Sun-Spots-Request@RICE.EDU.UUCP (06/20/86)
SUN-SPOTS DIGEST Thursday, 19 June 1986 Volume 4 : Issue 18
Today's Topics:
V-System release 6.0
Mount deadlocks
Suntools under 2.0 and SunView under 3.0 (3)
Sun 2 running 3.0 and TeX
Sun release 3.0 compiler bug
LCF lives
Suncore from Franz Lisp
How to track the cursor on the screen under SunView?
Audio Visual uses of Sun workstation output?
MACSYMA on SUN-3s?
Sun as a Timesharing Box?
MDQS on Sun 3.0?
Will Sun-3's run 68010 executables?
Need suppliers for new 9" Hitachi/NEC drives?
Need cheap paint program?
Atof bug fix?
------------------------------------------------------------------------
Date: 18 Jun 1986 1626-PDT (Wednesday)
From: Lance Berc <lance@su-pescadero.arpa>
Subject: V-System release 6.0
A new version of Stanford's V-System, release 6.0, is now available.
The V-System is a distributed operating system which runs on Sun and
MicroVax workstations. It's based on a distributed kernel that provides
lightweight processes and network transparent, high speed interprocess
communication. Utilities and applications include a high performance
window system, internet network protocol handlers, mouse oriented
editors, and the TeX family of document compilers.
Enhancements between release 5.0 and 6.0 include changes to naming &
authentication, group ipc, many more applications, and ports to DEC
MicroVax-II & Sun-2/50 and Sun-3 workstations.
Release information is available through the Stanford Office of
Technology Licensing. Their address is:
Office of Technology Licensing
Suite 250
350 Cambridge Avenue
Palo Alto, CA 94306
(415) 723-0651
Lance Berc
lance@pescadero.stanford.edu
-----------------------------
Date: Sat, 14 Jun 86 19:53:26 EDT
From: Barry Shein <bzs@BU-CS.BU.EDU>
Subject: mount deadlocks
Re: will cross-mounting file-systems cause deadlocks if the
two systems come up at about the same time.
Yes.
BUT, the good news is, simply move the line which does the mounts
to below the one which starts up NFSD in /etc/rc.local and the
problem seems to be solved. Note that if your fstab names local
mounts before NFS mounts (almost certainly should) the worst that
could happen is the far end has to wait for you to get to that
point. That is, '/etc/mount -vat nfs' should be about the last
thing in your /etc/rc.local file. Use your common sense, it wouldn't
be wierd to do local mounts early and only NFS mounts at the
end, whatever makes you happy but that's all that should be
involved.
-Barry Shein, Boston University
[I would suggest SUN change the default rc.local on the distribution
tape to be this way unless someone can come up with a reason why not.]
-----------------------------
Date: Thu, 12 Jun 86 16:50:49 -0500
From: Mark Weiser <mark@markssun.cs.umd.edu>
Subject: Suntools under 2.0 and SunView under 3.0 (1)
Panels are not a SunView feature, they appeared over a year ago.
From the Sun 2.0 Programmers Reference Manual for SunWindows (April 1985):
The option subwindow package is included in this release,
but will *not* be included in future releases of SunWindows.
We recommend that client programs instead use the panel
subwindow package.
That was a year's warning, not even "might not be included",
but "WILL not be included". Plenty of time to convert to panels
(especially since that same manual also gives a recipe for converting
options to panels which I had to use at one point and found pretty easy.)
-mark
-----
Spoken: Mark Weiser ARPA: mark@maryland Phone: +1-301-454-7817
CSNet: mark@umcp-cs UUCP: {seismo,allegra}!umcp-cs!mark
USPS: Computer Science Dept., University of Maryland, College Park, MD 20742
-----------------------------
Date: Mon, 16 Jun 86 14:03:49 bst
From: David England <mcvax!comp.lancs.ac.uk!de@seismo.CSS.GOV>
Subject: Suntools under 2.0 and SunView under 3.0 (2)
In the 2.0 Sunwindows Reference Manual App. F on Option Subwindows it says:
[1st para.]
