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

Pull ext4 updates from Ted Ts'o:

 - add GETFSMAP support

 - some performance improvements for very large file systems and for
   random write workloads into a preallocated file

 - bug fixes and cleanups.

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  jbd2: cleanup write flags handling from jbd2_write_superblock()
  ext4: mark superblock writes synchronous for nobarrier mounts
  ext4: inherit encryption xattr before other xattrs
  ext4: replace BUG_ON with WARN_ONCE in ext4_end_bio()
  ext4: avoid unnecessary transaction stalls during writeback
  ext4: preload block group descriptors
  ext4: make ext4_shutdown() static
  ext4: support GETFSMAP ioctls
  vfs: add common GETFSMAP ioctl definitions
  ext4: evict inline data when writing to memory map
  ext4: remove ext4_xattr_check_entry()
  ext4: rename ext4_xattr_check_names() to ext4_xattr_check_entries()
  ext4: merge ext4_xattr_list() into ext4_listxattr()
  ext4: constify static data that is never modified
  ext4: trim return value and 'dir' argument from ext4_insert_dentry()
  jbd2: fix dbench4 performance regression for 'nobarrier' mounts
  jbd2: Fix lockdep splat with generic/270 test
  mm: retry writepages() on ENOMEM when doing an data integrity writeback

fs/ext4/Makefile

@@ -4,11 +4,11 @@
 
 obj-$(CONFIG_EXT4_FS) += ext4.o
 
-ext4-y	:= balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o page-io.o \
-		ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o \
-		ext4_jbd2.o migrate.o mballoc.o block_validity.o move_extent.o \
-		mmp.o indirect.o extents_status.o xattr.o xattr_user.o \
-		xattr_trusted.o inline.o readpage.o sysfs.o
+ext4-y	:= balloc.o bitmap.o block_validity.o dir.o ext4_jbd2.o extents.o \
+		extents_status.o file.o fsmap.o fsync.o hash.o ialloc.o \
+		indirect.o inline.o inode.o ioctl.o mballoc.o migrate.o \
+		mmp.o move_extent.o namei.o page-io.o readpage.o resize.o \
+		super.o symlink.o sysfs.o xattr.o xattr_trusted.o xattr_user.o
 
 ext4-$(CONFIG_EXT4_FS_POSIX_ACL)	+= acl.o
 ext4-$(CONFIG_EXT4_FS_SECURITY)		+= xattr_security.o

fs/ext4/ext4.h

@@ -2356,8 +2356,7 @@ extern int ext4_find_dest_de(struct inode *dir, struct inode *inode,
 			     void *buf, int buf_size,
 			     struct ext4_filename *fname,
 			     struct ext4_dir_entry_2 **dest_de);
-int ext4_insert_dentry(struct inode *dir,
-		       struct inode *inode,
+void ext4_insert_dentry(struct inode *inode,
 			struct ext4_dir_entry_2 *de,
 			int buf_size,
 			struct ext4_filename *fname);
@@ -2366,7 +2365,7 @@ static inline void ext4_update_dx_flag(struct inode *inode)
 	if (!ext4_has_feature_dir_index(inode->i_sb))
 		ext4_clear_inode_flag(inode, EXT4_INODE_INDEX);
 }
-static unsigned char ext4_filetype_table[] = {
+static const unsigned char ext4_filetype_table[] = {
 	DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK
 };
 
@@ -3050,7 +3049,7 @@ extern int ext4_handle_dirty_dirent_node(handle_t *handle,
 					 struct inode *inode,
 					 struct buffer_head *bh);
 #define S_SHIFT 12
-static unsigned char ext4_type_by_mode[S_IFMT >> S_SHIFT] = {
+static const unsigned char ext4_type_by_mode[S_IFMT >> S_SHIFT] = {
 	[S_IFREG >> S_SHIFT]	= EXT4_FT_REG_FILE,
 	[S_IFDIR >> S_SHIFT]	= EXT4_FT_DIR,
 	[S_IFCHR >> S_SHIFT]	= EXT4_FT_CHRDEV,

