Skip to contentSkip to conversation

drivers/base/node: fold node register and unregister functions

PATCH In review Donet Tom 4 participants 10 messages Oct 14, 2025 – Oct 17, 2025

Subsystems (7)
lore
Share & tools
lore

Donet Tom

author
#0 lore raw

v2 cover drivers/base/node: fold node register and unregister functions

This change came from the discussion in [1]. It consists of two patches:

The first patch merges register_one_node() and register_node(), leaving a single register_node() function.

The second patch merges unregister_one_node() and unregister_node(), leaving a single unregister_node() function.

There is no functional change in these patches.

[1] https://lore.kernel.org/all/[email protected]/

---
v1 -> v2
- Changed commit description
- Removed extern from the function declarations of register_node() and unregister_node().

v1 - https://lore.kernel.org/all/[email protected]/

Donet Tom (2):
drivers/base/node: Fold register_node() into register_one_node() drivers/base/node: Fold unregister_node() into unregister_one_node()

 arch/powerpc/platforms/pseries/pci_dlpar.c |  2 +-
 arch/x86/mm/numa.c                         |  4 +-
 drivers/base/node.c                        | 90 +++++++++-------------
 include/linux/node.h                       | 10 +--
 mm/memory_hotplug.c                        |  8 +-
 mm/mm_init.c                               |  2 +-
 6 files changed, 49 insertions(+), 67 deletions(-)
Signature · 1 line
2.47.1
author
#1 lore raw

PATCH v2 1/2 drivers/base/node: Fold register_node() into register_one_node()

The function register_node() is only called from register_one_node(). This patch folds register_node() into its only caller and renames register_one_node() to register_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

---
 arch/powerpc/platforms/pseries/pci_dlpar.c |  2 +-
 arch/x86/mm/numa.c                         |  4 +-
 drivers/base/node.c                        | 52 +++++++++-------------
 include/linux/node.h                       |  4 +-
 mm/memory_hotplug.c                        |  4 +-
 mm/mm_init.c                               |  2 +-
 6 files changed, 28 insertions(+), 40 deletions(-)
arch/powerpc/platforms/pseries/pci_dlpar.c +1 -1
index aeb8633a3d00..8c77ec7980de 100644
@@ -29,7 +29,7 @@ struct pci_controller *init_phb_dynamic(struct device_node *dn)
 	nid = of_node_to_nid(dn);
 	if (likely((nid) >= 0)) {
 		if (!node_online(nid)) {
-			if (register_one_node(nid)) {
+			if (register_node(nid)) {
 				pr_err("PCI: Failed to register node %d\n", nid);
 			} else {
 				update_numa_distance(dn);
arch/x86/mm/numa.c +2 -2
index c24890c40138..7a97327140df 100644
@@ -262,7 +262,7 @@ void __init init_gi_nodes(void)
 	 * bringup_nonboot_cpus
 	 *  cpu_up
 	 *   __try_online_node
-	 *    register_one_node
+	 *    register_node
 	 * because node_subsys is not initialized yet.
 	 * TODO remove dependency on node_online
 	 */
@@ -303,7 +303,7 @@ void __init init_cpu_to_node(void)
 		 * bringup_nonboot_cpus
 		 *  cpu_up
 		 *   __try_online_node
-		 *    register_one_node
+		 *    register_node
 		 * because node_subsys is not initialized yet.
 		 * TODO remove dependency on node_online
 		 */
drivers/base/node.c +20 -32

