The patch titled
Subject: mm: multi-gen LRU: kill switch
has been added to the -mm mm-unstable branch. Its filename is
mm-multi-gen-lru-kill-switch.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-multi-gen-lru-kill-switch.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: Yu Zhao <yuzhao@xxxxxxxxxx>
Subject: mm: multi-gen LRU: kill switch
Date: Mon, 15 Aug 2022 01:13:29 -0600
Add /sys/kernel/mm/lru_gen/enabled as a kill switch. Components that
can be disabled include:
0x0001: the multi-gen LRU core
0x0002: walking page table, when arch_has_hw_pte_young() returns
true
0x0004: clearing the accessed bit in non-leaf PMD entries, when
CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG=y
[yYnN]: apply to all the components above
E.g.,
echo y >/sys/kernel/mm/lru_gen/enabled
cat /sys/kernel/mm/lru_gen/enabled
0x0007
echo 5 >/sys/kernel/mm/lru_gen/enabled
cat /sys/kernel/mm/lru_gen/enabled
0x0005
NB: the page table walks happen on the scale of seconds under heavy memory
pressure, in which case the mmap_lock contention is a lesser concern,
compared with the LRU lock contention and the I/O congestion. So far the
only well-known case of the mmap_lock contention happens on Android, due
to Scudo [1] which allocates several thousand VMAs for merely a few
hundred MBs. The SPF and the Maple Tree also have provided their own
assessments [2][3]. However, if walking page tables does worsen the
mmap_lock contention, the kill switch can be used to disable it. In this
case the multi-gen LRU will suffer a minor performance degradation, as
shown previously.
Clearing the accessed bit in non-leaf PMD entries can also be
disabled, since this behavior was not tested on x86 varieties other
than Intel and AMD.
[1] https://source.android.com/devices/tech/debug/scudo
[2] https://lore.kernel.org/r/20220128131006.67712-1-michel@xxxxxxxxxxxxxx/
[3] https://lore.kernel.org/r/20220426150616.3937571-1-Liam.Howlett@xxxxxxxxxx/
Link: https://lkml.kernel.org/r/20220815071332.627393-11-yuzhao@xxxxxxxxxx
Signed-off-by: Yu Zhao <yuzhao@xxxxxxxxxx>
Acked-by: Brian Geffon <bgeffon@xxxxxxxxxx>
Acked-by: Jan Alexander Steffens (heftig) <heftig@xxxxxxxxxxxxx>
Acked-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx>
Acked-by: Steven Barrett <steven@xxxxxxxxxxxx>
Acked-by: Suleiman Souhlal <suleiman@xxxxxxxxxx>
Tested-by: Daniel Byrne <djbyrne@xxxxxxx>
Tested-by: Donald Carr <d@xxxxxxxxxxxxxxx>
Tested-by: Holger Hoffstätte <holger@xxxxxxxxxxxxxxxxxxxxxx>
Tested-by: Konstantin Kharlamov <Hi-Angel@xxxxxxxxx>
Tested-by: Shuang Zhai <szhai2@xxxxxxxxxxxxxxxx>
Tested-by: Sofia Trinh <sofia.trinh@edi.works>
Tested-by: Vaibhav Jain <vaibhav@xxxxxxxxxxxxx>
Cc: Andi Kleen <ak@xxxxxxxxxxxxxxx>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@xxxxxxxxxxxxx>
Cc: Barry Song <baohua@xxxxxxxxxx>
Cc: Catalin Marinas <catalin.marinas@xxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>
Cc: Hillf Danton <hdanton@xxxxxxxx>
Cc: Jens Axboe <axboe@xxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Jonathan Corbet <corbet@xxxxxxx>
Cc: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Miaohe Lin <linmiaohe@xxxxxxxxxx>
Cc: Michael Larabel <Michael@xxxxxxxxxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Qi Zheng <zhengqi.arch@xxxxxxxxxxxxx>
Cc: Tejun Heo <tj@xxxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Will Deacon <will@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/cgroup.h | 15 +
include/linux/mm_inline.h | 15 +
include/linux/mmzone.h | 9 +
kernel/cgroup/cgroup-internal.h | 1
mm/Kconfig | 6
mm/vmscan.c | 231 +++++++++++++++++++++++++++++-
6 files changed, 268 insertions(+), 9 deletions(-)
--- a/include/linux/cgroup.h~mm-multi-gen-lru-kill-switch
+++ a/include/linux/cgroup.h
@@ -432,6 +432,18 @@ static inline void cgroup_put(struct cgr
css_put(&cgrp->self);
}
+extern struct mutex cgroup_mutex;
+
+static inline void cgroup_lock(void)
+{
+ mutex_lock(&cgroup_mutex);
+}
+
+static inline void cgroup_unlock(void)
+{
+ mutex_unlock(&cgroup_mutex);
+}
+
/**
* task_css_set_check - obtain a task's css_set with extra access conditions
* @task: the task to obtain css_set for
@@ -446,7 +458,6 @@ static inline void cgroup_put(struct cgr
* as locks used during the cgroup_subsys::attach() methods.
