public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Al Viro <viro@ZenIV.linux.org.uk>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Chris Mason <chris.mason@oracle.com>,
	Casey Schaufler <casey@schaufler-ca.com>
Subject: Re: Linux 3.2-rc6
Date: Sat, 17 Dec 2011 08:49:44 +0000	[thread overview]
Message-ID: <20111217084944.GV2203@ZenIV.linux.org.uk> (raw)
In-Reply-To: <CA+55aFzd6K2szJD5MLmy3Stcai9oAR=tRXZdQtKgVWBzDr5bdg@mail.gmail.com>

On Fri, Dec 16, 2011 at 06:53:17PM -0800, Linus Torvalds wrote:

> Casey Schaufler (1):
>       BTRFS: Establish i_ops before calling d_instantiate

I'm sorry, but that's nowhere near enough.  As I said in followup to
the original posting of that patch, it still leaves a race window
where inode is already associated with dentry but still doesn't have
->a_ops, etc. set up.  Hell, ->symlink() doesn't write the symlink
body until after that point...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 329a5c6..d9501f5 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -4590,10 +4590,6 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans,
 	int err = btrfs_add_link(trans, dir, inode,
 				 dentry->d_name.name, dentry->d_name.len,
 				 backref, index);
-	if (!err) {
-		d_instantiate(dentry, inode);
-		return 0;
-	}
 	if (err > 0)
 		err = -EEXIST;
 	return err;
@@ -4655,6 +4651,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry,
 	else {
 		init_special_inode(inode, inode->i_mode, rdev);
 		btrfs_update_inode(trans, root, inode);
+		d_instantiate(dentry, inode);
 	}
 out_unlock:
 	nr = trans->blocks_used;
@@ -4722,6 +4719,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry,
 		inode->i_mapping->a_ops = &btrfs_aops;
 		inode->i_mapping->backing_dev_info = &root->fs_info->bdi;
 		BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
+		d_instantiate(dentry, inode);
 	}
 out_unlock:
 	nr = trans->blocks_used;
@@ -4780,6 +4778,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir,
 		err = btrfs_update_inode(trans, root, inode);
 		BUG_ON(err);
 		btrfs_log_new_name(trans, inode, NULL, parent);
+		d_instantiate(dentry, inode);
 	}
 
 	nr = trans->blocks_used;
@@ -7244,6 +7243,8 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry,
 		drop_inode = 1;
 
 out_unlock:
+	if (!err)
+		d_instantiate(dentry, inode);
 	nr = trans->blocks_used;
 	btrfs_end_transaction_throttle(trans, root);
 	if (drop_inode) {

  parent reply	other threads:[~2011-12-17  8:49 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-12-17  2:53 Linux 3.2-rc6 Linus Torvalds
2011-12-17  7:29 ` Stephen Rothwell
2011-12-17  8:49 ` Al Viro [this message]
2011-12-17 15:36   ` Chris Mason

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=20111217084944.GV2203@ZenIV.linux.org.uk \
    --to=viro@zeniv.linux.org.uk \
    --cc=casey@schaufler-ca.com \
    --cc=chris.mason@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=torvalds@linux-foundation.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