"NOTE: The option subwindow package is included in this release but will
*not* be included in future releases of SunWindows. We recommend that client
programs instead use the panel subwindow package .... " etc.
This was also pointed out at Sun User group meetings by Sun
representatives in the UK. You had been warned.
Dave ..!mcvax!ukc!dcl-cs!de
"I like to go where the action is. Move in, move out. And no paper work."
-----------------------------
Date: Fri, 13 Jun 86 14:52:50 CDT
From: William LeFebvre <phil@titan.rice.edu>
Subject: egg on face, foot in mouth, general stupidity (3)
Well, what's a gentleman to do when he discovers that he has made a
complete fool of himself in front of thousands of people nation-wide?
As several people have pointed out to me, it was well documented in the
Sun release 2.0 manuals that option subwindows were going away and
would not be supported in future releases. My apologies to Sun and to
those I misled or otherwise offended.
As I look thru the Sun 2.0 manuals, it occurs to me that I have never
seen these things before. I suddenly realize that ALL of my SunWindow
hacking was done with the Release 1.1 manuals in my lap, and not 2.0!
No wonder the disappearing optionsw took me by surprise. But why is
this? I don't know---it may have something to do with the fact that
someone here at Rice hid the Sun 2.0 manuals in binders that say
"Digital" on the outside.
As for VorTeX, the people at Berkeley apparently did not heed the
warning any more than I did. But, an hour's worth of hacking or so
with a 2.0 manual in my lap, and I think I have a dvitool that will
compile and run under both 2.0 and 3.0 (SunWindows and SunView).
I really do like SunView alot better than the old stuff. The user
interface is nicer to use and the program interface is MUCH, MUCH
BETTER than the old one. No more of this declaring and initializing
300 different structures and calling 100 different routines that each
require you to give a different field of the structure the last routine
returned. Everything is much more straightforward.
Coming soon to a mailbox near you......a review of VorTeX.
William LeFebvre
Department of Computer Science
Rice University
<phil@Rice.edu>
-----------------------------
Date: Fri 13 Jun 86 14:54:33-PDT
From: Pierre MacKay <MACKAY@WASHINGTON.ARPA>
Subject: Sun 2 running 3.0 and TeX
Numerous reports confirm that there are serious bugs in Sun operating system
3.0 when it is brought up on a Sun 2. The most obvious for people trying to
bring up TeX and Metafont appear to center on the IO library, or at least in
that vicinity.
Both initex and inimf will compile, but inimf usually causes an instantaneous
core dump when it is invoked, and initex, though it will apparently run almost
to completion, hangs on the \dump phase. This is absolutely NOT a defect in
the TeX or METAFONT code. The very same code works in version 3.0 when it is
run on a SUN 3. It also works in versions of SUN 2 software before 3.0.
The problem is assuredly not limited to TeX and METAFONT, but it is
particularly crippling in the case of thos programs. The only thing I can
suggest, besides screams of protest, is that binaries from earlier versions
MAY work for now. They certainly run on a SUN 3, though you can't expect
to undump them for real convenience.
But this really MUST be fixed.
Pierre
-----------------------------
Date: Sat, 14 Jun 86 18:02:45 -0100
From: GIBBS Simon <lll-crg!seismo!mcvax!cui!simon@lll-tis-b.ARPA>
Subject: Sun release 3.0 compiler bug
WARNING - SUN 3 cc optimizer bug
Try compiling and executing the following program:
main() {
int i = 0;
char *p = " ";
register r1 = 0;
register char r2 = *(p+i);
if(r2 == 'x')
r1 = r1 - 1;
else
if(r2 == ' ')
r1 = 1;
printf("%d\n", r1);
}
it should print a 1.
