On 7/27/19 4:43 PM, Ken Moffat via blfs-dev wrote:
On Sat, Jul 27, 2019 at 04:09:30PM -0500, Bruce Dubbs via blfs-dev wrote:
On 7/27/19 3:52 PM, Ken Moffat via blfs-dev wrote:
On Sat, Jul 27, 2019 at 07:55:57PM +0000, DJ Lucas via blfs-dev wrote:
This is useful information, but I'm still a long way from building
an up to date system, so for the moment I can't say if dropping the
suid setting will work on any of my machines.
You don't have to build anything to test. You only need to add or remove
the suid bit on $XORG_PREFIX/libexec/Xorg abd run startx.
My most recent BLFS builds are nominally dated as 22nd April, i.e.
before the introduction of elogind - will give it a go later (ati,
intel), and perhaps also on an 8.4 where I switched to modesetting
from ati). But I suspect all of those might be too old, otherwise
why did we not drop enable-suid sooner ?
My understanding is that elogind is supposed to eliminate the need for
Xorg to have the suid bit set. I don't think having that set is a big
deal, but some people do.
One of the problems is whether or not a display manager like kdm is in
use. In that case, the DM comes up on tty1 by default and then when the
desktop environment comes up, it still uses tty1.
When we use startx, the system starts Xorg on the first tty that is
unused. Since we have agetty running on tty[1-6] by default, it starts
on tty7 which is
crw--w---- 1 root tty 4, 7 Jul 11 11:50 /dev/tty7
Something has to give a regular user permissions to rw tty7 and I think
elogind is supposed to do that. But it does not work for me unless Xorg
has the suid bit set.
Whether we need to pursue this or not is an open question.
-- Bruce
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
