[PATCH v3 0/7] mm, swap: remove swap slot cache

[PATCH v3 0/7] mm, swap: remove swap slot cache

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

Slot cache was initially introduced by commit 67afa38e012e ("mm/swap:
add cache for swap slots allocation") to reduce the lock contention
of si->lock.

Previous series "mm, swap: rework of swap allocator locks" [1] removed
swap slot cache for freeing path as freeing path no longer touches
si->lock in most cased. Allocation path also have slight to none
contention on si->lock since that series, but slot cache still helps
to reduce other overheads, like counters and the plist.

This series removes the slot cache from allocation path too, by using
the cluster as allocation fast path and also reduce other overheads.

Now slot cache is completely gone, the code is much simplified without
obvious feature or performance change, also clean up related workaround.
Also this should avoid other potential issues, e.g. the long pinning
of swap slots: swap slot cache pins swap slots with HAS_CACHE, causing
reclaim or allocation fail to use these slots on scanning.

The only behavior change is the swap device allocation rotation
mechanism, as explained in the patch "mm, swap: use percpu cluster
as allocation fast path".

Test results are looking good after deleting the swap slot cache:

- vm-scalability with: `usemem --init-time -O -y -x -R -31 1G`,
12G memory cgroup using simulated pmem as SWAP (32G pmem, 32 CPUs),
16 test runs for each case, measuring the total throughput:

                      Before (KB/s) (stdev)  After (KB/s) (stdev)
Random (4K):          424907.60 (24410.78)   414745.92  (34554.78)
Random (64K):         163308.82 (11635.72)   167314.50  (18434.99)
Sequential (4K, !-R): 6150056.79 (103205.90) 6321469.06 (115878.16)

- Build linux kernel with make -j96, using 4K folio with 1.5G memory
cgroup limit and 64K folio with 2G memory cgroup limit, on top of tmpfs,
12 test runs, measuring the system time:

                  Before (s) (stdev)  After (s) (stdev)
make -j96 (4K):   6445.69 (61.95)     6408.80 (69.46)
make -j96 (64K):  6841.71 (409.04)    6437.99 (435.55)

The performance is unchanged, slightly better in some cases.

[1] https://lore.kernel.org/linux-mm/20250113175732.48099-1-ryncsn@gmail.com/

---

V2: https://lore.kernel.org/linux-mm/20250224180212.22802-1-ryncsn@gmail.com/
Updates from V2:
- Make folio_alloc_swap() inline to fix build error [Stephen Rothwell]
- Flush the global percpu cluster cache on swapoff to prevent new swapon
  devices using the old invalid values. Based on:
  https://lore.kernel.org/linux-mm/CAMgjq7AkRmb5ote-VZErM_2UdEC575j9WcrstcQOypEb+T-DLA@mail.gmail.com/
- Minor update for patch 5/7: in slow path also try the local cluster
  first to avoid fragmentation. It's a intermediate patch change for easier
  testing if someone run into a bisect, the final code after the whole
  series applies is not changed.
- Need to call mem_cgroup_try_charge_swap even if swap allocation failed
  for cgroup events.
- Collect reviews and minor improvements [Baoquan He].

V1: https://lore.kernel.org/linux-mm/20250214175709.76029-1-ryncsn@gmail.com/
Updates from V1:
- Check the cluster with cluster_is_usable and cluster_is_empty in
  fast path too, improve performance and avoid fragmentation.
- Fix a build warning and error for !SWAP build reported by test bot.
- Global cluster array also record device for each order [Baoquan He]
- Adjust of comments and function name [Baoquan He]
- Collect Review-by [Baoquan He]
- Minor function style improvement [Matthew Wilcox]

Kairui Song (7):
  mm, swap: avoid reclaiming irrelevant swap cache
  mm, swap: drop the flag TTRS_DIRECT
  mm, swap: avoid redundant swap device pinning
  mm, swap: don't update the counter up-front
  mm, swap: use percpu cluster as allocation fast path
  mm, swap: remove swap slot cache
  mm, swap: simplify folio swap allocation

 include/linux/swap.h       |  22 +--
 include/linux/swap_slots.h |  28 ---
 mm/Makefile                |   2 +-
 mm/shmem.c                 |  21 +--
 mm/swap.h                  |   6 -
 mm/swap_slots.c            | 295 ------------------------------
 mm/swap_state.c            |  79 +--------
 mm/swapfile.c              | 355 ++++++++++++++++++++-----------------
 mm/vmscan.c                |  16 +-
 mm/zswap.c                 |   6 +
 10 files changed, 232 insertions(+), 598 deletions(-)
 delete mode 100644 include/linux/swap_slots.h
 delete mode 100644 mm/swap_slots.c

-- 
2.48.1

[PATCH v3 1/7] mm, swap: avoid reclaiming irrelevant swap cache

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

Swap allocator will do swap cache reclaim to recycle HAS_CACHE slots for
allocation.  It initiates the reclaim from the offset to be reclaimed and
looks up the corresponding folio.  The lookup process is lockless, so it's
possible the folio will be removed from the swap cache and given a
different swap entry before the reclaim locks the folio.  If it happens,
the reclaim will end up reclaiming an irrelevant folio, and return wrong
return value.

This shouldn't cause any problem with correctness or stability, but it is
indeed confusing and unexpected, and will increase fragmentation, decrease
performance.

Fix this by checking whether the folio is still pointing to the offset the
allocator want to reclaim before reclaiming it.

Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
---
 mm/swapfile.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index a7f60006c52c..5618cd1c4b03 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -210,6 +210,7 @@ static int __try_to_reclaim_swap(struct swap_info_struct *si,
 	int ret, nr_pages;
 	bool need_reclaim;
 
+again:
 	folio = filemap_get_folio(address_space, swap_cache_index(entry));
 	if (IS_ERR(folio))
 		return 0;
@@ -227,8 +228,16 @@ static int __try_to_reclaim_swap(struct swap_info_struct *si,
 	if (!folio_trylock(folio))
 		goto out;
 
-	/* offset could point to the middle of a large folio */
+	/*
+	 * Offset could point to the middle of a large folio, or folio
+	 * may no longer point to the expected offset before it's locked.
+	 */
 	entry = folio->swap;
+	if (offset < swp_offset(entry) || offset >= swp_offset(entry) + nr_pages) {
+		folio_unlock(folio);
+		folio_put(folio);
+		goto again;
+	}
 	offset = swp_offset(entry);
 
 	need_reclaim = ((flags & TTRS_ANYWAY) ||
-- 
2.48.1

[PATCH v3 2/7] mm, swap: drop the flag TTRS_DIRECT

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

This flag exists temporarily to allow the allocator to bypass the slot
cache during freeing, so reclaiming one slot will free the slot
immediately.

But now we have already removed slot cache usage on freeing, so this flag
has no effect now.

Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
---
 mm/swapfile.c | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 5618cd1c4b03..6f2de59c6355 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -158,8 +158,6 @@ static long swap_usage_in_pages(struct swap_info_struct *si)
 #define TTRS_UNMAPPED		0x2
 /* Reclaim the swap entry if swap is getting full */
 #define TTRS_FULL		0x4
-/* Reclaim directly, bypass the slot cache and don't touch device lock */
-#define TTRS_DIRECT		0x8
 
 static bool swap_only_has_cache(struct swap_info_struct *si,
 			      unsigned long offset, int nr_pages)
@@ -257,23 +255,8 @@ static int __try_to_reclaim_swap(struct swap_info_struct *si,
 	if (!need_reclaim)
 		goto out_unlock;
 
-	if (!(flags & TTRS_DIRECT)) {
-		/* Free through slot cache */
-		delete_from_swap_cache(folio);
-		folio_set_dirty(folio);
-		ret = nr_pages;
-		goto out_unlock;
-	}
-
-	xa_lock_irq(&address_space->i_pages);
-	__delete_from_swap_cache(folio, entry, NULL);
-	xa_unlock_irq(&address_space->i_pages);
-	folio_ref_sub(folio, nr_pages);
+	delete_from_swap_cache(folio);
 	folio_set_dirty(folio);
-
-	ci = lock_cluster(si, offset);
-	swap_entry_range_free(si, ci, entry, nr_pages);
-	unlock_cluster(ci);
 	ret = nr_pages;
 out_unlock:
 	folio_unlock(folio);
@@ -697,7 +680,7 @@ static bool cluster_reclaim_range(struct swap_info_struct *si,
 			offset++;
 			break;
 		case SWAP_HAS_CACHE:
-			nr_reclaim = __try_to_reclaim_swap(si, offset, TTRS_ANYWAY | TTRS_DIRECT);
+			nr_reclaim = __try_to_reclaim_swap(si, offset, TTRS_ANYWAY);
 			if (nr_reclaim > 0)
 				offset += nr_reclaim;
 			else
@@ -849,7 +832,7 @@ static void swap_reclaim_full_clusters(struct swap_info_struct *si, bool force)
 			if (READ_ONCE(map[offset]) == SWAP_HAS_CACHE) {
 				spin_unlock(&ci->lock);
 				nr_reclaim = __try_to_reclaim_swap(si, offset,
-								   TTRS_ANYWAY | TTRS_DIRECT);
+								   TTRS_ANYWAY);
 				spin_lock(&ci->lock);
 				if (nr_reclaim) {
 					offset += abs(nr_reclaim);
-- 
2.48.1

[PATCH v3 3/7] mm, swap: avoid redundant swap device pinning

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

Currently __read_swap_cache_async() has get/put_swap_device() calls to
increase/decrease a swap device reference to prevent swapoff.  While some
of its callers have already held the swap device reference, e.g in
do_swap_page() and shmem_swapin_folio() where __read_swap_cache_async()
will finally called.  Now there are only two callers not holding a swap
device reference, so make them hold a reference instead.  And drop the
get/put_swap_device calls in __read_swap_cache_async.  This should reduce
the overhead for swap in during page fault slightly.

Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
---
 mm/swap_state.c | 14 ++++++++------
 mm/zswap.c      |  6 ++++++
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/mm/swap_state.c b/mm/swap_state.c
index a54b035d6a6c..50840a2887a5 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -426,17 +426,13 @@ struct folio *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
 		struct mempolicy *mpol, pgoff_t ilx, bool *new_page_allocated,
 		bool skip_if_exists)
 {
-	struct swap_info_struct *si;
+	struct swap_info_struct *si = swp_swap_info(entry);
 	struct folio *folio;
 	struct folio *new_folio = NULL;
 	struct folio *result = NULL;
 	void *shadow = NULL;
 
 	*new_page_allocated = false;
-	si = get_swap_device(entry);
-	if (!si)
-		return NULL;
-
 	for (;;) {
 		int err;
 		/*
@@ -532,7 +528,6 @@ struct folio *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
 	put_swap_folio(new_folio, entry);
 	folio_unlock(new_folio);
 put_and_return:
-	put_swap_device(si);
 	if (!(*new_page_allocated) && new_folio)
 		folio_put(new_folio);
 	return result;
@@ -552,11 +547,16 @@ struct folio *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
 		struct vm_area_struct *vma, unsigned long addr,
 		struct swap_iocb **plug)
 {
+	struct swap_info_struct *si;
 	bool page_allocated;
 	struct mempolicy *mpol;
 	pgoff_t ilx;
 	struct folio *folio;
 
+	si = get_swap_device(entry);
+	if (!si)
+		return NULL;
+
 	mpol = get_vma_policy(vma, addr, 0, &ilx);
 	folio = __read_swap_cache_async(entry, gfp_mask, mpol, ilx,
 					&page_allocated, false);
@@ -564,6 +564,8 @@ struct folio *read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
 
 	if (page_allocated)
 		swap_read_folio(folio, plug);
+
+	put_swap_device(si);
 	return folio;
 }
 
diff --git a/mm/zswap.c b/mm/zswap.c
index 7d8d684e54d4..c470073c17cc 100644
--- a/mm/zswap.c
+++ b/mm/zswap.c
@@ -1055,15 +1055,21 @@ static int zswap_writeback_entry(struct zswap_entry *entry,
 	struct folio *folio;
 	struct mempolicy *mpol;
 	bool folio_was_allocated;
+	struct swap_info_struct *si;
 	struct writeback_control wbc = {
 		.sync_mode = WB_SYNC_NONE,
 	};
 	int ret = 0;
 
 	/* try to allocate swap cache folio */
+	si = get_swap_device(swpentry);
+	if (!si)
+		return -EEXIST;
+
 	mpol = get_task_policy(current);
 	folio = __read_swap_cache_async(swpentry, GFP_KERNEL, mpol,
 				NO_INTERLEAVE_INDEX, &folio_was_allocated, true);
+	put_swap_device(si);
 	if (!folio)
 		return -ENOMEM;
 
-- 
2.48.1

[PATCH v3 4/7] mm, swap: don't update the counter up-front

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

The counter update before allocation design was useful to avoid
unnecessary scan when device is full, so it will abort early if the
counter indicates the device is full.  But that is an uncommon case, and
now scanning of a full device is very fast, so the up-front update is not
helpful any more.

Remove it and simplify the slot allocation logic.

Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
---
 mm/swapfile.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 6f2de59c6355..db836670c334 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1201,22 +1201,10 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 	int order = swap_entry_order(entry_order);
 	unsigned long size = 1 << order;
 	struct swap_info_struct *si, *next;
-	long avail_pgs;
 	int n_ret = 0;
 	int node;
 
 	spin_lock(&swap_avail_lock);
-
-	avail_pgs = atomic_long_read(&nr_swap_pages) / size;
-	if (avail_pgs <= 0) {
-		spin_unlock(&swap_avail_lock);
-		goto noswap;
-	}
-
-	n_goal = min3((long)n_goal, (long)SWAP_BATCH, avail_pgs);
-
-	atomic_long_sub(n_goal * size, &nr_swap_pages);
-
 start_over:
 	node = numa_node_id();
 	plist_for_each_entry_safe(si, next, &swap_avail_heads[node], avail_lists[node]) {
@@ -1250,10 +1238,8 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 	spin_unlock(&swap_avail_lock);
 
 check_out:
-	if (n_ret < n_goal)
-		atomic_long_add((long)(n_goal - n_ret) * size,
-				&nr_swap_pages);
-noswap:
+	atomic_long_sub(n_ret * size, &nr_swap_pages);
+
 	return n_ret;
 }
 
-- 
2.48.1

[PATCH v3 5/7] mm, swap: use percpu cluster as allocation fast path

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

Current allocation workflow first traverses the plist with a global lock
held, after choosing a device, it uses the percpu cluster on that swap
device.  This commit moves the percpu cluster variable out of being tied
to individual swap devices, making it a global percpu variable, and will
be used directly for allocation as a fast path.

The global percpu cluster variable will never point to a HDD device, and
allocations on a HDD device are still globally serialized.

This improves the allocator performance and prepares for removal of the
slot cache in later commits.  There shouldn't be much observable behavior
change, except one thing: this changes how swap device allocation rotation
works.

Currently, each allocation will rotate the plist, and because of the
existence of slot cache (one order 0 allocation usually returns 64
entries), swap devices of the same priority are rotated for every 64 order
0 entries consumed.  High order allocations are different, they will
bypass the slot cache, and so swap device is rotated for every 16K, 32K,
or up to 2M allocation.

The rotation rule was never clearly defined or documented, it was changed
several times without mentioning.

After this commit, and once slot cache is gone in later commits, swap
device rotation will happen for every consumed cluster.  Ideally non-HDD
devices will be rotated if 2M space has been consumed for each order.
Fragmented clusters will rotate the device faster, which seems OK.  HDD
devices is rotated for every allocation regardless of the allocation
order, which should be OK too and trivial.

This commit also slightly changes allocation behaviour for slot cache.
The new added cluster allocation fast path may allocate entries from
different device to the slot cache, this is not observable from user
space, only impact performance very slightly, and slot cache will be just
gone in next commit, so this can be ignored.

Signed-off-by: Kairui Song <kasong@tencent.com>
---
 include/linux/swap.h |  11 ++-
 mm/swapfile.c        | 158 ++++++++++++++++++++++++++++++++-----------
 2 files changed, 121 insertions(+), 48 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 2fe91c293636..374bffc87427 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -284,12 +284,10 @@ enum swap_cluster_flags {
 #endif
 
 /*
- * We assign a cluster to each CPU, so each CPU can allocate swap entry from
- * its own cluster and swapout sequentially. The purpose is to optimize swapout
- * throughput.
+ * We keep using same cluster for rotational device so IO will be sequential.
+ * The purpose is to optimize SWAP throughput on these device.
  */
-struct percpu_cluster {
-	local_lock_t lock; /* Protect the percpu_cluster above */
+struct swap_sequential_cluster {
 	unsigned int next[SWAP_NR_ORDERS]; /* Likely next allocation offset */
 };
 
@@ -315,8 +313,7 @@ struct swap_info_struct {
 	atomic_long_t frag_cluster_nr[SWAP_NR_ORDERS];
 	unsigned int pages;		/* total of usable pages of swap */
 	atomic_long_t inuse_pages;	/* number of those currently in use */
-	struct percpu_cluster __percpu *percpu_cluster; /* per cpu's swap location */
-	struct percpu_cluster *global_cluster; /* Use one global cluster for rotating device */
+	struct swap_sequential_cluster *global_cluster; /* Use one global cluster for rotating device */
 	spinlock_t global_cluster_lock;	/* Serialize usage of global cluster */
 	struct rb_root swap_extent_root;/* root of the swap extent rbtree */
 	struct block_device *bdev;	/* swap device or bdev of swap file */
diff --git a/mm/swapfile.c b/mm/swapfile.c
index db836670c334..8b296c4c636b 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -116,6 +116,18 @@ static atomic_t proc_poll_event = ATOMIC_INIT(0);
 
 atomic_t nr_rotate_swap = ATOMIC_INIT(0);
 
+struct percpu_swap_cluster {
+	struct swap_info_struct *si[SWAP_NR_ORDERS];
+	unsigned long offset[SWAP_NR_ORDERS];
+	local_lock_t lock;
+};
+
+static DEFINE_PER_CPU(struct percpu_swap_cluster, percpu_swap_cluster) = {
+	.si = { NULL },
+	.offset = { SWAP_ENTRY_INVALID },
+	.lock = INIT_LOCAL_LOCK(),
+};
+
 static struct swap_info_struct *swap_type_to_swap_info(int type)
 {
 	if (type >= MAX_SWAPFILES)
@@ -539,7 +551,7 @@ static bool swap_do_scheduled_discard(struct swap_info_struct *si)
 		ci = list_first_entry(&si->discard_clusters, struct swap_cluster_info, list);
 		/*
 		 * Delete the cluster from list to prepare for discard, but keep
-		 * the CLUSTER_FLAG_DISCARD flag, there could be percpu_cluster
+		 * the CLUSTER_FLAG_DISCARD flag, percpu_swap_cluster could be
 		 * pointing to it, or ran into by relocate_cluster.
 		 */
 		list_del(&ci->list);
@@ -805,10 +817,12 @@ static unsigned int alloc_swap_scan_cluster(struct swap_info_struct *si,
 out:
 	relocate_cluster(si, ci);
 	unlock_cluster(ci);
-	if (si->flags & SWP_SOLIDSTATE)
-		__this_cpu_write(si->percpu_cluster->next[order], next);
-	else
+	if (si->flags & SWP_SOLIDSTATE) {
+		this_cpu_write(percpu_swap_cluster.offset[order], next);
+		this_cpu_write(percpu_swap_cluster.si[order], si);
+	} else {
 		si->global_cluster->next[order] = next;
+	}
 	return found;
 }
 
@@ -862,20 +876,18 @@ static void swap_reclaim_work(struct work_struct *work)
 }
 
 /*
- * Try to get swap entries with specified order from current cpu's swap entry
- * pool (a cluster). This might involve allocating a new cluster for current CPU
- * too.
+ * Try to allocate swap entries with specified order and try set a new
+ * cluster for current CPU too.
  */
 static unsigned long cluster_alloc_swap_entry(struct swap_info_struct *si, int order,
 					      unsigned char usage)
 {
 	struct swap_cluster_info *ci;
-	unsigned int offset, found = 0;
+	unsigned int offset = SWAP_ENTRY_INVALID, found = SWAP_ENTRY_INVALID;
 
 	if (si->flags & SWP_SOLIDSTATE) {
-		/* Fast path using per CPU cluster */
-		local_lock(&si->percpu_cluster->lock);
-		offset = __this_cpu_read(si->percpu_cluster->next[order]);
+		if (si == this_cpu_read(percpu_swap_cluster.si[order]))
+			offset = this_cpu_read(percpu_swap_cluster.offset[order]);
 	} else {
 		/* Serialize HDD SWAP allocation for each device. */
 		spin_lock(&si->global_cluster_lock);
@@ -973,9 +985,7 @@ static unsigned long cluster_alloc_swap_entry(struct swap_info_struct *si, int o
 		}
 	}
 done:
-	if (si->flags & SWP_SOLIDSTATE)
-		local_unlock(&si->percpu_cluster->lock);
-	else
+	if (!(si->flags & SWP_SOLIDSTATE))
 		spin_unlock(&si->global_cluster_lock);
 	return found;
 }
@@ -1196,6 +1206,51 @@ static bool get_swap_device_info(struct swap_info_struct *si)
 	return true;
 }
 
+/*
+ * Fast path try to get swap entries with specified order from current
+ * CPU's swap entry pool (a cluster).
+ */
+static int swap_alloc_fast(swp_entry_t entries[],
+			   unsigned char usage,
+			   int order, int n_goal)
+{
+	struct swap_cluster_info *ci;
+	struct swap_info_struct *si;
+	unsigned int offset, found;
+	int n_ret = 0;
+
+	n_goal = min(n_goal, SWAP_BATCH);
+
+	/*
+	 * Once allocated, swap_info_struct will never be completely freed,
+	 * so checking it's liveness by get_swap_device_info is enough.
+	 */
+	si = this_cpu_read(percpu_swap_cluster.si[order]);
+	offset = this_cpu_read(percpu_swap_cluster.offset[order]);
+	if (!si || !offset || !get_swap_device_info(si))
+		return 0;
+
+	while (offset) {
+		ci = lock_cluster(si, offset);
+		if (!cluster_is_usable(ci, order)) {
+			unlock_cluster(ci);
+			break;
+		}
+		if (cluster_is_empty(ci))
+			offset = cluster_offset(si, ci);
+		found = alloc_swap_scan_cluster(si, ci, offset, order, usage);
+		if (!found)
+			break;
+		entries[n_ret++] = swp_entry(si->type, found);
+		if (n_ret == n_goal)
+			break;
+		offset = this_cpu_read(percpu_swap_cluster.offset[order]);
+	}
+
+	put_swap_device(si);
+	return n_ret;
+}
+
 int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 {
 	int order = swap_entry_order(entry_order);
@@ -1204,19 +1259,36 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 	int n_ret = 0;
 	int node;
 
