[PATCH 0/6] Remove page_mapping()

[PATCH 0/6] Remove page_mapping()

[Matthew Wilcox (Oracle)]

There are only a few users left.  Convert them all to either call
folio_mapping() or just use folio->mapping directly.

Matthew Wilcox (Oracle) (6):
  fscrypt: Convert bh_get_inode_and_lblk_num to use a folio
  f2fs: Convert f2fs_clear_page_cache_dirty_tag to use a folio
  memory-failure: Remove calls to page_mapping()
  migrate: Expand the use of folio in __migrate_device_pages()
  userfault; Expand folio use in mfill_atomic_install_pte()
  mm: Remove page_mapping()

 fs/crypto/inline_crypt.c |  6 +++---
 fs/f2fs/data.c           |  5 +++--
 include/linux/pagemap.h  |  1 -
 mm/folio-compat.c        |  6 ------
 mm/memory-failure.c      |  6 ++++--
 mm/migrate_device.c      | 13 +++++--------
 mm/userfaultfd.c         |  5 ++---
 7 files changed, 17 insertions(+), 25 deletions(-)

-- 
2.43.0

[PATCH 1/6] fscrypt: Convert bh_get_inode_and_lblk_num to use a folio

[Matthew Wilcox (Oracle)]

Remove uses of page->index, page_mapping() and b_page.  Saves a call
to compound_head().

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/crypto/inline_crypt.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/crypto/inline_crypt.c b/fs/crypto/inline_crypt.c
index b4002aea7cdb..40de69860dcf 100644
--- a/fs/crypto/inline_crypt.c
+++ b/fs/crypto/inline_crypt.c
@@ -284,7 +284,7 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh,
 				      const struct inode **inode_ret,
 				      u64 *lblk_num_ret)
 {
-	struct page *page = bh->b_page;
+	struct folio *folio = bh->b_folio;
 	const struct address_space *mapping;
 	const struct inode *inode;
 
@@ -292,13 +292,13 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh,
 	 * The ext4 journal (jbd2) can submit a buffer_head it directly created
 	 * for a non-pagecache page.  fscrypt doesn't care about these.
 	 */
-	mapping = page_mapping(page);
+	mapping = folio_mapping(folio);
 	if (!mapping)
 		return false;
 	inode = mapping->host;
 
 	*inode_ret = inode;
-	*lblk_num_ret = ((u64)page->index << (PAGE_SHIFT - inode->i_blkbits)) +
+	*lblk_num_ret = ((u64)folio->index << (PAGE_SHIFT - inode->i_blkbits)) +
 			(bh_offset(bh) >> inode->i_blkbits);
 	return true;
 }
-- 
2.43.0

Re: [PATCH 1/6] fscrypt: Convert bh_get_inode_and_lblk_num to use a folio

[Eric Biggers]

On Tue, Apr 23, 2024 at 11:55:32PM +0100, Matthew Wilcox (Oracle) wrote:
> Remove uses of page->index, page_mapping() and b_page.  Saves a call
> to compound_head().
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>  fs/crypto/inline_crypt.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/fs/crypto/inline_crypt.c b/fs/crypto/inline_crypt.c
> index b4002aea7cdb..40de69860dcf 100644
> --- a/fs/crypto/inline_crypt.c
> +++ b/fs/crypto/inline_crypt.c
> @@ -284,7 +284,7 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh,
>  				      const struct inode **inode_ret,
>  				      u64 *lblk_num_ret)
>  {
> -	struct page *page = bh->b_page;
> +	struct folio *folio = bh->b_folio;
>  	const struct address_space *mapping;
>  	const struct inode *inode;
>  
> @@ -292,13 +292,13 @@ static bool bh_get_inode_and_lblk_num(const struct buffer_head *bh,
>  	 * The ext4 journal (jbd2) can submit a buffer_head it directly created
>  	 * for a non-pagecache page.  fscrypt doesn't care about these.
>  	 */
> -	mapping = page_mapping(page);
> +	mapping = folio_mapping(folio);
>  	if (!mapping)
>  		return false;
>  	inode = mapping->host;
>  
>  	*inode_ret = inode;
> -	*lblk_num_ret = ((u64)page->index << (PAGE_SHIFT - inode->i_blkbits)) +
> +	*lblk_num_ret = ((u64)folio->index << (PAGE_SHIFT - inode->i_blkbits)) +
>  			(bh_offset(bh) >> inode->i_blkbits);
>  	return true;
>  }

Reviewed-by: Eric Biggers <ebiggers@google.com>

- Eric

[PATCH 2/6] f2fs: Convert f2fs_clear_page_cache_dirty_tag to use a folio

[Matthew Wilcox (Oracle)]

Removes uses of page_mapping() and page_index().

