[PATCH v4 22/40] open: handle idmapped mounts

From: Christian Brauner <christian.brauner@ubuntu.com>
To: Alexander Viro <viro@zeniv.linux.org.uk>,
	Christoph Hellwig <hch@infradead.org>,
	linux-fsdevel@vger.kernel.org
Cc: Andy Lutomirski <luto@kernel.org>,
	Mimi Zohar <zohar@linux.ibm.com>,
	James Bottomley <James.Bottomley@hansenpartnership.com>,
	Andreas Dilger <adilger.kernel@dilger.ca>,
	containers@lists.linux-foundation.org,
	Christoph Hellwig <hch@lst.de>, Tycho Andersen <tycho@tycho.ws>,
	Paul Moore <paul@paul-moore.com>,
	Jonathan Corbet <corbet@lwn.net>,
	smbarber@chromium.org, linux-ext4@vger.kernel.org,
	Mrunal Patel <mpatel@redhat.com>,
	Kees Cook <keescook@chromium.org>, Arnd Bergmann <arnd@arndb.de>,
	selinux@vger.kernel.org, Josh Triplett <josh@joshtriplett.org>,
	Seth Forshee <seth.forshee@canonical.com>,
	Aleksa Sarai <cyphar@cyphar.com>,
	Lennart Poettering <lennart@poettering.net>,
	OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>,
	Geoffrey Thomas <geofft@ldpreload.com>,
	David Howells <dhowells@redhat.com>,
	John Johansen <john.johansen@canonical.com>,
	Theodore Tso <tytso@mit.edu>,
	Dmitry Kasatkin <dmitry.kasatkin@gmail.com>,
	Stephen Smalley <stephen.smalley.work@gmail.com>,
	linux-security-module@vger.kernel.org,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	linux-api@vger.kernel.org,
	Casey Schaufler <casey@schaufler-ca.com>,
	Alban Crequy <alban@kinvolk.io>,
	linux-integrity@vger.kernel.org, Todd Kjos <tkjos@google.com>
Subject: [PATCH v4 22/40] open: handle idmapped mounts
Date: Fri,  4 Dec 2020 00:57:18 +0100	[thread overview]
Message-ID: <20201203235736.3528991-23-christian.brauner@ubuntu.com> (raw)
In-Reply-To: <20201203235736.3528991-1-christian.brauner@ubuntu.com>

For core file operations such as changing directories or chrooting,
determining file access, changing mode or ownership the vfs will verify
that the caller is privileged over the inode. Extend the various helpers
to handle idmapped mounts. If the inode is accessed through an idmapped
mount it is mapped according to the mount's user namespace. Afterwards
the permissions checks are identical to non-idmapped mounts. When
changing file ownership we need to map the uid and gid from the mount's
user namespace. If the initial user namespace is passed nothing changes
so non-idmapped mounts will see identical behavior as before.

Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
---
/* v2 */
unchanged

/* v3 */
- David Howells <dhowells@redhat.com>:
  - Remove mnt_idmapped() check after removing mnt_idmapped() helper in earlier
    patches.

/* v4 */
- Serge Hallyn <serge@hallyn.com>:
  - Use "mnt_userns" to refer to a vfsmount's userns everywhere to make
    terminology consistent.
---
 fs/open.c | 29 ++++++++++++++++++++++-------
 1 file changed, 22 insertions(+), 7 deletions(-)

diff --git a/fs/open.c b/fs/open.c
index 648477aa7e0b..c928bc2dab0f 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -401,6 +401,7 @@ static const struct cred *access_override_creds(void)
 
 static long do_faccessat(int dfd, const char __user *filename, int mode, int flags)
 {
+	struct user_namespace *mnt_userns;
 	struct path path;
 	struct inode *inode;
 	int res;
@@ -441,7 +442,8 @@ static long do_faccessat(int dfd, const char __user *filename, int mode, int fla
 			goto out_path_release;
 	}
 
-	res = inode_permission(&init_user_ns, inode, mode | MAY_ACCESS);
+	mnt_userns = mnt_user_ns(path.mnt);
+	res = inode_permission(mnt_userns, inode, mode | MAY_ACCESS);
 	/* SuS v2 requires we report a read only fs too */
 	if (res || !(mode & S_IWOTH) || special_file(inode->i_mode))
 		goto out_path_release;
@@ -489,6 +491,7 @@ SYSCALL_DEFINE2(access, const char __user *, filename, int, mode)
 
