Commit 6b791bcc authored by Tao Ma's avatar Tao Ma Committed by Joel Becker

ocfs2: Adjust rightmost path in ocfs2_add_branch.

In ocfs2_add_branch, we use the rightmost rec of the leaf extent block
to generate the e_cpos for the newly added branch. In the most case, it
is OK but if the parent extent block's rightmost rec covers more clusters
than the leaf does, it will cause kernel panic if we insert some clusters
in it. The message is something like:
(7445,1):ocfs2_insert_at_leaf:3775 ERROR: bug expression:
le16_to_cpu(el->l_next_free_rec) >= le16_to_cpu(el->l_count)
(7445,1):ocfs2_insert_at_leaf:3775 ERROR: inode 66053, depth 0, count 28,
next free 28, rec.cpos 270, rec.clusters 1, insert.cpos 275, insert.clusters 1
 [<fa7ad565>] ? ocfs2_do_insert_extent+0xb58/0xda0 [ocfs2]
 [<fa7b08f2>] ? ocfs2_insert_extent+0x5bd/0x6ba [ocfs2]
 [<fa7b1b8b>] ? ocfs2_add_clusters_in_btree+0x37f/0x564 [ocfs2]
...

The panic can be easily reproduced by the following small test case
(with bs=512, cs=4K, and I remove all the error handling so that it looks
clear enough for reading).

int main(int argc, char **argv)
{
	int fd, i;
	char buf[5] = "test";

	fd = open(argv[1], O_RDWR|O_CREAT);

	for (i = 0; i < 30; i++) {
		lseek(fd, 40960 * i, SEEK_SET);
		write(fd, buf, 5);
	}

	ftruncate(fd, 1146880);

	lseek(fd, 1126400, SEEK_SET);
	write(fd, buf, 5);

	close(fd);

	return 0;
}

The reason of the panic is that:
the 30 writes and the ftruncate makes the file's extent list looks like:

	Tree Depth: 1   Count: 19   Next Free Rec: 1
	## Offset        Clusters       Block#
	0  0             280            86183
	SubAlloc Bit: 7   SubAlloc Slot: 0
	Blknum: 86183   Next Leaf: 0
	CRC32: 00000000   ECC: 0000
	Tree Depth: 0   Count: 28   Next Free Rec: 28
	## Offset        Clusters       Block#          Flags
	0  0             1              143368          0x0
	1  10            1              143376          0x0
	...
	26 260           1              143576          0x0
	27 270           1              143584          0x0

Now another write at 1126400(275 cluster) whiich will write at the gap
between 271 and 280 will trigger ocfs2_add_branch, but the result after
the function looks like:
	Tree Depth: 1   Count: 19   Next Free Rec: 2
	## Offset        Clusters       Block#
	0  0             280            86183
	1  271           0             143592
So the extent record is intersected and make the following operation bug out.

