1 files changed, 145 insertions, 112 deletions

diff --git a/fs/nilfs2/mdt.c b/fs/nilfs2/mdt.c
index 700870a92bc4..946b0d3534a5 100644
--- a/fs/nilfs2/mdt.c
+++ b/fs/nilfs2/mdt.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0+
 /*
- * mdt.c - meta data file for NILFS
+ * Meta data file for NILFS
  *
  * Copyright (C) 2005-2008 Nippon Telegraph and Telephone Corporation.
  *
@@ -33,7 +33,8 @@ nilfs_mdt_insert_new_block(struct inode *inode, unsigned long block,
 					      struct buffer_head *, void *))
 {
 	struct nilfs_inode_info *ii = NILFS_I(inode);
-	void *kaddr;
+	struct folio *folio = bh->b_folio;
+	void *from;
 	int ret;
 
 	/* Caller exclude read accesses using page lock */
@@ -47,12 +48,14 @@ nilfs_mdt_insert_new_block(struct inode *inode, unsigned long block,
 
 	set_buffer_mapped(bh);
 
-	kaddr = kmap_atomic(bh->b_page);
-	memset(kaddr + bh_offset(bh), 0, i_blocksize(inode));
+	/* Initialize block (block size > PAGE_SIZE not yet supported) */
+	from = kmap_local_folio(folio, offset_in_folio(folio, bh->b_data));
+	memset(from, 0, bh->b_size);
 	if (init_block)
-		init_block(inode, bh, kaddr);
-	flush_dcache_page(bh->b_page);
-	kunmap_atomic(kaddr);
+		init_block(inode, bh, from);
+	kunmap_local(from);
+
+	flush_dcache_folio(folio);
 
 	set_buffer_uptodate(bh);
 	mark_buffer_dirty(bh);
@@ -89,7 +92,6 @@ static int nilfs_mdt_create_block(struct inode *inode, unsigned long block,
 	if (buffer_uptodate(bh))
 		goto failed_bh;
 
-	bh->b_bdev = sb->s_bdev;
 	err = nilfs_mdt_insert_new_block(inode, block, bh, init_block);
 	if (likely(!err)) {
 		get_bh(bh);
@@ -97,8 +99,8 @@ static int nilfs_mdt_create_block(struct inode *inode, unsigned long block,
 	}
 
  failed_bh:
-	unlock_page(bh->b_page);
-	put_page(bh->b_page);
+	folio_unlock(bh->b_folio);
+	folio_put(bh->b_folio);
 	brelse(bh);
 
  failed_unlock:
@@ -111,8 +113,8 @@ static int nilfs_mdt_create_block(struct inode *inode, unsigned long block,
 }
 
 static int
-nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff,
-		       int mode, int mode_flags, struct buffer_head **out_bh)
+nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff, blk_opf_t opf,
+		       struct buffer_head **out_bh)
 {
 	struct buffer_head *bh;
 	__u64 blknum = 0;
@@ -126,12 +128,12 @@ nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff,
 	if (buffer_uptodate(bh))
 		goto out;
 
-	if (mode_flags & REQ_RAHEAD) {
+	if (opf & REQ_RAHEAD) {
 		if (!trylock_buffer(bh)) {
 			ret = -EBUSY;
 			goto failed_bh;
 		}
-	} else /* mode == READ */
+	} else /* opf == REQ_OP_READ */
 		lock_buffer(bh);
 
 	if (buffer_uptodate(bh)) {
@@ -148,17 +150,18 @@ nilfs_mdt_submit_block(struct inode *inode, unsigned long blkoff,
 
 	bh->b_end_io = end_buffer_read_sync;
 	get_bh(bh);
-	submit_bh(mode, mode_flags, bh);
+	submit_bh(opf, bh);
 	ret = 0;
 
-	trace_nilfs2_mdt_submit_block(inode, inode->i_ino, blkoff, mode);
+	trace_nilfs2_mdt_submit_block(inode, inode->i_ino, blkoff,
+				      opf & REQ_OP_MASK);
  out:
 	get_bh(bh);
 	*out_bh = bh;
 
  failed_bh:
-	unlock_page(bh->b_page);
-	put_page(bh->b_page);
+	folio_unlock(bh->b_folio);
+	folio_put(bh->b_folio);
 	brelse(bh);
  failed:
 	return ret;
@@ -172,7 +175,7 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
 	int i, nr_ra_blocks = NILFS_MDT_MAX_RA_BLOCKS;
 	int err;
 
