[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240624082011.4990-1-xuewen.yan@unisoc.com>
Date: Mon, 24 Jun 2024 16:20:09 +0800
From: Xuewen Yan <xuewen.yan@...soc.com>
To: <vincent.guittot@...aro.org>, <dietmar.eggemann@....com>,
<mingo@...hat.com>, <peterz@...radead.org>, <juri.lelli@...hat.com>,
<qyousef@...alina.io>
CC: <rostedt@...dmis.org>, <bsegall@...gle.com>, <mgorman@...e.de>,
<bristot@...hat.com>, <vschneid@...hat.com>,
<christian.loehle@....com>, <vincent.donnefort@....com>,
<ke.wang@...soc.com>, <di.shen@...soc.com>, <xuewen.yan94@...il.com>,
<linux-kernel@...r.kernel.org>
Subject: [PATCH V2 0/2] sched/fair: Some improvements to feec()
Patch 1 clamp the util with per-CPU instand of per-PD when calc pd_busy_time.
Patch 2 use actual_cpu_capacity when judging util_fits_cpu to cover the case
which rq-uclamp-max would cause some corner case wrong.
Changelog:
V1:
https://lore.kernel.org/all/20240606070645.3295-1-xuewen.yan@unisoc.com/
v2:
(1) remove the eenv->pd_cap capping in eenv_pd_busy_time() for patch-1;
(2) change commit-message for patch-1;
(3) add patch 2.
Xuewen Yan (2):
sched/fair: Prevent cpu_busy_time from exceeding actual_cpu_capacity
sched/fair: Use actual_cpu_capacity everywhere in util_fits_cpu()
kernel/sched/fair.c | 28 ++++++++++++++--------------
1 file changed, 14 insertions(+), 14 deletions(-)
--
2.25.1
Powered by blists - more mailing lists