Hi Andrew,
looking at the commit messages it seems the patch fixing overflow of
rec_len in ext3 directories when using 64KB blocksize got lost (I'm not
surprised given the churn of the patches in this area for 2.6.24-rc1 ;).
Anyway, the patch is attached. It should make it into 2.6.24 (or nasty
things would happen if somebody actually try 64KB blocksize in ext3).
Thanks
Honza
--
Jan Kara <[email protected]>
SUSE Labs, CR
---
With 64KB blocksize, a directory entry can have size 64KB which does not fit
into 16 bits we have for entry lenght. So we store 0xffff instead and convert
value when read from / written to disk. The patch also converts some places
to use ext3_next_entry() when we are changing them anyway.
Signed-off-by: Jan Kara <[email protected]>
diff -rupX /home/jack/.kerndiffexclude linux-2.6.24-rc1/fs/ext3/dir.c linux-2.6.24-rc1-1-ext3_64k_blocksize/fs/ext3/dir.c
--- linux-2.6.24-rc1/fs/ext3/dir.c 2007-10-24 20:43:56.000000000 +0200
+++ linux-2.6.24-rc1-1-ext3_64k_blocksize/fs/ext3/dir.c 2007-10-24 20:45:01.000000000 +0200
@@ -67,7 +67,7 @@ int ext3_check_dir_entry (const char * f
unsigned long offset)
{
const char * error_msg = NULL;
- const int rlen = le16_to_cpu(de->rec_len);
+ const int rlen = ext3_rec_len_from_disk(de->rec_len);
if (rlen < EXT3_DIR_REC_LEN(1))
error_msg = "rec_len is smaller than minimal";
@@ -173,10 +173,10 @@ revalidate:
* least that it is non-zero. A
* failure will be detected in the
* dirent test below. */
- if (le16_to_cpu(de->rec_len) <
+ if (ext3_rec_len_from_disk(de->rec_len) <
EXT3_DIR_REC_LEN(1))
break;
- i += le16_to_cpu(de->rec_len);
+ i += ext3_rec_len_from_disk(de->rec_len);
}
offset = i;
filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1))
@@ -197,7 +197,7 @@ revalidate:
ret = stored;
goto out;
}
- offset += le16_to_cpu(de->rec_len);
+ offset += ext3_rec_len_from_disk(de->rec_len);
if (le32_to_cpu(de->inode)) {
/* We might block in the next section
* if the data destination is
@@ -219,7 +219,7 @@ revalidate:
goto revalidate;
stored ++;
}
- filp->f_pos += le16_to_cpu(de->rec_len);
+ filp->f_pos += ext3_rec_len_from_disk(de->rec_len);
}
offset = 0;
brelse (bh);
diff -rupX /home/jack/.kerndiffexclude linux-2.6.24-rc1/fs/ext3/namei.c linux-2.6.24-rc1-1-ext3_64k_blocksize/fs/ext3/namei.c
--- linux-2.6.24-rc1/fs/ext3/namei.c 2007-10-24 20:43:56.000000000 +0200
+++ linux-2.6.24-rc1-1-ext3_64k_blocksize/fs/ext3/namei.c 2007-10-24 20:45:58.000000000 +0200
@@ -177,6 +177,15 @@ static int ext3_dx_add_entry(handle_t *h
struct inode *inode);
/*
+ * p is at least 6 bytes before the end of page
+ */
+static inline struct ext3_dir_entry_2 *ext3_next_entry(struct ext3_dir_entry_2 *p)
+{
+ return (struct ext3_dir_entry_2 *)((char*)p +
+ ext3_rec_len_from_disk(p->rec_len));
+}
+
+/*
* Future: use high four bits of block for coalesce-on-delete flags
* Mask them off for now.