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/f2fs/data.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 5d641fac02ba..9f74c867d790 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -4100,11 +4100,12 @@ const struct address_space_operations f2fs_dblock_aops = {
 
 void f2fs_clear_page_cache_dirty_tag(struct page *page)
 {
-	struct address_space *mapping = page_mapping(page);
+	struct folio *folio = page_folio(page);
+	struct address_space *mapping = folio->mapping;
 	unsigned long flags;
 
 	xa_lock_irqsave(&mapping->i_pages, flags);
-	__xa_clear_mark(&mapping->i_pages, page_index(page),
+	__xa_clear_mark(&mapping->i_pages, folio->index,
 						PAGECACHE_TAG_DIRTY);
 	xa_unlock_irqrestore(&mapping->i_pages, flags);
 }
-- 
2.43.0

Re: [PATCH 2/6] f2fs: Convert f2fs_clear_page_cache_dirty_tag to use a folio

[David Hildenbrand]

On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> Removes uses of page_mapping() and page_index().
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>   fs/f2fs/data.c | 5 +++--
>   1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> index 5d641fac02ba..9f74c867d790 100644
> --- a/fs/f2fs/data.c
> +++ b/fs/f2fs/data.c
> @@ -4100,11 +4100,12 @@ const struct address_space_operations f2fs_dblock_aops = {
>   
>   void f2fs_clear_page_cache_dirty_tag(struct page *page)
>   {
> -	struct address_space *mapping = page_mapping(page);
> +	struct folio *folio = page_folio(page);
> +	struct address_space *mapping = folio->mapping;
>   	unsigned long flags;
>   
>   	xa_lock_irqsave(&mapping->i_pages, flags);
> -	__xa_clear_mark(&mapping->i_pages, page_index(page),
> +	__xa_clear_mark(&mapping->i_pages, folio->index,
>   						PAGECACHE_TAG_DIRTY);
>   	xa_unlock_irqrestore(&mapping->i_pages, flags);
>   }

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb

[PATCH 3/6] memory-failure: Remove calls to page_mapping()

[Matthew Wilcox (Oracle)]

This is mostly just inlining page_mapping() into the two callers.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 mm/memory-failure.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index e065dd9be21e..62aa3db17854 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -216,6 +216,7 @@ EXPORT_SYMBOL_GPL(hwpoison_filter_flags_value);
 
 static int hwpoison_filter_dev(struct page *p)
 {
+	struct folio *folio = page_folio(p);
 	struct address_space *mapping;
 	dev_t dev;
 
@@ -223,7 +224,7 @@ static int hwpoison_filter_dev(struct page *p)
 	    hwpoison_filter_dev_minor == ~0U)
 		return 0;
 
-	mapping = page_mapping(p);
+	mapping = folio_mapping(folio);
 	if (mapping == NULL || mapping->host == NULL)
 		return -EINVAL;
 
@@ -1090,7 +1091,8 @@ static int me_pagecache_clean(struct page_state *ps, struct page *p)
  */
 static int me_pagecache_dirty(struct page_state *ps, struct page *p)
 {
-	struct address_space *mapping = page_mapping(p);
+	struct folio *folio = page_folio(p);
+	struct address_space *mapping = folio_mapping(folio);
 
 	/* TBD: print more information about the file. */
 	if (mapping) {
-- 
2.43.0

Re: [PATCH 3/6] memory-failure: Remove calls to page_mapping()

[Sidhartha Kumar]

