The patch titled
Subject: zswap: make shrinking memcg-aware
has been added to the -mm mm-unstable branch. Its filename is
zswap-make-shrinking-memcg-aware.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/zswap-make-shrinking-memcg-aware.patch
This patch will later appear in the mm-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Domenico Cerasuolo <cerasuolodomenico@xxxxxxxxx>
Subject: zswap: make shrinking memcg-aware
Date: Tue, 17 Oct 2023 16:21:49 -0700
Currently, we only have a single global LRU for zswap. This makes it
impossible to perform worload-specific shrinking - an memcg cannot
determine which pages in the pool it owns, and often ends up writing pages
from other memcgs. This issue has been previously observed in practice
and mitigated by simply disabling memcg-initiated shrinking:
https://lore.kernel.org/all/20230530232435.3097106-1-nphamcs@xxxxxxxxx/T/#u
This patch fully resolves the issue by replacing the global zswap LRU
with memcg- and NUMA-specific LRUs, and modify the reclaim logic:
a) When a store attempt hits an memcg limit, it now triggers a
synchronous reclaim attempt that, if successful, allows the new
hotter page to be accepted by zswap.
b) If the store attempt instead hits the global zswap limit, it will
trigger an asynchronous reclaim attempt, in which an memcg is
selected for reclaim in a round-robin-like fashion.
Link: https://lkml.kernel.org/r/20231017232152.2605440-3-nphamcs@xxxxxxxxx
Signed-off-by: Domenico Cerasuolo <cerasuolodomenico@xxxxxxxxx>
Co-developed-by: Nhat Pham <nphamcs@xxxxxxxxx>
Signed-off-by: Nhat Pham <nphamcs@xxxxxxxxx>
Cc: Dan Streetman <ddstreet@xxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: Muchun Song <muchun.song@xxxxxxxxx>
Cc: Roman Gushchin <roman.gushchin@xxxxxxxxx>
Cc: Seth Jennings <sjenning@xxxxxxxxxx>
Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
Cc: Shuah Khan <shuah@xxxxxxxxxx>
Cc: Vitaly Wool <vitaly.wool@xxxxxxxxxxxx>
Cc: Yosry Ahmed <yosryahmed@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/memcontrol.h | 5
mm/swap.h | 3
mm/swap_state.c | 17 ++-
mm/zswap.c | 179 ++++++++++++++++++++++++-----------
4 files changed, 147 insertions(+), 57 deletions(-)
--- a/include/linux/memcontrol.h~zswap-make-shrinking-memcg-aware
+++ a/include/linux/memcontrol.h
@@ -1179,6 +1179,11 @@ static inline struct mem_cgroup *page_me
return NULL;
}
+static inline struct mem_cgroup *get_mem_cgroup_from_objcg(struct obj_cgroup *objcg)
+{
+ return NULL;
+}
+
static inline bool folio_memcg_kmem(struct folio *folio)
{
return false;
--- a/mm/swap.h~zswap-make-shrinking-memcg-aware
+++ a/mm/swap.h
@@ -50,7 +50,8 @@ struct page *read_swap_cache_async(swp_e
struct page *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma,
unsigned long addr,
- bool *new_page_allocated);
+ bool *new_page_allocated,
+ bool fail_if_exists);
struct page *swap_cluster_readahead(swp_entry_t entry, gfp_t flag,
struct vm_fault *vmf);
struct page *swapin_readahead(swp_entry_t entry, gfp_t flag,
--- a/mm/swap_state.c~zswap-make-shrinking-memcg-aware
+++ a/mm/swap_state.c
@@ -411,7 +411,7 @@ struct folio *filemap_get_incore_folio(s
struct page *__read_swap_cache_async(swp_entry_t entry, gfp_t gfp_mask,
struct vm_area_struct *vma, unsigned long addr,
- bool *new_page_allocated)
+ bool *new_page_allocated, bool fail_if_exists)
{
struct swap_info_struct *si;
struct folio *folio;
@@ -469,6 +469,15 @@ struct page *__read_swap_cache_async(swp
goto fail_put_swap;
/*
+ * This check guards against a state that happens if a call
+ * to __read_swap_cache_async triggers a reclaim, if the
+ * reclaimer (zswap's writeback as of now) then decides to
+ * reclaim that same entry, then the subsequent call to
+ * __read_swap_cache_async would get stuck in this loop.
