Re: [RFC 5/6] mm, cpuset: always use seqlock when changing task's nodemask

[Vlastimil Babka]

On 04/12/2017 10:10 AM, Hillf Danton wrote:
> On April 11, 2017 10:06 PM Vlastimil Babka wrote: 
>>
>>  static void cpuset_change_task_nodemask(struct task_struct *tsk,
>>  					nodemask_t *newmems)
>>  {
>> -	bool need_loop;
>> -
>>  	task_lock(tsk);
>> -	/*
>> -	 * Determine if a loop is necessary if another thread is doing
>> -	 * read_mems_allowed_begin().  If at least one node remains unchanged and
>> -	 * tsk does not have a mempolicy, then an empty nodemask will not be
>> -	 * possible when mems_allowed is larger than a word.
>> -	 */
>> -	need_loop = task_has_mempolicy(tsk) ||
>> -			!nodes_intersects(*newmems, tsk->mems_allowed);
>>
>> -	if (need_loop) {
>> -		local_irq_disable();
>> -		write_seqcount_begin(&tsk->mems_allowed_seq);
>> -	}
>> +	local_irq_disable();
>> +	write_seqcount_begin(&tsk->mems_allowed_seq);
>>
>> -	nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems);
>>  	mpol_rebind_task(tsk, newmems);
>>  	tsk->mems_allowed = *newmems;
>>
>> -	if (need_loop) {
>> -		write_seqcount_end(&tsk->mems_allowed_seq);
>> -		local_irq_enable();
>> -	}
>> +	write_seqcount_end(&tsk->mems_allowed_seq);
>>
> Doubt if we'd listen irq again.

Ugh, thanks for catching this. Looks like my testing config didn't have
lockup detectors enabled.

>>  	task_unlock(tsk);
>>  }
>> --
>> 2.12.2
>