+	/* Fast path using percpu cluster */
+	local_lock(&percpu_swap_cluster.lock);
+	n_ret = swap_alloc_fast(swp_entries,
+				SWAP_HAS_CACHE,
+				order, n_goal);
+	if (n_ret == n_goal)
+		goto out;
+
+	n_goal = min_t(int, n_goal - n_ret, SWAP_BATCH);
+	/* Rotate the device and switch to a new cluster */
 	spin_lock(&swap_avail_lock);
 start_over:
 	node = numa_node_id();
 	plist_for_each_entry_safe(si, next, &swap_avail_heads[node], avail_lists[node]) {
-		/* requeue si to after same-priority siblings */
 		plist_requeue(&si->avail_lists[node], &swap_avail_heads[node]);
 		spin_unlock(&swap_avail_lock);
 		if (get_swap_device_info(si)) {
-			n_ret = scan_swap_map_slots(si, SWAP_HAS_CACHE,
-					n_goal, swp_entries, order);
+			/*
+			 * For order 0 allocation, try best to fill the request
+			 * as it's used by slot cache.
+			 *
+			 * For mTHP allocation, it always have n_goal == 1,
+			 * and falling a mTHP swapin will just make the caller
+			 * fallback to order 0 allocation, so just bail out.
+			 */
+			n_ret += scan_swap_map_slots(si, SWAP_HAS_CACHE, n_goal,
+					swp_entries + n_ret, order);
 			put_swap_device(si);
 			if (n_ret || size > 1)
-				goto check_out;
+				goto out;
 		}
 
 		spin_lock(&swap_avail_lock);
@@ -1234,12 +1306,10 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 		if (plist_node_empty(&next->avail_lists[node]))
 			goto start_over;
 	}
-
 	spin_unlock(&swap_avail_lock);
-
-check_out:
+out:
+	local_unlock(&percpu_swap_cluster.lock);
 	atomic_long_sub(n_ret * size, &nr_swap_pages);
-
 	return n_ret;
 }
 
@@ -2597,6 +2667,28 @@ static void wait_for_allocation(struct swap_info_struct *si)
 	}
 }
 
+/*
+ * Called after swap device's reference count is dead, so
+ * neither scan nor allocation will use it.
+ */
+static void flush_percpu_swap_cluster(struct swap_info_struct *si)
+{
+	int cpu, i;
+	struct swap_info_struct **pcp_si;
+
+	for_each_possible_cpu(cpu) {
+		pcp_si = per_cpu_ptr(percpu_swap_cluster.si, cpu);
+		/*
+		 * Invalidate the percpu swap cluster cache, si->users
+		 * is dead, so no new user will point to it, just flush
+		 * any existing user.
+		 */
+		for (i = 0; i < SWAP_NR_ORDERS; i++)
+			cmpxchg(&pcp_si[i], si, NULL);
+	}
+}
+
+
 SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 {
 	struct swap_info_struct *p = NULL;
@@ -2698,6 +2790,7 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 
 	flush_work(&p->discard_work);
 	flush_work(&p->reclaim_work);
+	flush_percpu_swap_cluster(p);
 
 	destroy_swap_extents(p);
 	if (p->flags & SWP_CONTINUED)
@@ -2725,8 +2818,6 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 	arch_swap_invalidate_area(p->type);
 	zswap_swapoff(p->type);
 	mutex_unlock(&swapon_mutex);
-	free_percpu(p->percpu_cluster);
-	p->percpu_cluster = NULL;
 	kfree(p->global_cluster);
 	p->global_cluster = NULL;
 	vfree(swap_map);
@@ -3125,7 +3216,7 @@ static struct swap_cluster_info *setup_clusters(struct swap_info_struct *si,
 	unsigned long nr_clusters = DIV_ROUND_UP(maxpages, SWAPFILE_CLUSTER);
 	struct swap_cluster_info *cluster_info;
 	unsigned long i, j, idx;
-	int cpu, err = -ENOMEM;
+	int err = -ENOMEM;
 
 	cluster_info = kvcalloc(nr_clusters, sizeof(*cluster_info), GFP_KERNEL);
 	if (!cluster_info)
@@ -3134,20 +3225,7 @@ static struct swap_cluster_info *setup_clusters(struct swap_info_struct *si,
 	for (i = 0; i < nr_clusters; i++)
 		spin_lock_init(&cluster_info[i].lock);
 
-	if (si->flags & SWP_SOLIDSTATE) {
-		si->percpu_cluster = alloc_percpu(struct percpu_cluster);
-		if (!si->percpu_cluster)
-			goto err_free;
-
-		for_each_possible_cpu(cpu) {
-			struct percpu_cluster *cluster;
-
-			cluster = per_cpu_ptr(si->percpu_cluster, cpu);
-			for (i = 0; i < SWAP_NR_ORDERS; i++)
-				cluster->next[i] = SWAP_ENTRY_INVALID;
-			local_lock_init(&cluster->lock);
-		}
-	} else {
+	if (!(si->flags & SWP_SOLIDSTATE)) {
 		si->global_cluster = kmalloc(sizeof(*si->global_cluster),
 				     GFP_KERNEL);
 		if (!si->global_cluster)
@@ -3424,8 +3502,6 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
 bad_swap_unlock_inode:
 	inode_unlock(inode);
 bad_swap:
-	free_percpu(si->percpu_cluster);
-	si->percpu_cluster = NULL;
 	kfree(si->global_cluster);
 	si->global_cluster = NULL;
 	inode = NULL;
-- 
2.48.1

[PATCH v3 6/7] mm, swap: remove swap slot cache

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

Slot cache is no longer needed now, removing it and all related code.

- vm-scalability with: `usemem --init-time -O -y -x -R -31 1G`,
12G memory cgroup using simulated pmem as SWAP (32G pmem, 32 CPUs),
16 test runs for each case, measuring the total throughput:

                      Before (KB/s) (stdev)  After (KB/s) (stdev)
Random (4K):          424907.60 (24410.78)   414745.92  (34554.78)
Random (64K):         163308.82 (11635.72)   167314.50  (18434.99)
Sequential (4K, !-R): 6150056.79 (103205.90) 6321469.06 (115878.16)

The performance changes are below noise level.

- Build linux kernel with make -j96, using 4K folio with 1.5G memory
cgroup limit and 64K folio with 2G memory cgroup limit, on top of tmpfs,
12 test runs, measuring the system time:

                  Before (s) (stdev)  After (s) (stdev)
make -j96 (4K):   6445.69 (61.95)     6408.80 (69.46)
make -j96 (64K):  6841.71 (409.04)    6437.99 (435.55)

Similar to above, 64k mTHP case showed a slight improvement.

Signed-off-by: Kairui Song <kasong@tencent.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
---
 include/linux/swap.h       |   3 -
 include/linux/swap_slots.h |  28 ----
 mm/Makefile                |   2 +-
 mm/swap_slots.c            | 295 -------------------------------------
 mm/swap_state.c            |   8 +-
 mm/swapfile.c              | 194 ++++++++----------------
 6 files changed, 67 insertions(+), 463 deletions(-)
 delete mode 100644 include/linux/swap_slots.h
 delete mode 100644 mm/swap_slots.c

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 374bffc87427..c5856dcc263a 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -465,7 +465,6 @@ void free_pages_and_swap_cache(struct encoded_page **, int);
 extern atomic_long_t nr_swap_pages;
 extern long total_swap_pages;
 extern atomic_t nr_rotate_swap;
-extern bool has_usable_swap(void);
 
 /* Swap 50% full? Release swapcache more aggressively.. */
 static inline bool vm_swap_full(void)
@@ -483,13 +482,11 @@ swp_entry_t folio_alloc_swap(struct folio *folio);
 bool folio_free_swap(struct folio *folio);
 void put_swap_folio(struct folio *folio, swp_entry_t entry);
 extern swp_entry_t get_swap_page_of_type(int);
-extern int get_swap_pages(int n, swp_entry_t swp_entries[], int order);
 extern int add_swap_count_continuation(swp_entry_t, gfp_t);
 extern void swap_shmem_alloc(swp_entry_t, int);
 extern int swap_duplicate(swp_entry_t);
 extern int swapcache_prepare(swp_entry_t entry, int nr);
 extern void swap_free_nr(swp_entry_t entry, int nr_pages);
-extern void swapcache_free_entries(swp_entry_t *entries, int n);
 extern void free_swap_and_cache_nr(swp_entry_t entry, int nr);
 int swap_type_of(dev_t device, sector_t offset);
 int find_first_swap(dev_t *device);
diff --git a/include/linux/swap_slots.h b/include/linux/swap_slots.h
deleted file mode 100644
index 840aec3523b2..000000000000
--- a/include/linux/swap_slots.h
+++ /dev/null
@@ -1,28 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _LINUX_SWAP_SLOTS_H
-#define _LINUX_SWAP_SLOTS_H
-
-#include <linux/swap.h>
-#include <linux/spinlock.h>
-#include <linux/mutex.h>
-
-#define SWAP_SLOTS_CACHE_SIZE			SWAP_BATCH
-#define THRESHOLD_ACTIVATE_SWAP_SLOTS_CACHE	(5*SWAP_SLOTS_CACHE_SIZE)
-#define THRESHOLD_DEACTIVATE_SWAP_SLOTS_CACHE	(2*SWAP_SLOTS_CACHE_SIZE)
-
-struct swap_slots_cache {
-	bool		lock_initialized;
-	struct mutex	alloc_lock; /* protects slots, nr, cur */
-	swp_entry_t	*slots;
-	int		nr;
-	int		cur;
-	int		n_ret;
-};
-
-void disable_swap_slots_cache_lock(void);
-void reenable_swap_slots_cache_unlock(void);
-void enable_swap_slots_cache(void);
-
-extern bool swap_slot_cache_enabled;
-
-#endif /* _LINUX_SWAP_SLOTS_H */
diff --git a/mm/Makefile b/mm/Makefile
index 4510a9869e77..e7f6bbf8ae5f 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -75,7 +75,7 @@ ifdef CONFIG_MMU
 	obj-$(CONFIG_ADVISE_SYSCALLS)	+= madvise.o
 endif
 
