Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Cruz Zhao <CruzZhao@linux.alibaba.com>
To:     mingo@redhat.com, peterz@infradead.org, juri.lelli@redhat.com,
        vincent.guittot@linaro.org, dietmar.eggemann@arm.com,
        rostedt@goodmis.org, bsegall@google.com, mgorman@suse.de,
        bristot@redhat.com
Cc:     linux-kernel@vger.kernel.org
Subject: [PATCH 0/3] sched/core: Optimize load balance of core scheduling
Date:   Tue, 28 Jun 2022 15:57:22 +0800
Message-Id: <1656403045-100840-1-git-send-email-CruzZhao@linux.alibaba.com>
Precedence: bulk

If the tasks with the same cookie are equal on SMT siblings, they can make
pairs when pick next task, and forceidle can be avoided.

In order to achieve this goal, we have to count how many tasks with this
cookie are in the runqueue. When counting we found a bug that task won't
enqueue into core tree when we update cookie of an uncookie'd task, so we
fix this bug first.

Cruz Zhao (3):
  sched/core: Fix the bug that task won't enqueue into core tree when
    update cookie
  sched/core: Introduce nr_running percpu for each cookie
  sched/core: Make tasks with the same cookie pairs on SMT siblings

 kernel/sched/core.c       |  7 +++++
 kernel/sched/core_sched.c | 18 ++++++------
 kernel/sched/fair.c       |  4 +--
 kernel/sched/sched.h      | 74 +++++++++++++++++++++++++++++++++++++++++++----
 4 files changed, 86 insertions(+), 17 deletions(-)

-- 
1.8.3.1