Merge tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

Pull ext4 updates from Ted Ts'o:
 "A large number of cleanups and bug fixes, with some (minor) journal
  optimizations"

[ This got sent to me before -rc1, but was stuck in my spam folder.   - Linus ]

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (67 commits)
  ext4: check s_chksum_driver when looking for bg csum presence
  ext4: move error report out of atomic context in ext4_init_block_bitmap()
  ext4: Replace open coded mdata csum feature to helper function
  ext4: delete useless comments about ext4_move_extents
  ext4: fix reservation overflow in ext4_da_write_begin
  ext4: add ext4_iget_normal() which is to be used for dir tree lookups
  ext4: don't orphan or truncate the boot loader inode
  ext4: grab missed write_count for EXT4_IOC_SWAP_BOOT
  ext4: optimize block allocation on grow indepth
  ext4: get rid of code duplication
  ext4: fix over-defensive complaint after journal abort
  ext4: fix return value of ext4_do_update_inode
  ext4: fix mmap data corruption when blocksize < pagesize
  vfs: fix data corruption when blocksize < pagesize for mmaped data
  ext4: fold ext4_nojournal_sops into ext4_sops
  ext4: support freezing ext2 (nojournal) file systems
  ext4: fold ext4_sync_fs_nojournal() into ext4_sync_fs()
  ext4: don't check quota format when there are no quota files
  jbd2: simplify calling convention around __jbd2_journal_clean_checkpoint_list
  jbd2: avoid pointless scanning of checkpoint lists
  ...

fs/buffer.c

@@ -993,7 +993,7 @@ init_page_buffers(struct page *page, struct block_device *bdev,
  */
 static int
 grow_dev_page(struct block_device *bdev, sector_t block,
-		pgoff_t index, int size, int sizebits)
+	      pgoff_t index, int size, int sizebits, gfp_t gfp)
 {
 	struct inode *inode = bdev->bd_inode;
 	struct page *page;
@@ -1002,8 +1002,8 @@ grow_dev_page(struct block_device *bdev, sector_t block,
 	int ret = 0;		/* Will call free_more_memory() */
 	gfp_t gfp_mask;
 
-	gfp_mask = mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS;
-	gfp_mask |= __GFP_MOVABLE;
+	gfp_mask = (mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS) | gfp;
+
 	/*
 	 * XXX: __getblk_slow() can not really deal with failure and
 	 * will endlessly loop on improvised global reclaim.  Prefer
@@ -1060,7 +1060,7 @@ failed:
  * that page was dirty, the buffers are set dirty also.
  */
 static int
-grow_buffers(struct block_device *bdev, sector_t block, int size)
+grow_buffers(struct block_device *bdev, sector_t block, int size, gfp_t gfp)
 {
 	pgoff_t index;
 	int sizebits;
@@ -1087,11 +1087,12 @@ grow_buffers(struct block_device *bdev, sector_t block, int size)
 	}
 
 	/* Create a page with the proper size buffers.. */
-	return grow_dev_page(bdev, block, index, size, sizebits);
+	return grow_dev_page(bdev, block, index, size, sizebits, gfp);
 }
 
-static struct buffer_head *
-__getblk_slow(struct block_device *bdev, sector_t block, int size)
+struct buffer_head *
+__getblk_slow(struct block_device *bdev, sector_t block,
+	     unsigned size, gfp_t gfp)
 {
 	/* Size must be multiple of hard sectorsize */
 	if (unlikely(size & (bdev_logical_block_size(bdev)-1) ||
@@ -1113,13 +1114,14 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size)
 		if (bh)
 			return bh;
 
-		ret = grow_buffers(bdev, block, size);
+		ret = grow_buffers(bdev, block, size, gfp);
 		if (ret < 0)
 			return NULL;
 		if (ret == 0)
 			free_more_memory();
 	}
 }
+EXPORT_SYMBOL(__getblk_slow);
 
 /*
  * The relationship between dirty buffers and dirty pages:
@@ -1373,24 +1375,25 @@ __find_get_block(struct block_device *bdev, sector_t block, unsigned size)
 EXPORT_SYMBOL(__find_get_block);
 
 /*
- * __getblk will locate (and, if necessary, create) the buffer_head
+ * __getblk_gfp() will locate (and, if necessary, create) the buffer_head
  * which corresponds to the passed block_device, block and size. The
  * returned buffer has its reference count incremented.
  *
- * __getblk() will lock up the machine if grow_dev_page's try_to_free_buffers()
- * attempt is failing.  FIXME, perhaps?
+ * __getblk_gfp() will lock up the machine if grow_dev_page's
+ * try_to_free_buffers() attempt is failing.  FIXME, perhaps?
  */
 struct buffer_head *
-__getblk(struct block_device *bdev, sector_t block, unsigned size)
+__getblk_gfp(struct block_device *bdev, sector_t block,
+	     unsigned size, gfp_t gfp)
 {
 	struct buffer_head *bh = __find_get_block(bdev, block, size);
 
 	might_sleep();
 	if (bh == NULL)
-		bh = __getblk_slow(bdev, block, size);
+		bh = __getblk_slow(bdev, block, size, gfp);
 	return bh;
 }
-EXPORT_SYMBOL(__getblk);
+EXPORT_SYMBOL(__getblk_gfp);
 
 /*
  * Do async read-ahead on a buffer..
@@ -1406,24 +1409,28 @@ void __breadahead(struct block_device *bdev, sector_t block, unsigned size)
 EXPORT_SYMBOL(__breadahead);
 
 /**
- *  __bread() - reads a specified block and returns the bh
+ *  __bread_gfp() - reads a specified block and returns the bh
  *  @bdev: the block_device to read from
  *  @block: number of block
  *  @size: size (in bytes) to read
- * 
+ *  @gfp: page allocation flag
+ *
  *  Reads a specified block, and returns buffer head that contains it.
+ *  The page cache can be allocated from non-movable area
+ *  not to prevent page migration if you set gfp to zero.
  *  It returns NULL if the block was unreadable.
  */
 struct buffer_head *
-__bread(struct block_device *bdev, sector_t block, unsigned size)
+__bread_gfp(struct block_device *bdev, sector_t block,
+		   unsigned size, gfp_t gfp)
 {
-	struct buffer_head *bh = __getblk(bdev, block, size);
+	struct buffer_head *bh = __getblk_gfp(bdev, block, size, gfp);
 
 	if (likely(bh) && !buffer_uptodate(bh))
 		bh = __bread_slow(bh);
 	return bh;
 }
-EXPORT_SYMBOL(__bread);
+EXPORT_SYMBOL(__bread_gfp);
 
 /*
  * invalidate_bh_lrus() is called rarely - but not only at unmount.
@@ -2082,6 +2089,7 @@ int generic_write_end(struct file *file, struct address_space *mapping,
 			struct page *page, void *fsdata)
 {
 	struct inode *inode = mapping->host;
+	loff_t old_size = inode->i_size;
 	int i_size_changed = 0;
 
 	copied = block_write_end(file, mapping, pos, len, copied, page, fsdata);
@@ -2101,6 +2109,8 @@ int generic_write_end(struct file *file, struct address_space *mapping,
 	unlock_page(page);
 	page_cache_release(page);
 
+	if (old_size < pos)
+		pagecache_isize_extended(inode, old_size, pos);
 	/*
 	 * Don't mark the inode dirty under page lock. First, it unnecessarily
 	 * makes the holding time of page lock longer. Second, it forces lock

fs/ext4/balloc.c

@@ -176,7 +176,7 @@ static unsigned int num_clusters_in_group(struct super_block *sb,
 }
 
