The quilt patch titled
Subject: zswap: memcg accounting
has been removed from the -mm tree. Its filename was
zswap-memcg-accounting.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Johannes Weiner <hannes@xxxxxxxxxxx>
Subject: zswap: memcg accounting
Applications can currently escape their cgroup memory containment when
zswap is enabled. This patch adds per-cgroup tracking and limiting of
zswap backend memory to rectify this.
The existing cgroup2 memory.stat file is extended to show zswap statistics
analogous to what's in meminfo and vmstat. Furthermore, two new control
files, memory.zswap.current and memory.zswap.max, are added to allow
tuning zswap usage on a per-workload basis. This is important since not
all workloads benefit from zswap equally; some even suffer compared to
disk swap when memory contents don't compress well. The optimal size of
the zswap pool, and the threshold for writeback, also depends on the size
of the workload's warm set.
The implementation doesn't use a traditional page_counter transaction.
zswap is unconventional as a memory consumer in that we only know the
amount of memory to charge once expensive compression has occurred. If
zwap is disabled or the limit is already exceeded we obviously don't want
to compress page upon page only to reject them all. Instead, the limit is
checked against current usage, then we compress and charge. This allows
some limit overrun, but not enough to matter in practice.
[hannes@xxxxxxxxxxx: fix for CONFIG_SLOB builds]
Link: https://lkml.kernel.org/r/YnwD14zxYjUJPc2w@xxxxxxxxxxx
[hannes@xxxxxxxxxxx: opt out of cgroups v1]
Link: https://lkml.kernel.org/r/Yn6it9mBYFA+/lTb@xxxxxxxxxxx
Link: https://lkml.kernel.org/r/20220510152847.230957-7-hannes@xxxxxxxxxxx
Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxx>
Cc: Roman Gushchin <guro@xxxxxx>
Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
Cc: Seth Jennings <sjenning@xxxxxxxxxx>
Cc: Dan Streetman <ddstreet@xxxxxxxx>
Cc: Minchan Kim <minchan@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
Documentation/admin-guide/cgroup-v2.rst | 21 ++
include/linux/memcontrol.h | 54 +++++
mm/memcontrol.c | 205 +++++++++++++++++++++-
mm/zswap.c | 37 +++
4 files changed, 302 insertions(+), 15 deletions(-)
--- a/Documentation/admin-guide/cgroup-v2.rst~zswap-memcg-accounting
+++ a/Documentation/admin-guide/cgroup-v2.rst
@@ -1354,6 +1354,12 @@ PAGE_SIZE multiple when read back.
Amount of cached filesystem data that is swap-backed,
such as tmpfs, shm segments, shared anonymous mmap()s
+ zswap
+ Amount of memory consumed by the zswap compression backend.
+
+ zswapped
+ Amount of application memory swapped out to zswap.
+
file_mapped
Amount of cached filesystem data mapped with mmap()
@@ -1544,6 +1550,21 @@ PAGE_SIZE multiple when read back.
higher than the limit for an extended period of time. This
reduces the impact on the workload and memory management.
+ memory.zswap.current
+ A read-only single value file which exists on non-root
+ cgroups.
+
+ The total amount of memory consumed by the zswap compression
+ backend.
+
+ memory.zswap.max
+ A read-write single value file which exists on non-root
+ cgroups. The default is "max".
+
+ Zswap usage hard limit. If a cgroup's zswap pool reaches this
+ limit, it will refuse to take any more stores before existing
+ entries fault back in or are written out to disk.
+
memory.pressure
A read-only nested-keyed file.
