Re: [RFC v3 02/10] Revert "lsm: Remove the socket_post_accept() hook"

Paul Moore wrote:

On Tuesday, May 03, 2011 10:28:24 PM Tetsuo Handa wrote:

quoted

Paul Moore wrote:

quoted

On Tuesday, May 03, 2011 10:24:15 AM Samir Bellabes wrote:

quoted

snet needs to reintroduce this hook, as it was designed to be: a hook
for updating security informations on objects.

Looking at this and 5/10 again, it seems that you should be able to do
what you need with the sock_graft() hook.  Am I missing something?

My apologies if we've already discussed this approach previously ...

Second problem is that genlmsg_unicast() might return -EAGAIN because we
can't sleep inside write_lock_bh()/write_unlock_bh().

Ah yes, the real problem.  I forgot that snet relied on a user space tool.  I 
tend to agree with others who have suggested this is not the right approach, 
but I understand why you want the post_accept() hook; thanks for reminding me.

However, it sounds that Samir says genlmsg_unicast() failure is not fatal.

Samir Bellabes wrote:

using snet_do_send_event() means that system is sending data to
userspace. the system is not waiting for a verdict from userspace.

If error occurs, we actually loose the information data.
I may be able to write a solution which try to send the data again, but
we need a exit solution for this loop (a number of try ?).

If genlmsg_unicast() failure is not fatal, snet doesn't need the
socket_post_accept hook. Samir, is genlmsg_unicast() failure fatal for snet?
(Although, I'd like to ask for revival of the hook for TOMOYO anyway.)