Skip to contentSkip to conversation

sched: select_task_rq_fair() cleanup/changes

PATCH In review Peter Zijlstra 1 participant 5 messages Nov 12, 2009 – Nov 13, 2009

lore
Share & tools
lore
#0 lore raw

cover sched: select_task_rq_fair() cleanup/changes

Some bits I came up with while trying to grok the recent changes in the mentioned function.

Please have a careful look.

#1 lore raw

PATCH 1/2 sched: cleanup select_task_rq_fair()

Clean up the new affine to idle sibling bits while trying to grok them. Should not have any function differences.

LKML-Reference: <new-submission>

---
 kernel/sched_fair.c |   73 ++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 51 insertions(+), 22 deletions(-)
kernel/sched_fair.c +51 -22
Index: linux-2.6/kernel/sched_fair.c
===================================================================
@@ -1345,6 +1345,41 @@ find_idlest_cpu(struct sched_group *grou
 }
 
 /*
+ * Try and locate an idle CPU in the sched_domain.
+ */
+static int
+select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target)
+{
+	int cpu = smp_processor_id();
+	int prev_cpu = task_cpu(p);
+	int i;
+
+	/*
+	 * If this domain spans both cpu and prev_cpu (see the SD_WAKE_AFFINE
+	 * test in select_task_rq_fair) and the prev_cpu is idle then that's
+	 * always a better target than the current cpu.
+	 */
+	if (target == cpu) {
+		if (!cpu_rq(prev_cpu)->cfs.nr_running)
+			target = prev_cpu;
+	}
+
+	/*
+	 * Otherwise, iterate the domain and find an elegible idle cpu.
+	 */
+	if (target == -1 || target == cpu) {
+		for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
+			if (!cpu_rq(i)->cfs.nr_running) {
+				target = i;
+				break;
+			}
+		}
+	}
+
+	return target;
+}
+
+/*
  * sched_balance_self: balance the current task (running on cpu) in domains
  * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and
  * SD_BALANCE_EXEC.
@@ -1399,36 +1434,30 @@ static int select_task_rq_fair(struct ta
 		}
 
 		if (want_affine && (tmp->flags & SD_WAKE_AFFINE)) {
-			int candidate = -1, i;
+			int target = -1;
 
+			/*
+			 * If both cpu and prev_cpu are part of this domain,
+			 * cpu is a valid SD_WAKE_AFFINE target.
+			 */
 			if (cpumask_test_cpu(prev_cpu, sched_domain_span(tmp)))
-				candidate = cpu;
+				target = cpu;
 
 			/*
-			 * Check for an idle shared cache.
+			 * If there's an idle sibling in this domain, make that
+			 * the wake_affine target instead of the current cpu.
+			 *
+			 * XXX: should we possibly do this outside of
+			 * WAKE_AFFINE, in case the shared cache domain is
+			 * smaller than the WAKE_AFFINE domain?
 			 */
-			if (tmp->flags & SD_PREFER_SIBLING) {
-				if (candidate == cpu) {
-					if (!cpu_rq(prev_cpu)->cfs.nr_running)
-						candidate = prev_cpu;
-				}
-
-				if (candidate == -1 || candidate == cpu) {
-					for_each_cpu(i, sched_domain_span(tmp)) {
-						if (!cpumask_test_cpu(i, &p->cpus_allowed))
-							continue;
-						if (!cpu_rq(i)->cfs.nr_running) {
-							candidate = i;
-							break;
-						}
-					}
-				}
-			}
+			if (tmp->flags & SD_PREFER_SIBLING)
+				target = select_idle_sibling(p, tmp, target);
 
-			if (candidate >= 0) {
+			if (target >= 0) {
 				affine_sd = tmp;
 				want_affine = 0;
-				cpu = candidate;
+				cpu = target;
 			}
 		}
 
#2 lore raw

[tip:sched/core] sched: Cleanup select_task_rq_fair()

Commit-ID:  a50bde5130f65733142b32975616427d0ea50856
Gitweb:     http://git.kernel.org/tip/a50bde5130f65733142b32975616427d0ea50856
Author:     Peter Zijlstra <[email protected]>
AuthorDate: Thu, 12 Nov 2009 15:55:28 +0100
Committer:  Ingo Molnar <[email protected]>
CommitDate: Fri, 13 Nov 2009 10:09:58 +0100

sched: Cleanup select_task_rq_fair()

Clean up the new affine to idle sibling bits while trying to grok them. Should not have any function differences.