*/
@@ -280,7 +289,7 @@ static struct stats dx_show_leaf(struct
space += EXT3_DIR_REC_LEN(de->name_len);
names++;
}
- de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
+ de = ext3_next_entry(de);
}
printk("(%i)\n", names);
return (struct stats) { names, space, 1 };
@@ -547,14 +556,6 @@ static int ext3_htree_next_block(struct
/*
- * p is at least 6 bytes before the end of page
- */
-static inline struct ext3_dir_entry_2 *ext3_next_entry(struct ext3_dir_entry_2 *p)
-{
- return (struct ext3_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len));
-}
-
-/*
* This function fills a red-black tree with information from a
* directory block. It returns the number directory entries loaded
* into the tree. If there is an error it is returned in err.
@@ -720,7 +721,7 @@ static int dx_make_map (struct ext3_dir_
cond_resched();
}
/* XXX: do we need to check rec_len == 0 case? -Chris */
- de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
+ de = ext3_next_entry(de);
}
return count;
}
@@ -822,7 +823,7 @@ static inline int search_dirblock(struct
return 1;
}
/* prevent looping on a bad block */
- de_len = le16_to_cpu(de->rec_len);
+ de_len = ext3_rec_len_from_disk(de->rec_len);
if (de_len <= 0)
return -1;
offset += de_len;
@@ -1130,7 +1131,7 @@ dx_move_dirents(char *from, char *to, st
rec_len = EXT3_DIR_REC_LEN(de->name_len);
memcpy (to, de, rec_len);
((struct ext3_dir_entry_2 *) to)->rec_len =
- cpu_to_le16(rec_len);
+ ext3_rec_len_to_disk(rec_len);
de->inode = 0;
map++;
to += rec_len;
@@ -1149,13 +1150,12 @@ static struct ext3_dir_entry_2* dx_pack_
prev = to = de;
while ((char*)de < base + size) {
- next = (struct ext3_dir_entry_2 *) ((char *) de +
- le16_to_cpu(de->rec_len));
+ next = ext3_next_entry(de);
if (de->inode && de->name_len) {
rec_len = EXT3_DIR_REC_LEN(de->name_len);
if (de > to)
memmove(to, de, rec_len);
- to->rec_len = cpu_to_le16(rec_len);
+ to->rec_len = ext3_rec_len_to_disk(rec_len);
prev = to;
to = (struct ext3_dir_entry_2 *) (((char *) to) + rec_len);
}
@@ -1229,8 +1229,8 @@ static struct ext3_dir_entry_2 *do_split
/* Fancy dance to stay within two buffers */
de2 = dx_move_dirents(data1, data2, map + split, count - split);
de = dx_pack_dirents(data1,blocksize);
- de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de);
- de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2);
+ de->rec_len = ext3_rec_len_to_disk(data1 + blocksize - (char *) de);
+ de2->rec_len = ext3_rec_len_to_disk(data2 + blocksize - (char *) de2);
dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data1, blocksize, 1));
dxtrace(dx_show_leaf (hinfo, (struct ext3_dir_entry_2 *) data2, blocksize, 1));
@@ -1300,7 +1300,7 @@ static int add_dirent_to_buf(handle_t *h
return -EEXIST;
}
nlen = EXT3_DIR_REC_LEN(de->name_len);
- rlen = le16_to_cpu(de->rec_len);
+ rlen = ext3_rec_len_from_disk(de->rec_len);
if ((de->inode? rlen - nlen: rlen) >= reclen)
break;
de = (struct ext3_dir_entry_2 *)((char *)de + rlen);
@@ -1319,11 +1319,11 @@ static int add_dirent_to_buf(handle_t *h
/* By now the buffer is marked for journaling */
