[RFC][PATCH] loginuid through procfs (+ a question)

[Serge Hallyn]

Sounds good.

On this topic, Stephen, were you going to send the loginuid inheritence
patch to lkml?

I guess I will rewrite the netlink patch quickly without
netlink_get_msgtype, send it back here today, and lacking seriously
negative feedback, send it to lkml tomorrow.

thanks,
-serge

On Fri, 2005-01-14 at 10:00 -0500, Stephen Smalley wrote:
> On Thu, 2005-01-13 at 18:00, Steve Grubb wrote:
> > I see places in the kernel setting -1 for uid on init, though. For example, 
> > look at security/keys/keyctl.c
> 
> Ah, good point.  And even the functions in kernel/sys.c for setting uids
> use (uid_t)-1 as a magic value to indicate that no change is desired.
> 
> Hence, possibly (uid_t)-1 is safe to use for the unset loginuid.
>  
-- 
Serge Hallyn <serue at us.ibm.com>