--- a/include/linux/memcontrol.h~zswap-memcg-accounting
+++ a/include/linux/memcontrol.h
@@ -35,6 +35,8 @@ enum memcg_stat_item {
MEMCG_PERCPU_B,
MEMCG_VMALLOC,
MEMCG_KMEM,
+ MEMCG_ZSWAP_B,
+ MEMCG_ZSWAPPED,
MEMCG_NR_STAT,
};
@@ -252,6 +254,10 @@ struct mem_cgroup {
/* Range enforcement for interrupt charges */
struct work_struct high_work;
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+ unsigned long zswap_max;
+#endif
+
unsigned long soft_limit;
/* vmpressure notifications */
@@ -1273,6 +1279,10 @@ struct mem_cgroup *mem_cgroup_from_css(s
return NULL;
}
+static inline void obj_cgroup_put(struct obj_cgroup *objcg)
+{
+}
+
static inline void mem_cgroup_put(struct mem_cgroup *memcg)
{
}
@@ -1694,6 +1704,7 @@ int __memcg_kmem_charge_page(struct page
void __memcg_kmem_uncharge_page(struct page *page, int order);
struct obj_cgroup *get_obj_cgroup_from_current(void);
+struct obj_cgroup *get_obj_cgroup_from_page(struct page *page);
int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size);
void obj_cgroup_uncharge(struct obj_cgroup *objcg, size_t size);
@@ -1730,6 +1741,20 @@ static inline int memcg_kmem_id(struct m
struct mem_cgroup *mem_cgroup_from_obj(void *p);
+static inline void count_objcg_event(struct obj_cgroup *objcg,
+ enum vm_event_item idx)
+{
+ struct mem_cgroup *memcg;
+
+ if (mem_cgroup_kmem_disabled())
+ return;
+
+ rcu_read_lock();
+ memcg = obj_cgroup_memcg(objcg);
+ count_memcg_events(memcg, idx, 1);
+ rcu_read_unlock();
+}
+
#else
static inline bool mem_cgroup_kmem_disabled(void)
{
@@ -1756,6 +1781,11 @@ static inline void __memcg_kmem_uncharge
{
}
+static inline struct obj_cgroup *get_obj_cgroup_from_page(struct page *page)
+{
+ return NULL;
+}
+
static inline bool memcg_kmem_enabled(void)
{
return false;
@@ -1771,6 +1801,30 @@ static inline struct mem_cgroup *mem_cgr
return NULL;
}
+static inline void count_objcg_event(struct obj_cgroup *objcg,
+ enum vm_event_item idx)
+{
+}
+
#endif /* CONFIG_MEMCG_KMEM */
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+bool obj_cgroup_may_zswap(struct obj_cgroup *objcg);
+void obj_cgroup_charge_zswap(struct obj_cgroup *objcg, size_t size);
+void obj_cgroup_uncharge_zswap(struct obj_cgroup *objcg, size_t size);
+#else
+static inline bool obj_cgroup_may_zswap(struct obj_cgroup *objcg)
+{
+ return true;
+}
+static inline void obj_cgroup_charge_zswap(struct obj_cgroup *objcg,
+ size_t size)
+{
+}
+static inline void obj_cgroup_uncharge_zswap(struct obj_cgroup *objcg,
+ size_t size)
+{
+}
+#endif
+
#endif /* _LINUX_MEMCONTROL_H */
--- a/mm/memcontrol.c~zswap-memcg-accounting
+++ a/mm/memcontrol.c
@@ -1398,6 +1398,10 @@ static const struct memory_stat memory_s
{ "sock", MEMCG_SOCK },
{ "vmalloc", MEMCG_VMALLOC },
{ "shmem", NR_SHMEM },
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+ { "zswap", MEMCG_ZSWAP_B },
+ { "zswapped", MEMCG_ZSWAPPED },
+#endif
{ "file_mapped", NR_FILE_MAPPED },
{ "file_dirty", NR_FILE_DIRTY },
{ "file_writeback", NR_WRITEBACK },
@@ -1432,6 +1436,7 @@ static int memcg_page_state_unit(int ite
{
switch (item) {
case MEMCG_PERCPU_B:
+ case MEMCG_ZSWAP_B:
