[PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[pratyush.brahma]

From: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>

The function `memory_add_physaddr_to_nid` seems a misnomer.
It does not to "add" a physical address to a NID mapping,
but rather it gets the NID associated with a given physical address.

Improve the semantic clarity of the API by renaming to a more
descriptive name.

Signed-off-by: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>
---
 arch/powerpc/include/asm/sparsemem.h | 4 ++--
 arch/powerpc/mm/mem.c                | 4 ++--
 arch/s390/include/asm/sparsemem.h    | 4 ++--
 drivers/acpi/acpi_memhotplug.c       | 2 +-
 drivers/acpi/nfit/core.c             | 2 +-
 drivers/base/memory.c                | 2 +-
 drivers/cxl/pmem.c                   | 2 +-
 drivers/dax/cxl.c                    | 2 +-
 drivers/hv/hv_balloon.c              | 6 +++---
 drivers/nvdimm/virtio_pmem.c         | 2 +-
 drivers/virtio/virtio_mem.c          | 2 +-
 drivers/xen/balloon.c                | 2 +-
 include/linux/numa.h                 | 6 +++---
 include/linux/numa_memblks.h         | 4 ++--
 mm/numa.c                            | 6 +++---
 mm/numa_memblks.c                    | 4 ++--
 16 files changed, 27 insertions(+), 27 deletions(-)

diff --git a/arch/powerpc/include/asm/sparsemem.h b/arch/powerpc/include/asm/sparsemem.h
index d072866842e4229159fcb6635745fd19a2788413..cdb5594616c8e32ab8ae3d209d0b2df20f3564f6 100644
--- a/arch/powerpc/include/asm/sparsemem.h
+++ b/arch/powerpc/include/asm/sparsemem.h
@@ -14,8 +14,8 @@
 
 #ifdef CONFIG_MEMORY_HOTPLUG
 extern int remove_section_mapping(unsigned long start, unsigned long end);
-extern int memory_add_physaddr_to_nid(u64 start);
-#define memory_add_physaddr_to_nid memory_add_physaddr_to_nid
+extern int memory_get_phys_to_nid(u64 start);
+#define memory_get_phys_to_nid memory_get_phys_to_nid
 
 #ifdef CONFIG_NUMA
 extern int hot_add_scn_to_nid(unsigned long scn_addr);
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 3ddbfdbfa9413a17d8321885724ce432d967005a..5665b2e2fc68df8634613c11706e1a9158a3a616 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -54,11 +54,11 @@ EXPORT_SYMBOL(__phys_mem_access_prot);
 static DEFINE_MUTEX(linear_mapping_mutex);
 
 #ifdef CONFIG_NUMA
-int memory_add_physaddr_to_nid(u64 start)
+int memory_get_phys_to_nid(u64 start)
 {
 	return hot_add_scn_to_nid(start);
 }
-EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
+EXPORT_SYMBOL_GPL(memory_get_phys_to_nid);
 #endif
 
 int __weak create_section_mapping(unsigned long start, unsigned long end,
diff --git a/arch/s390/include/asm/sparsemem.h b/arch/s390/include/asm/sparsemem.h
index 668dfc5de538742269af1b25151372506105fd4d..744c67dce4c106d063cb5436b37412045da0fd2a 100644
--- a/arch/s390/include/asm/sparsemem.h
+++ b/arch/s390/include/asm/sparsemem.h
@@ -7,11 +7,11 @@
 
