[PATCH 0/6] mm: preparatory patches for PMD level swap entries

[PATCH 0/6] mm: preparatory patches for PMD level swap entries

[Usama Arif]

This is the preparatory part of the PMD page table swapin work.  The
full PMD swap entry series has been split into two parts:

  1. this preparatory series, which contains the first 6 patches. Zi [1]
     and Lance [2] suggested to separate this out from the core series.
  2. the PMD swap entry core series, which depends on this one. I will
     send this once the preparatory series is merged in mm-new as v3
     as the combined is currently at v2 [1].
     I have not marked this prep series as v3, as its not really adding
     support for PMD swap entries.

This series does not introduce PMD swap entries and does not install any
new page-table entry type.  It only cleans up existing PMD softleaf
helpers and call sites so the follow-up PMD swap entry series can be
smaller and easier to review.

It should be safe to merge independently.  The patches are either helper
additions, refactors of existing open-coded logic, defensive checks that
preserve current migration/device-private behavior, or a mechanical
rename of the PMD softleaf Kconfig gate.  The follow-up series depends
on these helpers, but this series does not depend on the follow-up
series.

Patch breakdown:

  1. mm: add softleaf_to_pmd() and convert existing callers
     Add the PMD counterpart to softleaf_to_pte() and convert existing
     swp_entry_to_pmd() users that are constructing PMD softleaf
     entries.

  2. mm: extract mm_prepare_for_swap_entries() helper
     Hoist the "register mm with swapoff" double-checked-locking
     pattern out of try_to_unmap_one() and copy_nonpresent_pte() so
     future PMD-level users do not need another open-coded copy.

  3. fs/proc: use softleaf_has_pfn() in pagemap PMD walker
     Avoid assuming every non-present PMD softleaf entry encodes a PFN.
     Existing migration/device-private behavior is preserved.

  4. mm/huge_memory: move softleaf_to_folio() inside migration branch
     Keep the folio lookup in change_non_present_huge_pmd() scoped to
     the migration-entry branch that actually needs it.

  5. mm/migrate_device: move softleaf_to_folio() inside device-private
     branch
     Apply the same ordering cleanup to migrate_vma_collect_pmd(): only
     derive a folio after confirming the PMD entry is device-private.

  6. mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF
     Rename the architecture gate to describe what it actually enables:
     PMD softleaf entries.  Migration remains the only current user in
     this series; the follow-up series adds PMD swap entries.

The patches are based on akpm/mm-new as of 29 June
(git hash: 15654b0268cd65b4e673510c0377774e61b86397)

[1] https://lore.kernel.org/all/6E99CC4E-A026-4DE3-8A5A-34216771F521@nvidia.com/
[2] https://lore.kernel.org/all/b08cafbb-a4b7-4609-84ae-dbb2cfcfc8be@linux.dev/#t 
[3] https://lore.kernel.org/all/20260602142537.198755-1-usama.arif@linux.dev/

Usama Arif (6):
  mm: add softleaf_to_pmd() and convert existing callers
  mm: extract mm_prepare_for_swap_entries() helper
  fs/proc: use softleaf_has_pfn() in pagemap PMD walker
  mm/huge_memory: move softleaf_to_folio() inside migration branch
  mm/migrate_device: move softleaf_to_folio() inside device-private
    branch
  mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

 arch/arm64/Kconfig                           |  2 +-
 arch/arm64/include/asm/pgtable.h             |  4 +--
 arch/loongarch/Kconfig                       |  2 +-
 arch/powerpc/include/asm/book3s/64/pgtable.h |  2 +-
 arch/powerpc/platforms/Kconfig.cputype       |  2 +-
 arch/riscv/Kconfig                           |  2 +-
 arch/riscv/include/asm/pgtable.h             |  8 +++---
 arch/s390/Kconfig                            |  2 +-
 arch/s390/include/asm/pgtable.h              |  2 +-
 arch/x86/Kconfig                             |  2 +-
 arch/x86/include/asm/pgtable.h               |  2 +-
 fs/proc/task_mmu.c                           |  3 ++-
 include/linux/huge_mm.h                      |  2 +-
 include/linux/leafops.h                      | 28 +++++++++++++++++---
 include/linux/pgtable.h                      |  2 +-
 include/linux/swapops.h                      |  6 ++---
 mm/Kconfig                                   |  2 +-
 mm/debug_vm_pgtable.c                        | 12 ++++-----
 mm/hmm.c                                     |  4 +--
 mm/huge_memory.c                             | 17 ++++++------
 mm/internal.h                                | 13 +++++++++
 mm/memory.c                                  |  9 +------
 mm/migrate.c                                 |  4 +--
 mm/migrate_device.c                          | 19 +++++++------
 mm/rmap.c                                    |  9 ++-----
 25 files changed, 93 insertions(+), 67 deletions(-)

-- 
2.53.0-Meta

[PATCH 1/6] mm: add softleaf_to_pmd() and convert existing callers

[Usama Arif]

Add softleaf_to_pmd() as the PMD counterpart to softleaf_to_pte(),
completing the symmetry of the softleaf abstraction for page table
leaf entries.

The upcoming PMD swap entry support needs to construct PMD entries
from swap entries. Converting existing swp_entry_to_pmd() callers
to softleaf_to_pmd() in a prep patch keeps the feature patches
focused on new functionality rather than mixing refactoring with
new code.

Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 include/linux/leafops.h | 20 ++++++++++++++++++++
 mm/debug_vm_pgtable.c   |  4 ++--
 mm/huge_memory.c        | 12 ++++++------
 mm/migrate_device.c     |  2 +-
 4 files changed, 29 insertions(+), 9 deletions(-)

diff --git a/include/linux/leafops.h b/include/linux/leafops.h
index 992cd8bd8ed0..803d312437df 100644
--- a/include/linux/leafops.h
+++ b/include/linux/leafops.h
@@ -108,6 +108,21 @@ static inline softleaf_t softleaf_from_pmd(pmd_t pmd)
 	return swp_entry(__swp_type(arch_entry), __swp_offset(arch_entry));
 }
 
+/**
+ * softleaf_to_pmd() - Obtain a PMD entry from a leaf entry.
+ * @entry: Leaf entry.
+ *
+ * This generates an architecture-specific PMD entry that can be utilised to
+ * encode the metadata the leaf entry encodes.
+ *
+ * Returns: Architecture-specific PMD entry encoding leaf entry.
+ */
+static inline pmd_t softleaf_to_pmd(softleaf_t entry)
+{
+	/* Temporary until swp_entry_t eliminated. */
+	return swp_entry_to_pmd(entry);
+}
+
 #else
 
 static inline softleaf_t softleaf_from_pmd(pmd_t pmd)
@@ -115,6 +130,11 @@ static inline softleaf_t softleaf_from_pmd(pmd_t pmd)
 	return softleaf_mk_none();
 }
 
+static inline pmd_t softleaf_to_pmd(softleaf_t entry)
+{
+	return __pmd(0);
+}
+
 #endif
 
