Subject: X on Diamond Stealth 911/924 (Was: XFree86)
To: None <current-users@NetBSD.ORG>
From: Torsten Duwe (Emacs) <duwe@immd4.informatik.uni-erlangen.de>
List: current-users
Date: 01/18/1995 21:45:22
>>>>> "Rob" == Rob Windsor <windsor@ksu.ksu.edu> writes:

    Rob> Verily did John Evans write:
    >>> I just got a Diamond Stealth card, so I finally got around to
    >>> installing X... after finally making sure all the dot clock values,
    >>> monitor sync rates, etc, etc, etc, etc. were right, we eliminated all

Stealth '911 or '924 ? There is a well-known Problem with Diamond and their
clock generators. Don't supply clock values, let the X server probe for them !

[...]
    Rob> First thing, it doesn't like something in your fontpaths.
    Rob> Trial-and-error removal to find out if there's one specific one that
    Rob> is killing you.

To me it looked like the server already fixed that at run time.

[...]
    Rob> Check your kernel to see if you have the following options set:
    Rob> options XSERVER, UCONSOLE 
    Rob> options LKM,SYSVMSG,SYSVSEM,SYSVSHM

XSERVER: absolutely necessary - he had it in his kernel because a VT was
successfully allocated. UCONSOLE: only if you want a "console" window with
messages in it. SYSVSHM: X will tell you "using MIT SHM extensions" (or was
that the message generated bu doom ?-) anyway - that's what it's good for -
not necessary but will make your local progs run a little faster. The other
options: no influence on X AFAIK.

	Torsten

P.S.: this type of stuff should go to -help and port-i386...