 /* Initializes an uninitialized block bitmap */
-static void ext4_init_block_bitmap(struct super_block *sb,
+static int ext4_init_block_bitmap(struct super_block *sb,
 				   struct buffer_head *bh,
 				   ext4_group_t block_group,
 				   struct ext4_group_desc *gdp)
@@ -192,7 +192,6 @@ static void ext4_init_block_bitmap(struct super_block *sb,
 	/* If checksum is bad mark all blocks used to prevent allocation
 	 * essentially implementing a per-group read-only flag. */
 	if (!ext4_group_desc_csum_verify(sb, block_group, gdp)) {
-		ext4_error(sb, "Checksum bad for group %u", block_group);
 		grp = ext4_get_group_info(sb, block_group);
 		if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
 			percpu_counter_sub(&sbi->s_freeclusters_counter,
@@ -205,7 +204,7 @@ static void ext4_init_block_bitmap(struct super_block *sb,
 					   count);
 		}
 		set_bit(EXT4_GROUP_INFO_IBITMAP_CORRUPT_BIT, &grp->bb_state);
-		return;
+		return -EIO;
 	}
 	memset(bh->b_data, 0, sb->s_blocksize);
 
@@ -243,6 +242,7 @@ static void ext4_init_block_bitmap(struct super_block *sb,
 			     sb->s_blocksize * 8, bh->b_data);
 	ext4_block_bitmap_csum_set(sb, block_group, gdp, bh);
 	ext4_group_desc_csum_set(sb, block_group, gdp);
+	return 0;
 }
 
 /* Return the number of free blocks in a block group.  It is used when
@@ -438,11 +438,15 @@ ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
 	}
 	ext4_lock_group(sb, block_group);
 	if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
-		ext4_init_block_bitmap(sb, bh, block_group, desc);
+		int err;
+
+		err = ext4_init_block_bitmap(sb, bh, block_group, desc);
 		set_bitmap_uptodate(bh);
 		set_buffer_uptodate(bh);
 		ext4_unlock_group(sb, block_group);
 		unlock_buffer(bh);
+		if (err)
+			ext4_error(sb, "Checksum bad for grp %u", block_group);
 		return bh;
 	}
 	ext4_unlock_group(sb, block_group);
@@ -636,8 +640,7 @@ ext4_fsblk_t ext4_new_meta_blocks(handle_t *handle, struct inode *inode,
 	 * Account for the allocated meta blocks.  We will never
 	 * fail EDQUOT for metdata, but we do account for it.
 	 */
-	if (!(*errp) &&
-	    ext4_test_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED)) {
+	if (!(*errp) && (flags & EXT4_MB_DELALLOC_RESERVED)) {
 		spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
 		spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
 		dquot_alloc_block_nofail(inode,

fs/ext4/bitmap.c

@@ -24,8 +24,7 @@ int ext4_inode_bitmap_csum_verify(struct super_block *sb, ext4_group_t group,
 	__u32 provided, calculated;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-					EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return 1;
 
 	provided = le16_to_cpu(gdp->bg_inode_bitmap_csum_lo);
@@ -46,8 +45,7 @@ void ext4_inode_bitmap_csum_set(struct super_block *sb, ext4_group_t group,
 	__u32 csum;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-					EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return;
 
 	csum = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)bh->b_data, sz);
@@ -65,8 +63,7 @@ int ext4_block_bitmap_csum_verify(struct super_block *sb, ext4_group_t group,
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	int sz = EXT4_CLUSTERS_PER_GROUP(sb) / 8;
 
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-					EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return 1;
 
 	provided = le16_to_cpu(gdp->bg_block_bitmap_csum_lo);
@@ -91,8 +88,7 @@ void ext4_block_bitmap_csum_set(struct super_block *sb, ext4_group_t group,
 	__u32 csum;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-			EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return;
 
 	csum = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)bh->b_data, sz);

fs/ext4/dir.c

@@ -151,13 +151,11 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 					&file->f_ra, file,
 					index, 1);
 			file->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT;
-			bh = ext4_bread(NULL, inode, map.m_lblk, 0, &err);
+			bh = ext4_bread(NULL, inode, map.m_lblk, 0);
+			if (IS_ERR(bh))
+				return PTR_ERR(bh);
 		}
 
-		/*
-		 * We ignore I/O errors on directories so users have a chance
-		 * of recovering data when there's a bad sector
-		 */
 		if (!bh) {
 			if (!dir_has_error) {
 				EXT4_ERROR_FILE(file, 0,

fs/ext4/ext4.h

@@ -572,15 +572,15 @@ enum {
 
 /*
  * The bit position of these flags must not overlap with any of the
- * EXT4_GET_BLOCKS_*.  They are used by ext4_ext_find_extent(),
+ * EXT4_GET_BLOCKS_*.  They are used by ext4_find_extent(),
  * read_extent_tree_block(), ext4_split_extent_at(),
  * ext4_ext_insert_extent(), and ext4_ext_create_new_leaf().
  * EXT4_EX_NOCACHE is used to indicate that the we shouldn't be
  * caching the extents when reading from the extent tree while a
  * truncate or punch hole operation is in progress.
  */
-#define EXT4_EX_NOCACHE				0x0400
-#define EXT4_EX_FORCE_CACHE			0x0800
+#define EXT4_EX_NOCACHE				0x40000000
+#define EXT4_EX_FORCE_CACHE			0x20000000
 
 /*
  * Flags used by ext4_free_blocks
@@ -890,6 +890,7 @@ struct ext4_inode_info {
 	struct ext4_es_tree i_es_tree;
 	rwlock_t i_es_lock;
 	struct list_head i_es_lru;
+	unsigned int i_es_all_nr;	/* protected by i_es_lock */
 	unsigned int i_es_lru_nr;	/* protected by i_es_lock */
 	unsigned long i_touch_when;	/* jiffies of last accessing */
 
@@ -1174,6 +1175,9 @@ struct ext4_super_block {
 #define EXT4_MF_MNTDIR_SAMPLED	0x0001
 #define EXT4_MF_FS_ABORTED	0x0002	/* Fatal error detected */
 
+/* Number of quota types we support */
+#define EXT4_MAXQUOTAS 2
+
 /*
  * fourth extended-fs super-block data in memory
  */
@@ -1237,7 +1241,7 @@ struct ext4_sb_info {
 	u32 s_min_batch_time;
 	struct block_device *journal_bdev;
 #ifdef CONFIG_QUOTA
-	char *s_qf_names[MAXQUOTAS];		/* Names of quota files with journalled quota */
+	char *s_qf_names[EXT4_MAXQUOTAS];	/* Names of quota files with journalled quota */
 	int s_jquota_fmt;			/* Format of quota to use */
 #endif
 	unsigned int s_want_extra_isize; /* New inodes should reserve # bytes */
@@ -1330,8 +1334,7 @@ struct ext4_sb_info {
 	/* Reclaim extents from extent status tree */
 	struct shrinker s_es_shrinker;
 	struct list_head s_es_lru;
-	unsigned long s_es_last_sorted;
-	struct percpu_counter s_extent_cache_cnt;
+	struct ext4_es_stats s_es_stats;
 	struct mb_cache *s_mb_cache;
 	spinlock_t s_es_lru_lock ____cacheline_aligned_in_smp;
 
@@ -1399,7 +1402,6 @@ enum {
 	EXT4_STATE_EXT_MIGRATE,		/* Inode is migrating */
 	EXT4_STATE_DIO_UNWRITTEN,	/* need convert on dio done*/
 	EXT4_STATE_NEWENTRY,		/* File just added to dir */
-	EXT4_STATE_DELALLOC_RESERVED,	/* blks already reserved for delalloc */
 	EXT4_STATE_DIOREAD_LOCK,	/* Disable support for dio read
 					   nolocking */
 	EXT4_STATE_MAY_INLINE_DATA,	/* may have in-inode data */
@@ -2086,10 +2088,8 @@ extern int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
 extern int ext4_trim_fs(struct super_block *, struct fstrim_range *);
 
