Received-SPF: pass (google.com: domain of linux-kernel+bounces-214295-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Received-SPF: Pass (protection.outlook.com: domain of amd.com designates
 165.204.84.17 as permitted sender) receiver=protection.outlook.com;
 client-ip=165.204.84.17; helo=SATLEXMB04.amd.com; pr=C
Message-ID: <4c285c36-6893-c0c4-1061-0d7e16445e9d@amd.com>
Date: Fri, 14 Jun 2024 08:46:54 +0530
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
 Thunderbird/91.6.1
Subject: Re: [PATCH v2] sched/fair: Reschedule the cfs_rq when current is
 ineligible
Content-Language: en-US
From: K Prateek Nayak <kprateek.nayak@amd.com>
To: Chen Yu <yu.c.chen@intel.com>, Chunxin Zang <spring.cxz@gmail.com>
CC: Honglei Wang <jameshongleiwang@126.com>, <mingo@redhat.com>, Peter
 Zijlstra <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>, <vschneid@redhat.com>, Mike Galbraith <efault@gmx.de>,
	<linux-kernel@vger.kernel.org>, <yangchen11@lixiang.com>, Jerry Zhou
	<zhouchunhua@lixiang.com>, Chunxin Zang <zangchunxin@lixiang.com>
References: <20240529141806.16029-1-spring.cxz@gmail.com>
 <ZmKVjmuC4kGrUH5V@chenyu5-mobl2>
 <2E6EB0D6-D913-4205-B7DD-35EF4FA25667@gmail.com>
 <Zmrd2hwqHB8RQjqB@chenyu5-mobl2>
 <112FECA8-5D21-406F-814C-ACBE63351CBB@gmail.com>
 <ZmryuDqw2fEhHNni@chenyu5-mobl2>
 <9d83771f-3d12-25f0-c3f8-70dd7942e9ce@amd.com>
In-Reply-To: <9d83771f-3d12-25f0-c3f8-70dd7942e9ce@amd.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 8bit
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 14 Jun 2024 03:17:04.1141
 (UTC)
X-MS-Exchange-CrossTenant-Network-Message-Id: 2a0feaa8-f8ab-4dc4-47f1-08dc8c207708
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=3dd8961f-e488-4e60-8e11-a82d994e183d;Ip=[165.204.84.17];Helo=[SATLEXMB04.amd.com]
X-MS-Exchange-CrossTenant-AuthSource:
	BL6PEPF00020E60.namprd04.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Anonymous
X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SN7PR12MB6789

Hello Chenyu,

On 6/14/2024 8:41 AM, K Prateek Nayak wrote:
> Hello Chenyu,
> 
> Sorry, I'm a bit late to the thread but ...
> 
> On 6/13/2024 6:53 PM, Chen Yu wrote:
>> [..snip..]
>>>>
>>>> I wonder if we can only take care of the NO_RUN_TO_PARITY case? 
>>>> Something like this,
>>>>
>>>>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>>>>> index 03be0d1330a6..5e49a15bbdd3 100644
>>>>> --- a/kernel/sched/fair.c
>>>>> +++ b/kernel/sched/fair.c
>>>>> @@ -745,6 +745,21 @@ int entity_eligible(struct cfs_rq *cfs_rq, 
>>>>> struct sched_entity *se)
>>>>>         return vruntime_eligible(cfs_rq, se->vruntime);
>>>>> }
>>>>>
>>>>> +static bool check_entity_need_preempt(struct cfs_rq *cfs_rq, 
>>>>> struct sched_entity *se)
>>>>> +{
>>>> if (sched_feat(RUN_TO_PARITY) || cfs_rq->nr_running <= 1 ||
>>>>      !entity_eligible(cfs_rq, se))
>>>> return false;
>>>>
>>>> return true;
>>>>
>>>> Thoughts?
>>>>
>>>
>>> This does indeed look better. In that case, do I need to make the 
>>> changes this way and send
>>> out a version 3?
>>
>> If you mean the following changes, maybe we can continue the 
>> discussion here.
>> This is just my 2 cents, not sure what others think of it. Anyway, I 
>> can launch some tests.
>>
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index 8a5b1ae0aa55..c0fdb25f0695 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -744,6 +744,15 @@ int entity_eligible(struct cfs_rq *cfs_rq, struct 
>> sched_entity *se)
>>       return vruntime_eligible(cfs_rq, se->vruntime);
>>   }
>> +static bool check_curr_preempt(struct cfs_rq *cfs_rq, struct 
>> sched_entity *curr)
>> +{
>> +    if (sched_feat(RUN_TO_PARITY) || cfs_rq->nr_running <= 1 ||
>> +        !entity_eligible(cfs_rq, curr))
> 
> Shouldn't this return false if "entity_eligible(cfs_rq, curr)" returns 
> true since curr is still vruntime eligible on cfs_rq?
> 
> Would it better to have check_curr_preempt() as follows:
> 
>      if (sched_feat(RUN_TO_PARITY) || cfs_rq->nr_running <= 1)
>          return false;
> 
>      return entity_eligible(cfs_rq, curr);

The above return should have been:

	return !entity_eligible(cfs_rq, curr);

Which returns true once curr is not vruntime eligible on cfs_rq.
Sorry for the oversight.

> 
> which returns false if curr is ineligible and scheduler can go ahead and 
> and call schedule to evaluate the next best step?
> 
> Please let me know if I'm missing something.
> 
>> +        return false;
>> +
>> +    return true;
>> +}
>> +
>>   static u64 __update_min_vruntime(struct cfs_rq *cfs_rq, u64 vruntime)
>>   {
>>       u64 min_vruntime = cfs_rq->min_vruntime;
>> [..snip..]
> 

-- 
Thanks and Regards,
Prateek