This is a note to let you know that I've just added the patch titled Revert "af_unix: Fix splice-bind deadlock" to the 4.7-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: revert-af_unix-fix-splice-bind-deadlock.patch and it can be found in the queue-4.7 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From foo@baz Wed Sep 21 10:05:18 CEST 2016 From: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Date: Thu, 1 Sep 2016 14:56:49 -0700 Subject: Revert "af_unix: Fix splice-bind deadlock" From: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> [ Upstream commit 38f7bd94a97b542de86a2be9229289717e33a7a4 ] This reverts commit c845acb324aa85a39650a14e7696982ceea75dc1. It turns out that it just replaces one deadlock with another one: we can still get the wrong lock ordering with the readlock due to overlayfs calling back into the filesystem layer and still taking the vfs locks after the readlock. The proper solution ends up being to just split the readlock into two pieces: the bind lock (taken *outside* the vfs locks) and the IO lock (taken *inside* the filesystem locks). The two locks are independent anyway. Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Reviewed-by: Shmulik Ladkani <shmulik.ladkani@xxxxxxxxx> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- net/unix/af_unix.c | 68 +++++++++++++++++++++-------------------------------- 1 file changed, 27 insertions(+), 41 deletions(-) --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -953,20 +953,32 @@ fail: return NULL; } -static int unix_mknod(struct dentry *dentry, const struct path *path, umode_t mode, - struct path *res) +static int unix_mknod(const char *sun_path, umode_t mode, struct path *res) { - int err; - - err = security_path_mknod(path, dentry, mode, 0); + struct dentry *dentry; + struct path path; + int err = 0; + /* + * Get the parent directory, calculate the hash for last + * component. + */ + dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); + err = PTR_ERR(dentry); + if (IS_ERR(dentry)) + return err; + + /* + * All right, let's create it. + */ + err = security_path_mknod(&path, dentry, mode, 0); if (!err) { - err = vfs_mknod(d_inode(path->dentry), dentry, mode, 0); + err = vfs_mknod(d_inode(path.dentry), dentry, mode, 0); if (!err) { - res->mnt = mntget(path->mnt); + res->mnt = mntget(path.mnt); res->dentry = dget(dentry); } } - + done_path_create(&path, dentry); return err; } @@ -977,12 +989,10 @@ static int unix_bind(struct socket *sock struct unix_sock *u = unix_sk(sk); struct sockaddr_un *sunaddr = (struct sockaddr_un *)uaddr; char *sun_path = sunaddr->sun_path; - int err, name_err; + int err; unsigned int hash; struct unix_address *addr; struct hlist_head *list; - struct path path; - struct dentry *dentry; err = -EINVAL; if (sunaddr->sun_family != AF_UNIX) @@ -998,34 +1008,14 @@ static int unix_bind(struct socket *sock goto out; addr_len = err; - name_err = 0; - dentry = NULL; - if (sun_path[0]) { - /* Get the parent directory, calculate the hash for last - * component. - */ - dentry = kern_path_create(AT_FDCWD, sun_path, &path, 0); - - if (IS_ERR(dentry)) { - /* delay report until after 'already bound' check */ - name_err = PTR_ERR(dentry); - dentry = NULL; - } - } - err = mutex_lock_interruptible(&u->readlock); if (err) - goto out_path; + goto out; err = -EINVAL; if (u->addr) goto out_up; - if (name_err) { - err = name_err == -EEXIST ? -EADDRINUSE : name_err; - goto out_up; - } - err = -ENOMEM; addr = kmalloc(sizeof(*addr)+addr_len, GFP_KERNEL); if (!addr) @@ -1036,11 +1026,11 @@ static int unix_bind(struct socket *sock addr->hash = hash ^ sk->sk_type; atomic_set(&addr->refcnt, 1); - if (dentry) { - struct path u_path; + if (sun_path[0]) { + struct path path; umode_t mode = S_IFSOCK | (SOCK_INODE(sock)->i_mode & ~current_umask()); - err = unix_mknod(dentry, &path, mode, &u_path); + err = unix_mknod(sun_path, mode, &path); if (err) { if (err == -EEXIST) err = -EADDRINUSE; @@ -1048,9 +1038,9 @@ static int unix_bind(struct socket *sock goto out_up; } addr->hash = UNIX_HASH_SIZE; - hash = d_real_inode(dentry)->i_ino & (UNIX_HASH_SIZE - 1); + hash = d_real_inode(path.dentry)->i_ino & (UNIX_HASH_SIZE - 1); spin_lock(&unix_table_lock); - u->path = u_path; + u->path = path; list = &unix_socket_table[hash]; } else { spin_lock(&unix_table_lock); @@ -1073,10 +1063,6 @@ out_unlock: spin_unlock(&unix_table_lock); out_up: mutex_unlock(&u->readlock); -out_path: - if (dentry) - done_path_create(&path, dentry); - out: return err; } Patches currently in stable-queue which might be from torvalds@xxxxxxxxxxxxxxxxxxxx are queue-4.7/revert-af_unix-fix-splice-bind-deadlock.patch queue-4.7/af_unix-split-u-readlock-into-two-iolock-and-bindlock.patch -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html