 /* inode.c */
-struct buffer_head *ext4_getblk(handle_t *, struct inode *,
-						ext4_lblk_t, int, int *);
-struct buffer_head *ext4_bread(handle_t *, struct inode *,
-						ext4_lblk_t, int, int *);
+struct buffer_head *ext4_getblk(handle_t *, struct inode *, ext4_lblk_t, int);
+struct buffer_head *ext4_bread(handle_t *, struct inode *, ext4_lblk_t, int);
 int ext4_get_block_write(struct inode *inode, sector_t iblock,
 			 struct buffer_head *bh_result, int create);
 int ext4_get_block(struct inode *inode, sector_t iblock,
@@ -2109,6 +2109,7 @@ int do_journal_get_write_access(handle_t *handle,
 #define CONVERT_INLINE_DATA	 2
 
 extern struct inode *ext4_iget(struct super_block *, unsigned long);
+extern struct inode *ext4_iget_normal(struct super_block *, unsigned long);
 extern int  ext4_write_inode(struct inode *, struct writeback_control *);
 extern int  ext4_setattr(struct dentry *, struct iattr *);
 extern int  ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
@@ -2332,10 +2333,18 @@ extern int ext4_register_li_request(struct super_block *sb,
 static inline int ext4_has_group_desc_csum(struct super_block *sb)
 {
 	return EXT4_HAS_RO_COMPAT_FEATURE(sb,
-					  EXT4_FEATURE_RO_COMPAT_GDT_CSUM |
-					  EXT4_FEATURE_RO_COMPAT_METADATA_CSUM);
+					  EXT4_FEATURE_RO_COMPAT_GDT_CSUM) ||
+	       (EXT4_SB(sb)->s_chksum_driver != NULL);
 }
 
+static inline int ext4_has_metadata_csum(struct super_block *sb)
+{
+	WARN_ON_ONCE(EXT4_HAS_RO_COMPAT_FEATURE(sb,
+			EXT4_FEATURE_RO_COMPAT_METADATA_CSUM) &&
+		     !EXT4_SB(sb)->s_chksum_driver);
+
+	return (EXT4_SB(sb)->s_chksum_driver != NULL);
+}
 static inline ext4_fsblk_t ext4_blocks_count(struct ext4_super_block *es)
 {
 	return ((ext4_fsblk_t)le32_to_cpu(es->s_blocks_count_hi) << 32) |
@@ -2731,21 +2740,26 @@ extern int ext4_can_extents_be_merged(struct inode *inode,
 				      struct ext4_extent *ex1,
 				      struct ext4_extent *ex2);
 extern int ext4_ext_insert_extent(handle_t *, struct inode *,
-				  struct ext4_ext_path *,
+				  struct ext4_ext_path **,
 				  struct ext4_extent *, int);
-extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
-						  struct ext4_ext_path *,
-						  int flags);
+extern struct ext4_ext_path *ext4_find_extent(struct inode *, ext4_lblk_t,
+					      struct ext4_ext_path **,
+					      int flags);
 extern void ext4_ext_drop_refs(struct ext4_ext_path *);
 extern int ext4_ext_check_inode(struct inode *inode);
 extern int ext4_find_delalloc_range(struct inode *inode,
 				    ext4_lblk_t lblk_start,
 				    ext4_lblk_t lblk_end);
 extern int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk);
+extern ext4_lblk_t ext4_ext_next_allocated_block(struct ext4_ext_path *path);
 extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			__u64 start, __u64 len);
 extern int ext4_ext_precache(struct inode *inode);
 extern int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len);
+extern int ext4_swap_extents(handle_t *handle, struct inode *inode1,
+				struct inode *inode2, ext4_lblk_t lblk1,
+			     ext4_lblk_t lblk2,  ext4_lblk_t count,
+			     int mark_unwritten,int *err);
 
 /* move_extent.c */
 extern void ext4_double_down_write_data_sem(struct inode *first,
@@ -2755,8 +2769,6 @@ extern void ext4_double_up_write_data_sem(struct inode *orig_inode,
 extern int ext4_move_extents(struct file *o_filp, struct file *d_filp,
 			     __u64 start_orig, __u64 start_donor,
 			     __u64 len, __u64 *moved_len);
-extern int mext_next_extent(struct inode *inode, struct ext4_ext_path *path,
-			    struct ext4_extent **extent);
 
 /* page-io.c */
 extern int __init ext4_init_pageio(void);

fs/ext4/ext4_extents.h

@@ -123,6 +123,7 @@ find_ext4_extent_tail(struct ext4_extent_header *eh)
 struct ext4_ext_path {
 	ext4_fsblk_t			p_block;
 	__u16				p_depth;
+	__u16				p_maxdepth;
 	struct ext4_extent		*p_ext;
 	struct ext4_extent_idx		*p_idx;
 	struct ext4_extent_header	*p_hdr;

fs/ext4/ext4_jbd2.c

@@ -256,8 +256,8 @@ int __ext4_handle_dirty_metadata(const char *where, unsigned int line,
 	set_buffer_prio(bh);
 	if (ext4_handle_valid(handle)) {
 		err = jbd2_journal_dirty_metadata(handle, bh);
-		/* Errors can only happen if there is a bug */
-		if (WARN_ON_ONCE(err)) {
+		/* Errors can only happen due to aborted journal or a nasty bug */
+		if (!is_handle_aborted(handle) && WARN_ON_ONCE(err)) {
 			ext4_journal_abort_handle(where, line, __func__, bh,
 						  handle, err);
 			if (inode == NULL) {

fs/ext4/ext4_jbd2.h

@@ -102,9 +102,9 @@
 #define EXT4_QUOTA_INIT_BLOCKS(sb) 0
 #define EXT4_QUOTA_DEL_BLOCKS(sb) 0
 #endif
-#define EXT4_MAXQUOTAS_TRANS_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_TRANS_BLOCKS(sb))
-#define EXT4_MAXQUOTAS_INIT_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_INIT_BLOCKS(sb))
-#define EXT4_MAXQUOTAS_DEL_BLOCKS(sb) (MAXQUOTAS*EXT4_QUOTA_DEL_BLOCKS(sb))
+#define EXT4_MAXQUOTAS_TRANS_BLOCKS(sb) (EXT4_MAXQUOTAS*EXT4_QUOTA_TRANS_BLOCKS(sb))
+#define EXT4_MAXQUOTAS_INIT_BLOCKS(sb) (EXT4_MAXQUOTAS*EXT4_QUOTA_INIT_BLOCKS(sb))
+#define EXT4_MAXQUOTAS_DEL_BLOCKS(sb) (EXT4_MAXQUOTAS*EXT4_QUOTA_DEL_BLOCKS(sb))
 
 static inline int ext4_jbd2_credits_xattr(struct inode *inode)
 {

fs/ext4/extents_status.c

@@ -11,6 +11,8 @@
  */
 #include <linux/rbtree.h>
 #include <linux/list_sort.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
 #include "ext4.h"
 #include "extents_status.h"
 
@@ -313,19 +315,27 @@ ext4_es_alloc_extent(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len,
 	 */
 	if (!ext4_es_is_delayed(es)) {
 		EXT4_I(inode)->i_es_lru_nr++;
-		percpu_counter_inc(&EXT4_SB(inode->i_sb)->s_extent_cache_cnt);
+		percpu_counter_inc(&EXT4_SB(inode->i_sb)->
+					s_es_stats.es_stats_lru_cnt);
 	}
 
+	EXT4_I(inode)->i_es_all_nr++;
+	percpu_counter_inc(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt);
+
 	return es;
 }
 
 static void ext4_es_free_extent(struct inode *inode, struct extent_status *es)
 {
+	EXT4_I(inode)->i_es_all_nr--;
+	percpu_counter_dec(&EXT4_SB(inode->i_sb)->s_es_stats.es_stats_all_cnt);
+
 	/* Decrease the lru counter when this es is not delayed */
 	if (!ext4_es_is_delayed(es)) {
 		BUG_ON(EXT4_I(inode)->i_es_lru_nr == 0);
 		EXT4_I(inode)->i_es_lru_nr--;
-		percpu_counter_dec(&EXT4_SB(inode->i_sb)->s_extent_cache_cnt);
+		percpu_counter_dec(&EXT4_SB(inode->i_sb)->
+					s_es_stats.es_stats_lru_cnt);
 	}
 
 	kmem_cache_free(ext4_es_cachep, es);
@@ -426,7 +436,7 @@ static void ext4_es_insert_extent_ext_check(struct inode *inode,
 	unsigned short ee_len;
 	int depth, ee_status, es_status;
 
