linux-kernel - Re: [PATCH -V7 5/9] vfs: Add freadlink syscall

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20100513114351.793caf70@notabene.brown>
Date:	Thu, 13 May 2010 11:43:51 +1000
From:	Neil Brown <neilb@...e.de>
To:	"Aneesh Kumar K.V" <aneesh.kumar@...ux.vnet.ibm.com>
Cc:	hch@...radead.org, viro@...iv.linux.org.uk, adilger@....com,
	corbet@....net, serue@...ibm.com, linux-fsdevel@...r.kernel.org,
	sfrench@...ibm.com, philippe.deniel@....FR,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH -V7 5/9] vfs: Add freadlink syscall

On Wed, 12 May 2010 21:20:40 +0530
"Aneesh Kumar K.V" <aneesh.kumar@...ux.vnet.ibm.com> wrote:

> This enables to use open-by-handle and then get the link target
> details of a symlink using the fd returned by handle
> 

I find it very frustrating that a new syscall seems to be needed here.
We have 'readlinkat', and it should be enough.
How:  the 'dfd' has to be a 'directory', and the path name as to be non-empty.

The following patch allows 'path' to be NULL and in that case 'dfd' to be a
non-directory.  This allows readlinkat and faccessat (and probably others)
to be used on an fd with not following path name.

What do people think of this alternative?

NeilBrown

From: NeilBrown <neilb@...e.de>

Allow '*at' syscalls to have a NULL path name, thus using the 'fd' directly.

Move the 'is a directory' test from path_init to the start of link_path_walk
(which is always called after a successful path_init).
Remove the file_permission(MAY_EXEC) test from path_init as such a test
already exists in link_path_walk as exec_permission()
Allow user_path_at to use a NULL resulting in the dfd being the target object.

This effectively provides an 'f*' version for all syscalls with a '*at'
version. In particular: link, access, chmod, chown, stat, readlink, utimes

Some of those already have f* versions, but link, access, readlink all
benefit directly, and futimes code can be tidied up.

NULL is only allowed if 'dfd' is not ATFD_CWD to reduce possible
backward-compatibility issues.

openat is not effected by this change, even in non-O_CREAT instances.  Maybe
it should.

Signed-off-by: NeilBrown <neilb@...e.de>

diff --git a/fs/namei.c b/fs/namei.c
index 48e60a1..b9b091e 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -824,6 +824,11 @@ static int link_path_walk(const char *name, struct nameidata *nd)
 		goto return_reval;

 	inode = nd->path.dentry->d_inode;
+
+	err = -ENOTDIR;
+	if (!S_ISDIR(inode->i_mode))
+		goto return_err;
+
 	if (nd->depth)
 		lookup_flags = LOOKUP_FOLLOW | (nd->flags & LOOKUP_CONTINUE);

@@ -1030,14 +1035,6 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct namei

 		dentry = file->f_path.dentry;

-		retval = -ENOTDIR;
-		if (!S_ISDIR(dentry->d_inode->i_mode))
-			goto fput_fail;
-
-		retval = file_permission(file, MAY_EXEC);
-		if (retval)
-			goto fput_fail;
-
 		nd->path = file->f_path;
 		path_get(&file->f_path);

@@ -1241,17 +1238,22 @@ int user_path_at(int dfd, const char __user *name, unsigned flags,
 		 struct path *path)
 {
 	struct nameidata nd;
-	char *tmp = getname(name);
-	int err = PTR_ERR(tmp);
-	if (!IS_ERR(tmp)) {
-
-		BUG_ON(flags & LOOKUP_PARENT);
+	char *tmp;
+	int err;

-		err = do_path_lookup(dfd, tmp, flags, &nd);
-		putname(tmp);
-		if (!err)
-			*path = nd.path;
+	BUG_ON(flags & LOOKUP_PARENT);
+	if (name == NULL && dfd >= 0)
+		err = do_path_lookup(dfd, "", flags, &nd);
+	else {
+		tmp = getname(name);
+		err = PTR_ERR(tmp);
+		if (!IS_ERR(tmp)) {
+			err = do_path_lookup(dfd, tmp, flags, &nd);
+			putname(tmp);
+		}
 	}
+	if (!err)
+		*path = nd.path;
 	return err;
 }

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/