From: Stephen Smalley <sds@tycho.nsa.gov>
To: David Howells <dhowells@redhat.com>
Cc: linux-unionfs@vger.kernel.org, linux-kernel@vger.kernel.org,
drquigl <drquigl@tycho.nsa.gov>,
linux-security-module@vger.kernel.org, selinux@tycho.nsa.gov,
linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH 5/7] SELinux: Handle opening of a unioned file
Date: Tue, 16 Jun 2015 13:20:42 -0400 [thread overview]
Message-ID: <55805AEA.3050407@tycho.nsa.gov> (raw)
In-Reply-To: <7332.1434473388@warthog.procyon.org.uk>
On 06/16/2015 12:49 PM, David Howells wrote:
> Stephen Smalley <sds@tycho.nsa.gov> wrote:
>
>> It looks like commit 415103f9932d45f7927f4b17e3a9a13834cdb9a1 changed
>> selinux_inode_init_security()'s handling of SECURITY_FS_USE_MNTPOINT,
>> and this change was never propagated to selinux_dentry_init_security().
>> However, that commit also did not update
>> security/selinux/hooks.c:may_create()'s logic for computing the new file
>> label when checking CREATE permission, and therefore introduced a
>> potential inconsistency between the label used for the permission check
>> and the label assigned to the inode.
>>
>> That's why I suggested that we need a common helper for all three to
>> ensure consistency there.
>
> I think a common helper is harder than it seems. We need the parent dir in
> one of the cases the helper has to consider, but finding it is done in three
> different ways, depending on the caller:
>
> (1) dentry_init can just use ->d_parent as there's a lock held that prevents
> it changing (I think). This could use (2) instead, however.
>
> (2) file_open has to use dget_parent().
>
> (3) inode_init doesn't have any dentries, but rather has the object and
> parent inodes.
>
> If we don't mind file_open() always calling dget_parent(), then the common
> helper can take the dir inode.
>
> Also, thinking ahead to the possibility of bringing unionmount into the kernel
> at some point: union non-dir dentries that are not yet copied up have no inode
> attached, but rather fall through to the underlying lower inode in the VFS.
> This, however, gives us nowhere to hang the inode label. How expensive is the
> security_transition_sid() call?
Why are you talking about file_open()? It is may_create() that has the
duplicated logic, which has the parent dir passed to it by its callers
(selinux_inode_create, selinux_inode_symlink, selinux_inode_mkdir,
selinux_mknod). selinux_inode_init_security() also gets passed the
parent dir directly. Only selinux_dentry_init_security() has to use
d_parent, which as you say is safe, so it can just pass the resulting
dir to the helper.
Until a process writes to the file, we just want to use the lower inode
label, right? At the point a process writes to the file and a copy-up
is produced, we could perform a one-time computation, and then once the
upper inode is created, it should get set accordingly. So I wouldn't
think we would need to call security_transition_sid() frequently. If
so, we might want to do what was previously done in the userspace AVC in
libselinux, and start caching security_transition_sid() results in the
AVC itself and add an avc_transition_sid() interface (in the userspace
AVC, this is avc_compute_create()).
WARNING: multiple messages have this Message-ID (diff)
From: Stephen Smalley <sds@tycho.nsa.gov>
To: David Howells <dhowells@redhat.com>
Cc: linux-unionfs@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-fsdevel@vger.kernel.org,
linux-security-module@vger.kernel.org, selinux@tycho.nsa.gov,
drquigl <drquigl@tycho.nsa.gov>
Subject: Re: [PATCH 5/7] SELinux: Handle opening of a unioned file
Date: Tue, 16 Jun 2015 13:20:42 -0400 [thread overview]
Message-ID: <55805AEA.3050407@tycho.nsa.gov> (raw)
In-Reply-To: <7332.1434473388@warthog.procyon.org.uk>
On 06/16/2015 12:49 PM, David Howells wrote:
> Stephen Smalley <sds@tycho.nsa.gov> wrote:
>
>> It looks like commit 415103f9932d45f7927f4b17e3a9a13834cdb9a1 changed
>> selinux_inode_init_security()'s handling of SECURITY_FS_USE_MNTPOINT,
>> and this change was never propagated to selinux_dentry_init_security().