-	path = ext4_ext_find_extent(inode, es->es_lblk, NULL, EXT4_EX_NOCACHE);
+	path = ext4_find_extent(inode, es->es_lblk, NULL, EXT4_EX_NOCACHE);
 	if (IS_ERR(path))
 		return;
 
@@ -499,10 +509,8 @@ static void ext4_es_insert_extent_ext_check(struct inode *inode,
 		}
 	}
 out:
-	if (path) {
-		ext4_ext_drop_refs(path);
-		kfree(path);
-	}
+	ext4_ext_drop_refs(path);
+	kfree(path);
 }
 
 static void ext4_es_insert_extent_ind_check(struct inode *inode,
@@ -731,6 +739,7 @@ int ext4_es_lookup_extent(struct inode *inode, ext4_lblk_t lblk,
 			  struct extent_status *es)
 {
 	struct ext4_es_tree *tree;
+	struct ext4_es_stats *stats;
 	struct extent_status *es1 = NULL;
 	struct rb_node *node;
 	int found = 0;
@@ -767,11 +776,15 @@ int ext4_es_lookup_extent(struct inode *inode, ext4_lblk_t lblk,
 	}
 
 out:
+	stats = &EXT4_SB(inode->i_sb)->s_es_stats;
 	if (found) {
 		BUG_ON(!es1);
 		es->es_lblk = es1->es_lblk;
 		es->es_len = es1->es_len;
 		es->es_pblk = es1->es_pblk;
+		stats->es_stats_cache_hits++;
+	} else {
+		stats->es_stats_cache_misses++;
 	}
 
 	read_unlock(&EXT4_I(inode)->i_es_lock);
@@ -933,11 +946,16 @@ static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
 			    struct ext4_inode_info *locked_ei)
 {
 	struct ext4_inode_info *ei;
+	struct ext4_es_stats *es_stats;
 	struct list_head *cur, *tmp;
 	LIST_HEAD(skipped);
+	ktime_t start_time;
+	u64 scan_time;
 	int nr_shrunk = 0;
 	int retried = 0, skip_precached = 1, nr_skipped = 0;
 
+	es_stats = &sbi->s_es_stats;
+	start_time = ktime_get();
 	spin_lock(&sbi->s_es_lru_lock);
 
 retry:
@@ -948,7 +966,8 @@ retry:
 		 * If we have already reclaimed all extents from extent
 		 * status tree, just stop the loop immediately.
 		 */
-		if (percpu_counter_read_positive(&sbi->s_extent_cache_cnt) == 0)
+		if (percpu_counter_read_positive(
+				&es_stats->es_stats_lru_cnt) == 0)
 			break;
 
 		ei = list_entry(cur, struct ext4_inode_info, i_es_lru);
@@ -958,7 +977,7 @@ retry:
 		 * time.  Normally we try hard to avoid shrinking
 		 * precached inodes, but we will as a last resort.
 		 */
-		if ((sbi->s_es_last_sorted < ei->i_touch_when) ||
+		if ((es_stats->es_stats_last_sorted < ei->i_touch_when) ||
 		    (skip_precached && ext4_test_inode_state(&ei->vfs_inode,
 						EXT4_STATE_EXT_PRECACHED))) {
 			nr_skipped++;
@@ -992,7 +1011,7 @@ retry:
 	if ((nr_shrunk == 0) && nr_skipped && !retried) {
 		retried++;
 		list_sort(NULL, &sbi->s_es_lru, ext4_inode_touch_time_cmp);
-		sbi->s_es_last_sorted = jiffies;
+		es_stats->es_stats_last_sorted = jiffies;
 		ei = list_first_entry(&sbi->s_es_lru, struct ext4_inode_info,
 				      i_es_lru);
 		/*
@@ -1010,6 +1029,22 @@ retry:
 	if (locked_ei && nr_shrunk == 0)
 		nr_shrunk = __es_try_to_reclaim_extents(locked_ei, nr_to_scan);
 
+	scan_time = ktime_to_ns(ktime_sub(ktime_get(), start_time));
+	if (likely(es_stats->es_stats_scan_time))
+		es_stats->es_stats_scan_time = (scan_time +
+				es_stats->es_stats_scan_time*3) / 4;
+	else
+		es_stats->es_stats_scan_time = scan_time;
+	if (scan_time > es_stats->es_stats_max_scan_time)
+		es_stats->es_stats_max_scan_time = scan_time;
+	if (likely(es_stats->es_stats_shrunk))
+		es_stats->es_stats_shrunk = (nr_shrunk +
+				es_stats->es_stats_shrunk*3) / 4;
+	else
+		es_stats->es_stats_shrunk = nr_shrunk;
+
+	trace_ext4_es_shrink(sbi->s_sb, nr_shrunk, scan_time, skip_precached,
+			     nr_skipped, retried);
 	return nr_shrunk;
 }
 
@@ -1020,8 +1055,8 @@ static unsigned long ext4_es_count(struct shrinker *shrink,
 	struct ext4_sb_info *sbi;
 
 	sbi = container_of(shrink, struct ext4_sb_info, s_es_shrinker);
-	nr = percpu_counter_read_positive(&sbi->s_extent_cache_cnt);
-	trace_ext4_es_shrink_enter(sbi->s_sb, sc->nr_to_scan, nr);
+	nr = percpu_counter_read_positive(&sbi->s_es_stats.es_stats_lru_cnt);
+	trace_ext4_es_shrink_count(sbi->s_sb, sc->nr_to_scan, nr);
 	return nr;
 }
 
@@ -1033,31 +1068,160 @@ static unsigned long ext4_es_scan(struct shrinker *shrink,
 	int nr_to_scan = sc->nr_to_scan;
 	int ret, nr_shrunk;
 
-	ret = percpu_counter_read_positive(&sbi->s_extent_cache_cnt);
-	trace_ext4_es_shrink_enter(sbi->s_sb, nr_to_scan, ret);
+	ret = percpu_counter_read_positive(&sbi->s_es_stats.es_stats_lru_cnt);
+	trace_ext4_es_shrink_scan_enter(sbi->s_sb, nr_to_scan, ret);
 
 	if (!nr_to_scan)
 		return ret;
 
 	nr_shrunk = __ext4_es_shrink(sbi, nr_to_scan, NULL);
 
-	trace_ext4_es_shrink_exit(sbi->s_sb, nr_shrunk, ret);
+	trace_ext4_es_shrink_scan_exit(sbi->s_sb, nr_shrunk, ret);
 	return nr_shrunk;
 }
 