 /**
diff --git a/mm/debug_vm_pgtable.c b/mm/debug_vm_pgtable.c
index 23dc3ee09561..18411fb09aab 100644
--- a/mm/debug_vm_pgtable.c
+++ b/mm/debug_vm_pgtable.c
@@ -758,7 +758,7 @@ static void __init pmd_leaf_soft_dirty_tests(struct pgtable_debug_args *args)
 		return;
 
 	pr_debug("Validating PMD swap soft dirty\n");
-	pmd = swp_entry_to_pmd(args->leaf_entry);
+	pmd = softleaf_to_pmd(args->leaf_entry);
 	WARN_ON(!pmd_is_huge(pmd));
 	WARN_ON(!pmd_is_valid_softleaf(pmd));
 
@@ -829,7 +829,7 @@ static void __init pmd_softleaf_tests(struct pgtable_debug_args *args)
 		return;
 
 	pr_debug("Validating PMD swap\n");
-	pmd1 = swp_entry_to_pmd(args->leaf_entry);
+	pmd1 = softleaf_to_pmd(args->leaf_entry);
 	WARN_ON(!pmd_is_huge(pmd1));
 	WARN_ON(!pmd_is_valid_softleaf(pmd1));
 
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 2bccb0a53a0a..d51f53cfbe4f 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1819,7 +1819,7 @@ static void copy_huge_non_present_pmd(
 	if (softleaf_is_migration_write(entry) ||
 	    softleaf_is_migration_read_exclusive(entry)) {
 		entry = make_readable_migration_entry(swp_offset(entry));
-		pmd = swp_entry_to_pmd(entry);
+		pmd = softleaf_to_pmd(entry);
 		if (pmd_swp_soft_dirty(*src_pmd))
 			pmd = pmd_swp_mksoft_dirty(pmd);
 		if (pmd_swp_uffd_wp(*src_pmd))
@@ -1832,7 +1832,7 @@ static void copy_huge_non_present_pmd(
 		 */
 		if (softleaf_is_device_private_write(entry)) {
 			entry = make_readable_device_private_entry(swp_offset(entry));
-			pmd = swp_entry_to_pmd(entry);
+			pmd = softleaf_to_pmd(entry);
 
 			if (pmd_swp_soft_dirty(*src_pmd))
 				pmd = pmd_swp_mksoft_dirty(pmd);
@@ -2570,12 +2570,12 @@ static void change_non_present_huge_pmd(struct mm_struct *mm,
 			entry = make_readable_exclusive_migration_entry(swp_offset(entry));
 		else
 			entry = make_readable_migration_entry(swp_offset(entry));
-		newpmd = swp_entry_to_pmd(entry);
+		newpmd = softleaf_to_pmd(entry);
 		if (pmd_swp_soft_dirty(*pmd))
 			newpmd = pmd_swp_mksoft_dirty(newpmd);
 	} else if (softleaf_is_device_private_write(entry)) {
 		entry = make_readable_device_private_entry(swp_offset(entry));
-		newpmd = swp_entry_to_pmd(entry);
+		newpmd = softleaf_to_pmd(entry);
 		if (pmd_swp_uffd_wp(*pmd))
 			newpmd = pmd_swp_mkuffd_wp(newpmd);
 	} else {
@@ -4923,7 +4923,7 @@ int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
 	}
 
 	/* Set PMD. */
-	pmdswp = swp_entry_to_pmd(entry);
+	pmdswp = softleaf_to_pmd(entry);
 	if (softdirty)
 		pmdswp = pmd_swp_mksoft_dirty(pmdswp);
 	if (uffd_wp)
@@ -4976,7 +4976,7 @@ void remove_migration_pmd(struct page_vma_mapped_walk *pvmw, struct page *new)
 		else
 			entry = make_readable_device_private_entry(
 							page_to_pfn(new));
-		pmde = swp_entry_to_pmd(entry);
+		pmde = softleaf_to_pmd(entry);
 
 		if (pmd_swp_soft_dirty(*pvmw->pmd))
 			pmde = pmd_swp_mksoft_dirty(pmde);
diff --git a/mm/migrate_device.c b/mm/migrate_device.c
index ae39173d6a0e..aa948db49501 100644
--- a/mm/migrate_device.c
+++ b/mm/migrate_device.c
@@ -835,7 +835,7 @@ static int migrate_vma_insert_huge_pmd_page(struct migrate_vma *migrate,
 		else
 			swp_entry = make_readable_device_private_entry(
 						page_to_pfn(page));
-		entry = swp_entry_to_pmd(swp_entry);
+		entry = softleaf_to_pmd(swp_entry);
 	} else {
 		if (folio_is_zone_device(folio) &&
 		    !folio_is_device_coherent(folio)) {
-- 
2.53.0-Meta

[PATCH 2/6] mm: extract mm_prepare_for_swap_entries() helper

[Usama Arif]

When a swap entry is installed in a page table, the mm must be added
to init_mm.mmlist so that swapoff can find and unuse its swap entries.
This double-checked locking pattern is currently open-coded in
try_to_unmap_one() and copy_nonpresent_pte().

Move it into mm_prepare_for_swap_entries() in mm/internal.h and convert
both callers so it can be reused by upcoming PMD-level swap entry code
paths that also need to register the mm with swapoff.

copy_nonpresent_pte() previously inserted into &src_mm->mmlist rather
than &init_mm.mmlist, but the insertion point is irrelevant, mmlist
is a circular list and swapoff walks it entirely from init_mm.mmlist,
so only membership matters, not position.

Reviewed-by: Dev Jain <dev.jain@arm.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 mm/internal.h | 13 +++++++++++++
 mm/memory.c   |  9 +--------
 mm/rmap.c     |  7 +------
 3 files changed, 15 insertions(+), 14 deletions(-)

diff --git a/mm/internal.h b/mm/internal.h
index 8e0df3d1c6f9..0e3db9451e5c 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -1956,4 +1956,17 @@ static inline int get_sysctl_max_map_count(void)
 bool may_expand_vm(struct mm_struct *mm, const vma_flags_t *vma_flags,
 		   unsigned long npages);
 
+/*
+ * Ensure @mm is on the init_mm.mmlist so swapoff can find it.
+ */
+static inline void mm_prepare_for_swap_entries(struct mm_struct *mm)
+{
+	if (list_empty(&mm->mmlist)) {
+		spin_lock(&mmlist_lock);
+		if (list_empty(&mm->mmlist))
+			list_add(&mm->mmlist, &init_mm.mmlist);
+		spin_unlock(&mmlist_lock);
+	}
+}
+
 #endif	/* __MM_INTERNAL_H */
diff --git a/mm/memory.c b/mm/memory.c
index a3fcaf8cfe8f..6637c5b13c9b 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -953,14 +953,7 @@ copy_nonpresent_pte(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 		if (swap_dup_entry_direct(entry) < 0)
 			return -EIO;
 
