacity = DIV_ROUND_CLOSEST(power, SCHED_LOAD_SCALE);
 
-			if (nr_running/2 < capacity)
+			if (tmp->flags & SD_POWERSAVINGS_BALANCE)
+				nr_running /= 2;
+
+			if (nr_running < capacity)
 				break;
 		}