Re: [PATCH] mm:memcg: add __GFP_NOWARN in __memcg_schedule_kmem_cache_create

[Michal Hocko]

On Wed 18-04-18 22:23:28, Minchan Kim wrote:
> On Wed, Apr 18, 2018 at 09:54:37AM +0200, Michal Hocko wrote:
> > On Wed 18-04-18 16:41:17, Minchan Kim wrote:
> > > On Wed, Apr 18, 2018 at 09:20:02AM +0200, Michal Hocko wrote:
> > > > On Wed 18-04-18 11:29:12, Minchan Kim wrote:
> > [...]
> > > > > Let's not make user scared.
> > > > 
> > > > This is not a proper explanation. So what exactly happens when this
> > > > allocation fails? I would suggest something like the following
> > > > "
> > > > __memcg_schedule_kmem_cache_create tries to create a shadow slab cache
> > > > and the worker allocation failure is not really critical because we will
> > > > retry on the next kmem charge. We might miss some charges but that
> > > > shouldn't be critical. The excessive allocation failure report is not
> > > > very much helpful. Replace it with a rate limited single line output so
> > > > that we know that there is a lot of these failures and that we need to
> > > > do something about it in future.
> > > > "
> > > > 
> > > > With the last part to be implemented of course.
> > > 
> > > If you want to see warning and catch on it in future, I don't see any reason
> > > to change it. Because I didn't see any excessive warning output that it could
> > > make system slow unless we did ratelimiting.
> > 
> > Yeah, but a single line would be as much informative and less scary to
> > users.
> > 
> > > It was a just report from non-MM guys who have a concern that somethings
> > > might go wrong on the system. I just wanted them relax since it's not
> > > critical.
> > 
> > I do agree with __GFP_NOWARN but I think a single line warning is due
> > and helpful for further debugging.
> 
> Okay, no problem. However, I don't feel we need ratelimit at this moment.
> We can do when we got real report. Let's add just one line warning.
> However, I have no talent to write a poem to express with one line.
> Could you help me?

What about
	pr_info("Failed to create memcg slab cache. Report if you see floods of these\n");
 
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 671d07e73a3b..e26f85cac63f 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -2201,8 +2201,11 @@ static void __memcg_schedule_kmem_cache_create(struct mem_cgroup *memcg,
>         struct memcg_kmem_cache_create_work *cw;
> 
>         cw = kmalloc(sizeof(*cw), GFP_NOWAIT | __GFP_NOWARN);
> -       if (!cw)
> +       if (!cw) {
> +               pr_warn("Fail to create shadow slab cache for memcg but it's not critical.\n");
> +               pr_warn("If you see lots of this message, send an email to linux-mm@xxxxxxxxx\n");
>                 return;
> +       }
> 
>         css_get(&memcg->css);

-- 
Michal Hocko
SUSE Labs