-	err = nilfs_mdt_submit_block(inode, block, REQ_OP_READ, 0, &first_bh);
+	err = nilfs_mdt_submit_block(inode, block, REQ_OP_READ, &first_bh);
 	if (err == -EEXIST) /* internal code */
 		goto out;
 
@@ -182,8 +185,8 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
 	if (readahead) {
 		blkoff = block + 1;
 		for (i = 0; i < nr_ra_blocks; i++, blkoff++) {
-			err = nilfs_mdt_submit_block(inode, blkoff, REQ_OP_READ,
-						     REQ_RAHEAD, &bh);
+			err = nilfs_mdt_submit_block(inode, blkoff,
+						REQ_OP_READ | REQ_RAHEAD, &bh);
 			if (likely(!err || err == -EEXIST))
 				brelse(bh);
 			else if (err != -EBUSY)
@@ -199,7 +202,7 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
  out_no_wait:
 	err = -EIO;
 	if (!buffer_uptodate(first_bh)) {
-		nilfs_msg(inode->i_sb, KERN_ERR,
+		nilfs_err(inode->i_sb,
 			  "I/O error reading meta-data file (ino=%lu, block-offset=%lu)",
 			  inode->i_ino, block);
 		goto failed_bh;
@@ -223,20 +226,21 @@ static int nilfs_mdt_read_block(struct inode *inode, unsigned long block,
  * @out_bh: output of a pointer to the buffer_head
  *
  * nilfs_mdt_get_block() looks up the specified buffer and tries to create
- * a new buffer if @create is not zero.  On success, the returned buffer is
- * assured to be either existing or formatted using a buffer lock on success.
- * @out_bh is substituted only when zero is returned.
- *
- * Return Value: On success, it returns 0. On error, the following negative
- * error code is returned.
- *
- * %-ENOMEM - Insufficient memory available.
- *
- * %-EIO - I/O error
+ * a new buffer if @create is not zero.  If (and only if) this function
+ * succeeds, it stores a pointer to the retrieved buffer head in the location
+ * pointed to by @out_bh.
  *
- * %-ENOENT - the specified block does not exist (hole block)
+ * The retrieved buffer may be either an existing one or a newly allocated one.
+ * For a newly created buffer, if the callback function argument @init_block
+ * is non-NULL, the callback will be called with the buffer locked to format
+ * the block.
  *
- * %-EROFS - Read only filesystem (for create mode)
+ * Return: 0 on success, or one of the following negative error codes on
+ * failure:
+ * * %-EIO	- I/O error (including metadata corruption).
+ * * %-ENOENT	- The specified block does not exist (hole block).
+ * * %-ENOMEM	- Insufficient memory available.
+ * * %-EROFS	- Read only filesystem (for create mode).
  */
 int nilfs_mdt_get_block(struct inode *inode, unsigned long blkoff, int create,
 			void (*init_block)(struct inode *,
@@ -272,14 +276,11 @@ int nilfs_mdt_get_block(struct inode *inode, unsigned long blkoff, int create,
  * @out_bh, and block offset to @blkoff, respectively.  @out_bh and
  * @blkoff are substituted only when zero is returned.
  *
- * Return Value: On success, it returns 0. On error, the following negative
- * error code is returned.
- *
- * %-ENOMEM - Insufficient memory available.
- *
- * %-EIO - I/O error
- *
- * %-ENOENT - no block was found in the range
+ * Return: 0 on success, or one of the following negative error codes on
+ * failure:
+ * * %-EIO	- I/O error (including metadata corruption).
+ * * %-ENOENT	- No block was found in the range.
+ * * %-ENOMEM	- Insufficient memory available.
  */
 int nilfs_mdt_find_block(struct inode *inode, unsigned long start,
 			 unsigned long end, unsigned long *blkoff,
@@ -318,12 +319,11 @@ out:
  * @inode: inode of the meta data file
  * @block: block offset
  *
- * Return Value: On success, zero is returned.
- * On error, one of the following negative error code is returned.
- *
- * %-ENOMEM - Insufficient memory available.
- *
- * %-EIO - I/O error
+ * Return: 0 on success, or one of the following negative error codes on
+ * failure:
+ * * %-EIO	- I/O error (including metadata corruption).
+ * * %-ENOENT	- Non-existent block.
+ * * %-ENOMEM	- Insufficient memory available.
  */
 int nilfs_mdt_delete_block(struct inode *inode, unsigned long block)
 {
@@ -346,39 +346,35 @@ int nilfs_mdt_delete_block(struct inode *inode, unsigned long block)
  * nilfs_mdt_forget_block() clears a dirty flag of the specified buffer, and
  * tries to release the page including the buffer from a page cache.
  *
- * Return Value: On success, 0 is returned. On error, one of the following
- * negative error code is returned.
- *
- * %-EBUSY - page has an active buffer.
- *
- * %-ENOENT - page cache has no page addressed by the offset.
+ * Return: 0 on success, or one of the following negative error codes on
+ * failure:
+ * * %-EBUSY	- Page has an active buffer.
+ * * %-ENOENT	- Page cache has no page addressed by the offset.
  */
 int nilfs_mdt_forget_block(struct inode *inode, unsigned long block)
 {
-	pgoff_t index = (pgoff_t)block >>
-		(PAGE_SHIFT - inode->i_blkbits);
-	struct page *page;
-	unsigned long first_block;
+	pgoff_t index = block >> (PAGE_SHIFT - inode->i_blkbits);
+	struct folio *folio;
+	struct buffer_head *bh;
 	int ret = 0;
 	int still_dirty;
 