*/
#ifdef CONFIG_PROVE_RCU
-extern struct mutex cgroup_mutex;
extern spinlock_t css_set_lock;
#define task_css_set_check(task, __c) \
rcu_dereference_check((task)->cgroups, \
@@ -708,6 +719,8 @@ struct cgroup;
static inline u64 cgroup_id(const struct cgroup *cgrp) { return 1; }
static inline void css_get(struct cgroup_subsys_state *css) {}
static inline void css_put(struct cgroup_subsys_state *css) {}
+static inline void cgroup_lock(void) {}
+static inline void cgroup_unlock(void) {}
static inline int cgroup_attach_task_all(struct task_struct *from,
struct task_struct *t) { return 0; }
static inline int cgroupstats_build(struct cgroupstats *stats,
--- a/include/linux/mm_inline.h~mm-multi-gen-lru-kill-switch
+++ a/include/linux/mm_inline.h
@@ -106,10 +106,21 @@ static __always_inline enum lru_list fol
#ifdef CONFIG_LRU_GEN
+#ifdef CONFIG_LRU_GEN_ENABLED
static inline bool lru_gen_enabled(void)
{
- return true;
+ DECLARE_STATIC_KEY_TRUE(lru_gen_caps[NR_LRU_GEN_CAPS]);
+
+ return static_branch_likely(&lru_gen_caps[LRU_GEN_CORE]);
+}
+#else
+static inline bool lru_gen_enabled(void)
+{
+ DECLARE_STATIC_KEY_FALSE(lru_gen_caps[NR_LRU_GEN_CAPS]);
+
+ return static_branch_unlikely(&lru_gen_caps[LRU_GEN_CORE]);
}
+#endif
static inline bool lru_gen_in_fault(void)
{
@@ -222,7 +233,7 @@ static inline bool lru_gen_add_folio(str
VM_WARN_ON_ONCE_FOLIO(gen != -1, folio);
- if (folio_test_unevictable(folio))
+ if (folio_test_unevictable(folio) || !lrugen->enabled)
return false;
/*
* There are three common cases for this page:
--- a/include/linux/mmzone.h~mm-multi-gen-lru-kill-switch
+++ a/include/linux/mmzone.h
@@ -387,6 +387,13 @@ enum {
LRU_GEN_FILE,
};
+enum {
+ LRU_GEN_CORE,
+ LRU_GEN_MM_WALK,
+ LRU_GEN_NONLEAF_YOUNG,
+ NR_LRU_GEN_CAPS
+};
+
#define MIN_LRU_BATCH BITS_PER_LONG
#define MAX_LRU_BATCH (MIN_LRU_BATCH * 128)
@@ -428,6 +435,8 @@ struct lru_gen_struct {
/* can be modified without holding the LRU lock */
atomic_long_t evicted[NR_HIST_GENS][ANON_AND_FILE][MAX_NR_TIERS];
atomic_long_t refaulted[NR_HIST_GENS][ANON_AND_FILE][MAX_NR_TIERS];
+ /* whether the multi-gen LRU is enabled */
+ bool enabled;
};
enum {
--- a/kernel/cgroup/cgroup-internal.h~mm-multi-gen-lru-kill-switch
+++ a/kernel/cgroup/cgroup-internal.h
@@ -164,7 +164,6 @@ struct cgroup_mgctx {
#define DEFINE_CGROUP_MGCTX(name) \
struct cgroup_mgctx name = CGROUP_MGCTX_INIT(name)
-extern struct mutex cgroup_mutex;
extern spinlock_t css_set_lock;
extern struct cgroup_subsys *cgroup_subsys[];
extern struct list_head cgroup_roots;
--- a/mm/Kconfig~mm-multi-gen-lru-kill-switch
+++ a/mm/Kconfig
@@ -1133,6 +1133,12 @@ config LRU_GEN
help
A high performance LRU implementation to overcommit memory.