This patch just try to remove the gap before we add the new branch, so that
the root(branch) rightmost rec will cover the same right position. So in the
above case, before adding branch the tree will be changed to
	Tree Depth: 1   Count: 19   Next Free Rec: 1
	## Offset        Clusters       Block#
	0  0             271            86183
	SubAlloc Bit: 7   SubAlloc Slot: 0
	Blknum: 86183   Next Leaf: 0
	CRC32: 00000000   ECC: 0000
	Tree Depth: 0   Count: 28   Next Free Rec: 28
	## Offset        Clusters       Block#          Flags
	0  0             1              143368          0x0
	1  10            1              143376          0x0
	...
	26 260           1              143576          0x0
	27 270           1              143584          0x0
And after branch add, the tree looks like
	Tree Depth: 1   Count: 19   Next Free Rec: 2
	## Offset        Clusters       Block#
	0  0             271            86183
	1  271           0             143592
Signed-off-by: default avatarTao Ma <tao.ma@oracle.com>
Acked-by: default avatarMark Fasheh <mfasheh@suse.com>
Signed-off-by: default avatarJoel Becker <joel.becker@oracle.com>
parent e04cc15f
...@@ -475,6 +475,12 @@ struct ocfs2_path { ...@@ -475,6 +475,12 @@ struct ocfs2_path {
#define path_leaf_el(_path) ((_path)->p_node[(_path)->p_tree_depth].el) #define path_leaf_el(_path) ((_path)->p_node[(_path)->p_tree_depth].el)
#define path_num_items(_path) ((_path)->p_tree_depth + 1) #define path_num_items(_path) ((_path)->p_tree_depth + 1)
static int ocfs2_find_path(struct inode *inode, struct ocfs2_path *path,
u32 cpos);
static void ocfs2_adjust_rightmost_records(struct inode *inode,
handle_t *handle,
struct ocfs2_path *path,
struct ocfs2_extent_rec *insert_rec);
/* /*
* Reset the actual path elements so that we can re-use the structure * Reset the actual path elements so that we can re-use the structure
* to build another path. Generally, this involves freeing the buffer * to build another path. Generally, this involves freeing the buffer
...@@ -1012,6 +1018,54 @@ static inline u32 ocfs2_sum_rightmost_rec(struct ocfs2_extent_list *el) ...@@ -1012,6 +1018,54 @@ static inline u32 ocfs2_sum_rightmost_rec(struct ocfs2_extent_list *el)
ocfs2_rec_clusters(el, &el->l_recs[i]); ocfs2_rec_clusters(el, &el->l_recs[i]);
} }
/*
* Change range of the branches in the right most path according to the leaf
* extent block's rightmost record.
*/
static int ocfs2_adjust_rightmost_branch(handle_t *handle,
struct inode *inode,
struct ocfs2_extent_tree *et)
{
int status;
struct ocfs2_path *path = NULL;
struct ocfs2_extent_list *el;
struct ocfs2_extent_rec *rec;
path = ocfs2_new_path_from_et(et);
if (!path) {
status = -ENOMEM;
return status;
}
status = ocfs2_find_path(inode, path, UINT_MAX);
if (status < 0) {
mlog_errno(status);
goto out;
}
status = ocfs2_extend_trans(handle, path_num_items(path) +
handle->h_buffer_credits);
if (status < 0) {
mlog_errno(status);
goto out;
}
status = ocfs2_journal_access_path(inode, handle, path);
if (status < 0) {
mlog_errno(status);
goto out;
}
el = path_leaf_el(path);
rec = &el->l_recs[le32_to_cpu(el->l_next_free_rec) - 1];
ocfs2_adjust_rightmost_records(inode, handle, path, rec);
out:
ocfs2_free_path(path);
return status;
}
/* /*
* Add an entire tree branch to our inode. eb_bh is the extent block * Add an entire tree branch to our inode. eb_bh is the extent block
* to start at, if we don't want to start the branch at the dinode * to start at, if we don't want to start the branch at the dinode
...@@ -1038,7 +1092,7 @@ static int ocfs2_add_branch(struct ocfs2_super *osb, ...@@ -1038,7 +1092,7 @@ static int ocfs2_add_branch(struct ocfs2_super *osb,
struct ocfs2_extent_block *eb; struct ocfs2_extent_block *eb;
struct ocfs2_extent_list *eb_el; struct ocfs2_extent_list *eb_el;
struct ocfs2_extent_list *el; struct ocfs2_extent_list *el;
u32 new_cpos; u32 new_cpos, root_end;
mlog_entry_void(); mlog_entry_void();
...@@ -1055,6 +1109,27 @@ static int ocfs2_add_branch(struct ocfs2_super *osb, ...@@ -1055,6 +1109,27 @@ static int ocfs2_add_branch(struct ocfs2_super *osb,
new_blocks = le16_to_cpu(el->l_tree_depth); new_blocks = le16_to_cpu(el->l_tree_depth);
eb = (struct ocfs2_extent_block *)(*last_eb_bh)->b_data;
new_cpos = ocfs2_sum_rightmost_rec(&eb->h_list);
root_end = ocfs2_sum_rightmost_rec(et->et_root_el);
/*
* If there is a gap before the root end and the real end
* of the righmost leaf block, we need to remove the gap
* between new_cpos and root_end first so that the tree
* is consistent after we add a new branch(it will start
* from new_cpos).
*/
if (root_end > new_cpos) {
mlog(0, "adjust the cluster end from %u to %u\n",
root_end, new_cpos);
status = ocfs2_adjust_rightmost_branch(handle, inode, et);
if (status) {
mlog_errno(status);
goto bail;
}
}
/* allocate the number of new eb blocks we need */ /* allocate the number of new eb blocks we need */
new_eb_bhs = kcalloc(new_blocks, sizeof(struct buffer_head *), new_eb_bhs = kcalloc(new_blocks, sizeof(struct buffer_head *),
GFP_KERNEL); GFP_KERNEL);
...@@ -1071,9 +1146,6 @@ static int ocfs2_add_branch(struct ocfs2_super *osb, ...@@ -1071,9 +1146,6 @@ static int ocfs2_add_branch(struct ocfs2_super *osb,
goto bail; goto bail;
} }
eb = (struct ocfs2_extent_block *)(*last_eb_bh)->b_data;
new_cpos = ocfs2_sum_rightmost_rec(&eb->h_list);
/* Note: new_eb_bhs[new_blocks - 1] is the guy which will be /* Note: new_eb_bhs[new_blocks - 1] is the guy which will be
* linked with the rest of the tree. * linked with the rest of the tree.
* conversly, new_eb_bhs[0] is the new bottommost leaf. * conversly, new_eb_bhs[0] is the new bottommost leaf.
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment