Skip to contentSkip to conversation

mm/sparse-vmemmap: Provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

PATCH Applied Muchun Song 5 participants 20 messages Jun 1 – Jun 26

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

v3 cover mm/sparse-vmemmap: Provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

The weak vmemmap_set_pmd() and vmemmap_check_pmd() hooks are currently no-ops in the generic code, which leaves architectures that need PMD-level handling to open-code the same logic locally.

This series provides generic implementations for both helpers in mm/sparse-vmemmap.c. vmemmap_set_pmd() installs a huge PMD with PAGE_KERNEL protection, and vmemmap_check_pmd() verifies a present leaf PMD before reusing the existing vmemmap_verify() helper.

With those generic helpers in place, patches 2-5 remove the now redundant arch-specific implementations from arm64, riscv, loongarch, and sparc.

v2 -> v3:
- Replace BUG_ON() with WARN_ON_ONCE() in patch 1
- Add Will Deacon's Acked-by to patch 2

Muchun Song (5):
mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and
vmemmap_check_pmd()
arm64/mm: drop vmemmap_pmd helpers and use generic code riscv/mm: drop vmemmap_pmd helpers and use generic code loongarch/mm: drop vmemmap_check_pmd helper and use generic code sparc/mm: drop vmemmap_check_pmd helper and use generic code

 arch/arm64/mm/mmu.c      | 14 --------------
 arch/loongarch/mm/init.c | 11 -----------
 arch/riscv/mm/init.c     | 13 -------------
 arch/sparc/mm/init_64.c  | 11 -----------
 mm/sparse-vmemmap.c      |  7 ++++++-
 5 files changed, 6 insertions(+), 50 deletions(-)

base-commit: 7da7f07112610a520567421dd2ffcb51beaefbcc

Signature · 1 line
2.54.0
#1 lore raw

PATCH v3 1/5 mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

The two weak functions are currently no-ops on every architecture, forcing each platform that needs them to duplicate the same handful of lines. Provide a generic implementation:

- vmemmap_set_pmd() simply sets a huge PMD with PAGE_KERNEL protection.

- vmemmap_check_pmd() verifies that the PMD is present and leaf, then calls the existing vmemmap_verify() helper.

Architectures that need special handling can continue to override the weak symbols; everyone else gets the standard version for free.

---
v2->v3:
- Replace BUG_ON() with WARN_ON_ONCE() in vmemmap_set_pmd()

---
 mm/sparse-vmemmap.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
mm/sparse-vmemmap.c +6 -1
index 112ccf9c71ca..99e2be39671b 100644
@@ -386,12 +386,17 @@ int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end,
 void __weak __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
 				      unsigned long addr, unsigned long next)
 {
+	WARN_ON_ONCE(!pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL));
 }
 
 int __weak __meminit vmemmap_check_pmd(pmd_t *pmd, int node,
 				       unsigned long addr, unsigned long next)
 {
-	return 0;
+	if (!pmd_leaf(pmdp_get(pmd)))
+		return 0;
+	vmemmap_verify((pte_t *)pmd, node, addr, next);
+
+	return 1;
 }
 
 int __meminit vmemmap_populate_hugepages(unsigned long start, unsigned long end,
Signature · 1 line
2.54.0
maintainer Acked-by on v3 1/5
#2 lore raw

Re: [PATCH v3 1/5] mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

Quoted 25 lines · Muchun Song

The two weak functions are currently no-ops on every architecture, forcing each platform that needs them to duplicate the same handful of lines. Provide a generic implementation:

- vmemmap_set_pmd() simply sets a huge PMD with PAGE_KERNEL protection.

- vmemmap_check_pmd() verifies that the PMD is present and leaf, then calls the existing vmemmap_verify() helper.

Architectures that need special handling can continue to override the weak symbols; everyone else gets the standard version for free.

---
v2->v3:
- Replace BUG_ON() with WARN_ON_ONCE() in vmemmap_set_pmd()

---
 mm/sparse-vmemmap.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
mm/sparse-vmemmap.c +0 -0
index 112ccf9c71ca..99e2be39671b 100644

mm/sparse-vmemmap.c int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end, · line 389

Quoted 4 lines
mm/sparse-vmemmap.c +1 -0
@@ -386,12 +386,17 @@ int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end,
 void __weak __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
 				      unsigned long addr, unsigned long next)
 {
+	WARN_ON_ONCE(!pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL));

Not sure if a VM_WARN_ON_ONCE() would be appropriate. (then, we have to move the pmd_set_huge() out of the statement).