fs/ext4/fsmap.h

+/*
+ * Copyright (C) 2017 Oracle.  All Rights Reserved.
+ *
+ * Author: Darrick J. Wong <darrick.wong@oracle.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301, USA.
+ */
+#ifndef __EXT4_FSMAP_H__
+#define	__EXT4_FSMAP_H__
+
+struct fsmap;
+
+/* internal fsmap representation */
+struct ext4_fsmap {
+	struct list_head	fmr_list;
+	dev_t		fmr_device;	/* device id */
+	uint32_t	fmr_flags;	/* mapping flags */
+	uint64_t	fmr_physical;	/* device offset of segment */
+	uint64_t	fmr_owner;	/* owner id */
+	uint64_t	fmr_length;	/* length of segment, blocks */
+};
+
+struct ext4_fsmap_head {
+	uint32_t	fmh_iflags;	/* control flags */
+	uint32_t	fmh_oflags;	/* output flags */
+	unsigned int	fmh_count;	/* # of entries in array incl. input */
+	unsigned int	fmh_entries;	/* # of entries filled in (output). */
+
+	struct ext4_fsmap fmh_keys[2];	/* low and high keys */
+};
+
+void ext4_fsmap_from_internal(struct super_block *sb, struct fsmap *dest,
+		struct ext4_fsmap *src);
+void ext4_fsmap_to_internal(struct super_block *sb, struct ext4_fsmap *dest,
+		struct fsmap *src);
+
+/* fsmap to userspace formatter - copy to user & advance pointer */
+typedef int (*ext4_fsmap_format_t)(struct ext4_fsmap *, void *);
+
+int ext4_getfsmap(struct super_block *sb, struct ext4_fsmap_head *head,
+		ext4_fsmap_format_t formatter, void *arg);
+
+#define EXT4_QUERY_RANGE_ABORT		1
+#define EXT4_QUERY_RANGE_CONTINUE	0
+
+/*	fmr_owner special values for FS_IOC_GETFSMAP; some share w/ XFS */
+#define EXT4_FMR_OWN_FREE	FMR_OWN_FREE      /* free space */
+#define EXT4_FMR_OWN_UNKNOWN	FMR_OWN_UNKNOWN   /* unknown owner */
+#define EXT4_FMR_OWN_FS		FMR_OWNER('X', 1) /* static fs metadata */
+#define EXT4_FMR_OWN_LOG	FMR_OWNER('X', 2) /* journalling log */
+#define EXT4_FMR_OWN_INODES	FMR_OWNER('X', 5) /* inodes */
+#define EXT4_FMR_OWN_GDT	FMR_OWNER('f', 1) /* group descriptors */
+#define EXT4_FMR_OWN_RESV_GDT	FMR_OWNER('f', 2) /* reserved gdt blocks */
+#define EXT4_FMR_OWN_BLKBM	FMR_OWNER('f', 3) /* inode bitmap */
+#define EXT4_FMR_OWN_INOBM	FMR_OWNER('f', 4) /* block bitmap */
+
+#endif /* __EXT4_FSMAP_H__ */

fs/ext4/ialloc.c

@@ -1098,6 +1098,17 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
 	if (err)
 		goto fail_drop;
 
+	/*
+	 * Since the encryption xattr will always be unique, create it first so
+	 * that it's less likely to end up in an external xattr block and
+	 * prevent its deduplication.
+	 */
+	if (encrypt) {
+		err = fscrypt_inherit_context(dir, inode, handle, true);
+		if (err)
+			goto fail_free_drop;
+	}
+
 	err = ext4_init_acl(handle, inode, dir);
 	if (err)
 		goto fail_free_drop;
@@ -1119,12 +1130,6 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir,
 		ei->i_datasync_tid = handle->h_transaction->t_tid;
 	}
 
-	if (encrypt) {
-		err = fscrypt_inherit_context(dir, inode, handle, true);
-		if (err)
-			goto fail_free_drop;
-	}
-
 	err = ext4_mark_inode_dirty(handle, inode);
 	if (err) {
 		ext4_std_error(sb, err);

fs/ext4/inline.c

@@ -1034,7 +1034,7 @@ static int ext4_add_dirent_to_inline(handle_t *handle,
 	err = ext4_journal_get_write_access(handle, iloc->bh);
 	if (err)
 		return err;
-	ext4_insert_dentry(dir, inode, de, inline_size, fname);
+	ext4_insert_dentry(inode, de, inline_size, fname);
 
 	ext4_show_inline_dir(dir, iloc->bh, inline_start, inline_size);
 

fs/ext4/inode.c

@@ -1643,6 +1643,7 @@ struct mpage_da_data {
 	 */
 	struct ext4_map_blocks map;
 	struct ext4_io_submit io_submit;	/* IO submission data */
+	unsigned int do_map:1;
 };
 
 static void mpage_release_unused_pages(struct mpage_da_data *mpd,
@@ -2179,6 +2180,9 @@ static bool mpage_add_bh_to_extent(struct mpage_da_data *mpd, ext4_lblk_t lblk,
 
 	/* First block in the extent? */
 	if (map->m_len == 0) {
+		/* We cannot map unless handle is started... */
+		if (!mpd->do_map)
+			return false;
 		map->m_lblk = lblk;
 		map->m_len = 1;
 		map->m_flags = bh->b_state & BH_FLAGS;
@@ -2231,6 +2235,9 @@ static int mpage_process_page_bufs(struct mpage_da_data *mpd,
 			/* Found extent to map? */
 			if (mpd->map.m_len)
 				return 0;
+			/* Buffer needs mapping and handle is not started? */
+			if (!mpd->do_map)
+				return 0;
 			/* Everything mapped so far and we hit EOF */
 			break;
 		}
@@ -2747,6 +2754,29 @@ static int ext4_writepages(struct address_space *mapping,
 		tag_pages_for_writeback(mapping, mpd.first_page, mpd.last_page);
 	done = false;
 	blk_start_plug(&plug);
+
+	/*
+	 * First writeback pages that don't need mapping - we can avoid
+	 * starting a transaction unnecessarily and also avoid being blocked
+	 * in the block layer on device congestion while having transaction
+	 * started.
+	 */
+	mpd.do_map = 0;
+	mpd.io_submit.io_end = ext4_init_io_end(inode, GFP_KERNEL);
+	if (!mpd.io_submit.io_end) {
+		ret = -ENOMEM;
+		goto unplug;
+	}
+	ret = mpage_prepare_extent_to_map(&mpd);
+	/* Submit prepared bio */
+	ext4_io_submit(&mpd.io_submit);
+	ext4_put_io_end_defer(mpd.io_submit.io_end);
+	mpd.io_submit.io_end = NULL;
+	/* Unlock pages we didn't use */
+	mpage_release_unused_pages(&mpd, false);
+	if (ret < 0)
+		goto unplug;
+
 	while (!done && mpd.first_page <= mpd.last_page) {
 		/* For each extent of pages we use new io_end */
 		mpd.io_submit.io_end = ext4_init_io_end(inode, GFP_KERNEL);
@@ -2775,8 +2805,10 @@ static int ext4_writepages(struct address_space *mapping,
 				wbc->nr_to_write, inode->i_ino, ret);
 			/* Release allocated io_end */
 			ext4_put_io_end(mpd.io_submit.io_end);
+			mpd.io_submit.io_end = NULL;
 			break;
 		}
+		mpd.do_map = 1;
 