On 4/23/24 3:55 PM, Matthew Wilcox (Oracle) wrote:
> This is mostly just inlining page_mapping() into the two callers.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>   mm/memory-failure.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index e065dd9be21e..62aa3db17854 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -216,6 +216,7 @@ EXPORT_SYMBOL_GPL(hwpoison_filter_flags_value);
>   
>   static int hwpoison_filter_dev(struct page *p)
>   {
> +	struct folio *folio = page_folio(p);
>   	struct address_space *mapping;
>   	dev_t dev;
>   
> @@ -223,7 +224,7 @@ static int hwpoison_filter_dev(struct page *p)
>   	    hwpoison_filter_dev_minor == ~0U)
>   		return 0;
>   
> -	mapping = page_mapping(p);
> +	mapping = folio_mapping(folio);
>   	if (mapping == NULL || mapping->host == NULL)
>   		return -EINVAL;
>   
> @@ -1090,7 +1091,8 @@ static int me_pagecache_clean(struct page_state *ps, struct page *p)
>    */
>   static int me_pagecache_dirty(struct page_state *ps, struct page *p)
>   {
> -	struct address_space *mapping = page_mapping(p);
> +	struct folio *folio = page_folio(p);
> +	struct address_space *mapping = folio_mapping(folio);
>   
>   	/* TBD: print more information about the file. */
>   	if (mapping) {
Reviewed-by: Sidhartha Kumar <sidhartha.kumar@oracle.com>

Re: [PATCH 3/6] memory-failure: Remove calls to page_mapping()

[David Hildenbrand]

On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> This is mostly just inlining page_mapping() into the two callers.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb

Re: [PATCH 3/6] memory-failure: Remove calls to page_mapping()

[Miaohe Lin]

On 2024/4/24 6:55, Matthew Wilcox (Oracle) wrote:
> This is mostly just inlining page_mapping() into the two callers.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>

Acked-by: Miaohe Lin <linmiaohe@huawei.com>
Thanks.
.

> ---
>  mm/memory-failure.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index e065dd9be21e..62aa3db17854 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -216,6 +216,7 @@ EXPORT_SYMBOL_GPL(hwpoison_filter_flags_value);
>  
>  static int hwpoison_filter_dev(struct page *p)
>  {
> +	struct folio *folio = page_folio(p);
>  	struct address_space *mapping;
>  	dev_t dev;
>  
> @@ -223,7 +224,7 @@ static int hwpoison_filter_dev(struct page *p)
>  	    hwpoison_filter_dev_minor == ~0U)
>  		return 0;
>  
> -	mapping = page_mapping(p);
> +	mapping = folio_mapping(folio);
>  	if (mapping == NULL || mapping->host == NULL)
>  		return -EINVAL;
>  
> @@ -1090,7 +1091,8 @@ static int me_pagecache_clean(struct page_state *ps, struct page *p)
>   */
>  static int me_pagecache_dirty(struct page_state *ps, struct page *p)
>  {
> -	struct address_space *mapping = page_mapping(p);
> +	struct folio *folio = page_folio(p);
> +	struct address_space *mapping = folio_mapping(folio);
>  
>  	/* TBD: print more information about the file. */
>  	if (mapping) {
> 

[PATCH 4/6] migrate: Expand the use of folio in __migrate_device_pages()

[Matthew Wilcox (Oracle)]

Removes a few calls to compound_head() and a call to page_mapping().

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 mm/migrate_device.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/mm/migrate_device.c b/mm/migrate_device.c
index a68616c1965f..aecc71972a87 100644
--- a/mm/migrate_device.c
+++ b/mm/migrate_device.c
@@ -692,6 +692,7 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 		struct page *newpage = migrate_pfn_to_page(dst_pfns[i]);
 		struct page *page = migrate_pfn_to_page(src_pfns[i]);
 		struct address_space *mapping;
+		struct folio *folio;
 		int r;
 
 		if (!newpage) {
@@ -726,15 +727,12 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 			continue;
 		}
 
-		mapping = page_mapping(page);
+		folio = page_folio(page);
+		mapping = folio_mapping(folio);
 
 		if (is_device_private_page(newpage) ||
 		    is_device_coherent_page(newpage)) {
 			if (mapping) {
-				struct folio *folio;
-
-				folio = page_folio(page);
-
 				/*
 				 * For now only support anonymous memory migrating to
 				 * device private or coherent memory.
@@ -757,11 +755,10 @@ static void __migrate_device_pages(unsigned long *src_pfns,
 
 		if (migrate && migrate->fault_page == page)
 			r = migrate_folio_extra(mapping, page_folio(newpage),
-						page_folio(page),
-						MIGRATE_SYNC_NO_COPY, 1);
+						folio, MIGRATE_SYNC_NO_COPY, 1);
 		else
 			r = migrate_folio(mapping, page_folio(newpage),
-					page_folio(page), MIGRATE_SYNC_NO_COPY);
+					folio, MIGRATE_SYNC_NO_COPY);
 		if (r != MIGRATEPAGE_SUCCESS)
 			src_pfns[i] &= ~MIGRATE_PFN_MIGRATE;
 	}
-- 
2.43.0

Re: [PATCH 4/6] migrate: Expand the use of folio in __migrate_device_pages()

[David Hildenbrand]

On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> Removes a few calls to compound_head() and a call to page_mapping().
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb

[PATCH 5/6] userfault; Expand folio use in mfill_atomic_install_pte()

[Matthew Wilcox (Oracle)]

Call page_folio() a little earlier so we can use folio_mapping()
instead of page_mapping(), saving a call to compound_head().