+ */
+ if (fail_if_exists && err == -EEXIST)
+ goto fail_put_swap;
+ /*
* We might race against __delete_from_swap_cache(), and
* stumble across a swap_map entry whose SWAP_HAS_CACHE
* has not yet been cleared. Or race against another
@@ -530,7 +539,7 @@ struct page *read_swap_cache_async(swp_e
{
bool page_was_allocated;
struct page *retpage = __read_swap_cache_async(entry, gfp_mask,
- vma, addr, &page_was_allocated);
+ vma, addr, &page_was_allocated, false);
if (page_was_allocated)
swap_readpage(retpage, false, plug);
@@ -649,7 +658,7 @@ struct page *swap_cluster_readahead(swp_
/* Ok, do the async read-ahead now */
page = __read_swap_cache_async(
swp_entry(swp_type(entry), offset),
- gfp_mask, vma, addr, &page_allocated);
+ gfp_mask, vma, addr, &page_allocated, false);
if (!page)
continue;
if (page_allocated) {
@@ -815,7 +824,7 @@ static struct page *swap_vma_readahead(s
pte_unmap(pte);
pte = NULL;
page = __read_swap_cache_async(entry, gfp_mask, vma,
- addr, &page_allocated);
+ addr, &page_allocated, false);
if (!page)
continue;
if (page_allocated) {
--- a/mm/zswap.c~zswap-make-shrinking-memcg-aware
+++ a/mm/zswap.c
@@ -34,6 +34,7 @@
#include <linux/writeback.h>
#include <linux/pagemap.h>
#include <linux/workqueue.h>
+#include <linux/list_lru.h>
#include "swap.h"
#include "internal.h"
@@ -171,8 +172,8 @@ struct zswap_pool {
struct work_struct shrink_work;
struct hlist_node node;
char tfm_name[CRYPTO_MAX_ALG_NAME];
- struct list_head lru;
- spinlock_t lru_lock;
+ struct list_lru list_lru;
+ struct mem_cgroup *next_shrink;
};
/*
@@ -288,15 +289,25 @@ static void zswap_update_total_size(void
zswap_pool_total_size = total;
}
+static inline struct mem_cgroup *get_mem_cgroup_from_entry(struct zswap_entry *entry)
+{
+ return entry->objcg ? get_mem_cgroup_from_objcg(entry->objcg) : NULL;
+}
+
+static inline int entry_to_nid(struct zswap_entry *entry)
+{
+ return page_to_nid(virt_to_page(entry));
+}
+
/*********************************
* zswap entry functions
**********************************/
static struct kmem_cache *zswap_entry_cache;
-static struct zswap_entry *zswap_entry_cache_alloc(gfp_t gfp)
+static struct zswap_entry *zswap_entry_cache_alloc(gfp_t gfp, int nid)
{
struct zswap_entry *entry;
- entry = kmem_cache_alloc(zswap_entry_cache, gfp);
+ entry = kmem_cache_alloc_node(zswap_entry_cache, gfp, nid);
if (!entry)
return NULL;
entry->refcount = 1;
@@ -310,6 +321,27 @@ static void zswap_entry_cache_free(struc
}
/*********************************
+* lru functions
+**********************************/
+static bool zswap_lru_add(struct list_lru *list_lru, struct zswap_entry *entry)
+{
+ struct mem_cgroup *memcg = get_mem_cgroup_from_entry(entry);
+ bool added = __list_lru_add(list_lru, &entry->lru, entry_to_nid(entry), memcg);
+
+ mem_cgroup_put(memcg);
+ return added;
+}
+
+static bool zswap_lru_del(struct list_lru *list_lru, struct zswap_entry *entry)
+{
+ struct mem_cgroup *memcg = get_mem_cgroup_from_entry(entry);
+ bool removed = __list_lru_del(list_lru, &entry->lru, entry_to_nid(entry), memcg);
+
+ mem_cgroup_put(memcg);
+ return removed;
+}
+
+/*********************************
