Article 2930 of alt.sys.pdp10: Newsgroups: alt.sys.pdp8,alt.sys.pdp10 Path: nntp1.ba.best.com!news1.best.com!news.mathworks.com!nntprelay.mathworks.com!howland.erols.net!ix.netcom.com!netcom16!alderson From: alderson@netcom16.netcom.com (Richard M. Alderson III) Subject: Re: RTS8 images? In-Reply-To: jimc@zach1.tiac.net's message of 8 Jul 97 01:01:08 GMT Message-ID: Sender: alderson@netcom16.netcom.com Reply-To: alderson@netcom.com Organization: NETCOM On-line services References: Date: Tue, 8 Jul 1997 17:40:46 GMT Lines: 28 Xref: nntp1.ba.best.com alt.sys.pdp8:1960 alt.sys.pdp10:2930 In article jimc@zach1.tiac.net (James E. Carpenter) writes: >>And emulators for all of them are available; a friend of mine runs the PDP-10 >>emulator on his Mac PowerBook 145 (just to prove it could be done). >I think there are about 2 or 3 PDP-10 emulators being worked on but I don't >know of one that is available yet. But for something that isn't available I >sure hear about a lot of people having one! :-) Lemme see. Ken Harrenstien (KLH), still at Oracle I think, wrote a KS-10 emulator years ago; that's the one I've seen running on Macs (including the PB145). KLH later revised it to emulate a KL-10B, and tried to sell it through Digital as an Alpha Migration Tool; I don't know that anyone ever bought it from Digital. Then there's Stu Grossman, who wrote one while working at Stanford. I saw an early version running on his office Sparcstation; I know he's kept playing with it over the years. So that's three. There are others (one written in PDP-10 assembler, to emulate upwards to a new implementation of the architecture) of less use to the general populace. -- Rich Alderson You know the sort of thing that you can find in any dictionary of a strange language, and which so excites the amateur philo- logists, itching to derive one tongue from another that they know better: a word that is nearly the same in form and meaning as the corresponding word in English, or Latin, or Hebrew, or what not. --J. R. R. Tolkien, alderson@netcom.com _The Notion Club Papers_ Article 3074 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!su-news-hub1.bbnplanet.com!news.bbnplanet.com!news1.digital.com!pa.dec.com!usenet From: bob.supnik@digital.com (Bob Supnik) Newsgroups: comp.sys.dec,alt.sys.pdp8,alt.sys.pdp11,alt.sys.pdp10,alt.folklore.computers Subject: Computer History Simulators V2.3: new systems, software Date: 25 Jul 1997 22:20:08 GMT Organization: Digital Equipment Corporation Lines: 56 Message-ID: <5rb8qo$9ld@usenet.pa.dec.com> NNTP-Posting-Host: srcpc137.pa.dec.com Mime-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII X-Newsreader: WinVN 0.99.7 Xref: nntp1.ba.best.com comp.sys.dec:28306 alt.sys.pdp8:1986 alt.sys.pdp11:2626 alt.sys.pdp10:3074 alt.folklore.computers:89854 Computer History Simulators, V2.3 --------------------------------- In honor of the 40th anniversary of Digital Equipment Corporation on August 23, I am pleased to announce the availability of V2.3 of the Computer History simulators at the following FTP site: ftp://ftp.digital.com/pub/DEC/sim Subdirectory sources contains the simulator sources, and subdirectory software contains sample software for the simulators. All are compressed tar files. The Computer History simulators include the following systems: system sample software PDP-1 Lisp, courtesy L Peter Deutsch PDP-4 PDP-7 SIM8, courtesy David J Waks PDP-8 ESI-X, courtesy David J Waks Focal69, courtesy Digital Equipment Corporation OS-8, under license from Digital Equipment Corporation PDP-9 PDP-11 UNIX V5/6/7, under license from The Santa Cruz Operation PDP-15 Nova RDOS V7.5, under license from Data General Corporation IBM 1401 single card utilities The simulators operate on most versions of UNIX and on OpenVMS. The simulators compile under Visual C++ for Windows 95/Windows NT, but the terminal emulation is incomplete. Major changes and improvements for V2.3: 1. Inclusion of the PDP-1, additional sample software for the PDP-1, PDP-7, and PDP-8, and large disks (RP/RM series) for the PDP-11. 2. Endian independence. The simulator will now compile and operate correctly on big endian UNIX systems, without source modifications. 3. TERMIOS-based terminal handling. The default terminal handler is based on TERMIOS and will compile and operate correctly on most UNIX systems, without source modifications. 4. Revised simulated magtape format. The format for simulated magtape was revised to handle larger record sizes and reverse direction operations. A utility for converting from prior versions is included. The simulator documentation includes a full revisions list. /Bob Supnik Article 3200 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsxfer3.itd.umich.edu!infeed1.internetmci.com!newsfeed.internetmci.com!164.67.42.145!nntp.info.ucla.edu!news.bc.net!unixg.ubc.ca!alph02.triumf.ca!shoppa From: shoppa@alph02.triumf.ca (Tim Shoppa) Newsgroups: alt.sys.pdp10 Subject: Re: What's with the pdp-10 emulator? Date: 25 Aug 1997 14:17:46 GMT Organization: TRIUMF, Canada's National Meson Facility Lines: 43 Message-ID: <5ts46a$a7p$1@nntp.ucs.ubc.ca> References: <01bcafc2$e24acb40$29c906c6@hsnewman.ccsi.com> <5to3ud$ria$1@news.wizvax.net> <34016536.2FA8@cisco.com> NNTP-Posting-Host: alph02.triumf.ca Xref: nntp1.ba.best.com alt.sys.pdp10:3200 In article <34016536.2FA8@cisco.com>, Henry W. Miller wrote: > How about I/O ? Do you really need to emulate the I/O completely, >or do you make the I/O instructions a sort of emulator UUO, which would >map to a virtual device, like a large file on a disk to emulate a >virtual RP07? The conceptual idea of "container files" for disks and tapes is well developed in many other emulators. The difficulty with emulating a virtual RP is that if you want it to work with TOPS-10 or ITS, you also have to emulate the Massbus subsystems. This is extremely difficult, though not impossible, because many of the details necessary to do this were never publicly documented. (This didn't stop some from figuring it out from the schematics, though.) >> John Wilson wrote: >> D Bit >> (Author of emulators, but not THE emulators, of PDP-11, Nova, and Eclipse. >> But you won't be seeing the Nova or Eclipse emulators anywhere, and it *is* >> for legal reasons.) Interesting. Bob Supnik's Nova emulator appears to work just fine, and I've used it to run many pieces of RDOS software. What problem prevents you from releasing it, but allows Bob to do so? I'll take a wild guess at what the Eclipse problem is: the microcode, which is copyrightable? > Pity, and surprising (or maybe not) considering the history of >the DG line in regards to DEC computers of that era. Bob Supnik's PDP-8, PDP-11, PDP-1, Nova, and 18-bit (PDP-4, -7, -9, and -15) emulators (and a version of RDOS that runs on them) are available from ftp://ftp.digital.com/pub/DEC/sim Bob has succesfully persuaded the current owners of Unix to allow emulator use of Unix System 5, 6, and 7 binaries, and he has also gotten permission from Data General to include a release of RDOS 7.50 for use on the Nova emulator. Tim. (shoppa@triumf.ca) Article 3273 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.maxwell.syr.edu!news-peer.sprintlink.net!news.sprintlink.net!Sprint!uunet!in2.uu.net!198.6.201.7!quest.ccsi.com!not-for-mail From: "Harris S. Newman" Newsgroups: alt.sys.pdp10 Subject: More on the pdp-10 emulator Date: 6 Sep 1997 12:48:34 GMT Organization: Harris S. Newman Consulting Lines: 21 Message-ID: <01bcbac1$0b089ea0$50c906c6@hsnewman.ccsi.com> NNTP-Posting-Host: dialup25.ccsi.com X-Newsreader: Microsoft Internet News 4.70.1155 Xref: nntp1.ba.best.com alt.sys.pdp10:3273 It seems that Dec itself has the rights to the pdp-10 emulator now. This can be seen by the fact that they "gave" Gates a copy, as well as some other evidence I have uncovered this week. Ok, here is the question: If they (Dec) post on their FTP server (in the sim directory), the pdp emulators, why don't they also post the emulator they own for the pdp-10? Why do they post the full operating system images for the pdp-11, nova, etc but refuse to post the images for Tops? Bob.Supnik@digital.com is the person who seems to maintain the sim directory. He is also responsible for the Corporate Research and Advanced Development at Digital: http://www.research.digital.com/ He is also the person who wrote one of the pdp-11 emulators out there. I would like everyone out there who wishs to run the pdp-10 emulator to petition Digital to post the pdp-10 emulator and images they have. Any reasons they give not to would be crushed by the fact that they