-		/* make sure dst_mm is on swapoff's mmlist. */
-		if (unlikely(list_empty(&dst_mm->mmlist))) {
-			spin_lock(&mmlist_lock);
-			if (list_empty(&dst_mm->mmlist))
-				list_add(&dst_mm->mmlist,
-						&src_mm->mmlist);
-			spin_unlock(&mmlist_lock);
-		}
+		mm_prepare_for_swap_entries(dst_mm);
 		/* Mark the swap entry as shared. */
 		if (pte_swp_exclusive(orig_pte)) {
 			pte = pte_swp_clear_exclusive(orig_pte);
diff --git a/mm/rmap.c b/mm/rmap.c
index 1c77d5dc06e9..b93caabd186f 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -2304,12 +2304,7 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
 				set_pte_at(mm, address, pvmw.pte, pteval);
 				goto walk_abort;
 			}
-			if (list_empty(&mm->mmlist)) {
-				spin_lock(&mmlist_lock);
-				if (list_empty(&mm->mmlist))
-					list_add(&mm->mmlist, &init_mm.mmlist);
-				spin_unlock(&mmlist_lock);
-			}
+			mm_prepare_for_swap_entries(mm);
 			dec_mm_counter(mm, MM_ANONPAGES);
 			inc_mm_counter(mm, MM_SWAPENTS);
 			swp_pte = swp_entry_to_pte(entry);
-- 
2.53.0-Meta

Re: [PATCH 2/6] mm: extract mm_prepare_for_swap_entries() helper

[David Hildenbrand (Arm)]

On 6/30/26 18:34, Usama Arif wrote:
> When a swap entry is installed in a page table, the mm must be added
> to init_mm.mmlist so that swapoff can find and unuse its swap entries.
> This double-checked locking pattern is currently open-coded in
> try_to_unmap_one() and copy_nonpresent_pte().
> 
> Move it into mm_prepare_for_swap_entries() in mm/internal.h and convert
> both callers so it can be reused by upcoming PMD-level swap entry code
> paths that also need to register the mm with swapoff.
> 
> copy_nonpresent_pte() previously inserted into &src_mm->mmlist rather
> than &init_mm.mmlist, but the insertion point is irrelevant, mmlist
> is a circular list and swapoff walks it entirely from init_mm.mmlist,
> so only membership matters, not position.
> 
> Reviewed-by: Dev Jain <dev.jain@arm.com>
> Reviewed-by: Zi Yan <ziy@nvidia.com>
> Signed-off-by: Usama Arif <usama.arif@linux.dev>
> ---
>  mm/internal.h | 13 +++++++++++++
>  mm/memory.c   |  9 +--------
>  mm/rmap.c     |  7 +------
>  3 files changed, 15 insertions(+), 14 deletions(-)
> 
> diff --git a/mm/internal.h b/mm/internal.h
> index 8e0df3d1c6f9..0e3db9451e5c 100644
> --- a/mm/internal.h
> +++ b/mm/internal.h
> @@ -1956,4 +1956,17 @@ static inline int get_sysctl_max_map_count(void)
>  bool may_expand_vm(struct mm_struct *mm, const vma_flags_t *vma_flags,
>  		   unsigned long npages);
>  
> +/*
> + * Ensure @mm is on the init_mm.mmlist so swapoff can find it.
> + */

Comment fits into a single line and I don't think we need it (function name is
self-explaining enough).

> +static inline void mm_prepare_for_swap_entries(struct mm_struct *mm)
> +{
> +	if (list_empty(&mm->mmlist)) {
> +		spin_lock(&mmlist_lock);
> +		if (list_empty(&mm->mmlist))
> +			list_add(&mm->mmlist, &init_mm.mmlist);
> +		spin_unlock(&mmlist_lock);
> +	}
> +}
> +

Acked-by: David Hildenbrand (Arm) <david@kernel.org>

-- 
Cheers,

David

[PATCH 3/6] fs/proc: use softleaf_has_pfn() in pagemap PMD walker

[Usama Arif]

pagemap_pmd_range_thp() assumes that every non-present PMD is a
migration entry and unconditionally calls softleaf_to_page().  This
will crash on any non-present PMD type that does not encode a PFN,
such as the upcoming PMD-level swap entries.

Guard the page lookup with softleaf_has_pfn(), matching how
pte_to_pagemap_entry() already handles non-present PTEs.

Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 fs/proc/task_mmu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index d32408f7cd5e..1fb5acd88ad0 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -2129,7 +2129,8 @@ static int pagemap_pmd_range_thp(pmd_t *pmdp, unsigned long addr,
 			flags |= PM_SOFT_DIRTY;
 		if (pmd_swp_uffd_wp(pmd))
 			flags |= PM_UFFD_WP;
-		page = softleaf_to_page(entry);
+		if (softleaf_has_pfn(entry))
+			page = softleaf_to_page(entry);
 	}
 
 	if (page) {
-- 
2.53.0-Meta

[PATCH 4/6] mm/huge_memory: move softleaf_to_folio() inside migration branch

[Usama Arif]

change_non_present_huge_pmd() calls softleaf_to_folio() unconditionally
at the top of the function.  softleaf_to_folio() extracts a PFN from
the entry and converts it to a folio pointer, which is only meaningful
for migration and device_private entries that encode a real PFN.

A swap entry encodes a swap offset instead, so softleaf_to_folio()
would produce a bogus pointer and crash on mprotect() when a PMD swap
entry is present.

Move the call into the migration_write branch where the folio is
actually used, so the function is safe for any non-present PMD type.

Acked-by: David Hildenbrand (Arm) <david@kernel.org>
Reviewed-by: Dev Jain <dev.jain@arm.com>
Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 mm/huge_memory.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index d51f53cfbe4f..4aca9cdff73f 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2557,11 +2557,12 @@ static void change_non_present_huge_pmd(struct mm_struct *mm,
 		bool uffd_wp_resolve)
 {
 	softleaf_t entry = softleaf_from_pmd(*pmd);
-	const struct folio *folio = softleaf_to_folio(entry);
 	pmd_t newpmd;
 
 	VM_WARN_ON(!pmd_is_valid_softleaf(*pmd));
 	if (softleaf_is_migration_write(entry)) {
+		const struct folio *folio = softleaf_to_folio(entry);
+
 		/*
 		 * A protection check is difficult so
 		 * just be safe and disable write
-- 
2.53.0-Meta

[PATCH 5/6] mm/migrate_device: move softleaf_to_folio() inside device-private branch

[Usama Arif]

migrate_vma_collect_pmd() calls softleaf_to_folio() on a non-present
PMD before checking the entry's type.  softleaf_to_folio() converts
the entry's offset to a PFN, which is only meaningful for migration
or device-private entries.

A PMD swap entry's offset is a swap offset, not a PFN, so the
lookup would either return a bogus folio pointer or trip pfn_to_page
validation on a debug kernel.  In the non-device-private path the
returned folio is then unused (the OR short-circuits to
migrate_vma_collect_skip()), but the lookup itself is already
unsafe.

Move the softleaf_to_folio() call inside the device-private branch
where the folio is actually needed, mirroring the equivalent
change_non_present_huge_pmd() fix.

Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 mm/migrate_device.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/mm/migrate_device.c b/mm/migrate_device.c
index aa948db49501..36287e958b2a 100644
--- a/mm/migrate_device.c
+++ b/mm/migrate_device.c
@@ -166,11 +166,14 @@ static int migrate_vma_collect_huge_pmd(pmd_t *pmdp, unsigned long start,
 	} else if (!pmd_present(*pmdp)) {
 		const softleaf_t entry = softleaf_from_pmd(*pmdp);
 
-		folio = softleaf_to_folio(entry);
-
 		if (!softleaf_is_device_private(entry) ||
-			!(migrate->flags & MIGRATE_VMA_SELECT_DEVICE_PRIVATE) ||
-			(folio->pgmap->owner != migrate->pgmap_owner)) {
+		    !(migrate->flags & MIGRATE_VMA_SELECT_DEVICE_PRIVATE)) {
+			spin_unlock(ptl);
+			return migrate_vma_collect_skip(start, end, walk);
+		}
+
+		folio = softleaf_to_folio(entry);
+		if (folio->pgmap->owner != migrate->pgmap_owner) {
 			spin_unlock(ptl);
 			return migrate_vma_collect_skip(start, end, walk);
 		}
-- 
2.53.0-Meta

Re: [PATCH 5/6] mm/migrate_device: move softleaf_to_folio() inside device-private branch

[David Hildenbrand (Arm)]

On 6/30/26 18:34, Usama Arif wrote:
> migrate_vma_collect_pmd() calls softleaf_to_folio() on a non-present
> PMD before checking the entry's type.  softleaf_to_folio() converts
> the entry's offset to a PFN, which is only meaningful for migration
> or device-private entries.
> 
> A PMD swap entry's offset is a swap offset, not a PFN, so the
> lookup would either return a bogus folio pointer or trip pfn_to_page
> validation on a debug kernel.  In the non-device-private path the
> returned folio is then unused (the OR short-circuits to
> migrate_vma_collect_skip()), but the lookup itself is already
> unsafe.
> 
> Move the softleaf_to_folio() call inside the device-private branch
> where the folio is actually needed, mirroring the equivalent
> change_non_present_huge_pmd() fix.
> 
> Reviewed-by: Zi Yan <ziy@nvidia.com>
> Signed-off-by: Usama Arif <usama.arif@linux.dev>
> ---
>  mm/migrate_device.c | 11 +++++++----
>  1 file changed, 7 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/migrate_device.c b/mm/migrate_device.c
> index aa948db49501..36287e958b2a 100644
> --- a/mm/migrate_device.c
> +++ b/mm/migrate_device.c
> @@ -166,11 +166,14 @@ static int migrate_vma_collect_huge_pmd(pmd_t *pmdp, unsigned long start,
>  	} else if (!pmd_present(*pmdp)) {
>  		const softleaf_t entry = softleaf_from_pmd(*pmdp);
>  
> -		folio = softleaf_to_folio(entry);
> -
>  		if (!softleaf_is_device_private(entry) ||
> -			!(migrate->flags & MIGRATE_VMA_SELECT_DEVICE_PRIVATE) ||
> -			(folio->pgmap->owner != migrate->pgmap_owner)) {
> +		    !(migrate->flags & MIGRATE_VMA_SELECT_DEVICE_PRIVATE)) {
> +			spin_unlock(ptl);
> +			return migrate_vma_collect_skip(start, end, walk);
> +		}
> +
> +		folio = softleaf_to_folio(entry);
> +		if (folio->pgmap->owner != migrate->pgmap_owner) {
>  			spin_unlock(ptl);
>  			return migrate_vma_collect_skip(start, end, walk);
>  		}

LGTM

Acked-by: David Hildenbrand (Arm) <david@kernel.org>

-- 
Cheers,

David

[PATCH 6/6] mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

[Usama Arif]

CONFIG_ARCH_ENABLE_THP_MIGRATION started life gating just PMD-level
migration entries, but has grown to gate the entire PMD-level softleaf
machinery: migration entries, device-private entries, and soon swap
entries.

Rename CONFIG_ARCH_ENABLE_THP_MIGRATION to CONFIG_ARCH_SUPPORTS_PMD
_SOFTLEAF to make this clear.  This is a pure rename: the set of
selecting architectures (x86, arm64, s390, riscv, loongarch, and
powerpc on PPC_BOOK3S_64) and the gating semantics are unchanged.

No functional change intended.

Reviewed-by: Zi Yan <ziy@nvidia.com>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
 arch/arm64/Kconfig                           | 2 +-
 arch/arm64/include/asm/pgtable.h             | 4 ++--
 arch/loongarch/Kconfig                       | 2 +-
 arch/powerpc/include/asm/book3s/64/pgtable.h | 2 +-
 arch/powerpc/platforms/Kconfig.cputype       | 2 +-
 arch/riscv/Kconfig                           | 2 +-
 arch/riscv/include/asm/pgtable.h             | 8 ++++----
 arch/s390/Kconfig                            | 2 +-
 arch/s390/include/asm/pgtable.h              | 2 +-
 arch/x86/Kconfig                             | 2 +-
 arch/x86/include/asm/pgtable.h               | 2 +-
 include/linux/huge_mm.h                      | 2 +-
 include/linux/leafops.h                      | 8 ++++----
 include/linux/pgtable.h                      | 2 +-
 include/linux/swapops.h                      | 6 +++---
 mm/Kconfig                                   | 2 +-
 mm/debug_vm_pgtable.c                        | 8 ++++----
 mm/hmm.c                                     | 4 ++--
 mm/huge_memory.c                             | 2 +-
 mm/migrate.c                                 | 4 ++--
 mm/migrate_device.c                          | 6 +++---
 mm/rmap.c                                    | 2 +-
 22 files changed, 38 insertions(+), 38 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index b3afe0688919..3eb94f36c77e 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -17,7 +17,7 @@ config ARM64
 	select ARCH_ENABLE_HUGEPAGE_MIGRATION if HUGETLB_PAGE && MIGRATION
 	select ARCH_ENABLE_MEMORY_HOTPLUG
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
-	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
+	select ARCH_SUPPORTS_PMD_SOFTLEAF if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_CACHE_LINE_SIZE
 	select ARCH_HAS_CC_PLATFORM
 	select ARCH_HAS_CPU_CACHE_INVALIDATE_MEMREGION
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 27689c62bd25..984badfa9a74 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -1534,10 +1534,10 @@ static inline pmd_t pmdp_establish(struct vm_area_struct *vma,
 #define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
 #define __swp_entry_to_pte(swp)	((pte_t) { (swp).val })
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 #define __pmd_to_swp_entry(pmd)		((swp_entry_t) { pmd_val(pmd) })
 #define __swp_entry_to_pmd(swp)		__pmd((swp).val)
-#endif /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 /*
  * Ensure that there are not more swap files than can be encoded in the kernel
diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig
index d8d252325017..550142b7ee10 100644
--- a/arch/loongarch/Kconfig
+++ b/arch/loongarch/Kconfig
@@ -12,7 +12,7 @@ config LOONGARCH
 	select ARCH_NEEDS_DEFER_KASAN
 	select ARCH_DISABLE_KASAN_INLINE
 	select ARCH_ENABLE_MEMORY_HOTPLUG
-	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
+	select ARCH_SUPPORTS_PMD_SOFTLEAF if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_ACPI_TABLE_UPGRADE	if ACPI
 	select ARCH_HAS_CPU_FINALIZE_INIT
 	select ARCH_HAS_CURRENT_STACK_POINTER
diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
index e67e64ac6e8c..6f30aa8a6490 100644
--- a/arch/powerpc/include/asm/book3s/64/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
@@ -1060,7 +1060,7 @@ static inline pte_t *pmdp_ptep(pmd_t *pmd)
 #define pmd_mksoft_dirty(pmd)  pte_pmd(pte_mksoft_dirty(pmd_pte(pmd)))
 #define pmd_clear_soft_dirty(pmd) pte_pmd(pte_clear_soft_dirty(pmd_pte(pmd)))
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 #define pmd_swp_mksoft_dirty(pmd)	pte_pmd(pte_swp_mksoft_dirty(pmd_pte(pmd)))
 #define pmd_swp_soft_dirty(pmd)		pte_swp_soft_dirty(pmd_pte(pmd))
 #define pmd_swp_clear_soft_dirty(pmd)	pte_pmd(pte_swp_clear_soft_dirty(pmd_pte(pmd)))
diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
index bac02c83bb3e..4a0fa681bf98 100644
--- a/arch/powerpc/platforms/Kconfig.cputype
+++ b/arch/powerpc/platforms/Kconfig.cputype
@@ -112,7 +112,7 @@ config PPC_THP
        depends on PPC_RADIX_MMU || (PPC_64S_HASH_MMU && PAGE_SIZE_64KB)
        select HAVE_ARCH_TRANSPARENT_HUGEPAGE
        select HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
-       select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
+       select ARCH_SUPPORTS_PMD_SOFTLEAF if TRANSPARENT_HUGEPAGE
 