 #ifdef CONFIG_NUMA
 
-static inline int memory_add_physaddr_to_nid(u64 addr)
+static inline int memory_get_phys_to_nid(u64 addr)
 {
 	return 0;
 }
-#define memory_add_physaddr_to_nid memory_add_physaddr_to_nid
+#define memory_get_phys_to_nid memory_get_phys_to_nid
 
 static inline int phys_to_target_node(u64 start)
 {
diff --git a/drivers/acpi/acpi_memhotplug.c b/drivers/acpi/acpi_memhotplug.c
index d0c1a71007d0a3054608bec8fddc8e86bdffb78b..6f968243ffc3cc5864e73b914d1bf67748e30e66 100644
--- a/drivers/acpi/acpi_memhotplug.c
+++ b/drivers/acpi/acpi_memhotplug.c
@@ -183,7 +183,7 @@ static int acpi_memory_enable_device(struct acpi_memory_device *mem_device)
 			continue;
 		/* We want a single node for the whole memory group */
 		if (node < 0)
-			node = memory_add_physaddr_to_nid(info->start_addr);
+			node = memory_get_phys_to_nid(info->start_addr);
 		total_length += info->length;
 	}
 
diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index ae035b93da087812dee6ec47d9ef4aa97dc8e7bc..04c3ab311e4d5923aef50252efabd193c5fb7850 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -2630,7 +2630,7 @@ static int acpi_nfit_register_region(struct acpi_nfit_desc *acpi_desc,
 
 	/* Fallback to address based numa information if node lookup failed */
 	if (ndr_desc->numa_node == NUMA_NO_NODE) {
-		ndr_desc->numa_node = memory_add_physaddr_to_nid(spa->address);
+		ndr_desc->numa_node = memory_get_phys_to_nid(spa->address);
 		dev_info(acpi_desc->dev, "changing numa node from %d to %d for nfit region [%pa-%pa]",
 			NUMA_NO_NODE, ndr_desc->numa_node, &res.start, &res.end);
 	}
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 5c6c1d6bb59f1241a5f42a3396be1a8e2058c965..f657520855408804761afec379e3c0b2a238b239 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -599,7 +599,7 @@ static ssize_t probe_store(struct device *dev, struct device_attribute *attr,
 	if (ret)
 		return ret;
 
-	nid = memory_add_physaddr_to_nid(phys_addr);
+	nid = memory_get_phys_to_nid(phys_addr);
 	ret = __add_memory(nid, phys_addr,
 			   MIN_MEMORY_BLOCK_SIZE * sections_per_block,
 			   MHP_NONE);
diff --git a/drivers/cxl/pmem.c b/drivers/cxl/pmem.c
index e197883690efc11f60c31bd56aeb5695665d422b..471aed4e11241948e994d0b9d53600c147e38fb0 100644
--- a/drivers/cxl/pmem.c
+++ b/drivers/cxl/pmem.c
@@ -410,7 +410,7 @@ static int cxl_pmem_region_probe(struct device *dev)
 	ndr_desc.res = res;
 	ndr_desc.provider_data = cxlr_pmem;
 
-	ndr_desc.numa_node = memory_add_physaddr_to_nid(res->start);
+	ndr_desc.numa_node = memory_get_phys_to_nid(res->start);
 	ndr_desc.target_node = phys_to_target_node(res->start);
 	if (ndr_desc.target_node == NUMA_NO_NODE) {
 		ndr_desc.target_node = ndr_desc.numa_node;
diff --git a/drivers/dax/cxl.c b/drivers/dax/cxl.c
index 13cd94d32ff7a1d70af7821c1aecd7490302149d..35843791872ba466f571c022e1484816368a1198 100644
--- a/drivers/dax/cxl.c
+++ b/drivers/dax/cxl.c
@@ -15,7 +15,7 @@ static int cxl_dax_region_probe(struct device *dev)
 	struct dev_dax_data data;
 
 	if (nid == NUMA_NO_NODE)
-		nid = memory_add_physaddr_to_nid(cxlr_dax->hpa_range.start);
+		nid = memory_get_phys_to_nid(cxlr_dax->hpa_range.start);
 
 	dax_region = alloc_dax_region(dev, cxlr->id, &cxlr_dax->hpa_range, nid,
 				      PMD_SIZE, IORESOURCE_DAX_KMEM);
diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c
index 2b4080e51f97c30e7adb06faa56b1403bc08b53c..8878aac6da9cd3bd6f499eae2588e989b7219af3 100644
--- a/drivers/hv/hv_balloon.c
+++ b/drivers/hv/hv_balloon.c
@@ -725,7 +725,7 @@ static void hv_mem_hot_add(unsigned long start, unsigned long size,
 
 		reinit_completion(&dm_device.ol_waitevent);
 