+config LRU_GEN_ENABLED
+ bool "Enable by default"
+ depends on LRU_GEN
+ help
+ This option enables the multi-gen LRU by default.
+
config LRU_GEN_STATS
bool "Full stats for debugging"
depends on LRU_GEN
--- a/mm/vmscan.c~mm-multi-gen-lru-kill-switch
+++ a/mm/vmscan.c
@@ -51,6 +51,7 @@
#include <linux/psi.h>
#include <linux/pagewalk.h>
#include <linux/shmem_fs.h>
+#include <linux/ctype.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
@@ -3071,6 +3072,14 @@ static bool can_age_anon_pages(struct pg
#ifdef CONFIG_LRU_GEN
+#ifdef CONFIG_LRU_GEN_ENABLED
+DEFINE_STATIC_KEY_ARRAY_TRUE(lru_gen_caps, NR_LRU_GEN_CAPS);
+#define get_cap(cap) static_branch_likely(&lru_gen_caps[cap])
+#else
+DEFINE_STATIC_KEY_ARRAY_FALSE(lru_gen_caps, NR_LRU_GEN_CAPS);
+#define get_cap(cap) static_branch_unlikely(&lru_gen_caps[cap])
+#endif
+
/******************************************************************************
* shorthand helpers
******************************************************************************/
@@ -3948,7 +3957,8 @@ static void walk_pmd_range_locked(pud_t
goto next;
if (!pmd_trans_huge(pmd[i])) {
- if (IS_ENABLED(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG))
+ if (IS_ENABLED(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG) &&
+ get_cap(LRU_GEN_NONLEAF_YOUNG))
pmdp_test_and_clear_young(vma, addr, pmd + i);
goto next;
}
@@ -4046,10 +4056,12 @@ restart:
walk->mm_stats[MM_NONLEAF_TOTAL]++;
#ifdef CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG
- if (!pmd_young(val))
- continue;
+ if (get_cap(LRU_GEN_NONLEAF_YOUNG)) {
+ if (!pmd_young(val))
+ continue;
- walk_pmd_range_locked(pud, addr, vma, args, bitmap, &pos);
+ walk_pmd_range_locked(pud, addr, vma, args, bitmap, &pos);
+ }
#endif
if (!walk->force_scan && !test_bloom_filter(walk->lruvec, walk->max_seq, pmd + i))
continue;
@@ -4314,7 +4326,7 @@ static bool try_to_inc_max_seq(struct lr
* handful of PTEs. Spreading the work out over a period of time usually
* is less efficient, but it avoids bursty page faults.