 choice
 	prompt "CPU selection"
diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index 3f0a647218e4..bf7b271cccd8 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -22,7 +22,7 @@ config RISCV
 	select ARCH_ENABLE_HUGEPAGE_MIGRATION if HUGETLB_PAGE && MIGRATION
 	select ARCH_ENABLE_MEMORY_HOTPLUG if SPARSEMEM_VMEMMAP
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
-	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
+	select ARCH_SUPPORTS_PMD_SOFTLEAF if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_BINFMT_FLAT
 	select ARCH_HAS_CC_CAN_LINK
 	select ARCH_HAS_CURRENT_STACK_POINTER
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index 5d5756bda82e..2aa529e882d3 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -946,7 +946,7 @@ static inline pmd_t pmd_clear_soft_dirty(pmd_t pmd)
 	return pte_pmd(pte_clear_soft_dirty(pmd_pte(pmd)));
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 static inline bool pmd_swp_soft_dirty(pmd_t pmd)
 {
 	return pte_swp_soft_dirty(pmd_pte(pmd));
@@ -961,7 +961,7 @@ static inline pmd_t pmd_swp_clear_soft_dirty(pmd_t pmd)
 {
 	return pte_pmd(pte_swp_clear_soft_dirty(pmd_pte(pmd)));
 }
-#endif /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 #endif /* CONFIG_HAVE_ARCH_SOFT_DIRTY */
 
 static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
@@ -1208,10 +1208,10 @@ static inline pte_t pte_swp_clear_exclusive(pte_t pte)
 	return __pte(pte_val(pte) & ~_PAGE_SWP_EXCLUSIVE);
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 #define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val(pmd) })
 #define __swp_entry_to_pmd(swp) __pmd((swp).val)
-#endif /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 /*
  * In the RV64 Linux scheme, we give the user half of the virtual-address space
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 84404e6778d5..0719f68de0f3 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -82,7 +82,7 @@ config S390
 	select ARCH_CORRECT_STACKTRACE_ON_KRETPROBE
 	select ARCH_ENABLE_MEMORY_HOTPLUG if SPARSEMEM
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if PGTABLE_LEVELS > 2
-	select ARCH_ENABLE_THP_MIGRATION if TRANSPARENT_HUGEPAGE
+	select ARCH_SUPPORTS_PMD_SOFTLEAF if TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_CC_CAN_LINK
 	select ARCH_HAS_CPU_FINALIZE_INIT
 	select ARCH_HAS_CURRENT_STACK_POINTER
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 859ce7c7d454..6faccfa63b09 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -903,7 +903,7 @@ static inline pmd_t pmd_clear_soft_dirty(pmd_t pmd)
 	return clear_pmd_bit(pmd, __pgprot(_SEGMENT_ENTRY_SOFT_DIRTY));
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 #define pmd_swp_soft_dirty(pmd)		pmd_soft_dirty(pmd)
 #define pmd_swp_mksoft_dirty(pmd)	pmd_mksoft_dirty(pmd)
 #define pmd_swp_clear_soft_dirty(pmd)	pmd_clear_soft_dirty(pmd)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index bdad90f210e4..50e00d8417aa 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -70,7 +70,7 @@ config X86
 	select ARCH_ENABLE_HUGEPAGE_MIGRATION if X86_64 && HUGETLB_PAGE && MIGRATION
 	select ARCH_ENABLE_MEMORY_HOTPLUG if X86_64
 	select ARCH_ENABLE_SPLIT_PMD_PTLOCK if (PGTABLE_LEVELS > 2) && (X86_64 || X86_PAE)
-	select ARCH_ENABLE_THP_MIGRATION if X86_64 && TRANSPARENT_HUGEPAGE
+	select ARCH_SUPPORTS_PMD_SOFTLEAF if X86_64 && TRANSPARENT_HUGEPAGE
 	select ARCH_HAS_ACPI_TABLE_UPGRADE	if ACPI
 	select ARCH_HAS_CPU_ATTACK_VECTORS	if CPU_MITIGATIONS
 	select ARCH_HAS_CACHE_LINE_SIZE
diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h
index ac295ca6c92f..e0fd318d4004 100644
--- a/arch/x86/include/asm/pgtable.h
+++ b/arch/x86/include/asm/pgtable.h
@@ -1545,7 +1545,7 @@ static inline pte_t pte_swp_clear_soft_dirty(pte_t pte)
 	return pte_clear_flags(pte, _PAGE_SWP_SOFT_DIRTY);
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 static inline pmd_t pmd_swp_mksoft_dirty(pmd_t pmd)
 {
 	return pmd_set_flags(pmd, _PAGE_SWP_SOFT_DIRTY);
diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
index ad20f7f8c179..1487bf4af1a7 100644
--- a/include/linux/huge_mm.h
+++ b/include/linux/huge_mm.h
@@ -567,7 +567,7 @@ static inline struct folio *get_persistent_huge_zero_folio(void)
 
 static inline bool thp_migration_supported(void)
 {
-	return IS_ENABLED(CONFIG_ARCH_ENABLE_THP_MIGRATION);
+	return IS_ENABLED(CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF);
 }
 
 void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address,
diff --git a/include/linux/leafops.h b/include/linux/leafops.h
index 803d312437df..88888daeb018 100644
--- a/include/linux/leafops.h
+++ b/include/linux/leafops.h
@@ -81,7 +81,7 @@ static inline pte_t softleaf_to_pte(softleaf_t entry)
 	return swp_entry_to_pte(entry);
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 /**
  * softleaf_from_pmd() - Obtain a leaf entry from a PMD entry.
  * @pmd: PMD entry.
@@ -587,7 +587,7 @@ static inline bool pte_is_uffd_marker(pte_t pte)
 	return false;
 }
 