nlen = EXT3_DIR_REC_LEN(de->name_len);
- rlen = le16_to_cpu(de->rec_len);
+ rlen = ext3_rec_len_from_disk(de->rec_len);
if (de->inode) {
struct ext3_dir_entry_2 *de1 = (struct ext3_dir_entry_2 *)((char *)de + nlen);
- de1->rec_len = cpu_to_le16(rlen - nlen);
- de->rec_len = cpu_to_le16(nlen);
+ de1->rec_len = ext3_rec_len_to_disk(rlen - nlen);
+ de->rec_len = ext3_rec_len_to_disk(nlen);
de = de1;
}
de->file_type = EXT3_FT_UNKNOWN;
@@ -1400,17 +1400,18 @@ static int make_indexed_dir(handle_t *ha
/* The 0th block becomes the root, move the dirents out */
fde = &root->dotdot;
- de = (struct ext3_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len));
+ de = (struct ext3_dir_entry_2 *)((char *)fde +
+ ext3_rec_len_from_disk(fde->rec_len));
len = ((char *) root) + blocksize - (char *) de;
memcpy (data1, de, len);
de = (struct ext3_dir_entry_2 *) data1;
top = data1 + len;
- while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top)
+ while ((char *)(de2 = ext3_next_entry(de)) < top)
de = de2;
- de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de);
+ de->rec_len = ext3_rec_len_to_disk(data1 + blocksize - (char *) de);
/* Initialize the root; the dot dirents already exist */
de = (struct ext3_dir_entry_2 *) (&root->dotdot);
- de->rec_len = cpu_to_le16(blocksize - EXT3_DIR_REC_LEN(2));
+ de->rec_len = ext3_rec_len_to_disk(blocksize - EXT3_DIR_REC_LEN(2));
memset (&root->info, 0, sizeof(root->info));
root->info.info_length = sizeof(root->info);
root->info.hash_version = EXT3_SB(dir->i_sb)->s_def_hash_version;
@@ -1490,7 +1491,7 @@ static int ext3_add_entry (handle_t *han
return retval;
de = (struct ext3_dir_entry_2 *) bh->b_data;
de->inode = 0;
- de->rec_len = cpu_to_le16(blocksize);
+ de->rec_len = ext3_rec_len_to_disk(blocksize);
return add_dirent_to_buf(handle, dentry, inode, de, bh);
}
@@ -1553,7 +1554,7 @@ static int ext3_dx_add_entry(handle_t *h
goto cleanup;
node2 = (struct dx_node *)(bh2->b_data);
entries2 = node2->entries;
- node2->fake.rec_len = cpu_to_le16(sb->s_blocksize);
+ node2->fake.rec_len = ext3_rec_len_to_disk(sb->s_blocksize);
node2->fake.inode = 0;
BUFFER_TRACE(frame->bh, "get_write_access");
err = ext3_journal_get_write_access(handle, frame->bh);
@@ -1651,9 +1652,9 @@ static int ext3_delete_entry (handle_t *
BUFFER_TRACE(bh, "get_write_access");
ext3_journal_get_write_access(handle, bh);
if (pde)
- pde->rec_len =
- cpu_to_le16(le16_to_cpu(pde->rec_len) +
- le16_to_cpu(de->rec_len));
+ pde->rec_len = ext3_rec_len_to_disk(
+ ext3_rec_len_from_disk(pde->rec_len) +
+ ext3_rec_len_from_disk(de->rec_len));
else
de->inode = 0;
dir->i_version++;
@@ -1661,10 +1662,9 @@ static int ext3_delete_entry (handle_t *
ext3_journal_dirty_metadata(handle, bh);
return 0;
}
- i += le16_to_cpu(de->rec_len);
+ i += ext3_rec_len_from_disk(de->rec_len);
pde = de;
- de = (struct ext3_dir_entry_2 *)
- ((char *) de + le16_to_cpu(de->rec_len));
+ de = ext3_next_entry(de);
}
return -ENOENT;
}
@@ -1798,13 +1798,12 @@ retry:
de = (struct ext3_dir_entry_2 *) dir_block->b_data;