are posting ones for the pdp-11. Please use the above information in sending your messages to Dec. Article 3551 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsxfer3.itd.umich.edu!agate!bh From: bh@anarres.CS.Berkeley.EDU (Brian Harvey) Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Curious about 10s. Date: 12 Dec 1997 00:09:01 GMT Organization: University of California, Berkeley Lines: 5 Message-ID: <66pvat$93g$1@agate.berkeley.edu> References: <348B556D.167E@eng.pko.dec.com> <66gg3a$r8g$1@shell3.ba.best.com> NNTP-Posting-Host: anarres.cs.berkeley.edu Xref: nntp1.ba.best.com alt.sys.pdp10:3551 alt.folklore.computers:95570 "Daniel A. Seagraves" writes: >Hmm... How hard could it be? When KLH wrote his, he said that the hardest part was getting 36-bit floating point to perform exactly like a -10 on 32-bit IEEE machines. Article 3554 of alt.sys.pdp10: Path: nntp1.ba.best.com!not-for-mail From: inwap@best.com (Smith and O'Halloran) Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Curious about 10s. Date: 12 Dec 1997 10:23:05 -0800 Organization: Chez INWAP (people, computers, cats) Lines: 28 Message-ID: <66rve9$a0t$1@shell3.ba.best.com> References: <348B556D.167E@eng.pko.dec.com> <66gg3a$r8g$1@shell3.ba.best.com> NNTP-Posting-Host: shell3.ba.best.com X-Trace: 881950987 5000 inwap 206.184.139.132 Xref: nntp1.ba.best.com alt.sys.pdp10:3554 alt.folklore.computers:95615 In article , Daniel A. Seagraves wrote: >I should write an emulator myself, it'd be fun to have ITS on your >desktop... >Hmm... How hard could it be? >(Goes and retrieves that instruction list...) >Lemme see, 64 times 8... >(Counts on fingers&toes) >Umm... I retract that stupid question... > >Actually, given that the instructions are all one word and all in the same >spot, it doesn't sound too hard... and if the I/O instructions are all >you'd need to emulator for devices, it'd just be sheer work! It's been mentioned before in alt.sys.pdp10 that getting an emulator up and running to where it could run EDDT is not too hard. But in that case, you are running the user-mode instructions plus the read/write from the console terminal. But as soon as you attempt to enable interrupts, you've got a hold new can of worms to deal with. The goal is to run the ITS or other Monitor *WITH NO CHANGES*. That is, load SYSTEM.EXE verbatim off of an emulated RP06. This last part takes a couple of man-years of effort, and relies on information that is not documented in any DEC manual. -Joe -- INWAP.COM is Joe and Sally Smith, John and Chris O'Halloran and our cats See http://www.inwap.com/ for "ReBoot", PDP-10, and Clan MacLeod. Article 3561 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsxfer3.itd.umich.edu!news2.chicago.iagnet.net!iagnet.net!129.79.5.185!news.indiana.edu!news.iupui.edu!woodshed!mwood From: mwood@woodshed.iupui.edu (Mark H. Wood) Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Curious about 10s. Followup-To: alt.sys.pdp10,alt.folklore.computers Date: Sun, 14 Dec 1997 12:05:31 GMT Organization: La Petite Hackerie Lines: 38 Message-ID: References: <348B556D.167E@eng.pko.dec.com> <66gg3a$r8g$1@shell3.ba.best.com> <66pug4$jog$1@nntp.ucs.ubc.ca> NNTP-Posting-Host: mhw.oit.iupui.edu X-Newsreader: TIN [UNIX 1.3 950515BETA PL0] Xref: nntp1.ba.best.com alt.sys.pdp10:3561 alt.folklore.computers:95734 Tim Shoppa (shoppa@alph02.triumf.ca) wrote: : In article , : Daniel A. Seagraves wrote: : >Anyone have ITS on tapes? : : On RP05... : : >I should write an emulator myself, it'd be fun to have ITS on your : >desktop... : >Hmm... How hard could it be? : >(Goes and retrieves that instruction list...) : >Actually, given that the instructions are all one word and all in the same : >spot, it doesn't sound too hard... and if the I/O instructions are all : >you'd need to emulator for devices, it'd just be sheer work! : : No, it's more than just work. The details of the I/O subsystems : are arcane and hairy. It's a *lot* of work, with many iterations between : how the documents say it works, then realizing how the system actually : works! Well, this depends on what you want to emulate. One *could* design new, emulated "devices" that aren't so hairy, and add new drivers for them to the kernel. One would then know exactly how the new subsystems work and could be reasonably certain of having a usable system. If a usable system is one's goal, and all sources are available, it might not be unreasonably hard for a sharp team. If one insists on being able to emulate, say, a DTE, you've taken on quite a different project. Bit-for-bit and bug-for-bug emulation is not for the faint of heart. (OK, let the corrections roll. I did a fair amount of customization and debugging on DN87S, but didn't actually have to touch the DTE code on either side. Likewise with DAS79, and I don't even remember what 10/11 adaptor we had back then, I'm ashamed to say.) -- Mark H. Wood Speaking, as always, for himself MWOOD@INDYVAX.IUPUI.EDU Our nation suffers from too little leadership, and far too much management. Article 3564 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10,alt.folklore.computers Path: nntp1.ba.best.com!news1.best.com!nntprelay.mathworks.com!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!chippy.visi.com!news-out.visi.com!ix.netcom.com!netcom16!alderson From: alderson@netcom16.netcom.com (Richard M. Alderson III) Subject: Re: Curious about 10s. In-Reply-To: inwap@best.com's message of 15 Dec 1997 12:02:22 -0800 Message-ID: Sender: alderson@netcom16.netcom.com Reply-To: alderson@netcom.com Organization: NETCOM On-line services References: <66rve9$a0t$1@shell3.ba.best.com> <54zpm6ca8e.fsf@flipper.cisco.com> <6742ce$a1b$1@shell3.ba.best.com> Date: Wed, 17 Dec 1997 18:51:37 GMT Lines: 34 Xref: nntp1.ba.best.com alt.sys.pdp10:3564 alt.folklore.computers:95850 In article <6742ce$a1b$1@shell3.ba.best.com> inwap@best.com (Smith and O'Halloran) writes: >In article <54zpm6ca8e.fsf@flipper.cisco.com>, >William "Chops" Westfield wrote: >> The goal is to run the ITS or other Monitor *WITH NO CHANGES*. >>Why? I doubt that this was a goal for the TOAD-1, the SC-30, or the KL. >It was the goal of the SC-30. Systems Concepts bragged about being able to >run unmodified DSKB:SYSTEM.EXE or PS:MONITR.EXE on their hardware. That was certainly the *goal* of the SC-30M, though it wasn't entirely met--it was not *completely* bug-for-bug compatible with the KL10B, although it tried. Because the SC-30M was a pipelined architecture, something that should not have worked on the KL10B (but did) failed in the SC box: In the EXEC command-editor was located an instruction POP P,0(P) which caused, on the SC box, some very subtle errors. I fixed the EXEC at Stanford, rather than running off to Stewart to make this particular bogosity work. It most emphatically is *not* the goal for the TOAD-1, which is a PDP-10 but not fully compatible with the DEC systems: Larger address space, different I/O model, modern peripherals, etc. -- Rich Alderson You know the sort of thing that you can find in any dictionary of a strange language, and which so excites the amateur philo- logists, itching to derive one tongue from another that they know better: a word that is nearly the same in form and meaning as the corresponding word in English, or Latin, or Hebrew, or what not. --J. R. R. Tolkien, alderson@netcom.com _The Notion Club Papers_ Article 3567 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10,alt.folklore.computers Path: nntp1.ba.best.com!news1.best.com!