Sun-Spots-Request@RICE.EDU (William LeFebvre) (07/29/88)
SUN-SPOTS DIGEST Wednesday, 27 July 1988 Volume 6 : Issue 154 Today's Topics: Re: Error messages not appearing in console window on client Re: Transferring files from Unix to IBM via tape Re: Reading/Writing IBM tapes on a Sun New Sun networking mailing list 4.0 CAVEAT: /usr/include/sys/core.h (and "undump") Monitors and Eyestrain Return of the Revenge of the 68881 SLIP? Problems with Sun4 & OS4 & color displays ASN.1 replacement for XDR? tcsh on a Sun-4? Looking for a Sun compatible phone dialer Send contributions to: sun-spots@rice.edu Send subscription add/delete requests to: sun-spots-request@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@rice.edu" or mail the word "help" to the same address for more information. ---------------------------------------------------------------------- Date: Fri, 15 Jul 88 10:25:09 CDT From: Jim Knutson <knutson%sw.MCC.COM@mcc.com> Subject: Re: Error messages not appearing in console window on client I have had the same problem and it is definitely NOT due to multiple console windows and removing the wrong one. I have only seen it happen with a few st or si device errors usually closely associated with an error saying that the scsi bus is continually busy. I was more frustrated by the tape drive at the time so I can't remember exactly what the error messages were or if they always came out splattered across the screen. Jim Knutson knutson@mcc.com im4u!milano!knutson ------------------------------ Date: Fri, 15 Jul 88 10:48:14 PDT From: levine@nsc.nsc.com Subject: Re: Transferring files from Unix to IBM via tape I have written a program which runs on an IBM mainframe which will read a Unix 'tar' format tape and deblock the tape contents into IBM VM/CMS files. The source code and appropriate execware (scripts for the unfamiliar) are available for the asking. If there is sufficient interest, I will submit the files to sun-spots for general access. David LeVine 408-721-5317 Internet: levine@parns.nsc.com Uucp: {sun,decwrl,hplabs,uunet}!nsc!parns!levine Snail Mail: National Semiconductor Corporation P. O. Box 58090, MS D3677 2900 Semiconductor Drive Santa Clara, California 95052-8090 ------------------------------ Date: Fri, 15 Jul 88 12:21:32 EDT From: Bill Arbaugh <arbaugh@hqda-ai.arpa> Subject: Re: Reading/Writing IBM tapes on a Sun We have a PD program called ansitape that will read and write ASCII and EBCDIC labelled tapes. It was written by David Hayes who use to work with me here. Ansitape is availabe via anonymous ftp from: hqda-ai.arpa. The program has been used for a few years here and has saved us a great deal of time. Bill Arbaugh Phone: (202) 694-6912 UUCP: *!uunet!cos!hqda-ai!arbaugh ARPA: arbaugh@hqda-ai.arpa ------------------------------ Date: Fri, 15 Jul 88 12:11:23 EDT From: steve@umiacs.umd.edu (Steven D. Miller) Subject: New Sun networking mailing list This message is to announce the creation of a new mailing list devoted to Sun networking issues. (For those of you at the December SUG Conference, this is the list I was talking about then. My apologies for taking so long to set things up.) The addresses associated with the list are: Sun-Nets@brillig.umd.edu (submissions) Sun-Nets-Request@brillig.umd.edu (additions/deletions/admin stuff) Please address *all* requests for additions, deletions, and whatnot to Sun-Nets-Request@brillig.umd.edu. Otherwise, everyone on the list will see your junk mail, and many of them will be displeased... So far as I'm concerned, this is an open, unmoderated forum for any questions, issues, or concerns about Sun networking. Some sample topics for discussion include: NFS (note that there is a NFS list at, I think, nfs@bmc.com) TCP/IP RPC/XDR and RPC services Sun networking problems Sun networking bugs Sun network configuration (i.e., how many diskless machines can go on one Ethernet) Mail etc. As stated above, the list is unmoderated. That could conceivably change, though I might not have the time to serve as moderator. I will be keeping archives, and I will make them available for anonymous FTP. (More on this at some later date.) I do not at this time intend to bring up any sort of archive server. I hope this list can be useful for all concerned. -Steve Miller Networking Special Interest Group Coordinator, SUG Sun-Nets-Request@brillig.umd.edu Spoken: Steve Miller Domain: steve@mimsy.umd.edu UUCP: uunet!mimsy!steve Phone: +1-301-454-1808 USPS: UMIACS, Univ. of Maryland, College Park, MD 20742 ------------------------------ Date: Thu, 28 Jul 88 12:32:28 CDT From: William LeFebvre <phil@Rice.edu> Subject: 4.0 CAVEAT: /usr/include/sys/core.h (and "undump") [ Sorry that I got so