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 mm/userfaultfd.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
index d9dcc7d71a39..e6486923263c 100644
--- a/mm/userfaultfd.c
+++ b/mm/userfaultfd.c
@@ -180,9 +180,9 @@ int mfill_atomic_install_pte(pmd_t *dst_pmd,
 	pte_t _dst_pte, *dst_pte;
 	bool writable = dst_vma->vm_flags & VM_WRITE;
 	bool vm_shared = dst_vma->vm_flags & VM_SHARED;
-	bool page_in_cache = page_mapping(page);
 	spinlock_t *ptl;
-	struct folio *folio;
+	struct folio *folio = page_folio(page);
+	bool page_in_cache = folio_mapping(folio);
 
 	_dst_pte = mk_pte(page, dst_vma->vm_page_prot);
 	_dst_pte = pte_mkdirty(_dst_pte);
@@ -212,7 +212,6 @@ int mfill_atomic_install_pte(pmd_t *dst_pmd,
 	if (!pte_none_mostly(ptep_get(dst_pte)))
 		goto out_unlock;
 
-	folio = page_folio(page);
 	if (page_in_cache) {
 		/* Usually, cache pages are already added to LRU */
 		if (newly_allocated)
-- 
2.43.0

Re: [PATCH 5/6] userfault; Expand folio use in mfill_atomic_install_pte()

[David Hildenbrand]

On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> Call page_folio() a little earlier so we can use folio_mapping()
> instead of page_mapping(), saving a call to compound_head().
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb

[PATCH 6/6] mm: Remove page_mapping()

[Matthew Wilcox (Oracle)]

All callers are now converted, delete this compatibility wrapper.

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 include/linux/pagemap.h | 1 -
 mm/folio-compat.c       | 6 ------
 2 files changed, 7 deletions(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index b6f14e9a2d98..941f7ed714b9 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -399,7 +399,6 @@ static inline void filemap_nr_thps_dec(struct address_space *mapping)
 #endif
 }
 
-struct address_space *page_mapping(struct page *);
 struct address_space *folio_mapping(struct folio *);
 struct address_space *swapcache_mapping(struct folio *);
 
diff --git a/mm/folio-compat.c b/mm/folio-compat.c
index f31e0ce65b11..f05906006b3c 100644
--- a/mm/folio-compat.c
+++ b/mm/folio-compat.c
@@ -10,12 +10,6 @@
 #include <linux/swap.h>
 #include "internal.h"
 
-struct address_space *page_mapping(struct page *page)
-{
-	return folio_mapping(page_folio(page));
-}
-EXPORT_SYMBOL(page_mapping);
-
 void unlock_page(struct page *page)
 {
 	return folio_unlock(page_folio(page));
-- 
2.43.0

Re: [PATCH 6/6] mm: Remove page_mapping()

[David Hildenbrand]

On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> All callers are now converted, delete this compatibility wrapper.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>   include/linux/pagemap.h | 1 -
>   mm/folio-compat.c       | 6 ------
>   2 files changed, 7 deletions(-)
> 
> diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
> index b6f14e9a2d98..941f7ed714b9 100644
> --- a/include/linux/pagemap.h
> +++ b/include/linux/pagemap.h
> @@ -399,7 +399,6 @@ static inline void filemap_nr_thps_dec(struct address_space *mapping)
>   #endif
>   }
>   
> -struct address_space *page_mapping(struct page *);
>   struct address_space *folio_mapping(struct folio *);
>   struct address_space *swapcache_mapping(struct folio *);
>   
> diff --git a/mm/folio-compat.c b/mm/folio-compat.c
> index f31e0ce65b11..f05906006b3c 100644
> --- a/mm/folio-compat.c
> +++ b/mm/folio-compat.c
> @@ -10,12 +10,6 @@
>   #include <linux/swap.h>
>   #include "internal.h"
>   
> -struct address_space *page_mapping(struct page *page)
> -{
> -	return folio_mapping(page_folio(page));
> -}
> -EXPORT_SYMBOL(page_mapping);
> -
>   void unlock_page(struct page *page)
>   {
>   	return folio_unlock(page_folio(page));

