Merge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (48 commits)
  ext4: fix hot spins in mballoc after err_freebuddy and err_freemeta
  ext4: fix test ext_generic_write_end() copied return value
  ext3: fix test ext_generic_write_end() copied return value
  ext4: Move mballoc headers/structures to a seperate header file mballoc.h
  ext4: cleanup for compiling mballoc with verification and debugging #defines
  ext4: don't use ext4_error in ext4_check_descriptors
  ext4: mark inode dirty after initializing the extent tree
  ext4: update ctime and mtime for truncate with extents.
  ext4: Don't do GFP_NOFS allocations after taking ext4_lock_group
  ext4: move headers out of include/linux
  ext4: fix wrong gfp type under transaction
  ext4: Fix hang on umount with quotas when journal is aborted
  ext4: Fix update of mtime and ctime on rename
  jdb2: replace remaining __FUNCTION__ occurrences
  ext4: replace remaining __FUNCTION__ occurrences
  jbd2: only create debugfs and stats entries if init is successful
  jbd2: fix kernel-doc notation
  jbd2: replace potentially false assertion with if block
  jbd2: eliminate duplicated code in revocation table init/destroy functions
  jbd2: tidy up revoke cache initialisation and destruction
  ...

arch/arm/mm/mmu.c

@@ -35,6 +35,7 @@ extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
  * zero-initialized data and COW.
  */
 struct page *empty_zero_page;
+EXPORT_SYMBOL(empty_zero_page);
 
 /*
  * The pmd table for the upper-most set of pages.

arch/m68k/mm/init.c

@@ -69,6 +69,7 @@ void __init m68k_setup_node(int node)
  */
 
 void *empty_zero_page;
+EXPORT_SYMBOL(empty_zero_page);
 
 void show_mem(void)
 {

arch/sparc/kernel/sparc_ksyms.c

@@ -282,3 +282,5 @@ EXPORT_SYMBOL(do_BUG);
 
 /* Sun Power Management Idle Handler */
 EXPORT_SYMBOL(pm_idle);
+
+EXPORT_SYMBOL(empty_zero_page);

arch/sparc64/mm/init.c

@@ -160,6 +160,7 @@ extern unsigned int sparc_ramdisk_image;
 extern unsigned int sparc_ramdisk_size;
 
 struct page *mem_map_zero __read_mostly;
+EXPORT_SYMBOL(mem_map_zero);
 
 unsigned int sparc64_highest_unlocked_tlb_ent __read_mostly;
 

fs/ext3/inode.c

@@ -1261,10 +1261,11 @@ static int ext3_ordered_write_end(struct file *file,
 		new_i_size = pos + copied;
 		if (new_i_size > EXT3_I(inode)->i_disksize)
 			EXT3_I(inode)->i_disksize = new_i_size;
-		copied = ext3_generic_write_end(file, mapping, pos, len, copied,
+		ret2 = ext3_generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
-		if (copied < 0)
-			ret = copied;
+		copied = ret2;
+		if (ret2 < 0)
+			ret = ret2;
 	}
 	ret2 = ext3_journal_stop(handle);
 	if (!ret)
@@ -1289,10 +1290,11 @@ static int ext3_writeback_write_end(struct file *file,
 	if (new_i_size > EXT3_I(inode)->i_disksize)
 		EXT3_I(inode)->i_disksize = new_i_size;
 
-	copied = ext3_generic_write_end(file, mapping, pos, len, copied,
+	ret2 = ext3_generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
-	if (copied < 0)
-		ret = copied;
+	copied = ret2;
+	if (ret2 < 0)
+		ret = ret2;
 
 	ret2 = ext3_journal_stop(handle);
 	if (!ret)

fs/ext4/acl.c

@@ -9,8 +9,8 @@
 #include <linux/slab.h>
 #include <linux/capability.h>
 #include <linux/fs.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 #include "xattr.h"
 #include "acl.h"
 
@@ -37,7 +37,7 @@ ext4_acl_from_disk(const void *value, size_t size)
 		return ERR_PTR(-EINVAL);
 	if (count == 0)
 		return NULL;
-	acl = posix_acl_alloc(count, GFP_KERNEL);
+	acl = posix_acl_alloc(count, GFP_NOFS);
 	if (!acl)
 		return ERR_PTR(-ENOMEM);
 	for (n=0; n < count; n++) {
@@ -91,7 +91,7 @@ ext4_acl_to_disk(const struct posix_acl *acl, size_t *size)
 
 	*size = ext4_acl_size(acl->a_count);
 	ext_acl = kmalloc(sizeof(ext4_acl_header) + acl->a_count *
-			sizeof(ext4_acl_entry), GFP_KERNEL);
+			sizeof(ext4_acl_entry), GFP_NOFS);
 	if (!ext_acl)
 		return ERR_PTR(-ENOMEM);
 	ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION);
@@ -187,7 +187,7 @@ ext4_get_acl(struct inode *inode, int type)
 	}
 	retval = ext4_xattr_get(inode, name_index, "", NULL, 0);
 	if (retval > 0) {
-		value = kmalloc(retval, GFP_KERNEL);
+		value = kmalloc(retval, GFP_NOFS);
 		if (!value)
 			return ERR_PTR(-ENOMEM);
 		retval = ext4_xattr_get(inode, name_index, "", value, retval);
@@ -335,7 +335,7 @@ ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir)
 			if (error)
 				goto cleanup;
 		}
-		clone = posix_acl_clone(acl, GFP_KERNEL);
+		clone = posix_acl_clone(acl, GFP_NOFS);
 		error = -ENOMEM;
 		if (!clone)
 			goto cleanup;

fs/ext4/balloc.c

@@ -15,12 +15,12 @@
 #include <linux/capability.h>
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
-
+#include "ext4.h"
+#include "ext4_jbd2.h"
 #include "group.h"
+
 /*
  * balloc.c contains the blocks allocation and deallocation routines
  */
@@ -48,7 +48,6 @@ void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr,
 unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
 		 ext4_group_t block_group, struct ext4_group_desc *gdp)
 {
-	unsigned long start;
 	int bit, bit_max;
 	unsigned free_blocks, group_blocks;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -59,7 +58,7 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
 		/* 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(sbi, block_group, gdp)) {
-			ext4_error(sb, __FUNCTION__,
+			ext4_error(sb, __func__,
 				  "Checksum bad for group %lu\n", block_group);
 			gdp->bg_free_blocks_count = 0;
 			gdp->bg_free_inodes_count = 0;
@@ -106,11 +105,12 @@ unsigned ext4_init_block_bitmap(struct super_block *sb, struct buffer_head *bh,
 	free_blocks = group_blocks - bit_max;
 
 	if (bh) {
+		ext4_fsblk_t start;
+
 		for (bit = 0; bit < bit_max; bit++)
 			ext4_set_bit(bit, bh->b_data);
 
-		start = block_group * EXT4_BLOCKS_PER_GROUP(sb) +
-			le32_to_cpu(sbi->s_es->s_first_data_block);
+		start = ext4_group_first_block_no(sb, block_group);
 
 		/* Set bits for block and inode bitmaps, and inode table */
 		ext4_set_bit(ext4_block_bitmap(sb, gdp) - start, bh->b_data);
@@ -235,7 +235,7 @@ static int ext4_valid_block_bitmap(struct super_block *sb,
 		return 1;
 
 err_out:
-	ext4_error(sb, __FUNCTION__,
+	ext4_error(sb, __func__,
 			"Invalid block bitmap - "
 			"block_group = %d, block = %llu",
 			block_group, bitmap_blk);
@@ -264,7 +264,7 @@ read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 	bitmap_blk = ext4_block_bitmap(sb, desc);
 	bh = sb_getblk(sb, bitmap_blk);
 	if (unlikely(!bh)) {
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			    "Cannot read block bitmap - "
 			    "block_group = %d, block_bitmap = %llu",
 			    (int)block_group, (unsigned long long)bitmap_blk);
@@ -281,7 +281,7 @@ read_block_bitmap(struct super_block *sb, ext4_group_t block_group)
 	}
 	if (bh_submit_read(bh) < 0) {
 		put_bh(bh);
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			    "Cannot read block bitmap - "
 			    "block_group = %d, block_bitmap = %llu",
 			    (int)block_group, (unsigned long long)bitmap_blk);
@@ -360,7 +360,7 @@ restart:
 		BUG();
 }
 #define rsv_window_dump(root, verbose) \
-	__rsv_window_dump((root), (verbose), __FUNCTION__)
+	__rsv_window_dump((root), (verbose), __func__)
 #else
 #define rsv_window_dump(root, verbose) do {} while (0)
 #endif
@@ -740,7 +740,7 @@ do_more:
 		if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group),
 						bit + i, bitmap_bh->b_data)) {
 			jbd_unlock_bh_state(bitmap_bh);
-			ext4_error(sb, __FUNCTION__,
+			ext4_error(sb, __func__,
 				   "bit already cleared for block %llu",
 				   (ext4_fsblk_t)(block + i));
 			jbd_lock_bh_state(bitmap_bh);
@@ -752,9 +752,7 @@ do_more:
 	jbd_unlock_bh_state(bitmap_bh);
 
 	spin_lock(sb_bgl_lock(sbi, block_group));
-	desc->bg_free_blocks_count =
-		cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) +
-			group_freed);
+	le16_add_cpu(&desc->bg_free_blocks_count, group_freed);
 	desc->bg_checksum = ext4_group_desc_csum(sbi, block_group, desc);
 	spin_unlock(sb_bgl_lock(sbi, block_group));
 	percpu_counter_add(&sbi->s_freeblocks_counter, count);
@@ -1798,7 +1796,7 @@ allocated:
 			if (ext4_test_bit(grp_alloc_blk+i,
 					bh2jh(bitmap_bh)->b_committed_data)) {
 				printk("%s: block was unexpectedly set in "
-					"b_committed_data\n", __FUNCTION__);
+					"b_committed_data\n", __func__);
 			}
 		}
 	}
@@ -1823,8 +1821,7 @@ allocated:
 	spin_lock(sb_bgl_lock(sbi, group_no));
 	if (gdp->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT))
 		gdp->bg_flags &= cpu_to_le16(~EXT4_BG_BLOCK_UNINIT);
-	gdp->bg_free_blocks_count =
-			cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num);
+	le16_add_cpu(&gdp->bg_free_blocks_count, -num);
 	gdp->bg_checksum = ext4_group_desc_csum(sbi, group_no, gdp);
 	spin_unlock(sb_bgl_lock(sbi, group_no));
 	percpu_counter_sub(&sbi->s_freeblocks_counter, num);

fs/ext4/bitmap.c

@@ -9,7 +9,7 @@
 
 #include <linux/buffer_head.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
+#include "ext4.h"
 
 #ifdef EXT4FS_DEBUG
 

fs/ext4/dir.c

@@ -23,10 +23,10 @@
 
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
 #include <linux/buffer_head.h>
 #include <linux/slab.h>
 #include <linux/rbtree.h>
+#include "ext4.h"
 
 static unsigned char ext4_filetype_table[] = {
 	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
@@ -42,7 +42,7 @@ const struct file_operations ext4_dir_operations = {
 	.llseek		= generic_file_llseek,
 	.read		= generic_read_dir,
 	.readdir	= ext4_readdir,		/* we take BKL. needed?*/
-	.ioctl		= ext4_ioctl,		/* BKL held */
+	.unlocked_ioctl = ext4_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= ext4_compat_ioctl,
 #endif

include/linux/ext4_fs.h → fs/ext4/ext4.h

@@ -1,5 +1,5 @@
 /*
- *  linux/include/linux/ext4_fs.h
+ *  ext4.h
  *
  * Copyright (C) 1992, 1993, 1994, 1995
  * Remy Card (card@masi.ibp.fr)
@@ -13,14 +13,13 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  */
 
-#ifndef _LINUX_EXT4_FS_H
-#define _LINUX_EXT4_FS_H
+#ifndef _EXT4_H
+#define _EXT4_H
 
 #include <linux/types.h>
 #include <linux/blkdev.h>
 #include <linux/magic.h>
-
-#include <linux/ext4_fs_i.h>
+#include "ext4_i.h"
 
 /*
  * The second extended filesystem constants/structures
@@ -176,8 +175,7 @@ struct ext4_group_desc
 #define EXT4_BG_INODE_ZEROED	0x0004 /* On-disk itable initialized to zero */
 
 #ifdef __KERNEL__
-#include <linux/ext4_fs_i.h>
-#include <linux/ext4_fs_sb.h>
+#include "ext4_sb.h"
 #endif
 /*
  * Macro-instructions used to manage group descriptors
@@ -231,6 +229,7 @@ struct ext4_group_desc
 #define EXT4_TOPDIR_FL			0x00020000 /* Top of directory hierarchies*/
 #define EXT4_HUGE_FILE_FL               0x00040000 /* Set to each huge file */
 #define EXT4_EXTENTS_FL			0x00080000 /* Inode uses extents */
+#define EXT4_EXT_MIGRATE		0x00100000 /* Inode is migrating */
 #define EXT4_RESERVED_FL		0x80000000 /* reserved for ext4 lib */
 
 #define EXT4_FL_USER_VISIBLE		0x000BDFFF /* User visible flags */
@@ -1049,8 +1048,7 @@ extern int ext4_block_truncate_page(handle_t *handle, struct page *page,
 		struct address_space *mapping, loff_t from);
 
 /* ioctl.c */
-extern int ext4_ioctl (struct inode *, struct file *, unsigned int,
-		       unsigned long);
+extern long ext4_ioctl(struct file *, unsigned int, unsigned long);
 extern long ext4_compat_ioctl (struct file *, unsigned int, unsigned long);
 
 /* migrate.c */
@@ -1204,4 +1202,4 @@ extern int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode,
 			int extend_disksize);
 #endif	/* __KERNEL__ */
 
-#endif	/* _LINUX_EXT4_FS_H */
+#endif	/* _EXT4_H */

include/linux/ext4_fs_extents.h → fs/ext4/ext4_extents.h

@@ -16,10 +16,10 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-
  */
 
-#ifndef _LINUX_EXT4_EXTENTS
-#define _LINUX_EXT4_EXTENTS
+#ifndef _EXT4_EXTENTS
+#define _EXT4_EXTENTS
 
-#include <linux/ext4_fs.h>
+#include "ext4.h"
 
 /*
  * With AGGRESSIVE_TEST defined, the capacity of index/leaf blocks
@@ -228,5 +228,5 @@ extern int ext4_ext_search_left(struct inode *, struct ext4_ext_path *,
 extern int ext4_ext_search_right(struct inode *, struct ext4_ext_path *,
 						ext4_lblk_t *, ext4_fsblk_t *);
 extern void ext4_ext_drop_refs(struct ext4_ext_path *);
-#endif /* _LINUX_EXT4_EXTENTS */
+#endif /* _EXT4_EXTENTS */
 

include/linux/ext4_fs_i.h → fs/ext4/ext4_i.h

@@ -1,5 +1,5 @@
 /*
- *  linux/include/linux/ext4_fs_i.h
+ *  ext4_i.h
  *
  * Copyright (C) 1992, 1993, 1994, 1995
  * Remy Card (card@masi.ibp.fr)
@@ -13,8 +13,8 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  */
 
-#ifndef _LINUX_EXT4_FS_I
-#define _LINUX_EXT4_FS_I
+#ifndef _EXT4_I
+#define _EXT4_I
 
 #include <linux/rwsem.h>
 #include <linux/rbtree.h>
@@ -164,4 +164,4 @@ struct ext4_inode_info {
 	spinlock_t i_prealloc_lock;
 };
 
-#endif	/* _LINUX_EXT4_FS_I */
+#endif	/* _EXT4_I */

fs/ext4/ext4_jbd2.c

@@ -2,14 +2,14 @@
  * Interface between ext4 and JBD
  */
 
-#include <linux/ext4_jbd2.h>
+#include "ext4_jbd2.h"
 
 int __ext4_journal_get_undo_access(const char *where, handle_t *handle,
 				struct buffer_head *bh)
 {
 	int err = jbd2_journal_get_undo_access(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }
 
@@ -18,7 +18,7 @@ int __ext4_journal_get_write_access(const char *where, handle_t *handle,
 {
 	int err = jbd2_journal_get_write_access(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }
 
@@ -27,7 +27,7 @@ int __ext4_journal_forget(const char *where, handle_t *handle,
 {
 	int err = jbd2_journal_forget(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }
 
@@ -36,7 +36,7 @@ int __ext4_journal_revoke(const char *where, handle_t *handle,
 {
 	int err = jbd2_journal_revoke(handle, blocknr, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }
 
@@ -45,7 +45,7 @@ int __ext4_journal_get_create_access(const char *where,
 {
 	int err = jbd2_journal_get_create_access(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }
 
@@ -54,6 +54,6 @@ int __ext4_journal_dirty_metadata(const char *where,
 {
 	int err = jbd2_journal_dirty_metadata(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
+		ext4_journal_abort_handle(where, __func__, bh, handle, err);
 	return err;
 }

include/linux/ext4_jbd2.h → fs/ext4/ext4_jbd2.h

@@ -1,5 +1,5 @@
 /*
- * linux/include/linux/ext4_jbd2.h
+ * ext4_jbd2.h
  *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
  *
@@ -12,12 +12,12 @@
  * Ext4-specific journaling extensions.
  */
 
-#ifndef _LINUX_EXT4_JBD2_H
-#define _LINUX_EXT4_JBD2_H
+#ifndef _EXT4_JBD2_H
+#define _EXT4_JBD2_H
 
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
+#include "ext4.h"
 
 #define EXT4_JOURNAL(inode)	(EXT4_SB((inode)->i_sb)->s_journal)
 
@@ -228,4 +228,4 @@ static inline int ext4_should_writeback_data(struct inode *inode)
 	return 0;
 }
 
-#endif	/* _LINUX_EXT4_JBD2_H */
+#endif	/* _EXT4_JBD2_H */

include/linux/ext4_fs_sb.h → fs/ext4/ext4_sb.h

@@ -1,5 +1,5 @@
 /*
- *  linux/include/linux/ext4_fs_sb.h
+ *  ext4_sb.h
  *
  * Copyright (C) 1992, 1993, 1994, 1995
  * Remy Card (card@masi.ibp.fr)
@@ -13,8 +13,8 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  */
 