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!world!mbg From: mbg@world.std.com (Megan) Subject: Re: Curious about 10s. Message-ID: Organization: The World Public Access UNIX, Brookline, MA References: <66gg3a$r8g$1@shell3.ba.best.com> <66rve9$a0t$1@shell3.ba.best.com> <67a2cc$mbg$1@nntp4.u.washington.edu> Date: Thu, 18 Dec 1997 04:41:36 GMT Lines: 28 Xref: nntp1.ba.best.com alt.sys.pdp10:3567 alt.folklore.computers:95869 dpeschel@u.washington.edu (D. Peschel) writes: >I'm curious about the undocumented aspects of the -10 family. You've mentioned >interrupts; other people mentioned device drivers, single instructions that >have dangerous/subtle/mysterious effects, and pipelining. Weren't the -10's >generally asynchronous? >So: Are there other things that are undocumented? And since they _are_ >undocumented, how do people magically know how to explot them? (Perhaps all >the sneaky code was written at DEC, as opposed to MIT or Stanford?) It might be really good if someone could start collecting any such undocumented gotchas as people remember them, or encounter them (if anyone else out there is trying to write an emulator). This could be added to a pdp-10 faq... Megan Gentry Former RT-11 Developer +--------------------------------------+-------------------------------------+ | Megan Gentry | tcp/ip (work): gentry@zk3.dec.com | | Unix Support and Engineering Group | or: gentry@rusure.enet.dec.com | | Digital Equipment Corporation | (non-work): mbg@world.std.com | | 110 Spitbrook Rd. ZK03-2/T43 | URL http://world.std.com/~mbg/ | | Nashua, NH 03062 | "Still real-time after all these | | (603) 881 1055 | years." | +--------------------------------------+-------------------------------------+ Article 3568 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntprelay.mathworks.com!news-peer-east.sprintlink.net!news-peer.sprintlink.net!news-sea-19.sprintlink.net!news-in-west.sprintlink.net!news.sprintlink.net!Sprint!199.232.56.18!news.ultranet.com!not-for-mail From: "Carl R. Friend" Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Curious about 10s. Date: Thu, 18 Dec 1997 06:07:02 -0500 Organization: as little as possible! Lines: 56 Message-ID: <349903D6.1BDD2193@stoneweb.com> References: <66gg3a$r8g$1@shell3.ba.best.com> <66rve9$a0t$1@shell3.ba.best.com> <67a2cc$mbg$1@nntp4.u.washington.edu> NNTP-Posting-Host: zephyr.ultranet.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@ultra.net X-Ultra-Time: 18 Dec 1997 11:07:09 GMT X-Mailer: Mozilla 2.0 (X11; I; Linux 2.0.29 i586) Xref: nntp1.ba.best.com alt.sys.pdp10:3568 alt.folklore.computers:95878 D. Peschel, in article nr. <67a2cc$mbg$1@nntp4.u.washington.edu>, wrote: > > I'm curious about the undocumented aspects of the -10 family. You've > mentioned interrupts; other people mentioned device drivers, single > instructions that have dangerous/subtle/mysterious effects, and > pipelining. Weren't the -10's generally asynchronous? There was some limited pipelining done in the -10 line, at least in the later models. Usually this had to do with beginning a fetch before the previous instruction completed. The PDP-6 (the -10's progenitor) and the KA-10 were asynchronous machines, the KI-10 was a synchronous time-state machine, and the KL- and KS-10 were microcode implementations (pretty much by definition synchronous). > So: Are there other things that are undocumented? And since they > _are_ undocumented, how do people magically know how to explot them? One of the reasons that the -10 looms so large in hacker (_true_ hacker, that is - not the modern usage) lore is that one got to spend lots of time on them and learn all the little ins and outs of them. Usually an undocumented "oddity" (unlike now where an un- documented problem is a "bug" and a documented one a "feature") would be discovered by accident and some enterprising soul would find a way to exploit it to advantage (speed or code size). If a "thingy" was particularly nasty, DEC would issue an ECO to fix it (the floating point divide comes to mind). If it wasn't too bad, it'd get left alone. To give an idea of the level -10 "heads" went to, one never used the "JUMP" instruction to branch to another location, one used "JRST" (Jump and ResTore flags) with no flag bits set. It used fewer CPU cycles and, hence, was a smidgen faster than "JUMP". Note that this behaviour is _not_ a bug. > (Perhaps all the sneaky code was written at DEC, as opposed to MIT > or Stanford?) I suspect that "sneaky code" came from everywhere about equally. Sometimes finding a "bug" in a machine you can exploit to improve performance is a Good Thing. Of course being "bug-compatible" is _very_ difficult to get right in a simulator and usually requires a set of logic prints and microcode listings (if applicable) to model the code after. -- ______________________________________________________________________ | | | | Carl Richard Friend (UNIX Sysadmin) | West Boylston | | Minicomputer Collector / Enthusiast | Massachusetts, USA | | mailto:carl.friend@stoneweb.com | | | http://www.ultranet.com/~engelbrt/carl/museum | ICBM: N42:22 W71:47 | |________________________________________________|_____________________| Article 3960 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.northeast.verio.net!news-feed.inet.tele.dk!bofh.vszbr.cz!rill.news.pipex.net!pipex!server1.netnews.ja.net!server2.netnews.ja.net!spruce.sucs.soton.ac.uk!wapping.ecs.soton.ac.uk!news.port.ac.uk!news From: mike.meredith@port.ac.uk (Mike Meredith) Newsgroups: alt.sys.pdp10 Subject: Re: PDP10 emulator found! Date: 12 Jun 1998 13:33:07 GMT Organization: University of Portsmouth Lines: 19 Sender: msm@port.ac.uk Message-ID: <6lramj$ctf1@tech.port.ac.uk> References: <6lmi7l$c1t$1@quest.ccsi.com> <6lr6mj$gps$1@strato.ultra.net> NNTP-Posting-Host: nutmeg.hs.port.ac.uk Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Newsreader: knews 1.0b.0 Xref: nntp1.ba.best.com alt.sys.pdp10:3960 In article <6lr6mj$gps$1@strato.ultra.net>, jmfbahciv@aol.com writes: > In article <6lmi7l$c1t$1@quest.ccsi.com>, > hnewman@infinity.ccsi.com (Harris Newman) wrote: >>I have found a lisp emulator for a pdp10. It's at: >>http://www.csd.uu.se/kurs/ktek/KT-96/bin/pdp10.lisp >> >>Can anyone get Tops-10 running on this emulator? > > This doesn't make any sense. Do you know what you're talking > about? It almost does --- it's a simulator for the PDP-10 instruction set written in Lisp. Will Tops-10 run on it ? No. -- Mike Meredith (Informatics Officer @ The University of Portsmouth) Article 4057 of alt.sys.pdp10: From: Daniel Seagraves Newsgroups: alt.sys.pdp10 Subject: ka10 emulator updated. Date: Mon, 24 Aug 1998 13:56:26 -0500 Organization: Planet Digital Network Technologies Lines: 18 Approved: Why bother? Message-ID: NNTP-Posting-Host: bony.umtec.com Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: e10@cosmic.com Path: nntp1.ba.best.com!news1.best.com!nntprelay.mathworks.com!news.idt.net!news.vphos.net!news.bctel.net!sol.pdnt.net!bony.umtec.com!root Xref: nntp1.ba.best.com alt.sys.pdp10:4057 Just checked in the rest of the halfword instructions. This puts our list of implemented instructions at ... ... all of the halfword instructions, and EXCH. Cute. I also figured out what I did to FTP and the CVS daemon, and fixed it. Sorry :) Paranoia after a breakin... You (should) be able to fetch it from bony.umtec.com in /pub/ka10. Is anyone testing this cruft to make sure it works like it's supposed to? Daniel Seagraves | I'm an International Clandestine Arms Dealer! #!/bin/perl -sp0777i Newsgroups: alt.sys.pdp10 Subject: Another e10 update... Date: Thu, 27 Aug 1998 07:40:40 -0500 Organization: Planet Digital Network Technologies Lines: 11 Approved: Why bother? Message-ID: NNTP-Posting-Host: bony.umtec.com Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII To: e10@cosmic.com Xref: nntp1.ba.best.com alt.sys.pdp10:4058 Found the ir loading bug and clobbered it, and checked in op.c, so the CPU instruction trace works. Daniel Seagraves | I'm an International Clandestine Arms Dealer! #!/bin/perl -sp0777i Newsgroups: alt.sys.pdp10 Subject: Re: KL Console Commands Date: Wed, 7 Oct 1998 13:47:16 -0500 Organization: Planet Digital Network Technologies Lines: 34 Approved: Why bother? Message-ID: References: <3613e9a8.18494117@news.clark.net> <6uvli1$o2j$1@strato.ultra.net> <3615172c.19558816@news.clark.net> <36149B71.237C@s054.aone.net.au> <361b6174.71109341@news.clark.net> <361ba313.5952698@news.clark.net> NNTP-Posting-Host: bony.umtec.com Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <361ba313.5952698@news.clark.net> Xref: news3.best.com alt.sys.pdp10:4183 On Wed, 7 Oct 1998, Philip Gagner wrote: > The DC-10 was the old, ancient, and best of all simple, terminal > controller for PDP-10s. It came out around the same time as the KA-10. > It supported multiple serial terminals. You're probably right that the > I/O instruction was conventionally DATAO TTY, but I vaguely remember > it as conventionally DATAO CON, for DATAO Console. 'course it's all in > the assignments, and was a matter of taste. Hmm... Got any manuals for it? > This means you're really going to implement all the page-fault stuff > right? Wow! As I recall, TOPS-20 was a little sensitive to page fault > timings. I'm probably wrong about this, but as I recall ITS ran on the > modified page controller that MIT used--would it ever run on a vanilla > KL? I vaguely remember an effort to make it do that, but I never heard > whether anyone got it working. I have the DECsystem 10/20 Processor Refrence Manual from 36bit.org, and I plan on emulating everything in it that I can. And if I can't, I'll fake it, or find someone to explain it to me. Thing is, I have to haul ass and get this done before all the folks with the answers goes to the big bit-bucket in the sky... I'm at 43% done, and I'm not giving up anytime soon. (Also, I'm 19, so I've got the rest of my life to hack this to prefection.) Daniel Seagraves | I'm an International Clandestine Arms Dealer! #!