[PATCH v3 0/4] mm/vmscan: fix swappiness=max and clean up per-node proactive reclaim

[PATCH v3 0/4] mm/vmscan: fix swappiness=max and clean up per-node proactive reclaim

[Ridong]

From: Ridong Chen <chenridong@xiaomi.com>

Fixes and one cleanup.

Patch 1 fixes "swappiness=max": the anon-only test in get_scan_count()
sat after the "cannot reclaim anon" check, so when no anon was
reclaimable the request fell back to SCAN_FILE and evicted page cache
instead.

Patch 2 fixes reclaim_store() collapsing every error into -EAGAIN, so
callers can no longer tell an invalid argument from a busy interface;
propagate the real error code, matching the memcg path.

Patch 3 drops the now-unused gfp_mask parameter from __node_reclaim().

Patch 4 fixes the same "swappiness=max" issue for MGLRU.

---
v2 -> v3:
 - Update comments and commit message.

v1 -> v2:
 - Collected review tags for patches 1-3
 - Added patch 4 to fix the same issue in the MGLRU path.

Ridong Chen (4):
  mm/vmscan: fix anon-only reclaim evicting file pages when
    swappiness=max
  mm: vmscan: propagate real error code from per-node proactive reclaim
  mm: vmscan: drop unused gfp_mask parameter from __node_reclaim()
  mm/mglru: fix anon-only reclaim evicting file pages when
    swappiness=max

 mm/vmscan.c | 54 +++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 40 insertions(+), 14 deletions(-)

-- 
2.34.1

[PATCH v3 1/4] mm/vmscan: fix anon-only reclaim evicting file pages when swappiness=max

[Ridong]

From: Ridong Chen <chenridong@xiaomi.com>

As Qi mentioned [1], when swappiness=max (SWAPPINESS_ANON_ONLY) is set,
the reclaim logic is expected to reclaim anonymous pages exclusively.
However, due to the current ordering of checks in get_scan_count(),
file pages may still be evicted if can_reclaim_anon_pages() returns
false, which contradicts the semantics of SWAPPINESS_ANON_ONLY.

Reproducer in a cgroup holding 64M of file cache, with no swap configured:

  Before (file cache is wrongly evicted):
    # cat memory.stat
    anon 196608
    file 67178496
    pgscan_proactive 0
    # echo "64M swappiness=max" > memory.reclaim
    # cat memory.stat
    anon 208896
    file 4096                 <- page cache evicted
    pgsteal_proactive 16400
    pgscan_proactive 16400

  After (file cache is left intact):
    # cat memory.stat
    anon 200704
    file 67178496
    pgscan_proactive 0
    # echo "64M swappiness=max" > memory.reclaim
    -bash: echo: write error: Resource temporarily unavailable
    # cat memory.stat
    anon 208896
    file 67178496             <- page cache untouched
    pgsteal_proactive 0
    pgscan_proactive 0

Fix this by bailing out early when SWAPPINESS_ANON_ONLY is set and no
anonymous pages are reclaimable, before falling back to file reclaim.

[1] https://lore.kernel.org/cgroups/7ddf3eee-5fe2-45f7-8614-c8936a039e04@linux.dev/

Fixes: 68a1436bde00 ("mm: add swappiness=max arg to memory.reclaim for only anon reclaim")
Suggested-by: Qi Zheng <qi.zheng@linux.dev>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Reviewed-by: Qi Zheng <qi.zheng@linux.dev>
Reviewed-by: Barry Song <baohua@kernel.org>
Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
---
 mm/vmscan.c | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 35c3bb15ae96..2c689682b952 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2501,6 +2501,23 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 	enum scan_balance scan_balance;
 	enum lru_list lru;
 
+	/*
+	 * Proactive reclaim initiated by userspace for anonymous memory only.
+	 * SWAPPINESS_ANON_ONLY is set only on the proactive reclaim path, so
+	 * warn if it shows up elsewhere. When anon cannot be reclaimed (e.g.
+	 * no swap), bail out instead of falling back to evicting file pages,
+	 * which would violate the anon-only semantics.
+	 */
+	if (swappiness == SWAPPINESS_ANON_ONLY) {
+		WARN_ON_ONCE(!sc->proactive);
+		if (!can_reclaim_anon_pages(memcg, pgdat->node_id, sc)) {
+			memset(nr, 0, sizeof(*nr) * NR_LRU_LISTS);
+			return;
+		}
+		scan_balance = SCAN_ANON;
+		goto out;
+	}
+
 	/* If we have no swap space, do not bother scanning anon folios. */
 	if (!sc->may_swap || !can_reclaim_anon_pages(memcg, pgdat->node_id, sc)) {
 		scan_balance = SCAN_FILE;
@@ -2519,13 +2536,6 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 		goto out;
 	}
 