-		nid = memory_add_physaddr_to_nid(PFN_PHYS(start_pfn));
+		nid = memory_get_phys_to_nid(PFN_PHYS(start_pfn));
 		ret = add_memory(nid, PFN_PHYS((start_pfn)),
 				 HA_BYTES_IN_CHUNK, MHP_MERGE_RESOURCE);
 
@@ -1701,8 +1701,8 @@ static int hot_add_enabled(void)
 {
 	/*
 	 * Disable hot add on ARM64, because we currently rely on
-	 * memory_add_physaddr_to_nid() to get a node id of a hot add range,
-	 * however ARM64's memory_add_physaddr_to_nid() always return 0 and
+	 * memory_get_phys_to_nid() to get a node id of a hot add range,
+	 * however ARM64's memory_get_phys_to_nid() always return 0 and
 	 * DM_MEM_HOT_ADD_REQUEST doesn't have the NUMA node information for
 	 * add_memory().
 	 */
diff --git a/drivers/nvdimm/virtio_pmem.c b/drivers/nvdimm/virtio_pmem.c
index 2396d19ce5496974f8b93b54cc8c95e48dda103d..df6a3fd552d11a577d85708f80e7558ea83839d2 100644
--- a/drivers/nvdimm/virtio_pmem.c
+++ b/drivers/nvdimm/virtio_pmem.c
@@ -100,7 +100,7 @@ static int virtio_pmem_probe(struct virtio_device *vdev)
 
 	ndr_desc.res = &res;
 
-	ndr_desc.numa_node = memory_add_physaddr_to_nid(res.start);
+	ndr_desc.numa_node = memory_get_phys_to_nid(res.start);
 	ndr_desc.target_node = phys_to_target_node(res.start);
 	if (ndr_desc.target_node == NUMA_NO_NODE) {
 		ndr_desc.target_node = ndr_desc.numa_node;
diff --git a/drivers/virtio/virtio_mem.c b/drivers/virtio/virtio_mem.c
index 1688ecd69a0445d5c21d108a9a1d60577f96d7ac..6154f03a12c3efa65728f626e43b6270245d439c 100644
--- a/drivers/virtio/virtio_mem.c
+++ b/drivers/virtio/virtio_mem.c
@@ -2858,7 +2858,7 @@ static int virtio_mem_init(struct virtio_mem *vm)
 
 	/* Determine the nid for the device based on the lowest address. */
 	if (vm->nid == NUMA_NO_NODE)
-		vm->nid = memory_add_physaddr_to_nid(vm->addr);
+		vm->nid = memory_get_phys_to_nid(vm->addr);
 
 	dev_info(&vm->vdev->dev, "start address: 0x%llx", vm->addr);
 	dev_info(&vm->vdev->dev, "region size: 0x%llx", vm->region_size);
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index 2de37dcd75566fc1a03b75232cbe17fc0f53909d..93ca270ddd516ec11bc3f096eb518b9789d92664 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -284,7 +284,7 @@ static enum bp_state reserve_additional_memory(void)
 	if (!resource)
 		goto err;
 
-	nid = memory_add_physaddr_to_nid(resource->start);
+	nid = memory_get_phys_to_nid(resource->start);
 
 #ifdef CONFIG_XEN_HAVE_PVMMU
 	/*
diff --git a/include/linux/numa.h b/include/linux/numa.h
index e6baaf6051bcff6c23308d3b67f790053fbd29dc..ed65a20f39718a5a3157f6f9db60561f4418b000 100644
--- a/include/linux/numa.h
+++ b/include/linux/numa.h
@@ -33,8 +33,8 @@ int numa_nearest_node(int node, unsigned int state);
 
 int nearest_node_nodemask(int node, nodemask_t *mask);
 
