[v7 05/16] mm/huge_memory: implement device-private THP splitting

[Balbir Singh <balbirs@nvidia.com>]

Add support for splitting device-private THP folios, enabling fallback
to smaller page sizes when large page allocation or migration fails.

Key changes:
- split_huge_pmd(): Handle device-private PMD entries during splitting
- Preserve RMAP_EXCLUSIVE semantics for anonymous exclusive folios
- Skip RMP_USE_SHARED_ZEROPAGE for device-private entries as they
  don't support shared zero page semantics

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: David Hildenbrand <david@redhat.com>
Cc: Zi Yan <ziy@nvidia.com>
Cc: Joshua Hahn <joshua.hahnjy@gmail.com>
Cc: Rakie Kim <rakie.kim@sk.com>
Cc: Byungchul Park <byungchul@sk.com>
Cc: Gregory Price <gourry@gourry.net>
Cc: Ying Huang <ying.huang@linux.alibaba.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Cc: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: "Liam R. Howlett" <Liam.Howlett@oracle.com>
Cc: Nico Pache <npache@redhat.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Dev Jain <dev.jain@arm.com>
Cc: Barry Song <baohua@kernel.org>
Cc: Lyude Paul <lyude@redhat.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: Mika Penttilä <mpenttil@redhat.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Francois Dugast <francois.dugast@intel.com>
Signed-off-by: Balbir Singh <balbirs@nvidia.com>
---
 mm/huge_memory.c | 87 +++++++++++++++++++++++++++++++++++++++++-------
 mm/migrate.c     |  1 +
 2 files changed, 76 insertions(+), 12 deletions(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 483b8341ce22..05c68f5b5fe3 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2872,16 +2872,18 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	struct page *page;
 	pgtable_t pgtable;
 	pmd_t old_pmd, _pmd;
-	bool young, write, soft_dirty, pmd_migration = false, uffd_wp = false;
+	bool soft_dirty, uffd_wp = false, young = false, write = false;
 	bool anon_exclusive = false, dirty = false;
 	unsigned long addr;
 	pte_t *pte;
 	int i;
+	swp_entry_t entry;
 
 	VM_BUG_ON(haddr & ~HPAGE_PMD_MASK);
 	VM_BUG_ON_VMA(vma->vm_start > haddr, vma);
 	VM_BUG_ON_VMA(vma->vm_end < haddr + HPAGE_PMD_SIZE, vma);
-	VM_BUG_ON(!is_pmd_migration_entry(*pmd) && !pmd_trans_huge(*pmd));
+
+	VM_WARN_ON(!is_pmd_non_present_folio_entry(*pmd) && !pmd_trans_huge(*pmd));
 
 	count_vm_event(THP_SPLIT_PMD);
 
@@ -2929,20 +2931,51 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 		return __split_huge_zero_page_pmd(vma, haddr, pmd);
 	}
 
-	pmd_migration = is_pmd_migration_entry(*pmd);
-	if (unlikely(pmd_migration)) {
-		swp_entry_t entry;
 
+	if (is_pmd_migration_entry(*pmd)) {
 		old_pmd = *pmd;
 		entry = pmd_to_swp_entry(old_pmd);
 		page = pfn_swap_entry_to_page(entry);
+		folio = page_folio(page);
+
+		soft_dirty = pmd_swp_soft_dirty(old_pmd);
+		uffd_wp = pmd_swp_uffd_wp(old_pmd);
+
 		write = is_writable_migration_entry(entry);
 		if (PageAnon(page))
 			anon_exclusive = is_readable_exclusive_migration_entry(entry);
 		young = is_migration_entry_young(entry);
 		dirty = is_migration_entry_dirty(entry);
+	} else if (is_pmd_device_private_entry(*pmd)) {
+		old_pmd = *pmd;
+		entry = pmd_to_swp_entry(old_pmd);
+		page = pfn_swap_entry_to_page(entry);
+		folio = page_folio(page);
+
 		soft_dirty = pmd_swp_soft_dirty(old_pmd);
 		uffd_wp = pmd_swp_uffd_wp(old_pmd);
+
+		write = is_writable_device_private_entry(entry);
+		anon_exclusive = PageAnonExclusive(page);
+
+		/*
+		 * Device private THP should be treated the same as regular
+		 * folios w.r.t anon exclusive handling. See the comments for
+		 * folio handling and anon_exclusive below.
+		 */
+		if (freeze && anon_exclusive &&
+		    folio_try_share_anon_rmap_pmd(folio, page))
+			freeze = false;
+		if (!freeze) {
+			rmap_t rmap_flags = RMAP_NONE;
+
+			folio_ref_add(folio, HPAGE_PMD_NR - 1);
+			if (anon_exclusive)
+				rmap_flags |= RMAP_EXCLUSIVE;
+
+			folio_add_anon_rmap_ptes(folio, page, HPAGE_PMD_NR,
+						 vma, haddr, rmap_flags);
+		}
 	} else {
 		/*
 		 * Up to this point the pmd is present and huge and userland has
@@ -3026,11 +3059,11 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	 * Note that NUMA hinting access restrictions are not transferred to
 	 * avoid any possibility of altering permissions across VMAs.
 	 */
-	if (freeze || pmd_migration) {
-		for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
-			pte_t entry;
-			swp_entry_t swp_entry;
+	if (freeze || is_pmd_migration_entry(old_pmd)) {
+		pte_t entry;
+		swp_entry_t swp_entry;
 