-#ifndef _LINUX_EXT4_FS_SB
-#define _LINUX_EXT4_FS_SB
+#ifndef _EXT4_SB
+#define _EXT4_SB
 
 #ifdef __KERNEL__
 #include <linux/timer.h>
@@ -145,4 +145,4 @@ struct ext4_sb_info {
 	struct ext4_locality_group *s_locality_groups;
 };
 
-#endif	/* _LINUX_EXT4_FS_SB */
+#endif	/* _EXT4_SB */

fs/ext4/extents.c

@@ -32,7 +32,6 @@
 #include <linux/module.h>
 #include <linux/fs.h>
 #include <linux/time.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/jbd2.h>
 #include <linux/highuid.h>
 #include <linux/pagemap.h>
@@ -40,8 +39,9 @@
 #include <linux/string.h>
 #include <linux/slab.h>
 #include <linux/falloc.h>
-#include <linux/ext4_fs_extents.h>
 #include <asm/uaccess.h>
+#include "ext4_jbd2.h"
+#include "ext4_extents.h"
 
 
 /*
@@ -308,7 +308,7 @@ corrupted:
 }
 
 #define ext4_ext_check_header(inode, eh, depth)	\
-	__ext4_ext_check_header(__FUNCTION__, inode, eh, depth)
+	__ext4_ext_check_header(__func__, inode, eh, depth)
 
 #ifdef EXT_DEBUG
 static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
@@ -614,7 +614,7 @@ static int ext4_ext_insert_index(handle_t *handle, struct inode *inode,
 
 	ix->ei_block = cpu_to_le32(logical);
 	ext4_idx_store_pblock(ix, ptr);
-	curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1);
+	le16_add_cpu(&curp->p_hdr->eh_entries, 1);
 
 	BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries)
 			     > le16_to_cpu(curp->p_hdr->eh_max));
@@ -736,7 +736,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
 	}
 	if (m) {
 		memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m);
-		neh->eh_entries = cpu_to_le16(le16_to_cpu(neh->eh_entries)+m);
+		le16_add_cpu(&neh->eh_entries, m);
 	}
 
 	set_buffer_uptodate(bh);
@@ -753,8 +753,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
 		err = ext4_ext_get_access(handle, inode, path + depth);
 		if (err)
 			goto cleanup;
-		path[depth].p_hdr->eh_entries =
-		     cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m);
+		le16_add_cpu(&path[depth].p_hdr->eh_entries, -m);
 		err = ext4_ext_dirty(handle, inode, path + depth);
 		if (err)
 			goto cleanup;
@@ -817,8 +816,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
 		if (m) {
 			memmove(++fidx, path[i].p_idx - m,
 				sizeof(struct ext4_extent_idx) * m);
-			neh->eh_entries =
-				cpu_to_le16(le16_to_cpu(neh->eh_entries) + m);
+			le16_add_cpu(&neh->eh_entries, m);
 		}
 		set_buffer_uptodate(bh);
 		unlock_buffer(bh);
@@ -834,7 +832,7 @@ static int ext4_ext_split(handle_t *handle, struct inode *inode,
 			err = ext4_ext_get_access(handle, inode, path + i);
 			if (err)
 				goto cleanup;
-			path[i].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[i].p_hdr->eh_entries)-m);
+			le16_add_cpu(&path[i].p_hdr->eh_entries, -m);
 			err = ext4_ext_dirty(handle, inode, path + i);
 			if (err)
 				goto cleanup;
@@ -1369,7 +1367,7 @@ int ext4_ext_try_to_merge(struct inode *inode,
 				* sizeof(struct ext4_extent);
 			memmove(ex + 1, ex + 2, len);
 		}
-		eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries) - 1);
+		le16_add_cpu(&eh->eh_entries, -1);
 		merge_done = 1;
 		WARN_ON(eh->eh_entries == 0);
 		if (!eh->eh_entries)
@@ -1560,7 +1558,7 @@ has_space:
 		path[depth].p_ext = nearex;
 	}
 
-	eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)+1);
+	le16_add_cpu(&eh->eh_entries, 1);
 	nearex = path[depth].p_ext;
 	nearex->ee_block = newext->ee_block;
 	ext4_ext_store_pblock(nearex, ext_pblock(newext));
@@ -1699,7 +1697,7 @@ static int ext4_ext_rm_idx(handle_t *handle, struct inode *inode,
 	err = ext4_ext_get_access(handle, inode, path);
 	if (err)
 		return err;
-	path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1);
+	le16_add_cpu(&path->p_hdr->eh_entries, -1);
 	err = ext4_ext_dirty(handle, inode, path);
 	if (err)
 		return err;
@@ -1902,7 +1900,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode,
 		if (num == 0) {
 			/* this extent is removed; mark slot entirely unused */
 			ext4_ext_store_pblock(ex, 0);
-			eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1);
+			le16_add_cpu(&eh->eh_entries, -1);
 		}
 
 		ex->ee_block = cpu_to_le32(block);
@@ -1979,7 +1977,7 @@ static int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start)
 	 * We start scanning from right side, freeing all the blocks
 	 * after i_size and walking into the tree depth-wise.
 	 */
-	path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL);
+	path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_NOFS);
 	if (path == NULL) {
 		ext4_journal_stop(handle);
 		return -ENOMEM;
@@ -2138,6 +2136,82 @@ void ext4_ext_release(struct super_block *sb)
 #endif
 }
 
+static void bi_complete(struct bio *bio, int error)
+{
+	complete((struct completion *)bio->bi_private);
+}
+
+/* FIXME!! we need to try to merge to left or right after zero-out  */
+static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex)
+{
+	int ret = -EIO;
+	struct bio *bio;
+	int blkbits, blocksize;
+	sector_t ee_pblock;
+	struct completion event;
+	unsigned int ee_len, len, done, offset;
+
+
+	blkbits   = inode->i_blkbits;
+	blocksize = inode->i_sb->s_blocksize;
+	ee_len    = ext4_ext_get_actual_len(ex);
+	ee_pblock = ext_pblock(ex);
+
+	/* convert ee_pblock to 512 byte sectors */
+	ee_pblock = ee_pblock << (blkbits - 9);
+
+	while (ee_len > 0) {
+
+		if (ee_len > BIO_MAX_PAGES)
+			len = BIO_MAX_PAGES;
+		else
+			len = ee_len;
+
+		bio = bio_alloc(GFP_NOIO, len);
+		if (!bio)
+			return -ENOMEM;
+		bio->bi_sector = ee_pblock;
+		bio->bi_bdev   = inode->i_sb->s_bdev;
+
+		done = 0;
+		offset = 0;
+		while (done < len) {
+			ret = bio_add_page(bio, ZERO_PAGE(0),
+							blocksize, offset);
+			if (ret != blocksize) {
+				/*
+				 * We can't add any more pages because of
+				 * hardware limitations.  Start a new bio.
+				 */
+				break;
+			}
+			done++;
+			offset += blocksize;
+			if (offset >= PAGE_CACHE_SIZE)
+				offset = 0;
+		}
+
+		init_completion(&event);
+		bio->bi_private = &event;
+		bio->bi_end_io = bi_complete;
+		submit_bio(WRITE, bio);
+		wait_for_completion(&event);
+
+		if (test_bit(BIO_UPTODATE, &bio->bi_flags))
+			ret = 0;
+		else {
+			ret = -EIO;
+			break;
+		}
+		bio_put(bio);
+		ee_len    -= done;
+		ee_pblock += done  << (blkbits - 9);
+	}
+	return ret;
+}
+
+#define EXT4_EXT_ZERO_LEN 7
+
 /*
  * This function is called by ext4_ext_get_blocks() if someone tries to write
  * to an uninitialized extent. It may result in splitting the uninitialized
@@ -2154,7 +2228,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 						ext4_lblk_t iblock,
 						unsigned long max_blocks)
 {
-	struct ext4_extent *ex, newex;
+	struct ext4_extent *ex, newex, orig_ex;
 	struct ext4_extent *ex1 = NULL;
 	struct ext4_extent *ex2 = NULL;
 	struct ext4_extent *ex3 = NULL;
@@ -2173,10 +2247,26 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 	allocated = ee_len - (iblock - ee_block);
 	newblock = iblock - ee_block + ext_pblock(ex);
 	ex2 = ex;
+	orig_ex.ee_block = ex->ee_block;
+	orig_ex.ee_len   = cpu_to_le16(ee_len);
+	ext4_ext_store_pblock(&orig_ex, ext_pblock(ex));
 
 	err = ext4_ext_get_access(handle, inode, path + depth);
 	if (err)
 		goto out;
+	/* If extent has less than 2*EXT4_EXT_ZERO_LEN zerout directly */
+	if (ee_len <= 2*EXT4_EXT_ZERO_LEN) {
+		err =  ext4_ext_zeroout(inode, &orig_ex);
+		if (err)
+			goto fix_extent_len;
+		/* update the extent length and mark as initialized */
+		ex->ee_block = orig_ex.ee_block;
+		ex->ee_len   = orig_ex.ee_len;
+		ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+		ext4_ext_dirty(handle, inode, path + depth);
+		/* zeroed the full extent */
+		return allocated;
+	}
 
 	/* ex1: ee_block to iblock - 1 : uninitialized */
 	if (iblock > ee_block) {
@@ -2195,19 +2285,103 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 	/* ex3: to ee_block + ee_len : uninitialised */
 	if (allocated > max_blocks) {
 		unsigned int newdepth;
+		/* If extent has less than EXT4_EXT_ZERO_LEN zerout directly */
+		if (allocated <= EXT4_EXT_ZERO_LEN) {
+			/* Mark first half uninitialized.
+			 * Mark second half initialized and zero out the
+			 * initialized extent
+			 */
+			ex->ee_block = orig_ex.ee_block;
+			ex->ee_len   = cpu_to_le16(ee_len - allocated);
+			ext4_ext_mark_uninitialized(ex);
+			ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+			ext4_ext_dirty(handle, inode, path + depth);
+
+			ex3 = &newex;
+			ex3->ee_block = cpu_to_le32(iblock);
+			ext4_ext_store_pblock(ex3, newblock);
+			ex3->ee_len = cpu_to_le16(allocated);
+			err = ext4_ext_insert_extent(handle, inode, path, ex3);
+			if (err == -ENOSPC) {
+				err =  ext4_ext_zeroout(inode, &orig_ex);
+				if (err)
+					goto fix_extent_len;
+				ex->ee_block = orig_ex.ee_block;
+				ex->ee_len   = orig_ex.ee_len;
+				ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+				ext4_ext_dirty(handle, inode, path + depth);
+				/* zeroed the full extent */
+				return allocated;
+
+			} else if (err)
+				goto fix_extent_len;
+
+			/*
+			 * We need to zero out the second half because
+			 * an fallocate request can update file size and
+			 * converting the second half to initialized extent
+			 * implies that we can leak some junk data to user
+			 * space.
+			 */
+			err =  ext4_ext_zeroout(inode, ex3);
+			if (err) {
+				/*
+				 * We should actually mark the
+				 * second half as uninit and return error
+				 * Insert would have changed the extent
+				 */
+				depth = ext_depth(inode);
+				ext4_ext_drop_refs(path);
+				path = ext4_ext_find_extent(inode,
+								iblock, path);
+				if (IS_ERR(path)) {
+					err = PTR_ERR(path);
+					return err;
+				}
+				ex = path[depth].p_ext;
+				err = ext4_ext_get_access(handle, inode,
+								path + depth);
+				if (err)
+					return err;
+				ext4_ext_mark_uninitialized(ex);
+				ext4_ext_dirty(handle, inode, path + depth);
+				return err;
+			}
+
+			/* zeroed the second half */
+			return allocated;
+		}
 		ex3 = &newex;
 		ex3->ee_block = cpu_to_le32(iblock + max_blocks);
 		ext4_ext_store_pblock(ex3, newblock + max_blocks);
 		ex3->ee_len = cpu_to_le16(allocated - max_blocks);
 		ext4_ext_mark_uninitialized(ex3);
 		err = ext4_ext_insert_extent(handle, inode, path, ex3);
-		if (err)
-			goto out;
+		if (err == -ENOSPC) {
+			err =  ext4_ext_zeroout(inode, &orig_ex);
+			if (err)
+				goto fix_extent_len;
+			/* update the extent length and mark as initialized */
+			ex->ee_block = orig_ex.ee_block;
+			ex->ee_len   = orig_ex.ee_len;
+			ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+			ext4_ext_dirty(handle, inode, path + depth);
+			/* zeroed the full extent */
+			return allocated;
+
+		} else if (err)
+			goto fix_extent_len;
 		/*
 		 * The depth, and hence eh & ex might change
 		 * as part of the insert above.
 		 */
 		newdepth = ext_depth(inode);
+		/*
+		 * update the extent length after successfull insert of the
+		 * split extent
+		 */
+		orig_ex.ee_len = cpu_to_le16(ee_len -
+						ext4_ext_get_actual_len(ex3));
 		if (newdepth != depth) {
 			depth = newdepth;
 			ext4_ext_drop_refs(path);
@@ -2226,6 +2400,24 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 				goto out;
 		}
 		allocated = max_blocks;
+
+		/* If extent has less than EXT4_EXT_ZERO_LEN and we are trying
+		 * to insert a extent in the middle zerout directly
+		 * otherwise give the extent a chance to merge to left
+		 */
+		if (le16_to_cpu(orig_ex.ee_len) <= EXT4_EXT_ZERO_LEN &&
+							iblock != ee_block) {
+			err =  ext4_ext_zeroout(inode, &orig_ex);
+			if (err)
+				goto fix_extent_len;
+			/* update the extent length and mark as initialized */
+			ex->ee_block = orig_ex.ee_block;
+			ex->ee_len   = orig_ex.ee_len;
+			ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+			ext4_ext_dirty(handle, inode, path + depth);
+			/* zero out the first half */
+			return allocated;
+		}
 	}
 	/*
 	 * If there was a change of depth as part of the
@@ -2282,8 +2474,29 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 	goto out;
 insert:
 	err = ext4_ext_insert_extent(handle, inode, path, &newex);
+	if (err == -ENOSPC) {
+		err =  ext4_ext_zeroout(inode, &orig_ex);
+		if (err)
+			goto fix_extent_len;
+		/* update the extent length and mark as initialized */
+		ex->ee_block = orig_ex.ee_block;
+		ex->ee_len   = orig_ex.ee_len;
+		ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+		ext4_ext_dirty(handle, inode, path + depth);
+		/* zero out the first half */
+		return allocated;
+	} else if (err)
+		goto fix_extent_len;
 out:
 	return err ? err : allocated;
+
+fix_extent_len:
+	ex->ee_block = orig_ex.ee_block;
+	ex->ee_len   = orig_ex.ee_len;
+	ext4_ext_store_pblock(ex, ext_pblock(&orig_ex));
+	ext4_ext_mark_uninitialized(ex);
+	ext4_ext_dirty(handle, inode, path + depth);
+	return err;
 }
 
 /*
@@ -2393,8 +2606,20 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
 			}
 			if (create == EXT4_CREATE_UNINITIALIZED_EXT)
 				goto out;
-			if (!create)
+			if (!create) {
+				/*
+				 * We have blocks reserved already.  We
+				 * return allocated blocks so that delalloc
+				 * won't do block reservation for us.  But
+				 * the buffer head will be unmapped so that
+				 * a read from the block returns 0s.
+				 */
+				if (allocated > max_blocks)
+					allocated = max_blocks;
+				/* mark the buffer unwritten */
+				__set_bit(BH_Unwritten, &bh_result->b_state);
 				goto out2;
+			}
 
 			ret = ext4_ext_convert_to_initialized(handle, inode,
 								path, iblock,
@@ -2584,6 +2809,8 @@ out_stop:
 		ext4_orphan_del(handle, inode);
 
 	up_write(&EXT4_I(inode)->i_data_sem);
+	inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
+	ext4_mark_inode_dirty(handle, inode);
 	ext4_journal_stop(handle);
 }
 
@@ -2608,6 +2835,28 @@ int ext4_ext_writepage_trans_blocks(struct inode *inode, int num)
 	return needed;
 }
 
+static void ext4_falloc_update_inode(struct inode *inode,
+				int mode, loff_t new_size, int update_ctime)
+{
+	struct timespec now;
+
+	if (update_ctime) {
+		now = current_fs_time(inode->i_sb);
+		if (!timespec_equal(&inode->i_ctime, &now))
+			inode->i_ctime = now;
+	}
+	/*
+	 * Update only when preallocation was requested beyond
+	 * the file size.
+	 */
+	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
+				new_size > i_size_read(inode)) {
+		i_size_write(inode, new_size);
+		EXT4_I(inode)->i_disksize = new_size;
+	}
+
+}
+
 /*
  * preallocate space for a file. This implements ext4's fallocate inode
  * operation, which gets called from sys_fallocate system call.
@@ -2619,8 +2868,8 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
 {
 	handle_t *handle;
 	ext4_lblk_t block;
+	loff_t new_size;
 	unsigned long max_blocks;
-	ext4_fsblk_t nblocks = 0;
 	int ret = 0;
 	int ret2 = 0;
 	int retries = 0;
@@ -2639,9 +2888,12 @@ long ext4_fallocate(struct inode *inode, int mode, loff_t offset, loff_t len)
 		return -ENODEV;
 
 	block = offset >> blkbits;
+	/*
+	 * We can't just convert len to max_blocks because
+	 * If blocksize = 4096 offset = 3072 and len = 2048
+	 */
 	max_blocks = (EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits)
-			- block;
-
+							- block;
 	/*
 	 * credits to insert 1 extent into extent tree + buffers to be able to
 	 * modify 1 super block, 1 block bitmap and 1 group descriptor.
@@ -2657,7 +2909,6 @@ retry:
 			ret = PTR_ERR(handle);
 			break;
 		}
-
 		ret = ext4_get_blocks_wrap(handle, inode, block,
 					  max_blocks, &map_bh,
 					  EXT4_CREATE_UNINITIALIZED_EXT, 0);
@@ -2673,61 +2924,24 @@ retry:
 			ret2 = ext4_journal_stop(handle);
 			break;
 		}
-		if (ret > 0) {
-			/* check wrap through sign-bit/zero here */
-			if ((block + ret) < 0 || (block + ret) < block) {
-				ret = -EIO;
-				ext4_mark_inode_dirty(handle, inode);
-				ret2 = ext4_journal_stop(handle);
-				break;
-			}
-			if (buffer_new(&map_bh) && ((block + ret) >
-			    (EXT4_BLOCK_ALIGN(i_size_read(inode), blkbits)
-			    >> blkbits)))
-					nblocks = nblocks + ret;
-		}
-
-		/* Update ctime if new blocks get allocated */
-		if (nblocks) {
-			struct timespec now;
-
-			now = current_fs_time(inode->i_sb);
-			if (!timespec_equal(&inode->i_ctime, &now))
-				inode->i_ctime = now;
-		}
+		if ((block + ret) >= (EXT4_BLOCK_ALIGN(offset + len,
+						blkbits) >> blkbits))
+			new_size = offset + len;
+		else
+			new_size = (block + ret) << blkbits;
 