*/
- if (!arch_has_hw_pte_young()) {
+ if (!(arch_has_hw_pte_young() && get_cap(LRU_GEN_MM_WALK))) {
success = iterate_mm_list_nowalk(lruvec, max_seq);
goto done;
}
@@ -5051,6 +5063,211 @@ done:
}
/******************************************************************************
+ * state change
+ ******************************************************************************/
+
+static bool __maybe_unused state_is_valid(struct lruvec *lruvec)
+{
+ struct lru_gen_struct *lrugen = &lruvec->lrugen;
+
+ if (lrugen->enabled) {
+ enum lru_list lru;
+
+ for_each_evictable_lru(lru) {
+ if (!list_empty(&lruvec->lists[lru]))
+ return false;
+ }
+ } else {
+ int gen, type, zone;
+
+ for_each_gen_type_zone(gen, type, zone) {
+ if (!list_empty(&lrugen->lists[gen][type][zone]))
+ return false;
+
+ /* unlikely but not a bug when reset_batch_size() is pending */
+ VM_WARN_ON_ONCE(lrugen->nr_pages[gen][type][zone]);
+ }
+ }
+
+ return true;
+}
+
+static bool fill_evictable(struct lruvec *lruvec)
+{
+ enum lru_list lru;
+ int remaining = MAX_LRU_BATCH;
+
+ for_each_evictable_lru(lru) {
+ int type = is_file_lru(lru);
+ bool active = is_active_lru(lru);
+ struct list_head *head = &lruvec->lists[lru];
+
+ while (!list_empty(head)) {
+ bool success;
+ struct folio *folio = lru_to_folio(head);
+
+ VM_WARN_ON_ONCE_FOLIO(folio_test_unevictable(folio), folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_test_active(folio) != active, folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_is_file_lru(folio) != type, folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_lru_gen(folio) != -1, folio);
+
+ lruvec_del_folio(lruvec, folio);
+ success = lru_gen_add_folio(lruvec, folio, false);
+ VM_WARN_ON_ONCE(!success);
+
+ if (!--remaining)
+ return false;
+ }
+ }
+
+ return true;
+}
+
+static bool drain_evictable(struct lruvec *lruvec)
+{
+ int gen, type, zone;
+ int remaining = MAX_LRU_BATCH;
+
+ for_each_gen_type_zone(gen, type, zone) {
+ struct list_head *head = &lruvec->lrugen.lists[gen][type][zone];
+
+ while (!list_empty(head)) {
+ bool success;
+ struct folio *folio = lru_to_folio(head);
+
+ VM_WARN_ON_ONCE_FOLIO(folio_test_unevictable(folio), folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_test_active(folio), folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_is_file_lru(folio) != type, folio);
+ VM_WARN_ON_ONCE_FOLIO(folio_zonenum(folio) != zone, folio);
+
+ success = lru_gen_del_folio(lruvec, folio, false);
+ VM_WARN_ON_ONCE(!success);
+ lruvec_add_folio(lruvec, folio);
+
+ if (!--remaining)
+ return false;
+ }
+ }
+
+ return true;
+}
+
+static void lru_gen_change_state(bool enabled)
+{
+ static DEFINE_MUTEX(state_mutex);
+
+ struct mem_cgroup *memcg;
+
+ cgroup_lock();
+ cpus_read_lock();
+ get_online_mems();
+ mutex_lock(&state_mutex);
+
+ if (enabled == lru_gen_enabled())
+ goto unlock;
+
+ if (enabled)
+ static_branch_enable_cpuslocked(&lru_gen_caps[LRU_GEN_CORE]);
+ else
+ static_branch_disable_cpuslocked(&lru_gen_caps[LRU_GEN_CORE]);
+
+ memcg = mem_cgroup_iter(NULL, NULL, NULL);
+ do {
+ int nid;
+
+ for_each_node(nid) {
+ struct lruvec *lruvec = get_lruvec(memcg, nid);
+
+ if (!lruvec)
+ continue;
+
+ spin_lock_irq(&lruvec->lru_lock);
+
+ VM_WARN_ON_ONCE(!seq_is_valid(lruvec));
+ VM_WARN_ON_ONCE(!state_is_valid(lruvec));
+
+ lruvec->lrugen.enabled = enabled;
+