 		trace_ext4_da_write_pages(inode, mpd.first_page, mpd.wbc);
 		ret = mpage_prepare_extent_to_map(&mpd);
@@ -2807,6 +2839,7 @@ static int ext4_writepages(struct address_space *mapping,
 		if (!ext4_handle_valid(handle) || handle->h_sync == 0) {
 			ext4_journal_stop(handle);
 			handle = NULL;
+			mpd.do_map = 0;
 		}
 		/* Submit prepared bio */
 		ext4_io_submit(&mpd.io_submit);
@@ -2824,6 +2857,7 @@ static int ext4_writepages(struct address_space *mapping,
 			ext4_journal_stop(handle);
 		} else
 			ext4_put_io_end(mpd.io_submit.io_end);
+		mpd.io_submit.io_end = NULL;
 
 		if (ret == -ENOSPC && sbi->s_journal) {
 			/*
@@ -2839,6 +2873,7 @@ static int ext4_writepages(struct address_space *mapping,
 		if (ret)
 			break;
 	}
+unplug:
 	blk_finish_plug(&plug);
 	if (!ret && !cycled && wbc->nr_to_write > 0) {
 		cycled = 1;
@@ -5855,6 +5890,11 @@ int ext4_page_mkwrite(struct vm_fault *vmf)
 	file_update_time(vma->vm_file);
 
 	down_read(&EXT4_I(inode)->i_mmap_sem);
+
+	ret = ext4_convert_inline_data(inode);
+	if (ret)
+		goto out_ret;
+
 	/* Delalloc case is easy... */
 	if (test_opt(inode->i_sb, DELALLOC) &&
 	    !ext4_should_journal_data(inode) &&

fs/ext4/ioctl.c

@@ -19,6 +19,9 @@
 #include <linux/delay.h>
 #include "ext4_jbd2.h"
 #include "ext4.h"
+#include <linux/fsmap.h>
+#include "fsmap.h"
+#include <trace/events/ext4.h>
 
 /**
  * Swap memory between @a and @b for @len bytes.
@@ -443,7 +446,7 @@ static inline unsigned long ext4_xflags_to_iflags(__u32 xflags)
 	return iflags;
 }
 
-int ext4_shutdown(struct super_block *sb, unsigned long arg)
+static int ext4_shutdown(struct super_block *sb, unsigned long arg)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	__u32 flags;
@@ -489,6 +492,90 @@ int ext4_shutdown(struct super_block *sb, unsigned long arg)
 	return 0;
 }
 
+struct getfsmap_info {
+	struct super_block	*gi_sb;
+	struct fsmap_head __user *gi_data;
+	unsigned int		gi_idx;
+	__u32			gi_last_flags;
+};
+
+static int ext4_getfsmap_format(struct ext4_fsmap *xfm, void *priv)
+{
+	struct getfsmap_info *info = priv;
+	struct fsmap fm;
+
+	trace_ext4_getfsmap_mapping(info->gi_sb, xfm);
+
+	info->gi_last_flags = xfm->fmr_flags;
+	ext4_fsmap_from_internal(info->gi_sb, &fm, xfm);
+	if (copy_to_user(&info->gi_data->fmh_recs[info->gi_idx++], &fm,
+			sizeof(struct fsmap)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static int ext4_ioc_getfsmap(struct super_block *sb,
+			     struct fsmap_head __user *arg)
+{
+	struct getfsmap_info info = {0};
+	struct ext4_fsmap_head xhead = {0};
+	struct fsmap_head head;
+	bool aborted = false;
+	int error;
+
+	if (copy_from_user(&head, arg, sizeof(struct fsmap_head)))
+		return -EFAULT;
+	if (memchr_inv(head.fmh_reserved, 0, sizeof(head.fmh_reserved)) ||
+	    memchr_inv(head.fmh_keys[0].fmr_reserved, 0,
+		       sizeof(head.fmh_keys[0].fmr_reserved)) ||
+	    memchr_inv(head.fmh_keys[1].fmr_reserved, 0,
+		       sizeof(head.fmh_keys[1].fmr_reserved)))
+		return -EINVAL;
+	/*
+	 * ext4 doesn't report file extents at all, so the only valid
+	 * file offsets are the magic ones (all zeroes or all ones).
+	 */
+	if (head.fmh_keys[0].fmr_offset ||
+	    (head.fmh_keys[1].fmr_offset != 0 &&
+	     head.fmh_keys[1].fmr_offset != -1ULL))
+		return -EINVAL;
+
+	xhead.fmh_iflags = head.fmh_iflags;
+	xhead.fmh_count = head.fmh_count;
+	ext4_fsmap_to_internal(sb, &xhead.fmh_keys[0], &head.fmh_keys[0]);
+	ext4_fsmap_to_internal(sb, &xhead.fmh_keys[1], &head.fmh_keys[1]);
+
+	trace_ext4_getfsmap_low_key(sb, &xhead.fmh_keys[0]);
+	trace_ext4_getfsmap_high_key(sb, &xhead.fmh_keys[1]);
+
+	info.gi_sb = sb;
+	info.gi_data = arg;
+	error = ext4_getfsmap(sb, &xhead, ext4_getfsmap_format, &info);
+	if (error == EXT4_QUERY_RANGE_ABORT) {
+		error = 0;
+		aborted = true;
+	} else if (error)
+		return error;
+
+	/* If we didn't abort, set the "last" flag in the last fmx */
+	if (!aborted && info.gi_idx) {
+		info.gi_last_flags |= FMR_OF_LAST;
+		if (copy_to_user(&info.gi_data->fmh_recs[info.gi_idx - 1].fmr_flags,
+				 &info.gi_last_flags,
+				 sizeof(info.gi_last_flags)))
+			return -EFAULT;
+	}
+
+	/* copy back header */
+	head.fmh_entries = xhead.fmh_entries;
+	head.fmh_oflags = xhead.fmh_oflags;
+	if (copy_to_user(arg, &head, sizeof(struct fsmap_head)))
+		return -EFAULT;
+
+	return 0;
+}
+
 long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	struct inode *inode = file_inode(filp);
@@ -499,6 +586,8 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	ext4_debug("cmd = %u, arg = %lu\n", cmd, arg);
 
