Sun-Spots Digest, v6n279
William LeFebvre
Sun-Spots-Request at Rice.edu
Mon Oct 31 13:29:26 AEST 1988
SUN-SPOTS DIGEST Sunday, 30 October 1988 Volume 6 : Issue 279
Today's Topics:
Re: Dvi previewer to run under suntools
Re: FPU problems in Sun 4/110
Re: RBG monitor for my 150G; power line blues
Re: Remove booting a Sun-3/280 -- failure!
Re: TTY Windows in Sunview
GKS for X windows
Shift operations on a Sun 4
More on duping 3-1/2" disks on 386i
386i rlogin to sun3 problem
Quota system questions
emergency bootables?
Algebraic Calctool with Trig Functions?
single user while dumping?
Send contributions to: sun-spots at rice.edu
Send subscription add/delete requests to: sun-spots-request at rice.edu
Bitnet readers can subscribe directly with the CMS command:
TELL LISTSERV AT RICE SUBSCRIBE SUNSPOTS My Full Name
Recent backissues are available via anonymous FTP from "titan.rice.edu".
For volume X, issue Y, "get sun-spots/vXnY". They are also accessible
through the archive server: mail the request "send sun-spots vXnY" to
"archive-server at rice.edu" or mail the word "help" to the same address
for more information.
----------------------------------------------------------------------
Date: Tue, 25 Oct 88 15:45:09 -0700
From: greg at aerospace.aero.org
Subject: Re: Dvi previewer to run under suntools
A number of people have posted questions about a good dvi previewer for
SunView. The one we use is called "texsun" and it comes both as part of
the X.V11R2 user contributed distribution (it is hidden away with "texx",
just fix the makefile to also make texsun), and with the "TeX"
distribution from one of the TeX archives.
One really nice thing about texsun is that it uses "gf" format fonts,
which come with most TeX distributions. It also has a pretty nice mouse
interface (although it is not fancy).
snailmail: Greg Sassenrath ARPA: greg at aerospace.aero.org
M4-927
The Aerospace Corp.
2350 E. El Segundo Blvd.
El Segundo, CA 90245
(213)336-7305
------------------------------
Date: Tue, 25 Oct 88 15:40:06 EDT
From: russ at baklava.mitre.org (Russell Leighton)
Subject: Re: FPU problems in Sun 4/110
Earlier I posted a query about the disclaimer included in the README
documents that came with my 4/110. The disclaimer reads:
__________
FPU Option Deficiency
The following information applies only if you have an FPU equipped 4100
CPU board, and only if the board or FPU upgrade was ordered before October
1st 1988.
There are several deficient performance factors in FPU chips shipped prior
to October 1988. These factors will cause the chip to under- perform its
design specification during heavy "number crunching" or benchmarking
processes. There should be no indication of deficient performance during
normal operation of the system.
__________
Since the posting I received a few calls from Sun. Turns out that there is
not really a big problem. The FPU bug is with the transcedental functions
and results in a small slow down. Otherwise, things are fine and the
4/110 should perform at roughly 60-70% of a 4/280.
Russ.
ARPA: russ%yummy at gateway.mitre.org
Russell Leighton
M.S. Z406
MITRE Signal Processing Lab
7525 Colshire Dr.
McLean, Va. 22102
USA
------------------------------
Date: Tue, 25 Oct 88 14:40:43 CDT
From: cthombor at luke.d.umn.edu (Clark Thomborson)
Subject: Re: RBG monitor for my 150G; power line blues
I recently posted an item here, asking for info on RBG monitors to upgrade
a 3/150G. In the meantime, my 3/50 died, the victim of a power glitch.
Morals: 1) get yourself a line protector (any suggestions?) to avoid a
$1500 motherboard repair bill; 2) ask your system administrator to rework
your local mailer so that you can still receive mail if/when your home
machine dies.
I'm still interested in the RBG conversion, so please resend your info to
cthombor at gw.d.umn.edu. I received just one reply before the crash, from B
Wong, who suggests I order a monitor from Sun's spare parts price list.
Unfortunately, even with an educational discount, the prices on that list
seem inordinately high.