+		ext4_falloc_update_inode(inode, mode, new_size,
+						buffer_new(&map_bh));
 		ext4_mark_inode_dirty(handle, inode);
 		ret2 = ext4_journal_stop(handle);
 		if (ret2)
 			break;
 	}
-
-	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
+	if (ret == -ENOSPC &&
+			ext4_should_retry_alloc(inode->i_sb, &retries)) {
+		ret = 0;
 		goto retry;
-
-	/*
-	 * Time to update the file size.
-	 * Update only when preallocation was requested beyond the file size.
-	 */
-	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-	    (offset + len) > i_size_read(inode)) {
-		if (ret > 0) {
-			/*
-			 * if no error, we assume preallocation succeeded
-			 * completely
-			 */
-			i_size_write(inode, offset + len);
-			EXT4_I(inode)->i_disksize = i_size_read(inode);
-		} else if (ret < 0 && nblocks) {
-			/* Handle partial allocation scenario */
-			loff_t newsize;
-
-			newsize  = (nblocks << blkbits) + i_size_read(inode);
-			i_size_write(inode, EXT4_BLOCK_ALIGN(newsize, blkbits));
-			EXT4_I(inode)->i_disksize = i_size_read(inode);
-		}
 	}
-
 	mutex_unlock(&inode->i_mutex);
 	return ret > 0 ? ret2 : ret;
 }

fs/ext4/file.c

@@ -21,8 +21,8 @@
 #include <linux/time.h>
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
+#include "ext4.h"
+#include "ext4_jbd2.h"
 #include "xattr.h"
 #include "acl.h"
 
@@ -129,7 +129,7 @@ const struct file_operations ext4_file_operations = {
 	.write		= do_sync_write,
 	.aio_read	= generic_file_aio_read,
 	.aio_write	= ext4_file_write,
-	.ioctl		= ext4_ioctl,
+	.unlocked_ioctl = ext4_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl	= ext4_compat_ioctl,
 #endif

fs/ext4/fsync.c

@@ -27,8 +27,8 @@
 #include <linux/sched.h>
 #include <linux/writeback.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
+#include "ext4.h"
+#include "ext4_jbd2.h"
 
 /*
  * akpm: A new design for ext4_sync_file().
@@ -72,6 +72,9 @@ int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync)
 		goto out;
 	}
 
+	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+		goto out;
+
 	/*
 	 * The VFS has written the file data.  If the inode is unaltered
 	 * then we need not start a commit.

fs/ext4/hash.c

@@ -11,8 +11,8 @@
 
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
 #include <linux/cryptohash.h>
+#include "ext4.h"
 
 #define DELTA 0x9E3779B9
 

fs/ext4/ialloc.c

@@ -15,8 +15,6 @@
 #include <linux/time.h>
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/stat.h>
 #include <linux/string.h>
 #include <linux/quotaops.h>
@@ -25,7 +23,8 @@
 #include <linux/bitops.h>
 #include <linux/blkdev.h>
 #include <asm/byteorder.h>
-
+#include "ext4.h"
+#include "ext4_jbd2.h"
 #include "xattr.h"
 #include "acl.h"
 #include "group.h"
@@ -75,7 +74,7 @@ unsigned ext4_init_inode_bitmap(struct super_block *sb, struct buffer_head *bh,
 	/* If checksum is bad mark all blocks and inodes use to prevent
 	 * allocation, essentially implementing a per-group read-only flag. */
 	if (!ext4_group_desc_csum_verify(sbi, block_group, gdp)) {
-		ext4_error(sb, __FUNCTION__, "Checksum bad for group %lu\n",
+		ext4_error(sb, __func__, "Checksum bad for group %lu\n",
 			   block_group);
 		gdp->bg_free_blocks_count = 0;
 		gdp->bg_free_inodes_count = 0;
@@ -223,11 +222,9 @@ void ext4_free_inode (handle_t *handle, struct inode * inode)
 
 		if (gdp) {
 			spin_lock(sb_bgl_lock(sbi, block_group));
-			gdp->bg_free_inodes_count = cpu_to_le16(
-				le16_to_cpu(gdp->bg_free_inodes_count) + 1);
+			le16_add_cpu(&gdp->bg_free_inodes_count, 1);
 			if (is_directory)
-				gdp->bg_used_dirs_count = cpu_to_le16(
-				  le16_to_cpu(gdp->bg_used_dirs_count) - 1);
+				le16_add_cpu(&gdp->bg_used_dirs_count, -1);
 			gdp->bg_checksum = ext4_group_desc_csum(sbi,
 							block_group, gdp);
 			spin_unlock(sb_bgl_lock(sbi, block_group));
@@ -588,7 +585,7 @@ got:
 	ino++;
 	if ((group == 0 && ino < EXT4_FIRST_INO(sb)) ||
 	    ino > EXT4_INODES_PER_GROUP(sb)) {
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			   "reserved inode or inode > inodes count - "
 			   "block_group = %lu, inode=%lu", group,
 			   ino + group * EXT4_INODES_PER_GROUP(sb));
@@ -664,11 +661,9 @@ got:
 				cpu_to_le16(EXT4_INODES_PER_GROUP(sb) - ino);
 	}
 
-	gdp->bg_free_inodes_count =
-		cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1);
+	le16_add_cpu(&gdp->bg_free_inodes_count, -1);
 	if (S_ISDIR(mode)) {
-		gdp->bg_used_dirs_count =
-			cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1);
+		le16_add_cpu(&gdp->bg_used_dirs_count, 1);
 	}
 	gdp->bg_checksum = ext4_group_desc_csum(sbi, group, gdp);
 	spin_unlock(sb_bgl_lock(sbi, group));
@@ -744,23 +739,24 @@ got:
 	if (err)
 		goto fail_free_drop;
 
-	err = ext4_mark_inode_dirty(handle, inode);
-	if (err) {
-		ext4_std_error(sb, err);
-		goto fail_free_drop;
-	}
 	if (test_opt(sb, EXTENTS)) {
-		/* set extent flag only for directory and file */
-		if (S_ISDIR(mode) || S_ISREG(mode)) {
+		/* set extent flag only for diretory, file and normal symlink*/
+		if (S_ISDIR(mode) || S_ISREG(mode) || S_ISLNK(mode)) {
 			EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL;
 			ext4_ext_tree_init(handle, inode);
 			err = ext4_update_incompat_feature(handle, sb,
 					EXT4_FEATURE_INCOMPAT_EXTENTS);
 			if (err)
-				goto fail;
+				goto fail_free_drop;
 		}
 	}
 
+	err = ext4_mark_inode_dirty(handle, inode);
+	if (err) {
+		ext4_std_error(sb, err);
+		goto fail_free_drop;
+	}
+
 	ext4_debug("allocating inode %lu\n", inode->i_ino);
 	goto really_out;
 fail:
@@ -796,7 +792,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 
 	/* Error cases - e2fsck has already cleaned up for us */
 	if (ino > max_ino) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "bad orphan ino %lu!  e2fsck was run?", ino);
 		goto error;
 	}
@@ -805,7 +801,7 @@ struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino)
 	bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb);
 	bitmap_bh = read_inode_bitmap(sb, block_group);
 	if (!bitmap_bh) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "inode bitmap error for orphan %lu", ino);
 		goto error;
 	}
@@ -830,7 +826,7 @@ iget_failed:
 	err = PTR_ERR(inode);
 	inode = NULL;
 bad_orphan:
-	ext4_warning(sb, __FUNCTION__,
+	ext4_warning(sb, __func__,
 		     "bad orphan inode %lu!  e2fsck was run?", ino);
 	printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n",
 	       bit, (unsigned long long)bitmap_bh->b_blocknr,

fs/ext4/inode.c

@@ -25,7 +25,6 @@
 #include <linux/module.h>
 #include <linux/fs.h>
 #include <linux/time.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/jbd2.h>
 #include <linux/highuid.h>
 #include <linux/pagemap.h>
@@ -36,6 +35,7 @@
 #include <linux/mpage.h>
 #include <linux/uio.h>
 #include <linux/bio.h>
+#include "ext4_jbd2.h"
 #include "xattr.h"
 #include "acl.h"
 
@@ -93,7 +93,7 @@ int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode,
 	BUFFER_TRACE(bh, "call ext4_journal_revoke");
 	err = ext4_journal_revoke(handle, blocknr, bh);
 	if (err)
-		ext4_abort(inode->i_sb, __FUNCTION__,
+		ext4_abort(inode->i_sb, __func__,
 			   "error %d when attempting revoke", err);
 	BUFFER_TRACE(bh, "exit");
 	return err;
@@ -985,6 +985,16 @@ int ext4_get_blocks_wrap(handle_t *handle, struct inode *inode, sector_t block,
 	} else {
 		retval = ext4_get_blocks_handle(handle, inode, block,
 				max_blocks, bh, create, extend_disksize);
+
+		if (retval > 0 && buffer_new(bh)) {
+			/*
+			 * We allocated new blocks which will result in
+			 * i_data's format changing.  Force the migrate
+			 * to fail by clearing migrate flags
+			 */
+			EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags &
+							~EXT4_EXT_MIGRATE;
+		}
 	}
 	up_write((&EXT4_I(inode)->i_data_sem));
 	return retval;
@@ -1230,7 +1240,7 @@ int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh)
 {
 	int err = jbd2_journal_dirty_data(handle, bh);
 	if (err)
-		ext4_journal_abort_handle(__FUNCTION__, __FUNCTION__,
+		ext4_journal_abort_handle(__func__, __func__,
 						bh, handle, err);
 	return err;
 }
@@ -1301,10 +1311,11 @@ static int ext4_ordered_write_end(struct file *file,
 		new_i_size = pos + copied;
 		if (new_i_size > EXT4_I(inode)->i_disksize)
 			EXT4_I(inode)->i_disksize = new_i_size;
-		copied = ext4_generic_write_end(file, mapping, pos, len, copied,
+		ret2 = ext4_generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
-		if (copied < 0)
-			ret = copied;
+		copied = ret2;
+		if (ret2 < 0)
+			ret = ret2;
 	}
 	ret2 = ext4_journal_stop(handle);
 	if (!ret)
@@ -1329,10 +1340,11 @@ static int ext4_writeback_write_end(struct file *file,
 	if (new_i_size > EXT4_I(inode)->i_disksize)
 		EXT4_I(inode)->i_disksize = new_i_size;
 
-	copied = ext4_generic_write_end(file, mapping, pos, len, copied,
+	ret2 = ext4_generic_write_end(file, mapping, pos, len, copied,
 							page, fsdata);
-	if (copied < 0)
-		ret = copied;
+	copied = ret2;
+	if (ret2 < 0)
+		ret = ret2;
 
 	ret2 = ext4_journal_stop(handle);
 	if (!ret)
@@ -2501,12 +2513,10 @@ out_stop:
 static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
 		unsigned long ino, struct ext4_iloc *iloc)
 {
-	unsigned long desc, group_desc;
 	ext4_group_t block_group;
 	unsigned long offset;
 	ext4_fsblk_t block;
-	struct buffer_head *bh;
-	struct ext4_group_desc * gdp;
+	struct ext4_group_desc *gdp;
 
 	if (!ext4_valid_inum(sb, ino)) {
 		/*
@@ -2518,22 +2528,10 @@ static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb,
 	}
 
 	block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb);
-	if (block_group >= EXT4_SB(sb)->s_groups_count) {
-		ext4_error(sb,"ext4_get_inode_block","group >= groups count");
+	gdp = ext4_get_group_desc(sb, block_group, NULL);
+	if (!gdp)
 		return 0;
-	}
-	smp_rmb();
-	group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb);
-	desc = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1);
-	bh = EXT4_SB(sb)->s_group_desc[group_desc];
-	if (!bh) {
-		ext4_error (sb, "ext4_get_inode_block",
-			    "Descriptor not loaded");
-		return 0;
-	}
 
-	gdp = (struct ext4_group_desc *)((__u8 *)bh->b_data +
-		desc * EXT4_DESC_SIZE(sb));
 	/*
 	 * Figure out the offset within the block group inode table
 	 */
@@ -2976,7 +2974,8 @@ static int ext4_do_update_inode(handle_t *handle,
 	if (ext4_inode_blocks_set(handle, raw_inode, ei))
 		goto out_brelse;
 	raw_inode->i_dtime = cpu_to_le32(ei->i_dtime);
-	raw_inode->i_flags = cpu_to_le32(ei->i_flags);
+	/* clear the migrate flag in the raw_inode */
+	raw_inode->i_flags = cpu_to_le32(ei->i_flags & ~EXT4_EXT_MIGRATE);
 	if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
 	    cpu_to_le32(EXT4_OS_HURD))
 		raw_inode->i_file_acl_high =
@@ -3374,7 +3373,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode)
 				EXT4_I(inode)->i_state |= EXT4_STATE_NO_EXPAND;
 				if (mnt_count !=
 					le16_to_cpu(sbi->s_es->s_mnt_count)) {
-					ext4_warning(inode->i_sb, __FUNCTION__,
+					ext4_warning(inode->i_sb, __func__,
 					"Unable to expand inode %lu. Delete"
 					" some EAs or run e2fsck.",
 					inode->i_ino);
@@ -3415,7 +3414,7 @@ void ext4_dirty_inode(struct inode *inode)
 		current_handle->h_transaction != handle->h_transaction) {
 		/* This task has a transaction open against a different fs */
 		printk(KERN_EMERG "%s: transactions do not match!\n",
-		       __FUNCTION__);
+		       __func__);
 	} else {
 		jbd_debug(5, "marking dirty.  outer handle=%p\n",
 				current_handle);

fs/ext4/ioctl.c

@@ -10,17 +10,17 @@
 #include <linux/fs.h>
 #include <linux/jbd2.h>
 #include <linux/capability.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/time.h>
 #include <linux/compat.h>
 #include <linux/smp_lock.h>
 #include <linux/mount.h>
 #include <asm/uaccess.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 
-int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd,
-		unsigned long arg)
+long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
+	struct inode *inode = filp->f_dentry->d_inode;
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	unsigned int flags;
 	unsigned short rsv_window_size;
@@ -277,9 +277,6 @@ setversion_out:
 #ifdef CONFIG_COMPAT
 long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
-	struct inode *inode = file->f_path.dentry->d_inode;
-	int ret;
-
 	/* These are just misnamed, they actually get/put from/to user an int */
 	switch (cmd) {
 	case EXT4_IOC32_GETFLAGS:
@@ -319,9 +316,6 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	default:
 		return -ENOIOCTLCMD;
 	}
-	lock_kernel();
-	ret = ext4_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg));
-	unlock_kernel();
-	return ret;
+	return ext4_ioctl(file, cmd, (unsigned long) compat_ptr(arg));
 }
 #endif

fs/ext4/mballoc.c

@@ -21,21 +21,7 @@
  * mballoc.c contains the multiblocks allocation routines
  */
 
-#include <linux/time.h>
-#include <linux/fs.h>
-#include <linux/namei.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/quotaops.h>
-#include <linux/buffer_head.h>
-#include <linux/module.h>
-#include <linux/swap.h>
-#include <linux/proc_fs.h>
-#include <linux/pagemap.h>
-#include <linux/seq_file.h>
-#include <linux/version.h>
-#include "group.h"
-
+#include "mballoc.h"
 /*
  * MUSTDO:
  *   - test ext4_ext_search_left() and ext4_ext_search_right()
@@ -345,288 +331,6 @@
  *
  */
 
