Re: [PATCH v2.1 01/10] cpu: Introduce clear_tasks_mm_cpumask() helper

[Oleg Nesterov <oleg@redhat.com>]

On 03/24, Anton Vorontsov wrote:
>
> Many architctures clear tasks' mm_cpumask like this:
>
> 	read_lock(&tasklist_lock);
> 	for_each_process(p) {
> 		if (p->mm)
> 			cpumask_clear_cpu(cpu, mm_cpumask(p->mm));
> 	}
> 	read_unlock(&tasklist_lock);

Namely arm, powerpc, and sh.

> The code above has several problems, such as:
>
> 1. Working with task->mm w/o getting mm or grabing the task lock is
>    dangerous as ->mm might disappear (exit_mm() assigns NULL under
>    task_lock(), so tasklist lock is not enough).

This is not actually true for arm and sh, afaics. They do not even
need tasklist or rcu lock for for_each_process().

__cpu_disable() is called by __stop_machine(), we know that nobody
can preempt us and other CPUs can do nothing.

> 2. Checking for process->mm is not enough because process' main
>    thread may exit or detach its mm via use_mm(), but other threads
>    may still have a valid mm.

Yes,

> Also, Per Peter Zijlstra's idea, now we don't grab tasklist_lock in
> the new helper, instead we take the rcu read lock. We can do this
> because the function is called after the cpu is taken down and marked
> offline, so no new tasks will get this cpu set in their mm mask.

And only powerpc needs rcu_read_lock() and task_lock().

OTOH, I do not understand why powepc does this on CPU_DEAD...
And probably CPU_UP_CANCELED doesn't need to clear mm_cpumask().

That said, personally I think these patches are fine, the common
helper makes sense.

Oleg.