-	page = find_lock_page(inode->i_mapping, index);
-	if (!page)
+	folio = filemap_lock_folio(inode->i_mapping, index);
+	if (IS_ERR(folio))
 		return -ENOENT;
 
-	wait_on_page_writeback(page);
-
-	first_block = (unsigned long)index <<
-		(PAGE_SHIFT - inode->i_blkbits);
-	if (page_has_buffers(page)) {
-		struct buffer_head *bh;
+	folio_wait_writeback(folio);
 
-		bh = nilfs_page_get_nth_block(page, block - first_block);
+	bh = folio_buffers(folio);
+	if (bh) {
+		unsigned long first_block = index <<
+				(PAGE_SHIFT - inode->i_blkbits);
+		bh = get_nth_bh(bh, block - first_block);
 		nilfs_forget_buffer(bh);
 	}
-	still_dirty = PageDirty(page);
-	unlock_page(page);
-	put_page(page);
+	still_dirty = folio_test_dirty(folio);
+	folio_unlock(folio);
+	folio_put(folio);
 
 	if (still_dirty ||
 	    invalidate_inode_pages2_range(inode->i_mapping, index, index) != 0)
@@ -397,10 +393,10 @@ int nilfs_mdt_fetch_dirty(struct inode *inode)
 	return test_bit(NILFS_I_DIRTY, &ii->i_state);
 }
 
-static int
-nilfs_mdt_write_page(struct page *page, struct writeback_control *wbc)
+static int nilfs_mdt_write_folio(struct folio *folio,
+		struct writeback_control *wbc)
 {
-	struct inode *inode = page->mapping->host;
+	struct inode *inode = folio->mapping->host;
 	struct super_block *sb;
 	int err = 0;
 
@@ -408,16 +404,16 @@ nilfs_mdt_write_page(struct page *page, struct writeback_control *wbc)
 		/*
 		 * It means that filesystem was remounted in read-only
 		 * mode because of error or metadata corruption. But we
-		 * have dirty pages that try to be flushed in background.
-		 * So, here we simply discard this dirty page.
+		 * have dirty folios that try to be flushed in background.
+		 * So, here we simply discard this dirty folio.
 		 */
-		nilfs_clear_dirty_page(page, false);
-		unlock_page(page);
+		nilfs_clear_folio_dirty(folio);
+		folio_unlock(folio);
 		return -EROFS;
 	}
 
-	redirty_page_for_writepage(wbc, page);
-	unlock_page(page);
+	folio_redirty_for_writepage(wbc, folio);
+	folio_unlock(folio);
 
 	if (!inode)
 		return 0;
@@ -426,15 +422,27 @@ nilfs_mdt_write_page(struct page *page, struct writeback_control *wbc)
 
 	if (wbc->sync_mode == WB_SYNC_ALL)
 		err = nilfs_construct_segment(sb);
-	else if (wbc->for_reclaim)
-		nilfs_flush_segment(sb, inode->i_ino);
 