------------------------------
Date: Tue, 25 Oct 88 11:14:17 MDT
From: woods at handies.ucar.edu (Greg Woods)
Subject: Re: Remove booting a Sun-3/280 -- failure!
I have encountered (and beat my head against the wall over) this exact
problem on several occasions. What it comes down to is: the procedure
documented in the manual for booting tapeless workstations DOES NOT WORK
(at least, not on a class A or B network; Sun seems to think the whole
world is class C; ever try to run setup(8) with class B addressing?) The
only way to get it done is to reconfigure the server and set up the
tapeless workstation as a diskless client of that server. If your tape
server is not configured as a file server, you're truly hosed. If it is,
you could simply have your tapeless workstation pretend to be one of the
clients of the tape server (edit /etc/ethers appropriately on the tape
server). I don't know if this problem is present under 4.0 or not; I have
encountered it on Sun 4/280's running SYS4-3.2 and on Sun 3/280's with
3/50's with SCSI disks as tapeless clients running OS 3.5
--Greg
------------------------------
Date: Tue, 25 Oct 88 11:03:20 EDT
From: Ellery Chan <ellery at trantor.harris-atd.com>
Subject: Re: TTY Windows in Sunview
Paul R. Jordan writes:
> ...Is there a hard limitation on the number of TTY windows in one
> application ?...
The different SunView window objects use up file descriptors, of which
your program has a finite number (probably ~30). There is a handy table
on p.54 of the SunView Programmer's Guide that says:
Window Type/Package FD Usage
------------------- --------
Frame 1
Canvas 1
Textsw 3 + (2) { () = temporary }
TTYsw 2
Scrollbar 0
Window Manager (1)
Unix 3 (stdin/stdout/stderr)
Frame buffer 1
Selection Service 3 + (1)
You might just be running out of file descriptors. <sys/param.h> defines
the constant NOFILE as the max. open files per process. In a pinch,
closing std{in,out,err} would get you a couple back.
Since a TTY subwindow uses 2 descriptors, it makes sense that the number
of subwindows created by your test program decreases by 2*(#TTY's). But I
dunno why the descriptors are not being reclaimed when the TTYsw is
destroyed.
Ellery Y. Chan | Phone : (407) 729-3364
Advanced Technology Department | ARPA : ellery at trantor.harris-atd.com
Harris Corporation (GSS) | USmail: PO Box 37, MS 3A-1912
Melbourne, FL 32901 | Melbourne, FL 32902
------------------------------
Date: Tue, 25 Oct 88 07:41:00 CDT
From: sun!shell!jody at decwrl.dec.com (Jody Winston)
Subject: GKS for X windows
Reference: V6n252
> Is there a public domain or commercially available product which allows a
> program to make GKS calls in X-Windows version 11?
This is from: grogers at m.cs.uiuc.edu
===== Software Availability Announcement =====
The department of Computer Science and the Graphics Workstation
Research Group would like to announce the availability of the XGKS
graphics library. At this time it is available only to educational
institutions and only for use for internal research and instructional
purposes.
XGKS is a full implementation of the ANSI standard Graphical Kernel
System for use within the X Window System. XGKS provides for two
dimensional graphics using ANSI standards within the generally accepted
standard X window system. This software has been in use for
approximately nine months within the C.S. department for both research
and instructional purposes and has been rigorously tested.
Highlights:
* Full support for ANSI Standard GKS (level 2C)
* Support for color and monochrome displays
* Runs within the X Window System
* Support for multiple windows and displays
* Efficient implementation
* Rigorously tested
* Hides complexity of window system programming
Output Primitives:
* Polylines (control over color, width, and style)
* Markers (color, size, style)
* Text (fully transformable, color, 9 font styles)
* Polygons (application defined color patterns and colors)
* Cell Arrays (transformable pixel arrays)
Input Devices:
* Locator (tracking cross, rubberband line or box)
* Choice (menus or function buttons)
* Stroke (point streams)
* Valuator (range indicators)
* Pick (selection of output primitives)
* String (keyboard input)
XGKS supports all six logical input devices through the use of a mouse
and keyboard.