-	/* Proactive reclaim initiated by userspace for anonymous memory only */
-	if (swappiness == SWAPPINESS_ANON_ONLY) {
-		WARN_ON_ONCE(!sc->proactive);
-		scan_balance = SCAN_ANON;
-		goto out;
-	}
-
 	/*
 	 * Do not apply any pressure balancing cleverness when the
 	 * system is close to OOM, scan both anon and file equally
-- 
2.34.1

[PATCH v3 2/4] mm: vmscan: propagate real error code from per-node proactive reclaim

[Ridong]

From: Ridong Chen <chenridong@xiaomi.com>

It has been observed that per-node proactive reclaim always returns
-EAGAIN when any error occurs. As discussed in the mailing list [1],
the interface should distinguish between cases where no reclaimable
memory is left and where another entity is concurrently using the
same interface. Propagate the real error code, consistent with how
memcg proactive reclaim handles errors.

[1] https://lore.kernel.org/all/20250717235604.2atyx2aobwowpge3@offworld/T/#m3514718be82a31b05726a49da9b61fbfc69a589e

Fixes: b980077899ea ("mm: introduce per-node proactive reclaim interface")
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Reviewed-by: Qi Zheng <qi.zheng@linux.dev>
Reviewed-by: Barry Song <baohua@kernel.org>
Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
---
 mm/vmscan.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 2c689682b952..4b62d6304c49 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -8023,7 +8023,7 @@ static ssize_t reclaim_store(struct device *dev,
 	int ret, nid = dev->id;
 
 	ret = user_proactive_reclaim((char *)buf, NULL, NODE_DATA(nid));
-	return ret ? -EAGAIN : count;
+	return ret ? ret : count;
 }
 
 static DEVICE_ATTR_WO(reclaim);
-- 
2.34.1

[PATCH v3 3/4] mm: vmscan: drop unused gfp_mask parameter from __node_reclaim()

[Ridong]

From: Ridong Chen <chenridong@xiaomi.com>

Commit 57972c78e678 ("mm/vmscan: make __node_reclaim() more generic")
moved the scan_control construction out to the callers and passed the
struct in by pointer.  After that change every use of the gfp mask inside
__node_reclaim() goes through sc->gfp_mask, leaving the gfp_mask parameter
unused. Just remove the dead parameter and update the callers accordingly.
No functional change.

Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Muchun Song <muchun.song@linux.dev>
Reviewed-by: Barry Song <baohua@kernel.org>
Reviewed-by: Qi Zheng <qi.zheng@linux.dev>
Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
---
 mm/vmscan.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4b62d6304c49..859655fcf60d 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -7748,7 +7748,7 @@ static unsigned long node_pagecache_reclaimable(struct pglist_data *pgdat)
 /*
  * Try to free up some pages from this node through reclaim.
  */
-static unsigned long __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask,
+static unsigned long __node_reclaim(struct pglist_data *pgdat,
 				    unsigned long nr_pages,
 				    struct scan_control *sc)
 {
@@ -7840,7 +7840,7 @@ int node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned int order)
 	if (test_and_set_bit_lock(PGDAT_RECLAIM_LOCKED, &pgdat->flags))
 		return NODE_RECLAIM_NOSCAN;
 
-	ret = __node_reclaim(pgdat, gfp_mask, nr_pages, &sc) >= nr_pages;
+	ret = __node_reclaim(pgdat, nr_pages, &sc) >= nr_pages;
 	clear_bit_unlock(PGDAT_RECLAIM_LOCKED, &pgdat->flags);
 
 	if (ret)
@@ -7853,7 +7853,7 @@ int node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask, unsigned int order)
 
 #else
 
-static unsigned long __node_reclaim(struct pglist_data *pgdat, gfp_t gfp_mask,
+static unsigned long __node_reclaim(struct pglist_data *pgdat,
 				    unsigned long nr_pages,
 				    struct scan_control *sc)
 {
@@ -7955,8 +7955,7 @@ int user_proactive_reclaim(char *buf,
 						  &pgdat->flags))
 				return -EBUSY;
 
-			reclaimed = __node_reclaim(pgdat, gfp_mask,
-						   batch_size, &sc);
+			reclaimed = __node_reclaim(pgdat, batch_size, &sc);
 			clear_bit_unlock(PGDAT_RECLAIM_LOCKED, &pgdat->flags);
 		}
 
-- 
2.34.1

[PATCH v3 4/4] mm/mglru: fix anon-only reclaim evicting file pages when swappiness=max

[Ridong]

From: Ridong Chen <chenridong@xiaomi.com>

The previous patch fixed this issue for the traditional LRU. The same
problem exists in MGLRU [1]: when swappiness=max (SWAPPINESS_ANON_ONLY)
is set, reclaim is expected to evict anonymous pages exclusively, but
file pages can still be reclaimed when anonymous pages cannot be
reclaimed (e.g. no swap and no demotion target).