Signature · 3 lines
Cheers,

David
#3 lore raw

Re: [PATCH v3 1/5] mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

Quoted 25 lines · David Hildenbrand (Arm)
Quoted 20 lines · Muchun Song

The two weak functions are currently no-ops on every architecture, forcing each platform that needs them to duplicate the same handful of lines. Provide a generic implementation:

- vmemmap_set_pmd() simply sets a huge PMD with PAGE_KERNEL protection.

- vmemmap_check_pmd() verifies that the PMD is present and leaf, then calls the existing vmemmap_verify() helper.

Architectures that need special handling can continue to override the weak symbols; everyone else gets the standard version for free.

---
v2->v3:
- Replace BUG_ON() with WARN_ON_ONCE() in vmemmap_set_pmd()
---
mm/sparse-vmemmap.c | 7 ++++++-

1 file changed, 6 insertions(+), 1 deletion(-)
Quoted 13 lines

mm/sparse-vmemmap.c int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end, · line 389

Quoted 9 lines
mm/sparse-vmemmap.c +1 -0
index 112ccf9c71ca..99e2be39671b 100644
@@ -386,12 +386,17 @@ int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end,
 void __weak __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
      unsigned long addr, unsigned long next)
 {
+ 	WARN_ON_ONCE(!pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL));

Not sure if a VM_WARN_ON_ONCE() would be appropriate. (then, we have to move the pmd_set_huge() out of the statement).

I think it might be better to keep WARN_ON_ONCE here. This way, we can still monitor for warnings in production while keeping the code simple.

Quoted 2 lines

Thanks.

Quoted 6 lines
Signature · 4 lines
-- 
Cheers,

David
#4 lore raw

Re: [PATCH v3 1/5] mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

Quoted 39 lines · Muchun Song
Quoted 25 lines · David Hildenbrand (Arm)
Quoted 20 lines · Muchun Song

The two weak functions are currently no-ops on every architecture, forcing each platform that needs them to duplicate the same handful of lines. Provide a generic implementation:

- vmemmap_set_pmd() simply sets a huge PMD with PAGE_KERNEL protection.

- vmemmap_check_pmd() verifies that the PMD is present and leaf, then calls the existing vmemmap_verify() helper.

Architectures that need special handling can continue to override the weak symbols; everyone else gets the standard version for free.

---
v2->v3:
- Replace BUG_ON() with WARN_ON_ONCE() in vmemmap_set_pmd()
---
mm/sparse-vmemmap.c | 7 ++++++-

1 file changed, 6 insertions(+), 1 deletion(-)
Quoted 13 lines

mm/sparse-vmemmap.c int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end, · line 389

Quoted 9 lines
mm/sparse-vmemmap.c +1 -0
index 112ccf9c71ca..99e2be39671b 100644
@@ -386,12 +386,17 @@ int __meminit vmemmap_populate_hvo(unsigned long addr, unsigned long end,
 void __weak __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
      unsigned long addr, unsigned long next)
 {
+ 	WARN_ON_ONCE(!pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL));

Not sure if a VM_WARN_ON_ONCE() would be appropriate. (then, we have to move the pmd_set_huge() out of the statement).

Quoted 2 lines

I think it might be better to keep WARN_ON_ONCE here. This way, we can still monitor for warnings in production while keeping the code simple.

IIRC there was a discussion about this some time ago, given the fact that quite some people out there tend to have 'panic_on_warn', but then again, if we fail to set a PMD here it means something pretty nasty happened, so I am ok with that.

Signature · 2 lines
Oscar Salvador
SUSE Labs
Acked-by on v3 1/5
#5 lore raw

Re: [PATCH v3 1/5] mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

In reply to Muchun Song
Quoted 6 lines · Muchun Song

The two weak functions are currently no-ops on every architecture, forcing each platform that needs them to duplicate the same handful of lines. Provide a generic implementation:

- vmemmap_set_pmd() simply sets a huge PMD with PAGE_KERNEL protection.