-#ifndef memory_add_physaddr_to_nid
-int memory_add_physaddr_to_nid(u64 start);
+#ifndef memory_get_phys_to_nid
+int memory_get_phys_to_nid(u64 start);
 #endif
 
 #ifndef phys_to_target_node
@@ -54,7 +54,7 @@ static inline int nearest_node_nodemask(int node, nodemask_t *mask)
 	return NUMA_NO_NODE;
 }
 
-static inline int memory_add_physaddr_to_nid(u64 start)
+static inline int memory_get_phys_to_nid(u64 start)
 {
 	return 0;
 }
diff --git a/include/linux/numa_memblks.h b/include/linux/numa_memblks.h
index 991076cba7c5016d845eb40a2f9887f73fa83862..37cc0987e738f8aa2eb5c8f1ca2c94394def4780 100644
--- a/include/linux/numa_memblks.h
+++ b/include/linux/numa_memblks.h
@@ -53,8 +53,8 @@ static inline int numa_emu_cmdline(char *str)
 #ifdef CONFIG_NUMA_KEEP_MEMINFO
 extern int phys_to_target_node(u64 start);
 #define phys_to_target_node phys_to_target_node
-extern int memory_add_physaddr_to_nid(u64 start);
-#define memory_add_physaddr_to_nid memory_add_physaddr_to_nid
+extern int memory_get_phys_to_nid(u64 start);
+#define memory_get_phys_to_nid memory_get_phys_to_nid
 #endif /* CONFIG_NUMA_KEEP_MEMINFO */
 
 #endif /* CONFIG_NUMA_MEMBLKS */
diff --git a/mm/numa.c b/mm/numa.c
index 7d5e06fe5bd4a2790b83dd7dbe646617f6476d8c..f7b5ac8aea608368b75c606970187eb147ddf427 100644
--- a/mm/numa.c
+++ b/mm/numa.c
@@ -40,14 +40,14 @@ void __init alloc_offline_node_data(int nid)
 
 /* Stub functions: */
 
-#ifndef memory_add_physaddr_to_nid
-int memory_add_physaddr_to_nid(u64 start)
+#ifndef memory_get_phys_to_nid
+int memory_get_phys_to_nid(u64 start)
 {
 	pr_info_once("Unknown online node for memory at 0x%llx, assuming node 0\n",
 			start);
 	return 0;
 }
-EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
+EXPORT_SYMBOL_GPL(memory_get_phys_to_nid);
 #endif
 
 #ifndef phys_to_target_node
diff --git a/mm/numa_memblks.c b/mm/numa_memblks.c
index 541a99c4071a67e5b0ef66f4136dee268a880003..346039bd933390fb014931fc61ccd2f96a773c37 100644
--- a/mm/numa_memblks.c
+++ b/mm/numa_memblks.c
@@ -580,7 +580,7 @@ int phys_to_target_node(u64 start)
 }
 EXPORT_SYMBOL_GPL(phys_to_target_node);
 
-int memory_add_physaddr_to_nid(u64 start)
+int memory_get_phys_to_nid(u64 start)
 {
 	int nid = meminfo_to_nid(&numa_meminfo, start);
 
@@ -588,6 +588,6 @@ int memory_add_physaddr_to_nid(u64 start)
 		nid = numa_meminfo.blk[0].nid;
 	return nid;
 }
-EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
+EXPORT_SYMBOL_GPL(memory_get_phys_to_nid);
 
 #endif /* CONFIG_NUMA_KEEP_MEMINFO */

---
base-commit: 479058002c32b77acac43e883b92174e22c4be2d
change-id: 20250814-numa_memblks-b54a878820de

Best regards,
-- 
Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>

Re: [PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[David Hildenbrand]

On 18.08.25 08:41, pratyush.brahma@oss.qualcomm.com wrote:
> From: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>
> 
> The function `memory_add_physaddr_to_nid` seems a misnomer.
> It does not to "add" a physical address to a NID mapping,
> but rather it gets the NID associated with a given physical address.