 	switch (cmd) {
+	case FS_IOC_GETFSMAP:
+		return ext4_ioc_getfsmap(sb, (void __user *)arg);
 	case EXT4_IOC_GETFLAGS:
 		flags = ei->i_flags & EXT4_FL_USER_VISIBLE;
 		return put_user(flags, (int __user *) arg);
@@ -1007,6 +1096,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case EXT4_IOC_GET_ENCRYPTION_PWSALT:
 	case EXT4_IOC_GET_ENCRYPTION_POLICY:
 	case EXT4_IOC_SHUTDOWN:
+	case FS_IOC_GETFSMAP:
 		break;
 	default:
 		return -ENOIOCTLCMD;

fs/ext4/mballoc.c

@@ -357,7 +357,7 @@ static struct kmem_cache *ext4_free_data_cachep;
 #define NR_GRPINFO_CACHES 8
 static struct kmem_cache *ext4_groupinfo_caches[NR_GRPINFO_CACHES];
 
-static const char *ext4_groupinfo_slab_names[NR_GRPINFO_CACHES] = {
+static const char * const ext4_groupinfo_slab_names[NR_GRPINFO_CACHES] = {
 	"ext4_groupinfo_1k", "ext4_groupinfo_2k", "ext4_groupinfo_4k",
 	"ext4_groupinfo_8k", "ext4_groupinfo_16k", "ext4_groupinfo_32k",
 	"ext4_groupinfo_64k", "ext4_groupinfo_128k"
@@ -5277,3 +5277,52 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 	range->len = EXT4_C2B(EXT4_SB(sb), trimmed) << sb->s_blocksize_bits;
 	return ret;
 }
+
+/* Iterate all the free extents in the group. */
+int
+ext4_mballoc_query_range(
+	struct super_block		*sb,
+	ext4_group_t			group,
+	ext4_grpblk_t			start,
+	ext4_grpblk_t			end,
+	ext4_mballoc_query_range_fn	formatter,
+	void				*priv)
+{
+	void				*bitmap;
+	ext4_grpblk_t			next;
+	struct ext4_buddy		e4b;
+	int				error;
+
+	error = ext4_mb_load_buddy(sb, group, &e4b);
+	if (error)
+		return error;
+	bitmap = e4b.bd_bitmap;
+
+	ext4_lock_group(sb, group);
+
+	start = (e4b.bd_info->bb_first_free > start) ?
+		e4b.bd_info->bb_first_free : start;
+	if (end >= EXT4_CLUSTERS_PER_GROUP(sb))
+		end = EXT4_CLUSTERS_PER_GROUP(sb) - 1;
+
+	while (start <= end) {
+		start = mb_find_next_zero_bit(bitmap, end + 1, start);
+		if (start > end)
+			break;
+		next = mb_find_next_bit(bitmap, end + 1, start);
+
+		ext4_unlock_group(sb, group);
+		error = formatter(sb, group, start, next - start, priv);
+		if (error)
+			goto out_unload;
+		ext4_lock_group(sb, group);
+
+		start = next + 1;
+	}
+
+	ext4_unlock_group(sb, group);
+out_unload:
+	ext4_mb_unload_buddy(&e4b);
+
+	return error;
+}

fs/ext4/mballoc.h

@@ -199,4 +199,21 @@ static inline ext4_fsblk_t ext4_grp_offs_to_block(struct super_block *sb,
 	return ext4_group_first_block_no(sb, fex->fe_group) +
 		(fex->fe_start << EXT4_SB(sb)->s_cluster_bits);
 }
+
+typedef int (*ext4_mballoc_query_range_fn)(
+	struct super_block		*sb,
+	ext4_group_t			agno,
+	ext4_grpblk_t			start,
+	ext4_grpblk_t			len,
+	void				*priv);
+
+int
+ext4_mballoc_query_range(
+	struct super_block		*sb,
+	ext4_group_t			agno,
+	ext4_grpblk_t			start,
+	ext4_grpblk_t			end,
+	ext4_mballoc_query_range_fn	formatter,
+	void				*priv);
+
 #endif

fs/ext4/namei.c

@@ -1869,8 +1869,7 @@ int ext4_find_dest_de(struct inode *dir, struct inode *inode,
 	return res;
 }
 