LKML-Reference: <[email protected]>

---
 kernel/sched_fair.c |   73 +++++++++++++++++++++++++++++++++++---------------
 1 files changed, 51 insertions(+), 22 deletions(-)
kernel/sched_fair.c +51 -22
index e4d4483..a32df15 100644
@@ -1319,6 +1319,41 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 }
 
 /*
+ * Try and locate an idle CPU in the sched_domain.
+ */
+static int
+select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target)
+{
+	int cpu = smp_processor_id();
+	int prev_cpu = task_cpu(p);
+	int i;
+
+	/*
+	 * If this domain spans both cpu and prev_cpu (see the SD_WAKE_AFFINE
+	 * test in select_task_rq_fair) and the prev_cpu is idle then that's
+	 * always a better target than the current cpu.
+	 */
+	if (target == cpu) {
+		if (!cpu_rq(prev_cpu)->cfs.nr_running)
+			target = prev_cpu;
+	}
+
+	/*
+	 * Otherwise, iterate the domain and find an elegible idle cpu.
+	 */
+	if (target == -1 || target == cpu) {
+		for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
+			if (!cpu_rq(i)->cfs.nr_running) {
+				target = i;
+				break;
+			}
+		}
+	}
+
+	return target;
+}
+
+/*
  * sched_balance_self: balance the current task (running on cpu) in domains
  * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and
  * SD_BALANCE_EXEC.
@@ -1373,36 +1408,30 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag
 		}
 
 		if (want_affine && (tmp->flags & SD_WAKE_AFFINE)) {
-			int candidate = -1, i;
+			int target = -1;
 
+			/*
+			 * If both cpu and prev_cpu are part of this domain,
+			 * cpu is a valid SD_WAKE_AFFINE target.
+			 */
 			if (cpumask_test_cpu(prev_cpu, sched_domain_span(tmp)))
-				candidate = cpu;
+				target = cpu;
 
 			/*
-			 * Check for an idle shared cache.
+			 * If there's an idle sibling in this domain, make that
+			 * the wake_affine target instead of the current cpu.
+			 *
+			 * XXX: should we possibly do this outside of
+			 * WAKE_AFFINE, in case the shared cache domain is
+			 * smaller than the WAKE_AFFINE domain?
 			 */
-			if (tmp->flags & SD_PREFER_SIBLING) {
-				if (candidate == cpu) {
-					if (!cpu_rq(prev_cpu)->cfs.nr_running)
-						candidate = prev_cpu;
-				}
-
-				if (candidate == -1 || candidate == cpu) {
-					for_each_cpu(i, sched_domain_span(tmp)) {
-						if (!cpumask_test_cpu(i, &p->cpus_allowed))
-							continue;
-						if (!cpu_rq(i)->cfs.nr_running) {
-							candidate = i;
-							break;
-						}
-					}
-				}
-			}
+			if (tmp->flags & SD_PREFER_SIBLING)
+				target = select_idle_sibling(p, tmp, target);
 
-			if (candidate >= 0) {
+			if (target >= 0) {
 				affine_sd = tmp;
 				want_affine = 0;
-				cpu = candidate;
+				cpu = target;
 			}
 		}
 
#3 lore raw

PATCH 2/2 sched: More generic WAKE_AFFINE vs select_idle_sibling()

In reply to Peter Zijlstra

Instead of only considering SD_WAKE_AFFINE | SD_PREFER_SIBLING domains also allow all SD_PREFER_SIBLING domains below a SD_WAKE_AFFINE domain to change the affinity target.

LKML-Reference: <new-submission>

---
 kernel/sched_fair.c |   33 ++++++++++++++++-----------------
 1 file changed, 16 insertions(+), 17 deletions(-)
kernel/sched_fair.c +16 -17
Index: linux-2.6/kernel/sched_fair.c
===================================================================
@@ -1359,20 +1359,16 @@ select_idle_sibling(struct task_struct *
 	 * test in select_task_rq_fair) and the prev_cpu is idle then that's
 	 * always a better target than the current cpu.
 	 */
-	if (target == cpu) {
-		if (!cpu_rq(prev_cpu)->cfs.nr_running)
-			target = prev_cpu;
-	}
+	if (target == cpu && !cpu_rq(prev_cpu)->cfs.nr_running)
+		return prev_cpu;
 
 	/*
 	 * Otherwise, iterate the domain and find an elegible idle cpu.
 	 */
-	if (target == -1 || target == cpu) {
-		for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
-			if (!cpu_rq(i)->cfs.nr_running) {
-				target = i;
-				break;
-			}
+	for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
+		if (!cpu_rq(i)->cfs.nr_running) {
+			target = i;
+			break;
 		}
 	}
 