* rbtree functions
**********************************/
static struct zswap_entry *zswap_rb_search(struct rb_root *root, pgoff_t offset)
@@ -393,9 +425,7 @@ static void zswap_free_entry(struct zswa
if (!entry->length)
atomic_dec(&zswap_same_filled_pages);
else {
- spin_lock(&entry->pool->lru_lock);
- list_del(&entry->lru);
- spin_unlock(&entry->pool->lru_lock);
+ zswap_lru_del(&entry->pool->list_lru, entry);
zpool_free(zswap_find_zpool(entry), entry->handle);
zswap_pool_put(entry->pool);
}
@@ -629,21 +659,16 @@ static void zswap_invalidate_entry(struc
zswap_entry_put(tree, entry);
}
-static int zswap_reclaim_entry(struct zswap_pool *pool)
+static enum lru_status shrink_memcg_cb(struct list_head *item, struct list_lru_one *l,
+ spinlock_t *lock, void *arg)
{
- struct zswap_entry *entry;
+ struct zswap_entry *entry = container_of(item, struct zswap_entry, lru);
+ struct mem_cgroup *memcg;
struct zswap_tree *tree;
pgoff_t swpoffset;
- int ret;
+ enum lru_status ret = LRU_REMOVED_RETRY;
+ int writeback_result;
- /* Get an entry off the LRU */
- spin_lock(&pool->lru_lock);
- if (list_empty(&pool->lru)) {
- spin_unlock(&pool->lru_lock);
- return -EINVAL;
- }
- entry = list_last_entry(&pool->lru, struct zswap_entry, lru);
- list_del_init(&entry->lru);
/*
* Once the lru lock is dropped, the entry might get freed. The
* swpoffset is copied to the stack, and entry isn't deref'd again
@@ -651,28 +676,33 @@ static int zswap_reclaim_entry(struct zs
*/
swpoffset = swp_offset(entry->swpentry);
tree = zswap_trees[swp_type(entry->swpentry)];
- spin_unlock(&pool->lru_lock);
+ list_lru_isolate(l, item);
+ spin_unlock(lock);
/* Check for invalidate() race */
spin_lock(&tree->lock);
if (entry != zswap_rb_search(&tree->rbroot, swpoffset)) {
- ret = -EAGAIN;
goto unlock;
}
/* Hold a reference to prevent a free during writeback */
zswap_entry_get(entry);
spin_unlock(&tree->lock);
- ret = zswap_writeback_entry(entry, tree);
+ writeback_result = zswap_writeback_entry(entry, tree);
spin_lock(&tree->lock);
- if (ret) {
- /* Writeback failed, put entry back on LRU */
- spin_lock(&pool->lru_lock);
- list_move(&entry->lru, &pool->lru);
- spin_unlock(&pool->lru_lock);
+ if (writeback_result) {
+ zswap_reject_reclaim_fail++;
+ memcg = get_mem_cgroup_from_entry(entry);
+ spin_lock(lock);
+ /* we cannot use zswap_lru_add here, because it increments node's lru count */
+ list_lru_putback(&entry->pool->list_lru, item, entry_to_nid(entry), memcg);
+ spin_unlock(lock);
+ mem_cgroup_put(memcg);
+ ret = LRU_RETRY;
goto put_unlock;
}
+ zswap_written_back_pages++;
/*
* Writeback started successfully, the page now belongs to the
@@ -686,7 +716,36 @@ put_unlock:
zswap_entry_put(tree, entry);
unlock:
spin_unlock(&tree->lock);
- return ret ? -EAGAIN : 0;
+ spin_lock(lock);
+ return ret;
+}
+
+static int shrink_memcg(struct mem_cgroup *memcg)
+{
+ struct zswap_pool *pool;
+ int nid, shrunk = 0;
+
+ pool = zswap_pool_current_get();
+ if (!pool)
+ return -EINVAL;
+
+ /*
+ * Skip zombies because their LRUs are reparented and we would be
+ * reclaiming from the parent instead of the dead memcgroup.