-int ext4_insert_dentry(struct inode *dir,
-		       struct inode *inode,
+void ext4_insert_dentry(struct inode *inode,
 			struct ext4_dir_entry_2 *de,
 			int buf_size,
 			struct ext4_filename *fname)
@@ -1892,7 +1891,6 @@ int ext4_insert_dentry(struct inode *dir,
 	ext4_set_de_type(inode->i_sb, de, inode->i_mode);
 	de->name_len = fname_len(fname);
 	memcpy(de->name, fname_name(fname), fname_len(fname));
-	return 0;
 }
 
 /*
@@ -1928,11 +1926,8 @@ static int add_dirent_to_buf(handle_t *handle, struct ext4_filename *fname,
 		return err;
 	}
 
-	/* By now the buffer is marked for journaling. Due to crypto operations,
-	 * the following function call may fail */
-	err = ext4_insert_dentry(dir, inode, de, blocksize, fname);
-	if (err < 0)
-		return err;
+	/* By now the buffer is marked for journaling */
+	ext4_insert_dentry(inode, de, blocksize, fname);
 
 	/*
 	 * XXX shouldn't update any times until successful

fs/ext4/page-io.c

@@ -297,8 +297,17 @@ static void ext4_end_bio(struct bio *bio)
 {
 	ext4_io_end_t *io_end = bio->bi_private;
 	sector_t bi_sector = bio->bi_iter.bi_sector;
+	char b[BDEVNAME_SIZE];
 
-	BUG_ON(!io_end);
+	if (WARN_ONCE(!io_end, "io_end is NULL: %s: sector %Lu len %u err %d\n",
+		      bdevname(bio->bi_bdev, b),
+		      (long long) bio->bi_iter.bi_sector,
+		      (unsigned) bio_sectors(bio),
+		      bio->bi_error)) {
+		ext4_finish_bio(bio);
+		bio_put(bio);
+		return;
+	}
 	bio->bi_end_io = NULL;
 
 	if (bio->bi_error) {

fs/ext4/super.c

@@ -49,6 +49,7 @@
 #include "xattr.h"
 #include "acl.h"
 #include "mballoc.h"
+#include "fsmap.h"
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/ext4.h>
@@ -1230,7 +1231,7 @@ static const struct fscrypt_operations ext4_cryptops = {
 #endif
 
 #ifdef CONFIG_QUOTA
-static char *quotatypes[] = INITQFNAMES;
+static const char * const quotatypes[] = INITQFNAMES;
 #define QTYPE2NAME(t) (quotatypes[t])
 
 static int ext4_write_dquot(struct dquot *dquot);
@@ -1443,7 +1444,8 @@ static ext4_fsblk_t get_sb_block(void **data)
 }
 
 #define DEFAULT_JOURNAL_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3))
-static char deprecated_msg[] = "Mount option \"%s\" will be removed by %s\n"
+static const char deprecated_msg[] =
+	"Mount option \"%s\" will be removed by %s\n"
 	"Contact linux-ext4@vger.kernel.org if you think we should keep it.\n";
 
 #ifdef CONFIG_QUOTA
@@ -3898,6 +3900,12 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 
 	bgl_lock_init(sbi->s_blockgroup_lock);
 
+	/* Pre-read the descriptors into the buffer cache */
+	for (i = 0; i < db_count; i++) {
+		block = descriptor_loc(sb, logical_sb_block, i);
+		sb_breadahead(sb, block);
+	}
+
 	for (i = 0; i < db_count; i++) {
 		block = descriptor_loc(sb, logical_sb_block, i);
 		sbi->s_group_desc[i] = sb_bread_unmovable(sb, block);
@@ -4650,7 +4658,7 @@ static int ext4_commit_super(struct super_block *sb, int sync)
 	if (sync) {
 		unlock_buffer(sbh);
 		error = __sync_dirty_buffer(sbh,
-			test_opt(sb, BARRIER) ? REQ_FUA : REQ_SYNC);
+			REQ_SYNC | (test_opt(sb, BARRIER) ? REQ_FUA : 0));
 		if (error)
 			return error;
 

fs/ext4/sysfs.c

@@ -34,7 +34,7 @@ typedef enum {
 	ptr_ext4_super_block_offset,
 } attr_ptr_t;
 
-static const char *proc_dirname = "fs/ext4";
+static const char proc_dirname[] = "fs/ext4";
 static struct proc_dir_entry *ext4_proc_root;
 
 struct ext4_attr {
@@ -375,7 +375,7 @@ static const struct file_operations ext4_seq_##name##_fops = { \
 PROC_FILE_SHOW_DEFN(es_shrinker_info);
 PROC_FILE_SHOW_DEFN(options);
 