81 lines not on this page · raw message

include/linux/node.h +2 -2
index 866e3323f1fd..b7028d3ec3b4 100644
@@ -176,7 +176,7 @@ static inline int hotplug_node_notifier(notifier_fn_t fn, int pri)
 #ifdef CONFIG_NUMA
 extern void node_dev_init(void);
 /* Core of the node registration - only memory hotplug should use this */
-extern int register_one_node(int nid);
+int register_node(int nid);
 extern void unregister_one_node(int nid);
 extern int register_cpu_under_node(unsigned int cpu, unsigned int nid);
 extern int unregister_cpu_under_node(unsigned int cpu, unsigned int nid);
@@ -189,7 +189,7 @@ extern int register_memory_node_under_compute_node(unsigned int mem_nid,
 static inline void node_dev_init(void)
 {
 }
-static inline int register_one_node(int nid)
+static inline int register_node(int nid)
 {
 	return 0;
 }
mm/memory_hotplug.c +2 -2
index 0be83039c3b5..6c050d867031 100644
@@ -1311,7 +1311,7 @@ static int __try_online_node(int nid, bool set_node_online)
 
 	if (set_node_online) {
 		node_set_online(nid);
-		ret = register_one_node(nid);
+		ret = register_node(nid);
 		BUG_ON(ret);
 	}
 out:
@@ -1542,7 +1542,7 @@ int add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 		goto error_memblock_remove;
 	if (ret) {
 		node_set_online(nid);
-		ret = register_one_node(nid);
+		ret = register_node(nid);
 		if (WARN_ON(ret)) {
 			node_set_offline(nid);
 			goto error_memblock_remove;
mm/mm_init.c +1 -1
index 3db2dea7db4c..d9ef05c4dc5b 100644
@@ -1909,7 +1909,7 @@ void __init free_area_init(unsigned long *max_zone_pfn)
 		free_area_init_node(nid);
 
 		/*
-		 * No sysfs hierarchy will be created via register_one_node()
+		 * No sysfs hierarchy will be created via register_node()
 		 *for memory-less node because here it's not marked as N_MEMORY
 		 *and won't be set online later. The benefit is userspace
 		 *program won't be confused by sysfs files/directories of
Signature · 1 line
2.47.1
author
#7 lore raw

PATCH v2 2/2 drivers/base/node: Fold unregister_node() into unregister_one_node()

In reply to Donet Tom

The function unregister_node() is only called from unregister_one_node(). This patch folds unregister_node() into its only caller and renames unregister_one_node() to unregister_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

---
 drivers/base/node.c  | 38 +++++++++++++++++---------------------
 include/linux/node.h |  6 ++----
 mm/memory_hotplug.c  |  4 ++--
 3 files changed, 21 insertions(+), 27 deletions(-)
drivers/base/node.c +17 -21
index 4bcaa8006cd6..758b777dec1f 100644
@@ -676,23 +676,6 @@ static void node_device_release(struct device *dev)
 	kfree(to_node(dev));
 }
 
-/**
- * unregister_node - unregister a node device
- * @node: node going away
- *
- * Unregisters a node device @node.  All the devices on the node must be
- * unregistered before calling this function.
- */
-void unregister_node(struct node *node)
-{
-	hugetlb_unregister_node(node);
-	compaction_unregister_node(node);
-	reclaim_unregister_node(node);
-	node_remove_accesses(node);
-	node_remove_caches(node);
-	device_unregister(&node->dev);
-}
-
 struct node *node_devices[MAX_NUMNODES];
 
 /*
@@ -924,13 +907,26 @@ int register_node(int nid)
 
 	return error;
 }
-
-void unregister_one_node(int nid)
+/**
+ * unregister_node - unregister a node device
+ * @nid: nid of the node going away
+ *
+ * Unregisters the node device at node id  @nid.  All the devices on the
+ * node must be unregistered before calling this function.
+ */
+void unregister_node(int nid)
 {
-	if (!node_devices[nid])
+	struct node *node = node_devices[nid];
+
+	if (!node)
 		return;
 
-	unregister_node(node_devices[nid]);
+	hugetlb_unregister_node(node);
+	compaction_unregister_node(node);
+	reclaim_unregister_node(node);
+	node_remove_accesses(node);
+	node_remove_caches(node);
+	device_unregister(&node->dev);
 	node_devices[nid] = NULL;
 }
 
include/linux/node.h +2 -4
index b7028d3ec3b4..0269b064ba65 100644
@@ -132,8 +132,6 @@ static inline void register_memory_blocks_under_nodes(void)
 }
 #endif
 
