Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1267888 > unrolled thread

Re: [RFC 1/3] mm, oom: refactor oom detection

Started byMichal Hocko <mhocko@kernel.org>
First post2015-11-12 13:40 +0100
Last post2015-11-12 13:40 +0100
Articles 1 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [RFC 1/3] mm, oom: refactor oom detection Michal Hocko <mhocko@kernel.org> - 2015-11-12 13:40 +0100

#1267888 — Re: [RFC 1/3] mm, oom: refactor oom detection

FromMichal Hocko <mhocko@kernel.org>
Date2015-11-12 13:40 +0100
SubjectRe: [RFC 1/3] mm, oom: refactor oom detection
Message-ID<qtZku-rS-19@gated-at.bofh.it>
On Thu 29-10-15 16:17:13, mhocko@kernel.org wrote:
[...]
> @@ -3135,13 +3145,56 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
>  	if (gfp_mask & __GFP_NORETRY)
>  		goto noretry;
>  
> -	/* Keep reclaiming pages as long as there is reasonable progress */
> +	/*
> +	 * Do not retry high order allocations unless they are __GFP_REPEAT
> +	 * and even then do not retry endlessly.
> +	 */
>  	pages_reclaimed += did_some_progress;
> -	if ((did_some_progress && order <= PAGE_ALLOC_COSTLY_ORDER) ||
> -	    ((gfp_mask & __GFP_REPEAT) && pages_reclaimed < (1 << order))) {
> -		/* Wait for some write requests to complete then retry */
> -		wait_iff_congested(ac->preferred_zone, BLK_RW_ASYNC, HZ/50);
> -		goto retry;
> +	if (order > PAGE_ALLOC_COSTLY_ORDER) {
> +		if (!(gfp_mask & __GFP_REPEAT) || pages_reclaimed >= (1<<order))
> +			goto noretry;

This is not correct because we could fail __GFP_NOFAIL allocation. It
should do
		if (!(gfp_mask & __GFP_NOFAIL) &&
		   (!(gfp_mask & __GFP_REPEAT) || pages_reclaimed >= (1<<order)))
			goto noretry;

It looks rather ugly but it will get much better with patch3 which
reorganizes the code a bit

> +
> +		if (did_some_progress)
> +			goto retry;
> +	}

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

[toc] | [standalone]


Back to top | Article view | linux.kernel


csiph-web