-/*
- * with AGGRESSIVE_CHECK allocator runs consistency checks over
- * structures. these checks slow things down a lot
- */
-#define AGGRESSIVE_CHECK__
-
-/*
- * with DOUBLE_CHECK defined mballoc creates persistent in-core
- * bitmaps, maintains and uses them to check for double allocations
- */
-#define DOUBLE_CHECK__
-
-/*
- */
-#define MB_DEBUG__
-#ifdef MB_DEBUG
-#define mb_debug(fmt, a...)	printk(fmt, ##a)
-#else
-#define mb_debug(fmt, a...)
-#endif
-
-/*
- * with EXT4_MB_HISTORY mballoc stores last N allocations in memory
- * and you can monitor it in /proc/fs/ext4/<dev>/mb_history
- */
-#define EXT4_MB_HISTORY
-#define EXT4_MB_HISTORY_ALLOC		1	/* allocation */
-#define EXT4_MB_HISTORY_PREALLOC	2	/* preallocated blocks used */
-#define EXT4_MB_HISTORY_DISCARD		4	/* preallocation discarded */
-#define EXT4_MB_HISTORY_FREE		8	/* free */
-
-#define EXT4_MB_HISTORY_DEFAULT		(EXT4_MB_HISTORY_ALLOC | \
-					 EXT4_MB_HISTORY_PREALLOC)
-
-/*
- * How long mballoc can look for a best extent (in found extents)
- */
-#define MB_DEFAULT_MAX_TO_SCAN		200
-
-/*
- * How long mballoc must look for a best extent
- */
-#define MB_DEFAULT_MIN_TO_SCAN		10
-
-/*
- * How many groups mballoc will scan looking for the best chunk
- */
-#define MB_DEFAULT_MAX_GROUPS_TO_SCAN	5
-
-/*
- * with 'ext4_mb_stats' allocator will collect stats that will be
- * shown at umount. The collecting costs though!
- */
-#define MB_DEFAULT_STATS		1
-
-/*
- * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served
- * by the stream allocator, which purpose is to pack requests
- * as close each to other as possible to produce smooth I/O traffic
- * We use locality group prealloc space for stream request.
- * We can tune the same via /proc/fs/ext4/<parition>/stream_req
- */
-#define MB_DEFAULT_STREAM_THRESHOLD	16	/* 64K */
-
-/*
- * for which requests use 2^N search using buddies
- */
-#define MB_DEFAULT_ORDER2_REQS		2
-
-/*
- * default group prealloc size 512 blocks
- */
-#define MB_DEFAULT_GROUP_PREALLOC	512
-
-static struct kmem_cache *ext4_pspace_cachep;
-static struct kmem_cache *ext4_ac_cachep;
-
-#ifdef EXT4_BB_MAX_BLOCKS
-#undef EXT4_BB_MAX_BLOCKS
-#endif
-#define EXT4_BB_MAX_BLOCKS	30
-
-struct ext4_free_metadata {
-	ext4_group_t group;
-	unsigned short num;
-	ext4_grpblk_t  blocks[EXT4_BB_MAX_BLOCKS];
-	struct list_head list;
-};
-
-struct ext4_group_info {
-	unsigned long	bb_state;
-	unsigned long	bb_tid;
-	struct ext4_free_metadata *bb_md_cur;
-	unsigned short	bb_first_free;
-	unsigned short	bb_free;
-	unsigned short	bb_fragments;
-	struct		list_head bb_prealloc_list;
-#ifdef DOUBLE_CHECK
-	void		*bb_bitmap;
-#endif
-	unsigned short	bb_counters[];
-};
-
-#define EXT4_GROUP_INFO_NEED_INIT_BIT	0
-#define EXT4_GROUP_INFO_LOCKED_BIT	1
-
-#define EXT4_MB_GRP_NEED_INIT(grp)	\
-	(test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state)))
-
-
-struct ext4_prealloc_space {
-	struct list_head	pa_inode_list;
-	struct list_head	pa_group_list;
-	union {
-		struct list_head pa_tmp_list;
-		struct rcu_head	pa_rcu;
-	} u;
-	spinlock_t		pa_lock;
-	atomic_t		pa_count;
-	unsigned		pa_deleted;
-	ext4_fsblk_t		pa_pstart;	/* phys. block */
-	ext4_lblk_t		pa_lstart;	/* log. block */
-	unsigned short		pa_len;		/* len of preallocated chunk */
-	unsigned short		pa_free;	/* how many blocks are free */
-	unsigned short		pa_linear;	/* consumed in one direction
-						 * strictly, for grp prealloc */
-	spinlock_t		*pa_obj_lock;
-	struct inode		*pa_inode;	/* hack, for history only */
-};
-
-
-struct ext4_free_extent {
-	ext4_lblk_t fe_logical;
-	ext4_grpblk_t fe_start;
-	ext4_group_t fe_group;
-	int fe_len;
-};
-
-/*
- * Locality group:
- *   we try to group all related changes together
- *   so that writeback can flush/allocate them together as well
- */
-struct ext4_locality_group {
-	/* for allocator */
-	struct mutex		lg_mutex;	/* to serialize allocates */
-	struct list_head	lg_prealloc_list;/* list of preallocations */
-	spinlock_t		lg_prealloc_lock;
-};
-
-struct ext4_allocation_context {
-	struct inode *ac_inode;
-	struct super_block *ac_sb;
-
-	/* original request */
-	struct ext4_free_extent ac_o_ex;
-
-	/* goal request (after normalization) */
-	struct ext4_free_extent ac_g_ex;
-
-	/* the best found extent */
-	struct ext4_free_extent ac_b_ex;
-
-	/* copy of the bext found extent taken before preallocation efforts */
-	struct ext4_free_extent ac_f_ex;
-
-	/* number of iterations done. we have to track to limit searching */
-	unsigned long ac_ex_scanned;
-	__u16 ac_groups_scanned;
-	__u16 ac_found;
-	__u16 ac_tail;
-	__u16 ac_buddy;
-	__u16 ac_flags;		/* allocation hints */
-	__u8 ac_status;
-	__u8 ac_criteria;
-	__u8 ac_repeats;
-	__u8 ac_2order;		/* if request is to allocate 2^N blocks and
-				 * N > 0, the field stores N, otherwise 0 */
-	__u8 ac_op;		/* operation, for history only */
-	struct page *ac_bitmap_page;
-	struct page *ac_buddy_page;
-	struct ext4_prealloc_space *ac_pa;
-	struct ext4_locality_group *ac_lg;
-};
-
-#define AC_STATUS_CONTINUE	1
-#define AC_STATUS_FOUND		2
-#define AC_STATUS_BREAK		3
-
-struct ext4_mb_history {
-	struct ext4_free_extent orig;	/* orig allocation */
-	struct ext4_free_extent goal;	/* goal allocation */
-	struct ext4_free_extent result;	/* result allocation */
-	unsigned pid;
-	unsigned ino;
-	__u16 found;	/* how many extents have been found */
-	__u16 groups;	/* how many groups have been scanned */
-	__u16 tail;	/* what tail broke some buddy */
-	__u16 buddy;	/* buddy the tail ^^^ broke */
-	__u16 flags;
-	__u8 cr:3;	/* which phase the result extent was found at */
-	__u8 op:4;
-	__u8 merged:1;
-};
-
-struct ext4_buddy {
-	struct page *bd_buddy_page;
-	void *bd_buddy;
-	struct page *bd_bitmap_page;
-	void *bd_bitmap;
-	struct ext4_group_info *bd_info;
-	struct super_block *bd_sb;
-	__u16 bd_blkbits;
-	ext4_group_t bd_group;
-};
-#define EXT4_MB_BITMAP(e4b)	((e4b)->bd_bitmap)
-#define EXT4_MB_BUDDY(e4b)	((e4b)->bd_buddy)
-
-#ifndef EXT4_MB_HISTORY
-static inline void ext4_mb_store_history(struct ext4_allocation_context *ac)
-{
-	return;
-}
-#else
-static void ext4_mb_store_history(struct ext4_allocation_context *ac);
-#endif
-
-#define in_range(b, first, len)	((b) >= (first) && (b) <= (first) + (len) - 1)
-
-static struct proc_dir_entry *proc_root_ext4;
-struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
-ext4_fsblk_t ext4_new_blocks_old(handle_t *handle, struct inode *inode,
-			ext4_fsblk_t goal, unsigned long *count, int *errp);
-
-static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
-					ext4_group_t group);
-static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *);
-static void ext4_mb_free_committed_blocks(struct super_block *);
-static void ext4_mb_return_to_preallocation(struct inode *inode,
-					struct ext4_buddy *e4b, sector_t block,
-					int count);
-static void ext4_mb_put_pa(struct ext4_allocation_context *,
-			struct super_block *, struct ext4_prealloc_space *pa);
-static int ext4_mb_init_per_dev_proc(struct super_block *sb);
-static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
-
-
-static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group)
-{
-	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
-	bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
-}
-
-static inline void ext4_unlock_group(struct super_block *sb,
-					ext4_group_t group)
-{
-	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
-	bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
-}
-
-static inline int ext4_is_group_locked(struct super_block *sb,
-					ext4_group_t group)
-{
-	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
-
-	return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT,
-						&(grinfo->bb_state));
-}
-
-static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb,
-					struct ext4_free_extent *fex)
-{
-	ext4_fsblk_t block;
-
-	block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb)
-			+ fex->fe_start
-			+ le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
-	return block;
-}
-
 static inline void *mb_correct_addr_and_bit(int *bit, void *addr)
 {
 #if BITS_PER_LONG == 64
@@ -736,7 +440,7 @@ static void mb_free_blocks_double(struct inode *inode, struct ext4_buddy *e4b,
 			blocknr +=
 			    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
 
-			ext4_error(sb, __FUNCTION__, "double-free of inode"
+			ext4_error(sb, __func__, "double-free of inode"
 				   " %lu's block %llu(bit %u in group %lu)\n",
 				   inode ? inode->i_ino : 0, blocknr,
 				   first + i, e4b->bd_group);
@@ -898,17 +602,17 @@ static int __mb_check_buddy(struct ext4_buddy *e4b, char *file,
 	list_for_each(cur, &grp->bb_prealloc_list) {
 		ext4_group_t groupnr;
 		struct ext4_prealloc_space *pa;
-		pa = list_entry(cur, struct ext4_prealloc_space, group_list);
-		ext4_get_group_no_and_offset(sb, pa->pstart, &groupnr, &k);
+		pa = list_entry(cur, struct ext4_prealloc_space, pa_group_list);
+		ext4_get_group_no_and_offset(sb, pa->pa_pstart, &groupnr, &k);
 		MB_CHECK_ASSERT(groupnr == e4b->bd_group);
-		for (i = 0; i < pa->len; i++)
+		for (i = 0; i < pa->pa_len; i++)
 			MB_CHECK_ASSERT(mb_test_bit(k + i, buddy));
 	}
 	return 0;
 }
 #undef MB_CHECK_ASSERT
 #define mb_check_buddy(e4b) __mb_check_buddy(e4b,	\
-					__FILE__, __FUNCTION__, __LINE__)
+					__FILE__, __func__, __LINE__)
 #else
 #define mb_check_buddy(e4b)
 #endif
@@ -982,7 +686,7 @@ static void ext4_mb_generate_buddy(struct super_block *sb,
 	grp->bb_fragments = fragments;
 
 	if (free != grp->bb_free) {
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			"EXT4-fs: group %lu: %u blocks in bitmap, %u in gd\n",
 			group, free, grp->bb_free);
 		/*
@@ -1168,8 +872,9 @@ out:
 	return err;
 }
 
-static int ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
-		struct ext4_buddy *e4b)
+static noinline_for_stack int
+ext4_mb_load_buddy(struct super_block *sb, ext4_group_t group,
+					struct ext4_buddy *e4b)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct inode *inode = sbi->s_buddy_cache;
@@ -1367,7 +1072,7 @@ static int mb_free_blocks(struct inode *inode, struct ext4_buddy *e4b,
 			blocknr +=
 			    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
 
-			ext4_error(sb, __FUNCTION__, "double-free of inode"
+			ext4_error(sb, __func__, "double-free of inode"
 				   " %lu's block %llu(bit %u in group %lu)\n",
 				   inode ? inode->i_ino : 0, blocknr, block,
 				   e4b->bd_group);
@@ -1848,7 +1553,7 @@ static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 			 * free blocks even though group info says we
 			 * we have free blocks
 			 */
-			ext4_error(sb, __FUNCTION__, "%d free blocks as per "
+			ext4_error(sb, __func__, "%d free blocks as per "
 					"group info. But bitmap says 0\n",
 					free);
 			break;
@@ -1857,7 +1562,7 @@ static void ext4_mb_complex_scan_group(struct ext4_allocation_context *ac,
 		mb_find_extent(e4b, 0, i, ac->ac_g_ex.fe_len, &ex);
 		BUG_ON(ex.fe_len <= 0);
 		if (free < ex.fe_len) {
-			ext4_error(sb, __FUNCTION__, "%d free blocks as per "
+			ext4_error(sb, __func__, "%d free blocks as per "
 					"group info. But got %d blocks\n",
 					free, ex.fe_len);
 			/*
@@ -1965,7 +1670,8 @@ static int ext4_mb_good_group(struct ext4_allocation_context *ac,
 	return 0;
 }
 
-static int ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
+static noinline_for_stack int
+ext4_mb_regular_allocator(struct ext4_allocation_context *ac)
 {
 	ext4_group_t group;
 	ext4_group_t i;
@@ -2465,7 +2171,8 @@ static void ext4_mb_history_init(struct super_block *sb)
 	/* if we can't allocate history, then we simple won't use it */
 }
 
-static void ext4_mb_store_history(struct ext4_allocation_context *ac)
+static noinline_for_stack void
+ext4_mb_store_history(struct ext4_allocation_context *ac)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
 	struct ext4_mb_history h;
@@ -2565,13 +2272,13 @@ static int ext4_mb_init_backend(struct super_block *sb)
 		meta_group_info[j] = kzalloc(len, GFP_KERNEL);
 		if (meta_group_info[j] == NULL) {
 			printk(KERN_ERR "EXT4-fs: can't allocate buddy mem\n");
-			i--;
 			goto err_freebuddy;
 		}
 		desc = ext4_get_group_desc(sb, i, NULL);
 		if (desc == NULL) {
 			printk(KERN_ERR
 				"EXT4-fs: can't read descriptor %lu\n", i);
+			i++;
 			goto err_freebuddy;
 		}
 		memset(meta_group_info[j], 0, len);
@@ -2611,13 +2318,11 @@ static int ext4_mb_init_backend(struct super_block *sb)
 	return 0;
 
 err_freebuddy:
-	while (i >= 0) {
+	while (i-- > 0)
 		kfree(ext4_get_group_info(sb, i));
-		i--;
-	}
 	i = num_meta_group_infos;
 err_freemeta:
-	while (--i >= 0)
+	while (i-- > 0)
 		kfree(sbi->s_group_info[i]);
 	iput(sbi->s_buddy_cache);
 err_freesgi:
@@ -2801,7 +2506,8 @@ int ext4_mb_release(struct super_block *sb)
 	return 0;
 }
 
-static void ext4_mb_free_committed_blocks(struct super_block *sb)
+static noinline_for_stack void
+ext4_mb_free_committed_blocks(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	int err;
@@ -3021,7 +2727,8 @@ void exit_ext4_mballoc(void)
  * Check quota and mark choosed space (ac->ac_b_ex) non-free in bitmaps
  * Returns 0 if success or error code
  */
-static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
+static noinline_for_stack int
+ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 				handle_t *handle)
 {
 	struct buffer_head *bitmap_bh = NULL;
@@ -3070,7 +2777,7 @@ static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 			in_range(block, ext4_inode_table(sb, gdp),
 				EXT4_SB(sb)->s_itb_per_group)) {
 
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			   "Allocating block in system zone - block = %llu",
 			   block);
 	}
@@ -3094,9 +2801,7 @@ static int ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac,
 						ac->ac_b_ex.fe_group,
 						gdp));
 	}
-	gdp->bg_free_blocks_count =
-		cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)
-				- ac->ac_b_ex.fe_len);
+	le16_add_cpu(&gdp->bg_free_blocks_count, -ac->ac_b_ex.fe_len);
 	gdp->bg_checksum = ext4_group_desc_csum(sbi, ac->ac_b_ex.fe_group, gdp);
 	spin_unlock(sb_bgl_lock(sbi, ac->ac_b_ex.fe_group));
 	percpu_counter_sub(&sbi->s_freeblocks_counter, ac->ac_b_ex.fe_len);
@@ -3130,7 +2835,7 @@ static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac)
 		ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_stripe;
 	else
 		ac->ac_g_ex.fe_len = EXT4_SB(sb)->s_mb_group_prealloc;
-	mb_debug("#%u: goal %lu blocks for locality group\n",
+	mb_debug("#%u: goal %u blocks for locality group\n",
 		current->pid, ac->ac_g_ex.fe_len);
 }
 
@@ -3138,15 +2843,16 @@ static void ext4_mb_normalize_group_request(struct ext4_allocation_context *ac)
  * Normalization means making request better in terms of
  * size and alignment
  */
-static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
+static noinline_for_stack void
+ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 				struct ext4_allocation_request *ar)
 {
 	int bsbits, max;
 	ext4_lblk_t end;
-	struct list_head *cur;
 	loff_t size, orig_size, start_off;
 	ext4_lblk_t start, orig_start;
 	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
+	struct ext4_prealloc_space *pa;
 
 	/* do normalize only data requests, metadata requests
 	   do not need preallocation */
@@ -3232,12 +2938,9 @@ static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 
 	/* check we don't cross already preallocated blocks */
 	rcu_read_lock();
-	list_for_each_rcu(cur, &ei->i_prealloc_list) {
-		struct ext4_prealloc_space *pa;
+	list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
 		unsigned long pa_end;
 
-		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
-
 		if (pa->pa_deleted)
 			continue;
 		spin_lock(&pa->pa_lock);
@@ -3279,10 +2982,8 @@ static void ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 
 	/* XXX: extra loop to check we really don't overlap preallocations */
 	rcu_read_lock();
-	list_for_each_rcu(cur, &ei->i_prealloc_list) {
-		struct ext4_prealloc_space *pa;
+	list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
 		unsigned long pa_end;
-		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
 		spin_lock(&pa->pa_lock);
 		if (pa->pa_deleted == 0) {
 			pa_end = pa->pa_lstart + pa->pa_len;
@@ -3374,7 +3075,7 @@ static void ext4_mb_use_inode_pa(struct ext4_allocation_context *ac,
 	BUG_ON(pa->pa_free < len);
 	pa->pa_free -= len;
 
-	mb_debug("use %llu/%lu from inode pa %p\n", start, len, pa);
+	mb_debug("use %llu/%u from inode pa %p\n", start, len, pa);
 }
 
 /*
@@ -3404,12 +3105,12 @@ static void ext4_mb_use_group_pa(struct ext4_allocation_context *ac,
 /*
  * search goal blocks in preallocated space
  */
-static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
+static noinline_for_stack int
+ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
 {
 	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
 	struct ext4_locality_group *lg;
 	struct ext4_prealloc_space *pa;
-	struct list_head *cur;
 
 	/* only data can be preallocated */
 	if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
@@ -3417,8 +3118,7 @@ static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
 
 	/* first, try per-file preallocation */
 	rcu_read_lock();
-	list_for_each_rcu(cur, &ei->i_prealloc_list) {
-		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+	list_for_each_entry_rcu(pa, &ei->i_prealloc_list, pa_inode_list) {
 
