mm/rmap: Convert rmap_t into enum rmap_flags

Commit Message

Anshuman Khandual July 14, 2023, 6:14 a.m. UTC

rmap_t tracks bitwise positions for various page reverse map related flags.
enum could provide more compact representation. This converts these flags
into an enum listing, without any functional change.

Cc: Andrew Morton <[e-mail]>
Cc: "Matthew Wilcox (Oracle)" <[e-mail]>
Cc: "Kirill A. Shutemov" <[e-mail]>
Cc: Hugh Dickins <[e-mail]>
Cc: David Hildenbrand <[e-mail]>
Cc: [e-mail]
Cc: [e-mail]
Signed-off-by: Anshuman Khandual <[e-mail]>
---
This applies on v6.5-rc1

 include/linux/rmap.h | 26 +++++++-------------------
 mm/huge_memory.c     |  2 +-
 mm/memory.c          |  2 +-
 mm/migrate.c         |  2 +-
 mm/rmap.c            |  4 ++--
 mm/swapfile.c        |  2 +-
 6 files changed, 13 insertions(+), 25 deletions(-)

Comments

David Hildenbrand July 14, 2023, 6:46 a.m. UTC | #1

On 14.07.23 08:14, Anshuman Khandual wrote:
> rmap_t tracks bitwise positions for various page reverse map related flags.
> enum could provide more compact representation. This converts these flags
> into an enum listing, without any functional change.
> 
> Cc: Andrew Morton <[e-mail]>
> Cc: "Matthew Wilcox (Oracle)" <[e-mail]>
> Cc: "Kirill A. Shutemov" <[e-mail]>
> Cc: Hugh Dickins <[e-mail]>
> Cc: David Hildenbrand <[e-mail]>
> Cc: [e-mail]
> Cc: [e-mail]
> Signed-off-by: Anshuman Khandual <[e-mail]>
> ---
> This applies on v6.5-rc1

No,

please read Documentation/dev-tools/sparse.rst for "Using sparse for typechecking".

Patch

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index b87d01660412..aee6ee7ddac6 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -168,30 +168,18 @@  static inline void anon_vma_merge(struct vm_area_struct *vma,
 
 struct anon_vma *folio_get_anon_vma(struct folio *folio);
 
-/* RMAP flags, currently only relevant for some anon rmap operations. */
-typedef int __bitwise rmap_t;
-
-/*
- * No special request: if the page is a subpage of a compound page, it is
- * mapped via a PTE. The mapped (sub)page is possibly shared between processes.
- */
-#define RMAP_NONE		((__force rmap_t)0)
-
-/* The (sub)page is exclusive to a single process. */
-#define RMAP_EXCLUSIVE		((__force rmap_t)BIT(0))
-
-/*
- * The compound page is not mapped via PTEs, but instead via a single PMD and
- * should be accounted accordingly.
- */
-#define RMAP_COMPOUND		((__force rmap_t)BIT(1))
+enum rmap_flags {
+	RMAP_NONE	= 0x0,	/* No special request - (sub)page is mapped via a single PTE */
+	RMAP_EXCLUSIVE	= 0x1,	/* The (sub)page is exclusive to a single process */
+	RMAP_COMPOUND	= 0x2,	/* The compound page is mapped via a single PMD */
+};
 
 /*
  * rmap interfaces called when adding or removing pte of page
  */
 void page_move_anon_rmap(struct page *, struct vm_area_struct *);
 void page_add_anon_rmap(struct page *, struct vm_area_struct *,
-		unsigned long address, rmap_t flags);
+		unsigned long address, enum rmap_flags flags);
 void page_add_new_anon_rmap(struct page *, struct vm_area_struct *,
 		unsigned long address);
 void folio_add_new_anon_rmap(struct folio *, struct vm_area_struct *,
@@ -202,7 +190,7 @@  void page_remove_rmap(struct page *, struct vm_area_struct *,
 		bool compound);
 
 void hugepage_add_anon_rmap(struct page *, struct vm_area_struct *,
-		unsigned long address, rmap_t flags);
+		unsigned long address, enum rmap_flags flags);
 void hugepage_add_new_anon_rmap(struct folio *, struct vm_area_struct *,
 		unsigned long address);
 
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 965d845d73fc..79a790d1cfa8 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -3297,7 +3297,7 @@  void remove_migration_pmd(struct page_vma_mapped_walk *pvmw, struct page *new)
 		pmde = pmd_mkdirty(pmde);
 
 	if (PageAnon(new)) {
-		rmap_t rmap_flags = RMAP_COMPOUND;
+		enum rmap_flags rmap_flags = RMAP_COMPOUND;
 
 		if (!is_readable_migration_entry(entry))
 			rmap_flags |= RMAP_EXCLUSIVE;
diff --git a/mm/memory.c b/mm/memory.c
index 01f39e8144ef..1710aa6826d4 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3717,7 +3717,7 @@  vm_fault_t do_swap_page(struct vm_fault *vmf)
 	struct folio *swapcache, *folio = NULL;
 	struct page *page;
 	struct swap_info_struct *si = NULL;
-	rmap_t rmap_flags = RMAP_NONE;
+	enum rmap_flags rmap_flags = RMAP_NONE;
 	bool exclusive = false;
 	swp_entry_t entry;
 	pte_t pte;
diff --git a/mm/migrate.c b/mm/migrate.c
index 24baad2571e3..bdb73b11845a 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -186,7 +186,7 @@  static bool remove_migration_pte(struct folio *folio,
 	DEFINE_FOLIO_VMA_WALK(pvmw, old, vma, addr, PVMW_SYNC | PVMW_MIGRATION);
 
 	while (page_vma_mapped_walk(&pvmw)) {
-		rmap_t rmap_flags = RMAP_NONE;
+		enum rmap_flags rmap_flags = RMAP_NONE;
 		pte_t old_pte;
 		pte_t pte;
 		swp_entry_t entry;
diff --git a/mm/rmap.c b/mm/rmap.c
index 0c0d8857dfce..4d4c821d8e56 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1213,7 +1213,7 @@  static void __page_check_anon_rmap(struct page *page,
  * (but PageKsm is never downgraded to PageAnon).
  */
 void page_add_anon_rmap(struct page *page, struct vm_area_struct *vma,
-		unsigned long address, rmap_t flags)
+		unsigned long address, enum rmap_flags flags)
 {
 	struct folio *folio = page_folio(page);
 	atomic_t *mapped = &folio->_nr_pages_mapped;
@@ -2539,7 +2539,7 @@  void rmap_walk_locked(struct folio *folio, struct rmap_walk_control *rwc)
  * RMAP_COMPOUND is ignored.
  */
 void hugepage_add_anon_rmap(struct page *page, struct vm_area_struct *vma,
-			    unsigned long address, rmap_t flags)
+			    unsigned long address, enum rmap_flags flags)
 {
 	struct folio *folio = page_folio(page);
 	struct anon_vma *anon_vma = vma->anon_vma;
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 8e6dde68b389..3a9b1d8b3151 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1788,7 +1788,7 @@  static int unuse_pte(struct vm_area_struct *vma, pmd_t *pmd,
 	inc_mm_counter(vma->vm_mm, MM_ANONPAGES);
 	get_page(page);
 	if (page == swapcache) {
-		rmap_t rmap_flags = RMAP_NONE;
+		enum rmap_flags rmap_flags = RMAP_NONE;
 
 		/*
 		 * See do_swap_page(): PageWriteback() would be problematic.