The results we've had are:
compiler on OS: without -O with -O
4.2 1 1
Sun release 2.2 1 1
Sun release 3.0 1 0 ???
Simon Gibbs
Centre Universitaire d'Informatique
Universite de Geneve
-----------------------------
Date: Fri, 13 Jun 86 13:09:58 edt
From: young@ll-vlsi (George Young)
Subject: LCF lives
LCF does exist -- I talked to them last week. They can be reached at:
tel. (408) 221-0776
LCF International
2530 Berrryessa Rd. suite 507
San Jose, California 95132
They have 4MB multibus boards available immediately, VME boards are
in production, but will be 4-6 weeks for delivery.
All boards have lifetime warranty-- phone in report of a fault and they
will fed-express a new board, then you send back old board.
[I have no connection with LCF, just a brain damaged sun-1/100 starving
for memory]
-----------------------------
Date: Tue, 17 Jun 86 13:56:24 bst
From: David England <mcvax!comp.lancs.ac.uk!de@seismo.CSS.GOV>
Subject: Suncore from Franz Lisp
On our version on Franz Lisp these files exist in /usr/lib/lisp;
/usr/lib/lisp/suncore.demo.l
/usr/lib/lisp/suncore.l
/usr/lib/lisp/suncore.o
/usr/lib/lisp/suncoreinit.c
/usr/lib/lisp/suncoreinit.o
This is Franz Lisp opus 42.16.3 1985
(fasl '/usr/lib/lispsuncore.o) gives assess to SunCore routines which
are prefixed by 'sc:' e.g.
(sc:initialize_core sc:buffered sc:synchronous sc:twod)
(sc:initialize_view_surface viewsurf sc:false)
(sc:select_view_surface viewsurf)
(sc:set_window 0.0 100.0 0.0 100.0)
(sc:create_retained_segment 1)
;;; some drawing routines
(sc:close_retained_segment 1)
(sc:terminate_core)
I know Suncore and C but not lisp. This was borrowed from a lisp
hacking colleague.
Dave uucp: ...!mcvax!ukc!dcl-cs!de
arpa: de%lancs.comp@ucl-cs
"I like to go where the action is. Move in, move out. And no paper work."
-----------------------------
Date: Tue, 17 Jun 86 19:26:56 CDT
From: Bill Janssen <janssen@mcc.arpa>
Subject: How to track the cursor on the screen under SunView?
Question for you SunView hackers: how does one track the cursor on
the screen, when it goes outside the window it's associated with?
I'd like to get all mouse events, no matter where the cursor is on
the screen. I'd like to have the location reported in the window
coordinates, but screen coordinates will be fine too.
I'm using SunTools (release 3.0, I believe) on a Sun-3/160 with
a color monitor. I've tried (in order of frustration :-))
1. Setting the window attribute WIN_GRAB_ALL_INPUT to TRUE with
window_set(). When the mouse leaves the window boundary, events
stop happening until it comes back in. No good.
2. Calling "win_grabio()" and "win_releaseio()". Same no good
effects.
3. Using the complete "fullscreen" package described in Chapter 12
of the "SunView System Programmer's Manual". Same results.
4. Opening "/dev/mouse" and reading the thing directly. I DON'T
like to do this under SunTools! (and don't understand what I'm
reading, either :-))
Sun claims it can't be done unless SunTools is modified. Anyone
know differently? (And what DOES fullscreen do for one, anyway?)
Bill
-----------------------------
Date: Tue, 10 Jun 86 13:32:33 edt
From: Bob Marcus <harvard!wanginst!marcus@sally.utexas.edu>
Subject: Audio Visual uses of Sun workstation output?
We are interested in producing transparencies, videotapes, and 35mm slides
from Sun workstation output. We would appreciate information on hardware and
software packages which would give us this capability, along with any relevant
user experience. Thank you for your help.