-#if defined(CONFIG_ZONE_DEVICE) && defined(CONFIG_ARCH_ENABLE_THP_MIGRATION)
+#if defined(CONFIG_ZONE_DEVICE) && defined(CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF)
 
 /**
  * pmd_is_device_private_entry() - Check if PMD contains a device private swap
@@ -606,14 +606,14 @@ static inline bool pmd_is_device_private_entry(pmd_t pmd)
 	return softleaf_is_device_private(softleaf_from_pmd(pmd));
 }
 
-#else  /* CONFIG_ZONE_DEVICE && CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#else  /* CONFIG_ZONE_DEVICE && CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 static inline bool pmd_is_device_private_entry(pmd_t pmd)
 {
 	return false;
 }
 
-#endif /* CONFIG_ZONE_DEVICE && CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif /* CONFIG_ZONE_DEVICE && CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 /**
  * pmd_is_migration_entry() - Does this PMD entry encode a migration entry?
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index d52d2a976e5a..e38f069c1c91 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -1839,7 +1839,7 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
 #endif
 
 #ifdef CONFIG_HAVE_ARCH_SOFT_DIRTY
-#ifndef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifndef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 static inline pmd_t pmd_swp_mksoft_dirty(pmd_t pmd)
 {
 	return pmd;
diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index 8cfc966eae48..705a84154d28 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -321,7 +321,7 @@ static inline swp_entry_t make_guard_swp_entry(void)
 
 struct page_vma_mapped_walk;
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 extern int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
 		struct page *page);
 
@@ -338,7 +338,7 @@ static inline pmd_t swp_entry_to_pmd(swp_entry_t entry)
 	return __swp_entry_to_pmd(arch_entry);
 }
 
-#else  /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#else  /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 static inline int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
 		struct page *page)
 {
@@ -358,7 +358,7 @@ static inline pmd_t swp_entry_to_pmd(swp_entry_t entry)
 	return __pmd(0);
 }
 
-#endif  /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif  /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 #endif /* CONFIG_MMU */
 #endif /* _LINUX_SWAPOPS_H */
diff --git a/mm/Kconfig b/mm/Kconfig
index 9e0ca4824905..04fe5171bb8c 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -703,7 +703,7 @@ config DEVICE_MIGRATION
 config ARCH_ENABLE_HUGEPAGE_MIGRATION
 	bool
 
-config ARCH_ENABLE_THP_MIGRATION
+config ARCH_SUPPORTS_PMD_SOFTLEAF
 	bool
 
 config HUGETLB_PAGE_SIZE_VARIABLE
diff --git a/mm/debug_vm_pgtable.c b/mm/debug_vm_pgtable.c
index 18411fb09aab..507fbd1ae7e5 100644
--- a/mm/debug_vm_pgtable.c
+++ b/mm/debug_vm_pgtable.c
@@ -751,7 +751,7 @@ static void __init pmd_leaf_soft_dirty_tests(struct pgtable_debug_args *args)
 	pmd_t pmd;
 
 	if (!pgtable_supports_soft_dirty() ||
-	    !IS_ENABLED(CONFIG_ARCH_ENABLE_THP_MIGRATION))
+	    !IS_ENABLED(CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF))
 		return;
 
 	if (!has_transparent_hugepage())
@@ -819,7 +819,7 @@ static void __init pte_swap_tests(struct pgtable_debug_args *args)
 	WARN_ON(memcmp(&pte1, &pte2, sizeof(pte1)));
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 static void __init pmd_softleaf_tests(struct pgtable_debug_args *args)
 {
 	swp_entry_t arch_entry;
@@ -837,9 +837,9 @@ static void __init pmd_softleaf_tests(struct pgtable_debug_args *args)
 	pmd2 = __swp_entry_to_pmd(arch_entry);
 	WARN_ON(memcmp(&pmd1, &pmd2, sizeof(pmd1)));
 }
-#else  /* !CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#else  /* !CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 static void __init pmd_softleaf_tests(struct pgtable_debug_args *args) { }
-#endif /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 static void __init swap_migration_tests(struct pgtable_debug_args *args)
 {
diff --git a/mm/hmm.c b/mm/hmm.c
index c72c9ddfdb2f..4f3f627d2b47 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -331,7 +331,7 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr,
 	return hmm_vma_fault(addr, end, required_fault, walk);
 }
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 static int hmm_vma_handle_absent_pmd(struct mm_walk *walk, unsigned long start,
 				     unsigned long end, unsigned long *hmm_pfns,
 				     pmd_t pmd)
@@ -391,7 +391,7 @@ static int hmm_vma_handle_absent_pmd(struct mm_walk *walk, unsigned long start,
 		return -EFAULT;
 	return hmm_pfns_fill(start, end, range, HMM_PFN_ERROR);
 }
-#endif  /* CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#endif  /* CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 
 static int hmm_vma_walk_pmd(pmd_t *pmdp,
 			    unsigned long start,
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4aca9cdff73f..4b6e441ae931 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -4864,7 +4864,7 @@ static int __init split_huge_pages_debugfs(void)
 late_initcall(split_huge_pages_debugfs);
 #endif
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
 		struct page *page)
 {
diff --git a/mm/migrate.c b/mm/migrate.c
index 49e10feeb094..c48dc5127e6a 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -362,7 +362,7 @@ static bool remove_migration_pte(struct folio *folio,
 			idx = linear_page_index(vma, pvmw.address) - pvmw.pgoff;
 		new = folio_page(folio, idx);
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 		/* PMD-mapped THP migration entry */
 		if (!pvmw.pte) {
 			VM_BUG_ON_FOLIO(folio_test_hugetlb(folio) ||
@@ -545,7 +545,7 @@ void migration_entry_wait_huge(struct vm_area_struct *vma, unsigned long addr, p
 }
 #endif
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 void pmd_migration_entry_wait(struct mm_struct *mm, pmd_t *pmd)
 {
 	spinlock_t *ptl;
diff --git a/mm/migrate_device.c b/mm/migrate_device.c
index 36287e958b2a..2f8b646302c2 100644
--- a/mm/migrate_device.c
+++ b/mm/migrate_device.c
@@ -771,7 +771,7 @@ int migrate_vma_setup(struct migrate_vma *args)
 }
 EXPORT_SYMBOL(migrate_vma_setup);
 
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 /**
  * migrate_vma_insert_huge_pmd_page: Insert a huge folio into @migrate->vma->vm_mm
  * at @addr. folio is already allocated as a part of the migration process with
@@ -926,7 +926,7 @@ static int migrate_vma_split_unmapped_folio(struct migrate_vma *migrate,
 		migrate->src[i+idx] = migrate_pfn(pfn + i) | flags;
 	return ret;
 }
-#else /* !CONFIG_ARCH_ENABLE_THP_MIGRATION */
+#else /* !CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF */
 static int migrate_vma_insert_huge_pmd_page(struct migrate_vma *migrate,
 					 unsigned long addr,
 					 struct page *page,
@@ -947,7 +947,7 @@ static int migrate_vma_split_unmapped_folio(struct migrate_vma *migrate,
 static unsigned long migrate_vma_nr_pages(unsigned long *src)
 {
 	unsigned long nr = 1;
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 	if (*src & MIGRATE_PFN_COMPOUND)
 		nr = HPAGE_PMD_NR;
 #else
diff --git a/mm/rmap.c b/mm/rmap.c
index b93caabd186f..0fb7a1b82cf3 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -2472,7 +2472,7 @@ static bool try_to_migrate_one(struct folio *folio, struct vm_area_struct *vma,
 				page_vma_mapped_walk_restart(&pvmw);
 				continue;
 			}
-#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
+#ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
 			pmdval = pmdp_get(pvmw.pmd);
 			if (likely(pmd_present(pmdval)))
 				pfn = pmd_pfn(pmdval);
-- 
2.53.0-Meta

Re: [PATCH 6/6] mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

[David Hildenbrand (Arm)]

On 6/30/26 18:34, Usama Arif wrote:
> CONFIG_ARCH_ENABLE_THP_MIGRATION started life gating just PMD-level
> migration entries, but has grown to gate the entire PMD-level softleaf
> machinery: migration entries, device-private entries, and soon swap
> entries.

device-private really relies on migration support.

