From: Qing Wang <wangqing7171@gmail.com>
To: Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Juri Lelli <juri.lelli@redhat.com>,
Vincent Guittot <vincent.guittot@linaro.org>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Steven Rostedt <rostedt@goodmis.org>,
Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
Valentin Schneider <vschneid@redhat.com>,
K Prateek Nayak <kprateek.nayak@amd.com>,
Li Chen <chenl311@chinatelecom.cn>,
linux-kernel@vger.kernel.org
Cc: Qing Wang <wangqing7171@gmail.com>
Subject: [PATCH] sched/fair: Replace random newidle_balance with Bresenham accumulator
Date: Wed, 6 May 2026 10:44:55 +0800 [thread overview]
Message-ID: <20260506024455.2799069-1-wangqing7171@gmail.com> (raw)
In-Reply-To: <20260429085147.1029505-1-wangqing7171@gmail.com>
The current NI_RANDOM implementation uses a random dice roll to allow
newidle_balance attempts according to the success rate. There is a better
way to implememte it.
Replace the random dice with a Bresenham accumulator that distributes the
allowed attempts with uniformly spaced evenly across a 1024-step window:
Each step do those:
- Accumulate (1 + newidle_ratio) into newidle_window_pos.
- If the accumulator reaches 1024, allow the balance attempt and
subtract 1024.
This guarantees exactly (1 + newidle_ratio) newidle_balance per 1024 steps
and per newidle_balance with uniformly spaced for any ratio in [0, 1023].
Signed-off-by: Qing Wang <wangqing7171@gmail.com>
---
include/linux/sched/topology.h | 1 +
kernel/sched/fair.c | 18 +++++++++++++-----
kernel/sched/topology.c | 1 +
3 files changed, 15 insertions(+), 5 deletions(-)
diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h
index 36553e14866d..fd6de240fb2b 100644
--- a/include/linux/sched/topology.h
+++ b/include/linux/sched/topology.h
@@ -95,6 +95,7 @@ struct sched_domain {
unsigned int newidle_call;
unsigned int newidle_success;
unsigned int newidle_ratio;
+ unsigned int newidle_window_pos;
u64 newidle_stamp;
u64 max_newidle_lb_cost;
unsigned long last_decay_max_lb_cost;
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 69361c63353a..7874f795f62d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -12509,6 +12509,7 @@ static inline void update_newidle_stats(struct sched_domain *sd, unsigned int su
ratio += sd->newidle_success;
sd->newidle_ratio = min(1024, ratio);
+ sd->newidle_window_pos = 0;
sd->newidle_call /= 2;
sd->newidle_success /= 2;
}
@@ -13217,16 +13218,23 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf)
if (sched_feat(NI_RANDOM) && sd->newidle_ratio < 1024) {
/*
- * Throw a 1k sided dice; and only run
- * newidle_balance according to the success
- * rate.
+ * Base on accumulator of Bresenham's line algorithm
+ *
+ * Accumulate ratio each time and trigger balance
+ * when sum ≥ 1024.
+ *
+ * Ensures exactly ratio balances per 1024 calls
+ * and distributes ratio balance operations uniformly
+ * across every 1024 invocations.
