Re: [PATCH v2] mm: optimize the redundant loop of mm_update_next_owner()

[Michal Hocko]

On Thu 20-06-24 23:27:45, alexjlzheng@xxxxxxxxx wrote:
> From: Jinliang Zheng <alexjlzheng@xxxxxxxxxxx>
> 
> When mm_update_next_owner() is racing with swapoff (try_to_unuse()) or /proc or
> ptrace or page migration (get_task_mm()), it is impossible to find an
> appropriate task_struct in the loop whose mm_struct is the same as the target
> mm_struct.
> 
> If the above race condition is combined with the stress-ng-zombie and
> stress-ng-dup tests, such a long loop can easily cause a Hard Lockup in
> write_lock_irq() for tasklist_lock.
> 
> Recognize this situation in advance and exit early.
> 
> Signed-off-by: Jinliang Zheng <alexjlzheng@xxxxxxxxxxx>

Even if this is not really a full fix it is a useful stop gap to catch
at least some cases.

Acked-by: Michal Hocko <mhocko@xxxxxxxx>

> ---
> Changelog:
> 
> V2: Fix mm_update_owner_next() to mm_update_next_owner() in comment
> ---
>  kernel/exit.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/kernel/exit.c b/kernel/exit.c
> index f95a2c1338a8..81fcee45d630 100644
> --- a/kernel/exit.c
> +++ b/kernel/exit.c
> @@ -484,6 +484,8 @@ void mm_update_next_owner(struct mm_struct *mm)
>  	 * Search through everything else, we should not get here often.
>  	 */
>  	for_each_process(g) {
> +		if (atomic_read(&mm->mm_users) <= 1)
> +			break;
>  		if (g->flags & PF_KTHREAD)
>  			continue;
>  		for_each_thread(g, c) {
> -- 
> 2.39.3
> 

-- 
Michal Hocko
SUSE Labs