Linux-mm Archive on lore.kernel.org
 help / color / Atom feed
From: Mike Kravetz <mike.kravetz@oracle.com>
To: Hillf Danton <hdanton@sina.com>, Mel Gorman <mgorman@suse.de>
Cc: Vlastimil Babka <vbabka@suse.cz>,
	Michal Hocko <mhocko@kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	linux-kernel <linux-kernel@vger.kernel.org>,
	Johannes Weiner <hannes@cmpxchg.org>
Subject: Re: [Question] Should direct reclaim time be bounded?
Date: Fri, 12 Jul 2019 18:11:30 -0700
Message-ID: <c39e7cb3-204c-c4e3-fb43-7a37d91c0ccb@oracle.com> (raw)
In-Reply-To: <20190712054732.7264-1-hdanton@sina.com>

On 7/11/19 10:47 PM, Hillf Danton wrote:
> 
> On Thu, 11 Jul 2019 02:42:56 +0800 Mike Kravetz wrote:
>>
>> It is quite easy to hit the condition where:
>> nr_reclaimed == 0  && nr_scanned == 0 is true, but we skip the previous test
>>
> Then skipping check of __GFP_RETRY_MAYFAIL makes no sense in your case.
> It is restored in respin below.
> 
>> and the compaction check:
>> sc->nr_reclaimed < pages_for_compaction &&
>> 	inactive_lru_pages > pages_for_compaction
>> is true, so we return true before the below check of costly_fg_reclaim
>>
> This check is placed after COMPACT_SUCCESS; the latter is used to
> replace sc->nr_reclaimed < pages_for_compaction.
> 
> And dryrun detection is added based on the result of last round of
> shrinking of inactive pages, particularly when their number is large
> enough.
> 

Thanks Hillf.

This change does appear to eliminate the issue with stalls by
should_continue_reclaim returning true too often.  I need to think
some more about exactly what is impacted with the change.

With this change, the problem moves to compaction with should_compact_retry
returning true too often.  It is the same behavior seem when I simply removed
the __GFP_RETRY_MAYFAIL special casing in should_continue_reclaim.

At Mel's suggestion I removed the compaction_zonelist_suitable() call
from should_compact_retry.  This eliminated the compaction stalls.  Thanks
Mel.

With both changes, stalls appear to be eliminated.  This is promising.
I'll try to refine these approaches and continue testing.
-- 
Mike Kravetz

> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2571,18 +2571,6 @@ static inline bool should_continue_reclaim(struct pglist_data *pgdat,
>  			return false;
>  	}
> 
> -	/*
> -	 * If we have not reclaimed enough pages for compaction and the
> -	 * inactive lists are large enough, continue reclaiming
> -	 */
> -	pages_for_compaction = compact_gap(sc->order);
> -	inactive_lru_pages = node_page_state(pgdat, NR_INACTIVE_FILE);
> -	if (get_nr_swap_pages() > 0)
> -		inactive_lru_pages += node_page_state(pgdat, NR_INACTIVE_ANON);
> -	if (sc->nr_reclaimed < pages_for_compaction &&
> -			inactive_lru_pages > pages_for_compaction)
> -		return true;
> -
>  	/* If compaction would go ahead or the allocation would succeed, stop */
>  	for (z = 0; z <= sc->reclaim_idx; z++) {
>  		struct zone *zone = &pgdat->node_zones[z];
> @@ -2598,7 +2586,21 @@ static inline bool should_continue_reclaim(struct pglist_data *pgdat,
>  			;
>  		}
>  	}
> -	return true;
> +
> +	/*
> +	 * If we have not reclaimed enough pages for compaction and the
> +	 * inactive lists are large enough, continue reclaiming
> +	 */
> +	pages_for_compaction = compact_gap(sc->order);
> +	inactive_lru_pages = node_page_state(pgdat, NR_INACTIVE_FILE);
> +	if (get_nr_swap_pages() > 0)
> +		inactive_lru_pages += node_page_state(pgdat, NR_INACTIVE_ANON);
> +
> +	return inactive_lru_pages > pages_for_compaction &&
> +		/*
> +		 * avoid dryrun with plenty of inactive pages
> +		 */
> +		nr_scanned && nr_reclaimed;
>  }
> 
>  static bool pgdat_memcg_congested(pg_data_t *pgdat, struct mem_cgroup *memcg)
> --
> 


  reply index

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-12  5:47 Hillf Danton
2019-07-13  1:11 ` Mike Kravetz [this message]
  -- strict thread matches above, loose matches on Subject: below --
2019-07-11 15:44 Hillf Danton
2019-04-23  4:07 Mike Kravetz
2019-04-23  7:19 ` Michal Hocko
2019-04-23 16:39   ` Mike Kravetz
2019-04-24 14:35     ` Vlastimil Babka
2019-06-28 18:20       ` Mike Kravetz
2019-07-01  8:59         ` Mel Gorman
2019-07-02  3:15           ` Mike Kravetz
2019-07-03  9:43             ` Mel Gorman
2019-07-03 23:54               ` Mike Kravetz
2019-07-04 11:09                 ` Michal Hocko
2019-07-04 15:11                   ` Mike Kravetz
2019-07-08  5:19             ` Hillf Danton
2019-07-10 18:42             ` Mike Kravetz
2019-07-10 19:44               ` Michal Hocko
2019-07-10 23:36                 ` Mike Kravetz
2019-07-11  7:12                   ` Michal Hocko
2019-07-12  9:49                     ` Mel Gorman

Reply instructions:

You may reply publically to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c39e7cb3-204c-c4e3-fb43-7a37d91c0ccb@oracle.com \
    --to=mike.kravetz@oracle.com \
    --cc=hannes@cmpxchg.org \
    --cc=hdanton@sina.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=mhocko@kernel.org \
    --cc=vbabka@suse.cz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

Linux-mm Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-mm/0 linux-mm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-mm linux-mm/ https://lore.kernel.org/linux-mm \
		linux-mm@kvack.org linux-mm@archiver.kernel.org
	public-inbox-index linux-mm


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kvack.linux-mm


AGPL code for this site: git clone https://public-inbox.org/ public-inbox