-static struct ext4_proc_files {
+static const struct ext4_proc_files {
 	const char *name;
 	const struct file_operations *fops;
 } proc_files[] = {
@@ -388,7 +388,7 @@ static struct ext4_proc_files {
 int ext4_register_sysfs(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct ext4_proc_files *p;
+	const struct ext4_proc_files *p;
 	int err;
 
 	sbi->s_kobj.kset = &ext4_kset;
@@ -412,7 +412,7 @@ int ext4_register_sysfs(struct super_block *sb)
 void ext4_unregister_sysfs(struct super_block *sb)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
-	struct ext4_proc_files *p;
+	const struct ext4_proc_files *p;
 
 	if (sbi->s_proc) {
 		for (p = proc_files; p->name; p++)

fs/ext4/xattr.c

@@ -78,10 +78,8 @@ 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 dentry *dentry, char *buffer,
-			   size_t buffer_size);
 
-static const struct xattr_handler *ext4_xattr_handler_map[] = {
+static const struct xattr_handler * const ext4_xattr_handler_map[] = {
 	[EXT4_XATTR_INDEX_USER]		     = &ext4_xattr_user_handler,
 #ifdef CONFIG_EXT4_FS_POSIX_ACL
 	[EXT4_XATTR_INDEX_POSIX_ACL_ACCESS]  = &posix_acl_access_xattr_handler,
@@ -163,19 +161,8 @@ ext4_xattr_handler(int name_index)
 	return handler;
 }
 
-/*
- * Inode operation listxattr()
- *
- * d_inode(dentry)->i_mutex: don't care
- */
-ssize_t
-ext4_listxattr(struct dentry *dentry, char *buffer, size_t size)
-{
-	return ext4_xattr_list(dentry, buffer, size);
-}
-
 static int
-ext4_xattr_check_names(struct ext4_xattr_entry *entry, void *end,
+ext4_xattr_check_entries(struct ext4_xattr_entry *entry, void *end,
 			 void *value_start)
 {
 	struct ext4_xattr_entry *e = entry;
@@ -230,7 +217,7 @@ ext4_xattr_check_block(struct inode *inode, struct buffer_head *bh)
 		return -EFSCORRUPTED;
 	if (!ext4_xattr_block_csum_verify(inode, bh))
 		return -EFSBADCRC;
-	error = ext4_xattr_check_names(BFIRST(bh), bh->b_data + bh->b_size,
+	error = ext4_xattr_check_entries(BFIRST(bh), bh->b_data + bh->b_size,
 					 bh->b_data);
 	if (!error)
 		set_buffer_verified(bh);
@@ -246,7 +233,7 @@ __xattr_check_inode(struct inode *inode, struct ext4_xattr_ibody_header *header,
 	if (end - (void *)header < sizeof(*header) + sizeof(u32) ||
 	    (header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)))
 		goto errout;
-	error = ext4_xattr_check_names(IFIRST(header), end, IFIRST(header));
+	error = ext4_xattr_check_entries(IFIRST(header), end, IFIRST(header));
 errout:
 	if (error)
 		__ext4_error_inode(inode, function, line, 0,
@@ -257,20 +244,9 @@ __xattr_check_inode(struct inode *inode, struct ext4_xattr_ibody_header *header,
 #define xattr_check_inode(inode, header, end) \
 	__xattr_check_inode((inode), (header), (end), __func__, __LINE__)
 
-static inline int
-ext4_xattr_check_entry(struct ext4_xattr_entry *entry, size_t size)
-{
-	size_t value_size = le32_to_cpu(entry->e_value_size);
-
-	if (entry->e_value_block != 0 || value_size > size ||
-	    le16_to_cpu(entry->e_value_offs) + value_size > size)
-		return -EFSCORRUPTED;
-	return 0;
-}
-
 static int
 ext4_xattr_find_entry(struct ext4_xattr_entry **pentry, int name_index,
-		      const char *name, size_t size, int sorted)
+		      const char *name, int sorted)
 {
 	struct ext4_xattr_entry *entry;
 	size_t name_len;
@@ -290,8 +266,6 @@ ext4_xattr_find_entry(struct ext4_xattr_entry **pentry, int name_index,
 			break;
 	}
 	*pentry = entry;
-	if (!cmp && ext4_xattr_check_entry(entry, size))
-		return -EFSCORRUPTED;
 	return cmp ? -ENODATA : 0;
 }
 
@@ -319,7 +293,6 @@ 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(inode, bh)) {
-bad_block:
 		EXT4_ERROR_INODE(inode, "bad block %llu",
 				 EXT4_I(inode)->i_file_acl);
 		error = -EFSCORRUPTED;
@@ -327,9 +300,7 @@ ext4_xattr_block_get(struct inode *inode, int name_index, const char *name,
 	}
 	ext4_xattr_cache_insert(ext4_mb_cache, bh);
 	entry = BFIRST(bh);
-	error = ext4_xattr_find_entry(&entry, name_index, name, bh->b_size, 1);
-	if (error == -EFSCORRUPTED)
-		goto bad_block;
+	error = ext4_xattr_find_entry(&entry, name_index, name, 1);
 	if (error)
 		goto cleanup;
 	size = le32_to_cpu(entry->e_value_size);
@@ -366,13 +337,12 @@ ext4_xattr_ibody_get(struct inode *inode, int name_index, const char *name,
 		return error;
 	raw_inode = ext4_raw_inode(&iloc);
 	header = IHDR(inode, raw_inode);
-	entry = IFIRST(header);
 	end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size;
 	error = xattr_check_inode(inode, header, end);
 	if (error)
 		goto cleanup;
-	error = ext4_xattr_find_entry(&entry, name_index, name,
-				      end - (void *)entry, 0);
+	entry = IFIRST(header);
+	error = ext4_xattr_find_entry(&entry, name_index, name, 0);
 	if (error)
 		goto cleanup;
 	size = le32_to_cpu(entry->e_value_size);
@@ -519,7 +489,9 @@ ext4_xattr_ibody_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 }
 