>> However, that commit also did not update
>> security/selinux/hooks.c:may_create()'s logic for computing the new file
>> label when checking CREATE permission, and therefore introduced a
>> potential inconsistency between the label used for the permission check
>> and the label assigned to the inode.
>>
>> That's why I suggested that we need a common helper for all three to
>> ensure consistency there.
>
> I think a common helper is harder than it seems. We need the parent dir in
> one of the cases the helper has to consider, but finding it is done in three
> different ways, depending on the caller:
>
> (1) dentry_init can just use ->d_parent as there's a lock held that prevents
> it changing (I think). This could use (2) instead, however.
>
> (2) file_open has to use dget_parent().
>
> (3) inode_init doesn't have any dentries, but rather has the object and
> parent inodes.
>
> If we don't mind file_open() always calling dget_parent(), then the common
> helper can take the dir inode.
>
> Also, thinking ahead to the possibility of bringing unionmount into the kernel
> at some point: union non-dir dentries that are not yet copied up have no inode
> attached, but rather fall through to the underlying lower inode in the VFS.
> This, however, gives us nowhere to hang the inode label. How expensive is the
> security_transition_sid() call?
Why are you talking about file_open()? It is may_create() that has the
duplicated logic, which has the parent dir passed to it by its callers
(selinux_inode_create, selinux_inode_symlink, selinux_inode_mkdir,
selinux_mknod). selinux_inode_init_security() also gets passed the
parent dir directly. Only selinux_dentry_init_security() has to use
d_parent, which as you say is safe, so it can just pass the resulting
dir to the helper.
Until a process writes to the file, we just want to use the lower inode
label, right? At the point a process writes to the file and a copy-up
is produced, we could perform a one-time computation, and then once the
upper inode is created, it should get set accordingly. So I wouldn't
think we would need to call security_transition_sid() frequently. If
so, we might want to do what was previously done in the userspace AVC in
libselinux, and start caching security_transition_sid() results in the
AVC itself and add an avc_transition_sid() interface (in the userspace
AVC, this is avc_compute_create()).
next prev parent reply other threads:[~2015-06-16 17:21 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-05 15:42 [PATCH 0/7] Security: Provide unioned file support David Howells
2014-11-05 15:42 ` [PATCH 1/7] Security: Provide copy-up security hooks for unioned files David Howells
2014-11-06 17:46 ` Casey Schaufler
2014-11-07 14:49 ` David Howells
2014-11-07 14:49 ` David Howells
2014-11-07 21:22 ` Paul Moore
2014-11-07 21:22 ` Paul Moore
2014-11-07 22:10 ` David Howells
2014-11-07 22:10 ` David Howells
2014-11-10 15:28 ` Paul Moore
2014-11-10 15:28 ` Paul Moore
2014-11-05 15:42 ` [PATCH 2/7] Overlayfs: Use copy-up security hooks David Howells
2014-11-07 21:39 ` Paul Moore
2014-11-07 21:39 ` Paul Moore
2014-11-07 22:05 ` David Howells
2014-11-07 22:05 ` David Howells
2014-11-10 15:45 ` Paul Moore