-obj-$(CONFIG_SWAP)	+= page_io.o swap_state.o swapfile.o swap_slots.o
+obj-$(CONFIG_SWAP)	+= page_io.o swap_state.o swapfile.o
 obj-$(CONFIG_ZSWAP)	+= zswap.o
 obj-$(CONFIG_HAS_DMA)	+= dmapool.o
 obj-$(CONFIG_HUGETLBFS)	+= hugetlb.o
diff --git a/mm/swap_slots.c b/mm/swap_slots.c
deleted file mode 100644
index 9c7c171df7ba..000000000000
--- a/mm/swap_slots.c
+++ /dev/null
@@ -1,295 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-/*
- * Manage cache of swap slots to be used for and returned from
- * swap.
- *
- * Copyright(c) 2016 Intel Corporation.
- *
- * Author: Tim Chen <tim.c.chen@linux.intel.com>
- *
- * We allocate the swap slots from the global pool and put
- * it into local per cpu caches.  This has the advantage
- * of no needing to acquire the swap_info lock every time
- * we need a new slot.
- *
- * There is also opportunity to simply return the slot
- * to local caches without needing to acquire swap_info
- * lock.  We do not reuse the returned slots directly but
- * move them back to the global pool in a batch.  This
- * allows the slots to coalesce and reduce fragmentation.
- *
- * The swap entry allocated is marked with SWAP_HAS_CACHE
- * flag in map_count that prevents it from being allocated
- * again from the global pool.
- *
- * The swap slots cache is protected by a mutex instead of
- * a spin lock as when we search for slots with scan_swap_map,
- * we can possibly sleep.
- */
-
-#include <linux/swap_slots.h>
-#include <linux/cpu.h>
-#include <linux/cpumask.h>
-#include <linux/slab.h>
-#include <linux/vmalloc.h>
-#include <linux/mutex.h>
-#include <linux/mm.h>
-
-static DEFINE_PER_CPU(struct swap_slots_cache, swp_slots);
-static bool	swap_slot_cache_active;
-bool	swap_slot_cache_enabled;
-static bool	swap_slot_cache_initialized;
-static DEFINE_MUTEX(swap_slots_cache_mutex);
-/* Serialize swap slots cache enable/disable operations */
-static DEFINE_MUTEX(swap_slots_cache_enable_mutex);
-
-static void __drain_swap_slots_cache(void);
-
-#define use_swap_slot_cache (swap_slot_cache_active && swap_slot_cache_enabled)
-
-static void deactivate_swap_slots_cache(void)
-{
-	mutex_lock(&swap_slots_cache_mutex);
-	swap_slot_cache_active = false;
-	__drain_swap_slots_cache();
-	mutex_unlock(&swap_slots_cache_mutex);
-}
-
-static void reactivate_swap_slots_cache(void)
-{
-	mutex_lock(&swap_slots_cache_mutex);
-	swap_slot_cache_active = true;
-	mutex_unlock(&swap_slots_cache_mutex);
-}
-
-/* Must not be called with cpu hot plug lock */
-void disable_swap_slots_cache_lock(void)
-{
-	mutex_lock(&swap_slots_cache_enable_mutex);
-	swap_slot_cache_enabled = false;
-	if (swap_slot_cache_initialized) {
-		/* serialize with cpu hotplug operations */
-		cpus_read_lock();
-		__drain_swap_slots_cache();
-		cpus_read_unlock();
-	}
-}
-
-static void __reenable_swap_slots_cache(void)
-{
-	swap_slot_cache_enabled = has_usable_swap();
-}
-
-void reenable_swap_slots_cache_unlock(void)
-{
-	__reenable_swap_slots_cache();
-	mutex_unlock(&swap_slots_cache_enable_mutex);
-}
-
-static bool check_cache_active(void)
-{
-	long pages;
-
-	if (!swap_slot_cache_enabled)
-		return false;
-
-	pages = get_nr_swap_pages();
-	if (!swap_slot_cache_active) {
-		if (pages > num_online_cpus() *
-		    THRESHOLD_ACTIVATE_SWAP_SLOTS_CACHE)
-			reactivate_swap_slots_cache();
-		goto out;
-	}
-
-	/* if global pool of slot caches too low, deactivate cache */
-	if (pages < num_online_cpus() * THRESHOLD_DEACTIVATE_SWAP_SLOTS_CACHE)
-		deactivate_swap_slots_cache();
-out:
-	return swap_slot_cache_active;
-}
-
-static int alloc_swap_slot_cache(unsigned int cpu)
-{
-	struct swap_slots_cache *cache;
-	swp_entry_t *slots;
-
-	/*
-	 * Do allocation outside swap_slots_cache_mutex
-	 * as kvzalloc could trigger reclaim and folio_alloc_swap,
-	 * which can lock swap_slots_cache_mutex.
-	 */
-	slots = kvcalloc(SWAP_SLOTS_CACHE_SIZE, sizeof(swp_entry_t),
-			 GFP_KERNEL);
-	if (!slots)
-		return -ENOMEM;
-
-	mutex_lock(&swap_slots_cache_mutex);
-	cache = &per_cpu(swp_slots, cpu);
-	if (cache->slots) {
-		/* cache already allocated */
-		mutex_unlock(&swap_slots_cache_mutex);
-
-		kvfree(slots);
-
-		return 0;
-	}
-
-	if (!cache->lock_initialized) {
-		mutex_init(&cache->alloc_lock);
-		cache->lock_initialized = true;
-	}
-	cache->nr = 0;
-	cache->cur = 0;
-	cache->n_ret = 0;
-	/*
-	 * We initialized alloc_lock and free_lock earlier.  We use
-	 * !cache->slots or !cache->slots_ret to know if it is safe to acquire
-	 * the corresponding lock and use the cache.  Memory barrier below
-	 * ensures the assumption.
-	 */
-	mb();
-	cache->slots = slots;
-	mutex_unlock(&swap_slots_cache_mutex);
-	return 0;
-}
-
-static void drain_slots_cache_cpu(unsigned int cpu, bool free_slots)
-{
-	struct swap_slots_cache *cache;
-
-	cache = &per_cpu(swp_slots, cpu);
-	if (cache->slots) {
-		mutex_lock(&cache->alloc_lock);
-		swapcache_free_entries(cache->slots + cache->cur, cache->nr);
-		cache->cur = 0;
-		cache->nr = 0;
-		if (free_slots && cache->slots) {
-			kvfree(cache->slots);
-			cache->slots = NULL;
-		}
-		mutex_unlock(&cache->alloc_lock);
-	}
-}
-
-static void __drain_swap_slots_cache(void)
-{
-	unsigned int cpu;
-
-	/*
-	 * This function is called during
-	 *	1) swapoff, when we have to make sure no
-	 *	   left over slots are in cache when we remove
-	 *	   a swap device;
-	 *      2) disabling of swap slot cache, when we run low
-	 *	   on swap slots when allocating memory and need
-	 *	   to return swap slots to global pool.
-	 *
-	 * We cannot acquire cpu hot plug lock here as
-	 * this function can be invoked in the cpu
-	 * hot plug path:
-	 * cpu_up -> lock cpu_hotplug -> cpu hotplug state callback
-	 *   -> memory allocation -> direct reclaim -> folio_alloc_swap
-	 *   -> drain_swap_slots_cache
-	 *
-	 * Hence the loop over current online cpu below could miss cpu that
-	 * is being brought online but not yet marked as online.
-	 * That is okay as we do not schedule and run anything on a
-	 * cpu before it has been marked online. Hence, we will not
-	 * fill any swap slots in slots cache of such cpu.
-	 * There are no slots on such cpu that need to be drained.
-	 */
-	for_each_online_cpu(cpu)
-		drain_slots_cache_cpu(cpu, false);
-}
-
-static int free_slot_cache(unsigned int cpu)
-{
-	mutex_lock(&swap_slots_cache_mutex);
-	drain_slots_cache_cpu(cpu, true);
-	mutex_unlock(&swap_slots_cache_mutex);
-	return 0;
-}
-
-void enable_swap_slots_cache(void)
-{
-	mutex_lock(&swap_slots_cache_enable_mutex);
-	if (!swap_slot_cache_initialized) {
-		int ret;
-
-		ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "swap_slots_cache",
-					alloc_swap_slot_cache, free_slot_cache);
-		if (WARN_ONCE(ret < 0, "Cache allocation failed (%s), operating "
-				       "without swap slots cache.\n", __func__))
-			goto out_unlock;
-
-		swap_slot_cache_initialized = true;
-	}
-
-	__reenable_swap_slots_cache();
-out_unlock:
-	mutex_unlock(&swap_slots_cache_enable_mutex);
-}
-
-/* called with swap slot cache's alloc lock held */
-static int refill_swap_slots_cache(struct swap_slots_cache *cache)
-{
-	if (!use_swap_slot_cache)
-		return 0;
-
-	cache->cur = 0;
-	if (swap_slot_cache_active)
-		cache->nr = get_swap_pages(SWAP_SLOTS_CACHE_SIZE,
-					   cache->slots, 0);
-
-	return cache->nr;
-}
-
-swp_entry_t folio_alloc_swap(struct folio *folio)
-{
-	swp_entry_t entry;
-	struct swap_slots_cache *cache;
-
-	entry.val = 0;
-
-	if (folio_test_large(folio)) {
-		if (IS_ENABLED(CONFIG_THP_SWAP))
-			get_swap_pages(1, &entry, folio_order(folio));
-		goto out;
-	}
-
-	/*
-	 * Preemption is allowed here, because we may sleep
-	 * in refill_swap_slots_cache().  But it is safe, because
-	 * accesses to the per-CPU data structure are protected by the
-	 * mutex cache->alloc_lock.
-	 *
-	 * The alloc path here does not touch cache->slots_ret
-	 * so cache->free_lock is not taken.
-	 */
-	cache = raw_cpu_ptr(&swp_slots);
-
-	if (likely(check_cache_active() && cache->slots)) {
-		mutex_lock(&cache->alloc_lock);
-		if (cache->slots) {
-repeat:
-			if (cache->nr) {
-				entry = cache->slots[cache->cur];
-				cache->slots[cache->cur++].val = 0;
-				cache->nr--;
-			} else if (refill_swap_slots_cache(cache)) {
-				goto repeat;
-			}
-		}
-		mutex_unlock(&cache->alloc_lock);
-		if (entry.val)
-			goto out;
-	}
-
-	get_swap_pages(1, &entry, 0);
-out:
-	if (mem_cgroup_try_charge_swap(folio, entry)) {
-		put_swap_folio(folio, entry);
-		entry.val = 0;
-	}
-	return entry;
-}
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 50840a2887a5..2b5744e211cd 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -20,7 +20,6 @@
 #include <linux/blkdev.h>
 #include <linux/migrate.h>
 #include <linux/vmalloc.h>
-#include <linux/swap_slots.h>
 #include <linux/huge_mm.h>
 #include <linux/shmem_fs.h>
 #include "internal.h"
@@ -447,13 +446,8 @@ struct folio *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
 
 		/*
 		 * Just skip read ahead for unused swap slot.
-		 * During swap_off when swap_slot_cache is disabled,
-		 * we have to handle the race between putting
-		 * swap entry in swap cache and marking swap slot
-		 * as SWAP_HAS_CACHE.  That's done in later part of code or
-		 * else swap_off will be aborted if we return NULL.
 		 */
-		if (!swap_entry_swapped(si, entry) && swap_slot_cache_enabled)
+		if (!swap_entry_swapped(si, entry))
 			goto put_and_return;
 
 		/*
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 8b296c4c636b..9bd95173865d 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -37,7 +37,6 @@
 #include <linux/oom.h>
 #include <linux/swapfile.h>
 #include <linux/export.h>
-#include <linux/swap_slots.h>
 #include <linux/sort.h>
 #include <linux/completion.h>
 #include <linux/suspend.h>
@@ -885,16 +884,20 @@ static unsigned long cluster_alloc_swap_entry(struct swap_info_struct *si, int o
 	struct swap_cluster_info *ci;
 	unsigned int offset = SWAP_ENTRY_INVALID, found = SWAP_ENTRY_INVALID;
 
