[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240813224015.614146342@infradead.org>
Date: Wed, 14 Aug 2024 00:25:49 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: mingo@...nel.org,
tj@...nel.org,
void@...ifault.com
Cc: peterz@...radead.org,
juri.lelli@...hat.com,
vincent.guittot@...aro.org,
dietmar.eggemann@....com,
rostedt@...dmis.org,
bsegall@...gle.com,
mgorman@...e.de,
vschneid@...hat.com,
linux-kernel@...r.kernel.org,
joelaf@...gle.com
Subject: [PATCH 1/9] sched: Use set_next_task(.first) where required
Turns out the core_sched bits forgot to use the
set_next_task(.first=true) variant. Notably:
pick_next_task() := pick_task() + set_next_task(.first = true)
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
kernel/sched/core.c | 4 ++--
kernel/sched/sched.h | 4 ++++
2 files changed, 6 insertions(+), 2 deletions(-)
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -6010,7 +6010,7 @@ pick_next_task(struct rq *rq, struct tas
next = rq->core_pick;
if (next != prev) {
put_prev_task(rq, prev);
- set_next_task(rq, next);
+ set_next_task_first(rq, next);
}
rq->core_pick = NULL;
@@ -6184,7 +6184,7 @@ pick_next_task(struct rq *rq, struct tas
}
out_set_next:
- set_next_task(rq, next);
+ set_next_task_first(rq, next);
out:
if (rq->core->core_forceidle_count && next == rq->idle)
queue_core_balance(rq);
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -2363,6 +2363,10 @@ static inline void set_next_task(struct
next->sched_class->set_next_task(rq, next, false);
}
+static inline void set_next_task_first(struct rq *rq, struct task_struct *next)
+{
+ next->sched_class->set_next_task(rq, next, true);
+}
/*
* Helper to define a sched_class instance; each one is placed in a separate
Powered by blists - more mailing lists