-void ext4_es_register_shrinker(struct ext4_sb_info *sbi)
+static void *ext4_es_seq_shrinker_info_start(struct seq_file *seq, loff_t *pos)
 {
+	return *pos ? NULL : SEQ_START_TOKEN;
+}
+
+static void *
+ext4_es_seq_shrinker_info_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	return NULL;
+}
+
+static int ext4_es_seq_shrinker_info_show(struct seq_file *seq, void *v)
+{
+	struct ext4_sb_info *sbi = seq->private;
+	struct ext4_es_stats *es_stats = &sbi->s_es_stats;
+	struct ext4_inode_info *ei, *max = NULL;
+	unsigned int inode_cnt = 0;
+
+	if (v != SEQ_START_TOKEN)
+		return 0;
+
+	/* here we just find an inode that has the max nr. of objects */
+	spin_lock(&sbi->s_es_lru_lock);
+	list_for_each_entry(ei, &sbi->s_es_lru, i_es_lru) {
+		inode_cnt++;
+		if (max && max->i_es_all_nr < ei->i_es_all_nr)
+			max = ei;
+		else if (!max)
+			max = ei;
+	}
+	spin_unlock(&sbi->s_es_lru_lock);
+
+	seq_printf(seq, "stats:\n  %lld objects\n  %lld reclaimable objects\n",
+		   percpu_counter_sum_positive(&es_stats->es_stats_all_cnt),
+		   percpu_counter_sum_positive(&es_stats->es_stats_lru_cnt));
+	seq_printf(seq, "  %lu/%lu cache hits/misses\n",
+		   es_stats->es_stats_cache_hits,
+		   es_stats->es_stats_cache_misses);
+	if (es_stats->es_stats_last_sorted != 0)
+		seq_printf(seq, "  %u ms last sorted interval\n",
+			   jiffies_to_msecs(jiffies -
+					    es_stats->es_stats_last_sorted));
+	if (inode_cnt)
+		seq_printf(seq, "  %d inodes on lru list\n", inode_cnt);
+
+	seq_printf(seq, "average:\n  %llu us scan time\n",
+	    div_u64(es_stats->es_stats_scan_time, 1000));
+	seq_printf(seq, "  %lu shrunk objects\n", es_stats->es_stats_shrunk);
+	if (inode_cnt)
+		seq_printf(seq,
+		    "maximum:\n  %lu inode (%u objects, %u reclaimable)\n"
+		    "  %llu us max scan time\n",
+		    max->vfs_inode.i_ino, max->i_es_all_nr, max->i_es_lru_nr,
+		    div_u64(es_stats->es_stats_max_scan_time, 1000));
+
+	return 0;
+}
+
+static void ext4_es_seq_shrinker_info_stop(struct seq_file *seq, void *v)
+{
+}
+
+static const struct seq_operations ext4_es_seq_shrinker_info_ops = {
+	.start = ext4_es_seq_shrinker_info_start,
+	.next  = ext4_es_seq_shrinker_info_next,
+	.stop  = ext4_es_seq_shrinker_info_stop,
+	.show  = ext4_es_seq_shrinker_info_show,
+};
+
+static int
+ext4_es_seq_shrinker_info_open(struct inode *inode, struct file *file)
+{
+	int ret;
+
+	ret = seq_open(file, &ext4_es_seq_shrinker_info_ops);
+	if (!ret) {
+		struct seq_file *m = file->private_data;
+		m->private = PDE_DATA(inode);
+	}
+
+	return ret;
+}
+
+static int
+ext4_es_seq_shrinker_info_release(struct inode *inode, struct file *file)
+{
+	return seq_release(inode, file);
+}
+
+static const struct file_operations ext4_es_seq_shrinker_info_fops = {
+	.owner		= THIS_MODULE,
+	.open		= ext4_es_seq_shrinker_info_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= ext4_es_seq_shrinker_info_release,
+};
+
+int ext4_es_register_shrinker(struct ext4_sb_info *sbi)
+{
+	int err;
+
 	INIT_LIST_HEAD(&sbi->s_es_lru);
 	spin_lock_init(&sbi->s_es_lru_lock);
-	sbi->s_es_last_sorted = 0;
+	sbi->s_es_stats.es_stats_last_sorted = 0;
+	sbi->s_es_stats.es_stats_shrunk = 0;
+	sbi->s_es_stats.es_stats_cache_hits = 0;
+	sbi->s_es_stats.es_stats_cache_misses = 0;
+	sbi->s_es_stats.es_stats_scan_time = 0;
+	sbi->s_es_stats.es_stats_max_scan_time = 0;
+	err = percpu_counter_init(&sbi->s_es_stats.es_stats_all_cnt, 0, GFP_KERNEL);
+	if (err)
+		return err;
+	err = percpu_counter_init(&sbi->s_es_stats.es_stats_lru_cnt, 0, GFP_KERNEL);
+	if (err)
+		goto err1;
+
 	sbi->s_es_shrinker.scan_objects = ext4_es_scan;
 	sbi->s_es_shrinker.count_objects = ext4_es_count;
 	sbi->s_es_shrinker.seeks = DEFAULT_SEEKS;
-	register_shrinker(&sbi->s_es_shrinker);
+	err = register_shrinker(&sbi->s_es_shrinker);
+	if (err)
+		goto err2;
+
+	if (sbi->s_proc)
+		proc_create_data("es_shrinker_info", S_IRUGO, sbi->s_proc,
+				 &ext4_es_seq_shrinker_info_fops, sbi);
+
+	return 0;
+
+err2:
+	percpu_counter_destroy(&sbi->s_es_stats.es_stats_lru_cnt);
+err1:
+	percpu_counter_destroy(&sbi->s_es_stats.es_stats_all_cnt);
+	return err;
 }
 
 void ext4_es_unregister_shrinker(struct ext4_sb_info *sbi)
 {
+	if (sbi->s_proc)
+		remove_proc_entry("es_shrinker_info", sbi->s_proc);
+	percpu_counter_destroy(&sbi->s_es_stats.es_stats_all_cnt);
+	percpu_counter_destroy(&sbi->s_es_stats.es_stats_lru_cnt);
 	unregister_shrinker(&sbi->s_es_shrinker);
 }
 

fs/ext4/extents_status.h

@@ -64,6 +64,17 @@ struct ext4_es_tree {
 	struct extent_status *cache_es;	/* recently accessed extent */
 };
 
+struct ext4_es_stats {
+	unsigned long es_stats_last_sorted;
+	unsigned long es_stats_shrunk;
+	unsigned long es_stats_cache_hits;
+	unsigned long es_stats_cache_misses;
+	u64 es_stats_scan_time;
+	u64 es_stats_max_scan_time;
+	struct percpu_counter es_stats_all_cnt;
+	struct percpu_counter es_stats_lru_cnt;
+};
+
 extern int __init ext4_init_es(void);
 extern void ext4_exit_es(void);
 extern void ext4_es_init_tree(struct ext4_es_tree *tree);
@@ -138,7 +149,7 @@ static inline void ext4_es_store_pblock_status(struct extent_status *es,
 		       (pb & ~ES_MASK));
 }
 
-extern void ext4_es_register_shrinker(struct ext4_sb_info *sbi);
+extern int ext4_es_register_shrinker(struct ext4_sb_info *sbi);
 extern void ext4_es_unregister_shrinker(struct ext4_sb_info *sbi);
 extern void ext4_es_lru_add(struct inode *inode);
 extern void ext4_es_lru_del(struct inode *inode);

fs/ext4/ialloc.c

@@ -1011,8 +1011,7 @@ got:
 	spin_unlock(&sbi->s_next_gen_lock);
 
 	/* Precompute checksum seed for inode metadata */
-	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
-			EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
+	if (ext4_has_metadata_csum(sb)) {
 		__u32 csum;
 		__le32 inum = cpu_to_le32(inode->i_ino);
 		__le32 gen = cpu_to_le32(inode->i_generation);

fs/ext4/indirect.c

@@ -318,34 +318,24 @@ static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned int blks,
  *	ext4_alloc_block() (normally -ENOSPC). Otherwise we set the chain
  *	as described above and return 0.
  */
-static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
-			     ext4_lblk_t iblock, int indirect_blks,
-			     int *blks, ext4_fsblk_t goal,
-			     ext4_lblk_t *offsets, Indirect *branch)
+static int ext4_alloc_branch(handle_t *handle,
+			     struct ext4_allocation_request *ar,
+			     int indirect_blks, ext4_lblk_t *offsets,
+			     Indirect *branch)
 {
-	struct ext4_allocation_request	ar;
 	struct buffer_head *		bh;
 	ext4_fsblk_t			b, new_blocks[4];
 	__le32				*p;
 	int				i, j, err, len = 1;
 