 	return err;
 }
 
+static int nilfs_mdt_writeback(struct address_space *mapping,
+		struct writeback_control *wbc)
+{
+	struct folio *folio = NULL;
+	int error;
+
+	while ((folio = writeback_iter(mapping, wbc, folio, &error)))
+		error = nilfs_mdt_write_folio(folio, wbc);
+
+	return error;
+}
 
 static const struct address_space_operations def_mdt_aops = {
-	.writepage		= nilfs_mdt_write_page,
+	.dirty_folio		= block_dirty_folio,
+	.invalidate_folio	= block_invalidate_folio,
+	.writepages		= nilfs_mdt_writeback,
+	.migrate_folio		= buffer_migrate_folio_norefs,
 };
 
 static const struct inode_operations def_mdt_iops;
@@ -469,9 +477,18 @@ int nilfs_mdt_init(struct inode *inode, gfp_t gfp_mask, size_t objsz)
 void nilfs_mdt_clear(struct inode *inode)
 {
 	struct nilfs_mdt_info *mdi = NILFS_MDT(inode);
+	struct nilfs_shadow_map *shadow = mdi->mi_shadow;
 
 	if (mdi->mi_palloc_cache)
 		nilfs_palloc_destroy_cache(inode);
+
+	if (shadow) {
+		struct inode *s_inode = shadow->inode;
+
+		shadow->inode = NULL;
+		iput(s_inode);
+		mdi->mi_shadow = NULL;
+	}
 }
 
 /**
@@ -500,17 +517,22 @@ void nilfs_mdt_set_entry_size(struct inode *inode, unsigned int entry_size,
  * nilfs_mdt_setup_shadow_map - setup shadow map and bind it to metadata file
  * @inode: inode of the metadata file
  * @shadow: shadow mapping
+ *
+ * Return: 0 on success, or a negative error code on failure.
  */
 int nilfs_mdt_setup_shadow_map(struct inode *inode,
 			       struct nilfs_shadow_map *shadow)
 {
 	struct nilfs_mdt_info *mi = NILFS_MDT(inode);
+	struct inode *s_inode;
 
 	INIT_LIST_HEAD(&shadow->frozen_buffers);
-	address_space_init_once(&shadow->frozen_data);
-	nilfs_mapping_init(&shadow->frozen_data, inode);
-	address_space_init_once(&shadow->frozen_btnodes);
-	nilfs_mapping_init(&shadow->frozen_btnodes, inode);
+
+	s_inode = nilfs_iget_for_shadow(inode);
+	if (IS_ERR(s_inode))
+		return PTR_ERR(s_inode);
+
+	shadow->inode = s_inode;
 	mi->mi_shadow = shadow;
 	return 0;
 }
@@ -518,20 +540,23 @@ int nilfs_mdt_setup_shadow_map(struct inode *inode,
 /**
  * nilfs_mdt_save_to_shadow_map - copy bmap and dirty pages to shadow map
  * @inode: inode of the metadata file
+ *
+ * Return: 0 on success, or a negative error code on failure.
  */
 int nilfs_mdt_save_to_shadow_map(struct inode *inode)
 {
 	struct nilfs_mdt_info *mi = NILFS_MDT(inode);
 	struct nilfs_inode_info *ii = NILFS_I(inode);
 	struct nilfs_shadow_map *shadow = mi->mi_shadow;
+	struct inode *s_inode = shadow->inode;
 	int ret;
 
-	ret = nilfs_copy_dirty_pages(&shadow->frozen_data, inode->i_mapping);
+	ret = nilfs_copy_dirty_pages(s_inode->i_mapping, inode->i_mapping);
 	if (ret)
 		goto out;
 
-	ret = nilfs_copy_dirty_pages(&shadow->frozen_btnodes,
-				     &ii->i_btnode_cache);
+	ret = nilfs_copy_dirty_pages(NILFS_I(s_inode)->i_assoc_inode->i_mapping,
+				     ii->i_assoc_inode->i_mapping);
 	if (ret)
 		goto out;
 
@@ -544,17 +569,20 @@ int nilfs_mdt_freeze_buffer(struct inode *inode, struct buffer_head *bh)
 {
 	struct nilfs_shadow_map *shadow = NILFS_MDT(inode)->mi_shadow;
 	struct buffer_head *bh_frozen;
-	struct page *page;
+	struct folio *folio;
 	int blkbits = inode->i_blkbits;
 
