[PATCH v12 0/2] Make find_later_rq() choose a closer cpu in topology

[PATCH v12 0/2] Make find_later_rq() choose a closer cpu in topology

[Byungchul Park]

Change from v11
   -. Rebase onto the latest tip/sched/core
   -. make the word 'cpu' in comments be upper case as Ingo did
      in the commit 97fb7a0a89
   -. Avoid a compile warning caused by mismatching bwt const and
      non-const 'struct sched_domain' in find_cpu()

Change from v10
   -. modify a comment a bit as Steven suggested

Change from v9
   -. modify a comment a bit so to be more clear as Juri suggested

Change from v8
   -. add suggested-by Peterz
   -. add several comments

Change from v7
   -. fix a trivial typo
   -. modify commit messages to explain what it does more clearly
   -. simplify code with an existing macro

Change from v6
   -. add a comment about selection of fallback_cpu incase more than one exist
   -. modify a comment explaining what we do wrt PREFER_SIBLING

Change from v5
   -. exclude two patches already picked up by peterz
      (sched/deadline: Make find_later_rq() choose a closer cpu in topology)
      (sched/deadline: Change return value of cpudl_find())
   -. apply what peterz fixed for 'prefer sibling', into deadline and rt

Change from v4
   -. remove a patch that might cause huge lock contention
      (by spin lock(&cpudl.lock) in a hot path of scheduler)

Change from v3
   -. rename closest_cpu to best_cpu so that it align with rt
   -. protect referring cpudl.elements with cpudl.lock
   -. change return value of cpudl_find() to bool

Change from v2
   -. add support for SD_PREFER_SIBLING

Change from v1
   -. clean up the patch

Byungchul Park (2):
  sched/deadline: Add support for SD_PREFER_SIBLING on find_later_rq()
  sched/rt: Add support for SD_PREFER_SIBLING on find_lowest_rq()

 kernel/sched/deadline.c | 82 ++++++++++++++++++++++++++++++++++++++++++++-----
 kernel/sched/rt.c       | 80 +++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 152 insertions(+), 10 deletions(-)

-- 
1.9.1

*** BLURB HERE ***

Byungchul Park (2):
  sched/deadline: Add support for SD_PREFER_SIBLING on find_later_rq()
  sched/rt: Add support for SD_PREFER_SIBLING on find_lowest_rq()

 kernel/sched/deadline.c | 80 ++++++++++++++++++++++++++++++++++++++++++++-----
 kernel/sched/rt.c       | 78 +++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 148 insertions(+), 10 deletions(-)

-- 
1.9.1

[PATCH v12 1/2] sched/deadline: Add support for SD_PREFER_SIBLING on find_later_rq()

[Byungchul Park]

Hello Juri,

I've changed the code a little bit to avoid a compile warning caused by
'const' args of find_cpu(). Can I keep your Acked-by?

BEFORE:
static int find_cpu(const struct cpumask *mask,
		    const struct sched_domain *sd,
		    const struct sched_domain *prefer)

AFTER:
static int find_cpu(const struct cpumask *mask,
		    struct sched_domain *sd,
		    struct sched_domain *prefer)

(I temporarily removed the Acked-by you gave me.)
Acked-by: Juri Lelli <juri.lelli@arm.com>

-----8<-----
>From 5a4753e8c15369420a16fa04026f74ae5c9d377c Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@lge.com>
Date: Mon, 4 Jun 2018 16:46:56 +0900
Subject: [PATCH v12 1/2] sched/deadline: Add support for SD_PREFER_SIBLING on
 find_later_rq()

It would be better to try to check other siblings first if
SD_PREFER_SIBLING is flaged when pushing tasks - migration.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
 kernel/sched/deadline.c | 80 ++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 73 insertions(+), 7 deletions(-)

diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 1356afd..6130d40 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1853,12 +1853,33 @@ static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu
 
 static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask_dl);
 