-	/*
-	 * Set up for the direct block allocation
-	 */
-	memset(&ar, 0, sizeof(ar));
-	ar.inode = inode;
-	ar.len = *blks;
-	ar.logical = iblock;
-	if (S_ISREG(inode->i_mode))
-		ar.flags = EXT4_MB_HINT_DATA;
-
 	for (i = 0; i <= indirect_blks; i++) {
 		if (i == indirect_blks) {
-			ar.goal = goal;
-			new_blocks[i] = ext4_mb_new_blocks(handle, &ar, &err);
+			new_blocks[i] = ext4_mb_new_blocks(handle, ar, &err);
 		} else
-			goal = new_blocks[i] = ext4_new_meta_blocks(handle, inode,
-							goal, 0, NULL, &err);
+			ar->goal = new_blocks[i] = ext4_new_meta_blocks(handle,
+					ar->inode, ar->goal,
+					ar->flags & EXT4_MB_DELALLOC_RESERVED,
+					NULL, &err);
 		if (err) {
 			i--;
 			goto failed;
@@ -354,7 +344,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 		if (i == 0)
 			continue;
 
-		bh = branch[i].bh = sb_getblk(inode->i_sb, new_blocks[i-1]);
+		bh = branch[i].bh = sb_getblk(ar->inode->i_sb, new_blocks[i-1]);
 		if (unlikely(!bh)) {
 			err = -ENOMEM;
 			goto failed;
@@ -372,7 +362,7 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 		b = new_blocks[i];
 
 		if (i == indirect_blks)
-			len = ar.len;
+			len = ar->len;
 		for (j = 0; j < len; j++)
 			*p++ = cpu_to_le32(b++);
 
@@ -381,11 +371,10 @@ static int ext4_alloc_branch(handle_t *handle, struct inode *inode,
 		unlock_buffer(bh);
 
 		BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
-		err = ext4_handle_dirty_metadata(handle, inode, bh);
+		err = ext4_handle_dirty_metadata(handle, ar->inode, bh);
 		if (err)
 			goto failed;
 	}
-	*blks = ar.len;
 	return 0;
 failed:
 	for (; i >= 0; i--) {
@@ -396,10 +385,10 @@ failed:
 		 * existing before ext4_alloc_branch() was called.
 		 */
 		if (i > 0 && i != indirect_blks && branch[i].bh)
-			ext4_forget(handle, 1, inode, branch[i].bh,
+			ext4_forget(handle, 1, ar->inode, branch[i].bh,
 				    branch[i].bh->b_blocknr);
-		ext4_free_blocks(handle, inode, NULL, new_blocks[i],
-				 (i == indirect_blks) ? ar.len : 1, 0);
+		ext4_free_blocks(handle, ar->inode, NULL, new_blocks[i],
+				 (i == indirect_blks) ? ar->len : 1, 0);
 	}
 	return err;
 }
@@ -419,9 +408,9 @@ failed:
  * inode (->i_blocks, etc.). In case of success we end up with the full
  * chain to new block and return 0.
  */
-static int ext4_splice_branch(handle_t *handle, struct inode *inode,
-			      ext4_lblk_t block, Indirect *where, int num,
-			      int blks)
+static int ext4_splice_branch(handle_t *handle,
+			      struct ext4_allocation_request *ar,
+			      Indirect *where, int num)
 {
 	int i;
 	int err = 0;
@@ -446,9 +435,9 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode,
 	 * Update the host buffer_head or inode to point to more just allocated
 	 * direct blocks blocks
 	 */
-	if (num == 0 && blks > 1) {
+	if (num == 0 && ar->len > 1) {
 		current_block = le32_to_cpu(where->key) + 1;
-		for (i = 1; i < blks; i++)
+		for (i = 1; i < ar->len; i++)
 			*(where->p + i) = cpu_to_le32(current_block++);
 	}
 
@@ -465,14 +454,14 @@ static int ext4_splice_branch(handle_t *handle, struct inode *inode,
 		 */
 		jbd_debug(5, "splicing indirect only\n");
 		BUFFER_TRACE(where->bh, "call ext4_handle_dirty_metadata");
-		err = ext4_handle_dirty_metadata(handle, inode, where->bh);
+		err = ext4_handle_dirty_metadata(handle, ar->inode, where->bh);
 		if (err)
 			goto err_out;
 	} else {
 		/*
 		 * OK, we spliced it into the inode itself on a direct block.
 		 */
-		ext4_mark_inode_dirty(handle, inode);
+		ext4_mark_inode_dirty(handle, ar->inode);
 		jbd_debug(5, "splicing direct\n");
 	}
 	return err;
@@ -484,11 +473,11 @@ err_out:
 		 * need to revoke the block, which is why we don't
 		 * need to set EXT4_FREE_BLOCKS_METADATA.
 		 */
-		ext4_free_blocks(handle, inode, where[i].bh, 0, 1,
+		ext4_free_blocks(handle, ar->inode, where[i].bh, 0, 1,
 				 EXT4_FREE_BLOCKS_FORGET);
 	}
-	ext4_free_blocks(handle, inode, NULL, le32_to_cpu(where[num].key),
-			 blks, 0);
+	ext4_free_blocks(handle, ar->inode, NULL, le32_to_cpu(where[num].key),
+			 ar->len, 0);
 
 	return err;
 }
@@ -525,11 +514,11 @@ int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 			struct ext4_map_blocks *map,
 			int flags)
 {
+	struct ext4_allocation_request ar;
 	int err = -EIO;
 	ext4_lblk_t offsets[4];
 	Indirect chain[4];
 	Indirect *partial;
-	ext4_fsblk_t goal;
 	int indirect_blks;
 	int blocks_to_boundary = 0;
 	int depth;
@@ -579,7 +568,16 @@ int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 		return -ENOSPC;
 	}
 
-	goal = ext4_find_goal(inode, map->m_lblk, partial);
+	/* Set up for the direct block allocation */
+	memset(&ar, 0, sizeof(ar));
+	ar.inode = inode;
+	ar.logical = map->m_lblk;
+	if (S_ISREG(inode->i_mode))
+		ar.flags = EXT4_MB_HINT_DATA;
+	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
+		ar.flags |= EXT4_MB_DELALLOC_RESERVED;
+
+	ar.goal = ext4_find_goal(inode, map->m_lblk, partial);
 
 	/* the number of blocks need to allocate for [d,t]indirect blocks */
 	indirect_blks = (chain + depth) - partial - 1;
@@ -588,13 +586,13 @@ int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 	 * Next look up the indirect map to count the totoal number of
 	 * direct blocks to allocate for this branch.
 	 */
-	count = ext4_blks_to_allocate(partial, indirect_blks,
-				      map->m_len, blocks_to_boundary);
+	ar.len = ext4_blks_to_allocate(partial, indirect_blks,
+				       map->m_len, blocks_to_boundary);
+
 	/*
 	 * Block out ext4_truncate while we alter the tree
 	 */
-	err = ext4_alloc_branch(handle, inode, map->m_lblk, indirect_blks,
-				&count, goal,
+	err = ext4_alloc_branch(handle, &ar, indirect_blks,
 				offsets + (partial - chain), partial);
 
 	/*
@@ -605,14 +603,14 @@ int ext4_ind_map_blocks(handle_t *handle, struct inode *inode,
 	 * may need to return -EAGAIN upwards in the worst case.  --sct
 	 */
 	if (!err)
-		err = ext4_splice_branch(handle, inode, map->m_lblk,
-					 partial, indirect_blks, count);
+		err = ext4_splice_branch(handle, &ar, partial, indirect_blks);
 	if (err)
 		goto cleanup;
 
 	map->m_flags |= EXT4_MAP_NEW;
 
 	ext4_update_inode_fsync_trans(handle, inode, 1);
+	count = ar.len;
 got_it:
 	map->m_flags |= EXT4_MAP_MAPPED;
 	map->m_pblk = le32_to_cpu(chain[depth-1].key);

fs/ext4/inline.c

@@ -594,6 +594,7 @@ retry:
 	if (ret) {
 		unlock_page(page);
 		page_cache_release(page);
+		page = NULL;
 		ext4_orphan_add(handle, inode);
 		up_write(&EXT4_I(inode)->xattr_sem);
 		sem_held = 0;
@@ -613,7 +614,8 @@ retry:
 	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
 		goto retry;
 
