All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
To: linux-mm@kvack.org, akpm@linux-foundation.org
Cc: Wei Xu <weixugc@google.com>, Huang Ying <ying.huang@intel.com>,
	Yang Shi <shy828301@gmail.com>,
	Davidlohr Bueso <dave@stgolabs.net>,
	Tim C Chen <tim.c.chen@intel.com>,
	Michal Hocko <mhocko@kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Hesham Almatary <hesham.almatary@huawei.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Jonathan Cameron <Jonathan.Cameron@huawei.com>,
	Alistair Popple <apopple@nvidia.com>,
	Dan Williams <dan.j.williams@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	jvgediya.oss@gmail.com, Bharata B Rao <bharata@amd.com>,
	"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
Subject: [RFC PATCH 1/2] mm/demotion: Expose memory type details via sysfs
Date: Thu, 25 Aug 2022 14:53:24 +0530	[thread overview]
Message-ID: <20220825092325.381517-1-aneesh.kumar@linux.ibm.com> (raw)

This patch adds /sys/devices/virtual/memtier/ where all memory tier related
details can be found. All allocated memory types will be listed there as
/sys/devices/virtual/memtier/memtypeN/

The nodes which are part of a specific memory type can be listed via
/sys/devices/system/memtier/memtypeN/nodes.

The adistance value of a specific memory type can be listed via
/sys/devices/system/memtier/memtypeN/adistance.

A directory listing looks like:
:/sys/devices/virtual/memtier# tree memtype1
memtype1
├── adistance
├── nodes
├── subsystem -> ../../../../bus/memtier
└── uevent

Since we will be using struct device to expose details via sysfs, drop struct
kref and use struct device for refcounting the memtype.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
---
 include/linux/memory-tiers.h |  3 +-
 mm/memory-tiers.c            | 97 +++++++++++++++++++++++++++++++++---
 2 files changed, 91 insertions(+), 9 deletions(-)

diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
index ecd865922707..487209a572b2 100644
--- a/include/linux/memory-tiers.h
+++ b/include/linux/memory-tiers.h
@@ -6,6 +6,7 @@
 #include <linux/nodemask.h>
 #include <linux/kref.h>
 #include <linux/mmzone.h>
+#include <linux/device.h>
 /*
  * Each tier cover a abstrace distance chunk size of 128
  */
@@ -28,7 +29,7 @@ struct memory_dev_type {
 	int adistance;
 	/* Nodes of same abstract distance */
 	nodemask_t nodes;
-	struct kref kref;
+	struct device dev;
 };
 
 #ifdef CONFIG_NUMA
diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
index ba844fe9cc8c..9eef3bd8d134 100644
--- a/mm/memory-tiers.c
+++ b/mm/memory-tiers.c
@@ -4,6 +4,7 @@
 #include <linux/sysfs.h>
 #include <linux/kobject.h>
 #include <linux/memory.h>
+#include <linux/idr.h>
 #include <linux/memory-tiers.h>
 
 #include "internal.h"
@@ -31,6 +32,15 @@ static DEFINE_MUTEX(memory_tier_lock);
 static LIST_HEAD(memory_tiers);
 static struct memory_dev_type *node_memory_types[MAX_NUMNODES];
 static struct memory_dev_type *default_dram_type;
+
+#define MAX_MEMORY_TYPE_ID	20
+static DEFINE_IDR(memory_type_idr);
+#define to_memory_type(device) container_of(device, struct memory_dev_type, dev)
+static struct bus_type memory_tier_subsys = {
+	.name = "memtier",
+	.dev_name = "memtier",
+};
+
 #ifdef CONFIG_MIGRATION
 static int top_tier_adistance;
 /*
@@ -388,7 +398,7 @@ static inline void __init_node_memory_type(int node, struct memory_dev_type *mem
 {
 	if (!node_memory_types[node]) {
 		node_memory_types[node] = memtype;
-		kref_get(&memtype->kref);
+		get_device(&memtype->dev);
 	}
 }
 
@@ -460,33 +470,87 @@ static bool clear_node_memory_tier(int node)
 	return cleared;
 }
 
-static void release_memtype(struct kref *kref)
+static ssize_t nodes_show(struct device *dev,
+			  struct device_attribute *attr, char *buf)
 {
-	struct memory_dev_type *memtype;
+	int ret;
+	struct memory_dev_type *memtype = to_memory_type(dev);
 