-extern void unregister_node(struct node *node);
-
 struct node_notify {
 	int nid;
 };
@@ -177,7 +175,7 @@ static inline int hotplug_node_notifier(notifier_fn_t fn, int pri)
 extern void node_dev_init(void);
 /* Core of the node registration - only memory hotplug should use this */
 int register_node(int nid);
-extern void unregister_one_node(int nid);
+void unregister_node(int nid);
 extern int register_cpu_under_node(unsigned int cpu, unsigned int nid);
 extern int unregister_cpu_under_node(unsigned int cpu, unsigned int nid);
 extern void unregister_memory_block_under_nodes(struct memory_block *mem_blk);
@@ -193,7 +191,7 @@ static inline int register_node(int nid)
 {
 	return 0;
 }
-static inline int unregister_one_node(int nid)
+static inline int unregister_node(int nid)
 {
 	return 0;
 }
mm/memory_hotplug.c +2 -2
index 6c050d867031..94a8f6e8811a 100644
@@ -1596,7 +1596,7 @@ int add_memory_resource(int nid, struct resource *res, mhp_t mhp_flags)
 error:
 	if (new_node) {
 		node_set_offline(nid);
-		unregister_one_node(nid);
+		unregister_node(nid);
 	}
 error_memblock_remove:
 	if (IS_ENABLED(CONFIG_ARCH_KEEP_MEMBLOCK))
@@ -2201,7 +2201,7 @@ void try_offline_node(int nid)
 	 * node now.
 	 */
 	node_set_offline(nid);
-	unregister_one_node(nid);
+	unregister_node(nid);
 }
 EXPORT_SYMBOL(try_offline_node);
 
Signature · 1 line
2.47.1
#2 lore raw

Re: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()

In reply to Donet Tom
kernel test robot noticed the following build warnings:

Hi Donet,

kernel test robot noticed the following build warnings:

[auto build test WARNING on akpm-mm/mm-everything]
[also build test WARNING on powerpc/fixes linus/master v6.18-rc1 next-20251014]
[cannot apply to driver-core/driver-core-testing driver-core/driver-core-next driver-core/driver-core-linus powerpc/next tip/x86/mm rppt-memblock/for-next rppt-memblock/fixes] [If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Donet-Tom/drivers-base-node-Fold-register_node-into-register_one_node/20251015-000850
base:   https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
patch link:    https://lore.kernel.org/r/910853c9dd61f7a2190a56cba101e73e9c6859be.1760097207.git.donettom%40linux.ibm.com
patch subject: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()
config: s390-randconfig-001-20251015 (https://download.01.org/0day-ci/archive/20251015/[email protected]/config)
compiler: s390-linux-gcc (GCC) 12.5.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251015/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All warnings (new ones prefixed by >>):

>> Warning: drivers/base/node.c:889 function parameter 'nid' not described in 'register_node'
Signature · 2 lines
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
author
#3 lore raw

Re: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()

Quoted 8 lines · kernel test robot

Hi Donet,

kernel test robot noticed the following build warnings:

[auto build test WARNING on akpm-mm/mm-everything]
[also build test WARNING on powerpc/fixes linus/master v6.18-rc1 next-20251014]
[cannot apply to driver-core/driver-core-testing driver-core/driver-core-next driver-core/driver-core-linus powerpc/next tip/x86/mm rppt-memblock/for-next rppt-memblock/fixes] [If your patch is applied to the wrong git tree, kindly drop us a note.

This patch is based on the mm-new branch. It is not applying cleanly on target branch because the target branches are missing some patches that are present in mm-new.