-	block_commit_write(page, from, to);
+	if (page)
+		block_commit_write(page, from, to);
 out:
 	if (page) {
 		unlock_page(page);
@@ -1126,8 +1128,7 @@ static int ext4_finish_convert_inline_dir(handle_t *handle,
 	memcpy((void *)de, buf + EXT4_INLINE_DOTDOT_SIZE,
 		inline_size - EXT4_INLINE_DOTDOT_SIZE);
 
-	if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
-				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (ext4_has_metadata_csum(inode->i_sb))
 		csum_size = sizeof(struct ext4_dir_entry_tail);
 
 	inode->i_size = inode->i_sb->s_blocksize;

fs/ext4/inode.c

@@ -83,8 +83,7 @@ static int ext4_inode_csum_verify(struct inode *inode, struct ext4_inode *raw,
 
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_LINUX) ||
-	    !EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
-		EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	    !ext4_has_metadata_csum(inode->i_sb))
 		return 1;
 
 	provided = le16_to_cpu(raw->i_checksum_lo);
@@ -105,8 +104,7 @@ static void ext4_inode_csum_set(struct inode *inode, struct ext4_inode *raw,
 
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_LINUX) ||
-	    !EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
-		EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	    !ext4_has_metadata_csum(inode->i_sb))
 		return;
 
 	csum = ext4_inode_csum(inode, raw, ei);
@@ -224,16 +222,15 @@ void ext4_evict_inode(struct inode *inode)
 		goto no_delete;
 	}
 
-	if (!is_bad_inode(inode))
-		dquot_initialize(inode);
+	if (is_bad_inode(inode))
+		goto no_delete;
+	dquot_initialize(inode);
 
 	if (ext4_should_order_data(inode))
 		ext4_begin_ordered_truncate(inode, 0);
 	truncate_inode_pages_final(&inode->i_data);
 
 	WARN_ON(atomic_read(&EXT4_I(inode)->i_ioend_count));
-	if (is_bad_inode(inode))
-		goto no_delete;
 
 	/*
 	 * Protect us against freezing - iput() caller didn't have to have any
@@ -590,19 +587,11 @@ found:
 	/*
 	 * New blocks allocate and/or writing to unwritten extent
 	 * will possibly result in updating i_data, so we take
-	 * the write lock of i_data_sem, and call get_blocks()
+	 * the write lock of i_data_sem, and call get_block()
 	 * with create == 1 flag.
 	 */
 	down_write(&EXT4_I(inode)->i_data_sem);
 
-	/*
-	 * if the caller is from delayed allocation writeout path
-	 * we have already reserved fs blocks for allocation
-	 * let the underlying get_block() function know to
-	 * avoid double accounting
-	 */
-	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
-		ext4_set_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
 	/*
 	 * We need to check for EXT4 here because migrate
 	 * could have changed the inode type in between
@@ -631,8 +620,6 @@ found:
 			(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE))
 			ext4_da_update_reserve_space(inode, retval, 1);
 	}
-	if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
-		ext4_clear_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
 
 	if (retval > 0) {
 		unsigned int status;
@@ -734,11 +721,11 @@ int ext4_get_block(struct inode *inode, sector_t iblock,
  * `handle' can be NULL if create is zero
  */
 struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
-				ext4_lblk_t block, int create, int *errp)
+				ext4_lblk_t block, int create)
 {
 	struct ext4_map_blocks map;
 	struct buffer_head *bh;
-	int fatal = 0, err;
+	int err;
 
 	J_ASSERT(handle != NULL || create == 0);
 
@@ -747,21 +734,14 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
 	err = ext4_map_blocks(handle, inode, &map,
 			      create ? EXT4_GET_BLOCKS_CREATE : 0);
 
-	/* ensure we send some value back into *errp */
-	*errp = 0;
-
-	if (create && err == 0)
-		err = -ENOSPC;	/* should never happen */
+	if (err == 0)
+		return create ? ERR_PTR(-ENOSPC) : NULL;
 	if (err < 0)
-		*errp = err;
-	if (err <= 0)
-		return NULL;
+		return ERR_PTR(err);
 
 	bh = sb_getblk(inode->i_sb, map.m_pblk);
-	if (unlikely(!bh)) {
-		*errp = -ENOMEM;
-		return NULL;
-	}
+	if (unlikely(!bh))
+		return ERR_PTR(-ENOMEM);
 	if (map.m_flags & EXT4_MAP_NEW) {
 		J_ASSERT(create != 0);
 		J_ASSERT(handle != NULL);
@@ -775,44 +755,44 @@ struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
 		 */
 		lock_buffer(bh);
 		BUFFER_TRACE(bh, "call get_create_access");
-		fatal = ext4_journal_get_create_access(handle, bh);
-		if (!fatal && !buffer_uptodate(bh)) {
+		err = ext4_journal_get_create_access(handle, bh);
+		if (unlikely(err)) {
+			unlock_buffer(bh);
+			goto errout;
+		}
+		if (!buffer_uptodate(bh)) {
 			memset(bh->b_data, 0, inode->i_sb->s_blocksize);
 			set_buffer_uptodate(bh);
 		}
 		unlock_buffer(bh);
 		BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
 		err = ext4_handle_dirty_metadata(handle, inode, bh);
-		if (!fatal)
-			fatal = err;
-	} else {
+		if (unlikely(err))
+			goto errout;
+	} else
 		BUFFER_TRACE(bh, "not a new buffer");
-	}
-	if (fatal) {
-		*errp = fatal;
-		brelse(bh);
-		bh = NULL;
-	}
 	return bh;
+errout:
+	brelse(bh);
+	return ERR_PTR(err);
 }
 
 struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode,
-			       ext4_lblk_t block, int create, int *err)
+			       ext4_lblk_t block, int create)
 {
 	struct buffer_head *bh;
 
-	bh = ext4_getblk(handle, inode, block, create, err);
-	if (!bh)
+	bh = ext4_getblk(handle, inode, block, create);
+	if (IS_ERR(bh))
 		return bh;
-	if (buffer_uptodate(bh))
+	if (!bh || buffer_uptodate(bh))
 		return bh;
 	ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &bh);
 	wait_on_buffer(bh);
 	if (buffer_uptodate(bh))
 		return bh;
 	put_bh(bh);
-	*err = -EIO;
-	return NULL;
+	return ERR_PTR(-EIO);
 }
 
 int ext4_walk_page_buffers(handle_t *handle,
@@ -1536,7 +1516,7 @@ out_unlock:
 }
 
 /*
- * This is a special get_blocks_t callback which is used by
+ * This is a special get_block_t callback which is used by
  * ext4_da_write_begin().  It will either return mapped block or
  * reserve space for a single block.
  *
@@ -2011,12 +1991,10 @@ static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
 	 * in data loss.  So use reserved blocks to allocate metadata if
 	 * possible.
 	 *
-	 * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE if the blocks
-	 * in question are delalloc blocks.  This affects functions in many
-	 * different parts of the allocation call path.  This flag exists
-	 * primarily because we don't want to change *many* call functions, so
-	 * ext4_map_blocks() will set the EXT4_STATE_DELALLOC_RESERVED flag
-	 * once the inode's allocation semaphore is taken.
+	 * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE if
+	 * the blocks in question are delalloc blocks.  This indicates
+	 * that the blocks and quotas has already been checked when
+	 * the data was copied into the page cache.
 	 */
 	get_blocks_flags = EXT4_GET_BLOCKS_CREATE |
 			   EXT4_GET_BLOCKS_METADATA_NOFAIL;
@@ -2515,6 +2493,20 @@ static int ext4_nonda_switch(struct super_block *sb)
 	return 0;
 }
 
