Re: [PATCH v2 02/11] memcg: Reclaim when more than one page needed.

[Michal Hocko]

On Thu 09-08-12 17:01:10, Glauber Costa wrote:
[...]
> @@ -2317,18 +2318,18 @@ static int mem_cgroup_do_charge(struct mem_cgroup *memcg, gfp_t gfp_mask,
>  	} else
>  		mem_over_limit = mem_cgroup_from_res_counter(fail_res, res);
>  	/*
> -	 * nr_pages can be either a huge page (HPAGE_PMD_NR), a batch
> -	 * of regular pages (CHARGE_BATCH), or a single regular page (1).
> -	 *
>  	 * Never reclaim on behalf of optional batching, retry with a
>  	 * single page instead.
>  	 */
> -	if (nr_pages == CHARGE_BATCH)
> +	if (nr_pages > min_pages)
>  		return CHARGE_RETRY;

This is dangerous because THP charges will be retried now while they
previously failed with CHARGE_NOMEM which means that we will keep
attempting potentially endlessly.
Why cannot we simply do if (nr_pages < CHARGE_BATCH) and get rid of the
min_pages altogether?
Also the comment doesn't seem to be valid anymore.

>  
>  	if (!(gfp_mask & __GFP_WAIT))
>  		return CHARGE_WOULDBLOCK;
>  
> +	if (gfp_mask & __GFP_NORETRY)
> +		return CHARGE_NOMEM;
> +
>  	ret = mem_cgroup_reclaim(mem_over_limit, gfp_mask, flags);
>  	if (mem_cgroup_margin(mem_over_limit) >= nr_pages)
>  		return CHARGE_RETRY;
> @@ -2341,7 +2342,7 @@ static int mem_cgroup_do_charge(struct mem_cgroup *memcg, gfp_t gfp_mask,
>  	 * unlikely to succeed so close to the limit, and we fall back
>  	 * to regular pages anyway in case of failure.
>  	 */
> -	if (nr_pages == 1 && ret)
> +	if (nr_pages <= (1 << PAGE_ALLOC_COSTLY_ORDER) && ret)
>  		return CHARGE_RETRY;
>  
>  	/*
> @@ -2476,7 +2477,8 @@ again:
>  			nr_oom_retries = MEM_CGROUP_RECLAIM_RETRIES;
>  		}
>  
> -		ret = mem_cgroup_do_charge(memcg, gfp_mask, batch, oom_check);
> +		ret = mem_cgroup_do_charge(memcg, gfp_mask, batch, nr_pages,
> +		    oom_check);
>  		switch (ret) {
>  		case CHARGE_OK:
>  			break;
> -- 
> 1.7.11.2
> 
> --
> To unsubscribe from this list: send the line "unsubscribe cgroups" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/