The patch titled
Subject: mm/demotion: add pg_data_t member to track node memory tier details
has been added to the -mm mm-unstable branch. Its filename is
mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details.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: "Aneesh Kumar K.V" <aneesh.kumar@xxxxxxxxxxxxx>
Subject: mm/demotion: add pg_data_t member to track node memory tier details
Date: Thu, 18 Aug 2022 18:40:38 +0530
Also update different helpes to use NODE_DATA()->memtier. Since node
specific memtier can change based on the reassignment of NUMA node to a
different memory tiers, accessing NODE_DATA()->memtier needs to happen
under an rcu read lock or memory_tier_lock.
Link: https://lkml.kernel.org/r/20220818131042.113280-7-aneesh.kumar@xxxxxxxxxxxxx
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxx>
Reviewed-by: "Huang, Ying" <ying.huang@xxxxxxxxx>
Cc: Alistair Popple <apopple@xxxxxxxxxx>
Cc: Bharata B Rao <bharata@xxxxxxx>
Cc: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxx>
Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx>
Cc: Hesham Almatary <hesham.almatary@xxxxxxxxxx>
Cc: Jagdish Gediya <jvgediya.oss@xxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: Tim Chen <tim.c.chen@xxxxxxxxx>
Cc: Wei Xu <weixugc@xxxxxxxxxx>
Cc: Yang Shi <shy828301@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mmzone.h | 3 ++
mm/memory-tiers.c | 40 ++++++++++++++++++++++++++++++++++-----
2 files changed, 38 insertions(+), 5 deletions(-)
--- a/include/linux/mmzone.h~mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details
+++ a/include/linux/mmzone.h
@@ -1244,6 +1244,9 @@ typedef struct pglist_data {
/* Per-node vmstats */
struct per_cpu_nodestat __percpu *per_cpu_nodestats;
atomic_long_t vm_stat[NR_VM_NODE_STAT_ITEMS];
+#ifdef CONFIG_NUMA
+ struct memory_tier __rcu *memtier;
+#endif
} pg_data_t;
#define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages)
--- a/mm/memory-tiers.c~mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details
+++ a/mm/memory-tiers.c
@@ -4,6 +4,7 @@
#include <linux/sysfs.h>
#include <linux/kobject.h>
#include <linux/memory.h>
+#include <linux/mmzone.h>
#include <linux/memory-tiers.h>
#include "internal.h"
@@ -136,12 +137,18 @@ static struct memory_tier *find_create_m
static struct memory_tier *__node_get_memory_tier(int node)
{
- struct memory_dev_type *memtype;
+ pg_data_t *pgdat;
- memtype = node_memory_types[node];
- if (memtype && node_isset(node, memtype->nodes))
- return memtype->memtier;
- return NULL;
+ pgdat = NODE_DATA(node);
+ if (!pgdat)
+ return NULL;
+ /*
+ * Since we hold memory_tier_lock, we can avoid
+ * RCU read locks when accessing the details. No
+ * parallel updates are possible here.
+ */
+ return rcu_dereference_check(pgdat->memtier,
+ lockdep_is_held(&memory_tier_lock));
}
#ifdef CONFIG_MIGRATION
@@ -294,6 +301,8 @@ static struct memory_tier *set_node_memo
{
struct memory_tier *memtier;
struct memory_dev_type *memtype;
+ pg_data_t *pgdat = NODE_DATA(node);
+
lockdep_assert_held_once(&memory_tier_lock);
@@ -305,24 +314,45 @@ static struct memory_tier *set_node_memo
memtype = node_memory_types[node];
node_set(node, memtype->nodes);
memtier = find_create_memory_tier(memtype);
+ if (!IS_ERR(memtier))
+ rcu_assign_pointer(pgdat->memtier, memtier);
return memtier;
}
static void destroy_memory_tier(struct memory_tier *memtier)
{
list_del(&memtier->list);
+ /*
+ * synchronize_rcu in clear_node_memory_tier makes sure
+ * we don't have rcu access to this memory tier.
+ */
kfree(memtier);
}
static bool clear_node_memory_tier(int node)
{
bool cleared = false;
+ pg_data_t *pgdat;
struct memory_tier *memtier;
+ pgdat = NODE_DATA(node);
+ if (!pgdat)
+ return false;
+
+ /*
+ * Make sure that anybody looking at NODE_DATA who finds
+ * a valid memtier finds memory_dev_types with nodes still
+ * linked to the memtier. We achieve this by waiting for
+ * rcu read section to finish using synchronize_rcu.
+ * This also enables us to free the destroyed memory tier
+ * with kfree instead of kfree_rcu
+ */
memtier = __node_get_memory_tier(node);
if (memtier) {
struct memory_dev_type *memtype;
+ rcu_assign_pointer(pgdat->memtier, NULL);
+ synchronize_rcu();
memtype = node_memory_types[node];
node_clear(node, memtype->nodes);
if (nodes_empty(memtype->nodes)) {
_
Patches currently in -mm which might be from aneesh.kumar@xxxxxxxxxxxxx are
mm-demotion-add-support-for-explicit-memory-tiers.patch
mm-demotion-move-memory-demotion-related-code.patch
mm-demotion-add-hotplug-callbacks-to-handle-new-numa-node-onlined.patch
mm-demotion-dax-kmem-set-nodes-abstract-distance-to-memtier_default_dax_adistance.patch
mm-demotion-build-demotion-targets-based-on-explicit-memory-tiers.patch
mm-demotion-add-pg_data_t-member-to-track-node-memory-tier-details.patch
mm-demotion-drop-memtier-from-memtype.patch
mm-demotion-update-node_is_toptier-to-work-with-memory-tiers.patch
lib-nodemask-optimize-node_random-for-nodemask-with-single-numa-node.patch