-	if (si->flags & SWP_SOLIDSTATE) {
-		if (si == this_cpu_read(percpu_swap_cluster.si[order]))
-			offset = this_cpu_read(percpu_swap_cluster.offset[order]);
-	} else {
+	/*
+	 * Swapfile is not block device so unable
+	 * to allocate large entries.
+	 */
+	if (order && !(si->flags & SWP_BLKDEV))
+		return 0;
+
+	if (!(si->flags & SWP_SOLIDSTATE)) {
 		/* Serialize HDD SWAP allocation for each device. */
 		spin_lock(&si->global_cluster_lock);
 		offset = si->global_cluster->next[order];
-	}
+		if (offset == SWAP_ENTRY_INVALID)
+			goto new_cluster;
 
-	if (offset) {
 		ci = lock_cluster(si, offset);
 		/* Cluster could have been used by another order */
 		if (cluster_is_usable(ci, order)) {
@@ -1153,43 +1156,6 @@ static void swap_range_free(struct swap_info_struct *si, unsigned long offset,
 	swap_usage_sub(si, nr_entries);
 }
 
-static int scan_swap_map_slots(struct swap_info_struct *si,
-			       unsigned char usage, int nr,
-			       swp_entry_t slots[], int order)
-{
-	unsigned int nr_pages = 1 << order;
-	int n_ret = 0;
-
-	if (order > 0) {
-		/*
-		 * Should not even be attempting large allocations when huge
-		 * page swap is disabled.  Warn and fail the allocation.
-		 */
-		if (!IS_ENABLED(CONFIG_THP_SWAP) ||
-		    nr_pages > SWAPFILE_CLUSTER) {
-			VM_WARN_ON_ONCE(1);
-			return 0;
-		}
-
-		/*
-		 * Swapfile is not block device so unable
-		 * to allocate large entries.
-		 */
-		if (!(si->flags & SWP_BLKDEV))
-			return 0;
-	}
-
-	while (n_ret < nr) {
-		unsigned long offset = cluster_alloc_swap_entry(si, order, usage);
-
-		if (!offset)
-			break;
-		slots[n_ret++] = swp_entry(si->type, offset);
-	}
-
-	return n_ret;
-}
-
 static bool get_swap_device_info(struct swap_info_struct *si)
 {
 	if (!percpu_ref_tryget_live(&si->users))
@@ -1210,16 +1176,13 @@ static bool get_swap_device_info(struct swap_info_struct *si)
  * Fast path try to get swap entries with specified order from current
  * CPU's swap entry pool (a cluster).
  */
-static int swap_alloc_fast(swp_entry_t entries[],
+static int swap_alloc_fast(swp_entry_t *entry,
 			   unsigned char usage,
-			   int order, int n_goal)
+			   int order)
 {
 	struct swap_cluster_info *ci;
 	struct swap_info_struct *si;
-	unsigned int offset, found;
-	int n_ret = 0;
-
-	n_goal = min(n_goal, SWAP_BATCH);
+	unsigned int offset, found = SWAP_ENTRY_INVALID;
 
 	/*
 	 * Once allocated, swap_info_struct will never be completely freed,
@@ -1228,46 +1191,48 @@ static int swap_alloc_fast(swp_entry_t entries[],
 	si = this_cpu_read(percpu_swap_cluster.si[order]);
 	offset = this_cpu_read(percpu_swap_cluster.offset[order]);
 	if (!si || !offset || !get_swap_device_info(si))
-		return 0;
+		return false;
 
-	while (offset) {
-		ci = lock_cluster(si, offset);
-		if (!cluster_is_usable(ci, order)) {
-			unlock_cluster(ci);
-			break;
-		}
+	ci = lock_cluster(si, offset);
+	if (cluster_is_usable(ci, order)) {
 		if (cluster_is_empty(ci))
 			offset = cluster_offset(si, ci);
 		found = alloc_swap_scan_cluster(si, ci, offset, order, usage);
-		if (!found)
-			break;
-		entries[n_ret++] = swp_entry(si->type, found);
-		if (n_ret == n_goal)
-			break;
-		offset = this_cpu_read(percpu_swap_cluster.offset[order]);
+		if (found)
+			*entry = swp_entry(si->type, found);
+	} else {
+		unlock_cluster(ci);
 	}
 
 	put_swap_device(si);
-	return n_ret;
+	return !!found;
 }
 
-int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
+swp_entry_t folio_alloc_swap(struct folio *folio)
 {
-	int order = swap_entry_order(entry_order);
-	unsigned long size = 1 << order;
+	unsigned int order = folio_order(folio);
+	unsigned int size = 1 << order;
 	struct swap_info_struct *si, *next;
-	int n_ret = 0;
+	swp_entry_t entry = {};
+	unsigned long offset;
 	int node;
 
+	if (order) {
+		/*
+		 * Should not even be attempting large allocations when huge
+		 * page swap is disabled. Warn and fail the allocation.
+		 */
+		if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
+			VM_WARN_ON_ONCE(1);
+			return entry;
+		}
+	}
+
 	/* Fast path using percpu cluster */
 	local_lock(&percpu_swap_cluster.lock);
-	n_ret = swap_alloc_fast(swp_entries,
-				SWAP_HAS_CACHE,
-				order, n_goal);
-	if (n_ret == n_goal)
+	if (swap_alloc_fast(&entry, SWAP_HAS_CACHE, order))
 		goto out;
 
-	n_goal = min_t(int, n_goal - n_ret, SWAP_BATCH);
 	/* Rotate the device and switch to a new cluster */
 	spin_lock(&swap_avail_lock);
 start_over:
@@ -1276,18 +1241,13 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 		plist_requeue(&si->avail_lists[node], &swap_avail_heads[node]);
 		spin_unlock(&swap_avail_lock);
 		if (get_swap_device_info(si)) {
-			/*
-			 * For order 0 allocation, try best to fill the request
-			 * as it's used by slot cache.
-			 *
-			 * For mTHP allocation, it always have n_goal == 1,
-			 * and falling a mTHP swapin will just make the caller
-			 * fallback to order 0 allocation, so just bail out.
-			 */
-			n_ret += scan_swap_map_slots(si, SWAP_HAS_CACHE, n_goal,
-					swp_entries + n_ret, order);
+			offset = cluster_alloc_swap_entry(si, order, SWAP_HAS_CACHE);
 			put_swap_device(si);
-			if (n_ret || size > 1)
+			if (offset) {
+				entry = swp_entry(si->type, offset);
+				goto out;
+			}
+			if (order)
 				goto out;
 		}
 
@@ -1309,8 +1269,14 @@ int get_swap_pages(int n_goal, swp_entry_t swp_entries[], int entry_order)
 	spin_unlock(&swap_avail_lock);
 out:
 	local_unlock(&percpu_swap_cluster.lock);
-	atomic_long_sub(n_ret * size, &nr_swap_pages);
-	return n_ret;
+	/* Need to call this even if allocation failed, for MEMCG_SWAP_FAIL. */
+	if (mem_cgroup_try_charge_swap(folio, entry)) {
+		put_swap_folio(folio, entry);
+		entry.val = 0;
+	}
+	if (entry.val)
+		atomic_long_sub(size, &nr_swap_pages);
+	return entry;
 }
 
 static struct swap_info_struct *_swap_info_get(swp_entry_t entry)
@@ -1606,25 +1572,6 @@ void put_swap_folio(struct folio *folio, swp_entry_t entry)
 	unlock_cluster(ci);
 }
 
-void swapcache_free_entries(swp_entry_t *entries, int n)
-{
-	int i;
-	struct swap_cluster_info *ci;
-	struct swap_info_struct *si = NULL;
-
-	if (n <= 0)
-		return;
-
-	for (i = 0; i < n; ++i) {
-		si = _swap_info_get(entries[i]);
-		if (si) {
-			ci = lock_cluster(si, swp_offset(entries[i]));
-			swap_entry_range_free(si, ci, entries[i], 1);
-			unlock_cluster(ci);
-		}
-	}
-}
-
 int __swap_count(swp_entry_t entry)
 {
 	struct swap_info_struct *si = swp_swap_info(entry);
@@ -1865,6 +1812,7 @@ void free_swap_and_cache_nr(swp_entry_t entry, int nr)
 swp_entry_t get_swap_page_of_type(int type)
 {
 	struct swap_info_struct *si = swap_type_to_swap_info(type);
+	unsigned long offset;
 	swp_entry_t entry = {0};
 
 	if (!si)
@@ -1872,8 +1820,13 @@ swp_entry_t get_swap_page_of_type(int type)
 
 	/* This is called for allocating swap entry, not cache */
 	if (get_swap_device_info(si)) {
-		if ((si->flags & SWP_WRITEOK) && scan_swap_map_slots(si, 1, 1, &entry, 0))
-			atomic_long_dec(&nr_swap_pages);
+		if (si->flags & SWP_WRITEOK) {
+			offset = cluster_alloc_swap_entry(si, 0, 1);
+			if (offset) {
+				entry = swp_entry(si->type, offset);
+				atomic_long_dec(&nr_swap_pages);
+			}
+		}
 		put_swap_device(si);
 	}
 fail:
@@ -2634,21 +2587,6 @@ static void reinsert_swap_info(struct swap_info_struct *si)
 	spin_unlock(&swap_lock);
 }
 
-static bool __has_usable_swap(void)
-{
-	return !plist_head_empty(&swap_active_head);
-}
-
-bool has_usable_swap(void)
-{
-	bool ret;
-
-	spin_lock(&swap_lock);
-	ret = __has_usable_swap();
-	spin_unlock(&swap_lock);
-	return ret;
-}
-
 /*
  * Called after clearing SWP_WRITEOK, ensures cluster_alloc_range
  * see the updated flags, so there will be no more allocations.
@@ -2761,8 +2699,6 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 
 	wait_for_allocation(p);
 
-	disable_swap_slots_cache_lock();
-
 	set_current_oom_origin();
 	err = try_to_unuse(p->type);
 	clear_current_oom_origin();
@@ -2770,12 +2706,9 @@ SYSCALL_DEFINE1(swapoff, const char __user *, specialfile)
 	if (err) {
 		/* re-insert swap space back into swap_list */
 		reinsert_swap_info(p);
-		reenable_swap_slots_cache_unlock();
 		goto out_dput;
 	}
 
-	reenable_swap_slots_cache_unlock();
-
 	/*
 	 * Wait for swap operations protected by get/put_swap_device()
 	 * to complete.  Because of synchronize_rcu() here, all swap
@@ -3525,8 +3458,6 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
 		putname(name);
 	if (inode)
 		inode_unlock(inode);
-	if (!error)
-		enable_swap_slots_cache();
 	return error;
 }
 
@@ -3922,6 +3853,11 @@ static void free_swap_count_continuations(struct swap_info_struct *si)
 }
 
 #if defined(CONFIG_MEMCG) && defined(CONFIG_BLK_CGROUP)
+static bool __has_usable_swap(void)
+{
+	return !plist_head_empty(&swap_active_head);
+}
+
 void __folio_throttle_swaprate(struct folio *folio, gfp_t gfp)
 {
 	struct swap_info_struct *si, *next;
-- 
2.48.1

[PATCH v3 7/7] mm, swap: simplify folio swap allocation

[Kairui Song]

From: Kairui Song <kasong@tencent.com>

With slot cache gone, clean up the allocation helpers even more.
folio_alloc_swap will be the only entry for allocation and adding the
folio to swap cache (except suspend), making it opposite of
folio_free_swap.