 		/* all fields in this condition don't change,
 		 * so we can skip locking for them */
@@ -3450,8 +3150,7 @@ static int ext4_mb_use_preallocated(struct ext4_allocation_context *ac)
 		return 0;
 
 	rcu_read_lock();
-	list_for_each_rcu(cur, &lg->lg_prealloc_list) {
-		pa = list_entry(cur, struct ext4_prealloc_space, pa_inode_list);
+	list_for_each_entry_rcu(pa, &lg->lg_prealloc_list, pa_inode_list) {
 		spin_lock(&pa->pa_lock);
 		if (pa->pa_deleted == 0 && pa->pa_free >= ac->ac_o_ex.fe_len) {
 			atomic_inc(&pa->pa_count);
@@ -3571,7 +3270,8 @@ static void ext4_mb_put_pa(struct ext4_allocation_context *ac,
 /*
  * creates new preallocated space for given inode
  */
-static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
+static noinline_for_stack int
+ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
 {
 	struct super_block *sb = ac->ac_sb;
 	struct ext4_prealloc_space *pa;
@@ -3658,7 +3358,8 @@ static int ext4_mb_new_inode_pa(struct ext4_allocation_context *ac)
 /*
  * creates new preallocated space for locality group inodes belongs to
  */
-static int ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
+static noinline_for_stack int
+ext4_mb_new_group_pa(struct ext4_allocation_context *ac)
 {
 	struct super_block *sb = ac->ac_sb;
 	struct ext4_locality_group *lg;
@@ -3731,11 +3432,11 @@ static int ext4_mb_new_preallocation(struct ext4_allocation_context *ac)
  * the caller MUST hold group/inode locks.
  * TODO: optimize the case when there are no in-core structures yet
  */
-static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
-				struct buffer_head *bitmap_bh,
-				struct ext4_prealloc_space *pa)
+static noinline_for_stack int
+ext4_mb_release_inode_pa(struct ext4_buddy *e4b, struct buffer_head *bitmap_bh,
+			struct ext4_prealloc_space *pa,
+			struct ext4_allocation_context *ac)
 {
-	struct ext4_allocation_context *ac;
 	struct super_block *sb = e4b->bd_sb;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	unsigned long end;
@@ -3751,8 +3452,6 @@ static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
 	BUG_ON(group != e4b->bd_group && pa->pa_len != 0);
 	end = bit + pa->pa_len;
 
-	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
-
 	if (ac) {
 		ac->ac_sb = sb;
 		ac->ac_inode = pa->pa_inode;
@@ -3789,7 +3488,7 @@ static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
 			pa, (unsigned long) pa->pa_lstart,
 			(unsigned long) pa->pa_pstart,
 			(unsigned long) pa->pa_len);
-		ext4_error(sb, __FUNCTION__, "free %u, pa_free %u\n",
+		ext4_error(sb, __func__, "free %u, pa_free %u\n",
 						free, pa->pa_free);
 		/*
 		 * pa is already deleted so we use the value obtained
@@ -3797,22 +3496,19 @@ static int ext4_mb_release_inode_pa(struct ext4_buddy *e4b,
 		 */
 	}
 	atomic_add(free, &sbi->s_mb_discarded);
-	if (ac)
-		kmem_cache_free(ext4_ac_cachep, ac);
 
 	return err;
 }
 
-static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
-				struct ext4_prealloc_space *pa)
+static noinline_for_stack int
+ext4_mb_release_group_pa(struct ext4_buddy *e4b,
+				struct ext4_prealloc_space *pa,
+				struct ext4_allocation_context *ac)
 {
-	struct ext4_allocation_context *ac;
 	struct super_block *sb = e4b->bd_sb;
 	ext4_group_t group;
 	ext4_grpblk_t bit;
 
-	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
-
 	if (ac)
 		ac->ac_op = EXT4_MB_HISTORY_DISCARD;
 
@@ -3830,7 +3526,6 @@ static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
 		ac->ac_b_ex.fe_len = pa->pa_len;
 		ac->ac_b_ex.fe_logical = 0;
 		ext4_mb_store_history(ac);
-		kmem_cache_free(ext4_ac_cachep, ac);
 	}
 
 	return 0;
@@ -3845,12 +3540,14 @@ static int ext4_mb_release_group_pa(struct ext4_buddy *e4b,
  * - how many do we discard
  *   1) how many requested
  */
-static int ext4_mb_discard_group_preallocations(struct super_block *sb,
+static noinline_for_stack int
+ext4_mb_discard_group_preallocations(struct super_block *sb,
 					ext4_group_t group, int needed)
 {
 	struct ext4_group_info *grp = ext4_get_group_info(sb, group);
 	struct buffer_head *bitmap_bh = NULL;
 	struct ext4_prealloc_space *pa, *tmp;
+	struct ext4_allocation_context *ac;
 	struct list_head list;
 	struct ext4_buddy e4b;
 	int err;
@@ -3878,6 +3575,7 @@ static int ext4_mb_discard_group_preallocations(struct super_block *sb,
 	grp = ext4_get_group_info(sb, group);
 	INIT_LIST_HEAD(&list);
 
+	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
 repeat:
 	ext4_lock_group(sb, group);
 	list_for_each_entry_safe(pa, tmp,
@@ -3932,9 +3630,9 @@ repeat:
 		spin_unlock(pa->pa_obj_lock);
 
 		if (pa->pa_linear)
-			ext4_mb_release_group_pa(&e4b, pa);
+			ext4_mb_release_group_pa(&e4b, pa, ac);
 		else
-			ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa);
+			ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa, ac);
 
 		list_del(&pa->u.pa_tmp_list);
 		call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
@@ -3942,6 +3640,8 @@ repeat:
 
 out:
 	ext4_unlock_group(sb, group);
+	if (ac)
+		kmem_cache_free(ext4_ac_cachep, ac);
 	ext4_mb_release_desc(&e4b);
 	put_bh(bitmap_bh);
 	return free;
@@ -3962,6 +3662,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode)
 	struct super_block *sb = inode->i_sb;
 	struct buffer_head *bitmap_bh = NULL;
 	struct ext4_prealloc_space *pa, *tmp;
+	struct ext4_allocation_context *ac;
 	ext4_group_t group = 0;
 	struct list_head list;
 	struct ext4_buddy e4b;
@@ -3976,6 +3677,7 @@ void ext4_mb_discard_inode_preallocations(struct inode *inode)
 
 	INIT_LIST_HEAD(&list);
 
+	ac = kmem_cache_alloc(ext4_ac_cachep, GFP_NOFS);
 repeat:
 	/* first, collect all pa's in the inode */
 	spin_lock(&ei->i_prealloc_lock);
@@ -4040,7 +3742,7 @@ repeat:
 
 		ext4_lock_group(sb, group);
 		list_del(&pa->pa_group_list);
-		ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa);
+		ext4_mb_release_inode_pa(&e4b, bitmap_bh, pa, ac);
 		ext4_unlock_group(sb, group);
 
 		ext4_mb_release_desc(&e4b);
@@ -4049,6 +3751,8 @@ repeat:
 		list_del(&pa->u.pa_tmp_list);
 		call_rcu(&(pa)->u.pa_rcu, ext4_mb_pa_callback);
 	}
+	if (ac)
+		kmem_cache_free(ext4_ac_cachep, ac);
 }
 
 /*
@@ -4108,7 +3812,7 @@ static void ext4_mb_show_ac(struct ext4_allocation_context *ac)
 			printk(KERN_ERR "PA:%lu:%d:%u \n", i,
 							start, pa->pa_len);
 		}
-		ext4_lock_group(sb, i);
+		ext4_unlock_group(sb, i);
 
 		if (grp->bb_free == 0)
 			continue;
@@ -4167,7 +3871,8 @@ static void ext4_mb_group_or_file(struct ext4_allocation_context *ac)
 	mutex_lock(&ac->ac_lg->lg_mutex);
 }
 
-static int ext4_mb_initialize_context(struct ext4_allocation_context *ac,
+static noinline_for_stack int
+ext4_mb_initialize_context(struct ext4_allocation_context *ac,
 				struct ext4_allocation_request *ar)
 {
 	struct super_block *sb = ar->inode->i_sb;
@@ -4398,7 +4103,8 @@ static void ext4_mb_poll_new_transaction(struct super_block *sb,
 	ext4_mb_free_committed_blocks(sb);
 }
 
-static int ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
+static noinline_for_stack int
+ext4_mb_free_metadata(handle_t *handle, struct ext4_buddy *e4b,
 			  ext4_group_t group, ext4_grpblk_t block, int count)
 {
 	struct ext4_group_info *db = e4b->bd_info;
@@ -4489,7 +4195,7 @@ void ext4_mb_free_blocks(handle_t *handle, struct inode *inode,
 	if (block < le32_to_cpu(es->s_first_data_block) ||
 	    block + count < block ||
 	    block + count > ext4_blocks_count(es)) {
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			    "Freeing blocks not in datazone - "
 			    "block = %lu, count = %lu", block, count);
 		goto error_return;
@@ -4530,7 +4236,7 @@ do_more:
 	    in_range(block + count - 1, ext4_inode_table(sb, gdp),
 		      EXT4_SB(sb)->s_itb_per_group)) {
 
-		ext4_error(sb, __FUNCTION__,
+		ext4_error(sb, __func__,
 			   "Freeing blocks in system zone - "
 			   "Block = %lu, count = %lu", block, count);
 	}
@@ -4588,8 +4294,7 @@ do_more:
 	}
 
 	spin_lock(sb_bgl_lock(sbi, block_group));
-	gdp->bg_free_blocks_count =
-		cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) + count);
+	le16_add_cpu(&gdp->bg_free_blocks_count, count);
 	gdp->bg_checksum = ext4_group_desc_csum(sbi, block_group, gdp);
 	spin_unlock(sb_bgl_lock(sbi, block_group));
 	percpu_counter_add(&sbi->s_freeblocks_counter, count);

fs/ext4/mballoc.h

@@ -0,0 +1,304 @@
+/*
+ *  fs/ext4/mballoc.h
+ *
+ *  Written by: Alex Tomas <alex@clusterfs.com>
+ *
+ */
+#ifndef _EXT4_MBALLOC_H
+#define _EXT4_MBALLOC_H
+
+#include <linux/time.h>
+#include <linux/fs.h>
+#include <linux/namei.h>
+#include <linux/quotaops.h>
+#include <linux/buffer_head.h>
+#include <linux/module.h>
+#include <linux/swap.h>
+#include <linux/proc_fs.h>
+#include <linux/pagemap.h>
+#include <linux/seq_file.h>
+#include <linux/version.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
+#include "group.h"
+
+/*
+ * with AGGRESSIVE_CHECK allocator runs consistency checks over
+ * structures. these checks slow things down a lot
+ */
+#define AGGRESSIVE_CHECK__
+
+/*
+ * with DOUBLE_CHECK defined mballoc creates persistent in-core
+ * bitmaps, maintains and uses them to check for double allocations
+ */
+#define DOUBLE_CHECK__
+
+/*
+ */
+#define MB_DEBUG__
+#ifdef MB_DEBUG
+#define mb_debug(fmt, a...)	printk(fmt, ##a)
+#else
+#define mb_debug(fmt, a...)
+#endif
+
+/*
+ * with EXT4_MB_HISTORY mballoc stores last N allocations in memory
+ * and you can monitor it in /proc/fs/ext4/<dev>/mb_history
+ */
+#define EXT4_MB_HISTORY
+#define EXT4_MB_HISTORY_ALLOC		1	/* allocation */
+#define EXT4_MB_HISTORY_PREALLOC	2	/* preallocated blocks used */
+#define EXT4_MB_HISTORY_DISCARD		4	/* preallocation discarded */
+#define EXT4_MB_HISTORY_FREE		8	/* free */
+
+#define EXT4_MB_HISTORY_DEFAULT		(EXT4_MB_HISTORY_ALLOC | \
+					 EXT4_MB_HISTORY_PREALLOC)
+
+/*
+ * How long mballoc can look for a best extent (in found extents)
+ */
+#define MB_DEFAULT_MAX_TO_SCAN		200
+
+/*
+ * How long mballoc must look for a best extent
+ */
+#define MB_DEFAULT_MIN_TO_SCAN		10
+
+/*
+ * How many groups mballoc will scan looking for the best chunk
+ */
+#define MB_DEFAULT_MAX_GROUPS_TO_SCAN	5
+
+/*
+ * with 'ext4_mb_stats' allocator will collect stats that will be
+ * shown at umount. The collecting costs though!
+ */
+#define MB_DEFAULT_STATS		1
+
+/*
+ * files smaller than MB_DEFAULT_STREAM_THRESHOLD are served
+ * by the stream allocator, which purpose is to pack requests
+ * as close each to other as possible to produce smooth I/O traffic
+ * We use locality group prealloc space for stream request.
+ * We can tune the same via /proc/fs/ext4/<parition>/stream_req
+ */
+#define MB_DEFAULT_STREAM_THRESHOLD	16	/* 64K */
+
+/*
+ * for which requests use 2^N search using buddies
+ */
+#define MB_DEFAULT_ORDER2_REQS		2
+
+/*
+ * default group prealloc size 512 blocks
+ */
+#define MB_DEFAULT_GROUP_PREALLOC	512
+
+static struct kmem_cache *ext4_pspace_cachep;
+static struct kmem_cache *ext4_ac_cachep;
+
+#ifdef EXT4_BB_MAX_BLOCKS
+#undef EXT4_BB_MAX_BLOCKS
+#endif
+#define EXT4_BB_MAX_BLOCKS	30
+
+struct ext4_free_metadata {
+	ext4_group_t group;
+	unsigned short num;
+	ext4_grpblk_t  blocks[EXT4_BB_MAX_BLOCKS];
+	struct list_head list;
+};
+
+struct ext4_group_info {
+	unsigned long	bb_state;
+	unsigned long	bb_tid;
+	struct ext4_free_metadata *bb_md_cur;
+	unsigned short	bb_first_free;
+	unsigned short	bb_free;
+	unsigned short	bb_fragments;
+	struct		list_head bb_prealloc_list;
+#ifdef DOUBLE_CHECK
+	void		*bb_bitmap;
+#endif
+	unsigned short	bb_counters[];
+};
+
+#define EXT4_GROUP_INFO_NEED_INIT_BIT	0
+#define EXT4_GROUP_INFO_LOCKED_BIT	1
+
+#define EXT4_MB_GRP_NEED_INIT(grp)	\
+	(test_bit(EXT4_GROUP_INFO_NEED_INIT_BIT, &((grp)->bb_state)))
+
+
+struct ext4_prealloc_space {
+	struct list_head	pa_inode_list;
+	struct list_head	pa_group_list;
+	union {
+		struct list_head pa_tmp_list;
+		struct rcu_head	pa_rcu;
+	} u;
+	spinlock_t		pa_lock;
+	atomic_t		pa_count;
+	unsigned		pa_deleted;
+	ext4_fsblk_t		pa_pstart;	/* phys. block */
+	ext4_lblk_t		pa_lstart;	/* log. block */
+	unsigned short		pa_len;		/* len of preallocated chunk */
+	unsigned short		pa_free;	/* how many blocks are free */
+	unsigned short		pa_linear;	/* consumed in one direction
+						 * strictly, for grp prealloc */
+	spinlock_t		*pa_obj_lock;
+	struct inode		*pa_inode;	/* hack, for history only */
+};
+
+
+struct ext4_free_extent {
+	ext4_lblk_t fe_logical;
+	ext4_grpblk_t fe_start;
+	ext4_group_t fe_group;
+	int fe_len;
+};
+
+/*
+ * Locality group:
+ *   we try to group all related changes together
+ *   so that writeback can flush/allocate them together as well
+ */
+struct ext4_locality_group {
+	/* for allocator */
+	struct mutex		lg_mutex;	/* to serialize allocates */
+	struct list_head	lg_prealloc_list;/* list of preallocations */
+	spinlock_t		lg_prealloc_lock;
+};
+
+struct ext4_allocation_context {
+	struct inode *ac_inode;
+	struct super_block *ac_sb;
+
+	/* original request */
+	struct ext4_free_extent ac_o_ex;
+
+	/* goal request (after normalization) */
+	struct ext4_free_extent ac_g_ex;
+
+	/* the best found extent */
+	struct ext4_free_extent ac_b_ex;
+
+	/* copy of the bext found extent taken before preallocation efforts */
+	struct ext4_free_extent ac_f_ex;
+
+	/* number of iterations done. we have to track to limit searching */
+	unsigned long ac_ex_scanned;
+	__u16 ac_groups_scanned;
+	__u16 ac_found;
+	__u16 ac_tail;
+	__u16 ac_buddy;
+	__u16 ac_flags;		/* allocation hints */
+	__u8 ac_status;
+	__u8 ac_criteria;
+	__u8 ac_repeats;
+	__u8 ac_2order;		/* if request is to allocate 2^N blocks and
+				 * N > 0, the field stores N, otherwise 0 */
+	__u8 ac_op;		/* operation, for history only */
+	struct page *ac_bitmap_page;
+	struct page *ac_buddy_page;
+	struct ext4_prealloc_space *ac_pa;
+	struct ext4_locality_group *ac_lg;
+};
+
+#define AC_STATUS_CONTINUE	1
+#define AC_STATUS_FOUND		2
+#define AC_STATUS_BREAK		3
+
+struct ext4_mb_history {
+	struct ext4_free_extent orig;	/* orig allocation */
+	struct ext4_free_extent goal;	/* goal allocation */
+	struct ext4_free_extent result;	/* result allocation */
+	unsigned pid;
+	unsigned ino;
+	__u16 found;	/* how many extents have been found */
+	__u16 groups;	/* how many groups have been scanned */
+	__u16 tail;	/* what tail broke some buddy */
+	__u16 buddy;	/* buddy the tail ^^^ broke */
+	__u16 flags;
+	__u8 cr:3;	/* which phase the result extent was found at */
+	__u8 op:4;
+	__u8 merged:1;
+};
+
+struct ext4_buddy {
+	struct page *bd_buddy_page;
+	void *bd_buddy;
+	struct page *bd_bitmap_page;
+	void *bd_bitmap;
+	struct ext4_group_info *bd_info;
+	struct super_block *bd_sb;
+	__u16 bd_blkbits;
+	ext4_group_t bd_group;
+};
+#define EXT4_MB_BITMAP(e4b)	((e4b)->bd_bitmap)
+#define EXT4_MB_BUDDY(e4b)	((e4b)->bd_buddy)
+
+#ifndef EXT4_MB_HISTORY
+static inline void ext4_mb_store_history(struct ext4_allocation_context *ac)
+{
+	return;
+}
+#else
+static void ext4_mb_store_history(struct ext4_allocation_context *ac);
+#endif
+
+#define in_range(b, first, len)	((b) >= (first) && (b) <= (first) + (len) - 1)
+
+static struct proc_dir_entry *proc_root_ext4;
+struct buffer_head *read_block_bitmap(struct super_block *, ext4_group_t);
+
+static void ext4_mb_generate_from_pa(struct super_block *sb, void *bitmap,
+					ext4_group_t group);
+static void ext4_mb_poll_new_transaction(struct super_block *, handle_t *);
+static void ext4_mb_free_committed_blocks(struct super_block *);
+static void ext4_mb_return_to_preallocation(struct inode *inode,
+					struct ext4_buddy *e4b, sector_t block,
+					int count);
+static void ext4_mb_put_pa(struct ext4_allocation_context *,
+			struct super_block *, struct ext4_prealloc_space *pa);
+static int ext4_mb_init_per_dev_proc(struct super_block *sb);
+static int ext4_mb_destroy_per_dev_proc(struct super_block *sb);
+
+
+static inline void ext4_lock_group(struct super_block *sb, ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	bit_spin_lock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
+}
+
+static inline void ext4_unlock_group(struct super_block *sb,
+					ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	bit_spin_unlock(EXT4_GROUP_INFO_LOCKED_BIT, &(grinfo->bb_state));
+}
+
+static inline int ext4_is_group_locked(struct super_block *sb,
+					ext4_group_t group)
+{
+	struct ext4_group_info *grinfo = ext4_get_group_info(sb, group);
+
+	return bit_spin_is_locked(EXT4_GROUP_INFO_LOCKED_BIT,
+						&(grinfo->bb_state));
+}
+
+static ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb,
+					struct ext4_free_extent *fex)
+{
+	ext4_fsblk_t block;
+
+	block = (ext4_fsblk_t) fex->fe_group * EXT4_BLOCKS_PER_GROUP(sb)
+			+ fex->fe_start
+			+ le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block);
+	return block;
+}
+#endif

fs/ext4/migrate.c

@@ -13,8 +13,8 @@
  */
 
 #include <linux/module.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs_extents.h>
+#include "ext4_jbd2.h"
+#include "ext4_extents.h"
 