case NR_SLAB_RECLAIMABLE_B:
case NR_SLAB_UNRECLAIMABLE_B:
case WORKINGSET_REFAULT_ANON:
@@ -1512,6 +1517,13 @@ static char *memory_stat_format(struct m
seq_buf_printf(&s, "%s %lu\n", vm_event_name(PGLAZYFREED),
memcg_events(memcg, PGLAZYFREED));
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+ seq_buf_printf(&s, "%s %lu\n", vm_event_name(ZSWPIN),
+ memcg_events(memcg, ZSWPIN));
+ seq_buf_printf(&s, "%s %lu\n", vm_event_name(ZSWPOUT),
+ memcg_events(memcg, ZSWPOUT));
+#endif
+
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
seq_buf_printf(&s, "%s %lu\n", vm_event_name(THP_FAULT_ALLOC),
memcg_events(memcg, THP_FAULT_ALLOC));
@@ -2883,6 +2895,19 @@ struct mem_cgroup *mem_cgroup_from_obj(v
return page_memcg_check(folio_page(folio, 0));
}
+static struct obj_cgroup *__get_obj_cgroup_from_memcg(struct mem_cgroup *memcg)
+{
+ struct obj_cgroup *objcg = NULL;
+
+ for (; memcg != root_mem_cgroup; memcg = parent_mem_cgroup(memcg)) {
+ objcg = rcu_dereference(memcg->objcg);
+ if (objcg && obj_cgroup_tryget(objcg))
+ break;
+ objcg = NULL;
+ }
+ return objcg;
+}
+
__always_inline struct obj_cgroup *get_obj_cgroup_from_current(void)
{
struct obj_cgroup *objcg = NULL;
@@ -2896,15 +2921,32 @@ __always_inline struct obj_cgroup *get_o
memcg = active_memcg();
else
memcg = mem_cgroup_from_task(current);
-
- for (; memcg != root_mem_cgroup; memcg = parent_mem_cgroup(memcg)) {
- objcg = rcu_dereference(memcg->objcg);
- if (objcg && obj_cgroup_tryget(objcg))
- break;
- objcg = NULL;
- }
+ objcg = __get_obj_cgroup_from_memcg(memcg);
rcu_read_unlock();
+ return objcg;
+}
+
+struct obj_cgroup *get_obj_cgroup_from_page(struct page *page)
+{
+ struct obj_cgroup *objcg;
+
+ if (!memcg_kmem_enabled() || memcg_kmem_bypass())
+ return NULL;
+
+ if (PageMemcgKmem(page)) {
+ objcg = __folio_objcg(page_folio(page));
+ obj_cgroup_get(objcg);
+ } else {
+ struct mem_cgroup *memcg;
+ rcu_read_lock();
+ memcg = __folio_memcg(page_folio(page));
+ if (memcg)
+ objcg = __get_obj_cgroup_from_memcg(memcg);
+ else
+ objcg = NULL;
+ rcu_read_unlock();
+ }
return objcg;
}
@@ -5142,6 +5184,9 @@ mem_cgroup_css_alloc(struct cgroup_subsy
page_counter_set_high(&memcg->memory, PAGE_COUNTER_MAX);
memcg->soft_limit = PAGE_COUNTER_MAX;
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+ memcg->zswap_max = PAGE_COUNTER_MAX;
+#endif
page_counter_set_high(&memcg->swap, PAGE_COUNTER_MAX);
if (parent) {
memcg->swappiness = mem_cgroup_swappiness(parent);
@@ -7421,6 +7466,148 @@ static struct cftype memsw_files[] = {
{ }, /* terminate */
};
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+/**
+ * obj_cgroup_may_zswap - check if this cgroup can zswap
+ * @objcg: the object cgroup
+ *
+ * Check if the hierarchical zswap limit has been reached.
+ *
+ * This doesn't check for specific headroom, and it is not atomic
+ * either. But with zswap, the size of the allocation is only known
+ * once compression has occured, and this optimistic pre-check avoids
+ * spending cycles on compression when there is already no room left
+ * or zswap is disabled altogether somewhere in the hierarchy.