+/*
+ * Find the first CPU in: mask & sd & ~prefer
+ */
+static int find_cpu(const struct cpumask *mask,
+		    struct sched_domain *sd,
+		    struct sched_domain *prefer)
+{
+	int cpu;
+
+	for_each_cpu(cpu, mask) {
+		if (!cpumask_test_cpu(cpu, sched_domain_span(sd)))
+			continue;
+		if (prefer && cpumask_test_cpu(cpu, sched_domain_span(prefer)))
+			continue;
+		break;
+	}
+
+	return cpu;
+}
+
 static int find_later_rq(struct task_struct *task)
 {
-	struct sched_domain *sd;
+	struct sched_domain *sd, *prefer = NULL;
 	struct cpumask *later_mask = this_cpu_cpumask_var_ptr(local_cpu_mask_dl);
 	int this_cpu = smp_processor_id();
 	int cpu = task_cpu(task);
+	int fallback_cpu = -1;
 
 	/* Make sure the mask is initialized first */
 	if (unlikely(!later_mask))
@@ -1910,15 +1931,37 @@ static int find_later_rq(struct task_struct *task)
 				return this_cpu;
 			}
 
-			best_cpu = cpumask_first_and(later_mask,
-							sched_domain_span(sd));
 			/*
-			 * Last chance: if a CPU being in both later_mask
-			 * and current sd span is valid, that becomes our
-			 * choice. Of course, the latest possible CPU is
-			 * already under consideration through later_mask.
+			 * If a CPU exists that is in the later_mask and
+			 * the current sd span, but not in the prefer sd
+			 * span, then that becomes our choice.
+			 *
+			 * Of course, the latest possible CPU is already
+			 * under consideration through later_mask.
 			 */
+			best_cpu = find_cpu(later_mask, sd, prefer);
+
 			if (best_cpu < nr_cpu_ids) {
+				/*
+				 * If current domain is SD_PREFER_SIBLING
+				 * flaged, we have to try to check other
+				 * siblings first.
+				 */
+				if (sd->flags & SD_PREFER_SIBLING) {
+					prefer = sd;
+
+					/*
+					 * fallback_cpu should be one
+					 * in the closest domain among
+					 * SD_PREFER_SIBLING domains,
+					 * in case that more than one
+					 * SD_PREFER_SIBLING domains
+					 * exist in the hierachy.
+					 */
+					if (fallback_cpu == -1)
+						fallback_cpu = best_cpu;
+					continue;
+				}
 				rcu_read_unlock();
 				return best_cpu;
 			}
@@ -1927,6 +1970,29 @@ static int find_later_rq(struct task_struct *task)
 	rcu_read_unlock();
 
 	/*
+	 * If fallback_cpu is valid, all our guesses failed *except* for
+	 * SD_PREFER_SIBLING domain. Now, we can return the fallback CPU.
+	 *
+	 * XXX: Consider the following example, 4 cores SMT2 system:
+	 *
+	 *    LLC [0       -        7]
+	 *    SMT [0 1][2 3][4 5][6 7]
+	 *         o x  o x  x x  x x
+	 *
+	 *    where 'o': occupied and 'x': empty.
+	 *
+	 * A wakeup on CPU0 will exclude CPU1 and choose CPU3, since
+	 * CPU1 is in a SD_PREFER_SIBLING sd and CPU3 is not. However,
+	 * in this case, CPU4 would have been a better choice, since
+	 * CPU3 is a (SMT) thread of an already loaded core.
+	 *
+	 * Doing it 'right' is difficult and expensive. The current
+	 * solution is an acceptable approximation.
+	 */
+	if (fallback_cpu != -1)
+		return fallback_cpu;
+
+	/*
 	 * At this point, all our guesses failed, we just return
 	 * 'something', and let the caller sort the things out.
 	 */
-- 
1.9.1

[PATCH v12 2/2] sched/rt: Add support for SD_PREFER_SIBLING on find_lowest_rq()

[Byungchul Park]

Hello Steven,

I've changed the code a little bit to avoid a compile warning caused by
'const' args of find_cpu(). Can I keep your Reviewed-by?

