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 support for the CSUM_SEED feature which will allow future
  userspace utilities to change the file system's UUID without rewriting
  all of the file system metadata.

  A number of miscellaneous fixes, the most significant of which are in
  the ext4 encryption support.  Anyone wishing to use the encryption
  feature should backport all of the ext4 crypto patches up to 4.4 to
  get fixes to a memory leak and file system corruption bug.

  There are also cleanups in ext4's feature test macros and in ext4's
  sysfs support code"

* tag 'ext4_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (26 commits)
  fs/ext4: remove unnecessary new_valid_dev check
  ext4: fix abs() usage in ext4_mb_check_group_pa
  ext4: do not allow journal_opts for fs w/o journal
  ext4: explicit mount options parsing cleanup
  ext4, jbd2: ensure entering into panic after recording an error in superblock
  [PATCH] fix calculation of meta_bg descriptor backups
  ext4: fix potential use after free in __ext4_journal_stop
  jbd2: fix checkpoint list cleanup
  ext4: fix xfstest generic/269 double revoked buffer bug with bigalloc
  ext4: make the bitmap read routines return real error codes
  jbd2: clean up feature test macros with predicate functions
  ext4: clean up feature test macros with predicate functions
  ext4: call out CRC and corruption errors with specific error codes
  ext4: store checksum seed in superblock
  ext4: reserve code points for the project quota feature
  ext4: promote ext4 over ext2 in the default probe order
  jbd2: gate checksum calculations on crc driver presence, not sb flags
  ext4: use private version of page_zero_new_buffers() for data=journal mode
  ext4 crypto: fix bugs in ext4_encrypted_zeroout()
  ext4 crypto: replace some BUG_ON()'s with error checks
  ...

1 files changed, 61 insertions, 22 deletions

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 612fbcf76b5c..e8d620a484f6 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -378,7 +378,7 @@ static int __check_block_validity(struct inode *inode, const char *func,
 				 "lblock %lu mapped to illegal pblock "
 				 "(length %d)", (unsigned long) map->m_lblk,
 				 map->m_len);
-		return -EIO;
+		return -EFSCORRUPTED;
 	}
 	return 0;
 }
@@ -480,7 +480,7 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode,
 
 	/* We can handle the block number less than EXT_MAX_BLOCKS */
 	if (unlikely(map->m_lblk >= EXT_MAX_BLOCKS))
-		return -EIO;
+		return -EFSCORRUPTED;
 
 	/* Lookup extent status tree firstly */
 	if (ext4_es_lookup_extent(inode, map->m_lblk, &es)) {
@@ -965,7 +965,7 @@ static int ext4_block_write_begin(struct page *page, loff_t pos, unsigned len,
 	if (unlikely(err))
 		page_zero_new_buffers(page, from, to);
 	else if (decrypt)
-		err = ext4_decrypt_one(inode, page);
+		err = ext4_decrypt(page);
 	return err;
 }
 #endif
@@ -1181,6 +1181,38 @@ errout:
 	return ret ? ret : copied;
 }
 
+/*
+ * This is a private version of page_zero_new_buffers() which doesn't
+ * set the buffer to be dirty, since in data=journalled mode we need
+ * to call ext4_handle_dirty_metadata() instead.
+ */
+static void zero_new_buffers(struct page *page, unsigned from, unsigned to)
+{
+	unsigned int block_start = 0, block_end;
+	struct buffer_head *head, *bh;
+
+	bh = head = page_buffers(page);
+	do {
+		block_end = block_start + bh->b_size;
+		if (buffer_new(bh)) {
+			if (block_end > from && block_start < to) {
+				if (!PageUptodate(page)) {
+					unsigned start, size;
+
+					start = max(from, block_start);
+					size = min(to, block_end) - start;
+
+					zero_user(page, start, size);
+					set_buffer_uptodate(bh);
+				}
+				clear_buffer_new(bh);
+			}
+		}
+		block_start = block_end;
+		bh = bh->b_this_page;
+	} while (bh != head);
+}
+
 static int ext4_journalled_write_end(struct file *file,
 				     struct address_space *mapping,
 				     loff_t pos, unsigned len, unsigned copied,
@@ -1207,7 +1239,7 @@ static int ext4_journalled_write_end(struct file *file,
 		if (copied < len) {
 			if (!PageUptodate(page))
 				copied = 0;
-			page_zero_new_buffers(page, from+copied, to);
+			zero_new_buffers(page, from+copied, to);
 		}
 
 		ret = ext4_walk_page_buffers(handle, page_buffers(page), from,
@@ -1815,11 +1847,22 @@ static int ext4_writepage(struct page *page,
 	 * the page. But we may reach here when we do a journal commit via
 	 * journal_submit_inode_data_buffers() and in that case we must write
 	 * allocated buffers to achieve data=ordered mode guarantees.
+	 *
+	 * Also, if there is only one buffer per page (the fs block
+	 * size == the page size), if one buffer needs block
+	 * allocation or needs to modify the extent tree to clear the
+	 * unwritten flag, we know that the page can't be written at
+	 * all, so we might as well refuse the write immediately.
+	 * Unfortunately if the block size != page size, we can't as
+	 * easily detect this case using ext4_walk_page_buffers(), but
+	 * for the extremely common case, this is an optimization that
+	 * skips a useless round trip through ext4_bio_write_page().
 	 */
 	if (ext4_walk_page_buffers(NULL, page_bufs, 0, len, NULL,
 				   ext4_bh_delay_or_unwritten)) {
 		redirty_page_for_writepage(wbc, page);
-		if (current->flags & PF_MEMALLOC) {
+		if ((current->flags & PF_MEMALLOC) ||
+		    (inode->i_sb->s_blocksize == PAGE_CACHE_SIZE)) {
 			/*
 			 * For memory cleaning there's no point in writing only
 			 * some buffers. So just bail out. Warn if we came here
@@ -2599,8 +2642,7 @@ static int ext4_nonda_switch(struct super_block *sb)
 /* We always reserve for an inode update; the superblock could be there too */
 static int ext4_da_write_credits(struct inode *inode, loff_t pos, unsigned len)
 {
-	if (likely(EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
-				EXT4_FEATURE_RO_COMPAT_LARGE_FILE)))
+	if (likely(ext4_has_feature_large_file(inode->i_sb)))
 		return 1;
 