Storage:
* Segments (runtime storage of output pickable primitives)
* Metafiles (longterm file storage of graphical objects)
In addition to the standard segment facility of GKS, XGKS provides
storage of non-segment output primitives. This frees the application
from the complexity of window system management.
XGKS has been thoroughly tested on an IBM RT running Berkeley Unix, and
has been ported to the Encore Multmax and Sun workstation. This
software is available immediately from the department of Computer Science.
For additional information contact:
Greg Rogers
University of Illinois at Urbana-Champaign
Department of Computer Science
1304 W. Springfield Ave.
Urbana, IL 61801
(217) 333-6174
UUCP: {pur-ee,convex,inhp4}!uiucdcs!grogers
ARPA: grogers at cs.uiuc.edu
CSNET: grogers%uiuc at csnet-relay
I'd like to thank everyone who sent me this information.
Jody Winston jody at shell.uucp
...!{sun,psuvax1,bcm,rice,decwrl,cs.utexas.edu}!shell!jody
Shell Development Company, Bellaire Research Center
P.O. Box 481, Room 2202
Houston, TX 77001 (Voice: 713 663-2050)
------------------------------
Date: Tue, 25 Oct 88 10:14:09 CDT
From: execu!mike at cs.utexas.edu (mike mccants)
Subject: Shift operations on a Sun 4
The compilers translate i << j, i >> j, lshift(i,j), rshift(i,j) to RISC
instructions on the Sun4. Unlike a Sun3's 68020 CPU, the RISC CPU treats
the shift count modulo 32. In other words, i << 32 is no shift at all!
(68020 gives zero.)
[[ K&R, page 189, section 7.5: "The shift operators...group
left-to-right... The result is undefined if the right operand is
negative, or greater than or equal to the length of the object in bits."
In this case, the length of the left operand is 32. Any C program that
depends on i << 32 being anything other than garbage is a buggy program.
--wnl ]]
------------------------------
Date: Tue, 25 Oct 88 10:23:19 EDT
From: gfr%wolfgang at gateway.mitre.org (Glenn Roberts)
Subject: More on duping 3-1/2" disks on 386i
Reference: v6n272
I previously suggested that you could use cp to duplicate 3-1/2" disks on
roadrunners. Some of the disks I backed up this way failed when I tried
to use them. The procedure I ended up using was:
fdformat
dd if=/dev/rfd0c of=temp bs=512 count=2880
dd if=temp of=/dev/rfd0c bs=512 count=2880
which does a block-for-block copy.
This works but is SLOW (10 minutes per dd step or 20 minutes to backup a
disk, plus even more time if you have to fdformat the disk first!). I
wonder of there is a disk 'spin up' time that is affecting this?
Perfmeter seems to indicate 30 seconds between disk i/o's !!
Perhaps by the time this appears in Sun-Spots someone will have suggested
a faster way to to this.
- Glenn Roberts, MITRE Corporation, McLean VA
gfr%wolfgang at gateway.mitre.org
------------------------------
Date: Tue, 25 Oct 88 12:55:56 PDT
From: teg at orc.olivetti.com (Tom Griner)
Subject: 386i rlogin to sun3 problem
We get the error "rcmd: socket: Permission denied" when trying to rsh or
rlogin from our 386i's to our Sun3's. We have been running a few 386i's
as diskless clients of a 3/280 file server for a few months now. Sun does
not support this configuration (yet), but I managed to get it to work by
rewriting all of the rc scripts in etc and rearranging the 386i's file
system to work in the SunOS 4.0 heterogenius file arrangement. The
3/280's hosts.equiv and .rhosts contain proper entries, so the problem is
most probably something with the 386i. I have removed all of the 386i
specific rpc daemons that appeared in the 386i's /etc/inetd.conf. I think
that the inetd still has everything running that should permit rlogin to
work, but for some reason it doesn't.