BEFORE:
static int find_cpu(const struct cpumask *mask,
		    const struct sched_domain *sd,
		    const struct sched_domain *prefer)

AFTER:
static int find_cpu(const struct cpumask *mask,
		    struct sched_domain *sd,
		    struct sched_domain *prefer)

(I temporarily removed the Reviewed-by you gave me.)
Reviewed-by: Steven Rostedt (VMware) <rostedt@goodmis.org>

-----8<-----
>From 205b197043085947ae30cd939bc12e436c328fe5 Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@lge.com>
Date: Mon, 4 Jun 2018 16:47:45 +0900
Subject: [PATCH v12 2/2] sched/rt: Add support for SD_PREFER_SIBLING on
 find_lowest_rq()

It would be better to try to check other siblings first if
SD_PREFER_SIBLING is flaged when pushing tasks - migration.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
 kernel/sched/rt.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 75 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index ef3c4e6..b2aff1a 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1623,12 +1623,33 @@ static struct task_struct *pick_highest_pushable_task(struct rq *rq, int cpu)
 
 static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask);
 
+/*
+ * Find the first CPU in: mask & sd & ~prefer
+ */
+static int find_cpu(const struct cpumask *mask,
+		    struct sched_domain *sd,
+		    struct sched_domain *prefer)
+{
+	int cpu;
+
+	for_each_cpu(cpu, mask) {
+		if (!cpumask_test_cpu(cpu, sched_domain_span(sd)))
+			continue;
+		if (prefer && cpumask_test_cpu(cpu, sched_domain_span(prefer)))
+			continue;
+		break;
+	}
+
+	return cpu;
+}
+
 static int find_lowest_rq(struct task_struct *task)
 {
-	struct sched_domain *sd;
+	struct sched_domain *sd, *prefer = NULL;
 	struct cpumask *lowest_mask = this_cpu_cpumask_var_ptr(local_cpu_mask);
 	int this_cpu = smp_processor_id();
 	int cpu      = task_cpu(task);
+	int fallback_cpu = -1;
 
 	/* Make sure the mask is initialized first */
 	if (unlikely(!lowest_mask))
@@ -1673,9 +1694,37 @@ static int find_lowest_rq(struct task_struct *task)
 				return this_cpu;
 			}
 
-			best_cpu = cpumask_first_and(lowest_mask,
-						     sched_domain_span(sd));
+			/*
+			 * If a CPU exists that is in the lowest_mask and
+			 * the current sd span, but not in the prefer sd
+			 * span, then that becomes our choice.
+			 *
+			 * Of course, the lowest possible CPU is already
+			 * under consideration through lowest_mask.
+			 */
+			best_cpu = find_cpu(lowest_mask, sd, prefer);
+
 			if (best_cpu < nr_cpu_ids) {
+				/*
+				 * If current domain is SD_PREFER_SIBLING
+				 * flaged, we have to try to check other
+				 * siblings first.
+				 */
+				if (sd->flags & SD_PREFER_SIBLING) {
+					prefer = sd;
+
+					/*
+					 * fallback_cpu should be one
+					 * in the closest domain among
+					 * SD_PREFER_SIBLING domains,
+					 * in case that more than one
+					 * SD_PREFER_SIBLING domains
+					 * exist in the hierachy.
+					 */
+					if (fallback_cpu == -1)
+						fallback_cpu = best_cpu;
+					continue;
+				}
 				rcu_read_unlock();
 				return best_cpu;
 			}
@@ -1684,6 +1733,29 @@ static int find_lowest_rq(struct task_struct *task)
 	rcu_read_unlock();
 
 	/*
+	 * If fallback_cpu is valid, all our guesses failed *except* for
+	 * SD_PREFER_SIBLING domain. Now, we can return the fallback CPU.
+	 *
+	 * XXX: Consider the following example, 4 cores SMT2 system:
+	 *
+	 *    LLC [0       -        7]
+	 *    SMT [0 1][2 3][4 5][6 7]
+	 *         o x  o x  x x  x x
+	 *
+	 *    where 'o': occupied and 'x': empty.
+	 *
+	 * A wakeup on CPU0 will exclude CPU1 and choose CPU3, since
+	 * CPU1 is in a SD_PREFER_SIBLING sd and CPU3 is not. However,
+	 * in this case, CPU4 would have been a better choice, since
+	 * CPU3 is a (SMT) thread of an already loaded core.
+	 *
+	 * Doing it 'right' is difficult and expensive. The current
+	 * solution is an acceptable approximation.
+	 */
+	if (fallback_cpu != -1)
+		return fallback_cpu;
+
+	/*
 	 * And finally, if there were no matches within the domains
 	 * just give the caller *something* to work with from the compatible
 	 * locations.
-- 
1.9.1