+/* We always reserve for an inode update; the superblock could be there too */
+static int ext4_da_write_credits(struct inode *inode, loff_t pos, unsigned len)
+{
+	if (likely(EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
+				EXT4_FEATURE_RO_COMPAT_LARGE_FILE)))
+		return 1;
+
+	if (pos + len <= 0x7fffffffULL)
+		return 1;
+
+	/* We might need to update the superblock to set LARGE_FILE */
+	return 2;
+}
+
 static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 			       loff_t pos, unsigned len, unsigned flags,
 			       struct page **pagep, void **fsdata)
@@ -2565,7 +2557,8 @@ retry_grab:
 	 * of file which has an already mapped buffer.
 	 */
 retry_journal:
-	handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, 1);
+	handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE,
+				ext4_da_write_credits(inode, pos, len));
 	if (IS_ERR(handle)) {
 		page_cache_release(page);
 		return PTR_ERR(handle);
@@ -2658,10 +2651,7 @@ static int ext4_da_write_end(struct file *file,
 	if (copied && new_i_size > EXT4_I(inode)->i_disksize) {
 		if (ext4_has_inline_data(inode) ||
 		    ext4_da_should_update_i_disksize(page, end)) {
-			down_write(&EXT4_I(inode)->i_data_sem);
-			if (new_i_size > EXT4_I(inode)->i_disksize)
-				EXT4_I(inode)->i_disksize = new_i_size;
-			up_write(&EXT4_I(inode)->i_data_sem);
+			ext4_update_i_disksize(inode, new_i_size);
 			/* We need to mark inode dirty even if
 			 * new_i_size is less that inode->i_size
 			 * bu greater than i_disksize.(hint delalloc)
@@ -3936,8 +3926,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 		ei->i_extra_isize = 0;
 
 	/* Precompute checksum seed for inode metadata */
-	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
-			EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
+	if (ext4_has_metadata_csum(sb)) {
 		struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 		__u32 csum;
 		__le32 inum = cpu_to_le32(inode->i_ino);
@@ -4127,6 +4116,13 @@ bad_inode:
 	return ERR_PTR(ret);
 }
 
+struct inode *ext4_iget_normal(struct super_block *sb, unsigned long ino)
+{
+	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
+		return ERR_PTR(-EIO);
+	return ext4_iget(sb, ino);
+}
+
 static int ext4_inode_blocks_set(handle_t *handle,
 				struct ext4_inode *raw_inode,
 				struct ext4_inode_info *ei)
@@ -4226,7 +4222,8 @@ static int ext4_do_update_inode(handle_t *handle,
 	EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode);
 	EXT4_EINODE_SET_XTIME(i_crtime, ei, raw_inode);
 
-	if (ext4_inode_blocks_set(handle, raw_inode, ei)) {
+	err = ext4_inode_blocks_set(handle, raw_inode, ei);
+	if (err) {
 		spin_unlock(&ei->i_raw_lock);
 		goto out_brelse;
 	}
@@ -4536,8 +4533,12 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 				ext4_orphan_del(NULL, inode);
 				goto err_out;
 			}
-		} else
+		} else {
+			loff_t oldsize = inode->i_size;
+
 			i_size_write(inode, attr->ia_size);
+			pagecache_isize_extended(inode, oldsize, inode->i_size);
+		}
 
 		/*
 		 * Blocks are going to be removed from the inode. Wait

fs/ext4/ioctl.c

@@ -331,8 +331,7 @@ flags_out:
 		if (!inode_owner_or_capable(inode))
 			return -EPERM;
 
-		if (EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
-				EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
+		if (ext4_has_metadata_csum(inode->i_sb)) {
 			ext4_warning(sb, "Setting inode version is not "
 				     "supported with metadata_csum enabled.");
 			return -ENOTTY;
@@ -532,9 +531,17 @@ group_add_out:
 	}
 
 	case EXT4_IOC_SWAP_BOOT:
+	{
+		int err;
 		if (!(filp->f_mode & FMODE_WRITE))
 			return -EBADF;
-		return swap_inode_boot_loader(sb, inode);
+		err = mnt_want_write_file(filp);
+		if (err)
+			return err;
+		err = swap_inode_boot_loader(sb, inode);
+		mnt_drop_write_file(filp);
+		return err;
+	}
 
 	case EXT4_IOC_RESIZE_FS: {
 		ext4_fsblk_t n_blocks_count;

fs/ext4/mballoc.c

@@ -3155,9 +3155,8 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 			 "start %lu, size %lu, fe_logical %lu",
 			 (unsigned long) start, (unsigned long) size,
 			 (unsigned long) ac->ac_o_ex.fe_logical);
+		BUG();
 	}
-	BUG_ON(start + size <= ac->ac_o_ex.fe_logical &&
-			start > ac->ac_o_ex.fe_logical);
 	BUG_ON(size <= 0 || size > EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
 
 	/* now prepare goal request */
@@ -4410,14 +4409,7 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle,
 	if (IS_NOQUOTA(ar->inode))
 		ar->flags |= EXT4_MB_USE_ROOT_BLOCKS;
 
-	/*
-	 * For delayed allocation, we could skip the ENOSPC and
-	 * EDQUOT check, as blocks and quotas have been already
-	 * reserved when data being copied into pagecache.
-	 */
-	if (ext4_test_inode_state(ar->inode, EXT4_STATE_DELALLOC_RESERVED))
-		ar->flags |= EXT4_MB_DELALLOC_RESERVED;
-	else {
+	if ((ar->flags & EXT4_MB_DELALLOC_RESERVED) == 0) {
 		/* Without delayed allocation we need to verify
 		 * there is enough free blocks to do block allocation
 		 * and verify allocation doesn't exceed the quota limits.
@@ -4528,8 +4520,7 @@ out:
 	if (inquota && ar->len < inquota)
 		dquot_free_block(ar->inode, EXT4_C2B(sbi, inquota - ar->len));
 	if (!ar->len) {
-		if (!ext4_test_inode_state(ar->inode,
-					   EXT4_STATE_DELALLOC_RESERVED))
+		if ((ar->flags & EXT4_MB_DELALLOC_RESERVED) == 0)
 			/* release all the reserved blocks if non delalloc */
 			percpu_counter_sub(&sbi->s_dirtyclusters_counter,
 						reserv_clstrs);

fs/ext4/migrate.c

@@ -41,8 +41,7 @@ static int finish_range(handle_t *handle, struct inode *inode,
 	ext4_ext_store_pblock(&newext, lb->first_pblock);
 	/* Locking only for convinience since we are operating on temp inode */
 	down_write(&EXT4_I(inode)->i_data_sem);
-	path = ext4_ext_find_extent(inode, lb->first_block, NULL, 0);
-
+	path = ext4_find_extent(inode, lb->first_block, NULL, 0);
 	if (IS_ERR(path)) {
 		retval = PTR_ERR(path);
 		path = NULL;
@@ -81,13 +80,11 @@ static int finish_range(handle_t *handle, struct inode *inode,
 				goto err_out;
 		}
 	}
-	retval = ext4_ext_insert_extent(handle, inode, path, &newext, 0);
+	retval = ext4_ext_insert_extent(handle, inode, &path, &newext, 0);
 err_out:
 	up_write((&EXT4_I(inode)->i_data_sem));
-	if (path) {
-		ext4_ext_drop_refs(path);
-		kfree(path);
-	}
+	ext4_ext_drop_refs(path);
+	kfree(path);
 	lb->first_pblock = 0;
 	return retval;
 }

fs/ext4/mmp.c

@@ -20,8 +20,7 @@ static __le32 ext4_mmp_csum(struct super_block *sb, struct mmp_struct *mmp)
 
 static int ext4_mmp_csum_verify(struct super_block *sb, struct mmp_struct *mmp)
 {
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return 1;
 
 	return mmp->mmp_checksum == ext4_mmp_csum(sb, mmp);
@@ -29,8 +28,7 @@ static int ext4_mmp_csum_verify(struct super_block *sb, struct mmp_struct *mmp)
 
 static void ext4_mmp_csum_set(struct super_block *sb, struct mmp_struct *mmp)
 {
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-				       EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
+	if (!ext4_has_metadata_csum(sb))
 		return;
 
 	mmp->mmp_checksum = ext4_mmp_csum(sb, mmp);