linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Swapnil Sapkal <swapnil.sapkal@amd.com>
To: <mingo@redhat.com>, <peterz@infradead.org>,
	<juri.lelli@redhat.com>, <vincent.guittot@linaro.org>
Cc: <dietmar.eggemann@arm.com>, <rostedt@goodmis.org>,
	<bsegall@google.com>, <mgorman@suse.de>, <bristot@redhat.com>,
	<vschneid@redhat.com>, <iamjoonsoo.kim@lge.com>,
	<linux-kernel@vger.kernel.org>, <gautham.shenoy@amd.com>,
	<kprateek.nayak@amd.com>, <wyes.karny@amd.com>,
	Swapnil Sapkal <swapnil.sapkal@amd.com>
Subject: [PATCH 2/2] sched/fair: Cleanup in migrate_degrades_locality() to improve readability
Date: Wed, 14 Jun 2023 10:22:24 +0000	[thread overview]
Message-ID: <20230614102224.12555-3-swapnil.sapkal@amd.com> (raw)
In-Reply-To: <20230614102224.12555-1-swapnil.sapkal@amd.com>

The migrate_degrades_locality() returns tristate value whether
the migration will improve locality, degrades locality or no
impact. Handle this return values with enum to improve the
readability.

Signed-off-by: Swapnil Sapkal <swapnil.sapkal@amd.com>
---
 kernel/sched/fair.c | 69 +++++++++++++++++++++++++++++----------------
 1 file changed, 44 insertions(+), 25 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 9a8e5dcbe7e6..06813ce5356e 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -8443,45 +8443,52 @@ static int task_hot(struct task_struct *p, struct lb_env *env)
 	return delta < (s64)sysctl_sched_migration_cost;
 }
 
+enum migration_impact {
+	/* if task migration is not affected by locality */
+	no_change = -1,
+
+	/* if task migration improves locality i.e migration preferred */
+	improves_locality = 0,
+
+	/* if task migration degrades locality */
+	degrades_locality = 1
+};
+
 #ifdef CONFIG_NUMA_BALANCING
-/*
- * Returns 1, if task migration degrades locality
- * Returns 0, if task migration improves locality i.e migration preferred.
- * Returns -1, if task migration is not affected by locality.
- */
-static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
+static enum migration_impact
+migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 {
 	struct numa_group *numa_group = rcu_dereference(p->numa_group);
 	unsigned long src_weight, dst_weight;
 	int src_nid, dst_nid, dist;
 
 	if (!static_branch_likely(&sched_numa_balancing))
-		return -1;
+		return no_change;
 
 	if (!p->numa_faults || !(env->sd->flags & SD_NUMA))
-		return -1;
+		return no_change;
 
 	src_nid = cpu_to_node(env->src_cpu);
 	dst_nid = cpu_to_node(env->dst_cpu);
 
 	if (src_nid == dst_nid)
-		return -1;
+		return no_change;
 
 	/* Migrating away from the preferred node is always bad. */
 	if (src_nid == p->numa_preferred_nid) {
 		if (env->src_rq->nr_running > env->src_rq->nr_preferred_running)
-			return 1;
+			return degrades_locality;
 		else
-			return -1;
+			return no_change;
 	}
 
 	/* Encourage migration to the preferred node. */
 	if (dst_nid == p->numa_preferred_nid)
-		return 0;
+		return improves_locality;
 
 	/* Leaving a core idle is often worse than degrading locality. */
 	if (env->idle == CPU_IDLE)
-		return -1;
+		return no_change;
 
 	dist = node_distance(src_nid, dst_nid);
 	if (numa_group) {
@@ -8492,14 +8499,14 @@ static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 		dst_weight = task_weight(p, dst_nid, dist);
 	}
 
-	return dst_weight < src_weight;
+	return (dst_weight < src_weight) ? degrades_locality : improves_locality;
 }
 
 #else
-static inline int migrate_degrades_locality(struct task_struct *p,
-					     struct lb_env *env)
+static inline enum migration_impact
+migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 {
-	return -1;
+	return no_change;
 }
 #endif
 
@@ -8509,7 +8516,7 @@ static inline int migrate_degrades_locality(struct task_struct *p,
 static
 int can_migrate_task(struct task_struct *p, struct lb_env *env, int *tsk_cache_hot)
 {
-	int degrades_locality;
+	enum migration_impact migration_impact;
 
 	lockdep_assert_rq_held(env->src_rq);
 
@@ -8578,18 +8585,30 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env, int *tsk_cache_h
 	if (env->flags & LBF_ACTIVE_LB)
 		return 1;
 
-	degrades_locality = migrate_degrades_locality(p, env);
-	if (degrades_locality == -1)
+	migration_impact = migrate_degrades_locality(p, env);
+
+	switch (migration_impact) {
+	case no_change:
 		*tsk_cache_hot = task_hot(p, env);
-	else
-		*tsk_cache_hot = degrades_locality;
+		break;
+
+	case degrades_locality:
+		*tsk_cache_hot = 1;
+		break;
+
+	case improves_locality:
+		*tsk_cache_hot = 0;
+		break;
+	}
+
+	if (!(*tsk_cache_hot))
+		return 1;
 
 	/*
-	 * Can migrate a hot task only after the attempts to reach balance
+	 * Can migrate a task only after the attempts to reach balance
 	 * without the task have exceeded the cache_nice_tries threshold.
 	 */
-	if (!(*tsk_cache_hot) ||
-		env->sd->nr_balance_failed > env->sd->cache_nice_tries)
+	if (env->sd->nr_balance_failed > env->sd->cache_nice_tries)
 		return 1;
 
 	schedstat_inc(p->stats.nr_failed_migrations_hot);
-- 
2.34.1


  parent reply	other threads:[~2023-06-14 10:26 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-14 10:22 [PATCH 0/2] Cleanup and Fix for wrong accounting of migrated cache hot tasks Swapnil Sapkal
2023-06-14 10:22 ` [PATCH 1/2] sched/fair: Fix value reported by hot tasks pulled in /proc/schedstat Swapnil Sapkal
2023-06-19  9:22   ` Peter Zijlstra
2023-06-21  4:38     ` Swapnil Sapkal
2023-06-14 10:22 ` Swapnil Sapkal [this message]
2023-06-19  9:45   ` [PATCH 2/2] sched/fair: Cleanup in migrate_degrades_locality() to improve readability Peter Zijlstra
2023-06-21  4:44     ` Swapnil Sapkal

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230614102224.12555-3-swapnil.sapkal@amd.com \
    --to=swapnil.sapkal@amd.com \
    --cc=bristot@redhat.com \
    --cc=bsegall@google.com \
    --cc=dietmar.eggemann@arm.com \
    --cc=gautham.shenoy@amd.com \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=juri.lelli@redhat.com \
    --cc=kprateek.nayak@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.com \
    --cc=wyes.karny@amd.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).