/bin/perl -sp0777i Newsgroups: alt.sys.pdp10 Subject: Re: KL Console Commands Date: Wed, 7 Oct 1998 13:32:03 -0700 Organization: Networks & Distributed Computing Lines: 42 Message-ID: References: <3613e9a8.18494117@news.clark.net> <6uvli1$o2j$1@strato.ultra.net> <3615172c.19558816@news.clark.net> <36149B71.237C@s054.aone.net.au> <361b6174.71109341@news.clark.net> <361ba313.5952698@news.clark.net> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 907792326 28140 (None) 140.142.17.38 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: bqt In-Reply-To: Xref: news3.best.com alt.sys.pdp10:4187 On Wed, 7 Oct 1998, Daniel Seagraves wrote: > I have the DECsystem 10/20 Processor Refrence Manual from 36bit.org, > and I plan on emulating everything in it that I can. And if I can't, I'll > fake it, or find someone to explain it to me. Thing is, I have to haul > ass and get this done before all the folks with the answers goes to the > big bit-bucket in the sky... I'm at 43% done, and I'm not giving up > anytime soon. (Also, I'm 19, so I've got the rest of my life to hack this > to prefection.) With all due respect, and with absolutely no intent to discourage you or dampen your youthful enthusiasm, I must inform you that your "43% done" is extremely optimistic. You may have implemented 43% of the user mode machine instructions, but that part of the task is less than 20% of the overall work necessary to produce an emulator that can run any of the operating systems. Or, more accurately, an emulator that can run EDDT (meaning that you have basic console I/O done) is about 20% of the task. I/O and the virtual memory system is a *huge* task, and often you are not going to find the answers from reading the manual; you will have to analyze EDDT and operating system sources. It may also be helpful to analyze microcode sources as well; in particular, you may think that you've implemented floating point correctly, until you actually try running real software and find that you have not. I wish you the best of luck in your project, but having observed the progress of Stu's incomplete emulator and Ken's completed emulator, I strongly urge you to prepare yourself for much more work than you anticipate. I would hate to see you become discouraged when reality sets in (and it will) and abandon the project. You're probably about 8% done. Good luck. -- Mark -- * RCW 19.149 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 4188 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!feed1.news.rcn.net!rcn!howland.erols.net!news.idt.net!nntp.farm.idt.net!news From: "Chris Ward" Newsgroups: alt.sys.pdp10 Subject: Re: KL Console Commands Date: Wed, 7 Oct 1998 17:11:27 -0400 Organization: IDT (Best News In The World) Lines: 41 Message-ID: <6vglh3$85b@nnrp2.farm.idt.net> References: <3613e9a8.18494117@news.clark.net> <6uvli1$o2j$1@strato.ultra.net> <3615172c.19558816@news.clark.net> <36149B71.237C@s054.aone.net.au> <361b6174.71109341@news.clark.net> <361ba313.5952698@news.clark.net> NNTP-Posting-Host: ppp-11.ts-4-bay.hck.idt.net X-Newsreader: Microsoft Outlook Express 4.72.3115.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Xref: news3.best.com alt.sys.pdp10:4188 Mark Crispin wrote in message ... >With all due respect, and with absolutely no intent to discourage you or >dampen your youthful enthusiasm, I must inform you that your "43% done" is >extremely optimistic. You may have implemented 43% of the user mode >machine instructions, but that part of the task is less than 20% of the >overall work necessary to produce an emulator that can run any of the >operating systems. > ... >I wish you the best of luck in your project, but having observed the >progress of Stu's incomplete emulator and Ken's completed emulator, I >strongly urge you to prepare yourself for much more work than you >anticipate. I would hate to see you become discouraged when reality >sets in (and it will) and abandon the project. > >You're probably about 8% done. Good luck. > >-- Mark -- I do not have Mr. Crispin's experience in the 10, but I do have to agree with him on the size of the task you have set before you. Have you thought of implementing a KS-10? It was a later machine, and though it did not have as much capacity or thoroughput as the KL, there were lots fewer options. I think it could run TOPS-10 and TOPS20, and if someone could run ITS, that would cover the audience nicely. The simpler machine, and the less complex the interface to the hardware the better if you have to emulate it. I think the KS-10 was microcoded, and emulating the microcode might be a good route too. The KL was the big boy on the block though, and there will be a lot of people who would not settle for an emulation of anything less. But as far as I can tell there is a bigger chance of you getting on a real KS-10 than a KL-10 in order to compare your results, power and housing costs being what they are. Chris Ward If you were building a machine from scratch, a PDP-10 like machine (in the fashionable 32 or 64 bits) might be made to work quite well. Article 4604 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newshub.northeast.verio.net!europa.clark.net!206.55.3.15!news.clark.net!not-for-mail From: "Timothy Stark" Newsgroups: alt.sys.pdp10 Subject: Looking for TOPS-10/20 OS software and layered products... Date: Sun, 20 Dec 1998 22:24:48 -0500 Organization: Verio East Lines: 27 Message-ID: <75kf4u$42d$1@callisto.clark.net> NNTP-Posting-Host: shark7.clark.net X-Newsreader: Microsoft Outlook Express 4.72.3110.1 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 Xref: news3.best.com alt.sys.pdp10:4604 Hello Folks: Thank you for your few answers quickly. I found http://www.36bit.org for much informaton about DEC systems and PDP-10 emulator software that still is under developement. Megan wrote statements in private message: > If you can find the hardware (and some people do have some available) > you can get a hobbyist license for any -10 software you can lay your > hands on. Oh, the hardware is dino-sized - too large for my home! Without my ownship of any hardware, I can't get any software or manuals? How about TOPS-10/20 operating systems software and any layered products? That's why I am looking for them to can run pdp-10 emulators. I can develop my own PDP-10 emulator under 'C' language. I found some manuals (PDF format) on this web site. I can sign license agreement to get manuals regardless of ownship of dino-sized hardware? right? If so, I would appericate that. Thank you! -- Tim Stark Article 4611 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10 Subject: Re: Looking for TOPS-10/20 OS software and layered products... References: <75kf4u$42d$1@callisto.clark.net> <367eb231.0@news.spies.com> Organization: D Bit, Troy, NY From: wilson@dbit.com (John Wilson) NNTP-Posting-Host: dbit.dbit.com Message-ID: <367ecc99.0@news.wizvax.net> Date: 21 Dec 1998 17:32:57 -0500 X-Trace: 21 Dec 1998 17:32:57 -0500, dbit.dbit.com Lines: 22 Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson Xref: news3.best.com alt.sys.pdp10:4611 In article <367eb231.0@news.spies.com>, Al Kossow wrote: >From article <75kf4u$42d$1@callisto.clark.net>, by "Timothy Stark" : >> I found some manuals (PDF format) on this web site. I can sign license >> agreement to get manuals regardless of ownship of dino-sized hardware? >> right? > >I don't beleive so. Eric Smith worked out this agreement with DEC so that >home owners of the real hardware had a legal way of possessing the software >for the hardware that they own (actually GETTING the software in any form >has proven to be very difficult, though..). It does not cover distribution >of the manuals or code for people to don't have the hardware. However, back in 1985 DEC issued a notice allowing anyone (actually the phrasing is "Digital customers" so if you aren't one, buy a printer ribbon or something!) to copy discontinued software documentation (only!). This was inside the front cover of one of those wonderful software documentation catalogs they used to send out, the key passage is quoted in: http://www.dbit.com/pub/pdp8/doc/README John Wilson D Bit Article 4658 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: double, double, TOIL, and trouble! Date: Thu, 21 Jan 1999 13:47:54 -0800 Organization: Networks & Distributed Computing Lines: 17 Sender: mrc@Tomobiki-Cho.CAC.Washington.EDU Message-ID: NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 916955278 36450 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: jms Xref: news3.best.com alt.sys.pdp10:4658 TOIL (Ten On an Intel Laptop) is no longer a joke; it is reality. I finally bought a PC notebook (an NEC 120LT, a 100MHz Pentium-class CPU with 32MB RAM and 2GB hard drive for $1000), and spent a couple of hours porting the PDP-10 emulator (an old, KS-only, version of klh10) to Windows 98. It was surprisingly easy, once I found the Windows console routines. It runs TOPS-20 at about twice the speed of a real 2020. This is without any tuning; among other things I'm using POSIX calls. I think that I/O (especially console I/O) would run much better using native Windows calls. -- Mark -- * RCW 19.149 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 4664 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!remarQ73!supernews.com!remarQ69!not-for-mail From: pdonini@zoomnet.net (Patrick Donini) Newsgroups: alt.sys.pdp10 Subject: Re: double, double, TOIL, and trouble! Date: Sat, 23 Jan 1999 00:40:46 GMT Organization: Posted via RemarQ, http://www.remarQ.com - Discussions start here! Lines: 36 Message-ID: <36a91891.3752816@news.zoomnet.net> References: NNTP-Posting-Host: 208.19.128.230 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: 917052048 WX0JS5BPZ80E6D013C usenet52.supernews.com X-Complaints-To: newsabuse@remarQ.com X-Newsreader: Forte Agent 1.5/32.452 Xref: news3.best.com alt.sys.pdp10:4664 On Thu, 21 Jan 1999 13:47:54 -0800, Mark Crispin wrote: >TOIL (Ten On an Intel Laptop) is no longer a joke; it is reality. > >I finally bought a PC notebook (an NEC 120LT, a 100MHz Pentium-class CPU >with 32MB RAM and 2GB hard drive for $1000), and spent a couple of hours >porting the PDP-10 emulator (an old, KS-only, version of klh10) to Windows >98. It was surprisingly easy, once I found the Windows console routines. > >It runs TOPS-20 at about twice the speed of a real 2020. This is without >any tuning; among other things I'm using POSIX calls. I think that I/O >(especially console I/O) would run much better using native Windows calls. > >-- Mark -- Mark Where can I get My copy... Just got rid of a dozen KL10's and have a couple of SC's running TOPS-10 which replaced them. Two DN20 cabnets, replace half a room of equipment, 9track tape drive, 8mm tape drives, 30 disk drives. Thanks Patrick Donini pdonini@zoomnet.net !=================================! ! Patrick M. Donini ! ! pdonini@zoomnet.net ! ! http://www.zoomnet.net/~pdonini ! !=================================! Article 4838 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: E10-KL10 new version released. Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 23 Date: Mon, 26 Apr 1999 17:12:04 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 925164382 198.199.189.6 (Mon, 26 Apr 1999 17:06:22 CDT) NNTP-Posting-Date: Mon, 26 Apr 1999 17:06:22 CDT Xref: news3.best.com alt.sys.pdp10:4838 The long-in-progress Rewrite 4 of E10-KL10 is released. This version fixes more than a few screws present in the first draft. For one, the CPU is no longer in the main loop, it's in it's own function. The state of this function is controlled by a micropc variable - One could say it was "microcoded". This means instructions are now interruptable, by the PI system as well as the console :) Also, the halfword memory system was flushed. Memory is now an array of unsigned long longs. These are much easier to deal with. Other than that, this is a complete recode. It's mostly untested at the moment as I try catching up to where I was originally. It currently lives at ftp://bony.umtec.com/pub/kl10. Have fun. Daniel Seagraves | I'm an International Clandestine Arms Dealer! #!/bin/perl -sp0777i Subject: This is bad, very bad... Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 21 Date: Wed, 15 Sep 1999 10:36:31 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 937409457 198.199.189.6 (Wed, 15 Sep 1999 10:30:57 CDT) NNTP-Posting-Date: Wed, 15 Sep 1999 10:30:57 CDT Xref: news3.best.com alt.fan.sailor-moon:318412 alt.sys.pdp10:5881 Okay, due to a large screw-up on my part[1] and a larger screw-up on my stepdad's part[2], I may lose my job today. If this happens, I'll lose my inet access for a little while. The E10KL10 page and site will go down, but I have backups, so it's OK. But, I won't be seeing anyone for awhile. Just figured I'd let everyone know... [1] I got really busy and didn't clock in for a lot of the last pay period. Yes, I know how stupid that is. Anyway, I'm not getting paid. (This has happened before.) [2] He's going to call and harass my boss. He thinks this is illegal. Either way, I'm either going to get fired over him bothering my boss, or my stepdad will make me quit. Like I said, this is not good. Daniel Seagraves | I'm an International Clandestine Arms Dealer! "In the name of the moon, will you punish me?" - Crayon Shin-Chan(?),SMS #104 "There is nothing more dangerous than a resourceful idiot." - Dilbert "Who hired you guys, anyway?" -- 8-year-old Sailor Venus to the DiC SM VAs. SailorMoon Into Eternity! | Usagi's Stormtroopers Local #42 | 36 BITS 4EVER! Article 5988 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news1.best.com!nntp.primenet.com!nntp.gctr.net!newsfeed.icl.net!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: Re: Here's another update... Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 95 Date: Tue, 28 Sep 1999 16:23:46 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 938553485 198.199.189.6 (Tue, 28 Sep 1999 16:18:05 CDT) NNTP-Posting-Date: Tue, 28 Sep 1999 16:18:05 CDT Xref: news3.best.com alt.sys.pdp10:5988 On Tue, 28 Sep 1999 jmfbahciv@aol.com wrote: [snip] > You're going to have to be more specific than that. The people > I know who can help you don't want to start cold and don't have > time to start cold. Which DDT are you talking about? DDT.EXE, > EDDT, VMDDT. Give a scenario of the failure. Be specific. Gomen, my fudge. Details follow: DDT is not one of DEC's, sorry. The first DDT I got source for was ITS's Exec DDT. It's loadable to the emulator as a SBLK format binary, and it's the file ddt.sbk on the FTP site. The MIDAS listing file (Read: source code) is ddt.lst, and ddt.asc is the result of the 36bit-to-ASCII program I ran on TOAD to get the binary to my PC. Anyway, here's how to reproduce my screw. Load the emulator: ----- bony:/kl10# decsystem10 NEW-KL10 SIM CONSOLE 127 Setting up the terminal KL> ----- Now type "RS ddt.sbk". The emulator will load the SBLK file of DDT, and go back to the not-running prompt. Then, start the emulator at 774000 (The DDT entry point). ------ KL> RS ddt.sbk [Snipped SBLK loader messages] KL> ST 774000 ------ You will get a few newlines after a few seconds, and then DDT is waiting for a command. I want to open location 774000 and see what's in it, so type "774000/". DDT will type a tab character and hang here. ------ 774000/ _ <- It hangs with the cursor here. ------ At this point, push ^\ (Control-slash). The emulator will give you the running prompt, so you have to halt the processor. ------ 774000/ %CONSOLE MODE KL: HA %REQUESTING CPU HALT KL: CPU: Console requested stop at 195 %DECSYSTEM-10 NOT RUNNING CPU: DPB 5 777103 | PC 400200,,774224 %CONSOLE MODE KL> ------ At this point I'm quite unable to determine exactly what happens. It's in the symbol-table lookup probably. I'm currently trying to check every instruction I've coded so far. I already tried single-stepping the whole thing and I never noticed anything. > And, I doubt that these people are going to want to read the > code as you have it. That's not how debugging happens. When > you ask people to "look at your code", you're asking them to > do a code review. TOPS10 just didn't do that but TOPS20 did. > DDT was a -10 product. Was it ever converted to do only -20 stuff? > How was it built for the -20 tapes? I dunno. You have to keep in mind that outside of TDs, I've never seen a live PDP-10. All I have is documentation for TOPS-20, documentation for the CPU, and the sources for ITS and TOPS-10. And some old KA-era documentation for TOPS-10 and the KA processor. Besides, I wasn't asking for people to look at MY code. My code is known broken, and I code in such a manner that even I can't understand how it works afterward ^_^. I'm asking for someone to poke through Exec DDT and see where instruction(s) don't do what they're supposed to. I can find my faults from that. Probably. Daniel Seagraves | I'm an International Clandestine Arms Dealer! "In the name of the moon, will you punish me?" - Crayon Shin-Chan(?),SMS #104 "There is nothing more dangerous than a resourceful idiot." - Dilbert "Who hired you guys, anyway?" -- 8-year-old Sailor Venus to the DiC SM VAs. SailorMoon Into Eternity! | Usagi's Stormtroopers Local #42 | 36 BITS 4EVER! Article 6005 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!144.212.100.101.MISMATCH!