Tom Griner
teg at orc.olivetti.com
------------------------------
Date: Tue, 25 Oct 88 13:12:29 CDT
From: "Eric C. Pearce" <pearce at tycho.yerkes.uchicago.edu>
Subject: Quota system questions
I have recently tried to turn on our dist quota system (Sun OS 3.5,
4.2BSD). Several questions have poped up and the system seems somewhat
buggy. The questions are:
1. Do I need to run quotacheck each time I edit quota's with edquota?
Seems that edquota'ing gets things all out of sync and the partitions
quotas files stops updating.
2. Although we are not limiting inodes, the system cannot seem to count
files very well. For example, starting with an empty directory...
=======
tycho% ls
tycho% quota -v
Disk quotas for pearce (uid 157):
Filesystem usage quota limit timeleft files quota limit timele
/usr2 1460 37000 74000 96 0 0
tycho% touch a b c d e
tycho% ls
a b c d e
tycho% quota -v
Disk quotas for pearce (uid 157):
Filesystem usage quota limit timeleft files quota limit timele
/usr2 1460 37000 74000 96 0 0
tycho% rm a b c d e
tycho% quota -v
Disk quotas for pearce (uid 157):
Filesystem usage quota limit timeleft files quota limit timele
/usr2 1460 37000 74000 91 0 0
=====
Notice that the number of files did not increase when we created 5 null
files, but *did* decrease by 5 when we rm'ed them. Can a user take
himself down to zero files this way?
3. How well does the system keep track of usage, particularly since it
seems to mess up files counts this bad?
Any help would be greatly appreciated (pearce at tycho.yerkes.uchicago.edu).
Thanks.
- Ecp
Eric C. Pearce, Yerkes Observatory, University of Chicago.
pearce at tycho.yerkes.uchicago.edu
pearce%tycho.yerkes.uchicago.edu at oddjob.uchicago.edu
------------------------------
Date: Tue, 25 Oct 88 13:24:42 EDT
From: albert at mssun7.msi.cornell.edu (Jay Albert)
Subject: emergency bootables?
As a new SysAdmin, I am nervous about our disks. I would like to be able
to reboot our Sun3/260 server and 4 Sun3/50 clients off of any of them if
one or more fail. Currently, the output of df looks like:
Filesystem kbytes used avail capacity Mounted on
/dev/xy0a 7511 3998 2761 59% /
/dev/xy0f 5071 3848 715 84% /pub.MC68020
/dev/xy0h 113657 63243 39048 62% /usr.MC68020
/dev/xy3c 260874 211557 23229 90% /usr.MC68020/mc3
/dev/xy2c 260874 198129 36657 84% /usr.MC68020/local
/dev/xy1c 260874 207690 27096 88% /usr.MC68020/mc1
I have copied our kernel vmunix, as well as diag and boot, into
directories in /usr.MC68020/mc1, /usr.MC68020/local, and /usr.MC68030/mc3.
Will they be usable if Doomsday comes, and is there anything else I should
do? We don't have distribution tapes handy, although there are other
machines we could try to boot from over our local network. But booting
from the other disks seems easier and more reliable.
------------------------------
Date: Tue, 25 Oct 88 13:23:06 EDT
From: mfrydenb at server1.prime.com (Mark Frydenberg)
Subject: Algebraic Calctool with Trig Functions?
Does anyone know of a calctool with operator precedence (i.e., 2+3*4+5 =
19, NOT 25) and trig functions, logarithms, and other assorted goodies?
Thanks,
Mark Frydenberg
mfrydenb at server1.prime.com
Software Engr. / Geometry-Surface Modelling Group
Computervision, A Division of Prime Computer, Inc.
201 Burlington Road MS 4-1 Bedford, MA 01730
------------------------------
Date: Tue, 25 Oct 88 13:34:55 EDT
From: albert at mssun7.msi.cornell.edu (Jay Albert)
Subject: single user while dumping?
The manuals all seem to say that you "should" be in single-user mode while
making dumps, especially level 0 dumps. I have made some level 0's while
in multi-user mode, with a handfull of people on (over several hours).
Should I throw them away, or are they probably fine? Dump didn't complain
about anything. I haven't tried to restore anything off them yet.
------------------------------
End of SUN-Spots Digest
***********************
More information about the Comp.sys.sun
mailing list