See do_huge_pmd_device_private() where we migrate back to CPU memory.

The whole concept is pased on migration. So that's why it currently makes sense.

So "has grown to" is not quite accurate. It's a different story with swap
entries (that can exist even without migration support).


> Rename CONFIG_ARCH_ENABLE_THP_MIGRATION to CONFIG_ARCH_SUPPORTS_PMD
> _SOFTLEAF to make this clear.  This is a pure rename: the set of
> selecting architectures (x86, arm64, s390, riscv, loongarch, and
> powerpc on PPC_BOOK3S_64) and the gating semantics are unchanged.

Hm, that makes me wonder what it would take to make this all just be implicitly
supported by CONFIG_TRANSPARENT_HUGEPAGE.

Looks like someone would have the courage to touch stuff like arc, mips and
sparc64 (+arm and x86, but that's less of a concern :) ).

So yeah, renaming the config option makes sense. The config space is the wild west.

We have

	PGTABLE_HAS_HUGE_LEAVES

and stuff like

	HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD
	HAVE_ARCH_SOFT_DIRTY
	ARCH_HAS_GIGANTIC_PAGE
	ARCH_HAS_PTE_SPECIAL

We do have a limited number of "SUPPORTS" (which is longer than HAS), but I
would either use


	ARCH_HAS_PMD_SOFTLEAVES

or

	HAVE_ARCH_PMD_SOFTLEAVES


Apart from that LGTM.

-- 
Cheers,

David

Re: [PATCH 6/6] mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

[Zi Yan]

On 1 Jul 2026, at 16:03, David Hildenbrand (Arm) wrote:

> On 6/30/26 18:34, Usama Arif wrote:
>> CONFIG_ARCH_ENABLE_THP_MIGRATION started life gating just PMD-level
>> migration entries, but has grown to gate the entire PMD-level softleaf
>> machinery: migration entries, device-private entries, and soon swap
>> entries.
>
> device-private really relies on migration support.
>
> See do_huge_pmd_device_private() where we migrate back to CPU memory.
>
> The whole concept is pased on migration. So that's why it currently makes sense.
>
> So "has grown to" is not quite accurate. It's a different story with swap
> entries (that can exist even without migration support).
>
>
>> Rename CONFIG_ARCH_ENABLE_THP_MIGRATION to CONFIG_ARCH_SUPPORTS_PMD
>> _SOFTLEAF to make this clear.  This is a pure rename: the set of
>> selecting architectures (x86, arm64, s390, riscv, loongarch, and
>> powerpc on PPC_BOOK3S_64) and the gating semantics are unchanged.
>
> Hm, that makes me wonder what it would take to make this all just be implicitly
> supported by CONFIG_TRANSPARENT_HUGEPAGE.
>
> Looks like someone would have the courage to touch stuff like arc, mips and
> sparc64 (+arm and x86, but that's less of a concern :) ).

You mean me back in 2018[1]? ;)

At least I got a no go for s390.


[1] https://lore.kernel.org/linux-mm/20180426142804.180152-1-zi.yan@sent.com/


Best Regards,
Yan, Zi

Re: [PATCH 6/6] mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

[David Hildenbrand (Arm)]

On 7/1/26 22:39, Zi Yan wrote:
> On 1 Jul 2026, at 16:03, David Hildenbrand (Arm) wrote:
> 
>> On 6/30/26 18:34, Usama Arif wrote:
>>> CONFIG_ARCH_ENABLE_THP_MIGRATION started life gating just PMD-level
>>> migration entries, but has grown to gate the entire PMD-level softleaf
>>> machinery: migration entries, device-private entries, and soon swap
>>> entries.
>>
>> device-private really relies on migration support.
>>
>> See do_huge_pmd_device_private() where we migrate back to CPU memory.
>>
>> The whole concept is pased on migration. So that's why it currently makes sense.
>>
>> So "has grown to" is not quite accurate. It's a different story with swap
>> entries (that can exist even without migration support).
>>
>>
>>> Rename CONFIG_ARCH_ENABLE_THP_MIGRATION to CONFIG_ARCH_SUPPORTS_PMD
>>> _SOFTLEAF to make this clear.  This is a pure rename: the set of
>>> selecting architectures (x86, arm64, s390, riscv, loongarch, and
>>> powerpc on PPC_BOOK3S_64) and the gating semantics are unchanged.
>>
>> Hm, that makes me wonder what it would take to make this all just be implicitly
>> supported by CONFIG_TRANSPARENT_HUGEPAGE.
>>
>> Looks like someone would have the courage to touch stuff like arc, mips and
>> sparc64 (+arm and x86, but that's less of a concern :) ).
> 
> You mean me back in 2018[1]? ;)
> 
> At least I got a no go for s390.

hah! and 8 years later it seems to exist upstream :)

commit 10dd5a0009898ba35eafeb6087e5c83b84742ff1
Author: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
Date:   Tue Jul 29 13:16:19 2025 +0200

    s390/mm: Enable THP_SWAP and THP_MIGRATION

    After hugetlbfs PTE_MARKER support for s390 introduced region-third and
    segment table swap entries, it is now possible to also enable THP_SWAP
    and THP_MIGRATION for s390.

So, with that out of the way, could you revive your patchset?