 /*
  * The contiguous blocks details which can be
@@ -327,7 +327,7 @@ static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data)
 }
 
 static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
-				struct inode *tmp_inode)
+						struct inode *tmp_inode)
 {
 	int retval;
 	__le32	i_data[3];
@@ -339,7 +339,7 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 	 * i_data field of the original inode
 	 */
 	retval = ext4_journal_extend(handle, 1);
-	if (retval != 0) {
+	if (retval) {
 		retval = ext4_journal_restart(handle, 1);
 		if (retval)
 			goto err_out;
@@ -350,6 +350,18 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 	i_data[2] = ei->i_data[EXT4_TIND_BLOCK];
 
 	down_write(&EXT4_I(inode)->i_data_sem);
+	/*
+	 * if EXT4_EXT_MIGRATE is cleared a block allocation
+	 * happened after we started the migrate. We need to
+	 * fail the migrate
+	 */
+	if (!(EXT4_I(inode)->i_flags & EXT4_EXT_MIGRATE)) {
+		retval = -EAGAIN;
+		up_write(&EXT4_I(inode)->i_data_sem);
+		goto err_out;
+	} else
+		EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags &
+							~EXT4_EXT_MIGRATE;
 	/*
 	 * We have the extent map build with the tmp inode.
 	 * Now copy the i_data across
@@ -508,6 +520,17 @@ int ext4_ext_migrate(struct inode *inode, struct file *filp,
 	 * switch the inode format to prevent read.
 	 */
 	mutex_lock(&(inode->i_mutex));
+	/*
+	 * Even though we take i_mutex we can still cause block allocation
+	 * via mmap write to holes. If we have allocated new blocks we fail
+	 * migrate.  New block allocation will clear EXT4_EXT_MIGRATE flag.
+	 * The flag is updated with i_data_sem held to prevent racing with
+	 * block allocation.
+	 */
+	down_read((&EXT4_I(inode)->i_data_sem));
+	EXT4_I(inode)->i_flags = EXT4_I(inode)->i_flags | EXT4_EXT_MIGRATE;
+	up_read((&EXT4_I(inode)->i_data_sem));
+
 	handle = ext4_journal_start(inode, 1);
 
 	ei = EXT4_I(inode);
@@ -559,9 +582,15 @@ err_out:
 		 * tmp_inode
 		 */
 		free_ext_block(handle, tmp_inode);
-	else
-		retval = ext4_ext_swap_inode_data(handle, inode,
-							tmp_inode);
+	else {
+		retval = ext4_ext_swap_inode_data(handle, inode, tmp_inode);
+		if (retval)
+			/*
+			 * if we fail to swap inode data free the extent
+			 * details of the tmp inode
+			 */
+			free_ext_block(handle, tmp_inode);
+	}
 
 	/* We mark the tmp_inode dirty via ext4_ext_tree_init. */
 	if (ext4_journal_extend(handle, 1) != 0)

fs/ext4/namei.c

@@ -28,14 +28,14 @@
 #include <linux/pagemap.h>
 #include <linux/jbd2.h>
 #include <linux/time.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/fcntl.h>
 #include <linux/stat.h>
 #include <linux/string.h>
 #include <linux/quotaops.h>
 #include <linux/buffer_head.h>
 #include <linux/bio.h>
+#include "ext4.h"
+#include "ext4_jbd2.h"
 
 #include "namei.h"
 #include "xattr.h"
@@ -57,10 +57,15 @@ static struct buffer_head *ext4_append(handle_t *handle,
 
 	*block = inode->i_size >> inode->i_sb->s_blocksize_bits;
 
-	if ((bh = ext4_bread(handle, inode, *block, 1, err))) {
+	bh = ext4_bread(handle, inode, *block, 1, err);
+	if (bh) {
 		inode->i_size += inode->i_sb->s_blocksize;
 		EXT4_I(inode)->i_disksize = inode->i_size;
-		ext4_journal_get_write_access(handle,bh);
+		*err = ext4_journal_get_write_access(handle, bh);
+		if (*err) {
+			brelse(bh);
+			bh = NULL;
+		}
 	}
 	return bh;
 }
@@ -348,7 +353,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	if (root->info.hash_version != DX_HASH_TEA &&
 	    root->info.hash_version != DX_HASH_HALF_MD4 &&
 	    root->info.hash_version != DX_HASH_LEGACY) {
-		ext4_warning(dir->i_sb, __FUNCTION__,
+		ext4_warning(dir->i_sb, __func__,
 			     "Unrecognised inode hash code %d",
 			     root->info.hash_version);
 		brelse(bh);
@@ -362,7 +367,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	hash = hinfo->hash;
 
 	if (root->info.unused_flags & 1) {
-		ext4_warning(dir->i_sb, __FUNCTION__,
+		ext4_warning(dir->i_sb, __func__,
 			     "Unimplemented inode hash flags: %#06x",
 			     root->info.unused_flags);
 		brelse(bh);
@@ -371,7 +376,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	}
 
 	if ((indirect = root->info.indirect_levels) > 1) {
-		ext4_warning(dir->i_sb, __FUNCTION__,
+		ext4_warning(dir->i_sb, __func__,
 			     "Unimplemented inode hash depth: %#06x",
 			     root->info.indirect_levels);
 		brelse(bh);
@@ -384,7 +389,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 
 	if (dx_get_limit(entries) != dx_root_limit(dir,
 						   root->info.info_length)) {
-		ext4_warning(dir->i_sb, __FUNCTION__,
+		ext4_warning(dir->i_sb, __func__,
 			     "dx entry: limit != root limit");
 		brelse(bh);
 		*err = ERR_BAD_DX_DIR;
@@ -396,7 +401,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 	{
 		count = dx_get_count(entries);
 		if (!count || count > dx_get_limit(entries)) {
-			ext4_warning(dir->i_sb, __FUNCTION__,
+			ext4_warning(dir->i_sb, __func__,
 				     "dx entry: no count or count > limit");
 			brelse(bh);
 			*err = ERR_BAD_DX_DIR;
@@ -441,7 +446,7 @@ dx_probe(struct dentry *dentry, struct inode *dir,
 			goto fail2;
 		at = entries = ((struct dx_node *) bh->b_data)->entries;
 		if (dx_get_limit(entries) != dx_node_limit (dir)) {
-			ext4_warning(dir->i_sb, __FUNCTION__,
+			ext4_warning(dir->i_sb, __func__,
 				     "dx entry: limit != node limit");
 			brelse(bh);
 			*err = ERR_BAD_DX_DIR;
@@ -457,7 +462,7 @@ fail2:
 	}
 fail:
 	if (*err == ERR_BAD_DX_DIR)
-		ext4_warning(dir->i_sb, __FUNCTION__,
+		ext4_warning(dir->i_sb, __func__,
 			     "Corrupt dir inode %ld, running e2fsck is "
 			     "recommended.", dir->i_ino);
 	return NULL;
@@ -914,7 +919,7 @@ restart:
 		wait_on_buffer(bh);
 		if (!buffer_uptodate(bh)) {
 			/* read error, skip block & hope for the best */
-			ext4_error(sb, __FUNCTION__, "reading directory #%lu "
+			ext4_error(sb, __func__, "reading directory #%lu "
 				   "offset %lu", dir->i_ino,
 				   (unsigned long)block);
 			brelse(bh);
@@ -1007,7 +1012,7 @@ static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry,
 		retval = ext4_htree_next_block(dir, hash, frame,
 					       frames, NULL);
 		if (retval < 0) {
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 			     "error reading index page in directory #%lu",
 			     dir->i_ino);
 			*err = retval;
@@ -1532,7 +1537,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
 
 		if (levels && (dx_get_count(frames->entries) ==
 			       dx_get_limit(frames->entries))) {
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 				     "Directory index full!");
 			err = -ENOSPC;
 			goto cleanup;
@@ -1860,11 +1865,11 @@ static int empty_dir (struct inode * inode)
 	if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) ||
 	    !(bh = ext4_bread (NULL, inode, 0, 0, &err))) {
 		if (err)
-			ext4_error(inode->i_sb, __FUNCTION__,
+			ext4_error(inode->i_sb, __func__,
 				   "error %d reading directory #%lu offset 0",
 				   err, inode->i_ino);
 		else
-			ext4_warning(inode->i_sb, __FUNCTION__,
+			ext4_warning(inode->i_sb, __func__,
 				     "bad directory (dir #%lu) - no data block",
 				     inode->i_ino);
 		return 1;
@@ -1893,7 +1898,7 @@ static int empty_dir (struct inode * inode)
 				offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err);
 			if (!bh) {
 				if (err)
-					ext4_error(sb, __FUNCTION__,
+					ext4_error(sb, __func__,
 						   "error %d reading directory"
 						   " #%lu offset %lu",
 						   err, inode->i_ino, offset);
@@ -2217,6 +2222,8 @@ retry:
 			goto out_stop;
 		}
 	} else {
+		/* clear the extent format for fast symlink */
+		EXT4_I(inode)->i_flags &= ~EXT4_EXTENTS_FL;
 		inode->i_op = &ext4_fast_symlink_inode_operations;
 		memcpy((char*)&EXT4_I(inode)->i_data,symname,l);
 		inode->i_size = l-1;
@@ -2347,6 +2354,9 @@ static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry,
 					      EXT4_FEATURE_INCOMPAT_FILETYPE))
 			new_de->file_type = old_de->file_type;
 		new_dir->i_version++;
+		new_dir->i_ctime = new_dir->i_mtime =
+					ext4_current_time(new_dir);
+		ext4_mark_inode_dirty(handle, new_dir);
 		BUFFER_TRACE(new_bh, "call ext4_journal_dirty_metadata");
 		ext4_journal_dirty_metadata(handle, new_bh);
 		brelse(new_bh);

fs/ext4/resize.c

@@ -11,11 +11,10 @@
 
 #define EXT4FS_DEBUG
 
-#include <linux/ext4_jbd2.h>
-
 #include <linux/errno.h>
 #include <linux/slab.h>
 
+#include "ext4_jbd2.h"
 #include "group.h"
 
 #define outside(b, first, last)	((b) < (first) || (b) >= (last))
@@ -50,63 +49,63 @@ static int verify_group_input(struct super_block *sb,
 
 	ext4_get_group_no_and_offset(sb, start, NULL, &offset);
 	if (group != sbi->s_groups_count)
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Cannot add at group %u (only %lu groups)",
 			     input->group, sbi->s_groups_count);
 	else if (offset != 0)
-			ext4_warning(sb, __FUNCTION__, "Last group not full");
+			ext4_warning(sb, __func__, "Last group not full");
 	else if (input->reserved_blocks > input->blocks_count / 5)
-		ext4_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)",
+		ext4_warning(sb, __func__, "Reserved blocks too high (%u)",
 			     input->reserved_blocks);
 	else if (free_blocks_count < 0)
-		ext4_warning(sb, __FUNCTION__, "Bad blocks count %u",
+		ext4_warning(sb, __func__, "Bad blocks count %u",
 			     input->blocks_count);
 	else if (!(bh = sb_bread(sb, end - 1)))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Cannot read last block (%llu)",
 			     end - 1);
 	else if (outside(input->block_bitmap, start, end))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Block bitmap not in group (block %llu)",
 			     (unsigned long long)input->block_bitmap);
 	else if (outside(input->inode_bitmap, start, end))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Inode bitmap not in group (block %llu)",
 			     (unsigned long long)input->inode_bitmap);
 	else if (outside(input->inode_table, start, end) ||
 	         outside(itend - 1, start, end))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Inode table not in group (blocks %llu-%llu)",
 			     (unsigned long long)input->inode_table, itend - 1);
 	else if (input->inode_bitmap == input->block_bitmap)
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Block bitmap same as inode bitmap (%llu)",
 			     (unsigned long long)input->block_bitmap);
 	else if (inside(input->block_bitmap, input->inode_table, itend))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Block bitmap (%llu) in inode table (%llu-%llu)",
 			     (unsigned long long)input->block_bitmap,
 			     (unsigned long long)input->inode_table, itend - 1);
 	else if (inside(input->inode_bitmap, input->inode_table, itend))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Inode bitmap (%llu) in inode table (%llu-%llu)",
 			     (unsigned long long)input->inode_bitmap,
 			     (unsigned long long)input->inode_table, itend - 1);
 	else if (inside(input->block_bitmap, start, metaend))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Block bitmap (%llu) in GDT table"
 			     " (%llu-%llu)",
 			     (unsigned long long)input->block_bitmap,
 			     start, metaend - 1);
 	else if (inside(input->inode_bitmap, start, metaend))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Inode bitmap (%llu) in GDT table"
 			     " (%llu-%llu)",
 			     (unsigned long long)input->inode_bitmap,
 			     start, metaend - 1);
 	else if (inside(input->inode_table, start, metaend) ||
 	         inside(itend - 1, start, metaend))
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Inode table (%llu-%llu) overlaps"
 			     "GDT table (%llu-%llu)",
 			     (unsigned long long)input->inode_table,
@@ -368,7 +367,7 @@ static int verify_reserved_gdb(struct super_block *sb,
 	while ((grp = ext4_list_backups(sb, &three, &five, &seven)) < end) {
 		if (le32_to_cpu(*p++) !=
 		    grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 				     "reserved GDT %llu"
 				     " missing grp %d (%llu)",
 				     blk, grp,
@@ -424,7 +423,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	 */
 	if (EXT4_SB(sb)->s_sbh->b_blocknr !=
 	    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			"won't resize using backup superblock at %llu",
 			(unsigned long long)EXT4_SB(sb)->s_sbh->b_blocknr);
 		return -EPERM;
@@ -448,7 +447,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 
 	data = (__le32 *)dind->b_data;
 	if (le32_to_cpu(data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)]) != gdblock) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "new group %u GDT block %llu not reserved",
 			     input->group, gdblock);
 		err = -EINVAL;
@@ -469,10 +468,10 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		goto exit_dindj;
 
 	n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
-			GFP_KERNEL);
+			GFP_NOFS);
 	if (!n_group_desc) {
 		err = -ENOMEM;
-		ext4_warning (sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			      "not enough memory for %lu groups", gdb_num + 1);
 		goto exit_inode;
 	}
@@ -502,8 +501,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	EXT4_SB(sb)->s_gdb_count++;
 	kfree(o_group_desc);
 
-	es->s_reserved_gdt_blocks =
-		cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1);
+	le16_add_cpu(&es->s_reserved_gdt_blocks, -1);
 	ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh);
 
 	return 0;
@@ -553,7 +551,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	int res, i;
 	int err;
 
-	primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_KERNEL);
+	primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_NOFS);
 	if (!primary)
 		return -ENOMEM;
 
@@ -571,7 +569,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	/* Get each reserved primary GDT block and verify it holds backups */
 	for (res = 0; res < reserved_gdb; res++, blk++) {
 		if (le32_to_cpu(*data) != blk) {
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 				     "reserved block %llu"
 				     " not at offset %ld",
 				     blk,
@@ -715,7 +713,7 @@ static void update_backups(struct super_block *sb,
 	 */
 exit_err:
 	if (err) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "can't update backup for group %lu (err %d), "
 			     "forcing fsck on next reboot", group, err);
 		sbi->s_mount_state &= ~EXT4_VALID_FS;
@@ -755,33 +753,33 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 
 	if (gdb_off == 0 && !EXT4_HAS_RO_COMPAT_FEATURE(sb,
 					EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "Can't resize non-sparse filesystem further");
 		return -EPERM;
 	}
 
 	if (ext4_blocks_count(es) + input->blocks_count <
 	    ext4_blocks_count(es)) {
-		ext4_warning(sb, __FUNCTION__, "blocks_count overflow\n");
+		ext4_warning(sb, __func__, "blocks_count overflow\n");
 		return -EINVAL;
 	}
 
 	if (le32_to_cpu(es->s_inodes_count) + EXT4_INODES_PER_GROUP(sb) <
 	    le32_to_cpu(es->s_inodes_count)) {
-		ext4_warning(sb, __FUNCTION__, "inodes_count overflow\n");
+		ext4_warning(sb, __func__, "inodes_count overflow\n");
 		return -EINVAL;
 	}
 
 	if (reserved_gdb || gdb_off == 0) {
 		if (!EXT4_HAS_COMPAT_FEATURE(sb,
 					     EXT4_FEATURE_COMPAT_RESIZE_INODE)){
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 				     "No reserved GDT blocks, can't resize");
 			return -EPERM;
 		}
 		inode = ext4_iget(sb, EXT4_RESIZE_INO);
 		if (IS_ERR(inode)) {
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 				     "Error opening resize inode");
 			return PTR_ERR(inode);
 		}
@@ -810,7 +808,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 
 	lock_super(sb);
 	if (input->group != sbi->s_groups_count) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "multiple resizers run on filesystem!");
 		err = -EBUSY;
 		goto exit_journal;
@@ -877,8 +875,7 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input)
 	 */
 	ext4_blocks_count_set(es, ext4_blocks_count(es) +
 		input->blocks_count);
-	es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) +
-		EXT4_INODES_PER_GROUP(sb));
+	le32_add_cpu(&es->s_inodes_count, EXT4_INODES_PER_GROUP(sb));
 
