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;
Message-ID: <f0f73a14-8738-6a17-b442-ec111a3b1064@linux.intel.com>
Date:   Mon, 10 Oct 2022 09:14:50 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.9.1
Subject: Re: [PATCH -next] sched/core: Fix the bug that traversal in
 sched_group_cookie_match is wrong
Content-Language: en-US
To:     Lin Shengwang <linshengwang1@huawei.com>, peterz@infradead.org,
        mingo@redhat.com, juri.lelli@redhat.com, vincent.guittot@linaro.org
Cc:     dietmar.eggemann@arm.com, rostedt@goodmis.org, bsegall@google.com,
        mgorman@suse.de, bristot@redhat.com, vschneid@redhat.com,
        linux-kernel@vger.kernel.org, cj.chengjian@huawei.com
References: <20221008022709.642-1-linshengwang1@huawei.com>
From:   Aubrey Li <aubrey.li@linux.intel.com>
In-Reply-To: <20221008022709.642-1-linshengwang1@huawei.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 10/8/22 10:27, Lin Shengwang wrote:
> In commit 97886d9dcd86 ("sched: Migration changes for core scheduling"),
> sched_group_cookie_match() was added to help finding cookie matched
> group, but was inconsistent with the actual purpose.
> 
> Using cpu_rq(cpu) instead of rq to fix the bug.
> 
> Fixes: 97886d9dcd86 ("sched: Migration changes for core scheduling")
> Signed-off-by: Lin Shengwang <linshengwang1@huawei.com>
> ---
>  kernel/sched/sched.h | 18 +++++++++---------
>  1 file changed, 9 insertions(+), 9 deletions(-)
> 
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 23c6f9f42ba1..1ba602139840 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -1182,6 +1182,14 @@ static inline bool is_migration_disabled(struct task_struct *p)
>  #endif
>  }
>  
> +DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
> +
> +#define cpu_rq(cpu)		(&per_cpu(runqueues, (cpu)))
> +#define this_rq()		this_cpu_ptr(&runqueues)
> +#define task_rq(p)		cpu_rq(task_cpu(p))
> +#define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
> +#define raw_rq()		raw_cpu_ptr(&runqueues)
> +
>  struct sched_group;
>  #ifdef CONFIG_SCHED_CORE
>  static inline struct cpumask *sched_group_span(struct sched_group *sg);
> @@ -1269,7 +1277,7 @@ static inline bool sched_group_cookie_match(struct rq *rq,
>  		return true;
>  
>  	for_each_cpu_and(cpu, sched_group_span(group), p->cpus_ptr) {
> -		if (sched_core_cookie_match(rq, p))
> +		if (sched_core_cookie_match(cpu_rq(cpu), p))>  			return true;
>  	}
>  	return false;
This looks a proper fix to me. Thanks!

-Aubrey