Signed-off-by: Kairui Song <kasong@tencent.com>
---
 include/linux/swap.h |   8 ++--
 mm/shmem.c           |  21 +++-----
 mm/swap.h            |   6 ---
 mm/swap_state.c      |  57 ----------------------
 mm/swapfile.c        | 111 ++++++++++++++++++++++++++++---------------
 mm/vmscan.c          |  16 ++++++-
 6 files changed, 95 insertions(+), 124 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index c5856dcc263a..9c99eee160f9 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -478,7 +478,7 @@ static inline long get_nr_swap_pages(void)
 }
 
 extern void si_swapinfo(struct sysinfo *);
-swp_entry_t folio_alloc_swap(struct folio *folio);
+int folio_alloc_swap(struct folio *folio, gfp_t gfp_mask);
 bool folio_free_swap(struct folio *folio);
 void put_swap_folio(struct folio *folio, swp_entry_t entry);
 extern swp_entry_t get_swap_page_of_type(int);
@@ -586,11 +586,9 @@ static inline int swp_swapcount(swp_entry_t entry)
 	return 0;
 }
 
-static inline swp_entry_t folio_alloc_swap(struct folio *folio)
+static inline int folio_alloc_swap(struct folio *folio, gfp_t gfp_mask)
 {
-	swp_entry_t entry;
-	entry.val = 0;
-	return entry;
+	return -EINVAL;
 }
 
 static inline bool folio_free_swap(struct folio *folio)
diff --git a/mm/shmem.c b/mm/shmem.c
index 1eed26bf8ae5..7b738d8d6581 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1546,7 +1546,6 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc)
 	struct inode *inode = mapping->host;
 	struct shmem_inode_info *info = SHMEM_I(inode);
 	struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb);
-	swp_entry_t swap;
 	pgoff_t index;
 	int nr_pages;
 	bool split = false;
@@ -1628,14 +1627,6 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc)
 		folio_mark_uptodate(folio);
 	}
 
-	swap = folio_alloc_swap(folio);
-	if (!swap.val) {
-		if (nr_pages > 1)
-			goto try_split;
-
-		goto redirty;
-	}
-
 	/*
 	 * Add inode to shmem_unuse()'s list of swapped-out inodes,
 	 * if it's not already there.  Do it now before the folio is
@@ -1648,20 +1639,20 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc)
 	if (list_empty(&info->swaplist))
 		list_add(&info->swaplist, &shmem_swaplist);
 
-	if (add_to_swap_cache(folio, swap,
-			__GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN,
-			NULL) == 0) {
+	if (!folio_alloc_swap(folio, __GFP_HIGH | __GFP_NOMEMALLOC | __GFP_NOWARN)) {
 		shmem_recalc_inode(inode, 0, nr_pages);
-		swap_shmem_alloc(swap, nr_pages);
-		shmem_delete_from_page_cache(folio, swp_to_radix_entry(swap));
+		swap_shmem_alloc(folio->swap, nr_pages);
+		shmem_delete_from_page_cache(folio, swp_to_radix_entry(folio->swap));
 
 		mutex_unlock(&shmem_swaplist_mutex);
 		BUG_ON(folio_mapped(folio));
 		return swap_writepage(&folio->page, wbc);
 	}
 
+	list_del_init(&info->swaplist);
 	mutex_unlock(&shmem_swaplist_mutex);
-	put_swap_folio(folio, swap);
+	if (nr_pages > 1)
+		goto try_split;
 redirty:
 	folio_mark_dirty(folio);
 	if (wbc->for_reclaim)
diff --git a/mm/swap.h b/mm/swap.h
index ad2f121de970..0abb68091b4f 100644
--- a/mm/swap.h
+++ b/mm/swap.h
@@ -50,7 +50,6 @@ static inline pgoff_t swap_cache_index(swp_entry_t entry)
 }
 
 void show_swap_cache_info(void);
-bool add_to_swap(struct folio *folio);
 void *get_shadow_from_swap_cache(swp_entry_t entry);
 int add_to_swap_cache(struct folio *folio, swp_entry_t entry,
 		      gfp_t gfp, void **shadowp);
@@ -163,11 +162,6 @@ struct folio *filemap_get_incore_folio(struct address_space *mapping,
 	return filemap_get_folio(mapping, index);
 }
 
-static inline bool add_to_swap(struct folio *folio)
-{
-	return false;
-}
-
 static inline void *get_shadow_from_swap_cache(swp_entry_t entry)
 {
 	return NULL;
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 2b5744e211cd..68fd981b514f 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -166,63 +166,6 @@ void __delete_from_swap_cache(struct folio *folio,
 	__lruvec_stat_mod_folio(folio, NR_SWAPCACHE, -nr);
 }
 
-/**
- * add_to_swap - allocate swap space for a folio
- * @folio: folio we want to move to swap
- *
- * Allocate swap space for the folio and add the folio to the
- * swap cache.
- *
- * Context: Caller needs to hold the folio lock.
- * Return: Whether the folio was added to the swap cache.
- */
-bool add_to_swap(struct folio *folio)
-{
-	swp_entry_t entry;
-	int err;
-
-	VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);
-	VM_BUG_ON_FOLIO(!folio_test_uptodate(folio), folio);
-
-	entry = folio_alloc_swap(folio);
-	if (!entry.val)
-		return false;
-
-	/*
-	 * XArray node allocations from PF_MEMALLOC contexts could
-	 * completely exhaust the page allocator. __GFP_NOMEMALLOC
-	 * stops emergency reserves from being allocated.
-	 *
-	 * TODO: this could cause a theoretical memory reclaim
-	 * deadlock in the swap out path.
-	 */
-	/*
-	 * Add it to the swap cache.
-	 */
-	err = add_to_swap_cache(folio, entry,
-			__GFP_HIGH|__GFP_NOMEMALLOC|__GFP_NOWARN, NULL);
-	if (err)
-		goto fail;
-	/*
-	 * Normally the folio will be dirtied in unmap because its
-	 * pte should be dirty. A special case is MADV_FREE page. The
-	 * page's pte could have dirty bit cleared but the folio's
-	 * SwapBacked flag is still set because clearing the dirty bit
-	 * and SwapBacked flag has no lock protected. For such folio,
-	 * unmap will not set dirty bit for it, so folio reclaim will
-	 * not write the folio out. This can cause data corruption when
-	 * the folio is swapped in later. Always setting the dirty flag
-	 * for the folio solves the problem.
-	 */
-	folio_mark_dirty(folio);
-
-	return true;
-
-fail:
-	put_swap_folio(folio, entry);
-	return false;
-}
-
 /*
  * This must be called only on folios that have
  * been verified to be in the swap cache and locked.
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 9bd95173865d..2eff8b51a945 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1176,9 +1176,8 @@ static bool get_swap_device_info(struct swap_info_struct *si)
  * Fast path try to get swap entries with specified order from current
  * CPU's swap entry pool (a cluster).
  */
-static int swap_alloc_fast(swp_entry_t *entry,
-			   unsigned char usage,
-			   int order)
+static bool swap_alloc_fast(swp_entry_t *entry,
+			    int order)
 {
 	struct swap_cluster_info *ci;
 	struct swap_info_struct *si;
@@ -1197,7 +1196,7 @@ static int swap_alloc_fast(swp_entry_t *entry,
 	if (cluster_is_usable(ci, order)) {
 		if (cluster_is_empty(ci))
 			offset = cluster_offset(si, ci);
-		found = alloc_swap_scan_cluster(si, ci, offset, order, usage);
+		found = alloc_swap_scan_cluster(si, ci, offset, order, SWAP_HAS_CACHE);
 		if (found)
 			*entry = swp_entry(si->type, found);
 	} else {
@@ -1208,47 +1207,30 @@ static int swap_alloc_fast(swp_entry_t *entry,
 	return !!found;
 }
 
-swp_entry_t folio_alloc_swap(struct folio *folio)
+/* Rotate the device and switch to a new cluster */
+static bool swap_alloc_slow(swp_entry_t *entry,
+			    int order)
 {
-	unsigned int order = folio_order(folio);
-	unsigned int size = 1 << order;
-	struct swap_info_struct *si, *next;
-	swp_entry_t entry = {};
-	unsigned long offset;
 	int node;
+	unsigned long offset;
+	struct swap_info_struct *si, *next;
 
-	if (order) {
-		/*
-		 * Should not even be attempting large allocations when huge
-		 * page swap is disabled. Warn and fail the allocation.
-		 */
-		if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
-			VM_WARN_ON_ONCE(1);
-			return entry;
-		}
-	}
-
-	/* Fast path using percpu cluster */
-	local_lock(&percpu_swap_cluster.lock);
-	if (swap_alloc_fast(&entry, SWAP_HAS_CACHE, order))
-		goto out;
-
-	/* Rotate the device and switch to a new cluster */
+	node = numa_node_id();
 	spin_lock(&swap_avail_lock);
 start_over:
-	node = numa_node_id();
 	plist_for_each_entry_safe(si, next, &swap_avail_heads[node], avail_lists[node]) {
+		/* Rotate the device and switch to a new cluster */
 		plist_requeue(&si->avail_lists[node], &swap_avail_heads[node]);
 		spin_unlock(&swap_avail_lock);
 		if (get_swap_device_info(si)) {
 			offset = cluster_alloc_swap_entry(si, order, SWAP_HAS_CACHE);
 			put_swap_device(si);
 			if (offset) {
-				entry = swp_entry(si->type, offset);
-				goto out;
+				*entry = swp_entry(si->type, offset);
+				return true;
 			}
 			if (order)
-				goto out;
+				return false;
 		}
 
 		spin_lock(&swap_avail_lock);
@@ -1267,16 +1249,67 @@ swp_entry_t folio_alloc_swap(struct folio *folio)
 			goto start_over;
 	}
 	spin_unlock(&swap_avail_lock);
-out:
+	return false;
+}
+
+/**
+ * folio_alloc_swap - allocate swap space for a folio
+ * @folio: folio we want to move to swap
+ * @gfp: gfp mask for shadow nodes
+ *
+ * Allocate swap space for the folio and add the folio to the
+ * swap cache.
+ *
+ * Context: Caller needs to hold the folio lock.
+ * Return: Whether the folio was added to the swap cache.
+ */
+int folio_alloc_swap(struct folio *folio, gfp_t gfp)
+{
+	unsigned int order = folio_order(folio);
+	unsigned int size = 1 << order;
+	swp_entry_t entry = {};
+
+	VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);
+	VM_BUG_ON_FOLIO(!folio_test_uptodate(folio), folio);
+
+	/*
+	 * Should not even be attempting large allocations when huge
+	 * page swap is disabled. Warn and fail the allocation.
+	 */
+	if (order && (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER)) {
+		VM_WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
+
+	local_lock(&percpu_swap_cluster.lock);
+	if (!swap_alloc_fast(&entry, order))
+		swap_alloc_slow(&entry, order);
 	local_unlock(&percpu_swap_cluster.lock);
+
 	/* Need to call this even if allocation failed, for MEMCG_SWAP_FAIL. */
-	if (mem_cgroup_try_charge_swap(folio, entry)) {
-		put_swap_folio(folio, entry);
-		entry.val = 0;
-	}
-	if (entry.val)
-		atomic_long_sub(size, &nr_swap_pages);
-	return entry;
+	if (mem_cgroup_try_charge_swap(folio, entry))
+		goto out_free;
+
+	if (!entry.val)
+		return -ENOMEM;
+
+	/*
+	 * XArray node allocations from PF_MEMALLOC contexts could
+	 * completely exhaust the page allocator. __GFP_NOMEMALLOC
+	 * stops emergency reserves from being allocated.
+	 *
+	 * TODO: this could cause a theoretical memory reclaim
+	 * deadlock in the swap out path.
+	 */
+	if (add_to_swap_cache(folio, entry, gfp | __GFP_NOMEMALLOC, NULL))
+		goto out_free;
+
+	atomic_long_sub(size, &nr_swap_pages);
+	return 0;
+
+out_free:
+	put_swap_folio(folio, entry);
+	return -ENOMEM;
 }
 
 static struct swap_info_struct *_swap_info_get(swp_entry_t entry)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 84ec20f12200..2bc740637a6c 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1289,7 +1289,7 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 					    split_folio_to_list(folio, folio_list))
 						goto activate_locked;
 				}
-				if (!add_to_swap(folio)) {
+				if (folio_alloc_swap(folio, __GFP_HIGH | __GFP_NOWARN)) {
 					int __maybe_unused order = folio_order(folio);
 
 					if (!folio_test_large(folio))
@@ -1305,9 +1305,21 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 					}
 #endif
 					count_mthp_stat(order, MTHP_STAT_SWPOUT_FALLBACK);
-					if (!add_to_swap(folio))
+					if (folio_alloc_swap(folio, __GFP_HIGH | __GFP_NOWARN))
 						goto activate_locked_split;
 				}
+				/*
+				 * Normally the folio will be dirtied in unmap because its
+				 * pte should be dirty. A special case is MADV_FREE page. The
+				 * page's pte could have dirty bit cleared but the folio's
+				 * SwapBacked flag is still set because clearing the dirty bit
+				 * and SwapBacked flag has no lock protected. For such folio,
+				 * unmap will not set dirty bit for it, so folio reclaim will
+				 * not write the folio out. This can cause data corruption when
+				 * the folio is swapped in later. Always setting the dirty flag
+				 * for the folio solves the problem.
+				 */
+				folio_mark_dirty(folio);
 			}
 		}
 
-- 
2.48.1

Re: [PATCH v3 6/7] mm, swap: remove swap slot cache

[Heiko Carstens]

Hi Kairui,

On Fri, Mar 14, 2025 at 12:59:34AM +0800, Kairui Song wrote:
> From: Kairui Song <kasong@tencent.com>
> 
> Slot cache is no longer needed now, removing it and all related code.
...
> Signed-off-by: Kairui Song <kasong@tencent.com>
> Reviewed-by: Baoquan He <bhe@redhat.com>
> ---
>  include/linux/swap.h       |   3 -
>  include/linux/swap_slots.h |  28 ----
>  mm/Makefile                |   2 +-
>  mm/swap_slots.c            | 295 -------------------------------------
>  mm/swap_state.c            |   8 +-
>  mm/swapfile.c              | 194 ++++++++----------------
>  6 files changed, 67 insertions(+), 463 deletions(-)
>  delete mode 100644 include/linux/swap_slots.h
>  delete mode 100644 mm/swap_slots.c
...
> diff --git a/mm/swapfile.c b/mm/swapfile.c
...
> +swp_entry_t folio_alloc_swap(struct folio *folio)
>  {
> -	int order = swap_entry_order(entry_order);
> -	unsigned long size = 1 << order;
> +	unsigned int order = folio_order(folio);
> +	unsigned int size = 1 << order;
>  	struct swap_info_struct *si, *next;
> -	int n_ret = 0;
> +	swp_entry_t entry = {};
> +	unsigned long offset;
>  	int node;
>  
> +	if (order) {
> +		/*
> +		 * Should not even be attempting large allocations when huge
> +		 * page swap is disabled. Warn and fail the allocation.
> +		 */
> +		if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
> +			VM_WARN_ON_ONCE(1);
> +			return entry;
> +		}
> +	}

This warning triggers on s390. CONFIG_THP_SWAP is disabled and order
is 8 when this triggers (reproduced with ltp's swapon01 test case):

------------[ cut here ]------------
WARNING: CPU: 1 PID: 895 at mm/swapfile.c:1227 folio_alloc_swap+0x438/0x440
Modules linked in:
CPU: 1 UID: 0 PID: 895 Comm: swapon01 Not tainted 6.14.0-rc6-00227-g0ff67f990bd4-dirty #25
Hardware name: IBM 3931 A01 704 (z/VM 7.4.0)
Krnl PSW : 0704d00180000000 000003ffe051210c (folio_alloc_swap+0x43c/0x440)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:1 PM:0 RI:0 EA:3
Krnl GPRS: 0000000080000000 0000000000000001 0000000000000013 0000000000070000
           0000000000000006 fffffef40e9da000 0000000000000000 0000037202fc4000
           0000037f00000100 0000000000000100 0000037fe2e4b770 0000037202fc4000
           0000000000000000 0000000000000000 000003ffe0512108 0000037fe2e4b3c8
Krnl Code: 000003ffe05120fe: b9160044            llgfr   %r4,%r4
           000003ffe0512102: c0e5ffdf8c0b        brasl   %r14,000003ffe0103918
          #000003ffe0512108: af000000            mc      0,0
          >000003ffe051210c: a7f4fe94            brc     15,000003ffe0511e34
           000003ffe0512110: c0040069ce74        brcl    0,000003ffe124bdf8
           000003ffe0512116: eb8ff0580024        stmg    %r8,%r15,88(%r15)
           000003ffe051211c: b90400ef            lgr     %r14,%r15
           000003ffe0512120: e3f0ffb8ff71        lay     %r15,-72(%r15)
Call Trace:
 [<000003ffe051210c>] folio_alloc_swap+0x43c/0x440 
 [<000003ffe050afa6>] add_to_swap+0x56/0xf0 
 [<000003ffe045fdc0>] shrink_folio_list+0xe80/0x13b0 
 [<000003ffe0461946>] shrink_inactive_list+0x1a6/0x550 
 [<000003ffe04624a2>] shrink_lruvec+0x2b2/0x410 
 [<000003ffe0462840>] shrink_node_memcgs+0x240/0x2d0 
 [<000003ffe0462986>] shrink_node+0xb6/0x3e0 
 [<000003ffe046302a>] do_try_to_free_pages+0xda/0x610 
 [<000003ffe0464d2c>] try_to_free_mem_cgroup_pages+0x14c/0x2a0 
 [<000003ffe0568270>] try_charge_memcg+0x220/0x5d0 
 [<000003ffe056867a>] charge_memcg+0x5a/0x270 
 [<000003ffe056a484>] __mem_cgroup_charge+0x44/0x80 
 [<000003ffe04acf20>] alloc_anon_folio+0x280/0x610 
 [<000003ffe04ad45a>] do_anonymous_page+0x1aa/0x5e0 
 [<000003ffe04af4c4>] __handle_mm_fault+0x244/0x500 
 [<000003ffe04af820>] handle_mm_fault+0xa0/0x170 
 [<000003ffe01533f8>] do_exception+0x1d8/0x4a0 
 [<000003ffe11fb92a>] __do_pgm_check+0x13a/0x220 
 [<000003ffe120c3ce>] pgm_check_handler+0x11e/0x170 
---[ end trace 0000000000000000 ]---

Re: [PATCH v3 6/7] mm, swap: remove swap slot cache

[Kairui Song]

On Mon, Apr 28, 2025 at 9:53 PM Heiko Carstens <hca@linux.ibm.com> wrote:
>
> Hi Kairui,
>
> On Fri, Mar 14, 2025 at 12:59:34AM +0800, Kairui Song wrote:
> > From: Kairui Song <kasong@tencent.com>
> >
> > Slot cache is no longer needed now, removing it and all related code.
> ...
> > Signed-off-by: Kairui Song <kasong@tencent.com>
> > Reviewed-by: Baoquan He <bhe@redhat.com>
> > ---
> >  include/linux/swap.h       |   3 -
> >  include/linux/swap_slots.h |  28 ----
> >  mm/Makefile                |   2 +-
> >  mm/swap_slots.c            | 295 -------------------------------------
> >  mm/swap_state.c            |   8 +-
> >  mm/swapfile.c              | 194 ++++++++----------------
> >  6 files changed, 67 insertions(+), 463 deletions(-)
> >  delete mode 100644 include/linux/swap_slots.h
> >  delete mode 100644 mm/swap_slots.c
> ...
> > diff --git a/mm/swapfile.c b/mm/swapfile.c
> ...
> > +swp_entry_t folio_alloc_swap(struct folio *folio)
> >  {
> > -     int order = swap_entry_order(entry_order);
> > -     unsigned long size = 1 << order;
> > +     unsigned int order = folio_order(folio);
> > +     unsigned int size = 1 << order;
> >       struct swap_info_struct *si, *next;
> > -     int n_ret = 0;
> > +     swp_entry_t entry = {};
> > +     unsigned long offset;
> >       int node;
> >
> > +     if (order) {
> > +             /*
> > +              * Should not even be attempting large allocations when huge
> > +              * page swap is disabled. Warn and fail the allocation.
> > +              */
> > +             if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
> > +                     VM_WARN_ON_ONCE(1);
> > +                     return entry;
> > +             }
> > +     }
>
> This warning triggers on s390. CONFIG_THP_SWAP is disabled and order
> is 8 when this triggers (reproduced with ltp's swapon01 test case):

Hi Heiko,

Thanks for the report.

