From: Nick Piggin <npiggin@suse.de>
To: Chris Mason <chris.mason@oracle.com>, linux-btrfs@vger.kernel.org
Subject: [patch] btrfs: fix inode rbtree corruption
Date: Tue, 18 Aug 2009 18:45:42 +0200 [thread overview]
Message-ID: <20090818164542.GB30325@wotan.suse.de> (raw)
Hi,
Ran into a problem stress testing my btrfs truncate conversion attempt...
Unfortunately it was an existing btrfs problem. Fortunately I think I
was able to fix it.
Thanks,
Nick
--
btrfs: fix inode rbtree corruption
Node may not be inserted over existing node. This causes inode tree
corruption and I was seeing crashes in inode_tree_del which I can not
reproduce after this patch.
The other way to fix this would be to tie inode lifetime in the rbtree
with inode while not in freeing state. I had a look at this but it is
not so trivial at this point. At least this patch gets things working again.
Signed-off-by: Nick Piggin <npiggin@suse.de>
---
fs/btrfs/inode.c | 21 ++++++++++++++-------
1 file changed, 14 insertions(+), 7 deletions(-)
Index: linux-2.6/fs/btrfs/inode.c
===================================================================
--- linux-2.6.orig/fs/btrfs/inode.c
+++ linux-2.6/fs/btrfs/inode.c
@@ -3099,8 +3099,12 @@ static void inode_tree_add(struct inode
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_inode *entry;
- struct rb_node **p = &root->inode_tree.rb_node;
- struct rb_node *parent = NULL;
+ struct rb_node **p;
+ struct rb_node *parent;
+
+again:
+ p = &root->inode_tree.rb_node;
+ parent = NULL;
spin_lock(&root->inode_lock);
while (*p) {
@@ -3108,13 +3112,16 @@ static void inode_tree_add(struct inode
entry = rb_entry(parent, struct btrfs_inode, rb_node);
if (inode->i_ino < entry->vfs_inode.i_ino)
- p = &(*p)->rb_left;
+ p = &parent->rb_left;
else if (inode->i_ino > entry->vfs_inode.i_ino)
- p = &(*p)->rb_right;
+ p = &parent->rb_right;
else {
WARN_ON(!(entry->vfs_inode.i_state &
(I_WILL_FREE | I_FREEING | I_CLEAR)));
- break;
+ rb_erase(parent, &root->inode_tree);
+ RB_CLEAR_NODE(parent);
+ spin_unlock(&root->inode_lock);
+ goto again;
}
}
rb_link_node(&BTRFS_I(inode)->rb_node, parent, p);
@@ -3126,12 +3133,12 @@ static void inode_tree_del(struct inode
{
struct btrfs_root *root = BTRFS_I(inode)->root;
+ spin_lock(&root->inode_lock);
if (!RB_EMPTY_NODE(&BTRFS_I(inode)->rb_node)) {
- spin_lock(&root->inode_lock);
rb_erase(&BTRFS_I(inode)->rb_node, &root->inode_tree);
- spin_unlock(&root->inode_lock);
RB_CLEAR_NODE(&BTRFS_I(inode)->rb_node);
}
+ spin_unlock(&root->inode_lock);
}
static noinline void init_btrfs_i(struct inode *inode)
next reply other threads:[~2009-08-18 16:45 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-18 16:45 Nick Piggin [this message]
2009-08-18 18:56 ` [patch] btrfs: fix inode rbtree corruption Yan, Zheng
2009-08-18 21:19 ` Jens Axboe
2009-08-19 8:45 ` Nick Piggin
2009-08-19 8:46 ` Jens Axboe
2009-08-19 8:52 ` Nick Piggin
2009-08-19 8:59 ` Jens Axboe
2009-08-20 13:23 ` Nick Piggin
2009-08-20 13:51 ` Yan, Zheng
2009-08-20 22:07 ` Jens Axboe
2009-08-21 0:55 ` Yan, Zheng
2009-08-21 6:20 ` Jens Axboe
2009-08-21 8:06 ` Yan, Zheng
2009-08-21 8:10 ` Jens Axboe
2009-08-19 8:56 ` Yan, Zheng
2009-08-19 9:04 ` Nick Piggin
2009-08-19 9:34 ` Yan, Zheng
2009-08-19 10:47 ` Nick Piggin
2009-08-19 12:00 ` Yan, Zheng
2009-08-19 8:32 ` Nick Piggin
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=20090818164542.GB30325@wotan.suse.de \
--to=npiggin@suse.de \
--cc=chris.mason@oracle.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