+ */
+ if (memcg && !mem_cgroup_online(memcg))
+ goto out;
+
+ for_each_node_state(nid, N_NORMAL_MEMORY) {
+ unsigned long nr_to_walk = 1;
+
+ if (list_lru_walk_one(&pool->list_lru, nid, memcg, &shrink_memcg_cb,
+ NULL, &nr_to_walk))
+ shrunk++;
+ }
+out:
+ zswap_pool_put(pool);
+ return shrunk ? 0 : -EAGAIN;
}
static void shrink_worker(struct work_struct *w)
@@ -695,10 +754,13 @@ static void shrink_worker(struct work_st
shrink_work);
int ret, failures = 0;
+ /* global reclaim will select cgroup in a round-robin fashion. */
do {
- ret = zswap_reclaim_entry(pool);
+ pool->next_shrink = mem_cgroup_iter(NULL, pool->next_shrink, NULL);
+
+ ret = shrink_memcg(pool->next_shrink);
+
if (ret) {
- zswap_reject_reclaim_fail++;
if (ret != -EAGAIN)
break;
if (++failures == MAX_RECLAIM_RETRIES)
@@ -764,8 +826,7 @@ static struct zswap_pool *zswap_pool_cre
*/
kref_init(&pool->kref);
INIT_LIST_HEAD(&pool->list);
- INIT_LIST_HEAD(&pool->lru);
- spin_lock_init(&pool->lru_lock);
+ list_lru_init_memcg(&pool->list_lru, NULL);
INIT_WORK(&pool->shrink_work, shrink_worker);
zswap_pool_debug("created", pool);
@@ -831,6 +892,9 @@ static void zswap_pool_destroy(struct zs
cpuhp_state_remove_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node);
free_percpu(pool->acomp_ctx);
+ list_lru_destroy(&pool->list_lru);
+ if (pool->next_shrink)
+ mem_cgroup_put(pool->next_shrink);
for (i = 0; i < ZSWAP_NR_ZPOOLS; i++)
zpool_destroy_pool(pool->zpools[i]);
kfree(pool);
@@ -1076,7 +1140,7 @@ static int zswap_writeback_entry(struct
/* try to allocate swap cache page */
page = __read_swap_cache_async(swpentry, GFP_KERNEL, NULL, 0,
- &page_was_allocated);
+ &page_was_allocated, true);
if (!page) {
ret = -ENOMEM;
goto fail;
@@ -1142,7 +1206,6 @@ static int zswap_writeback_entry(struct
/* start writeback */
__swap_writepage(page, &wbc);
put_page(page);
- zswap_written_back_pages++;
return ret;
@@ -1199,8 +1262,10 @@ bool zswap_store(struct folio *folio)
struct scatterlist input, output;
struct crypto_acomp_ctx *acomp_ctx;
struct obj_cgroup *objcg = NULL;
+ struct mem_cgroup *memcg = NULL;
struct zswap_pool *pool;
struct zpool *zpool;
+ int lru_alloc_ret;
unsigned int dlen = PAGE_SIZE;
unsigned long handle, value;
char *buf;
@@ -1230,15 +1295,15 @@ bool zswap_store(struct folio *folio)
zswap_invalidate_entry(tree, dupentry);
}
spin_unlock(&tree->lock);
-
- /*
- * XXX: zswap reclaim does not work with cgroups yet. Without a
- * cgroup-aware entry LRU, we will push out entries system-wide based on
- * local cgroup limits.
