linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Josef Bacik <jbacik@fb.com>
To: <linux-btrfs@vger.kernel.org>
Subject: [PATCH] Btrfs-progs: check blocks when checking fs roots
Date: Thu, 2 Oct 2014 15:11:13 -0400	[thread overview]
Message-ID: <1412277073-9559-1-git-send-email-jbacik@fb.com> (raw)

If check block fails during the extent tree checks we could evict the extent
buffer from cache, so the next time we go to read it for the fs_tree checks we
could miss the fact that its bogus and blow up in strange and interesting ways.
Fix this by running the leaf/node checks on all blocks in the fs root so we know
we're looking at valid blocks.  Thanks,

Signed-off-by: Josef Bacik <jbacik@fb.com>
---
 cmds-check.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/cmds-check.c b/cmds-check.c
index 5b8417c..db779ae 100644
--- a/cmds-check.c
+++ b/cmds-check.c
@@ -1376,6 +1376,8 @@ static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path,
 
 		next = btrfs_find_tree_block(root, bytenr, blocksize);
 		if (!next || !btrfs_buffer_uptodate(next, ptr_gen)) {
+			enum btrfs_tree_block_status status;
+
 			free_extent_buffer(next);
 			reada_walk_down(root, cur, path->slots[*level]);
 			next = read_tree_block(root, bytenr, blocksize,
@@ -1384,6 +1386,16 @@ static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path,
 				err = -EIO;
 				goto out;
 			}
+
+			if (btrfs_is_leaf(next))
+				status = btrfs_check_leaf(root, NULL, next);
+			else
+				status = btrfs_check_node(root, NULL, next);
+			if (status != BTRFS_TREE_BLOCK_CLEAN) {
+				free_extent_buffer(next);
+				err = -EIO;
+				goto out;
+			}
 		}
 
 		*level = *level - 1;
@@ -2078,6 +2090,7 @@ static int check_fs_root(struct btrfs_root *root,
 	struct shared_node root_node;
 	struct root_record *rec;
 	struct btrfs_root_item *root_item = &root->root_item;
+	enum btrfs_tree_block_status status;
 
 	if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
 		rec = get_root_rec(root_cache, root->root_key.objectid);
@@ -2096,6 +2109,14 @@ static int check_fs_root(struct btrfs_root *root,
 	wc->active_node = level;
 	wc->root_level = level;
 
+	/* We may not have checked the root block, lets do that now */
+	if (btrfs_is_leaf(root->node))
+		status = btrfs_check_leaf(root, NULL, root->node);
+	else
+		status = btrfs_check_node(root, NULL, root->node);
+	if (status != BTRFS_TREE_BLOCK_CLEAN)
+		return -EIO;
+
 	if (btrfs_root_refs(root_item) > 0 ||
 	    btrfs_disk_key_objectid(&root_item->drop_progress) == 0) {
 		path.nodes[level] = root->node;
-- 
1.8.3.1


                 reply	other threads:[~2014-10-02 19:11 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1412277073-9559-1-git-send-email-jbacik@fb.com \
    --to=jbacik@fb.com \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).