linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Dan Schatzberg <schatzberg.dan@gmail.com>
To: Johannes Weiner <hannes@cmpxchg.org>,
	Roman Gushchin <roman.gushchin@linux.dev>,
	Yosry Ahmed <yosryahmed@google.com>, Huan Yang <link@vivo.com>
Cc: linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
	linux-mm@kvack.org, Michal Hocko <mhocko@kernel.org>,
	Shakeel Butt <shakeelb@google.com>,
	Muchun Song <muchun.song@linux.dev>,
	Andrew Morton <akpm@linux-foundation.org>,
	David Hildenbrand <david@redhat.com>,
	Matthew Wilcox <willy@infradead.org>,
	Huang Ying <ying.huang@intel.com>,
	Kefeng Wang <wangkefeng.wang@huawei.com>,
	Peter Xu <peterx@redhat.com>,
	"Vishal Moola (Oracle)" <vishal.moola@gmail.com>,
	Yue Zhao <findns94@gmail.com>, Hugh Dickins <hughd@google.com>
Subject: [PATCH 1/1] mm: add swapiness= arg to memory.reclaim
Date: Thu, 30 Nov 2023 07:36:54 -0800	[thread overview]
Message-ID: <20231130153658.527556-2-schatzberg.dan@gmail.com> (raw)
In-Reply-To: <20231130153658.527556-1-schatzberg.dan@gmail.com>

Allow proactive reclaimers to submit an additional swappiness=<val>
argument to memory.reclaim. This overrides the global or per-memcg
swappiness setting for that reclaim attempt.

For example:

echo "2M swappiness=0" > /sys/fs/cgroup/memory.reclaim

will perform reclaim on the rootcg with a swappiness setting of 0 (no
swap) regardless of the vm.swappiness sysctl setting.

Signed-off-by: Dan Schatzberg <schatzberg.dan@gmail.com>
---
 include/linux/swap.h |  3 ++-
 mm/memcontrol.c      | 55 +++++++++++++++++++++++++++++++++++---------
 mm/vmscan.c          | 13 +++++++++--
 3 files changed, 57 insertions(+), 14 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index f6dd6575b905..c6e309199f10 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -410,7 +410,8 @@ extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
 extern unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
 						  unsigned long nr_pages,
 						  gfp_t gfp_mask,
-						  unsigned int reclaim_options);
+						  unsigned int reclaim_options,
+						  int *swappiness);
 extern unsigned long mem_cgroup_shrink_node(struct mem_cgroup *mem,
 						gfp_t gfp_mask, bool noswap,
 						pg_data_t *pgdat,
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 1c1061df9cd1..ba1c89455ab0 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -63,6 +63,7 @@
 #include <linux/resume_user_mode.h>
 #include <linux/psi.h>
 #include <linux/seq_buf.h>
+#include <linux/parser.h>
 #include <linux/sched/isolation.h>
 #include "internal.h"
 #include <net/sock.h>
@@ -2449,7 +2450,7 @@ static unsigned long reclaim_high(struct mem_cgroup *memcg,
 		psi_memstall_enter(&pflags);
 		nr_reclaimed += try_to_free_mem_cgroup_pages(memcg, nr_pages,
 							gfp_mask,
-							MEMCG_RECLAIM_MAY_SWAP);
+							MEMCG_RECLAIM_MAY_SWAP, NULL);
 		psi_memstall_leave(&pflags);
 	} while ((memcg = parent_mem_cgroup(memcg)) &&
 		 !mem_cgroup_is_root(memcg));
@@ -2740,7 +2741,7 @@ static int try_charge_memcg(struct mem_cgroup *memcg, gfp_t gfp_mask,
 
 	psi_memstall_enter(&pflags);
 	nr_reclaimed = try_to_free_mem_cgroup_pages(mem_over_limit, nr_pages,
-						    gfp_mask, reclaim_options);
+						    gfp_mask, reclaim_options, NULL);
 	psi_memstall_leave(&pflags);
 
 	if (mem_cgroup_margin(mem_over_limit) >= nr_pages)
@@ -3660,7 +3661,7 @@ static int mem_cgroup_resize_max(struct mem_cgroup *memcg,
 		}
 
 		if (!try_to_free_mem_cgroup_pages(memcg, 1, GFP_KERNEL,
-					memsw ? 0 : MEMCG_RECLAIM_MAY_SWAP)) {
+					memsw ? 0 : MEMCG_RECLAIM_MAY_SWAP, NULL)) {
 			ret = -EBUSY;
 			break;
 		}
@@ -3774,7 +3775,7 @@ static int mem_cgroup_force_empty(struct mem_cgroup *memcg)
 			return -EINTR;
 
 		if (!try_to_free_mem_cgroup_pages(memcg, 1, GFP_KERNEL,
-						  MEMCG_RECLAIM_MAY_SWAP))
+						  MEMCG_RECLAIM_MAY_SWAP, NULL))
 			nr_retries--;
 	}
 
@@ -6720,7 +6721,7 @@ static ssize_t memory_high_write(struct kernfs_open_file *of,
 		}
 
 		reclaimed = try_to_free_mem_cgroup_pages(memcg, nr_pages - high,
-					GFP_KERNEL, MEMCG_RECLAIM_MAY_SWAP);
+					GFP_KERNEL, MEMCG_RECLAIM_MAY_SWAP, NULL);
 
 		if (!reclaimed && !nr_retries--)
 			break;
@@ -6769,7 +6770,7 @@ static ssize_t memory_max_write(struct kernfs_open_file *of,
 
 		if (nr_reclaims) {
 			if (!try_to_free_mem_cgroup_pages(memcg, nr_pages - max,
-					GFP_KERNEL, MEMCG_RECLAIM_MAY_SWAP))
+					GFP_KERNEL, MEMCG_RECLAIM_MAY_SWAP, NULL))
 				nr_reclaims--;
 			continue;
 		}
@@ -6895,6 +6896,16 @@ static ssize_t memory_oom_group_write(struct kernfs_open_file *of,
 	return nbytes;
 }
 
+enum {
+	MEMORY_RECLAIM_SWAPPINESS = 0,
+	MEMORY_RECLAIM_NULL,
+};
+
+static const match_table_t if_tokens = {
+	{ MEMORY_RECLAIM_SWAPPINESS, "swappiness=%d"},
+	{ MEMORY_RECLAIM_NULL, NULL },
+};
+
 static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
 			      size_t nbytes, loff_t off)
 {
@@ -6902,12 +6913,33 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
 	unsigned int nr_retries = MAX_RECLAIM_RETRIES;
 	unsigned long nr_to_reclaim, nr_reclaimed = 0;
 	unsigned int reclaim_options;
-	int err;
+	char *old_buf, *start;
+	substring_t args[MAX_OPT_ARGS];
+	int swappiness = -1;
 
 	buf = strstrip(buf);
-	err = page_counter_memparse(buf, "", &nr_to_reclaim);
-	if (err)
-		return err;
+
+	old_buf = buf;
+	nr_to_reclaim = memparse(buf, &buf) / PAGE_SIZE;
+	if (buf == old_buf)
+		return -EINVAL;
+
+	buf = strstrip(buf);
+
+	while ((start = strsep(&buf, " ")) != NULL) {
+		if (!strlen(start))
+			continue;
+		switch (match_token(start, if_tokens, args)) {
+		case MEMORY_RECLAIM_SWAPPINESS:
+			if (match_int(&args[0], &swappiness))
+				return -EINVAL;
+			if (swappiness < 0 || swappiness > 200)
+				return -EINVAL;
+			break;
+		default:
+			return -EINVAL;
+		}
+	}
 
 	reclaim_options	= MEMCG_RECLAIM_MAY_SWAP | MEMCG_RECLAIM_PROACTIVE;
 	while (nr_reclaimed < nr_to_reclaim) {
@@ -6926,7 +6958,8 @@ static ssize_t memory_reclaim(struct kernfs_open_file *of, char *buf,
 
 		reclaimed = try_to_free_mem_cgroup_pages(memcg,
 					min(nr_to_reclaim - nr_reclaimed, SWAP_CLUSTER_MAX),
-					GFP_KERNEL, reclaim_options);
+					GFP_KERNEL, reclaim_options,
+					swappiness == -1 ? NULL : &swappiness);
 
 		if (!reclaimed && !nr_retries--)
 			return -EAGAIN;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 506f8220c5fe..546704ea01e1 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -136,6 +136,9 @@ struct scan_control {
 	/* Always discard instead of demoting to lower tier memory */
 	unsigned int no_demotion:1;
 