Reviewed-by: David Hildenbrand <david@redhat.com>

-- 
Cheers,

David / dhildenb

Re: [PATCH 6/6] mm: Remove page_mapping()

[Andrew Morton]

On Wed, 24 Apr 2024 13:55:32 +0200 David Hildenbrand <david@redhat.com> wrote:

> On 24.04.24 00:55, Matthew Wilcox (Oracle) wrote:
> > All callers are now converted, delete this compatibility wrapper.
> > 

For some reason,

mm/hugetlb.c: In function 'hugetlb_page_mapping_lock_write':
mm/hugetlb.c:2164:41: error: implicit declaration of function 'page_mapping'; did you mean 'page_mapped'? [-Werror=implicit-function-declaration]
 2164 |         struct address_space *mapping = page_mapping(hpage);
      |                                         ^~~~~~~~~~~~
      |                                         page_mapped
mm/hugetlb.c:2164:41: error: initialization of 'struct address_space *' from 'int' makes pointer from integer without a cast [-Werror=int-conversion]


I'll disable "mm: Remove page_mapping()" pending review of the below,
please.


From: Andrew Morton <akpm@linux-foundation.org>
Subject: mm: convert hugetlb_page_mapping_lock_write() to hugetlb_folio_mapping_lock_write
Date: Wed Apr 24 04:20:30 PM PDT 2024

Convert this to use folios, so we can remove page_mapping()

Cc: David Hildenbrand <david@redhat.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/hugetlb.h |    6 +++---
 mm/hugetlb.c            |    6 +++---
 mm/memory-failure.c     |    4 ++--
 mm/migrate.c            |    2 +-
 4 files changed, 9 insertions(+), 9 deletions(-)

--- a/mm/hugetlb.c~mm-convert-hugetlb_page_mapping_lock_write-to-hugetlb_folio_mapping_lock_write
+++ a/mm/hugetlb.c
@@ -2155,13 +2155,13 @@ static bool prep_compound_gigantic_folio
 /*
  * Find and lock address space (mapping) in write mode.
  *
- * Upon entry, the page is locked which means that page_mapping() is
+ * Upon entry, the folio is locked which means that folio_mapping() is
  * stable.  Due to locking order, we can only trylock_write.  If we can
  * not get the lock, simply return NULL to caller.
  */
-struct address_space *hugetlb_page_mapping_lock_write(struct page *hpage)
+struct address_space *hugetlb_folio_mapping_lock_write(struct folio *folio)
 {
-	struct address_space *mapping = page_mapping(hpage);
+	struct address_space *mapping = folio_mapping(folio);
 
 	if (!mapping)
 		return mapping;
--- a/mm/memory-failure.c~mm-convert-hugetlb_page_mapping_lock_write-to-hugetlb_folio_mapping_lock_write
+++ a/mm/memory-failure.c
@@ -1595,7 +1595,7 @@ static bool hwpoison_user_mappings(struc
 	 * XXX: the dirty test could be racy: set_page_dirty() may not always
 	 * be called inside page lock (it's recommended but not enforced).
 	 */
-	mapping = page_mapping(hpage);
+	mapping = folio_mapping(folio);
 	if (!(flags & MF_MUST_KILL) && !PageDirty(hpage) && mapping &&
 	    mapping_can_writeback(mapping)) {
 		if (page_mkclean(hpage)) {
@@ -1622,7 +1622,7 @@ static bool hwpoison_user_mappings(struc
 		 * TTU_RMAP_LOCKED to indicate we have taken the lock
 		 * at this higher level.
 		 */
-		mapping = hugetlb_page_mapping_lock_write(hpage);
+		mapping = hugetlb_folio_mapping_lock_write(folio);
 		if (mapping) {
 			try_to_unmap(folio, ttu|TTU_RMAP_LOCKED);
 			i_mmap_unlock_write(mapping);
--- a/include/linux/hugetlb.h~mm-convert-hugetlb_page_mapping_lock_write-to-hugetlb_folio_mapping_lock_write
+++ a/include/linux/hugetlb.h
@@ -178,7 +178,7 @@ bool hugetlbfs_pagecache_present(struct
 				 struct vm_area_struct *vma,
 				 unsigned long address);
 
