Re: [PATCH v13 2/2] Add oom victim's memcg to the oom context information

[David Rientjes]

On Fri, 13 Jul 2018, ufo19890607@xxxxxxxxx wrote:

> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index 531b2c86d4db..7fbd389ea779 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -434,10 +434,11 @@ static void dump_header(struct oom_control *oc, struct task_struct *p)
>  			oom_constraint_text[oc->constraint],
>  			nodemask_pr_args(oc->nodemask));
>  	cpuset_print_current_mems_allowed();
> +	mem_cgroup_print_oom_context(oc->memcg, p);
>  	pr_cont(",task=%s,pid=%5d,uid=%5d\n", p->comm, p->pid,
>  		from_kuid(&init_user_ns, task_uid(p)));
>  	if (is_memcg_oom(oc))
> -		mem_cgroup_print_oom_info(oc->memcg, p);
> +		mem_cgroup_print_oom_meminfo(oc->memcg);
>  	else {
>  		show_mem(SHOW_MEM_FILTER_NODES, oc->nodemask);
>  		if (is_dump_unreclaim_slabs())

Ugh, could we please not pad the pid and uid with spaces?  I don't think 
it achieves anything and just makes regex less robust.

Otherwise, looks good!