far behind in digests. ] We just spent half of yesterday trying to get TeX and friends working on a 3/60 running 4.0. In the process we discovered that undump does not quite work correctly when compiled under 4.0, even with the fixes posted here. It would pretend to work, but the resulting executable would core dump before doing anything. In the process of tracking down the problem, I discovered that the executable from a 3.5 system worked just fine (much to my surprise). This got me suspicious. I resorted to comparing the ".s" files produced by a "cc -S". Most of the changes were in the label names. So I decided to assemble and link the 3.5 ".s". It *worked*! "AHA! A compiler bug." Well, that was a hasty conclusion. Closer inspection of the ".s" diff revealed a vastly different constant: the size of a struct core. The problem is in the difference between the systems' <sys/core.h> files, specifically the definition of a "struct core". Under 3.5, the last few lines of the definition are: #ifdef sun struct fp_status c_fpstatus; /* External FPP state, if any */ struct fpa_regs c_fparegs; /* FPA registers, if any */ int c_pad[CORE_PADLEN]; /* see comment above */ #else sun But under 4.0, those lines are replaced with: #ifdef FPU struct fpu c_fpu; /* external FPU state */ #endif And where is FPU defined? Not in <sys/param.h> where you would expect it, but in <machine/reg.h>. That's not too bad, except that the #define is inside a #ifdef KERNEL! So, FPU was not defined when I compiled undump. Now this strikes me as a little odd. Technically, I cannot get a reliable definition of "struct core" from the include files unless "KERNEL" is defined. I'm not sure I like that. I'm tempted to report this as a bug, but I'm not really sure it is. Anyway, thought you net-folks would appreciate the warning. I "fixed" the problem with a kluge: I added -DFPU to the flags given to "cc". William LeFebvre Sun-Spots moderator Department of Computer Science Rice University <phil@Rice.edu> ------------------------------ Date: Fri, 15 Jul 88 12:07:15 EDT From: ric@rioja.cc.umich.edu (Richard Campbell) Subject: Monitors and Eyestrain In v6n140, Jeremy Isserlis writes: > The only other problem [with monitors and eyestrain]is that the height > of the screen above the desk, being raised some seven inches, > is uncomfortable, particularly as one's paper work is not that high! > I suppose the answer is to put the machine on a lower table? I have noticed similar problems with the awkward placement of workstations on my desktop. Not only does the height above the desktop seem to result in a strain on the eyes and the neck, but it takes up a large volume of space. Also, after a recent visit to my eye doctor, I was told that due to differences in visual acuity between my right and left eye, I should really get bifocal glasses. While sound medical advice, I had to reject it and purchase "reading only" glasses; trying to view a large screen through the lower, small "reading" half of bifocals would have caused *extreme* strain on my eyes and a good deal of muscle fatigue from the odd angle of my head such viewing would cause. I do, however, have a solution to the problem. It is a relatively simple carpentry project to hack together a desktop that has a surface which slopes away from you. Onto this surface you can place either the monitor or even an entire workstation. The point is to place the surface of the monitor at a more natural and convenient angle for reading. (You must, of course, be careful of the monitor's center-of-gravity and make sure that it won't tip over.) The first release of this desk quickly confirmed that my posture was better and typing was easier due to the keyboard appearing closer to the screen images. It was also more conveniant for me to place papers and manuals nearer the screen for reference. And visitors to my desk could much more easily read the screen while standing up. The only problems were that the screen reflected overhead flourescent lighting (which should be turned off anyway) and that if the workstation itself is placed on the sloping desktop it is much harder to manipulate the cables in the rear (the next revision will put only the monitor on the slope and place the workstation on the floor). + / \ / \ / \ / \ / \ / \ ---------+ / \ / keyboard| / \ +-------------+ / \ | | \ + +-------------+/\ \ / | |\/ \ \ / | | \ \ \ / | | \ \ / \ / | | \ \ / / \ / | | \ \ / \ / | | \ \ \ / | | \ \ ^ | | \ \ | | \ \ | | \ \ | | \/ | | \+-+ | | | | | | | | | | | | | | | | +-+----------------------------------+-+ Richard Campbell ------------------------------ Date: Fri, 15 Jul 88 12:43:34 EDT From: valdis@sun.mcs.clarkson.edu Subject: Return of the Revenge of the 68881 OK, Ok already.. Thanks for all the mail concerning the -f68881 flag. It didn't take me long to figure out that you need the -f68881 flag on the LINK step as well. I finally ended up bagging the idea, and used -fswitch instead - that's still faster than -fsoft, and doesn't require a flag EVERY TIME you link against the library. (Too bad 'csh' doesn't provide a facility similar to 'sh's "/etc/profile" where I could park a SETENV (and no, I don't need a patch - we don't have source :-( )) The only part of the whole thing that threw me for a loop is that a casual reading of the Floating Point Programmer's Guide seems to indicate that if you only specify one floating-point option, that it can figure out which to use. I guess there's a world of difference between getting a diagnostic if two are used, and it being able to figure it out for just one.... Valdis Kletnieks ------------------------------ Date: Fri, 15 Jul 88 02:55:45 EDT From: sunvice!cordis!gls@sun.com (Gls Gary_Schaps_2157) Subject: SLIP? I would appreciate any leads to a public domain version of SLIP. Gary L. Schaps Voice: (305) 551-2157 / 800-327-2490 ext. 2157 Cordis Corporation P.O. Box 025700 ML7A E-mail: sun!sunvice!cordis!gschaps Miami, FL 33102-5700 gatech!uflorida!novavax!cordis!gschaps [[ Try the sun-spots archives! Under "sun-source" there is a shar file called "slip.shar". It can be retrieved via anonymous FTP from the host "titan.rice.edu" or via the archive server. For more information about the archive server, send a mail message containing the word "help" to the address "archive-server@rice.edu". --wnl ]] ------------------------------ Date: 15 Jul 88 09:19 PDT From: Stanley's Tool Works <Lanning.pa@xerox.com> Subject: Problems with Sun4 & OS4 & color displays I've recently "upgraded" my Sun4 w color display to OS4. I use quotes because I'm not sure this is an improvement. The first thing I noticed is the nice new alert windows in Sunview. Only problem is that the cursor is invisible when the alert window pops up. Well, not quite -- it (the cursor) actually punchs a hole in the alert window, letting you see the window below it (the alert window), or just white if there's nothing under it. Makes it kind of hard to figure out where you're pointing. The other problem is with X: I recompiled Xv11 from scratch, and it's given me nothing but grief. Most tools die immediatly; xterm works unless I try to add a scroll bar. I'm no C hacker, and I really don't want to try to debug this myself. At least GnuEmacs works. Does any of this sound familiar? More to the point, what can I do to fix it? ---smL [[ Hmmm....On a black and white monitor (a 3/60), the alert window works "normally". That is, the cursor stays black. I won't say that it works "correctly" because I don't like the fact that it is instantly transported to the default button and then instantly transported back to its original position when I click. I don't like it bouncing around like that. --wnl ]] ------------------------------ Date: Fri, 15 Jul 88 09:37:27 EDT From: wang!ellison@decvax.dec.com Subject: ASN.1 replacement for XDR? I am looking for a public domain implementation of ASN.1 which can replace the XDR and RPCGEN pieces of RPC. Can anyone help? Mark Ellison Wang Labs 014-590 ..!decvax!wang!ellison One Industrial Av (617) 967-2583 Lowell, MA 01851 ------------------------------ Date: Fri, 15 Jul 88 11:55:54 edt From: trevor@grasp.cis.upenn.edu (Trevor Darrell) Subject: tcsh on a Sun-4? Has anyone had any sucess bringing up tcsh under SunOS 4.0 on a Sun-4? I have used the 4.3 tcsh diffs distributed recently on comp.sources.unix, applying them to BSD 4.3 csh source. No Luck. Are the SunOs 4.0 csh sources different then the BSD 4.3 source? Anyone have any tips about tcsh and SunOS 4.0? --trevor trevor@grasp.cis.upenn.edu ------------------------------ Date: Fri, 15 Jul 88 10:22:05 PDT From: (Lorenzo Aguilar) <aguilar@tsca.istc.sri.com> Subject: Looking for a Sun compatible phone dialer I will appreciate information about vendors offering Sun (specifically 3/50) compatible phone dialers. I want a dialer that converts an ASCII string (a phone number) into dialing tones that can be fed into a phone via an standard modular jack (do not want the kind that produces tone sounds). Probably the dialer will take input from a serial port. If I find such a thing, I will write a little problem that will let me type a name on my keyboard, lookup the corresponding phone number in a file and then pass it to the dialer. Thanks Lorenzo Aguilar SRI aguilar@tsca.istc.sri.com 415-859-4118 ------------------------------ End of SUN-Spots Digest ***********************