+ */
+bool obj_cgroup_may_zswap(struct obj_cgroup *objcg)
+{
+ struct mem_cgroup *memcg, *original_memcg;
+ bool ret = true;
+
+ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+ return true;
+
+ original_memcg = get_mem_cgroup_from_objcg(objcg);
+ for (memcg = original_memcg; memcg != root_mem_cgroup;
+ memcg = parent_mem_cgroup(memcg)) {
+ unsigned long max = READ_ONCE(memcg->zswap_max);
+ unsigned long pages;
+
+ if (max == PAGE_COUNTER_MAX)
+ continue;
+ if (max == 0) {
+ ret = false;
+ break;
+ }
+
+ cgroup_rstat_flush(memcg->css.cgroup);
+ pages = memcg_page_state(memcg, MEMCG_ZSWAP_B) / PAGE_SIZE;
+ if (pages < max)
+ continue;
+ ret = false;
+ break;
+ }
+ mem_cgroup_put(original_memcg);
+ return ret;
+}
+
+/**
+ * obj_cgroup_charge_zswap - charge compression backend memory
+ * @objcg: the object cgroup
+ * @size: size of compressed object
+ *
+ * This forces the charge after obj_cgroup_may_swap() allowed
+ * compression and storage in zwap for this cgroup to go ahead.
+ */
+void obj_cgroup_charge_zswap(struct obj_cgroup *objcg, size_t size)
+{
+ struct mem_cgroup *memcg;
+
+ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+ return;
+
+ VM_WARN_ON_ONCE(!(current->flags & PF_MEMALLOC));
+
+ /* PF_MEMALLOC context, charging must succeed */
+ if (obj_cgroup_charge(objcg, GFP_KERNEL, size))
+ VM_WARN_ON_ONCE(1);
+
+ rcu_read_lock();
+ memcg = obj_cgroup_memcg(objcg);
+ mod_memcg_state(memcg, MEMCG_ZSWAP_B, size);
+ mod_memcg_state(memcg, MEMCG_ZSWAPPED, 1);
+ rcu_read_unlock();
+}
+
+/**
+ * obj_cgroup_uncharge_zswap - uncharge compression backend memory
+ * @objcg: the object cgroup
+ * @size: size of compressed object
+ *
+ * Uncharges zswap memory on page in.
+ */
+void obj_cgroup_uncharge_zswap(struct obj_cgroup *objcg, size_t size)
+{
+ struct mem_cgroup *memcg;
+
+ if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+ return;
+
+ obj_cgroup_uncharge(objcg, size);
+
+ rcu_read_lock();
+ memcg = obj_cgroup_memcg(objcg);
+ mod_memcg_state(memcg, MEMCG_ZSWAP_B, -size);
+ mod_memcg_state(memcg, MEMCG_ZSWAPPED, -1);
+ rcu_read_unlock();
+}
+
+static u64 zswap_current_read(struct cgroup_subsys_state *css,
+ struct cftype *cft)
+{
+ cgroup_rstat_flush(css->cgroup);
+ return memcg_page_state(mem_cgroup_from_css(css), MEMCG_ZSWAP_B);
+}
+
+static int zswap_max_show(struct seq_file *m, void *v)
+{
+ return seq_puts_memcg_tunable(m,
+ READ_ONCE(mem_cgroup_from_seq(m)->zswap_max));
+}
+
+static ssize_t zswap_max_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ struct mem_cgroup *memcg = mem_cgroup_from_css(of_css(of));
+ unsigned long max;
+ int err;
+
+ buf = strstrip(buf);
+ err = page_counter_memparse(buf, "max", &max);
+ if (err)
+ return err;
+
+ xchg(&memcg->zswap_max, max);
+
+ return nbytes;
+}
+
+static struct cftype zswap_files[] = {
+ {
+ .name = "zswap.current",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .read_u64 = zswap_current_read,
+ },
+ {
