9p: attach-per-user

The 9P2000 protocol requires the authentication and permission checks to be
done in the file server. For that reason every user that accesses the file
server tree has to authenticate and attach to the server separately.
Multiple users can share the same connection to the server.

Currently v9fs does a single attach and executes all I/O operations as a
single user. This makes using v9fs in multiuser environment unsafe as it
depends on the client doing the permission checking.

This patch improves the 9P2000 support by allowing every user to attach
separately. The patch defines three modes of access (new mount option
'access'):

- attach-per-user (access=user) (default mode for 9P2000.u)
 If a user tries to access a file served by v9fs for the first time, v9fs
 sends an attach command to the server (Tattach) specifying the user. If
 the attach succeeds, the user can access the v9fs tree.
 As there is no uname->uid (string->integer) mapping yet, this mode works
 only with the 9P2000.u dialect.

- allow only one user to access the tree (access=<uid>)
 Only the user with uid can access the v9fs tree. Other users that attempt
 to access it will get EPERM error.

- do all operations as a single user (access=any) (default for 9P2000)
 V9fs does a single attach and all operations are done as a single user.
 If this mode is selected, the v9fs behavior is identical with the current
 one.

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>

1 files changed, 52 insertions, 15 deletions

diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index 68f82be3bf3..89ee0bace41 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -91,6 +91,8 @@ enum {
 	Opt_legacy, Opt_nodevmap,
 	/* Cache options */
 	Opt_cache_loose,
+	/* Access options */
+	Opt_access,
 	/* Error token */
 	Opt_err
 };
@@ -108,6 +110,7 @@ static match_table_t tokens = {
 	{Opt_nodevmap, "nodevmap"},
 	{Opt_cache_loose, "cache=loose"},
 	{Opt_cache_loose, "loose"},
+	{Opt_access, "access=%s"},
 	{Opt_err, NULL}
 };
 
@@ -125,10 +128,10 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
 	char *p;
 	int option;
 	int ret;
+	char *s, *e;
 
 	/* setup defaults */
 	v9ses->maxdata = 8192;
-	v9ses->flags = V9FS_EXTENDED;
 	v9ses->afid = ~0;
 	v9ses->debug = 0;
 	v9ses->cache = 0;
@@ -172,10 +175,10 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
 			v9ses->trans = v9fs_match_trans(&args[0]);
 			break;
 		case Opt_uname:
-			match_strcpy(v9ses->name, &args[0]);
+			match_strcpy(v9ses->uname, &args[0]);
 			break;
 		case Opt_remotename:
-			match_strcpy(v9ses->remotename, &args[0]);
+			match_strcpy(v9ses->aname, &args[0]);
 			break;
 		case Opt_legacy:
 			v9ses->flags &= ~V9FS_EXTENDED;
@@ -186,6 +189,22 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
 		case Opt_cache_loose:
 			v9ses->cache = CACHE_LOOSE;
 			break;
+
+		case Opt_access:
+			s = match_strdup(&args[0]);
+			v9ses->flags &= ~V9FS_ACCESS_MASK;
+			if (strcmp(s, "user") == 0)
+				v9ses->flags |= V9FS_ACCESS_USER;
+			else if (strcmp(s, "any") == 0)
+				v9ses->flags |= V9FS_ACCESS_ANY;
+			else {
+				v9ses->flags |= V9FS_ACCESS_SINGLE;
+				v9ses->uid = simple_strtol(s, &e, 10);
+				if (*e != '\0')
+					v9ses->uid = ~0;
+			}
+			break;
+
 		default:
 			continue;
 		}
@@ -207,21 +226,22 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 	struct p9_trans *trans = NULL;
 	struct p9_fid *fid;
 
-	v9ses->name = __getname();
-	if (!v9ses->name)
+	v9ses->uname = __getname();
+	if (!v9ses->uname)
 		return ERR_PTR(-ENOMEM);
 
-	v9ses->remotename = __getname();
-	if (!v9ses->remotename) {
-		__putname(v9ses->name);
+	v9ses->aname = __getname();
+	if (!v9ses->aname) {
+		__putname(v9ses->uname);
 		return ERR_PTR(-ENOMEM);
 	}
 
-	strcpy(v9ses->name, V9FS_DEFUSER);
-	strcpy(v9ses->remotename, V9FS_DEFANAME);
+	v9ses->flags = V9FS_EXTENDED | V9FS_ACCESS_USER;
+	strcpy(v9ses->uname, V9FS_DEFUSER);
+	strcpy(v9ses->aname, V9FS_DEFANAME);
+	v9ses->uid = ~0;
 	v9ses->dfltuid = V9FS_DEFUID;
 	v9ses->dfltgid = V9FS_DEFGID;
-
 	v9ses->options = kstrdup(data, GFP_KERNEL);
 	v9fs_parse_options(v9ses);
 
@@ -255,8 +275,20 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 		goto error;
 	}
 
-	fid = p9_client_attach(v9ses->clnt, NULL, v9ses->name,
-							v9ses->remotename);
+	if (!v9ses->clnt->dotu)
+		v9ses->flags &= ~V9FS_EXTENDED;
+
+	/* for legacy mode, fall back to V9FS_ACCESS_ANY */
+	if (!v9fs_extended(v9ses) &&
+		((v9ses->flags&V9FS_ACCESS_MASK) == V9FS_ACCESS_USER)) {
+
+		v9ses->flags &= ~V9FS_ACCESS_MASK;
+		v9ses->flags |= V9FS_ACCESS_ANY;
+		v9ses->uid = ~0;
+	}
+
+	fid = p9_client_attach(v9ses->clnt, NULL, v9ses->uname, ~0,
+							v9ses->aname);
 	if (IS_ERR(fid)) {
 		retval = PTR_ERR(fid);
 		fid = NULL;
@@ -264,6 +296,11 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
 		goto error;
 	}
 
+	if ((v9ses->flags & V9FS_ACCESS_MASK) == V9FS_ACCESS_SINGLE)
+		fid->uid = v9ses->uid;
+	else
+		fid->uid = ~0;
+
 	return fid;
 
 error:
@@ -284,8 +321,8 @@ void v9fs_session_close(struct v9fs_session_info *v9ses)
 		v9ses->clnt = NULL;
 	}
 
-	__putname(v9ses->name);
-	__putname(v9ses->remotename);
+	__putname(v9ses->uname);
+	__putname(v9ses->aname);
 	kfree(v9ses->options);
 }