 /*
- * ext4_xattr_list()
+ * Inode operation listxattr()
+ *
+ * d_inode(dentry)->i_rwsem: don't care
  *
  * Copy a list of attribute names into the buffer
  * provided, or compute the buffer size required.
@@ -528,8 +500,8 @@ ext4_xattr_ibody_list(struct dentry *dentry, char *buffer, size_t buffer_size)
  * Returns a negative error number on failure, or the number of bytes
  * used / required on success.
  */
-static int
-ext4_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
+ssize_t
+ext4_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 {
 	int ret, ret2;
 
@@ -804,7 +776,7 @@ ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i,
 		bs->s.end = bs->bh->b_data + bs->bh->b_size;
 		bs->s.here = bs->s.first;
 		error = ext4_xattr_find_entry(&bs->s.here, i->name_index,
-					      i->name, bs->bh->b_size, 1);
+					      i->name, 1);
 		if (error && error != -ENODATA)
 			goto cleanup;
 		bs->s.not_found = error;
@@ -1076,8 +1048,7 @@ int ext4_xattr_ibody_find(struct inode *inode, struct ext4_xattr_info *i,
 			return error;
 		/* Find the named attribute. */
 		error = ext4_xattr_find_entry(&is->s.here, i->name_index,
-					      i->name, is->s.end -
-					      (void *)is->s.base, 0);
+					      i->name, 0);
 		if (error && error != -ENODATA)
 			return error;
 		is->s.not_found = error;

fs/jbd2/journal.c

@@ -700,8 +700,21 @@ int jbd2_log_wait_commit(journal_t *journal, tid_t tid)
 {
 	int err = 0;
 
+	read_lock(&journal->j_state_lock);
+#ifdef CONFIG_PROVE_LOCKING
+	/*
+	 * Some callers make sure transaction is already committing and in that
+	 * case we cannot block on open handles anymore. So don't warn in that
+	 * case.
+	 */
+	if (tid_gt(tid, journal->j_commit_sequence) &&
+	    (!journal->j_committing_transaction ||
+	     journal->j_committing_transaction->t_tid != tid)) {
+		read_unlock(&journal->j_state_lock);
 		jbd2_might_wait_for_commit(journal);
 		read_lock(&journal->j_state_lock);
+	}
+#endif
 #ifdef CONFIG_JBD2_DEBUG
 	if (!tid_geq(journal->j_commit_request, tid)) {
 		printk(KERN_ERR
@@ -922,7 +935,8 @@ int __jbd2_update_log_tail(journal_t *journal, tid_t tid, unsigned long block)
 	 * space and if we lose sb update during power failure we'd replay
 	 * old transaction with possibly newly overwritten data.
 	 */
-	ret = jbd2_journal_update_sb_log_tail(journal, tid, block, REQ_FUA);
+	ret = jbd2_journal_update_sb_log_tail(journal, tid, block,
+					      REQ_SYNC | REQ_FUA);
 	if (ret)
 		goto out;
 
@@ -1323,7 +1337,7 @@ static int journal_reset(journal_t *journal)
 		jbd2_journal_update_sb_log_tail(journal,
 						journal->j_tail_sequence,
 						journal->j_tail,
-						REQ_FUA);
+						REQ_SYNC | REQ_FUA);
 		mutex_unlock(&journal->j_checkpoint_mutex);
 	}
 	return jbd2_journal_start_thread(journal);
@@ -1463,7 +1477,7 @@ void jbd2_journal_update_sb_errno(journal_t *journal)
 	sb->s_errno    = cpu_to_be32(journal->j_errno);
 	read_unlock(&journal->j_state_lock);
 
-	jbd2_write_superblock(journal, REQ_FUA);
+	jbd2_write_superblock(journal, REQ_SYNC | REQ_FUA);
 }
 EXPORT_SYMBOL(jbd2_journal_update_sb_errno);
 