With SWAPPINESS_ANON_ONLY, get_type_to_scan() always returns
LRU_GEN_ANON and for_each_evictable_type() only iterates the anon type.
But if get_nr_to_scan() does not bail out, evict_folios() still runs and
isolate_folios() falls back to scanning file pages in the !scanned case:

  shrink_one
    try_to_shrink_lruvec
      get_swappiness   // returns SWAPPINESS_ANON_ONLY for swappiness=max
      get_nr_to_scan
      evict_folios
        isolate_folios // falls back to file type when !scanned

This wrongly evicts file pages, violating the anon-only semantics.

Fix it the same way as the traditional LRU: keep returning
SWAPPINESS_ANON_ONLY in get_swappiness(), and return 0 from
get_nr_to_scan() when SWAPPINESS_ANON_ONLY is set but anon pages cannot
be reclaimed. This protects file pages from being reclaimed, and as a
side effect avoids the useless scan work.

The test result:
Before fix:

  # cat /sys/kernel/mm/lru_gen/enabled
  0x0007
  # cat memory.stat
  anon 204800
  file 67108864
  ...
  pgsteal_proactive 0
  pgscan_proactive 0

  # echo "64M swappiness=max" > memory.reclaim
  # cat memory.stat
  anon 208896
  file 0
  ...
  pgsteal_proactive 16384
  pgscan_proactive 16384

After fix:

  # cat memory.stat
  anon 188416
  file 67215360
  kernel 1970176
  ...
  pgsteal_proactive 0
  pgscan_proactive 0

  # echo "64M swappiness=max" > memory.reclaim
  -bash: echo: write error: Resource temporarily unavailable
  # cat memory.stat
  anon 204800
  file 67215360
  ...
  pgsteal_proactive 0
  pgscan_proactive 0

[1] https://sashiko.dev/#/patchset/20260717113300.214717-1-ridong.chen@linux.dev

Fixes: 68a1436bde00 ("mm: add swappiness=max arg to memory.reclaim for only anon reclaim")
Acked-by: Qi Zheng <qi.zheng@linux.dev>
Signed-off-by: Ridong Chen <chenridong@xiaomi.com>
---
 mm/vmscan.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 859655fcf60d..20cb93fd3da8 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2708,6 +2708,10 @@ static int get_swappiness(struct lruvec *lruvec, struct scan_control *sc)
 {
 	struct mem_cgroup *memcg = lruvec_memcg(lruvec);
 	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
+	int swappiness = sc_swappiness(sc, memcg);
+
+	if (swappiness == SWAPPINESS_ANON_ONLY)
+		return swappiness;
 
 	if (!sc->may_swap)
 		return 0;
@@ -2716,7 +2720,7 @@ static int get_swappiness(struct lruvec *lruvec, struct scan_control *sc)
 	    mem_cgroup_get_nr_swap_pages(memcg) < MIN_LRU_BATCH)
 		return 0;
 
-	return sc_swappiness(sc, memcg);
+	return swappiness;
 }
 
 static int get_nr_gens(struct lruvec *lruvec, int type)
@@ -4921,6 +4925,19 @@ static long get_nr_to_scan(struct lruvec *lruvec, struct scan_control *sc,
 			   struct mem_cgroup *memcg, int swappiness)
 {
 	unsigned long nr_to_scan, evictable;
+	struct pglist_data *pgdat = lruvec_pgdat(lruvec);
+	/*
+	 * Proactive reclaim initiated by userspace for anonymous memory only.
+	 * SWAPPINESS_ANON_ONLY is set only on the proactive reclaim path, so
+	 * warn if it shows up elsewhere. When anon cannot be reclaimed (e.g.
+	 * no swap), return 0 instead of falling through and evicting file
+	 * pages, which would violate the anon-only semantics.
+	 */
+	if (swappiness == SWAPPINESS_ANON_ONLY &&
+	    !can_reclaim_anon_pages(memcg, pgdat->node_id, sc)) {
+		WARN_ON_ONCE(!sc->proactive);
+		return 0;
+	}
 
 	evictable = lruvec_evictable_size(lruvec, swappiness);
 
-- 
2.34.1