You probably misunderstood what the function is used for: memory hotplug 
aka "memory_add".

This patch is making matters worse by stripping that detail, unfortunately.


-- 
Cheers

David / dhildenb

Re: [PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[Pratyush Brahma]

On Mon, Aug 18, 2025 at 12:29 PM David Hildenbrand <david@redhat.com> wrote:
>
> On 18.08.25 08:41, pratyush.brahma@oss.qualcomm.com wrote:
> > From: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>
> >
> > The function `memory_add_physaddr_to_nid` seems a misnomer.
> > It does not to "add" a physical address to a NID mapping,
> > but rather it gets the NID associated with a given physical address.
>
> You probably misunderstood what the function is used for: memory hotplug
> aka "memory_add".
Thanks for your feedback. I get the part about memory hotplug here but
using memory_add still seems a little odd as it doesn't truly reflect
what this api is doing.
However, I agree that my current suggestion
may not be the perfect choice for the name, so I'm open to suggestions.

Perhaps, something like "memory_add_get_nid_by_phys" may work here?
>
> This patch is making matters worse by stripping that detail, unfortunately.
>
>
> --
> Cheers
>
> David / dhildenb
>
Thanks and Regards
Pratyush

Re: [PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[David Hildenbrand]

On 18.08.25 10:27, Pratyush Brahma wrote:
> On Mon, Aug 18, 2025 at 12:29 PM David Hildenbrand <david@redhat.com> wrote:
>>
>> On 18.08.25 08:41, pratyush.brahma@oss.qualcomm.com wrote:
>>> From: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>
>>>
>>> The function `memory_add_physaddr_to_nid` seems a misnomer.
>>> It does not to "add" a physical address to a NID mapping,
>>> but rather it gets the NID associated with a given physical address.
>>
>> You probably misunderstood what the function is used for: memory hotplug
>> aka "memory_add".
> Thanks for your feedback. I get the part about memory hotplug here but
> using memory_add still seems a little odd as it doesn't truly reflect
> what this api is doing.
> However, I agree that my current suggestion
> may not be the perfect choice for the name, so I'm open to suggestions.
> 
> Perhaps, something like "memory_add_get_nid_by_phys" may work here?

I don't think this name is really any better and worth the churn :(

-- 
Cheers

David / dhildenb

Re: [PATCH] mm/numa: Rename memory_add_physaddr_to_nid to memory_get_phys_to_nid

[Pratyush Brahma]

On Mon, Aug 18, 2025 at 2:01 PM David Hildenbrand <david@redhat.com> wrote:
>
> On 18.08.25 10:27, Pratyush Brahma wrote:
> > On Mon, Aug 18, 2025 at 12:29 PM David Hildenbrand <david@redhat.com> wrote:
> >>
> >> On 18.08.25 08:41, pratyush.brahma@oss.qualcomm.com wrote:
> >>> From: Pratyush Brahma <pratyush.brahma@oss.qualcomm.com>
> >>>
> >>> The function `memory_add_physaddr_to_nid` seems a misnomer.
> >>> It does not to "add" a physical address to a NID mapping,
> >>> but rather it gets the NID associated with a given physical address.
> >>
> >> You probably misunderstood what the function is used for: memory hotplug
> >> aka "memory_add".
> > Thanks for your feedback. I get the part about memory hotplug here but
> > using memory_add still seems a little odd as it doesn't truly reflect
> > what this api is doing.
> > However, I agree that my current suggestion
> > may not be the perfect choice for the name, so I'm open to suggestions.
> >
> > Perhaps, something like "memory_add_get_nid_by_phys" may work here?
>
> I don't think this name is really any better and worth the churn :(
>
Sure. Thanks for taking a look. Will drop this.
> --
> Cheers
>
> David / dhildenb
>
Thanks & Regards
Pratyush