From: Mike Kravetz <mike.kravetz@oracle.com>
To: Vlastimil Babka <vbabka@suse.cz>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc: Hillf Danton <hdanton@sina.com>, Michal Hocko <mhocko@kernel.org>,
Mel Gorman <mgorman@suse.de>,
Johannes Weiner <hannes@cmpxchg.org>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [RFC PATCH 2/3] mm, compaction: use MIN_COMPACT_COSTLY_PRIORITY everywhere for costly orders
Date: Thu, 1 Aug 2019 13:33:02 -0700 [thread overview]
Message-ID: <2f1d6779-2b87-4699-abf7-0aa59a2e74d9@oracle.com> (raw)
In-Reply-To: <89ba8e07-b0f8-4334-070e-02fbdfc361e3@suse.cz>
On 8/1/19 6:01 AM, Vlastimil Babka wrote:
> Could you try testing the patch below instead? It should hopefully
> eliminate the stalls. If it makes hugepage allocation give up too early,
> we'll know we have to involve __GFP_RETRY_MAYFAIL in allowing the
> MIN_COMPACT_PRIORITY priority. Thanks!
Thanks. This patch does eliminate the stalls I was seeing.
In my testing, there is little difference in how many hugetlb pages are
allocated. It does not appear to be giving up/failing too early. But,
this is only with __GFP_RETRY_MAYFAIL. The real concern would with THP
requests. Any suggestions on how to test that?
--
Mike Kravetz
> ----8<----
> diff --git a/include/linux/compaction.h b/include/linux/compaction.h
> index 9569e7c786d3..b8bfe8d5d2e9 100644
> --- a/include/linux/compaction.h
> +++ b/include/linux/compaction.h
> @@ -129,11 +129,7 @@ static inline bool compaction_failed(enum compact_result result)
> return false;
> }
>
> -/*
> - * Compaction has backed off for some reason. It might be throttling or
> - * lock contention. Retrying is still worthwhile.
> - */
> -static inline bool compaction_withdrawn(enum compact_result result)
> +static inline bool compaction_needs_reclaim(enum compact_result result)
> {
> /*
> * Compaction backed off due to watermark checks for order-0
> @@ -142,6 +138,15 @@ static inline bool compaction_withdrawn(enum compact_result result)
> if (result == COMPACT_SKIPPED)
> return true;
>
> + return false;
> +}
> +
> +/*
> + * Compaction has backed off for some reason. It might be throttling or
> + * lock contention. Retrying is still worthwhile.
> + */
> +static inline bool compaction_withdrawn(enum compact_result result)
> +{
> /*
> * If compaction is deferred for high-order allocations, it is
> * because sync compaction recently failed. If this is the case
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 272c6de1bf4e..3dfce1f79112 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -3965,6 +3965,11 @@ should_compact_retry(struct alloc_context *ac, int order, int alloc_flags,
> if (compaction_failed(compact_result))
> goto check_priority;
>
> + if (compaction_needs_reclaim(compact_result)) {
> + ret = compaction_zonelist_suitable(ac, order, alloc_flags);
> + goto out;
> + }
> +
> /*
> * make sure the compaction wasn't deferred or didn't bail out early
> * due to locks contention before we declare that we should give up.
> @@ -3972,8 +3977,7 @@ should_compact_retry(struct alloc_context *ac, int order, int alloc_flags,
> * compaction.
> */
> if (compaction_withdrawn(compact_result)) {
> - ret = compaction_zonelist_suitable(ac, order, alloc_flags);
> - goto out;
> + goto check_priority;
> }
>
> /*
>
next prev parent reply other threads:[~2019-08-01 20:33 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-24 17:50 [RFC PATCH 0/3] fix hugetlb page allocation stalls Mike Kravetz
2019-07-24 17:50 ` [RFC PATCH 1/3] mm, reclaim: make should_continue_reclaim perform dryrun detection Mike Kravetz
2019-07-25 8:05 ` Mel Gorman
2019-07-26 8:12 ` Mel Gorman
2019-07-31 11:08 ` Vlastimil Babka
2019-07-31 12:25 ` Mel Gorman
2019-07-31 21:11 ` Mike Kravetz
2019-08-01 8:44 ` Vlastimil Babka
2019-07-24 17:50 ` [RFC PATCH 2/3] mm, compaction: use MIN_COMPACT_COSTLY_PRIORITY everywhere for costly orders Mike Kravetz
2019-07-25 8:06 ` Mel Gorman
2019-07-31 12:06 ` Vlastimil Babka
2019-07-31 20:30 ` Mike Kravetz
2019-08-01 13:01 ` Vlastimil Babka
2019-08-01 20:33 ` Mike Kravetz [this message]
2019-08-02 10:20 ` Vlastimil Babka
2019-08-02 12:05 ` Vlastimil Babka
2019-08-02 17:44 ` Mike Kravetz
2019-07-24 17:50 ` [RFC PATCH 3/3] hugetlbfs: don't retry when pool page allocations start to fail Mike Kravetz
2019-07-25 8:13 ` Mel Gorman
2019-07-25 17:15 ` Mike Kravetz
2019-07-25 22:43 ` Mel Gorman
2019-07-31 13:23 ` Vlastimil Babka
2019-07-31 21:13 ` Mike Kravetz
Reply instructions:
You may reply publicly 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=2f1d6779-2b87-4699-abf7-0aa59a2e74d9@oracle.com \
--to=mike.kravetz@oracle.com \
--cc=akpm@linux-foundation.org \
--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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).