@@ -1730,7 +1744,7 @@ int jbd2_journal_destroy(journal_t *journal)
 			write_unlock(&journal->j_state_lock);
 
 			jbd2_mark_journal_empty(journal,
-					REQ_PREFLUSH | REQ_FUA);
+					REQ_SYNC | REQ_PREFLUSH | REQ_FUA);
 			mutex_unlock(&journal->j_checkpoint_mutex);
 		} else
 			err = -EIO;
@@ -1989,7 +2003,7 @@ int jbd2_journal_flush(journal_t *journal)
 	 * the magic code for a fully-recovered superblock.  Any future
 	 * commits of data to the journal will restore the current
 	 * s_start value. */
-	jbd2_mark_journal_empty(journal, REQ_FUA);
+	jbd2_mark_journal_empty(journal, REQ_SYNC | REQ_FUA);
 	mutex_unlock(&journal->j_checkpoint_mutex);
 	write_lock(&journal->j_state_lock);
 	J_ASSERT(!journal->j_running_transaction);
@@ -2035,7 +2049,7 @@ int jbd2_journal_wipe(journal_t *journal, int write)
 	if (write) {
 		/* Lock to make assertions happy... */
 		mutex_lock(&journal->j_checkpoint_mutex);
-		jbd2_mark_journal_empty(journal, REQ_FUA);
+		jbd2_mark_journal_empty(journal, REQ_SYNC | REQ_FUA);
 		mutex_unlock(&journal->j_checkpoint_mutex);
 	}
 

include/trace/events/ext4.h

@@ -15,6 +15,7 @@ struct ext4_inode_info;
 struct mpage_da_data;
 struct ext4_map_blocks;
 struct extent_status;
+struct ext4_fsmap;
 
 #define EXT4_I(inode) (container_of(inode, struct ext4_inode_info, vfs_inode))
 
@@ -2529,6 +2530,79 @@ TRACE_EVENT(ext4_es_shrink,
 		  __entry->scan_time, __entry->nr_skipped, __entry->retried)
 );
 
+/* fsmap traces */
+DECLARE_EVENT_CLASS(ext4_fsmap_class,
+	TP_PROTO(struct super_block *sb, u32 keydev, u32 agno, u64 bno, u64 len,
+		 u64 owner),
+	TP_ARGS(sb, keydev, agno, bno, len, owner),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(dev_t, keydev)
+		__field(u32, agno)
+		__field(u64, bno)
+		__field(u64, len)
+		__field(u64, owner)
+	),
+	TP_fast_assign(
+		__entry->dev = sb->s_bdev->bd_dev;
+		__entry->keydev = new_decode_dev(keydev);
+		__entry->agno = agno;
+		__entry->bno = bno;
+		__entry->len = len;
+		__entry->owner = owner;
+	),
+	TP_printk("dev %d:%d keydev %d:%d agno %u bno %llu len %llu owner %lld\n",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  MAJOR(__entry->keydev), MINOR(__entry->keydev),
+		  __entry->agno,
+		  __entry->bno,
+		  __entry->len,
+		  __entry->owner)
+)
+#define DEFINE_FSMAP_EVENT(name) \
+DEFINE_EVENT(ext4_fsmap_class, name, \
+	TP_PROTO(struct super_block *sb, u32 keydev, u32 agno, u64 bno, u64 len, \
+		 u64 owner), \
+	TP_ARGS(sb, keydev, agno, bno, len, owner))
+DEFINE_FSMAP_EVENT(ext4_fsmap_low_key);
+DEFINE_FSMAP_EVENT(ext4_fsmap_high_key);
+DEFINE_FSMAP_EVENT(ext4_fsmap_mapping);
+
+DECLARE_EVENT_CLASS(ext4_getfsmap_class,
+	TP_PROTO(struct super_block *sb, struct ext4_fsmap *fsmap),
+	TP_ARGS(sb, fsmap),
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(dev_t, keydev)
+		__field(u64, block)
+		__field(u64, len)
+		__field(u64, owner)
+		__field(u64, flags)
+	),
+	TP_fast_assign(
+		__entry->dev = sb->s_bdev->bd_dev;
+		__entry->keydev = new_decode_dev(fsmap->fmr_device);
+		__entry->block = fsmap->fmr_physical;
+		__entry->len = fsmap->fmr_length;
+		__entry->owner = fsmap->fmr_owner;
+		__entry->flags = fsmap->fmr_flags;
+	),
+	TP_printk("dev %d:%d keydev %d:%d block %llu len %llu owner %lld flags 0x%llx\n",
+		  MAJOR(__entry->dev), MINOR(__entry->dev),
+		  MAJOR(__entry->keydev), MINOR(__entry->keydev),
+		  __entry->block,
+		  __entry->len,
+		  __entry->owner,
+		  __entry->flags)
+)
+#define DEFINE_GETFSMAP_EVENT(name) \
+DEFINE_EVENT(ext4_getfsmap_class, name, \
+	TP_PROTO(struct super_block *sb, struct ext4_fsmap *fsmap), \
+	TP_ARGS(sb, fsmap))
+DEFINE_GETFSMAP_EVENT(ext4_getfsmap_low_key);
+DEFINE_GETFSMAP_EVENT(ext4_getfsmap_high_key);
+DEFINE_GETFSMAP_EVENT(ext4_getfsmap_mapping);
+
 #endif /* _TRACE_EXT4_H */
 
 /* This part must be outside protection */

mm/page-writeback.c

@@ -2352,10 +2352,16 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
 
 	if (wbc->nr_to_write <= 0)
 		return 0;
+	while (1) {
 		if (mapping->a_ops->writepages)
 			ret = mapping->a_ops->writepages(mapping, wbc);
 		else
 			ret = generic_writepages(mapping, wbc);
+		if ((ret != -ENOMEM) || (wbc->sync_mode != WB_SYNC_ALL))
+			break;
+		cond_resched();
+		congestion_wait(BLK_RW_ASYNC, HZ/50);
+	}
 	return ret;
 }