-	memtype = container_of(kref, struct memory_dev_type, kref);
+	mutex_lock(&memory_tier_lock);
+	ret = sysfs_emit(buf, "%*pbl\n", nodemask_pr_args(&memtype->nodes));
+	mutex_unlock(&memory_tier_lock);
+	return ret;
+}
+static DEVICE_ATTR_RO(nodes);
+
+static ssize_t adistance_show(struct device *dev,
+			      struct device_attribute *attr, char *buf)
+{
+	int ret;
+	struct memory_dev_type *memtype = to_memory_type(dev);
+
+	mutex_lock(&memory_tier_lock);
+	ret = sysfs_emit(buf, "%d\n", memtype->adistance);
+	mutex_unlock(&memory_tier_lock);
+	return ret;
+}
+static DEVICE_ATTR_RO(adistance);
+
+static struct attribute *memtype_dev_attrs[] = {
+	&dev_attr_nodes.attr,
+	&dev_attr_adistance.attr,
+	NULL
+};
+
+static const struct attribute_group memtype_dev_group = {
+	.attrs = memtype_dev_attrs,
+};
+
+static const struct attribute_group *memtype_dev_groups[] = {
+	&memtype_dev_group,
+	NULL
+};
+
+static void memtype_device_release(struct device *dev)
+{
+	struct memory_dev_type *memtype = to_memory_type(dev);
+
+	idr_remove(&memory_type_idr, memtype->dev.id);
 	kfree(memtype);
 }
 
 struct memory_dev_type *alloc_memory_type(int adistance)
 {
+	int ret;
 	struct memory_dev_type *memtype;
 
-	memtype = kmalloc(sizeof(*memtype), GFP_KERNEL);
+	memtype = kzalloc(sizeof(*memtype), GFP_KERNEL);
 	if (!memtype)
 		return ERR_PTR(-ENOMEM);
 
 	memtype->adistance = adistance;
 	INIT_LIST_HEAD(&memtype->tier_sibiling);
 	memtype->nodes  = NODE_MASK_NONE;
-	kref_init(&memtype->kref);
+	memtype->dev.id = idr_alloc(&memory_type_idr, NULL,
+				 1, MAX_MEMORY_TYPE_ID + 1, GFP_KERNEL);
+	memtype->dev.bus = &memory_tier_subsys;
+	memtype->dev.release = memtype_device_release;
+	memtype->dev.groups = memtype_dev_groups;
+	dev_set_name(&memtype->dev, "%s%d", "memtype", memtype->dev.id);
+
+	ret = device_register(&memtype->dev);
+	if (ret) {
+		put_device(&memtype->dev);
+		return ERR_PTR(ret);
+	}
+
 	return memtype;
 }
 EXPORT_SYMBOL_GPL(alloc_memory_type);
 
 void destroy_memory_type(struct memory_dev_type *memtype)
 {
-	kref_put(&memtype->kref, release_memtype);
+	device_unregister(&memtype->dev);
 }
 EXPORT_SYMBOL_GPL(destroy_memory_type);
 
@@ -504,7 +568,7 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
 	mutex_lock(&memory_tier_lock);
 	if (node_memory_types[node] == memtype) {
 		node_memory_types[node] = NULL;
-		kref_put(&memtype->kref, release_memtype);
+		put_device(&memtype->dev);
 	}
 	mutex_unlock(&memory_tier_lock);
 }
@@ -582,6 +646,23 @@ static int __init memory_tier_init(void)
 }
 subsys_initcall(memory_tier_init);
 
+/*
+ * initialize sysfs subsys in core_initcall so that
+ * other kernel components can do alloc_memory_type
+ * before memory_tier_init
+ */
+static int __init memory_tier_sysfs_init(void)
+{
+	int err;
+
+	err = subsys_virtual_register(&memory_tier_subsys, NULL);
+	if (err)
+		return err;
+
+	return 0;
+}
+core_initcall(memory_tier_sysfs_init);
+
 bool numa_demotion_enabled = false;
 
 #ifdef CONFIG_MIGRATION
-- 
2.37.2


             reply	other threads:[~2022-08-25  9:24 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-25  9:23 Aneesh Kumar K.V [this message]
2022-08-25  9:23 ` [RFC PATCH 2/2] mm/demotion: Expose memory tier details via sysfs Aneesh Kumar K.V
2022-08-26  4:31   ` Huang, Ying
2022-08-26  1:50 ` [RFC PATCH 1/2] mm/demotion: Expose memory type " Huang, Ying
2022-08-26  2:37   ` Aneesh Kumar K V
2022-08-26  8:00     ` Wei Xu
2022-08-26  8:05       ` Aneesh Kumar K V
2022-08-26  9:15         ` Wei Xu
2022-08-28 16:20           ` Aneesh Kumar K.V

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220825092325.381517-1-aneesh.kumar@linux.ibm.com \
    --to=aneesh.kumar@linux.ibm.com \
    --cc=Jonathan.Cameron@huawei.com \
    --cc=akpm@linux-foundation.org \
    --cc=apopple@nvidia.com \
    --cc=bharata@amd.com \
    --cc=dan.j.williams@intel.com \
    --cc=dave.hansen@intel.com \
    --cc=dave@stgolabs.net \
    --cc=hannes@cmpxchg.org \
    --cc=hesham.almatary@huawei.com \
    --cc=jvgediya.oss@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@kernel.org \
    --cc=shy828301@gmail.com \
    --cc=tim.c.chen@intel.com \
    --cc=weixugc@google.com \
    --cc=ying.huang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.