[419] in linux-security and linux-alert archive
Re: console security (was Re: problem wi
daemon@ATHENA.MIT.EDU (Zygo Blaxell)
Sat Oct 14 11:04:10 1995
From: Zygo Blaxell <zblaxell@calum.csclub.uwaterloo.ca>
To: ats@hurd193.wustl.edu (Alan Shutko)
Date: Sat, 14 Oct 1995 06:04:14 -0400 (EDT)
Cc: tytso@MIT.EDU, RDMiller@legislate.com, kjh@seas.smu.edu,
owner-linux-security@tarsier.cv.nrao.edu,
linux-security@tarsier.cv.nrao.edu
In-Reply-To: <199510120326.WAA28922@hubert.wustl.edu> from "Alan Shutko" at Oct 11, 95 10:26:12 pm
Quoted from Alan Shutko:
>
> >>>>> "Zygo" == Zygo Blaxell <zblaxell@calum.csclub.uwaterloo.ca> writes:
>
> Zygo> A neat side-benefit of having SAK blow away XFree86 with extreme
> Zygo> prejudice is that when you exit X window sessions you don't
>
> Quick question: I have set up SAK after hearing this discussion.
> However, it doesn't work on X or SVGA apps like executor-svga. Any
> ideas?
There's a kernel patch posted to some linux-*@vger list a while ago that
does the SAK handling at a lower level than all the other keysyms. If
you can't find it in the standard mailing list archives, I might be able
to still find it in mine...it's been a while since I needed the patches,
having changed jobs and security policies since then.
> Zygo> On most hardware putting more functionality in the kernel would
> Zygo> be the most elegant thing to do; however, with an Intel-based
> Zygo> platform with N+1 video cards available, the kernel will only
> Zygo> ever have as many as N video drivers...
>
> That's a clear case for defining a solid video card interface for the
> kernel and writing lots of kernel modules for specific cards. SVGAlib
> and XFree86 both have to do lots of work to support lots of cards; by
> making them kernel modules we could drastically reduce it. Do the
> work once rather than twice.
Sigh...perhaps it's time for drivers/vga...
