linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
To: linux-mm@kvack.org
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>,
	Mel Gorman <mgorman@suse.de>
Subject: [RFC PATCH 11/14] mm: Free folios in a batch in shrink_folio_list()
Date: Fri, 25 Aug 2023 14:59:15 +0100	[thread overview]
Message-ID: <20230825135918.4164671-12-willy@infradead.org> (raw)
In-Reply-To: <20230825135918.4164671-1-willy@infradead.org>

Use free_unref_page_batch() to free the folios.  This may increase
the numer of IPIs from calling try_to_unmap_flush() more often,
but that's going to be very workload-dependent.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
---
 mm/vmscan.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 6f13394b112e..965c429847fd 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1706,14 +1706,15 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 		struct pglist_data *pgdat, struct scan_control *sc,
 		struct reclaim_stat *stat, bool ignore_references)
 {
+	struct folio_batch free_folios;
 	LIST_HEAD(ret_folios);
-	LIST_HEAD(free_folios);
 	LIST_HEAD(demote_folios);
 	unsigned int nr_reclaimed = 0;
 	unsigned int pgactivate = 0;
 	bool do_demote_pass;
 	struct swap_iocb *plug = NULL;
 
+	folio_batch_init(&free_folios);
 	memset(stat, 0, sizeof(*stat));
 	cond_resched();
 	do_demote_pass = can_demote(pgdat->node_id, sc);
@@ -2111,14 +2112,11 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 		 */
 		nr_reclaimed += nr_pages;
 
-		/*
-		 * Is there need to periodically free_folio_list? It would
-		 * appear not as the counts should be low
-		 */
-		if (unlikely(folio_test_large(folio)))
-			destroy_large_folio(folio);
-		else
-			list_add(&folio->lru, &free_folios);
+		if (folio_batch_add(&free_folios, folio) == 0) {
+			mem_cgroup_uncharge_folios(&free_folios);
+			try_to_unmap_flush();
+			free_unref_folios(&free_folios);
+		}
 		continue;
 
 activate_locked_split:
@@ -2182,9 +2180,9 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 
 	pgactivate = stat->nr_activate[0] + stat->nr_activate[1];
 
-	mem_cgroup_uncharge_list(&free_folios);
+	mem_cgroup_uncharge_folios(&free_folios);
 	try_to_unmap_flush();
-	free_unref_page_list(&free_folios);
+	free_unref_folios(&free_folios);
 
 	list_splice(&ret_folios, folio_list);
 	count_vm_events(PGACTIVATE, pgactivate);
-- 
2.40.1



  parent reply	other threads:[~2023-08-25 13:59 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-25 13:59 [RFC PATCH 00/14] Rearrange batched folio freeing Matthew Wilcox (Oracle)
2023-08-25 13:59 ` [RFC PATCH 01/14] mm: Make folios_put() the basis of release_pages() Matthew Wilcox (Oracle)
2023-08-31 14:21   ` Ryan Roberts
2023-09-01  3:58     ` Matthew Wilcox
2023-09-01  8:14       ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 02/14] mm: Convert free_unref_page_list() to use folios Matthew Wilcox (Oracle)
2023-08-31 14:29   ` Ryan Roberts
2023-09-01  4:03     ` Matthew Wilcox
2023-09-01  8:15       ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 03/14] mm: Add free_unref_folios() Matthew Wilcox (Oracle)
2023-08-31 14:39   ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 04/14] mm: Use folios_put() in __folio_batch_release() Matthew Wilcox (Oracle)
2023-08-31 14:41   ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 05/14] memcg: Add mem_cgroup_uncharge_folios() Matthew Wilcox (Oracle)
2023-08-31 14:49   ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 06/14] mm: Remove use of folio list from folios_put() Matthew Wilcox (Oracle)
2023-08-31 14:53   ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 07/14] mm: Use free_unref_folios() in put_pages_list() Matthew Wilcox (Oracle)
2023-08-25 13:59 ` [RFC PATCH 08/14] mm: use __page_cache_release() in folios_put() Matthew Wilcox (Oracle)
2023-08-25 13:59 ` [RFC PATCH 09/14] mm: Handle large folios in free_unref_folios() Matthew Wilcox (Oracle)
2023-08-31 15:21   ` Ryan Roberts
2023-09-01  4:09     ` Matthew Wilcox
2023-08-25 13:59 ` [RFC PATCH 10/14] mm: Allow non-hugetlb large folios to be batch processed Matthew Wilcox (Oracle)
2023-08-31 15:28   ` Ryan Roberts
2023-09-01  4:10     ` Matthew Wilcox
2023-08-25 13:59 ` Matthew Wilcox (Oracle) [this message]
2023-09-04  3:43   ` [RFC PATCH 11/14] mm: Free folios in a batch in shrink_folio_list() Matthew Wilcox
2024-01-05 17:00     ` Matthew Wilcox
2023-08-25 13:59 ` [RFC PATCH 12/14] mm: Free folios directly in move_folios_to_lru() Matthew Wilcox (Oracle)
2023-08-31 15:46   ` Ryan Roberts
2023-09-01  4:16     ` Matthew Wilcox
2023-08-25 13:59 ` [RFC PATCH 13/14] memcg: Remove mem_cgroup_uncharge_list() Matthew Wilcox (Oracle)
2023-08-31 18:26   ` Ryan Roberts
2023-08-25 13:59 ` [RFC PATCH 14/14] mm: Remove free_unref_page_list() Matthew Wilcox (Oracle)
2023-08-31 18:27   ` Ryan Roberts
2023-08-30 18:50 ` [RFC PATCH 15/18] mm: Convert free_pages_and_swap_cache() to use folios_put() Matthew Wilcox (Oracle)
2023-08-30 18:50 ` [RFC PATCH 16/18] mm: Use a folio in __collapse_huge_page_copy_succeeded() Matthew Wilcox (Oracle)
2023-08-30 18:50 ` [RFC PATCH 17/18] mm: Convert free_swap_cache() to take a folio Matthew Wilcox (Oracle)
2023-08-31 18:49   ` Ryan Roberts
2023-08-30 18:50 ` [RFC PATCH 18/18] mm: Add pfn_range_put() Matthew Wilcox (Oracle)
2023-08-31 19:03   ` Ryan Roberts
2023-09-01  4:27     ` Matthew Wilcox
2023-09-01  7:59       ` Ryan Roberts
2023-09-04 13:25 ` [RFC PATCH 00/14] Rearrange batched folio freeing Ryan Roberts
2023-09-05 13:15   ` Matthew Wilcox
2023-09-05 13:26     ` Ryan Roberts
2023-09-05 14:00       ` Matthew Wilcox
2023-09-06  3:48         ` Matthew Wilcox
2023-09-06 10:23           ` Ryan Roberts

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=20230825135918.4164671-12-willy@infradead.org \
    --to=willy@infradead.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    /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).