-- 
Cheers,

David

Re: [PATCH 6/6] mm: rename ARCH_ENABLE_THP_MIGRATION to ARCH_SUPPORTS_PMD_SOFTLEAF

[Zi Yan]

On 1 Jul 2026, at 16:50, David Hildenbrand (Arm) wrote:

> On 7/1/26 22:39, Zi Yan wrote:
>> On 1 Jul 2026, at 16:03, David Hildenbrand (Arm) wrote:
>>
>>> On 6/30/26 18:34, Usama Arif wrote:
>>>> CONFIG_ARCH_ENABLE_THP_MIGRATION started life gating just PMD-level
>>>> migration entries, but has grown to gate the entire PMD-level softleaf
>>>> machinery: migration entries, device-private entries, and soon swap
>>>> entries.
>>>
>>> device-private really relies on migration support.
>>>
>>> See do_huge_pmd_device_private() where we migrate back to CPU memory.
>>>
>>> The whole concept is pased on migration. So that's why it currently makes sense.
>>>
>>> So "has grown to" is not quite accurate. It's a different story with swap
>>> entries (that can exist even without migration support).
>>>
>>>
>>>> Rename CONFIG_ARCH_ENABLE_THP_MIGRATION to CONFIG_ARCH_SUPPORTS_PMD
>>>> _SOFTLEAF to make this clear.  This is a pure rename: the set of
>>>> selecting architectures (x86, arm64, s390, riscv, loongarch, and
>>>> powerpc on PPC_BOOK3S_64) and the gating semantics are unchanged.
>>>
>>> Hm, that makes me wonder what it would take to make this all just be implicitly
>>> supported by CONFIG_TRANSPARENT_HUGEPAGE.
>>>
>>> Looks like someone would have the courage to touch stuff like arc, mips and
>>> sparc64 (+arm and x86, but that's less of a concern :) ).
>>
>> You mean me back in 2018[1]? ;)
>>
>> At least I got a no go for s390.
>
> hah! and 8 years later it seems to exist upstream :)
>
> commit 10dd5a0009898ba35eafeb6087e5c83b84742ff1
> Author: Gerald Schaefer <gerald.schaefer@linux.ibm.com>
> Date:   Tue Jul 29 13:16:19 2025 +0200
>
>     s390/mm: Enable THP_SWAP and THP_MIGRATION
>
>     After hugetlbfs PTE_MARKER support for s390 introduced region-third and
>     segment table swap entries, it is now possible to also enable THP_SWAP
>     and THP_MIGRATION for s390.
>
> So, with that out of the way, could you revive your patchset?
>
Sure. Will put that on my TODO list.

Best Regards,
Yan, Zi

Re: [PATCH 0/6] mm: preparatory patches for PMD level swap entries

[Andrew Morton]

On Tue, 30 Jun 2026 09:34:37 -0700 Usama Arif <usama.arif@linux.dev> wrote:

> This is the preparatory part of the PMD page table swapin work. 

Thanks.

This comes reasonably reviewed, so I'll queue it in mm-new.

Why?

- Human review coverage and AI review make me believe this isn't the
  final version at all, but I do believe this is something we want in
  7.2, and getting it under test early will help to push it along.

- Getting it in there early means that others will base their work on
  material which will probably be upstreamed this cycle, so this gives
  them a more accurate base against which to work.

Sashiko did find a lot to talk about, some of it pre-existing so can
people who work on this code please take a look at

	https://sashiko.dev/#/patchset/20260630164143.1595669-1-usama.arif@linux.dev

Re: [PATCH 0/6] mm: preparatory patches for PMD level swap entries

[Lorenzo Stoakes]

On Tue, Jun 30, 2026 at 12:50:24PM -0700, Andrew Morton wrote:
> On Tue, 30 Jun 2026 09:34:37 -0700 Usama Arif <usama.arif@linux.dev> wrote:
>
> > This is the preparatory part of the PMD page table swapin work.
>
> Thanks.
>
> This comes reasonably reviewed, so I'll queue it in mm-new.
>
> Why?
>
> - Human review coverage and AI review make me believe this isn't the
>   final version at all, but I do believe this is something we want in
>   7.2, and getting it under test early will help to push it along.
>
> - Getting it in there early means that others will base their work on
>   material which will probably be upstreamed this cycle, so this gives
>   them a more accurate base against which to work.
>
> Sashiko did find a lot to talk about, some of it pre-existing so can
> people who work on this code please take a look at
>
> 	https://sashiko.dev/#/patchset/20260630164143.1595669-1-usama.arif@linux.dev
>
>

Hm, my understanding was that mm-new is where we take everything submitted for
early testing?

Did something change?

Generally anything submitted early in the cycle is expected to land in the cycle
unless there's significant adverse review or other reasons it can't be taken.

Thanks, Lorenzo

Re: [PATCH 0/6] mm: preparatory patches for PMD level swap entries

[Usama Arif]

On 30/06/2026 20:50, Andrew Morton wrote:
> On Tue, 30 Jun 2026 09:34:37 -0700 Usama Arif <usama.arif@linux.dev> wrote:
> 
>> This is the preparatory part of the PMD page table swapin work. 
> 
> Thanks.
> 
> This comes reasonably reviewed, so I'll queue it in mm-new.
> 
> Why?
> 
> - Human review coverage and AI review make me believe this isn't the
>   final version at all, but I do believe this is something we want in
>   7.2, and getting it under test early will help to push it along.
> 
> - Getting it in there early means that others will base their work on
>   material which will probably be upstreamed this cycle, so this gives
>   them a more accurate base against which to work.
> 
> Sashiko did find a lot to talk about, some of it pre-existing so can
> people who work on this code please take a look at
> 
> 	https://sashiko.dev/#/patchset/20260630164143.1595669-1-usama.arif@linux.dev
> 
> 

Thanks Andrew! I think most of the stuff sashiko pointed to is not
introduced by this series or can be ignored.

I will wait a few days and then send the PMD core series on latest
mm-new.

Re: [PATCH 0/6] mm: preparatory patches for PMD level swap entries

[Usama Arif]

On 30/06/2026 20:50, Andrew Morton wrote:
> On Tue, 30 Jun 2026 09:34:37 -0700 Usama Arif <usama.arif@linux.dev> wrote:
> 
>> This is the preparatory part of the PMD page table swapin work. 
> 
> Thanks.
> 
> This comes reasonably reviewed, so I'll queue it in mm-new.
> 
> Why?
> 
> - Human review coverage and AI review make me believe this isn't the
>   final version at all, but I do believe this is something we want in
>   7.2, and getting it under test early will help to push it along.
> 
> - Getting it in there early means that others will base their work on
>   material which will probably be upstreamed this cycle, so this gives
>   them a more accurate base against which to work.
> 
> Sashiko did find a lot to talk about, some of it pre-existing so can
> people who work on this code please take a look at
> 
> 	https://sashiko.dev/#/patchset/20260630164143.1595669-1-usama.arif@linux.dev
> 
> 

There is one pre-existing bug pointed by sashiko that is not related to my series
that I think needs to be fixed. I have sent it in
https://lore.kernel.org/all/20260701140638.840773-1-usama.arif@linux.dev/

Re: [PATCH 0/6] mm: preparatory patches for PMD level swap entries

[David Hildenbrand (Arm)]

On 6/30/26 18:34, Usama Arif wrote:
> This is the preparatory part of the PMD page table swapin work.  The
> full PMD swap entry series has been split into two parts:
> 
>   1. this preparatory series, which contains the first 6 patches. Zi [1]
>      and Lance [2] suggested to separate this out from the core series.
>   2. the PMD swap entry core series, which depends on this one. I will
>      send this once the preparatory series is merged in mm-new as v3
>      as the combined is currently at v2 [1].
>      I have not marked this prep series as v3, as its not really adding
>      support for PMD swap entries.
> 
> This series does not introduce PMD swap entries and does not install any
> new page-table entry type.  It only cleans up existing PMD softleaf
> helpers and call sites so the follow-up PMD swap entry series can be
> smaller and easier to review.
> 
> It should be safe to merge independently.  

Agreed, thanks for working on this and splitting this part out!

-- 
Cheers,

David