-struct address_space *hugetlb_page_mapping_lock_write(struct page *hpage);
+struct address_space *hugetlb_folio_mapping_lock_write(struct folio *folio);
 
 extern int sysctl_hugetlb_shm_group;
 extern struct list_head huge_boot_pages[MAX_NUMNODES];
@@ -297,8 +297,8 @@ static inline unsigned long hugetlb_tota
 	return 0;
 }
 
-static inline struct address_space *hugetlb_page_mapping_lock_write(
-							struct page *hpage)
+static inline struct address_space *hugetlb_folio_mapping_lock_write(
+							struct folio *folio)
 {
 	return NULL;
 }
--- a/mm/migrate.c~mm-convert-hugetlb_page_mapping_lock_write-to-hugetlb_folio_mapping_lock_write
+++ a/mm/migrate.c
@@ -1425,7 +1425,7 @@ static int unmap_and_move_huge_page(new_
 			 * semaphore in write mode here and set TTU_RMAP_LOCKED
 			 * to let lower levels know we have taken the lock.
 			 */
-			mapping = hugetlb_page_mapping_lock_write(&src->page);
+			mapping = hugetlb_folio_mapping_lock_write(src);
 			if (unlikely(!mapping))
 				goto unlock_put_anon;
 
_

Re: [PATCH 6/6] mm: Remove page_mapping()

[Matthew Wilcox]

On Wed, Apr 24, 2024 at 04:34:23PM -0700, Andrew Morton wrote:
> For some reason,
> 
> mm/hugetlb.c: In function 'hugetlb_page_mapping_lock_write':
> mm/hugetlb.c:2164:41: error: implicit declaration of function 'page_mapping'; did you mean 'page_mapped'? [-Werror=implicit-function-declaration]
>  2164 |         struct address_space *mapping = page_mapping(hpage);
>       |                                         ^~~~~~~~~~~~
>       |                                         page_mapped
> mm/hugetlb.c:2164:41: error: initialization of 'struct address_space *' from 'int' makes pointer from integer without a cast [-Werror=int-conversion]
> 
> 
> I'll disable "mm: Remove page_mapping()" pending review of the below,
> please.

Looks pretty similar to
https://lore.kernel.org/linux-mm/20240412193510.2356957-7-willy@infradead.org/

Sorry, I thought you'd picked up that series; I think it's fully
reviewed by now?

Re: [f2fs-dev] [PATCH 0/6] Remove page_mapping()

[patchwork-bot+f2fs]

Hello:

This series was applied to jaegeuk/f2fs.git (dev)
by Andrew Morton <akpm@linux-foundation.org>:

On Tue, 23 Apr 2024 23:55:31 +0100 you wrote:
> There are only a few users left.  Convert them all to either call
> folio_mapping() or just use folio->mapping directly.
> 
> Matthew Wilcox (Oracle) (6):
>   fscrypt: Convert bh_get_inode_and_lblk_num to use a folio
>   f2fs: Convert f2fs_clear_page_cache_dirty_tag to use a folio
>   memory-failure: Remove calls to page_mapping()
>   migrate: Expand the use of folio in __migrate_device_pages()
>   userfault; Expand folio use in mfill_atomic_install_pte()
>   mm: Remove page_mapping()
> 
> [...]

Here is the summary with links:
  - [f2fs-dev,1/6] fscrypt: Convert bh_get_inode_and_lblk_num to use a folio
    https://git.kernel.org/jaegeuk/f2fs/c/262f014dd7de
  - [f2fs-dev,2/6] f2fs: Convert f2fs_clear_page_cache_dirty_tag to use a folio
    https://git.kernel.org/jaegeuk/f2fs/c/196ad49cd626
  - [f2fs-dev,3/6] memory-failure: Remove calls to page_mapping()
    https://git.kernel.org/jaegeuk/f2fs/c/89f5c54b2281
  - [f2fs-dev,4/6] migrate: Expand the use of folio in __migrate_device_pages()
    https://git.kernel.org/jaegeuk/f2fs/c/e18a9faf06c2
  - [f2fs-dev,5/6] userfault; Expand folio use in mfill_atomic_install_pte()
    https://git.kernel.org/jaegeuk/f2fs/c/a568b4126b20
  - [f2fs-dev,6/6] mm: Remove page_mapping()
    (no matching commit)

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html