Steve Boudreau
AV Specialist
Wang Institute of Graduate Studies
-----------------------------
Date: Fri, 13 Jun 86 09:31 EST
From: Diana Wegner <WEGNER%RMATH1%gmr.com@CSNET-RELAY.ARPA>
Subject: MACSYMA on SUN-3s?
Does anyone have experience running MACSYMA on a Sun-3? We will
soon be receiving five diskless Suns and a file server with color
monitor and are considering running MACSYMA on the file server
node. We are interested in any information available.
Thanks in advance,
Diana Wegner
Mathematics Department
General Motors Research Laboratories
csnet address: wegner@gmr.com
-----------------------------
Date: Fri, 13 Jun 86 21:10:42 edt
From: Ken Mandelberg <km%emory.csnet@CSNET-RELAY.ARPA>
Subject: Sun as a Timesharing Box?
Is anyone using a Sun for straight Unix Timesharing with many users?
What would it take to configure a Sun system to handle say 40 students
simultaneously doing course work?
Ken Mandelberg
Emory University
Dept of Math and CS
Atlanta, Ga 30322
{akgua,sb1,gatech,decvax}!emory!km USENET
km@emory CSNET
km.emory@csnet-relay ARPANET
-----------------------------
Date: 13 Jun 86 14:55:44 PDT (Fri)
From: trwrb!trwspp!spp2!urban@ucbvax.berkeley.edu
Subject: MDQS on Sun 3.0?
Has anyone gotten BRL's `MDQS' spooling system working under version 3.0?
Some of our users have reported that the "netsend" routine stopped working
when they started running 3.0.
Mike
-----------------------------
Date: Mon, 16 Jun 86 14:57:18 PDT
From: speck@vlsi.caltech.edu (Don Speck)
Subject: Will Sun-3's run 68010 executables?
This summer, half of our Sun-2's will be upgraded to Sun-3's CPUs.
The Sun Unix 3.0 documentation claims that we need to keep separate
/usr.68010 and /usr.68020 directories, but we don't have disk space
for it. Is it really true that 68020's won't run 68010 binaries?
I thought the 68020 was supposed to be upward compatible! What is
it that goes wrong? Page size mismatch?
Don Speck speck@vlsi.caltech.edu seismo!cit-vax!speck
-----------------------------
Date: Mon, 16 Jun 86 17:25:48 pdt
From: bake@ee.UCLA.EDU (Dr Rich Baker)
Subject: Need suppliers for new 9" Hitachi/NEC drives?
Hitachi has a new 9" 525 MB (unformatted) drive, the DK-815, which
sports an 18 ms average access time. We are considering it for
use on a Sun file server.
We understand NEC has a similar product. Both fit in 1/2 the space
of the typical Fujitsu Eagle.
(1) Anyone using these drives?
(2) Who are your suppliers?
I'll summarize responses to Sun-Spots.
Rich Baker, EE Dept, UCLA
bake@ee.ucla.edu
ucbvax!ucla-cs!bake
-----------------------------
Date: Tue, 17 Jun 86 15:39:49 pdt
From: bake@ee.UCLA.EDU (Dr Rich Baker)
Subject: Need cheap paint program?
We need a good program for creating figures for technical publications
(IEEE, conferences) that runs on the Sun and creates ditroff or
postscript for laserprinting. It should be able to do block diagrams,
schematics, simple charts, etc., with typeset legends.
What do all you folks use? I'll post your suggestions.
Rich Baker, Asst. Prof., EE Dept, UCLA
bake@ee.ucla.edu
[ucbvax!]ucla-cs!bake
-----------------------------
Date: 18 Jun 1986 12:05:22 EDT (Wednesday)
From: T. Michael Louden (MS W422) <louden@mitre-gateway.arpa>
Subject: atof bug?
Does anyone have a fix for the atof function? I find that although the manual
says it returns a double you only get 6 decimal place accuracy. Somewhere it
must be doing float math.
-----------------------------
End of SUN-Spots Digest
***********************