*/
- u32 d1k = sched_rng() % 1024;
weight = 1 + sd->newidle_ratio;
- if (d1k > weight) {
+
+ sd->newidle_window_pos += weight;
+ if (sd->newidle_window_pos < 1024) {
update_newidle_stats(sd, 0);
continue;
}
+ sd->newidle_window_pos -= 1024;
weight = (1024 + weight/2) / weight;
}
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index 5847b83d9d55..0d8fa413f66c 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -1708,6 +1708,7 @@ sd_init(struct sched_domain_topology_level *tl,
.newidle_call = 512,
.newidle_success = 256,
.newidle_ratio = 512,
+ .newidle_window_pos = 0,
.newidle_stamp = now,
.max_newidle_lb_cost = 0,
--
2.34.1
next prev parent reply other threads:[~2026-05-06 2:45 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-07 16:06 [PATCH 0/4] sched: The newidle balance regression Peter Zijlstra
2025-11-07 16:06 ` [PATCH 1/4] sched/fair: Revert max_newidle_lb_cost bump Peter Zijlstra
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 2/4] sched/fair: Small cleanup to sched_balance_newidle() Peter Zijlstra
2025-11-10 13:55 ` Dietmar Eggemann
2025-11-10 14:04 ` Peter Zijlstra
2025-11-12 14:37 ` Shrikanth Hegde
2025-11-12 14:42 ` Peter Zijlstra
2025-11-12 15:08 ` Peter Zijlstra
2025-11-12 15:28 ` Shrikanth Hegde
2025-11-14 9:49 ` Peter Zijlstra
2025-11-14 10:22 ` Vincent Guittot
2025-11-14 11:05 ` Peter Zijlstra
2025-11-14 13:11 ` Vincent Guittot
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 3/4] sched/fair: Small cleanup to update_newidle_cost() Peter Zijlstra
2025-11-14 12:19 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2025-11-07 16:06 ` [PATCH 4/4] sched/fair: Proportional newidle balance Peter Zijlstra
2025-11-10 13:55 ` Dietmar Eggemann
2025-11-11 9:07 ` Adam Li
2025-11-11 9:20 ` Peter Zijlstra
2025-11-12 12:04 ` Adam Li
2025-11-12 13:41 ` Peter Zijlstra
2025-11-12 15:42 ` Shrikanth Hegde
2025-11-14 9:35 ` Peter Zijlstra
2025-11-14 12:18 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2025-11-17 16:23 ` tip-bot2 for Peter Zijlstra
2026-01-18 20:46 ` [PATCH 4/4] " Mario Roy
2026-01-23 10:50 ` Peter Zijlstra
2026-01-23 11:03 ` Peter Zijlstra
2026-01-23 12:24 ` K Prateek Nayak
2026-01-28 4:08 ` K Prateek Nayak
2026-01-27 4:15 ` Mario Roy
2026-01-27 10:40 ` Peter Zijlstra
2026-01-27 15:17 ` Peter Zijlstra
2026-01-30 1:44 ` Mario Roy
2026-01-30 4:14 ` Mario Roy
2026-02-24 9:13 ` [tip: sched/core] sched/fair: More complex proportional " tip-bot2 for Peter Zijlstra
2026-01-25 12:22 ` [PATCH 4/4] sched/fair: Proportional " Mohamed Abuelfotoh, Hazem
2026-01-27 8:44 ` Peter Zijlstra
2026-01-28 15:48 ` Mohamed Abuelfotoh, Hazem
2026-01-29 9:19 ` Peter Zijlstra
2026-01-29 9:24 ` Peter Zijlstra
2026-01-30 16:12 ` Mohamed Abuelfotoh, Hazem
2026-01-30 13:16 ` Mohamed Abuelfotoh, Hazem
2026-02-02 10:51 ` Peter Zijlstra
2026-02-02 11:07 ` Mohamed Abuelfotoh, Hazem
2026-02-04 12:45 ` Mohamed Abuelfotoh, Hazem
2026-02-04 13:27 ` Peter Zijlstra
2026-02-04 13:59 ` Mohamed Abuelfotoh, Hazem
2026-02-04 14:05 ` Peter Zijlstra
2026-02-04 22:48 ` Mohamed Abuelfotoh, Hazem
2026-01-27 8:50 ` Peter Zijlstra
2026-01-27 9:13 ` Peter Zijlstra
2026-01-28 16:24 ` Mohamed Abuelfotoh, Hazem
2026-01-28 16:03 ` Mohamed Abuelfotoh, Hazem
2026-04-29 8:51 ` Qing Wang
2026-05-06 2:44 ` Qing Wang [this message]
2025-11-10 19:47 ` [PATCH 0/4] sched: The newidle balance regression Chris Mason
2025-11-11 19:08 ` Josh Don
2025-11-12 21:59 ` Chris Mason
2025-11-14 9:37 ` Peter Zijlstra
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=20260506024455.2799069-1-wangqing7171@gmail.com \
--to=wangqing7171@gmail.com \
--cc=bsegall@google.com \
--cc=chenl311@chinatelecom.cn \
--cc=dietmar.eggemann@arm.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 \
/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