[lkml]   [2004]   [Apr]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Subject[CHECKER] warnings in fs/ext3/namei.c (2.4.19) where disk read errors get ignored, causing non-empty dir to be deleted

We checked EXT3 filesystem on 2.4.19 recently and found 2 cases that look
like bugs. For both of the cases, disk read errors are ignored, which
appears to cause a non-empty directory to be wrongly deleted or a dir to
contain more than one entries with identical names.

I'm not sure if they are real bugs or not, so your confirmations
/clarifications are appericated.

Please let me know if anything isn't clear

all warnings are in file fs/ext3/namei.c

[BUG] A non-empty dir may be deleted because ext3_read errors are ignored
by ext3_find_entry. empty_dir is called whenenver ext3_rmdir tries to
remove a directory.

static int empty_dir (struct inode * inode)
bh = ext3_bread (NULL, inode,
offset >> EXT3_BLOCK_SIZE_BITS(sb), 0, &err);
if (!bh) {
#if 0
ext3_error (sb, "empty_dir",
"directory #%lu contains a hole at offset %lu",
inode->i_ino, offset);
offset += sb->s_blocksize;
ERROR ---> continue;
de = (struct ext3_dir_entry_2 *) bh->b_data;

[BUG] A dir may end up containing more than one entries with identical
names because because disk read errors are ignored by ext3_find_entry.
ext3_find_entry is called by lots of other ext3 functions (ext3_add_entry,
ext3_unlink, ext3_rename)

static struct buffer_head * ext3_find_entry (struct dentry *dentry,
struct ext3_dir_entry_2 ** res_dir)
if ((bh = bh_use[ra_ptr++]) == NULL)
goto next;
if (!buffer_uptodate(bh)) {
/* read error, skip block & hope for the best */
ERROR ---> goto next;

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 14:02    [W:0.027 / U:1.480 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site