Quoted 7 lines

- vmemmap_check_pmd() verifies that the PMD is present and leaf, then calls the existing vmemmap_verify() helper.

Architectures that need special handling can continue to override the weak symbols; everyone else gets the standard version for free.

Signature · 2 lines
Oscar Salvador
SUSE Labs
#6 lore raw

PATCH v3 2/5 arm64/mm: drop vmemmap_pmd helpers and use generic code

In reply to Muchun Song

The generic implementations now suffice; remove the arm64 copies.

---
v2->v3:
- Collect Acked-by from Will Deacon

---
 arch/arm64/mm/mmu.c | 14 --------------
 1 file changed, 14 deletions(-)
arch/arm64/mm/mmu.c +0 -14
index 24388e5c727d..ea10b0ebecd7 100644
@@ -1776,20 +1776,6 @@ static void free_empty_tables(unsigned long addr, unsigned long end,
 }
 #endif
 
-void __meminit vmemmap_set_pmd(pmd_t *pmdp, void *p, int node,
-			       unsigned long addr, unsigned long next)
-{
-	pmd_set_huge(pmdp, __pa(p), __pgprot(PROT_SECT_NORMAL));
-}
-
-int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
-				unsigned long addr, unsigned long next)
-{
-	vmemmap_verify((pte_t *)pmdp, node, addr, next);
-
-	return pmd_leaf(READ_ONCE(*pmdp));
-}
-
 int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
 		struct vmem_altmap *altmap)
 {
Signature · 1 line
2.54.0
Reviewed-by on v3 2/5
#8 lore raw

Re: [PATCH v3 2/5] arm64/mm: drop vmemmap_pmd helpers and use generic code

In reply to Muchun Song
Quoted 4 lines · Muchun Song

The generic implementations now suffice; remove the arm64 copies.

Quoted 36 lines

---
v2->v3:
- Collect Acked-by from Will Deacon

---
 arch/arm64/mm/mmu.c | 14 --------------
 1 file changed, 14 deletions(-)
arch/arm64/mm/mmu.c +0 -14
index 24388e5c727d..ea10b0ebecd7 100644
@@ -1776,20 +1776,6 @@ static void free_empty_tables(unsigned long addr, unsigned long end,
 }
 #endif
 
-void __meminit vmemmap_set_pmd(pmd_t *pmdp, void *p, int node,
-			       unsigned long addr, unsigned long next)
-{
-	pmd_set_huge(pmdp, __pa(p), __pgprot(PROT_SECT_NORMAL));
-}
-
-int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
-				unsigned long addr, unsigned long next)
-{
-	vmemmap_verify((pte_t *)pmdp, node, addr, next);
-
-	return pmd_leaf(READ_ONCE(*pmdp));
-}
-
 int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
 		struct vmem_altmap *altmap)
 {
Signature · 1 line
2.54.0
Signature · 2 lines
Oscar Salvador
SUSE Labs
#9 lore raw

PATCH v3 3/5 riscv/mm: drop vmemmap_pmd helpers and use generic code

In reply to Muchun Song

The generic implementations now suffice; remove the riscv copies.

---
 arch/riscv/mm/init.c | 13 -------------
 1 file changed, 13 deletions(-)
arch/riscv/mm/init.c +0 -13
index 885f1db4e9bf..5f680eb83e86 100644
@@ -1359,19 +1359,6 @@ void __init misc_mem_init(void)
 }
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
-void __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
-			       unsigned long addr, unsigned long next)
-{
-	pmd_set_huge(pmd, virt_to_phys(p), PAGE_KERNEL);
-}
-
-int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
-				unsigned long addr, unsigned long next)
-{
-	vmemmap_verify((pte_t *)pmdp, node, addr, next);
-	return 1;
-}
-
 int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
 			       struct vmem_altmap *altmap)
 {
Signature · 1 line
2.54.0
#12 lore raw

PATCH v3 4/5 loongarch/mm: drop vmemmap_check_pmd helper and use generic code

In reply to Muchun Song

The generic implementations now suffice; remove the loongarch copies.