@@ -1433,7 +1429,12 @@ static int select_task_rq_fair(struct ta
 				want_sd = 0;
 		}
 
-		if (want_affine && (tmp->flags & SD_WAKE_AFFINE)) {
+		/*
+		 * While iterating the domains looking for a spanning
+		 * WAKE_AFFINE domain, adjust the affine target to any idle cpu
+		 * in cache sharing domains along the way.
+		 */
+		if (want_affine) {
 			int target = -1;
 
 			/*
@@ -1446,17 +1447,15 @@ static int select_task_rq_fair(struct ta
 			/*
 			 * If there's an idle sibling in this domain, make that
 			 * the wake_affine target instead of the current cpu.
-			 *
-			 * XXX: should we possibly do this outside of
-			 * WAKE_AFFINE, in case the shared cache domain is
-			 * smaller than the WAKE_AFFINE domain?
 			 */
 			if (tmp->flags & SD_PREFER_SIBLING)
 				target = select_idle_sibling(p, tmp, target);
 
 			if (target >= 0) {
-				affine_sd = tmp;
-				want_affine = 0;
+				if (tmp->flags & SD_WAKE_AFFINE) {
+					affine_sd = tmp;
+					want_affine = 0;
+				}
 				cpu = target;
 			}
 		}
#4 lore raw

[tip:sched/core] sched: More generic WAKE_AFFINE vs select_idle_sibling()

Commit-ID:  fe3bcfe1f6c1fc4ea7706ac2d05e579fd9092682
Gitweb:     http://git.kernel.org/tip/fe3bcfe1f6c1fc4ea7706ac2d05e579fd9092682
Author:     Peter Zijlstra <[email protected]>
AuthorDate: Thu, 12 Nov 2009 15:55:29 +0100
Committer:  Ingo Molnar <[email protected]>
CommitDate: Fri, 13 Nov 2009 10:09:59 +0100

sched: More generic WAKE_AFFINE vs select_idle_sibling()

Instead of only considering SD_WAKE_AFFINE | SD_PREFER_SIBLING domains also allow all SD_PREFER_SIBLING domains below a SD_WAKE_AFFINE domain to change the affinity target.

LKML-Reference: <[email protected]>

---
 kernel/sched_fair.c |   33 ++++++++++++++++-----------------
 1 files changed, 16 insertions(+), 17 deletions(-)
kernel/sched_fair.c +16 -17
index a32df15..f28a267 100644
@@ -1333,20 +1333,16 @@ select_idle_sibling(struct task_struct *p, struct sched_domain *sd, int target)
 	 * test in select_task_rq_fair) and the prev_cpu is idle then that's
 	 * always a better target than the current cpu.
 	 */
-	if (target == cpu) {
-		if (!cpu_rq(prev_cpu)->cfs.nr_running)
-			target = prev_cpu;
-	}
+	if (target == cpu && !cpu_rq(prev_cpu)->cfs.nr_running)
+		return prev_cpu;
 
 	/*
 	 * Otherwise, iterate the domain and find an elegible idle cpu.
 	 */
-	if (target == -1 || target == cpu) {
-		for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
-			if (!cpu_rq(i)->cfs.nr_running) {
-				target = i;
-				break;
-			}
+	for_each_cpu_and(i, sched_domain_span(sd), &p->cpus_allowed) {
+		if (!cpu_rq(i)->cfs.nr_running) {
+			target = i;
+			break;
 		}
 	}
 
@@ -1407,7 +1403,12 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag
 				want_sd = 0;
 		}
 
-		if (want_affine && (tmp->flags & SD_WAKE_AFFINE)) {
+		/*
+		 * While iterating the domains looking for a spanning
+		 * WAKE_AFFINE domain, adjust the affine target to any idle cpu
+		 * in cache sharing domains along the way.
+		 */
+		if (want_affine) {
 			int target = -1;
 
 			/*
@@ -1420,17 +1421,15 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag
 			/*
 			 * If there's an idle sibling in this domain, make that
 			 * the wake_affine target instead of the current cpu.
-			 *
-			 * XXX: should we possibly do this outside of
-			 * WAKE_AFFINE, in case the shared cache domain is
-			 * smaller than the WAKE_AFFINE domain?
 			 */
 			if (tmp->flags & SD_PREFER_SIBLING)
 				target = select_idle_sibling(p, tmp, target);
 
 			if (target >= 0) {
-				affine_sd = tmp;
-				want_affine = 0;
+				if (tmp->flags & SD_WAKE_AFFINE) {
+					affine_sd = tmp;
+					want_affine = 0;
+				}
 				cpu = target;
 			}
 		}