+ .name = "zswap.max",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .seq_show = zswap_max_show,
+ .write = zswap_max_write,
+ },
+ { } /* terminate */
+};
+#endif /* CONFIG_MEMCG_KMEM && CONFIG_ZSWAP */
+
/*
* If mem_cgroup_swap_init() is implemented as a subsys_initcall()
* instead of a core_initcall(), this could mean cgroup_memory_noswap still
@@ -7439,7 +7626,9 @@ static int __init mem_cgroup_swap_init(v
WARN_ON(cgroup_add_dfl_cftypes(&memory_cgrp_subsys, swap_files));
WARN_ON(cgroup_add_legacy_cftypes(&memory_cgrp_subsys, memsw_files));
-
+#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
+ WARN_ON(cgroup_add_dfl_cftypes(&memory_cgrp_subsys, zswap_files));
+#endif
return 0;
}
core_initcall(mem_cgroup_swap_init);
--- a/mm/zswap.c~zswap-memcg-accounting
+++ a/mm/zswap.c
@@ -188,6 +188,7 @@ struct zswap_entry {
unsigned long handle;
unsigned long value;
};
+ struct obj_cgroup *objcg;
};
struct zswap_header {
@@ -359,6 +360,10 @@ static void zswap_rb_erase(struct rb_roo
*/
static void zswap_free_entry(struct zswap_entry *entry)
{
+ if (entry->objcg) {
+ obj_cgroup_uncharge_zswap(entry->objcg, entry->length);
+ obj_cgroup_put(entry->objcg);
+ }
if (!entry->length)
atomic_dec(&zswap_same_filled_pages);
else {
@@ -1096,6 +1101,8 @@ static int zswap_frontswap_store(unsigne
struct zswap_entry *entry, *dupentry;
struct scatterlist input, output;
struct crypto_acomp_ctx *acomp_ctx;
+ struct obj_cgroup *objcg = NULL;
+ struct zswap_pool *pool;
int ret;
unsigned int hlen, dlen = PAGE_SIZE;
unsigned long handle, value;
@@ -1115,17 +1122,15 @@ static int zswap_frontswap_store(unsigne
goto reject;
}
+ objcg = get_obj_cgroup_from_page(page);
+ if (objcg && !obj_cgroup_may_zswap(objcg))
+ goto shrink;
+
/* reclaim space if needed */
if (zswap_is_full()) {
- struct zswap_pool *pool;
-
zswap_pool_limit_hit++;
zswap_pool_reached_full = true;
- pool = zswap_pool_last_get();
- if (pool)
- queue_work(shrink_wq, &pool->shrink_work);
- ret = -ENOMEM;
- goto reject;
+ goto shrink;
}
if (zswap_pool_reached_full) {
@@ -1227,6 +1232,13 @@ static int zswap_frontswap_store(unsigne
entry->length = dlen;
insert_entry:
+ entry->objcg = objcg;
+ if (objcg) {
+ obj_cgroup_charge_zswap(objcg, entry->length);
+ /* Account before objcg ref is moved to tree */
+ count_objcg_event(objcg, ZSWPOUT);
+ }
+
/* map */
spin_lock(&tree->lock);
do {
@@ -1253,7 +1265,16 @@ put_dstmem:
freepage:
zswap_entry_cache_free(entry);
reject:
+ if (objcg)
+ obj_cgroup_put(objcg);
return ret;
+
+shrink:
+ pool = zswap_pool_last_get();
+ if (pool)
+ queue_work(shrink_wq, &pool->shrink_work);
+ ret = -ENOMEM;
+ goto reject;
}
/*
@@ -1326,6 +1347,8 @@ static int zswap_frontswap_load(unsigned
BUG_ON(ret);
stats:
count_vm_event(ZSWPIN);
+ if (entry->objcg)
+ count_objcg_event(entry->objcg, ZSWPIN);
freeentry:
spin_lock(&tree->lock);
zswap_entry_put(tree, entry);
_
Patches currently in -mm which might be from hannes@xxxxxxxxxxx are
revert-mm-vmscan-never-demote-for-memcg-reclaim.patch