---
 arch/loongarch/mm/init.c | 11 -----------
 1 file changed, 11 deletions(-)
arch/loongarch/mm/init.c +0 -11
index 687980b6e91f..3407030f3e7a 100644
@@ -140,17 +140,6 @@ void __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
 	set_pmd_at(&init_mm, addr, pmd, entry);
 }
 
-int __meminit vmemmap_check_pmd(pmd_t *pmd, int node,
-				unsigned long addr, unsigned long next)
-{
-	int huge = pmd_val(pmdp_get(pmd)) & _PAGE_HUGE;
-
-	if (huge)
-		vmemmap_verify((pte_t *)pmd, node, addr, next);
-
-	return huge;
-}
-
 int __meminit vmemmap_populate(unsigned long start, unsigned long end,
 			       int node, struct vmem_altmap *altmap)
 {
Signature · 1 line
2.54.0
maintainer Reviewed-by on v3 4/5
#13 lore raw

Re: [PATCH v3 4/5] loongarch/mm: drop vmemmap_check_pmd helper and use generic code

Quoted 1 line · Muchun Song

The generic implementations now suffice; remove the loongarch copies.

You are only removing a single function? So "copy"

Signature · 3 lines
Cheers,

David
#14 lore raw

Re: [PATCH v3 4/5] loongarch/mm: drop vmemmap_check_pmd helper and use generic code

Quoted 8 lines · David Hildenbrand (Arm)
Quoted 1 line · Muchun Song

The generic implementations now suffice; remove the loongarch copies.

You are only removing a single function? So "copy"

Yes. I copied the commit message from other commit and forgot to update accordingly.

Quoted 3 lines

Thanks.

Quoted 5 lines
Signature · 4 lines
-- 
Cheers,

David
#16 lore raw

PATCH v3 5/5 sparc/mm: drop vmemmap_check_pmd helper and use generic code

In reply to Muchun Song

The generic implementations now suffice; remove the sparc copies.

---
 arch/sparc/mm/init_64.c | 11 -----------
 1 file changed, 11 deletions(-)
arch/sparc/mm/init_64.c +0 -11
index 3b679b1d1d72..103db4683b16 100644
@@ -2559,17 +2559,6 @@ void __meminit vmemmap_set_pmd(pmd_t *pmd, void *p, int node,
 	pmd_val(*pmd) = pte_base | __pa(p);
 }
 
-int __meminit vmemmap_check_pmd(pmd_t *pmdp, int node,
-				unsigned long addr, unsigned long next)
-{
-	int large = pmd_leaf(*pmdp);
-
-	if (large)
-		vmemmap_verify((pte_t *)pmdp, node, addr, next);
-
-	return large;
-}
-
 int __meminit vmemmap_populate(unsigned long vstart, unsigned long vend,
 			       int node, struct vmem_altmap *altmap)
 {
Signature · 1 line
2.54.0
#19 lore raw
In reply to Muchun Song
This series was applied to riscv/linux.git (fixes) by Andrew Morton <[email protected]>:

Hello:

This series was applied to riscv/linux.git (fixes) by Andrew Morton <[email protected]>:

Quoted 10 lines · On Mon, 1 Jun 2026 16:48:39 +0800 you wrote:

The weak vmemmap_set_pmd() and vmemmap_check_pmd() hooks are currently no-ops in the generic code, which leaves architectures that need PMD-level handling to open-code the same logic locally.

This series provides generic implementations for both helpers in mm/sparse-vmemmap.c. vmemmap_set_pmd() installs a huge PMD with PAGE_KERNEL protection, and vmemmap_check_pmd() verifies a present leaf PMD before reusing the existing vmemmap_verify() helper.

[...]

Here is the summary with links:
- [v3,1/5] mm/sparse-vmemmap: provide generic vmemmap_set_pmd() and vmemmap_check_pmd()

    https://git.kernel.org/riscv/c/0b6073ff1574
  - [v3,2/5] arm64/mm: drop vmemmap_pmd helpers and use generic code
    https://git.kernel.org/riscv/c/f521f198b50a
  - [v3,3/5] riscv/mm: drop vmemmap_pmd helpers and use generic code
    https://git.kernel.org/riscv/c/abff0ecf7602
  - [v3,4/5] loongarch/mm: drop vmemmap_check_pmd helper and use generic code
    https://git.kernel.org/riscv/c/ecca7da924b1
  - [v3,5/5] sparc/mm: drop vmemmap_check_pmd helper and use generic code
    https://git.kernel.org/riscv/c/d3d58e946900

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html