 	/*
 	 * We need to protect s_groups_count against other CPUs seeing
@@ -977,13 +974,13 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 			" too large to resize to %llu blocks safely\n",
 			sb->s_id, n_blocks_count);
 		if (sizeof(sector_t) < 8)
-			ext4_warning(sb, __FUNCTION__,
+			ext4_warning(sb, __func__,
 			"CONFIG_LBD not enabled\n");
 		return -EINVAL;
 	}
 
 	if (n_blocks_count < o_blocks_count) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "can't shrink FS - resize aborted");
 		return -EBUSY;
 	}
@@ -992,7 +989,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	ext4_get_group_no_and_offset(sb, o_blocks_count, NULL, &last);
 
 	if (last == 0) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "need to use ext2online to resize further");
 		return -EPERM;
 	}
@@ -1000,7 +997,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	add = EXT4_BLOCKS_PER_GROUP(sb) - last;
 
 	if (o_blocks_count + add < o_blocks_count) {
-		ext4_warning(sb, __FUNCTION__, "blocks_count overflow");
+		ext4_warning(sb, __func__, "blocks_count overflow");
 		return -EINVAL;
 	}
 
@@ -1008,7 +1005,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 		add = n_blocks_count - o_blocks_count;
 
 	if (o_blocks_count + add < n_blocks_count)
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "will only finish group (%llu"
 			     " blocks, %u new)",
 			     o_blocks_count + add, add);
@@ -1016,7 +1013,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	/* See if the device is actually as big as what was requested */
 	bh = sb_bread(sb, o_blocks_count + add -1);
 	if (!bh) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "can't read last block, resize aborted");
 		return -ENOSPC;
 	}
@@ -1028,13 +1025,13 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 	handle = ext4_journal_start_sb(sb, 3);
 	if (IS_ERR(handle)) {
 		err = PTR_ERR(handle);
-		ext4_warning(sb, __FUNCTION__, "error %d on journal start",err);
+		ext4_warning(sb, __func__, "error %d on journal start", err);
 		goto exit_put;
 	}
 
 	lock_super(sb);
 	if (o_blocks_count != ext4_blocks_count(es)) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "multiple resizers run on filesystem!");
 		unlock_super(sb);
 		ext4_journal_stop(handle);
@@ -1044,7 +1041,7 @@ int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es,
 
 	if ((err = ext4_journal_get_write_access(handle,
 						 EXT4_SB(sb)->s_sbh))) {
-		ext4_warning(sb, __FUNCTION__,
+		ext4_warning(sb, __func__,
 			     "error %d on journal write access", err);
 		unlock_super(sb);
 		ext4_journal_stop(handle);

fs/ext4/super.c

@@ -21,8 +21,6 @@
 #include <linux/fs.h>
 #include <linux/time.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
-#include <linux/ext4_jbd2.h>
 #include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/blkdev.h>
@@ -38,9 +36,10 @@
 #include <linux/seq_file.h>
 #include <linux/log2.h>
 #include <linux/crc16.h>
-
 #include <asm/uaccess.h>
 
+#include "ext4.h"
+#include "ext4_jbd2.h"
 #include "xattr.h"
 #include "acl.h"
 #include "namei.h"
@@ -135,7 +134,7 @@ handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks)
 	 * take the FS itself readonly cleanly. */
 	journal = EXT4_SB(sb)->s_journal;
 	if (is_journal_aborted(journal)) {
-		ext4_abort(sb, __FUNCTION__,
+		ext4_abort(sb, __func__,
 			   "Detected aborted journal");
 		return ERR_PTR(-EROFS);
 	}
@@ -355,7 +354,7 @@ void ext4_update_dynamic_rev(struct super_block *sb)
 	if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV)
 		return;
 
-	ext4_warning(sb, __FUNCTION__,
+	ext4_warning(sb, __func__,
 		     "updating to rev %d because of new feature flag, "
 		     "running e2fsck is recommended",
 		     EXT4_DYNAMIC_REV);
@@ -945,8 +944,8 @@ static match_table_t tokens = {
 	{Opt_mballoc, "mballoc"},
 	{Opt_nomballoc, "nomballoc"},
 	{Opt_stripe, "stripe=%u"},
-	{Opt_err, NULL},
 	{Opt_resize, "resize"},
+	{Opt_err, NULL},
 };
 
 static ext4_fsblk_t get_sb_block(void **data)
@@ -1388,11 +1387,11 @@ static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es,
 		 * a plain journaled filesystem we can keep it set as
 		 * valid forever! :)
 		 */
-	es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS);
+	es->s_state &= cpu_to_le16(~EXT4_VALID_FS);
 #endif
 	if (!(__s16) le16_to_cpu(es->s_max_mnt_count))
 		es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT);
-	es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1);
+	le16_add_cpu(&es->s_mnt_count, 1);
 	es->s_mtime = cpu_to_le32(get_seconds());
 	ext4_update_dynamic_rev(sb);
 	EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER);
@@ -1485,36 +1484,33 @@ static int ext4_check_descriptors(struct super_block *sb)
 		block_bitmap = ext4_block_bitmap(sb, gdp);
 		if (block_bitmap < first_block || block_bitmap > last_block)
 		{
-			ext4_error (sb, "ext4_check_descriptors",
-				    "Block bitmap for group %lu"
-				    " not in group (block %llu)!",
-				    i, block_bitmap);
+			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
+			       "Block bitmap for group %lu not in group "
+			       "(block %llu)!", i, block_bitmap);
 			return 0;
 		}
 		inode_bitmap = ext4_inode_bitmap(sb, gdp);
 		if (inode_bitmap < first_block || inode_bitmap > last_block)
 		{
-			ext4_error (sb, "ext4_check_descriptors",
-				    "Inode bitmap for group %lu"
-				    " not in group (block %llu)!",
-				    i, inode_bitmap);
+			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
+			       "Inode bitmap for group %lu not in group "
+			       "(block %llu)!", i, inode_bitmap);
 			return 0;
 		}
 		inode_table = ext4_inode_table(sb, gdp);
 		if (inode_table < first_block ||
 		    inode_table + sbi->s_itb_per_group - 1 > last_block)
 		{
-			ext4_error (sb, "ext4_check_descriptors",
-				    "Inode table for group %lu"
-				    " not in group (block %llu)!",
-				    i, inode_table);
+			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
+			       "Inode table for group %lu not in group "
+			       "(block %llu)!", i, inode_table);
 			return 0;
 		}
 		if (!ext4_group_desc_csum_verify(sbi, i, gdp)) {
-			ext4_error(sb, __FUNCTION__,
-				   "Checksum for group %lu failed (%u!=%u)\n",
-				    i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
-				    gdp)), le16_to_cpu(gdp->bg_checksum));
+			printk(KERN_ERR "EXT4-fs: ext4_check_descriptors: "
+			       "Checksum for group %lu failed (%u!=%u)\n",
+			       i, le16_to_cpu(ext4_group_desc_csum(sbi, i,
+			       gdp)), le16_to_cpu(gdp->bg_checksum));
 			return 0;
 		}
 		if (!flexbg_flag)
@@ -1594,8 +1590,8 @@ static void ext4_orphan_cleanup (struct super_block * sb,
 	while (es->s_last_orphan) {
 		struct inode *inode;
 
-		if (!(inode =
-		      ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) {
+		inode = ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan));
+		if (IS_ERR(inode)) {
 			es->s_last_orphan = 0;
 			break;
 		}
@@ -1605,7 +1601,7 @@ static void ext4_orphan_cleanup (struct super_block * sb,
 		if (inode->i_nlink) {
 			printk(KERN_DEBUG
 				"%s: truncating inode %lu to %Ld bytes\n",
-				__FUNCTION__, inode->i_ino, inode->i_size);
+				__func__, inode->i_ino, inode->i_size);
 			jbd_debug(2, "truncating inode %lu to %Ld bytes\n",
 				  inode->i_ino, inode->i_size);
 			ext4_truncate(inode);
@@ -1613,7 +1609,7 @@ static void ext4_orphan_cleanup (struct super_block * sb,
 		} else {
 			printk(KERN_DEBUG
 				"%s: deleting unreferenced inode %lu\n",
-				__FUNCTION__, inode->i_ino);
+				__func__, inode->i_ino);
 			jbd_debug(2, "deleting unreferenced inode %lu\n",
 				  inode->i_ino);
 			nr_orphans++;
@@ -2699,9 +2695,9 @@ static void ext4_clear_journal_err(struct super_block * sb,
 		char nbuf[16];
 
 		errstr = ext4_decode_error(sb, j_errno, nbuf);
-		ext4_warning(sb, __FUNCTION__, "Filesystem error recorded "
+		ext4_warning(sb, __func__, "Filesystem error recorded "
 			     "from previous mount: %s", errstr);
-		ext4_warning(sb, __FUNCTION__, "Marking fs in need of "
+		ext4_warning(sb, __func__, "Marking fs in need of "
 			     "filesystem check.");
 
 		EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS;
@@ -2828,7 +2824,7 @@ static int ext4_remount (struct super_block * sb, int * flags, char * data)
 	}
 
 	if (sbi->s_mount_opt & EXT4_MOUNT_ABORT)
-		ext4_abort(sb, __FUNCTION__, "Abort forced by user");
+		ext4_abort(sb, __func__, "Abort forced by user");
 
 	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
 		((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0);
@@ -3040,8 +3036,14 @@ static int ext4_dquot_drop(struct inode *inode)
 
 	/* We may delete quota structure so we need to reserve enough blocks */
 	handle = ext4_journal_start(inode, 2*EXT4_QUOTA_DEL_BLOCKS(inode->i_sb));
-	if (IS_ERR(handle))
+	if (IS_ERR(handle)) {
+		/*
+		 * We call dquot_drop() anyway to at least release references
+		 * to quota structures so that umount does not hang.
+		 */
+		dquot_drop(inode);
 		return PTR_ERR(handle);
+	}
 	ret = dquot_drop(inode);
 	err = ext4_journal_stop(handle);
 	if (!ret)

fs/ext4/symlink.c

@@ -19,8 +19,8 @@
 
 #include <linux/fs.h>
 #include <linux/jbd2.h>
-#include <linux/ext4_fs.h>
 #include <linux/namei.h>
+#include "ext4.h"
 #include "xattr.h"
 
 static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd)

fs/ext4/xattr.c

@@ -53,11 +53,11 @@
 #include <linux/init.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
 #include <linux/mbcache.h>
 #include <linux/quotaops.h>
 #include <linux/rwsem.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 #include "xattr.h"
 #include "acl.h"
 
@@ -92,6 +92,8 @@ static struct buffer_head *ext4_xattr_cache_find(struct inode *,
 						 struct mb_cache_entry **);
 static void ext4_xattr_rehash(struct ext4_xattr_header *,
 			      struct ext4_xattr_entry *);
+static int ext4_xattr_list(struct inode *inode, char *buffer,
+			   size_t buffer_size);
 
 static struct mb_cache *ext4_xattr_cache;
 
@@ -225,7 +227,7 @@ ext4_xattr_block_get(struct inode *inode, int name_index, const char *name,
 	ea_bdebug(bh, "b_count=%d, refcount=%d",
 		atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount));
 	if (ext4_xattr_check_block(bh)) {
-bad_block:	ext4_error(inode->i_sb, __FUNCTION__,
+bad_block:	ext4_error(inode->i_sb, __func__,
 			   "inode %lu: bad block %llu", inode->i_ino,
 			   EXT4_I(inode)->i_file_acl);
 		error = -EIO;
@@ -367,7 +369,7 @@ ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
 	ea_bdebug(bh, "b_count=%d, refcount=%d",
 		atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount));
 	if (ext4_xattr_check_block(bh)) {
-		ext4_error(inode->i_sb, __FUNCTION__,
+		ext4_error(inode->i_sb, __func__,
 			   "inode %lu: bad block %llu", inode->i_ino,
 			   EXT4_I(inode)->i_file_acl);
 		error = -EIO;
@@ -420,7 +422,7 @@ cleanup:
  * Returns a negative error number on failure, or the number of bytes
  * used / required on success.
  */
-int
+static int
 ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size)
 {
 	int i_error, b_error;
@@ -484,8 +486,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 		get_bh(bh);
 		ext4_forget(handle, 1, inode, bh, bh->b_blocknr);
 	} else {
-		BHDR(bh)->h_refcount = cpu_to_le32(
-				le32_to_cpu(BHDR(bh)->h_refcount) - 1);
+		le32_add_cpu(&BHDR(bh)->h_refcount, -1);
 		error = ext4_journal_dirty_metadata(handle, bh);
 		if (IS_SYNC(inode))
 			handle->h_sync = 1;
@@ -660,7 +661,7 @@ ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i,
 			atomic_read(&(bs->bh->b_count)),
 			le32_to_cpu(BHDR(bs->bh)->h_refcount));
 		if (ext4_xattr_check_block(bs->bh)) {
-			ext4_error(sb, __FUNCTION__,
+			ext4_error(sb, __func__,
 				"inode %lu: bad block %llu", inode->i_ino,
 				EXT4_I(inode)->i_file_acl);
 			error = -EIO;
@@ -738,7 +739,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 				ce = NULL;
 			}
 			ea_bdebug(bs->bh, "cloning");
-			s->base = kmalloc(bs->bh->b_size, GFP_KERNEL);
+			s->base = kmalloc(bs->bh->b_size, GFP_NOFS);
 			error = -ENOMEM;
 			if (s->base == NULL)
 				goto cleanup;
@@ -750,7 +751,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 		}
 	} else {
 		/* Allocate a buffer where we construct the new block. */
-		s->base = kzalloc(sb->s_blocksize, GFP_KERNEL);
+		s->base = kzalloc(sb->s_blocksize, GFP_NOFS);
 		/* assert(header == s->base) */
 		error = -ENOMEM;
 		if (s->base == NULL)
@@ -789,8 +790,7 @@ inserted:
 				if (error)
 					goto cleanup_dquot;
 				lock_buffer(new_bh);
-				BHDR(new_bh)->h_refcount = cpu_to_le32(1 +
-					le32_to_cpu(BHDR(new_bh)->h_refcount));
+				le32_add_cpu(&BHDR(new_bh)->h_refcount, 1);
 				ea_bdebug(new_bh, "reusing; refcount now=%d",
 					le32_to_cpu(BHDR(new_bh)->h_refcount));
 				unlock_buffer(new_bh);
@@ -808,10 +808,8 @@ inserted:
 			get_bh(new_bh);
 		} else {
 			/* We need to allocate a new block */
-			ext4_fsblk_t goal = le32_to_cpu(
-					EXT4_SB(sb)->s_es->s_first_data_block) +
-				(ext4_fsblk_t)EXT4_I(inode)->i_block_group *
-				EXT4_BLOCKS_PER_GROUP(sb);
+			ext4_fsblk_t goal = ext4_group_first_block_no(sb,
+						EXT4_I(inode)->i_block_group);
 			ext4_fsblk_t block = ext4_new_block(handle, inode,
 							goal, &error);
 			if (error)
@@ -863,7 +861,7 @@ cleanup_dquot:
 	goto cleanup;
 
 bad_block:
-	ext4_error(inode->i_sb, __FUNCTION__,
+	ext4_error(inode->i_sb, __func__,
 		   "inode %lu: bad block %llu", inode->i_ino,
 		   EXT4_I(inode)->i_file_acl);
 	goto cleanup;
@@ -1166,7 +1164,7 @@ retry:
 		if (!bh)
 			goto cleanup;
 		if (ext4_xattr_check_block(bh)) {
-			ext4_error(inode->i_sb, __FUNCTION__,
+			ext4_error(inode->i_sb, __func__,
 				"inode %lu: bad block %llu", inode->i_ino,
 				EXT4_I(inode)->i_file_acl);
 			error = -EIO;
@@ -1341,14 +1339,14 @@ ext4_xattr_delete_inode(handle_t *handle, struct inode *inode)
 		goto cleanup;
 	bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl);
 	if (!bh) {
-		ext4_error(inode->i_sb, __FUNCTION__,
+		ext4_error(inode->i_sb, __func__,
 			"inode %lu: block %llu read error", inode->i_ino,
 			EXT4_I(inode)->i_file_acl);
 		goto cleanup;
 	}
 	if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) ||
 	    BHDR(bh)->h_blocks != cpu_to_le32(1)) {
-		ext4_error(inode->i_sb, __FUNCTION__,
+		ext4_error(inode->i_sb, __func__,
 			"inode %lu: bad block %llu", inode->i_ino,
 			EXT4_I(inode)->i_file_acl);
 		goto cleanup;
@@ -1475,7 +1473,7 @@ again:
 		}
 		bh = sb_bread(inode->i_sb, ce->e_block);
 		if (!bh) {
-			ext4_error(inode->i_sb, __FUNCTION__,
+			ext4_error(inode->i_sb, __func__,
 				"inode %lu: block %lu read error",
 				inode->i_ino, (unsigned long) ce->e_block);
 		} else if (le32_to_cpu(BHDR(bh)->h_refcount) >=

fs/ext4/xattr.h

@@ -74,7 +74,6 @@ extern struct xattr_handler ext4_xattr_security_handler;
 extern ssize_t ext4_listxattr(struct dentry *, char *, size_t);
 
 extern int ext4_xattr_get(struct inode *, int, const char *, void *, size_t);
-extern int ext4_xattr_list(struct inode *, char *, size_t);
 extern int ext4_xattr_set(struct inode *, int, const char *, const void *, size_t, int);
 extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int);
 