+	/* Swappiness value for reclaim, if NULL use memcg/global value */
+	int *swappiness;
+
 	/* Allocation order */
 	s8 order;
 
@@ -2327,7 +2330,8 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	unsigned long anon_cost, file_cost, total_cost;
-	int swappiness = mem_cgroup_swappiness(memcg);
+	int swappiness = sc->swappiness ?
+		*sc->swappiness : mem_cgroup_swappiness(memcg);
 	u64 fraction[ANON_AND_FILE];
 	u64 denominator = 0;	/* gcc */
 	enum scan_balance scan_balance;
@@ -2608,6 +2612,9 @@ static int get_swappiness(struct lruvec *lruvec, struct scan_control *sc)
 	    mem_cgroup_get_nr_swap_pages(memcg) < MIN_LRU_BATCH)
 		return 0;
 
+	if (sc->swappiness)
+		return *sc->swappiness;
+
 	return mem_cgroup_swappiness(memcg);
 }
 
@@ -6433,7 +6440,8 @@ unsigned long mem_cgroup_shrink_node(struct mem_cgroup *memcg,
 unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
 					   unsigned long nr_pages,
 					   gfp_t gfp_mask,
-					   unsigned int reclaim_options)
+					   unsigned int reclaim_options,
+					   int *swappiness)
 {
 	unsigned long nr_reclaimed;
 	unsigned int noreclaim_flag;
@@ -6448,6 +6456,7 @@ unsigned long try_to_free_mem_cgroup_pages(struct mem_cgroup *memcg,
 		.may_unmap = 1,
 		.may_swap = !!(reclaim_options & MEMCG_RECLAIM_MAY_SWAP),
 		.proactive = !!(reclaim_options & MEMCG_RECLAIM_PROACTIVE),
+		.swappiness = swappiness,
 	};
 	/*
 	 * Traverse the ZONELIST_FALLBACK zonelist of the current node to put
-- 
2.34.1



  reply	other threads:[~2023-11-30 15:38 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-30 15:36 [PATCH 0/1] Add swappiness argument " Dan Schatzberg
2023-11-30 15:36 ` Dan Schatzberg [this message]
2023-11-30 21:33   ` [PATCH 1/1] mm: add swapiness= arg " Andrew Morton
2023-11-30 21:46     ` Dan Schatzberg
2023-12-01  1:56   ` Huan Yang
2023-12-01  2:05     ` Yosry Ahmed
2023-12-01  2:13       ` Huan Yang
2023-12-01  2:17         ` Yosry Ahmed
2023-12-01  2:24           ` Huan Yang
2023-11-30 15:57 ` [PATCH 0/1] Add swappiness argument " Michal Hocko
2023-11-30 16:56   ` Johannes Weiner
2023-11-30 18:49     ` Shakeel Butt
2023-11-30 19:47     ` Dan Schatzberg
2023-11-30 20:30       ` Shakeel Butt
2023-11-30 21:37         ` Dan Schatzberg
2023-11-30 21:52           ` Shakeel Butt
2023-12-01  9:33     ` Michal Hocko
2023-12-01 15:49       ` Dan Schatzberg
2023-12-01 17:09       ` Johannes Weiner
2023-12-04 15:23         ` Michal Hocko
2023-12-05 16:19           ` Johannes Weiner
2023-12-07 18:57         ` Michal Koutný
2023-11-30 18:44 ` Shakeel Butt
2023-11-30 18:54   ` Matthew Wilcox
2023-11-30 19:39     ` Johannes Weiner
2023-11-30 19:49   ` Johannes Weiner
2023-11-30 19:50   ` Dan Schatzberg
2023-12-06 16:28 [PATCH V2 " Dan Schatzberg
2023-12-06 16:28 ` [PATCH 1/1] mm: add swapiness= arg " Dan Schatzberg
2023-12-07 19:00   ` Michal Koutný

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=20231130153658.527556-2-schatzberg.dan@gmail.com \
    --to=schatzberg.dan@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=david@redhat.com \
    --cc=findns94@gmail.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=link@vivo.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=muchun.song@linux.dev \
    --cc=peterx@redhat.com \
    --cc=roman.gushchin@linux.dev \
    --cc=shakeelb@google.com \
    --cc=vishal.moola@gmail.com \
    --cc=wangkefeng.wang@huawei.com \
    --cc=willy@infradead.org \
    --cc=ying.huang@intel.com \
    --cc=yosryahmed@google.com \
    /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