 	if (pos + len <= 0x7fffffffULL)
@@ -3393,7 +3435,7 @@ static int __ext4_block_zero_page_range(handle_t *handle,
 			/* We expect the key to be set. */
 			BUG_ON(!ext4_has_encryption_key(inode));
 			BUG_ON(blocksize != PAGE_CACHE_SIZE);
-			WARN_ON_ONCE(ext4_decrypt_one(inode, page));
+			WARN_ON_ONCE(ext4_decrypt(page));
 		}
 	}
 	if (ext4_should_journal_data(inode)) {
@@ -3820,7 +3862,7 @@ static int __ext4_get_inode_loc(struct inode *inode,
 
 	iloc->bh = NULL;
 	if (!ext4_valid_inum(sb, inode->i_ino))
-		return -EIO;
+		return -EFSCORRUPTED;
 
 	iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb);
 	gdp = ext4_get_group_desc(sb, iloc->block_group, NULL);
@@ -4006,8 +4048,7 @@ static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
 	struct inode *inode = &(ei->vfs_inode);
 	struct super_block *sb = inode->i_sb;
 
-	if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
-				EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
+	if (ext4_has_feature_huge_file(sb)) {
 		/* we are using combined 48 bit field */
 		i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 |
 					le32_to_cpu(raw_inode->i_blocks_lo);
@@ -4068,7 +4109,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 			EXT4_ERROR_INODE(inode, "bad extra_isize (%u != %u)",
 				EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize,
 				EXT4_INODE_SIZE(inode->i_sb));
-			ret = -EIO;
+			ret = -EFSCORRUPTED;
 			goto bad_inode;
 		}
 	} else
@@ -4088,7 +4129,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 
 	if (!ext4_inode_csum_verify(inode, raw_inode, ei)) {
 		EXT4_ERROR_INODE(inode, "checksum invalid");
-		ret = -EIO;
+		ret = -EFSBADCRC;
 		goto bad_inode;
 	}
 
@@ -4130,7 +4171,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 	inode->i_blocks = ext4_inode_blocks(raw_inode, ei);
 	ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo);
-	if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT))
+	if (ext4_has_feature_64bit(sb))
 		ei->i_file_acl |=
 			((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
 	inode->i_size = ext4_isize(raw_inode);
@@ -4203,7 +4244,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	    !ext4_data_block_valid(EXT4_SB(sb), ei->i_file_acl, 1)) {
 		EXT4_ERROR_INODE(inode, "bad extended attribute block %llu",
 				 ei->i_file_acl);
-		ret = -EIO;
+		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	} else if (!ext4_has_inline_data(inode)) {
 		if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
@@ -4254,7 +4295,7 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
 	} else if (ino == EXT4_BOOT_LOADER_INO) {
 		make_bad_inode(inode);
 	} else {
-		ret = -EIO;
+		ret = -EFSCORRUPTED;
 		EXT4_ERROR_INODE(inode, "bogus i_mode (%o)", inode->i_mode);
 		goto bad_inode;
 	}
@@ -4272,7 +4313,7 @@ bad_inode:
 struct inode *ext4_iget_normal(struct super_block *sb, unsigned long ino)
 {
 	if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO)
-		return ERR_PTR(-EIO);
+		return ERR_PTR(-EFSCORRUPTED);
 	return ext4_iget(sb, ino);
 }
 
@@ -4294,7 +4335,7 @@ static int ext4_inode_blocks_set(handle_t *handle,
 		ext4_clear_inode_flag(inode, EXT4_INODE_HUGE_FILE);
 		return 0;
 	}
-	if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE))
+	if (!ext4_has_feature_huge_file(sb))
 		return -EFBIG;
 
 	if (i_blocks <= 0xffffffffffffULL) {
@@ -4455,8 +4496,7 @@ static int ext4_do_update_inode(handle_t *handle,
 		need_datasync = 1;
 	}
 	if (ei->i_disksize > 0x7fffffffULL) {
-		if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
-				EXT4_FEATURE_RO_COMPAT_LARGE_FILE) ||
+		if (!ext4_has_feature_large_file(sb) ||
 				EXT4_SB(sb)->s_es->s_rev_level ==
 		    cpu_to_le32(EXT4_GOOD_OLD_REV))
 			set_large_file = 1;
@@ -4505,8 +4545,7 @@ static int ext4_do_update_inode(handle_t *handle,
 		if (err)
 			goto out_brelse;
 		ext4_update_dynamic_rev(sb);
-		EXT4_SET_RO_COMPAT_FEATURE(sb,
-					   EXT4_FEATURE_RO_COMPAT_LARGE_FILE);
+		ext4_set_feature_large_file(sb);
 		ext4_handle_sync(handle);
 		err = ext4_handle_dirty_super(handle, sb);
 	}