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!d2 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Here's another update... Date: Fri, 01 Oct 99 09:13:17 GMT Organization: UltraNet Communications, Inc. Lines: 37 Message-ID: <7t25ig$hum$2@winter.news.rcn.net> References: X-Trace: ad3C44u9oFqvNnD+8svjr5abA6oB0ckYfobLb+5tWpw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 1 Oct 1999 11:22:56 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: news3.best.com alt.sys.pdp10:6005 In article , Daniel Seagraves wrote: > >On Tue, 28 Sep 1999 jmfbahciv@aol.com wrote: > >[snip] > >> You're going to have to be more specific than that. The people >> I know who can help you don't want to start cold and don't have >> time to start cold. Which DDT are you talking about? DDT.EXE, >> EDDT, VMDDT. Give a scenario of the failure. Be specific. > >Gomen, my fudge. Details follow: > >DDT is not one of DEC's, sorry. The first DDT I got source for was ITS's >Exec DDT. It's loadable to the emulator as a SBLK format binary, and it's >the file ddt.sbk on the FTP site. The MIDAS listing file (Read: source >code) is ddt.lst, and ddt.asc is the result of the 36bit-to-ASCII program >I ran on TOAD to get the binary to my PC. Anyway, here's how to reproduce >my screw. Load the emulator: Sigh! It looks like nobody wants to mess with this can of worms. I have no idea what an ITS DDT is. I do believe (but am not sure) that DEC's EDDT (which is already in SYSTEM.EXE) is incestuously involved with the TOPS10 monitor. In other words, you seem to be trying to mix apples and oranges to produce grapes. :-) I don't know if there's any ITS gurus that read this newsgroup so I can't tell if they ever picked up our DDT sources for their use after EXE, VM, and extended addressing implementations. /BAH Subtract a hundred and four for e-mail. Article 6060 of alt.sys.pdp10: Message-ID: <37FC5DE0.169A2D6E@jetnet.ab.ca> Date: Thu, 07 Oct 1999 08:46:24 +0000 From: Ben Franchuk X-Mailer: Mozilla 4.7 [en] (X11; I; Linux 2.2.10 i586) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Linux likes Dec. Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 207.153.6.60 X-Trace: 7 Oct 1999 02:43:03 -0600, 207.153.6.60 Organization: OA Internet Lines: 13 Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!newsfeed.direct.ca!news.oanet.com!207.153.6.60 Xref: news3.best.com alt.sys.pdp10:6060 I found a link to this on "FreshMeat" a source for new linux software. @http://bony.umtec.com/pdp10.html "This page is for the e10 PDP-10 emulators. Current target (for me at least) is the KL10 Model B. The distribution can be found here. bony.umtec.com/pub/kl10. Current status is here as well." Enjoy. Ben. -- ----------------------------------------------------------------- "Everything should be made as simple as possible, but not simpler." Albert Einstein Article 6111 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: E10 132 announcement Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 41 Date: Mon, 25 Oct 1999 16:38:09 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 940887127 198.199.189.6 (Mon, 25 Oct 1999 16:32:07 CDT) NNTP-Posting-Date: Mon, 25 Oct 1999 16:32:07 CDT Xref: news3.best.com alt.sys.pdp10:6111 E10-KL10 132 is now out. This one has the (hopefully last) fix to the IO system, more RH10 code (No, it still doesn't work.), some code for the TM10A (It will become a TM10B when I finally figure out the [CENSORED] DF10), and other miscellaneous. This time around, if we start ITS at the GO: point (Skipping the SALV checks), it gets as far as turning on paging, with only one undefined instruction along the way. (It's XCTR). Paging isn't written yet, so you get a microcode stop. (The emulator will stop itself.) Speaking of paging - Does anyone know how KL10 ITS paging works? I have the microcode and can try figuring it out from that, but if anyone else knows this already and can explain it, that will save me a lot of time. (I have no idea how the microcode works. I tried reading it for some other reason, and I just confused the hell out of myself. I bet writing it was a real pain in the ass... gate this to that and skip here and so on and so forth...) SALV itself doesn't go yet. The RH10 heavily screwed up. I have absolutely no documentation for it, just the source to SALV and the comments therein. ^_^ Also, the tape stuff uses a soon-to-be-heavily-hacked copy of tapeio.[c,h], which John Wilson and Eric Smith wrote. The effect of this is that tapes from John's tape-manipulation stuff (t10backup, itstar) will (maybe) be readable to the emulator. I have to hack up some of the unimplemented stuff, and get rid of the exit()s in it. Other than that it should work mostly out-of-the-tarfile. Thanks guys. Other than that, moving right along here. This is going to be cutting it incredibly close if I make my deadline! ^_^ (For those who don't know, my goal is to have an ITS running in time for 2000.) Daniel Seagraves | I'm an International Clandestine Arms Dealer! "In the name of the moon, will you punish me?" - Crayon Shin-Chan(?),SMS #104 "There is nothing more dangerous than a resourceful idiot." - Dilbert "Who hired you guys, anyway?" -- 8-year-old Sailor Venus to the DiC SM VAs. SailorMoon Into Eternity! | Usagi's Stormtroopers Local #42 | 36 BITS 4EVER! Article 6122 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.icl.net!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: Re: Stu's emulator - help! Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 28 Date: Wed, 27 Oct 1999 14:32:33 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 941052382 198.199.189.6 (Wed, 27 Oct 1999 14:26:22 CDT) NNTP-Posting-Date: Wed, 27 Oct 1999 14:26:22 CDT Xref: news3.best.com alt.sys.pdp10:6122 On Wed, 27 Oct 1999 jmfbahciv@aol.com wrote: > But he's not running our software. So anything he asks is moot > w.r.t. to us. I just couldn't figure out why he was bothering > with TM10As and RH10s. I finally figured it out when he said > ITS. No wonder he's having problems :-). No, He's running Stu's emulator, which runs TOPS-10. However, Stu's emulator isn't freely available. I'm the one with ITS. ^_^ He's having other problems entirely. He's trying to run TOPS-10 so he can play DECWAR. > Damn, I better say... This isn't a slam on ITS; there > were good reasons we said > we wouldn't support all that old gear with SMP ship. Bleh, when I figure out what I'm doing I'll run TOPS-10 (And maybe TOPS-20, I dunno if it's worth the trouble) too. Daniel Seagraves | I'm an International Clandestine Arms Dealer! "In the name of the moon, will you punish me?" - Crayon Shin-Chan(?),SMS #104 "There is nothing more dangerous than a resourceful idiot." - Dilbert "Who hired you guys, anyway?" -- 8-year-old Sailor Venus to the DiC SM VAs. SailorMoon Into Eternity! | Usagi's Stormtroopers Local #42 | 36 BITS 4EVER! Article 6165 of alt.sys.pdp10: Sender: jcrippen@mailhost.integritysi.com Newsgroups: alt.sys.pdp10 Subject: MIDAS, LISP availability? From: james@fredbox.com (James A. Crippen) Date: 11 Nov 1999 12:22:21 -0900 Message-ID: Lines: 20 X-Newsreader: Gnus v5.6.45/XEmacs 21.1 - "Big Bend" NNTP-Posting-Host: banana.arctic.net X-Trace: 11 Nov 1999 15:24:40 -0500, banana.arctic.net Organization: Newshosting.com Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!news-feed.fnsi.net!news.siscom.net!news.arctic.net!mailhost.integritysi.com Xref: news3.best.com alt.sys.pdp10:6165 I'd like to know if anyone has a copy of the source to MIDAS from ITS. I've checked all the directories that Alan has graciously made available (SYSDOC, SYSTEM, .TECO., .INFO., KSHACK, EMACS, EMACS1) but none of them contain what I want. The reason I want them is so I can implement a MIDAS-like assembler for E10 (Daniel Seagraves's KL10 emulator) to be used in UN*X userland. If the idea pans out then I might even write an {X,FSF}Emacs mode for it too (I don't intend to translate the TECO version unless someone can convince me otherwise). I'm also ISO the sources to MacLisp. If anyone knows where I could get source to NCOMPLR, COMPLR, LIBLSP, ets., I'd be very, very grateful. I want them because I'm a Lisp and Scheme