Re: [PATCH] sched: __set_cpus_allowed_ptr(): Check cpus_mask, not cpus_ptr

[Valentin Schneider <valentin.schneider@arm.com>]

On 17/06/20 13:17, Sebastian Andrzej Siewior wrote:
> From: Scott Wood <swood@redhat.com>
>
> This function is concerned with the long-term cpu mask, not the
> transitory mask the task might have while migrate disabled.  Before
> this patch, if a task was migrate disabled at the time
> __set_cpus_allowed_ptr() was called, and the new mask happened to be
> equal to the cpu that the task was running on, then the mask update
> would be lost.
>
> Signed-off-by: Scott Wood <swood@redhat.com>
> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
>  kernel/sched/core.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -1637,7 +1637,7 @@ static int __set_cpus_allowed_ptr(struct
>               goto out;
>       }
>
> -	if (cpumask_equal(p->cpus_ptr, new_mask))
> +	if (cpumask_equal(&p->cpus_mask, new_mask))
>               goto out;
>
>       /*

Makes sense, but what about the rest of the checks? Further down there is

        /* Can the task run on the task's current CPU? If so, we're done */
        if (cpumask_test_cpu(task_cpu(p), new_mask))
                goto out;

If the task is currently migrate disabled and for some stupid reason it
gets affined elsewhere, we could try to move it out - which AFAICT we don't
want to do because migrate disabled. So I suppose you'd want an extra
bailout condition here when the task is migrate disabled.

ISTR in RT you do re-check the affinity and potentially move the task away
when re-enabling migration, so that should work out all fine.