 SYSCALL_DEFINE1(chdir, const char __user *, filename)
 {
+	struct user_namespace *mnt_userns;
 	struct path path;
 	int error;
 	unsigned int lookup_flags = LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
@@ -497,7 +500,8 @@ SYSCALL_DEFINE1(chdir, const char __user *, filename)
 	if (error)
 		goto out;
 
-	error = inode_permission(&init_user_ns, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
+	mnt_userns = mnt_user_ns(path.mnt);
+	error = inode_permission(mnt_userns, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
 	if (error)
 		goto dput_and_out;
 
@@ -515,6 +519,7 @@ SYSCALL_DEFINE1(chdir, const char __user *, filename)
 
 SYSCALL_DEFINE1(fchdir, unsigned int, fd)
 {
+	struct user_namespace *mnt_userns;
 	struct fd f = fdget_raw(fd);
 	int error;
 
@@ -526,7 +531,8 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd)
 	if (!d_can_lookup(f.file->f_path.dentry))
 		goto out_putf;
 
-	error = inode_permission(&init_user_ns, file_inode(f.file), MAY_EXEC | MAY_CHDIR);
+	mnt_userns = mnt_user_ns(f.file->f_path.mnt);
+	error = inode_permission(mnt_userns, file_inode(f.file), MAY_EXEC | MAY_CHDIR);
 	if (!error)
 		set_fs_pwd(current->fs, &f.file->f_path);
 out_putf:
@@ -537,6 +543,7 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd)
 
 SYSCALL_DEFINE1(chroot, const char __user *, filename)
 {
+	struct user_namespace *mnt_userns;
 	struct path path;
 	int error;
 	unsigned int lookup_flags = LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
@@ -545,7 +552,8 @@ SYSCALL_DEFINE1(chroot, const char __user *, filename)
 	if (error)
 		goto out;
 
-	error = inode_permission(&init_user_ns, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
+	mnt_userns = mnt_user_ns(path.mnt);
+	error = inode_permission(mnt_userns, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
 	if (error)
 		goto dput_and_out;
 
@@ -570,6 +578,7 @@ SYSCALL_DEFINE1(chroot, const char __user *, filename)
 
 int chmod_common(const struct path *path, umode_t mode)
 {
+	struct user_namespace *mnt_userns;
 	struct inode *inode = path->dentry->d_inode;
 	struct inode *delegated_inode = NULL;
 	struct iattr newattrs;
@@ -585,7 +594,8 @@ int chmod_common(const struct path *path, umode_t mode)
 		goto out_unlock;
 	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
 	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
-	error = notify_change(&init_user_ns, path->dentry, &newattrs, &delegated_inode);
+	mnt_userns = mnt_user_ns(path->mnt);
+	error = notify_change(mnt_userns, path->dentry, &newattrs, &delegated_inode);
 out_unlock:
 	inode_unlock(inode);
 	if (delegated_inode) {
@@ -646,6 +656,7 @@ SYSCALL_DEFINE2(chmod, const char __user *, filename, umode_t, mode)
 
 int chown_common(const struct path *path, uid_t user, gid_t group)
 {
+	struct user_namespace *mnt_userns;
 	struct inode *inode = path->dentry->d_inode;
 	struct inode *delegated_inode = NULL;
 	int error;
@@ -656,6 +667,10 @@ int chown_common(const struct path *path, uid_t user, gid_t group)
 	uid = make_kuid(current_user_ns(), user);
 	gid = make_kgid(current_user_ns(), group);
 
+	mnt_userns = mnt_user_ns(path->mnt);
+	uid = kuid_from_mnt(mnt_userns, uid);
+	gid = kgid_from_mnt(mnt_userns, gid);
+
 retry_deleg:
 	newattrs.ia_valid =  ATTR_CTIME;
 	if (user != (uid_t) -1) {
@@ -676,7 +691,7 @@ int chown_common(const struct path *path, uid_t user, gid_t group)
 	inode_lock(inode);
 	error = security_path_chown(path, uid, gid);
 	if (!error)
-		error = notify_change(&init_user_ns, path->dentry, &newattrs, &delegated_inode);
+		error = notify_change(mnt_userns, path->dentry, &newattrs, &delegated_inode);
 	inode_unlock(inode);
 	if (delegated_inode) {
 		error = break_deleg_wait(&delegated_inode);
@@ -1133,7 +1148,7 @@ struct file *filp_open(const char *filename, int flags, umode_t mode)
 {
 	struct filename *name = getname_kernel(filename);
 	struct file *file = ERR_CAST(name);
-	
+
 	if (!IS_ERR(name)) {
 		file = file_open_name(name, flags, mode);
 		putname(name);
-- 
2.29.2

_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/containers