>
> ------------[ cut here ]------------
> WARNING: CPU: 1 PID: 895 at mm/swapfile.c:1227 folio_alloc_swap+0x438/0x440
> Modules linked in:
> CPU: 1 UID: 0 PID: 895 Comm: swapon01 Not tainted 6.14.0-rc6-00227-g0ff67f990bd4-dirty #25
> Hardware name: IBM 3931 A01 704 (z/VM 7.4.0)
> Krnl PSW : 0704d00180000000 000003ffe051210c (folio_alloc_swap+0x43c/0x440)
>            R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:1 PM:0 RI:0 EA:3
> Krnl GPRS: 0000000080000000 0000000000000001 0000000000000013 0000000000070000
>            0000000000000006 fffffef40e9da000 0000000000000000 0000037202fc4000
>            0000037f00000100 0000000000000100 0000037fe2e4b770 0000037202fc4000
>            0000000000000000 0000000000000000 000003ffe0512108 0000037fe2e4b3c8
> Krnl Code: 000003ffe05120fe: b9160044            llgfr   %r4,%r4
>            000003ffe0512102: c0e5ffdf8c0b        brasl   %r14,000003ffe0103918
>           #000003ffe0512108: af000000            mc      0,0
>           >000003ffe051210c: a7f4fe94            brc     15,000003ffe0511e34
>            000003ffe0512110: c0040069ce74        brcl    0,000003ffe124bdf8
>            000003ffe0512116: eb8ff0580024        stmg    %r8,%r15,88(%r15)
>            000003ffe051211c: b90400ef            lgr     %r14,%r15
>            000003ffe0512120: e3f0ffb8ff71        lay     %r15,-72(%r15)
> Call Trace:
>  [<000003ffe051210c>] folio_alloc_swap+0x43c/0x440
>  [<000003ffe050afa6>] add_to_swap+0x56/0xf0
>  [<000003ffe045fdc0>] shrink_folio_list+0xe80/0x13b0
>  [<000003ffe0461946>] shrink_inactive_list+0x1a6/0x550
>  [<000003ffe04624a2>] shrink_lruvec+0x2b2/0x410
>  [<000003ffe0462840>] shrink_node_memcgs+0x240/0x2d0
>  [<000003ffe0462986>] shrink_node+0xb6/0x3e0
>  [<000003ffe046302a>] do_try_to_free_pages+0xda/0x610
>  [<000003ffe0464d2c>] try_to_free_mem_cgroup_pages+0x14c/0x2a0
>  [<000003ffe0568270>] try_charge_memcg+0x220/0x5d0
>  [<000003ffe056867a>] charge_memcg+0x5a/0x270
>  [<000003ffe056a484>] __mem_cgroup_charge+0x44/0x80
>  [<000003ffe04acf20>] alloc_anon_folio+0x280/0x610
>  [<000003ffe04ad45a>] do_anonymous_page+0x1aa/0x5e0
>  [<000003ffe04af4c4>] __handle_mm_fault+0x244/0x500
>  [<000003ffe04af820>] handle_mm_fault+0xa0/0x170
>  [<000003ffe01533f8>] do_exception+0x1d8/0x4a0
>  [<000003ffe11fb92a>] __do_pgm_check+0x13a/0x220
>  [<000003ffe120c3ce>] pgm_check_handler+0x11e/0x170
> ---[ end trace 0000000000000000 ]---
>

The !CONFIG_THP_SWAP check existed before because slot cache should
reject high order allocation. But slot cache is gone, so large
allocation will directly go to the allocator.

It was not a meaningful WARN in the first place, and now the allocator
should just fail silently for high order allocation, that's totally
fine and expected and will just inform the caller to split the folio.

I'll just change the WARN_ON condition to `if (order && size >
SWAPFILE_CLUSTER)` then, this should silence the WARN.

Re: [PATCH v3 6/7] mm, swap: remove swap slot cache

[Heiko Carstens]

On Mon, Apr 28, 2025 at 11:31:59PM +0800, Kairui Song wrote:
> On Mon, Apr 28, 2025 at 9:53 PM Heiko Carstens <hca@linux.ibm.com> wrote:
> > > +     if (order) {
> > > +             /*
> > > +              * Should not even be attempting large allocations when huge
> > > +              * page swap is disabled. Warn and fail the allocation.
> > > +              */
> > > +             if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
> > > +                     VM_WARN_ON_ONCE(1);
> > > +                     return entry;
> > > +             }
> > > +     }
> 
> The !CONFIG_THP_SWAP check existed before because slot cache should
> reject high order allocation. But slot cache is gone, so large
> allocation will directly go to the allocator.
> 
> It was not a meaningful WARN in the first place, and now the allocator
> should just fail silently for high order allocation, that's totally
> fine and expected and will just inform the caller to split the folio.
> 
> I'll just change the WARN_ON condition to `if (order && size >
> SWAPFILE_CLUSTER)` then, this should silence the WARN.

If I understand your suggestion correctly then this would be the
resulting code:

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 2eff8b51a945..5a7797143948 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1276,7 +1276,7 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp)
         * Should not even be attempting large allocations when huge
         * page swap is disabled. Warn and fail the allocation.
         */
-       if (order && (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER)) {
+       if (order && size > SWAPFILE_CLUSTER) {
                VM_WARN_ON_ONCE(1);
                return -EINVAL;
        }

However, with that change I get this splat (and a few more) instead:

------------[ cut here ]------------
DEBUG_LOCKS_WARN_ON(l->owner != current)
WARNING: CPU: 4 PID: 934 at ./include/linux/local_lock_internal.h:52 folio_alloc_swap+0x22e/0x560
Modules linked in:
CPU: 4 UID: 0 PID: 934 Comm: swapon01 Not tainted 6.15.0-rc4-00021-gca91b9500108-dirty #2 PREEMPT 
Hardware name: IBM 3931 A01 704 (z/VM 7.4.0)
Krnl PSW : 0704f00180000000 000003ffe05231b2 (folio_alloc_swap+0x232/0x560)
           R:0 T:1 IO:1 EX:1 Key:0 M:1 W:0 P:0 AS:3 CC:3 PM:0 RI:0 EA:3
Krnl GPRS: 000003ff80000002 0000000000000003 0000000000000028 000003ffe2fe6c90
           0000000000000003 000003ff80000003 000003ffe19fc538 000003720322c000
           0000000000000000 000002f2ab38f000 000002f200000100 000003720322c000
           0000000000000000 000002f3f0eda278 000003ffe05231ae 0000037febbdb3f8
Krnl Code: 000003ffe05231a2: c020008e69bc        larl    %r2,000003ffe16f051a
           000003ffe05231a8: c0e5ffe2f4dc        brasl   %r14,000003ffe0181b60
          #000003ffe05231ae: af000000            mc      0,0
          >000003ffe05231b2: a7f4ff92            brc     15,000003ffe05230d6
           000003ffe05231b6: a7080001            lhi     %r0,1
           000003ffe05231ba: a7a80001            lhi     %r10,1
           000003ffe05231be: a7980000            lhi     %r9,0
           000003ffe05231c2: a7f4ff04            brc     15,000003ffe0522fca
Call Trace:
 [<000003ffe05231b2>] folio_alloc_swap+0x232/0x560 
 [<000003ffe046d2d2>] shrink_folio_list+0xe02/0x12d0 
 [<000003ffe046edc8>] shrink_inactive_list+0x188/0x5a0 
 [<000003ffe046f7d4>] shrink_lruvec+0x104/0x400 
 [<000003ffe046fd0c>] shrink_node_memcgs+0x23c/0x2c0 
 [<000003ffe046fe3a>] shrink_node+0xaa/0x420 
 [<000003ffe0470500>] do_try_to_free_pages+0xd0/0x5c0 
 [<000003ffe0472144>] try_to_free_mem_cgroup_pages+0x144/0x290 
 [<000003ffe057b33a>] try_charge_memcg+0x1ca/0x420 
 [<000003ffe057b5ea>] charge_memcg+0x5a/0x1a0 
 [<000003ffe057d5d4>] __mem_cgroup_charge+0x44/0x160 
 [<000003ffe04bb6ac>] alloc_anon_folio+0x27c/0x5d0 
 [<000003ffe04bbbac>] do_anonymous_page+0x1ac/0x7b0 
 [<000003ffe04bdec2>] __handle_mm_fault+0x212/0x4c0 
 [<000003ffe04be20c>] handle_mm_fault+0x9c/0x230 
 [<000003ffe015422c>] do_exception+0x1dc/0x540 
 [<000003ffe125c560>] __do_pgm_check+0x130/0x220 
 [<000003ffe126e3ae>] pgm_check_handler+0x11e/0x170 
INFO: lockdep is turned off.
Last Breaking-Event-Address:
 [<000003ffe0181c96>] __warn_printk+0x136/0x140
irq event stamp: 4120751
hardirqs last  enabled at (4120751): [<000003ffe126d0ee>] _raw_spin_unlock_irq+0x3e/0x80
hardirqs last disabled at (4120750): [<000003ffe126cca8>] _raw_spin_lock_irq+0x98/0xd0
softirqs last  enabled at (4104986): [<000003ffe018c1fc>] handle_softirqs+0x2fc/0x550
softirqs last disabled at (4104969): [<000003ffe018b9c6>] __irq_exit_rcu+0x126/0x140
---[ end trace 0000000000000000 ]---

Re: [PATCH v3 6/7] mm, swap: remove swap slot cache

[Kairui Song]

On Tue, Apr 29, 2025 at 3:31 PM Heiko Carstens <hca@linux.ibm.com> wrote:
>
> On Mon, Apr 28, 2025 at 11:31:59PM +0800, Kairui Song wrote:
> > On Mon, Apr 28, 2025 at 9:53 PM Heiko Carstens <hca@linux.ibm.com> wrote:
> > > > +     if (order) {
> > > > +             /*
> > > > +              * Should not even be attempting large allocations when huge
> > > > +              * page swap is disabled. Warn and fail the allocation.
> > > > +              */
> > > > +             if (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER) {
> > > > +                     VM_WARN_ON_ONCE(1);
> > > > +                     return entry;
> > > > +             }
> > > > +     }
> >
> > The !CONFIG_THP_SWAP check existed before because slot cache should
> > reject high order allocation. But slot cache is gone, so large
> > allocation will directly go to the allocator.
> >
> > It was not a meaningful WARN in the first place, and now the allocator
> > should just fail silently for high order allocation, that's totally
> > fine and expected and will just inform the caller to split the folio.
> >
> > I'll just change the WARN_ON condition to `if (order && size >
> > SWAPFILE_CLUSTER)` then, this should silence the WARN.
>
> If I understand your suggestion correctly then this would be the
> resulting code:
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 2eff8b51a945..5a7797143948 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1276,7 +1276,7 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp)
>          * Should not even be attempting large allocations when huge
>          * page swap is disabled. Warn and fail the allocation.
>          */
> -       if (order && (!IS_ENABLED(CONFIG_THP_SWAP) || size > SWAPFILE_CLUSTER)) {
> +       if (order && size > SWAPFILE_CLUSTER) {
>                 VM_WARN_ON_ONCE(1);
>                 return -EINVAL;
>         }
>
> However, with that change I get this splat (and a few more) instead:

Sorry my bad, the allocator needs to fail silencely, not ignore and go
on. So it should be:

diff --git a/mm/swapfile.c b/mm/swapfile.c
index e727021b8e2c..b86637cfb17a 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1272,13 +1272,22 @@ int folio_alloc_swap(struct folio *folio, gfp_t gfp)
        VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);
        VM_BUG_ON_FOLIO(!folio_test_uptodate(folio), folio);

-       /*
-        * Should not even be attempting large allocations when huge
-        * page swap is disabled. Warn and fail the allocation.
-        */
-       if (order && (!IS_ENABLED(CONFIG_THP_SWAP) || size >
SWAPFILE_CLUSTER)) {
-               VM_WARN_ON_ONCE(1);
-               return -EINVAL;
+       if (order) {
+               /*
+                * Reject large allocation when THP_SWAP is disabled,
+                * the caller should split the folio and try again.
+                */
+               if (!IS_ENABLED(CONFIG_THP_SWAP))
+                       return -EAGAIN;
+
+               /*
+                * Allocation size should never exceed cluster size
+                * (HPAGE_PMD_SIZE).
+                */
+               if (size > SWAPFILE_CLUSTER) {
+                       VM_WARN_ON_ONCE(1);
+                       return -EINVAL;
+               }
        }

        local_lock(&percpu_swap_cluster.lock);

---

I've tested locally and it seems to work well, I'll send a patch to fix it.