+ while (!(enabled ? fill_evictable(lruvec) : drain_evictable(lruvec))) {
+ spin_unlock_irq(&lruvec->lru_lock);
+ cond_resched();
+ spin_lock_irq(&lruvec->lru_lock);
+ }
+
+ spin_unlock_irq(&lruvec->lru_lock);
+ }
+
+ cond_resched();
+ } while ((memcg = mem_cgroup_iter(NULL, memcg, NULL)));
+unlock:
+ mutex_unlock(&state_mutex);
+ put_online_mems();
+ cpus_read_unlock();
+ cgroup_unlock();
+}
+
+/******************************************************************************
+ * sysfs interface
+ ******************************************************************************/
+
+static ssize_t show_enabled(struct kobject *kobj, struct kobj_attribute *attr, char *buf)
+{
+ unsigned int caps = 0;
+
+ if (get_cap(LRU_GEN_CORE))
+ caps |= BIT(LRU_GEN_CORE);
+
+ if (arch_has_hw_pte_young() && get_cap(LRU_GEN_MM_WALK))
+ caps |= BIT(LRU_GEN_MM_WALK);
+
+ if (IS_ENABLED(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG) && get_cap(LRU_GEN_NONLEAF_YOUNG))
+ caps |= BIT(LRU_GEN_NONLEAF_YOUNG);
+
+ return snprintf(buf, PAGE_SIZE, "0x%04x\n", caps);
+}
+
+static ssize_t store_enabled(struct kobject *kobj, struct kobj_attribute *attr,
+ const char *buf, size_t len)
+{
+ int i;
+ unsigned int caps;
+
+ if (tolower(*buf) == 'n')
+ caps = 0;
+ else if (tolower(*buf) == 'y')
+ caps = -1;
+ else if (kstrtouint(buf, 0, &caps))
+ return -EINVAL;
+
+ for (i = 0; i < NR_LRU_GEN_CAPS; i++) {
+ bool enabled = caps & BIT(i);
+
+ if (i == LRU_GEN_CORE)
+ lru_gen_change_state(enabled);
+ else if (enabled)
+ static_branch_enable(&lru_gen_caps[i]);
+ else
+ static_branch_disable(&lru_gen_caps[i]);
+ }
+
+ return len;
+}
+
+static struct kobj_attribute lru_gen_enabled_attr = __ATTR(
+ enabled, 0644, show_enabled, store_enabled
+);
+
+static struct attribute *lru_gen_attrs[] = {
+ &lru_gen_enabled_attr.attr,
+ NULL
+};
+
+static struct attribute_group lru_gen_attr_group = {
+ .name = "lru_gen",
+ .attrs = lru_gen_attrs,
+};
+
+/******************************************************************************
* initialization
******************************************************************************/
@@ -5060,6 +5277,7 @@ void lru_gen_init_lruvec(struct lruvec *
struct lru_gen_struct *lrugen = &lruvec->lrugen;
lrugen->max_seq = MIN_NR_GENS + 1;
+ lrugen->enabled = lru_gen_enabled();
for_each_gen_type_zone(gen, type, zone)
INIT_LIST_HEAD(&lrugen->lists[gen][type][zone]);
@@ -5099,6 +5317,9 @@ static int __init init_lru_gen(void)
BUILD_BUG_ON(MIN_NR_GENS + 1 >= MAX_NR_GENS);
BUILD_BUG_ON(BIT(LRU_GEN_WIDTH) <= MAX_NR_GENS);
+ if (sysfs_create_group(mm_kobj, &lru_gen_attr_group))
+ pr_err("lru_gen: failed to create sysfs group\n");
+
return 0;
};
late_initcall(init_lru_gen);
_
Patches currently in -mm which might be from yuzhao@xxxxxxxxxx are
mm-x86-arm64-add-arch_has_hw_pte_young.patch
mm-x86-add-config_arch_has_nonleaf_pmd_young.patch
mm-vmscanc-refactor-shrink_node.patch
revert-include-linux-mm_inlineh-fold-__update_lru_size-into-its-sole-caller.patch
mm-multi-gen-lru-groundwork.patch
mm-multi-gen-lru-minimal-implementation.patch
mm-multi-gen-lru-exploit-locality-in-rmap.patch
mm-multi-gen-lru-support-page-table-walks.patch
mm-multi-gen-lru-optimize-multiple-memcgs.patch
mm-multi-gen-lru-kill-switch.patch
mm-multi-gen-lru-thrashing-prevention.patch
mm-multi-gen-lru-debugfs-interface.patch
mm-multi-gen-lru-admin-guide.patch
mm-multi-gen-lru-design-doc.patch