- */
objcg = get_obj_cgroup_from_folio(folio);
- if (objcg && !obj_cgroup_may_zswap(objcg))
- goto reject;
+ if (objcg && !obj_cgroup_may_zswap(objcg)) {
+ memcg = get_mem_cgroup_from_objcg(objcg);
+ if (shrink_memcg(memcg)) {
+ mem_cgroup_put(memcg);
+ goto reject;
+ }
+ mem_cgroup_put(memcg);
+ }
/* reclaim space if needed */
if (zswap_is_full()) {
@@ -1254,10 +1319,15 @@ bool zswap_store(struct folio *folio)
zswap_pool_reached_full = false;
}
+ pool = zswap_pool_current_get();
+ if (!pool)
+ goto reject;
+
/* allocate entry */
- entry = zswap_entry_cache_alloc(GFP_KERNEL);
+ entry = zswap_entry_cache_alloc(GFP_KERNEL, page_to_nid(page));
if (!entry) {
zswap_reject_kmemcache_fail++;
+ zswap_pool_put(pool);
goto reject;
}
@@ -1269,6 +1339,7 @@ bool zswap_store(struct folio *folio)
entry->length = 0;
entry->value = value;
atomic_inc(&zswap_same_filled_pages);
+ zswap_pool_put(pool);
goto insert_entry;
}
kunmap_atomic(src);
@@ -1278,9 +1349,15 @@ bool zswap_store(struct folio *folio)
goto freepage;
/* if entry is successfully added, it keeps the reference */
- entry->pool = zswap_pool_current_get();
- if (!entry->pool)
- goto freepage;
+ entry->pool = pool;
+ if (objcg) {
+ memcg = get_mem_cgroup_from_objcg(objcg);
+ lru_alloc_ret = memcg_list_lru_alloc(memcg, &pool->list_lru, GFP_KERNEL);
+ mem_cgroup_put(memcg);
+
+ if (lru_alloc_ret)
+ goto freepage;
+ }
/* compress */
acomp_ctx = raw_cpu_ptr(entry->pool->acomp_ctx);
@@ -1358,9 +1435,8 @@ insert_entry:
zswap_invalidate_entry(tree, dupentry);
}
if (entry->length) {
- spin_lock(&entry->pool->lru_lock);
- list_add(&entry->lru, &entry->pool->lru);
- spin_unlock(&entry->pool->lru_lock);
+ INIT_LIST_HEAD(&entry->lru);
+ zswap_lru_add(&pool->list_lru, entry);
}
spin_unlock(&tree->lock);
@@ -1373,8 +1449,8 @@ insert_entry:
put_dstmem:
mutex_unlock(acomp_ctx->mutex);
- zswap_pool_put(entry->pool);
freepage:
+ zswap_pool_put(entry->pool);
zswap_entry_cache_free(entry);
reject:
if (objcg)
@@ -1467,9 +1543,8 @@ freeentry:
zswap_invalidate_entry(tree, entry);
folio_mark_dirty(folio);
} else if (entry->length) {
- spin_lock(&entry->pool->lru_lock);
- list_move(&entry->lru, &entry->pool->lru);
- spin_unlock(&entry->pool->lru_lock);
+ zswap_lru_del(&entry->pool->list_lru, entry);
+ zswap_lru_add(&entry->pool->list_lru, entry);
}
zswap_entry_put(tree, entry);
spin_unlock(&tree->lock);
_
Patches currently in -mm which might be from cerasuolodomenico@xxxxxxxxx are
zswap-make-shrinking-memcg-aware.patch
mm-memcg-add-per-memcg-zswap-writeback-stat.patch
selftests-cgroup-update-per-memcg-zswap-writeback-selftest.patch
