[PATCH] sched: remove HZ_BW feature hedge

[PATCH] sched: remove HZ_BW feature hedge

[Phil Auld]

As a hedge against unexpected user issues commit 88c56cfeaec4
("sched/fair: Block nohz tick_stop when cfs bandwidth in use")
included a scheduler feature to disable the new functionality.
It's been a few releases (v6.6) and no screams, so remove it.

Signed-off-by: Phil Auld <pauld@redhat.com>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Ben Segall <bsegall@google.com>
Cc: Valentin Schneider <vschneid@redhat.com>
---
 kernel/sched/core.c     | 2 +-
 kernel/sched/fair.c     | 2 +-
 kernel/sched/features.h | 2 --
 3 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 7019a40457a6..d7afcee8e2ae 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1269,7 +1269,7 @@ bool sched_can_stop_tick(struct rq *rq)
 	 * dequeued by migrating while the constrained task continues to run.
 	 * E.g. going from 2->1 without going through pick_next_task().
 	 */
-	if (sched_feat(HZ_BW) && __need_bw_check(rq, rq->curr)) {
+	if (__need_bw_check(rq, rq->curr)) {
 		if (cfs_task_bw_constrained(rq->curr))
 			return false;
 	}
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index c62805dbd608..06e026f10cb4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6549,7 +6549,7 @@ static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p)
 {
 	int cpu = cpu_of(rq);
 
-	if (!sched_feat(HZ_BW) || !cfs_bandwidth_used())
+	if (!cfs_bandwidth_used())
 		return;
 
 	if (!tick_nohz_full_cpu(cpu))
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 143f55df890b..929021fd6bc4 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -85,5 +85,3 @@ SCHED_FEAT(WA_BIAS, true)
 SCHED_FEAT(UTIL_EST, true)
 
 SCHED_FEAT(LATENCY_WARN, false)
-
-SCHED_FEAT(HZ_BW, true)
-- 
2.39.3

Re: [PATCH] sched: remove HZ_BW feature hedge

[Valentin Schneider]

On 15/05/24 09:37, Phil Auld wrote:
> As a hedge against unexpected user issues commit 88c56cfeaec4
> ("sched/fair: Block nohz tick_stop when cfs bandwidth in use")
> included a scheduler feature to disable the new functionality.
> It's been a few releases (v6.6) and no screams, so remove it.
>
> Signed-off-by: Phil Auld <pauld@redhat.com>
> Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Vincent Guittot <vincent.guittot@linaro.org>
> Cc: Juri Lelli <juri.lelli@redhat.com>
> Cc: Ben Segall <bsegall@google.com>
> Cc: Valentin Schneider <vschneid@redhat.com>

Looks like it's made it in a few distro releases already, so I'd say go for
it.

Reviewed-by: Valentin Schneider <vschneid@redhat.com>

[tip: sched/core] sched: remove HZ_BW feature hedge

[tip-bot2 for Phil Auld]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     a58501fb8320d6232507f722b4c9dcd4e03362ee
Gitweb:        https://git.kernel.org/tip/a58501fb8320d6232507f722b4c9dcd4e03362ee
Author:        Phil Auld <pauld@redhat.com>
AuthorDate:    Wed, 15 May 2024 09:37:05 -04:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Mon, 29 Jul 2024 12:22:34 +02:00

sched: remove HZ_BW feature hedge

As a hedge against unexpected user issues commit 88c56cfeaec4
("sched/fair: Block nohz tick_stop when cfs bandwidth in use")
included a scheduler feature to disable the new functionality.
It's been a few releases (v6.6) and no screams, so remove it.

Signed-off-by: Phil Auld <pauld@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <vschneid@redhat.com>
Link: https://lore.kernel.org/r/20240515133705.3632915-1-pauld@redhat.com
---
 kernel/sched/core.c     | 2 +-
 kernel/sched/fair.c     | 2 +-
 kernel/sched/features.h | 2 --
 3 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index db5823f..0a71050 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1269,7 +1269,7 @@ bool sched_can_stop_tick(struct rq *rq)
 	 * dequeued by migrating while the constrained task continues to run.
 	 * E.g. going from 2->1 without going through pick_next_task().
 	 */
-	if (sched_feat(HZ_BW) && __need_bw_check(rq, rq->curr)) {
+	if (__need_bw_check(rq, rq->curr)) {
 		if (cfs_task_bw_constrained(rq->curr))
 			return false;
 	}
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index e8cdfeb..02694fc 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6555,7 +6555,7 @@ static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p)
 {
 	int cpu = cpu_of(rq);
 
-	if (!sched_feat(HZ_BW) || !cfs_bandwidth_used())
+	if (!cfs_bandwidth_used())
 		return;
 
 	if (!tick_nohz_full_cpu(cpu))
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 143f55d..929021f 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -85,5 +85,3 @@ SCHED_FEAT(WA_BIAS, true)
 SCHED_FEAT(UTIL_EST, true)
 
 SCHED_FEAT(LATENCY_WARN, false)
-
-SCHED_FEAT(HZ_BW, true)