@@ -98,12 +97,6 @@ ext4_xattr_get(struct inode *inode, int name_index, const char *name,
 	return -EOPNOTSUPP;
 }
 
-static inline int
-ext4_xattr_list(struct inode *inode, void *buffer, size_t size)
-{
-	return -EOPNOTSUPP;
-}
-
 static inline int
 ext4_xattr_set(struct inode *inode, int name_index, const char *name,
 	       const void *value, size_t size, int flags)

fs/ext4/xattr_security.c

@@ -6,9 +6,9 @@
 #include <linux/module.h>
 #include <linux/string.h>
 #include <linux/fs.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
 #include <linux/security.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 #include "xattr.h"
 
 static size_t

fs/ext4/xattr_trusted.c

@@ -9,8 +9,8 @@
 #include <linux/string.h>
 #include <linux/capability.h>
 #include <linux/fs.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 #include "xattr.h"
 
 #define XATTR_TRUSTED_PREFIX "trusted."

fs/ext4/xattr_user.c

@@ -8,8 +8,8 @@
 #include <linux/module.h>
 #include <linux/string.h>
 #include <linux/fs.h>
-#include <linux/ext4_jbd2.h>
-#include <linux/ext4_fs.h>
+#include "ext4_jbd2.h"
+#include "ext4.h"
 #include "xattr.h"
 
 #define XATTR_USER_PREFIX "user."

fs/jbd2/commit.c

@@ -519,22 +519,6 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 
 	jbd_debug (3, "JBD: commit phase 2\n");
 
-	/*
-	 * First, drop modified flag: all accesses to the buffers
-	 * will be tracked for a new trasaction only -bzzz
-	 */
-	spin_lock(&journal->j_list_lock);
-	if (commit_transaction->t_buffers) {
-		new_jh = jh = commit_transaction->t_buffers->b_tnext;
-		do {
-			J_ASSERT_JH(new_jh, new_jh->b_modified == 1 ||
-					new_jh->b_modified == 0);
-			new_jh->b_modified = 0;
-			new_jh = new_jh->b_tnext;
-		} while (new_jh != jh);
-	}
-	spin_unlock(&journal->j_list_lock);
-
 	/*
 	 * Now start flushing things to disk, in the order they appear
 	 * on the transaction lists.  Data blocks go first.
@@ -584,6 +568,9 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	stats.u.run.rs_blocks = commit_transaction->t_outstanding_credits;
 	stats.u.run.rs_blocks_logged = 0;
 
+	J_ASSERT(commit_transaction->t_nr_buffers <=
+		 commit_transaction->t_outstanding_credits);
+
 	descriptor = NULL;
 	bufs = 0;
 	while (commit_transaction->t_buffers) {

fs/jbd2/journal.c

@@ -534,7 +534,7 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid)
 	if (!tid_geq(journal->j_commit_request, tid)) {
 		printk(KERN_EMERG
 		       "%s: error: j_commit_request=%d, tid=%d\n",
-		       __FUNCTION__, journal->j_commit_request, tid);
+		       __func__, journal->j_commit_request, tid);
 	}
 	spin_unlock(&journal->j_state_lock);
 #endif
@@ -599,7 +599,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr,
 
 			printk(KERN_ALERT "%s: journal block not found "
 					"at offset %lu on %s\n",
-				__FUNCTION__,
+				__func__,
 				blocknr,
 				bdevname(journal->j_dev, b));
 			err = -EIO;
@@ -997,13 +997,14 @@ fail:
  */
 
 /**
- *  journal_t * jbd2_journal_init_dev() - creates an initialises a journal structure
+ *  journal_t * jbd2_journal_init_dev() - creates and initialises a journal structure
  *  @bdev: Block device on which to create the journal
  *  @fs_dev: Device which hold journalled filesystem for this journal.
  *  @start: Block nr Start of journal.
  *  @len:  Length of the journal in blocks.
  *  @blocksize: blocksize of journalling device
- *  @returns: a newly created journal_t *
+ *
+ *  Returns: a newly created journal_t *
  *
  *  jbd2_journal_init_dev creates a journal which maps a fixed contiguous
  *  range of blocks on an arbitrary block device.
@@ -1027,7 +1028,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
 	journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
 	if (!journal->j_wbuf) {
 		printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
-			__FUNCTION__);
+			__func__);
 		kfree(journal);
 		journal = NULL;
 		goto out;
@@ -1083,7 +1084,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
 	journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL);
 	if (!journal->j_wbuf) {
 		printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n",
-			__FUNCTION__);
+			__func__);
 		kfree(journal);
 		return NULL;
 	}
@@ -1092,7 +1093,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
 	/* If that failed, give up */
 	if (err) {
 		printk(KERN_ERR "%s: Cannnot locate journal superblock\n",
-		       __FUNCTION__);
+		       __func__);
 		kfree(journal);
 		return NULL;
 	}
@@ -1178,7 +1179,7 @@ int jbd2_journal_create(journal_t *journal)
 		 */
 		printk(KERN_EMERG
 		       "%s: creation of journal on external device!\n",
-		       __FUNCTION__);
+		       __func__);
 		BUG();
 	}
 
@@ -1976,9 +1977,10 @@ static int journal_init_jbd2_journal_head_cache(void)
 
 static void jbd2_journal_destroy_jbd2_journal_head_cache(void)
 {
-	J_ASSERT(jbd2_journal_head_cache != NULL);
-	kmem_cache_destroy(jbd2_journal_head_cache);
-	jbd2_journal_head_cache = NULL;
+	if (jbd2_journal_head_cache) {
+		kmem_cache_destroy(jbd2_journal_head_cache);
+		jbd2_journal_head_cache = NULL;
+	}
 }
 
 /*
@@ -1997,7 +1999,7 @@ static struct journal_head *journal_alloc_journal_head(void)
 		jbd_debug(1, "out of memory for journal_head\n");
 		if (time_after(jiffies, last_warning + 5*HZ)) {
 			printk(KERN_NOTICE "ENOMEM in %s, retrying.\n",
-			       __FUNCTION__);
+			       __func__);
 			last_warning = jiffies;
 		}
 		while (!ret) {
@@ -2134,13 +2136,13 @@ static void __journal_remove_journal_head(struct buffer_head *bh)
 			if (jh->b_frozen_data) {
 				printk(KERN_WARNING "%s: freeing "
 						"b_frozen_data\n",
-						__FUNCTION__);
+						__func__);
 				jbd2_free(jh->b_frozen_data, bh->b_size);
 			}
 			if (jh->b_committed_data) {
 				printk(KERN_WARNING "%s: freeing "
 						"b_committed_data\n",
-						__FUNCTION__);
+						__func__);
 				jbd2_free(jh->b_committed_data, bh->b_size);
 			}
 			bh->b_private = NULL;
@@ -2305,10 +2307,12 @@ static int __init journal_init(void)
 	BUILD_BUG_ON(sizeof(struct journal_superblock_s) != 1024);
 
 	ret = journal_init_caches();
-	if (ret != 0)
+	if (ret == 0) {
+		jbd2_create_debugfs_entry();
+		jbd2_create_jbd_stats_proc_entry();
+	} else {
 		jbd2_journal_destroy_caches();
-	jbd2_create_debugfs_entry();
-	jbd2_create_jbd_stats_proc_entry();
+	}
 	return ret;
 }
 

fs/jbd2/revoke.c

@@ -139,7 +139,7 @@ repeat:
 oom:
 	if (!journal_oom_retry)
 		return -ENOMEM;
-	jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__);
+	jbd_debug(1, "ENOMEM in %s, retrying\n", __func__);
 	yield();
 	goto repeat;
 }
@@ -167,138 +167,121 @@ static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal,
 	return NULL;
 }
 
+void jbd2_journal_destroy_revoke_caches(void)
+{
+	if (jbd2_revoke_record_cache) {
+		kmem_cache_destroy(jbd2_revoke_record_cache);
+		jbd2_revoke_record_cache = NULL;
+	}
+	if (jbd2_revoke_table_cache) {
+		kmem_cache_destroy(jbd2_revoke_table_cache);
+		jbd2_revoke_table_cache = NULL;
+	}
+}
+
 int __init jbd2_journal_init_revoke_caches(void)
 {
+	J_ASSERT(!jbd2_revoke_record_cache);
+	J_ASSERT(!jbd2_revoke_table_cache);
+
 	jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record",
 					   sizeof(struct jbd2_revoke_record_s),
 					   0,
 					   SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
 					   NULL);
 	if (!jbd2_revoke_record_cache)
-		return -ENOMEM;
+		goto record_cache_failure;
 
 	jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table",
 					   sizeof(struct jbd2_revoke_table_s),
 					   0, SLAB_TEMPORARY, NULL);
-	if (!jbd2_revoke_table_cache) {
-		kmem_cache_destroy(jbd2_revoke_record_cache);
-		jbd2_revoke_record_cache = NULL;
-		return -ENOMEM;
-	}
+	if (!jbd2_revoke_table_cache)
+		goto table_cache_failure;
 	return 0;
+table_cache_failure:
+	jbd2_journal_destroy_revoke_caches();
+record_cache_failure:
+		return -ENOMEM;
 }
 
-void jbd2_journal_destroy_revoke_caches(void)
+static struct jbd2_revoke_table_s *jbd2_journal_init_revoke_table(int hash_size)
 {
-	kmem_cache_destroy(jbd2_revoke_record_cache);
-	jbd2_revoke_record_cache = NULL;
-	kmem_cache_destroy(jbd2_revoke_table_cache);
-	jbd2_revoke_table_cache = NULL;
-}
-
-/* Initialise the revoke table for a given journal to a given size. */
-
-int jbd2_journal_init_revoke(journal_t *journal, int hash_size)
-{
-	int shift, tmp;
+	int shift = 0;
+	int tmp = hash_size;
+	struct jbd2_revoke_table_s *table;
 
-	J_ASSERT (journal->j_revoke_table[0] == NULL);
+	table = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL);
+	if (!table)
+		goto out;
 
-	shift = 0;
-	tmp = hash_size;
 	while((tmp >>= 1UL) != 0UL)
 		shift++;
 
-	journal->j_revoke_table[0] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL);
-	if (!journal->j_revoke_table[0])
-		return -ENOMEM;
-	journal->j_revoke = journal->j_revoke_table[0];
-
-	/* Check that the hash_size is a power of two */
-	J_ASSERT(is_power_of_2(hash_size));
-
-	journal->j_revoke->hash_size = hash_size;
-
-	journal->j_revoke->hash_shift = shift;
-
-	journal->j_revoke->hash_table =
+	table->hash_size = hash_size;
+	table->hash_shift = shift;
+	table->hash_table =
 		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
-	if (!journal->j_revoke->hash_table) {
-		kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]);
-		journal->j_revoke = NULL;
-		return -ENOMEM;
+	if (!table->hash_table) {
+		kmem_cache_free(jbd2_revoke_table_cache, table);
+		table = NULL;
+		goto out;
 	}
 
 	for (tmp = 0; tmp < hash_size; tmp++)
-		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+		INIT_LIST_HEAD(&table->hash_table[tmp]);
 
-	journal->j_revoke_table[1] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL);
-	if (!journal->j_revoke_table[1]) {
-		kfree(journal->j_revoke_table[0]->hash_table);
-		kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]);
-		return -ENOMEM;
+out:
+	return table;
+}
+
+static void jbd2_journal_destroy_revoke_table(struct jbd2_revoke_table_s *table)
+{
+	int i;
+	struct list_head *hash_list;
+
+	for (i = 0; i < table->hash_size; i++) {
+		hash_list = &table->hash_table[i];
+		J_ASSERT(list_empty(hash_list));
 	}
 
-	journal->j_revoke = journal->j_revoke_table[1];
+	kfree(table->hash_table);
+	kmem_cache_free(jbd2_revoke_table_cache, table);
+}
 
-	/* Check that the hash_size is a power of two */
+/* Initialise the revoke table for a given journal to a given size. */
+int jbd2_journal_init_revoke(journal_t *journal, int hash_size)
+{
+	J_ASSERT(journal->j_revoke_table[0] == NULL);
 	J_ASSERT(is_power_of_2(hash_size));
 
-	journal->j_revoke->hash_size = hash_size;
-
-	journal->j_revoke->hash_shift = shift;
+	journal->j_revoke_table[0] = jbd2_journal_init_revoke_table(hash_size);
+	if (!journal->j_revoke_table[0])
+		goto fail0;
 
-	journal->j_revoke->hash_table =
-		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
-	if (!journal->j_revoke->hash_table) {
-		kfree(journal->j_revoke_table[0]->hash_table);
-		kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]);
-		kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[1]);
-		journal->j_revoke = NULL;
-		return -ENOMEM;
-	}
+	journal->j_revoke_table[1] = jbd2_journal_init_revoke_table(hash_size);
+	if (!journal->j_revoke_table[1])
+		goto fail1;
 
-	for (tmp = 0; tmp < hash_size; tmp++)
-		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+	journal->j_revoke = journal->j_revoke_table[1];
 
 	spin_lock_init(&journal->j_revoke_lock);
 
 	return 0;
-}
 
-/* Destoy a journal's revoke table.  The table must already be empty! */
+fail1:
+	jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]);
+fail0:
+	return -ENOMEM;
+}
 
+/* Destroy a journal's revoke table.  The table must already be empty! */
 void jbd2_journal_destroy_revoke(journal_t *journal)
 {
-	struct jbd2_revoke_table_s *table;
-	struct list_head *hash_list;
-	int i;
-
-	table = journal->j_revoke_table[0];
-	if (!table)
-		return;
-
-	for (i=0; i<table->hash_size; i++) {
-		hash_list = &table->hash_table[i];
-		J_ASSERT (list_empty(hash_list));
-	}
-
-	kfree(table->hash_table);
-	kmem_cache_free(jbd2_revoke_table_cache, table);
-	journal->j_revoke = NULL;
-
-	table = journal->j_revoke_table[1];
-	if (!table)
-		return;
-
-	for (i=0; i<table->hash_size; i++) {
-		hash_list = &table->hash_table[i];
-		J_ASSERT (list_empty(hash_list));
-	}
-
-	kfree(table->hash_table);
-	kmem_cache_free(jbd2_revoke_table_cache, table);
 	journal->j_revoke = NULL;
+	if (journal->j_revoke_table[0])
+		jbd2_journal_destroy_revoke_table(journal->j_revoke_table[0]);
+	if (journal->j_revoke_table[1])
+		jbd2_journal_destroy_revoke_table(journal->j_revoke_table[1]);
 }
 
 

fs/jbd2/transaction.c

@@ -617,6 +617,12 @@ repeat:
 	    jh->b_next_transaction == transaction)
 		goto done;
 
+	/*
+	 * this is the first time this transaction is touching this buffer,
+	 * reset the modified flag
+	 */
+       jh->b_modified = 0;
+
 	/*
 	 * If there is already a copy-out version of this buffer, then we don't
 	 * need to make another one
@@ -690,7 +696,7 @@ repeat:
 				if (!frozen_buffer) {
 					printk(KERN_EMERG
 					       "%s: OOM for frozen_buffer\n",
-					       __FUNCTION__);
+					       __func__);
 					JBUFFER_TRACE(jh, "oom!");
 					error = -ENOMEM;
 					jbd_lock_bh_state(bh);
@@ -829,9 +835,16 @@ int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 
 	if (jh->b_transaction == NULL) {
 		jh->b_transaction = transaction;
+
+		/* first access by this transaction */
+		jh->b_modified = 0;
+
 		JBUFFER_TRACE(jh, "file as BJ_Reserved");
 		__jbd2_journal_file_buffer(jh, transaction, BJ_Reserved);
 	} else if (jh->b_transaction == journal->j_committing_transaction) {
+		/* first access by this transaction */
+		jh->b_modified = 0;
+
 		JBUFFER_TRACE(jh, "set next transaction");
 		jh->b_next_transaction = transaction;
 	}
@@ -901,7 +914,7 @@ repeat:
 		committed_data = jbd2_alloc(jh2bh(jh)->b_size, GFP_NOFS);
 		if (!committed_data) {
 			printk(KERN_EMERG "%s: No memory for committed data\n",
-				__FUNCTION__);
+				__func__);
 			err = -ENOMEM;
 			goto out;
 		}
@@ -1230,6 +1243,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 	struct journal_head *jh;
 	int drop_reserve = 0;
 	int err = 0;
+	int was_modified = 0;
 
 	BUFFER_TRACE(bh, "entry");
 
@@ -1248,6 +1262,9 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		goto not_jbd;
 	}
 
+	/* keep track of wether or not this transaction modified us */
+	was_modified = jh->b_modified;
+
 	/*
 	 * The buffer's going from the transaction, we must drop
 	 * all references -bzzz
@@ -1265,7 +1282,12 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 
 		JBUFFER_TRACE(jh, "belongs to current transaction: unfile");
 
-		drop_reserve = 1;
+		/*
+		 * we only want to drop a reference if this transaction
+		 * modified the buffer
+		 */
+		if (was_modified)
+			drop_reserve = 1;
 
 		/*
 		 * We are no longer going to journal this buffer.
@@ -1305,7 +1327,13 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh)
 		if (jh->b_next_transaction) {
 			J_ASSERT(jh->b_next_transaction == transaction);
 			jh->b_next_transaction = NULL;
-			drop_reserve = 1;
+
+			/*
+			 * only drop a reference if this transaction modified
+			 * the buffer
+			 */
+			if (was_modified)
+				drop_reserve = 1;
 		}
 	}
 
@@ -1434,7 +1462,8 @@ int jbd2_journal_stop(handle_t *handle)
 	return err;
 }
 
-/**int jbd2_journal_force_commit() - force any uncommitted transactions
+/**
+ * int jbd2_journal_force_commit() - force any uncommitted transactions
  * @journal: journal to force
  *
  * For synchronous operations: force any uncommitted transactions
@@ -2077,7 +2106,7 @@ void __jbd2_journal_refile_buffer(struct journal_head *jh)
 	jh->b_transaction = jh->b_next_transaction;
 	jh->b_next_transaction = NULL;
 	__jbd2_journal_file_buffer(jh, jh->b_transaction,
-				was_dirty ? BJ_Metadata : BJ_Reserved);
+				jh->b_modified ? BJ_Metadata : BJ_Reserved);
 	J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING);
 
 	if (was_dirty)