2014-11-10 15:45 ` Paul Moore
2014-11-05 15:42 ` [PATCH 3/7] SELinux: Stub in copy-up handling David Howells
2014-11-07 21:44 ` Paul Moore
2014-11-07 21:44 ` Paul Moore
2014-11-07 22:08 ` David Howells
2014-11-07 22:08 ` David Howells
2014-11-10 15:47 ` Paul Moore
2014-11-10 15:47 ` Paul Moore
2014-11-05 15:42 ` [PATCH 4/7] Security: Pass the union-layer file path into security_file_open() David Howells
2014-11-05 15:43 ` [PATCH 5/7] SELinux: Handle opening of a unioned file David Howells
2014-11-05 16:35 ` Stephen Smalley
2014-11-06 12:03 ` David Howells
2014-11-06 12:03 ` David Howells
2014-11-06 13:13 ` Stephen Smalley
2014-11-06 13:13 ` Stephen Smalley
2014-11-06 13:34 ` David Howells
2014-11-06 13:34 ` David Howells
2014-11-27 14:15 ` David Howells
2014-11-27 14:15 ` David Howells
2014-11-06 12:27 ` David Howells
2014-11-06 12:27 ` David Howells
2014-11-06 12:27 ` David Howells
2014-11-27 17:25 ` David Howells
2014-11-27 17:25 ` David Howells
2015-06-12 15:30 ` David Howells
2015-06-12 15:30 ` David Howells
2015-06-15 12:57 ` Stephen Smalley
2015-06-15 12:57 ` Stephen Smalley
2015-06-16 9:41 ` David Howells
2015-06-16 9:41 ` David Howells
2015-06-16 16:49 ` David Howells
2015-06-16 16:49 ` David Howells
2015-06-16 17:20 ` Stephen Smalley [this message]
2015-06-16 17:20 ` Stephen Smalley
2015-06-16 21:34 ` David Howells
2015-06-16 21:34 ` David Howells
2015-06-17 14:44 ` Stephen Smalley
2015-06-17 14:44 ` Stephen Smalley
2015-06-18 10:15 ` David Howells
2015-06-18 10:15 ` David Howells
2015-06-18 12:48 ` Stephen Smalley
2015-06-18 12:48 ` Stephen Smalley
2015-06-18 15:26 ` David Howells
2015-06-18 15:26 ` David Howells
2015-06-18 10:32 ` David Howells
2015-06-18 10:32 ` David Howells
2015-06-18 12:16 ` Stephen Smalley
2015-06-18 12:16 ` Stephen Smalley
2014-11-05 15:43 ` [PATCH 6/7] SELinux: The copy-up operation must have read permission on the lower file David Howells
2014-11-05 16:43 ` Stephen Smalley
2014-11-05 17:54 ` Stephen Smalley
2014-11-06 13:39 ` Stephen Smalley
2014-11-27 14:17 ` David Howells
2014-11-27 14:17 ` David Howells
2014-11-27 14:21 ` David Howells
2014-11-27 14:21 ` David Howells
2014-11-27 14:21 ` David Howells
2014-11-05 15:43 ` [PATCH 7/7] SELinux: Check against union and lower labels for file ops on lower files David Howells
2014-11-06 17:35 ` [PATCH 0/7] Security: Provide unioned file support Casey Schaufler
2014-11-06 17:35 ` Casey Schaufler
2014-11-06 17:58 ` David Howells
2014-11-06 17:58 ` David Howells
2014-11-06 18:40 ` Casey Schaufler
2014-11-06 18:40 ` Casey Schaufler
2014-11-07 15:21 ` David Howells
2014-11-07 15:21 ` David Howells
2014-11-07 18:54 ` Daniel J Walsh
2014-11-07 18:54 ` Daniel J Walsh
2014-11-09 1:31 ` Casey Schaufler
2014-11-09 1:31 ` Casey Schaufler
2014-11-10 13:59 ` Daniel J Walsh
2014-11-10 13:59 ` Daniel J Walsh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=55805AEA.3050407@tycho.nsa.gov \
--to=sds@tycho.nsa.gov \
--cc=dhowells@redhat.com \
--cc=drquigl@tycho.nsa.gov \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-security-module@vger.kernel.org \
--cc=linux-unionfs@vger.kernel.org \
--cc=selinux@tycho.nsa.gov \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.