+		for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
 			if (write)
 				swp_entry = make_writable_migration_entry(
 							page_to_pfn(page + i));
@@ -3049,7 +3082,33 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 				entry = pte_swp_mksoft_dirty(entry);
 			if (uffd_wp)
 				entry = pte_swp_mkuffd_wp(entry);
+			VM_WARN_ON(!pte_none(ptep_get(pte + i)));
+			set_pte_at(mm, addr, pte + i, entry);
+		}
+	} else if (is_pmd_device_private_entry(old_pmd)) {
+		pte_t entry;
+		swp_entry_t swp_entry;
 
+		for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
+			/*
+			 * anon_exclusive was already propagated to the relevant
+			 * pages corresponding to the pte entries when freeze
+			 * is false.
+			 */
+			if (write)
+				swp_entry = make_writable_device_private_entry(
+							page_to_pfn(page + i));
+			else
+				swp_entry = make_readable_device_private_entry(
+							page_to_pfn(page + i));
+			/*
+			 * Young and dirty bits are not progated via swp_entry
+			 */
+			entry = swp_entry_to_pte(swp_entry);
+			if (soft_dirty)
+				entry = pte_swp_mksoft_dirty(entry);
+			if (uffd_wp)
+				entry = pte_swp_mkuffd_wp(entry);
 			VM_WARN_ON(!pte_none(ptep_get(pte + i)));
 			set_pte_at(mm, addr, pte + i, entry);
 		}
@@ -3076,7 +3135,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	}
 	pte_unmap(pte);
 
-	if (!pmd_migration)
+	if (!is_pmd_migration_entry(*pmd))
 		folio_remove_rmap_pmd(folio, page, vma);
 	if (freeze)
 		put_page(page);
@@ -3089,7 +3148,7 @@ void split_huge_pmd_locked(struct vm_area_struct *vma, unsigned long address,
 			   pmd_t *pmd, bool freeze)
 {
 	VM_WARN_ON_ONCE(!IS_ALIGNED(address, HPAGE_PMD_SIZE));
-	if (pmd_trans_huge(*pmd) || is_pmd_migration_entry(*pmd))
+	if (pmd_trans_huge(*pmd) || is_pmd_non_present_folio_entry(*pmd))
 		__split_huge_pmd_locked(vma, pmd, address, freeze);
 }
 
@@ -3268,6 +3327,9 @@ static void lru_add_split_folio(struct folio *folio, struct folio *new_folio,
 	VM_BUG_ON_FOLIO(folio_test_lru(new_folio), folio);
 	lockdep_assert_held(&lruvec->lru_lock);
 
+	if (folio_is_device_private(folio))
+		return;
+
 	if (list) {
 		/* page reclaim is reclaiming a huge page */
 		VM_WARN_ON(folio_test_lru(folio));
@@ -3885,8 +3947,9 @@ static int __folio_split(struct folio *folio, unsigned int new_order,
 	if (nr_shmem_dropped)
 		shmem_uncharge(mapping->host, nr_shmem_dropped);
 
-	if (!ret && is_anon)
+	if (!ret && is_anon && !folio_is_device_private(folio))
 		remap_flags = RMP_USE_SHARED_ZEROPAGE;
+
 	remap_page(folio, 1 << order, remap_flags);
 
 	/*
diff --git a/mm/migrate.c b/mm/migrate.c
index ce83c2c3c287..11fbfe905e3c 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -307,6 +307,7 @@ static bool try_to_map_unused_to_zeropage(struct page_vma_mapped_walk *pvmw,
 	VM_BUG_ON_PAGE(!PageAnon(page), page);
 	VM_BUG_ON_PAGE(!PageLocked(page), page);
 	VM_BUG_ON_PAGE(pte_present(ptep_get(pvmw->pte)), page);
+	VM_WARN_ON_ONCE_FOLIO(folio_is_device_private(folio), folio);
 
 	if (folio_test_mlocked(folio) || (pvmw->vma->vm_flags & VM_LOCKED) ||
 	    mm_forbids_zeropage(pvmw->vma->vm_mm))
-- 
2.51.0