Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Christian Brauner <christian@brauner.io>
To:     linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Cc:     cjwatson@ubuntu.com, ebiederm@xmission.com,
        viro@zeniv.linux.org.uk, serge@hallyn.com,
        Christian Brauner <christian@brauner.io>
Subject: [PATCH 0/1] getxattr: use correct xattr length
Date:   Thu,  7 Jun 2018 13:43:47 +0200
Message-Id: <20180607114348.23667-1-christian@brauner.io>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hey,

When running in a container with a user namespace, if you call getxattr
with name = "system.posix_acl_access" and size % 8 != 4, then getxattr
silently skips the user namespace fixup that it normally does resulting in
un-fixed-up data being returned.
I think that the analysis that this is caused by
posix_acl_fix_xattr_to_user() being passed the total buffer size and not
the actual size of the xattr as returned by vfs_getxattr() is correct.
So this commit passes the actual length of the xattr as returned by
vfs_getxattr() down.

It might be worth considering making posix_acl_fix_xattr_userns() and
posix_acl_fix_xattr_to_user() return errors. Such that e.g.
posix_acl_xattr_count() errors can be caught in getxattr(). If that's
something we want I can send a follow-up patch.

More details of this can be found in
https://bugzilla.kernel.org/show_bug.cgi?id=199945

Thanks!
Christian

Christian Brauner (1):
  getxattr: use correct xattr length

 fs/xattr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

-- 
2.17.0