-	page = grab_cache_page(&shadow->frozen_data, bh->b_page->index);
-	if (!page)
-		return -ENOMEM;
+	folio = filemap_grab_folio(shadow->inode->i_mapping,
+			bh->b_folio->index);
+	if (IS_ERR(folio))
+		return PTR_ERR(folio);
 
-	if (!page_has_buffers(page))
-		create_empty_buffers(page, 1 << blkbits, 0);
+	bh_frozen = folio_buffers(folio);
+	if (!bh_frozen)
+		bh_frozen = create_empty_buffers(folio, 1 << blkbits, 0);
 
-	bh_frozen = nilfs_page_get_nth_block(page, bh_offset(bh) >> blkbits);
+	bh_frozen = get_nth_bh(bh_frozen,
+			       offset_in_folio(folio, bh->b_data) >> blkbits);
 
 	if (!buffer_uptodate(bh_frozen))
 		nilfs_copy_buffer(bh_frozen, bh);
@@ -566,8 +594,8 @@ int nilfs_mdt_freeze_buffer(struct inode *inode, struct buffer_head *bh)
 		brelse(bh_frozen); /* already frozen */
 	}
 
-	unlock_page(page);
-	put_page(page);
+	folio_unlock(folio);
+	folio_put(folio);
 	return 0;
 }
 
@@ -576,17 +604,20 @@ nilfs_mdt_get_frozen_buffer(struct inode *inode, struct buffer_head *bh)
 {
 	struct nilfs_shadow_map *shadow = NILFS_MDT(inode)->mi_shadow;
 	struct buffer_head *bh_frozen = NULL;
-	struct page *page;
+	struct folio *folio;
 	int n;
 
-	page = find_lock_page(&shadow->frozen_data, bh->b_page->index);
-	if (page) {
-		if (page_has_buffers(page)) {
-			n = bh_offset(bh) >> inode->i_blkbits;
-			bh_frozen = nilfs_page_get_nth_block(page, n);
+	folio = filemap_lock_folio(shadow->inode->i_mapping,
+			bh->b_folio->index);
+	if (!IS_ERR(folio)) {
+		bh_frozen = folio_buffers(folio);
+		if (bh_frozen) {
+			n = offset_in_folio(folio, bh->b_data) >>
+				inode->i_blkbits;
+			bh_frozen = get_nth_bh(bh_frozen, n);
 		}
-		unlock_page(page);
-		put_page(page);
+		folio_unlock(folio);
+		folio_put(folio);
 	}
 	return bh_frozen;
 }
@@ -619,11 +650,12 @@ void nilfs_mdt_restore_from_shadow_map(struct inode *inode)
 	if (mi->mi_palloc_cache)
 		nilfs_palloc_clear_cache(inode);
 
-	nilfs_clear_dirty_pages(inode->i_mapping, true);
-	nilfs_copy_back_pages(inode->i_mapping, &shadow->frozen_data);
+	nilfs_clear_dirty_pages(inode->i_mapping);
+	nilfs_copy_back_pages(inode->i_mapping, shadow->inode->i_mapping);
 
-	nilfs_clear_dirty_pages(&ii->i_btnode_cache, true);
-	nilfs_copy_back_pages(&ii->i_btnode_cache, &shadow->frozen_btnodes);
+	nilfs_clear_dirty_pages(ii->i_assoc_inode->i_mapping);
+	nilfs_copy_back_pages(ii->i_assoc_inode->i_mapping,
+			      NILFS_I(shadow->inode)->i_assoc_inode->i_mapping);
 
 	nilfs_bmap_restore(ii->i_bmap, &shadow->bmap_store);
 
@@ -638,10 +670,11 @@ void nilfs_mdt_clear_shadow_map(struct inode *inode)
 {
 	struct nilfs_mdt_info *mi = NILFS_MDT(inode);
 	struct nilfs_shadow_map *shadow = mi->mi_shadow;
+	struct inode *shadow_btnc_inode = NILFS_I(shadow->inode)->i_assoc_inode;
 
 	down_write(&mi->mi_sem);
 	nilfs_release_frozen_buffers(shadow);
-	truncate_inode_pages(&shadow->frozen_data, 0);
-	truncate_inode_pages(&shadow->frozen_btnodes, 0);
+	truncate_inode_pages(shadow->inode->i_mapping, 0);
+	truncate_inode_pages(shadow_btnc_inode->i_mapping, 0);
 	up_write(&mi->mi_sem);
 }