The unix_dgram routines add secdata to socket messages,
but the unix_stream routines do not. I have added the
two missing lines of code.
Signed-off-by: Pat Kane <redacted>
---
net/unix/af_unix.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
From: David Miller <davem@davemloft.net> Date: 2011-03-23 02:31:29
From: Pat Kane <redacted>
Date: Tue, 22 Mar 2011 19:38:37 -0500
The unix_dgram routines add secdata to socket messages,
but the unix_stream routines do not. I have added the
two missing lines of code.
Signed-off-by: Pat Kane <redacted>
The security hooks appear to be only intended to operate on datagram
sockets, and as such I think the omission of UNIX stream sockets was
very much on purpose.
The SELINUX hook implementations even have "_dgram()" in their names.
Catherine Zhang added to CC: as she last made modifications to these
hooks.
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Eric Paris <eparis@redhat.com> Date: 2011-03-23 15:57:37
On Tue, 2011-03-22 at 19:32 -0700, David Miller wrote:
From: Pat Kane <redacted>
Date: Tue, 22 Mar 2011 19:38:37 -0500
quoted
The unix_dgram routines add secdata to socket messages,
but the unix_stream routines do not. I have added the
two missing lines of code.
Signed-off-by: Pat Kane <redacted>
The security hooks appear to be only intended to operate on datagram
sockets, and as such I think the omission of UNIX stream sockets was
very much on purpose.
The SELINUX hook implementations even have "_dgram()" in their names.
Catherine Zhang added to CC: as she last made modifications to these
hooks.
And I'll add Paul Moore as I think he understands the intersection
of /net and /security better than anyone.
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
On Tue, 2011-03-22 at 19:32 -0700, David Miller wrote:
quoted
From: Pat Kane <redacted>
Date: Tue, 22 Mar 2011 19:38:37 -0500
quoted
The unix_dgram routines add secdata to socket messages,
but the unix_stream routines do not. I have added the
two missing lines of code.
Signed-off-by: Pat Kane <redacted>
The security hooks appear to be only intended to operate on datagram
sockets, and as such I think the omission of UNIX stream sockets was
very much on purpose.
The SELINUX hook implementations even have "_dgram()" in their names.
Catherine Zhang added to CC: as she last made modifications to these
hooks.
And I'll add Paul Moore as I think he understands the intersection
of /net and /security better than anyone.
Paul is definitely the man on this. I've also added the LSM list,
as while SELinux is the only current user of secdata that may not
always be the case.
--
1.7.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
>> On Tue, 2011-03-22 at 19:32 -0700, David Miller wrote:
>>> The SELINUX hook implementations even have "_dgram()" in their names.
The LSM hook that I am having problems with, and that the patch fixes
is "secid_to_secctx()" not "socket_getpeersec_dgram()".
Pat
---
From: Paul Moore <hidden> Date: 2011-03-23 22:16:51
On Wednesday, March 23, 2011 4:23:35 PM Pat Kane wrote:
>> On Tue, 2011-03-22 at 19:32 -0700, David Miller wrote:
>>> The SELINUX hook implementations even have "_dgram()" in their names.
The LSM hook that I am having problems with, and that the patch fixes
is "secid_to_secctx()" not "socket_getpeersec_dgram()".
Can you explain the problem you are having? I'm specifically interested in
who is calling secid_to_secctx() on a AF_UNIX stream packet.
Adding the secid token to a AF_UNIX stream packet isn't likely to be the right
solution, although until we hear what the core problem is, I'm not sure we can
offer any fixes.
--
paul moore
linux @ hp