1 files changed, 134 insertions, 62 deletions

diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 68368ba99321..02ebb2f238af 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -7,28 +7,44 @@
 #include <linux/refcount.h>
 #include <linux/fiemap.h>
 #include <linux/btrfs_tree.h>
-#include "compression.h"
+#include <linux/spinlock.h>
+#include <linux/atomic.h>
+#include <linux/rwsem.h>
+#include <linux/list.h>
+#include <linux/slab.h>
+#include "messages.h"
 #include "ulist.h"
 #include "misc.h"
 
+struct page;
+struct file;
+struct folio;
+struct inode;
+struct fiemap_extent_info;
+struct readahead_control;
+struct address_space;
+struct writeback_control;
+struct extent_io_tree;
+struct extent_map_tree;
+struct extent_state;
+struct btrfs_block_group;
+struct btrfs_fs_info;
+struct btrfs_inode;
+struct btrfs_root;
 struct btrfs_trans_handle;
+struct btrfs_tree_parent_check;
 
 enum {
 	EXTENT_BUFFER_UPTODATE,
 	EXTENT_BUFFER_DIRTY,
-	EXTENT_BUFFER_CORRUPT,
-	/* this got triggered by readahead */
-	EXTENT_BUFFER_READAHEAD,
 	EXTENT_BUFFER_TREE_REF,
 	EXTENT_BUFFER_STALE,
 	EXTENT_BUFFER_WRITEBACK,
-	/* read IO error */
-	EXTENT_BUFFER_READ_ERR,
 	EXTENT_BUFFER_UNMAPPED,
-	EXTENT_BUFFER_IN_TREE,
 	/* write IO error */
 	EXTENT_BUFFER_WRITE_ERR,
-	EXTENT_BUFFER_NO_CHECK,
+	/* Indicate the extent buffer is written zeroed out (for zoned) */
+	EXTENT_BUFFER_ZONED_ZEROOUT,
 	/* Indicate that extent buffer pages a being read */
 	EXTENT_BUFFER_READING,
 };
@@ -43,10 +59,10 @@ enum {
 };
 
 /*
- * page->private values.  Every page that is controlled by the extent
- * map has page->private set to one.
+ * Folio private values.  Every page that is controlled by the extent map has
+ * folio private set to this value.
  */
-#define EXTENT_PAGE_PRIVATE 1
+#define EXTENT_FOLIO_PRIVATE			1
 
 /*
  * The extent buffer bitmap operations are done with byte granularity instead of
@@ -56,17 +72,12 @@ enum {
  *    single word in a bitmap may straddle two pages in the extent buffer.
  */
 #define BIT_BYTE(nr) ((nr) / BITS_PER_BYTE)
-#define BYTE_MASK ((1 << BITS_PER_BYTE) - 1)
+#define BYTE_MASK ((1U << BITS_PER_BYTE) - 1)
 #define BITMAP_FIRST_BYTE_MASK(start) \
 	((BYTE_MASK << ((start) & (BITS_PER_BYTE - 1))) & BYTE_MASK)
 #define BITMAP_LAST_BYTE_MASK(nbits) \
 	(BYTE_MASK >> (-(nbits) & (BITS_PER_BYTE - 1)))
 
-struct btrfs_root;
-struct btrfs_inode;
-struct btrfs_fs_info;
-struct extent_io_tree;
-struct btrfs_tree_parent_check;
 
 int __init extent_buffer_init_cachep(void);
 void __cold extent_buffer_free_cachep(void);
@@ -74,22 +85,36 @@ void __cold extent_buffer_free_cachep(void);
 #define INLINE_EXTENT_BUFFER_PAGES     (BTRFS_MAX_METADATA_BLOCKSIZE / PAGE_SIZE)
 struct extent_buffer {
 	u64 start;
-	unsigned long len;
+	u32 len;
+	u32 folio_size;
 	unsigned long bflags;
 	struct btrfs_fs_info *fs_info;
+
+	/*
+	 * The address where the eb can be accessed without any cross-page handling.
+	 * This can be NULL if not possible.
+	 */
+	void *addr;
+
 	spinlock_t refs_lock;
-	atomic_t refs;
+	refcount_t refs;
 	int read_mirror;
-	struct rcu_head rcu_head;
-	pid_t lock_owner;
 	/* >= 0 if eb belongs to a log tree, -1 otherwise */
 	s8 log_index;
+	u8 folio_shift;
+	struct rcu_head rcu_head;
 
 	struct rw_semaphore lock;
 
-	struct page *pages[INLINE_EXTENT_BUFFER_PAGES];
+	/*
+	 * Pointers to all the folios of the extent buffer.
+	 *
+	 * For now the folio is always order 0 (aka, a single page).
+	 */
+	struct folio *folios[INLINE_EXTENT_BUFFER_PAGES];
 #ifdef CONFIG_BTRFS_DEBUG
 	struct list_head leak_list;
+	pid_t lock_owner;
 #endif
 };
 
