public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Wen Congyang <wency@cn.fujitsu.com>
To: David Rientjes <rientjes@google.com>
Cc: Tang Chen <tangchen@cn.fujitsu.com>,
	mingo@redhat.com, peterz@infradead.org, miaox@cn.fujitsu.com,
	linux-kernel@vger.kernel.org, linux-numa@vger.kernel.org
Subject: Re: [PATCH] Do not use cpu_to_node() to find an offlined cpu's node.
Date: Tue, 09 Oct 2012 16:34:18 +0800	[thread overview]
Message-ID: <5073E18A.2090203@cn.fujitsu.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1210082308230.360@chino.kir.corp.google.com>

At 10/09/2012 02:21 PM, David Rientjes Wrote:
> On Mon, 8 Oct 2012, Tang Chen wrote:
> 
>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>> index 66b36ab..e76dce9 100644
>> --- a/kernel/sched/core.c
>> +++ b/kernel/sched/core.c
>> @@ -1263,18 +1263,24 @@ EXPORT_SYMBOL_GPL(kick_process);
>>   */
>>  static int select_fallback_rq(int cpu, struct task_struct *p)
>>  {
>> -	const struct cpumask *nodemask = cpumask_of_node(cpu_to_node(cpu));
>> +	int nid = cpu_to_node(cpu);
>> +	const struct cpumask *nodemask = NULL;
>>  	enum { cpuset, possible, fail } state = cpuset;
>>  	int dest_cpu;
>>  
>> -	/* Look for allowed, online CPU in same node. */
>> -	for_each_cpu(dest_cpu, nodemask) {
>> -		if (!cpu_online(dest_cpu))
>> -			continue;
>> -		if (!cpu_active(dest_cpu))
>> -			continue;
>> -		if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
>> -			return dest_cpu;
>> +	/* If the cpu has been offlined, its nid was set to -1. */
>> +	if (nid != -1) {
> 
> NUMA_NO_NODE.
> 
> Eek, the nid shouldn't be -1 yet, though, for cpu hotplug since this 
> should be called at CPU_DYING level and migrate_tasks() still sees a valid 
> cpu.

the cpu's node is set when the cpu is hotpluged(not online), and it will
be cleared when the cpu is hotremoved(This patch is in akpm tree):
https://lkml.org/lkml/2012/9/3/39

I guess the task is in sleep state when the cpu is offlined, and it doesn't
be migrated to another cpu.

Thanks
Wen Congyang

> 
> On x86, cpumask_of_node() is always guaranteed to return a valid cpumask 
> after boot so presumably this is a problem in some non-x86 arch code and 
> isn't actually a sched problem.
> 
>> +		nodemask = cpumask_of_node(nid);
>> +
>> +		/* Look for allowed, online CPU in same node. */
>> +		for_each_cpu(dest_cpu, nodemask) {
>> +			if (!cpu_online(dest_cpu))
>> +				continue;
>> +			if (!cpu_active(dest_cpu))
>> +				continue;
>> +			if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p)))
>> +				return dest_cpu;
>> +		}
>>  	}
>>  
>>  	for (;;) {
> 


  reply	other threads:[~2012-10-09  8:28 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-08  2:59 [PATCH] Do not use cpu_to_node() to find an offlined cpu's node Tang Chen
2012-10-09  6:21 ` David Rientjes
2012-10-09  8:34   ` Wen Congyang [this message]
2012-10-09  8:39     ` Tang Chen
2012-10-09 10:04       ` David Rientjes
2012-10-09 10:22         ` Wen Congyang
2012-10-09 20:00           ` David Rientjes
2012-10-09 10:57 ` Peter Zijlstra
2012-10-09 20:36   ` David Rientjes
2012-10-09 20:47     ` Peter Zijlstra
2012-10-09 23:27       ` David Rientjes
2012-10-10  2:06         ` Wen Congyang
2012-10-10  3:48           ` Wen Congyang
2012-10-10  9:10         ` Peter Zijlstra
2012-10-10  9:33           ` Wen Congyang
2012-10-10  9:51             ` Peter Zijlstra
2012-10-10 10:10               ` Wen Congyang
2012-10-10 10:07                 ` Peter Zijlstra
2012-10-10 20:30           ` David Rientjes
2012-10-10 20:37             ` Andrew Morton
2012-10-10 20:57               ` David Rientjes
2012-10-18  0:52                 ` David Rientjes
2012-10-18  2:51                   ` Tang Chen
2012-10-18  3:29                     ` David Rientjes
2012-10-19 11:18                       ` 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=5073E18A.2090203@cn.fujitsu.com \
    --to=wency@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-numa@vger.kernel.org \
    --cc=miaox@cn.fujitsu.com \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rientjes@google.com \
    --cc=tangchen@cn.fujitsu.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