Quoted 19 lines

And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Donet-Tom/drivers-base-node-Fold-register_node-into-register_one_node/20251015-000850
base:   https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
patch link:    https://lore.kernel.org/r/910853c9dd61f7a2190a56cba101e73e9c6859be.1760097207.git.donettom%40linux.ibm.com
patch subject: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()
config: s390-randconfig-001-20251015 (https://download.01.org/0day-ci/archive/20251015/[email protected]/config)
compiler: s390-linux-gcc (GCC) 12.5.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251015/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All warnings (new ones prefixed by >>):

Quoted 1 line
Quoted 1 line

Warning: drivers/base/node.c:889 function parameter 'nid' not described in 'register_node'

Acked-by on v2 1/2
#4 lore raw

Re: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()

In reply to Donet Tom
Quoted 10 lines · Donet Tom

The function register_node() is only called from register_one_node(). This patch folds register_node() into its only caller and renames register_one_node() to register_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

---

[...]

drivers/base/node.c void register_memory_blocks_under_node_hotplug(int nid, unsigned long start_pfn, · line 886

Quoted 12 lines

/**
* unregister_node - unregister a node device
* @node: node going away

drivers/base/node.c +3 -1
@@ -907,7 +880,13 @@ void register_memory_blocks_under_node_hotplug(int nid, unsigned long start_pfn,
  }
  #endif /* CONFIG_MEMORY_HOTPLUG */
 
-int register_one_node(int nid)
+/**
+ * register_node - Initialize and register the node device.
+ * @nid - Node number to use when creating the device.
 

I assume that should be "@nid: ..." to silence the warning.

Signature · 3 lines
Cheers

David / dhildenb
Acked-by on v2 1/2
#6 lore raw

Re: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()

In reply to Donet Tom
Quoted 9 lines · On Tue, 14 Oct 2025 21:09:16 +0530 Donet Tom <[email protected]> wrote:

The function register_node() is only called from register_one_node(). This patch folds register_node() into its only caller and renames register_one_node() to register_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

Thanks,
SJ

[...]

Acked-by on v2 2/2
#8 lore raw

Re: [PATCH v2 2/2] drivers/base/node: Fold unregister_node() into unregister_one_node()

In reply to Donet Tom
Quoted 10 lines · On Tue, 14 Oct 2025 21:09:17 +0530 Donet Tom <[email protected]> wrote:

The function unregister_node() is only called from unregister_one_node(). This patch folds unregister_node() into its only caller and renames unregister_one_node() to unregister_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

Quoted 35 lines
---
 drivers/base/node.c  | 38 +++++++++++++++++---------------------
 include/linux/node.h |  6 ++----
 mm/memory_hotplug.c  |  4 ++--
 3 files changed, 21 insertions(+), 27 deletions(-)
drivers/base/node.c +0 -17
index 4bcaa8006cd6..758b777dec1f 100644
@@ -676,23 +676,6 @@ static void node_device_release(struct device *dev)
 	kfree(to_node(dev));
 }
 
-/**
- * unregister_node - unregister a node device
- * @node: node going away
- *
- * Unregisters a node device @node.  All the devices on the node must be
- * unregistered before calling this function.
- */
-void unregister_node(struct node *node)
-{
-	hugetlb_unregister_node(node);
-	compaction_unregister_node(node);
-	reclaim_unregister_node(node);
-	node_remove_accesses(node);
-	node_remove_caches(node);
-	device_unregister(&node->dev);
-}
-
 struct node *node_devices[MAX_NUMNODES];
 
 /*

drivers/base/node.c int register_node(int nid) · line 914

Quoted 10 lines
drivers/base/node.c +5 -2
@@ -924,13 +907,26 @@ int register_node(int nid)
 
 	return error;
 }
-
-void unregister_one_node(int nid)
+/**
+ * unregister_node - unregister a node device
+ * @nid: nid of the node going away
+ *
+ * Unregisters the node device at node id  @nid.  All the devices on the

Nit. Why two spaces before "@nid"?

[...]