fanatic as well as a retrocomputing freak, and would thus like to merge the two psychoses. My eventual goal is to make MacLisp work on E10 so that I can play with archaic AI programs and learn more about the 'Golden Age' from the source (bad pun). BTW, this message has been posted on the e10 mailing list as well. ;james Article 6260 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!news.mindspring.net!newsfeed.mindspring.net!firehose.mindspring.com!207.69.128.190!alderson From: alderson@netcom2.netcom.com (Richard M. Alderson III) Newsgroups: alt.sys.pdp10 Subject: Re: Does a DECSYSTEM20 emulator exist? Date: 20 Jan 2000 20:35:40 GMT Organization: NETCOM On-line services Lines: 13 Message-ID: References: <3876AC22.E77A11FA@bellatlantic.net> <857e7j$hb8$1@nnrp1.deja.com> Reply-To: alderson@netcom.com NNTP-Posting-Host: c7.b7.09.66 X-Server-Date: 20 Jan 2000 20:36:39 GMT In-reply-to: hsnewman@my-deja.com's message of Sat, 08 Jan 2000 13:31:31 GMT Xref: news3.best.com alt.sys.pdp10:6260 In article <857e7j$hb8$1@nnrp1.deja.com> hsnewman@my-deja.com writes: >Additionally, there were several emulators developed for sale. All of these >seem to run Tops-20 only. I would personally want a Tops-10 emulator. "Several"? I can think of only 2--Harrenstien's and Grossman's. Further, both emulate the PDP-10 hardware, not the monitor, so either will run Tops-10 if you can lay hands on it. Rich Alderson Last LOTS Tops-20 Systems Programmer, 1984-1991 Current maintainer, MIT TECO EMACS (v. 170) Customer Advocacy, XKL LLC, 1993-now last name @ XKL dot COM Chief systems administrator, XKL LLC, 1998-now Article 6262 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!144.212.100.101.MISMATCH!newsfeed.mathworks.com!sunqbc.risq.qc.ca!newsfeed.cwix.com!newsfeed.nyc.globix.net!uunet!nyc.uu.net!ffx.uu.net!ffx2nh5!not-for-mail From: Tim Shoppa Newsgroups: alt.sys.pdp10 Subject: Re: Does a DECSYSTEM20 emulator exist? Date: Thu, 20 Jan 2000 17:50:37 -0400 Organization: Trailing Edge Technology Lines: 30 Message-ID: <38874AED.5E691430@trailing-edge.com> References: <3876AC22.E77A11FA@bellatlantic.net> <857e7j$hb8$1@nnrp1.deja.com> NNTP-Posting-Host: timaxp.trailing-edge.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: ffx2nh5.news.uu.net 948408638 23208 63.73.218.130 (20 Jan 2000 22:50:38 GMT) X-Complaints-To: news@ffx2nh5.news.uu.net NNTP-Posting-Date: 20 Jan 2000 22:50:38 GMT X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) Xref: news3.best.com alt.sys.pdp10:6262 Richard M. Alderson III wrote: > > In article <857e7j$hb8$1@nnrp1.deja.com> hsnewman@my-deja.com writes: > > >Additionally, there were several emulators developed for sale. All of these > >seem to run Tops-20 only. I would personally want a Tops-10 emulator. > > "Several"? I can think of only 2--Harrenstien's and Grossman's. > Further, both emulate the PDP-10 hardware, not the monitor, > so either will run Tops-10 if you can lay hands on it. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Anyone who is covered by the hobbyist license is welcome to lay their hands on it at http://pdp-10.trailing-edge.com/ Sources are there, too, handy for those of us struggling to get our PDP-10 emulators ready for release. While there, check out the several hundred megabytes of TOPS-10 and TOPS-20 freeware. (I've got the abstracts to the numerically ordered DECUS entries in hardcopy, they're going to appear on the HTML indices and CD-ROM Real Soon Now.) -- Tim Shoppa Email: shoppa@trailing-edge.com Trailing Edge Technology WWW: http://www.trailing-edge.com/ 7328 Bradley Blvd Voice: 301-767-5917 Bethesda, MD, USA 20817 Fax: 301-767-5927 Article 6261 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10 Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.tli.de!grolier!uunet!ams.uu.net!ffx.uu.net!world!mbg From: mbg@world.std.com (Megan) Subject: Re: Does a DECSYSTEM20 emulator exist? Message-ID: Date: Thu, 20 Jan 2000 22:35:03 GMT References: <3876AC22.E77A11FA@bellatlantic.net> <857e7j$hb8$1@nnrp1.deja.com> Organization: The World Public Access UNIX, Brookline, MA Lines: 30 Xref: news3.best.com alt.sys.pdp10:6261 alderson@netcom2.netcom.com (Richard M. Alderson III) writes: >In article <857e7j$hb8$1@nnrp1.deja.com> hsnewman@my-deja.com writes: >>Additionally, there were several emulators developed for sale. All of these >>seem to run Tops-20 only. I would personally want a Tops-10 emulator. >"Several"? I can think of only 2--Harrenstien's and Grossman's. Further, both >emulate the PDP-10 hardware, not the monitor, so either will run Tops-10 if you >can lay hands on it. But Harrentstien's isn't available, is it... and there *are* a couple more efforts going to create a pdp-10 emulator. I can think of three others right off the bat... one of which is Daniel Seagraves', who posts here occasionally... the two others I know don't want to be revealed at this time. As for TOPS-10, check out the archive site at trailing-edge.com... Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | "pdp-11 programmer - some assembler | | (603) 884 1055 | required." - mbg | +--------------------------------+-------------------------------------+ Article 6271 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Re: Does a DECSYSTEM20 emulator exist? References: <3876AC22.E77A11FA@bellatlantic.net> <857e7j$hb8$1@nnrp1.deja.com> <38874AED.5E691430@trailing-edge.com> <38883206.5B20738@sics.se> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 21 Jan 2000 17:34:22 -0800 Message-ID: Lines: 10 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 21 Jan 2000 18:17:19 -0800, ruckus.brouhaha.com Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: news3.best.com alt.sys.pdp10:6271 Per Danielsson writes: > Can someone tell me how to get the license? I tried sending a form > to the adress specified at http://pdp-10.trailing-edge.com/#license > but my mail bounced. If you mean that you were not able to snail-mail a printed and signed copy to DEC/Compaq, that's no longer necessary. Bob Supnik provided a revised license which removes that requirement, which can be seen at http://www.36bit.org/dec/license.html Article 6292 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!netnews.com!howland.erols.net!newsfeed.mindspring.net.MISMATCH!news.mindspring.net!newsfeed.mindspring.net!firehose.mindspring.com!207.69.128.190!alderson From: alderson@netcom2.netcom.com (Richard M. Alderson III) Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators Date: 25 Jan 2000 01:55:12 GMT Organization: NETCOM On-line services Lines: 18 Message-ID: References: <20000122064720.02777.00001011@ng-ft1.aol.com> <388bf682$0$228@nntp1.ba.best.com> Reply-To: alderson@netcom.com NNTP-Posting-Host: c7.b7.09.66 X-Server-Date: 25 Jan 2000 01:56:43 GMT In-reply-to: Christopher C Stacy's message of Mon, 24 Jan 2000 15:41:01 GMT Xref: news3.best.com alt.sys.pdp10:6292 In article Christopher C Stacy writes: >I wasn't aware that KLH's PDP-10 emulator was able to run TOPS-10. >I thought the microcode choices were between ITS and TOPS-20. I think that was true for the version of the emulator that did a KS-10. He later did a version that was supposed to become an official Migration Tool for the 36-bit to Alpha VMS customers; I don't think there were any of those to pay for it, and I think he took it back from Digital. It definitely did a KL, and it definitely did Tops-10 and Tops-20--I was at the DECUS BOF that talked about it. Rich Alderson Last LOTS Tops-20 Systems Programmer, 1984-1991 Current maintainer, MIT TECO EMACS (v. 170) Customer Advocacy, XKL LLC, 1993-now last name @ XKL dot COM Chief systems administrator, XKL LLC, 1998-now Article 6310 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators References: <20000129094039.25683.00000303@ng-cf1.aol.com> Organization: D Bit, Troy, NY From: wilson@dbit.com (John Wilson) NNTP-Posting-Host: dbit.dbit.com Message-ID: <38931fcd_2@news.wizvax.net> Date: 29 Jan 2000 12:13:49 -0500 X-Trace: 29 Jan 2000 12:13:49 -0500, dbit.dbit.com Lines: 18 XPident: wilson XPident: Unknown Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newspeer.monmouth.com!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson Xref: news3.best.com alt.sys.pdp10:6310 In article <20000129094039.25683.00000303@ng-cf1.aol.com>, JMFBAH wrote: >And what version of microcode are you talking about? 