@@ -100,6 +125,13 @@ struct btrfs_eb_write_context {
 	struct btrfs_block_group *zoned_bg;
 };
 
+static inline unsigned long offset_in_eb_folio(const struct extent_buffer *eb,
+					       u64 start)
+{
+	ASSERT(eb->folio_size);
+	return start & (eb->folio_size - 1);
+}
+
 /*
  * Get the correct offset inside the page of extent buffer.
  *
@@ -108,29 +140,43 @@ struct btrfs_eb_write_context {
  *
  * Will handle both sectorsize == PAGE_SIZE and sectorsize < PAGE_SIZE cases.
  */
-static inline size_t get_eb_offset_in_page(const struct extent_buffer *eb,
-					   unsigned long offset)
+static inline size_t get_eb_offset_in_folio(const struct extent_buffer *eb,
+					    unsigned long offset)
 {
 	/*
-	 * For sectorsize == PAGE_SIZE case, eb->start will always be aligned
-	 * to PAGE_SIZE, thus adding it won't cause any difference.
+	 * 1) sectorsize == PAGE_SIZE and nodesize >= PAGE_SIZE case
+	 *    1.1) One large folio covering the whole eb
+	 *	   The eb->start is aligned to folio size, thus adding it
+	 *	   won't cause any difference.
+	 *    1.2) Several page sized folios
+	 *	   The eb->start is aligned to folio (page) size, thus
+	 *	   adding it won't cause any difference.
 	 *
-	 * For sectorsize < PAGE_SIZE, we must only read the data that belongs
-	 * to the eb, thus we have to take the eb->start into consideration.
+	 * 2) sectorsize < PAGE_SIZE and nodesize < PAGE_SIZE case
+	 *    In this case there would only be one page sized folio, and there
+	 *    may be several different extent buffers in the page/folio.
+	 *    We need to add eb->start to properly access the offset inside
+	 *    that eb.
 	 */
-	return offset_in_page(offset + eb->start);
+	return offset_in_folio(eb->folios[0], offset + eb->start);
 }
 
-static inline unsigned long get_eb_page_index(unsigned long offset)
+static inline unsigned long get_eb_folio_index(const struct extent_buffer *eb,
+					       unsigned long offset)
 {
 	/*
-	 * For sectorsize == PAGE_SIZE case, plain >> PAGE_SHIFT is enough.
+	 * 1) sectorsize == PAGE_SIZE and nodesize >= PAGE_SIZE case
+	 *    1.1) One large folio covering the whole eb.
+	 *	   the folio_shift would be large enough to always make us
+	 *	   return 0 as index.
+	 *    1.2) Several page sized folios
+	 *         The folio_shift would be PAGE_SHIFT, giving us the correct
+	 *         index.
 	 *
-	 * For sectorsize < PAGE_SIZE case, we only support 64K PAGE_SIZE,
-	 * and have ensured that all tree blocks are contained in one page,
-	 * thus we always get index == 0.
+	 * 2) sectorsize < PAGE_SIZE and nodesize < PAGE_SIZE case
+	 *    The folio would only be page sized, and always give us 0 as index.
 	 */
-	return offset >> PAGE_SHIFT;
+	return offset >> eb->folio_shift;
 }
 
 /*
@@ -162,6 +208,11 @@ static inline struct extent_changeset *extent_changeset_alloc(void)
 	return ret;
 }
 
+static inline void extent_changeset_prealloc(struct extent_changeset *changeset, gfp_t gfp_mask)
+{
+	ulist_prealloc(&changeset->range_changed, gfp_mask);
+}
+
 static inline void extent_changeset_release(struct extent_changeset *changeset)
 {
 	if (!changeset)
@@ -178,29 +229,23 @@ static inline void extent_changeset_free(struct extent_changeset *changeset)
 	kfree(changeset);
 }
 
-struct extent_map_tree;
-
-int try_release_extent_mapping(struct page *page, gfp_t mask);
-int try_release_extent_buffer(struct page *page);
+bool try_release_extent_mapping(struct folio *folio, gfp_t mask);
+int try_release_extent_buffer(struct folio *folio);
 
 int btrfs_read_folio(struct file *file, struct folio *folio);
-void extent_write_locked_range(struct inode *inode, struct page *locked_page,
+void extent_write_locked_range(struct inode *inode, const struct folio *locked_folio,
 			       u64 start, u64 end, struct writeback_control *wbc,
 			       bool pages_dirty);
-int extent_writepages(struct address_space *mapping,
-		      struct writeback_control *wbc);
+int btrfs_writepages(struct address_space *mapping, struct writeback_control *wbc);
 int btree_write_cache_pages(struct address_space *mapping,
 			    struct writeback_control *wbc);
-void extent_readahead(struct readahead_control *rac);
-int extent_fiemap(struct btrfs_inode *inode, struct fiemap_extent_info *fieinfo,
-		  u64 start, u64 len);
-int set_page_extent_mapped(struct page *page);
-void clear_page_extent_mapped(struct page *page);
+void btrfs_btree_wait_writeback_range(struct btrfs_fs_info *fs_info, u64 start, u64 end);
+void btrfs_readahead(struct readahead_control *rac);
+int set_folio_extent_mapped(struct folio *folio);
+void clear_folio_extent_mapped(struct folio *folio);
 
 struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info,
 					  u64 start, u64 owner_root, int level);
-struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
-						  u64 start, unsigned long len);
 struct extent_buffer *alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info,
 						u64 start);
 struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src);
@@ -208,17 +253,23 @@ struct extent_buffer *find_extent_buffer(struct btrfs_fs_info *fs_info,
 					 u64 start);
 void free_extent_buffer(struct extent_buffer *eb);
 void free_extent_buffer_stale(struct extent_buffer *eb);
-#define WAIT_NONE	0
-#define WAIT_COMPLETE	1
-#define WAIT_PAGE_LOCK	2
-int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num,
-			     struct btrfs_tree_parent_check *parent_check);
-void wait_on_extent_buffer_writeback(struct extent_buffer *eb);
+int read_extent_buffer_pages(struct extent_buffer *eb, int mirror_num,
+			     const struct btrfs_tree_parent_check *parent_check);
+int read_extent_buffer_pages_nowait(struct extent_buffer *eb, int mirror_num,
+				    const struct btrfs_tree_parent_check *parent_check);
+
+static inline void wait_on_extent_buffer_writeback(struct extent_buffer *eb)
+{
+	wait_on_bit_io(&eb->bflags, EXTENT_BUFFER_WRITEBACK,
+		       TASK_UNINTERRUPTIBLE);
+}
+
 void btrfs_readahead_tree_block(struct btrfs_fs_info *fs_info,
 				u64 bytenr, u64 owner_root, u64 gen, int level);
 void btrfs_readahead_node_child(struct extent_buffer *node, int slot);
 
-static inline int num_extent_pages(const struct extent_buffer *eb)
+/* Note: this can be used in for loops without caching the value in a variable. */
+static inline int __pure num_extent_pages(const struct extent_buffer *eb)
 {
 	/*
 	 * For sectorsize == PAGE_SIZE case, since nodesize is always aligned to
@@ -230,6 +281,24 @@ static inline int num_extent_pages(const struct extent_buffer *eb)
 	return (eb->len >> PAGE_SHIFT) ?: 1;
 }
 
+/*
+ * This can only be determined at runtime by checking eb::folios[0].
+ *
+ * As we can have either one large folio covering the whole eb
+ * (either nodesize <= PAGE_SIZE, or high order folio), or multiple
+ * single-paged folios.
+ *
+ * Note: this can be used in for loops without caching the value in a variable.
+ */
+static inline int __pure num_extent_folios(const struct extent_buffer *eb)
+{
+	if (!eb->folios[0])
+		return 0;
+	if (folio_order(eb->folios[0]))
+		return 1;
+	return num_extent_pages(eb);
+}
+
 static inline int extent_buffer_uptodate(const struct extent_buffer *eb)
 {
 	return test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags);
@@ -275,8 +344,8 @@ void memmove_extent_buffer(const struct extent_buffer *dst,
 			   unsigned long len);
 void memzero_extent_buffer(const struct extent_buffer *eb, unsigned long start,
 			   unsigned long len);
-int extent_buffer_test_bit(const struct extent_buffer *eb, unsigned long start,
-			   unsigned long pos);
+bool extent_buffer_test_bit(const struct extent_buffer *eb, unsigned long start,
+			    unsigned long pos);
 void extent_buffer_bitmap_set(const struct extent_buffer *eb, unsigned long start,
 			      unsigned long pos, unsigned long len);
 void extent_buffer_bitmap_clear(const struct extent_buffer *eb,
@@ -285,20 +354,23 @@ void extent_buffer_bitmap_clear(const struct extent_buffer *eb,
 void set_extent_buffer_dirty(struct extent_buffer *eb);
 void set_extent_buffer_uptodate(struct extent_buffer *eb);
 void clear_extent_buffer_uptodate(struct extent_buffer *eb);
-void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end);
 void extent_clear_unlock_delalloc(struct btrfs_inode *inode, u64 start, u64 end,
-				  struct page *locked_page,
+				  const struct folio *locked_folio,
+				  struct extent_state **cached,
 				  u32 bits_to_clear, unsigned long page_ops);
 int extent_invalidate_folio(struct extent_io_tree *tree,
 			    struct folio *folio, size_t offset);
 void btrfs_clear_buffer_dirty(struct btrfs_trans_handle *trans,
 			      struct extent_buffer *buf);
 
-int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array);
+int btrfs_alloc_page_array(unsigned int nr_pages, struct page **page_array,
+			   bool nofail);
+int btrfs_alloc_folio_array(unsigned int nr_folios, unsigned int order,
+			    struct folio **folio_array);
 
 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
 bool find_lock_delalloc_range(struct inode *inode,
-			     struct page *locked_page, u64 *start,
+			      struct folio *locked_folio, u64 *start,
 			     u64 *end);
 #endif
 struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info,