Thanks,
SJ

author
#9 lore raw

Re: [PATCH v2 2/2] drivers/base/node: Fold unregister_node() into unregister_one_node()

Quoted 50 lines · SeongJae Park
Quoted 10 lines · On Tue, 14 Oct 2025 21:09:17 +0530 Donet Tom<[email protected]> wrote:

The function unregister_node() is only called from unregister_one_node(). This patch folds unregister_node() into its only caller and renames unregister_one_node() to unregister_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

Quoted 36 lines
---
  drivers/base/node.c  | 38 +++++++++++++++++---------------------
  include/linux/node.h |  6 ++----
  mm/memory_hotplug.c  |  4 ++--
  3 files changed, 21 insertions(+), 27 deletions(-)
drivers/base/node.c +0 -17
index 4bcaa8006cd6..758b777dec1f 100644
@@ -676,23 +676,6 @@ static void node_device_release(struct device *dev)
  	kfree(to_node(dev));
  }
 
-/**
- * unregister_node - unregister a node device
- * @node: node going away
- *
- * Unregisters a node device @node.  All the devices on the node must be
- * unregistered before calling this function.
- */
-void unregister_node(struct node *node)
-{
-	hugetlb_unregister_node(node);
-	compaction_unregister_node(node);
-	reclaim_unregister_node(node);
-	node_remove_accesses(node);
-	node_remove_caches(node);
-	device_unregister(&node->dev);
-}
-
  struct node *node_devices[MAX_NUMNODES];
 
  /*
@@ -924,13 +907,26 @@ int register_node(int nid)
 
Quoted 10 lines

drivers/base/node.c int register_node(int nid) · line 914

Quoted 9 lines
drivers/base/node.c +5 -2
  	return error;
  }
-
-void unregister_one_node(int nid)
+/**
+ * unregister_node - unregister a node device
+ * @nid: nid of the node going away
+ *
+ * Unregisters the node device at node id  @nid.  All the devices on the

Nit. Why two spaces before "@nid"?

Thank you for pointing this out.

Andrew could you please help apply the below fixlet on top of this patch?

From 0d0345847ad33607c7ebae3798ef37ff16a20fd0 Mon Sep 17 00:00:00 2001
From: Donet Tom<[email protected]> Date: Fri, 17 Oct 2025 10:27:54 +0530
Subject: [PATCH] [fixlet] drivers/base/node: Remove extra space in description

This patch removes an extra space before @nid and "All".

 drivers/base/node.c | 2 +-

 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/base/node.c +0 -0
index 63b223a3b38e..00cf4532f121 100644
@@ -911,7 +911,7 @@ int register_node(int nid)

  * unregister_node - unregister a node device
  * @nid: nid of the node going away
  *
- * Unregisters the node device at node id  @nid.  All the devices on the
+ * Unregisters the node device at node id @nid. All the devices on the
  * node must be unregistered before calling this function. */
 void unregister_node(int nid)
--
2.51.0

Quoted 6 lines

[...]

Thanks,
SJ

author
#5 lore raw

Re: [PATCH v2 1/2] drivers/base/node: Fold register_node() into register_one_node()

In reply to David Hildenbrand
Quoted 13 lines · David Hildenbrand
Quoted 10 lines · Donet Tom

The function register_node() is only called from register_one_node(). This patch folds register_node() into its only caller and renames register_one_node() to register_node().

This reduces unnecessary indirection and simplifies the code structure. No functional changes are introduced.

---

[...]

Quoted 16 lines

drivers/base/node.c void · line 887

Quoted 12 lines

  /**
   * unregister_node - unregister a node device
   * @node: node going away

drivers/base/node.c +3 -0
@@ -907,7 +880,13 @@ void 
 register_memory_blocks_under_node_hotplug(int nid, unsigned long start_pfn,
   }
   #endif /* CONFIG_MEMORY_HOTPLUG */
   -int register_one_node(int nid)
+/**
+ * register_node - Initialize and register the node device.
+ * @nid - Node number to use when creating the device.
 

I assume that should be "@nid: ..." to silence the warning.

Thank you, David, for pointing this out. Andrew has already applied the fix.

Quoted 3 lines