de->inode = cpu_to_le32(inode->i_ino);
de->name_len = 1;
- de->rec_len = cpu_to_le16(EXT3_DIR_REC_LEN(de->name_len));
+ de->rec_len = ext3_rec_len_to_disk(EXT3_DIR_REC_LEN(de->name_len));
strcpy (de->name, ".");
ext3_set_de_type(dir->i_sb, de, S_IFDIR);
- de = (struct ext3_dir_entry_2 *)
- ((char *) de + le16_to_cpu(de->rec_len));
+ de = ext3_next_entry(de);
de->inode = cpu_to_le32(dir->i_ino);
- de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT3_DIR_REC_LEN(1));
+ de->rec_len = ext3_rec_len_to_disk(inode->i_sb->s_blocksize-EXT3_DIR_REC_LEN(1));
de->name_len = 2;
strcpy (de->name, "..");
ext3_set_de_type(dir->i_sb, de, S_IFDIR);
@@ -1856,8 +1855,7 @@ static int empty_dir (struct inode * ino
return 1;
}
de = (struct ext3_dir_entry_2 *) bh->b_data;
- de1 = (struct ext3_dir_entry_2 *)
- ((char *) de + le16_to_cpu(de->rec_len));
+ de1 = ext3_next_entry(de);
if (le32_to_cpu(de->inode) != inode->i_ino ||
!le32_to_cpu(de1->inode) ||
strcmp (".", de->name) ||
@@ -1868,9 +1866,9 @@ static int empty_dir (struct inode * ino
brelse (bh);
return 1;
}
- offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len);
- de = (struct ext3_dir_entry_2 *)
- ((char *) de1 + le16_to_cpu(de1->rec_len));
+ offset = ext3_rec_len_from_disk(de->rec_len) +
+ ext3_rec_len_from_disk(de1->rec_len);
+ de = ext3_next_entry(de1);
while (offset < inode->i_size ) {
if (!bh ||
(void *) de >= (void *) (bh->b_data+sb->s_blocksize)) {
@@ -1899,9 +1897,8 @@ static int empty_dir (struct inode * ino
brelse (bh);
return 0;
}
- offset += le16_to_cpu(de->rec_len);
- de = (struct ext3_dir_entry_2 *)
- ((char *) de + le16_to_cpu(de->rec_len));
+ offset += ext3_rec_len_from_disk(de->rec_len);
+ de = ext3_next_entry(de);
}
brelse (bh);
return 1;
@@ -2255,8 +2252,7 @@ retry:
}
#define PARENT_INO(buffer) \
- ((struct ext3_dir_entry_2 *) ((char *) buffer + \
- le16_to_cpu(((struct ext3_dir_entry_2 *) buffer)->rec_len)))->inode
+ (ext3_next_entry((struct ext3_dir_entry_2 *)(buffer))->inode)
/*
* Anybody can rename anything with this: the permission checks are left to the
diff -rupX /home/jack/.kerndiffexclude linux-2.6.24-rc1/include/linux/ext3_fs.h linux-2.6.24-rc1-1-ext3_64k_blocksize/include/linux/ext3_fs.h
--- linux-2.6.24-rc1/include/linux/ext3_fs.h 2007-10-24 20:44:03.000000000 +0200
+++ linux-2.6.24-rc1-1-ext3_64k_blocksize/include/linux/ext3_fs.h 2007-10-24 20:45:01.000000000 +0200
@@ -656,6 +656,26 @@ struct ext3_dir_entry_2 {
#define EXT3_DIR_ROUND (EXT3_DIR_PAD - 1)
#define EXT3_DIR_REC_LEN(name_len) (((name_len) + 8 + EXT3_DIR_ROUND) & \
~EXT3_DIR_ROUND)
+#define EXT3_MAX_REC_LEN ((1<<16)-1)
+
+static inline unsigned ext3_rec_len_from_disk(__le16 dlen)
+{
+ unsigned len = le16_to_cpu(dlen);
+
+ if (len == EXT3_MAX_REC_LEN)
+ return 1 << 16;
+ return len;
+}
+
+static inline __le16 ext3_rec_len_to_disk(unsigned len)
+{
+ if (len == (1 << 16))
+ return cpu_to_le16(EXT3_MAX_REC_LEN);
+ else if (len > (1 << 16))
+ BUG();
+ return cpu_to_le16(len);
+}
+
/*
* Hash Tree Directory indexing
* (c) Daniel Phillips, 2001