240? (I >think that's the correct number.) And where did you get a copy >of the development pack? Do you still have it? The KL microcode sources and assembler used to be downloadable from MIT-MC. I remember asking Don Lewine why DEC didn't freak out and sue MIT for leaving what was then a $25,000 DEC product lying around where anyone could copy it, and I liked his answer -- the $25,000 price tag is actually paying for the right to gripe to DEC about the micromachine bugs that a user would surely run into when trying to run something other than DEC's own microcode. DEC didn't have to pay any attention to bug reports from pirates, and as piracy goes, stealing the microcode isn't such a big deal, since it's only useful if you have a KL10 anyway and $25K is small change compared to what you paid for that. John Wilson D Bit Article 6313 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!feed1.news.rcn.net!rcn!not-for-mail From: "Carl R. Friend" Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators Date: Sat, 29 Jan 2000 21:38:01 -0500 Organization: as little as possible! Lines: 30 Message-ID: <3893A409.DA0CE677@prescienttech.com> References: <20000129094039.25683.00000303@ng-cf1.aol.com> <3893629C.DE234E85@bellatlantic.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: 8WL6RTE5XYZ1OTBzFrDqs0jGGKNrvqgtz/fC5WJLR5o= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 30 Jan 2000 02:38:11 GMT X-Accept-Language: en X-Mailer: Mozilla 4.5 [en] (X11; I; Linux 2.0.29 i586) Xref: news3.best.com alt.sys.pdp10:6313 HerbalGypsy/justbob wrote: > > 2060 was faster than a 780. We called it 1 MIPS. > Vax folks rated 780 at 1 VUP. KL was faster than that. That all depends. Running without its cache, a KL-10 was actually slower than the KA-10. I tried this using the timing diags back in the early '80s when I still had unfettered access to all 3 (KA, KI, and KL) types. I'll agree that a tightly-written program on the KL would beat the 11/780 every time, but one needs to remember that the 11/780 had better memory bandwidth than the -10 had. The "VUP" is useless as a performance measure -- it's not portable between architectures. It was strictly a marketing measure, so treat it as such. MIPS isn't too much better; it's more a measure of what the main- store can provide. A 1-microsecond-cycle core store can happily yield one MIPS (with no DMA/channel accesses, of course). -- +------------------------------------------------+---------------------+ | Carl Richard Friend (UNIX Sysadmin) | West Boylston | | Minicomputer Collector / Enthusiast | Massachusetts, USA | | mailto:crfriend@ma.ultranet.com +---------------------+ | http://www.ultranet.com/~crfriend/museum | ICBM: N42:22 W71:47 | +------------------------------------------------+---------------------+ Article 6314 of alt.sys.pdp10: Path: news3.best.com!nntp1.ba.best.com!not-for-mail Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators References: <20000129094039.25683.00000303@ng-cf1.aol.com> Organization: Chez Inwap From: inwap@best.com (Joe Smith) Date: 30 Jan 2000 07:18:38 GMT Lines: 22 Message-ID: <3893e5ce$0$220@nntp1.ba.best.com> NNTP-Posting-Host: shell3.ba.best.com X-Trace: nntp1.ba.best.com 949216718 220 inwap@206.184.139.134 Xref: news3.best.com alt.sys.pdp10:6314 In article , Daniel Seagraves wrote: >we mean the KL hardware + microcode. I gather that a KL10B can run either >the TOPS-10 or TOPS-20 microcode, without hardware changes, is that right? Yep. Consider a KL-1091 or an upgraded KL-1090: TOPS-10 7.01 - runs with older "TOPS-10" (KI paging) microcode. TOPS-10 7.02 - runs with newer "TOPS-20" (KL paging) microcode; uses extended addressing in Exec mode. No support for nonzero sections in User mode, although you could execute in the ACs in section 1. TOPS-10 7.03 - full support for extended addressing in User mode. All three ran on the same hardware. Or you could swap out the RP06 disk pack and run TOPS-20 on that hardware. (This is assuming that you did a full reset, forcing the console front-end to reload the microcode from the new boot pack.) -Joe -- INWAP.COM is Joe Smith, Sally Smith and our cat Murdock. See http://www.inwap.com/ for PDP-10, "ReBoot", "Shadow Raiders"/"War Planets" Article 6326 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators Date: Mon, 7 Feb 2000 11:40:10 -0800 Organization: Networks & Distributed Computing Lines: 15 Message-ID: References: <389b6dbf$0$226@nntp1.ba.best.com> <20000206093051.23572.00001532@ng-fu1.aol.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 949952422 25696 (None) 140.142.17.38 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: eon In-Reply-To: <20000206093051.23572.00001532@ng-fu1.aol.com> Xref: news3.best.com alt.sys.pdp10:6326 On 6 Feb 2000, JMFBAH wrote: > Need I point out that we didn't ship a SYSTEM.EXE? (Don't bother > me with the boot tape.) And that was because everybody had a > different hardware configuration. That, Barb, was a TOPS-10 problem. TOPS-20 shipped pre-built monitors. <<>> -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 6329 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!nntp2.deja.com!nnrp1.deja.com!not-for-mail From: hsnewman@my-deja.com Newsgroups: alt.sys.pdp10 Subject: Emulator Discussion Date: Thu, 10 Feb 2000 15:35:34 GMT Organization: Deja.com - Before you buy. Lines: 40 Message-ID: <87uls6$dvi$1@nnrp1.deja.com> NNTP-Posting-Host: 38.241.169.11 X-Article-Creation-Date: Thu Feb 10 15:35:34 2000 GMT X-Http-User-Agent: Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt) X-Http-Proxy: 1.0 PROXY:5000, 1.0 x30.deja.com:80 (Squid/1.1.22) for client 38.241.169.11 X-MyDeja-Info: XMYDJUIDhsnewman Xref: news3.best.com alt.sys.pdp10:6329 Here is my two cents on the emulator projects: My goal of running an emulator is to run decwars on the internet FREE. I have a T3 available, and could set this up easily. My requirements are that the emulator supports networking, and runs Tops-10. * Thanks to Tim Shoppa, the distributions for Tops-10, Tops-20, tools, and microcode are on the net for easy downloading, along with lots of other software. * From previous postings, it seems that each version of the OS may be hardware specific. * I have a version of an windows emulator which runs Tops-20, not in a networking environment, and has no tape support. (klh's) Are the developers writing the emulator to one of the OS's on Shoppa's site? This would allow a large # of people to use the emulator. Is there a version of Tops-10 which supports telnet or rlogin access? I know of several developers: Daniel Seagraves, Ken Harrenstein, Stu Grossman, and Megan Gentry. Finally, there seems to be two sites with emulator packages on it (less than complete) 1) Daniel Segraves ftp://bony.umtec.com/pub/kl10/ 2) Megan Gentry: ftp://ftp.std.com/ftp/pub/mbg/pdp10/ Note, several of the files on Megan's site are protected and cannot be downloaded. KX10 is Stu Grossman's and Sim10 is Megan's. Alas, the source for Ken's emulator is not released. I hope this discussion will assist the developers in achieving their goal! Sent via Deja.com http://www.deja.com/ Before you buy. Article 6330 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!su-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.berkeley.edu!newsxfer3.itd.umich.edu!srvr1.engin.umich.edu!news.cc.utah.edu!cs.utexas.edu!geraldo.cc.utexas.edu!not-for-mail From: "Noah Smith" Newsgroups: alt.sys.pdp10 Subject: Re: Emulator Discussion Date: Thu, 10 Feb 2000 11:33:23 -0600 Organization: The University of Texas at Austin; Austin, Texas Lines: 22 Message-ID: <87utml$ssr$1@geraldo.cc.utexas.edu> References: <87uls6$dvi$1@nnrp1.deja.com> NNTP-Posting-Host: noah.biosci.utexas.edu X-Trace: geraldo.cc.utexas.edu 950204949 29595 146.6.127.120 (10 Feb 2000 17:49:09 GMT) X-Complaints-To: abuse@cc.utexas.edu NNTP-Posting-Date: 10 Feb 2000 17:49:09 GMT X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.5600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.5600 Xref: news3.best.com alt.sys.pdp10:6330 Hi everyone, I have been trying to follow this discussion with exactly the same idea, getting decwars on the internet, as a reasonable project to learn more about the PDP-10 and keep it alive. A lot of the hardware discussion is over my head but it does seem like the first step is getting TOPS-10 up.?. In 96 there was some discussion of this and I vaguely remember someone here in Texas had a webpage about their PDP-10/decwars project but I got sidetracked long ago. The discussion here has really been great and it is appreciated! It would be great to collaborate on this project! I picked up a VAXstation with the vague notion that it would be aesthetically nice to try and work with DEC hardware but any platform will do. Is it feasible for TOPS-10 to either support IP or tunnel the terminal interface over IP? Cheers, Noah Article 6337 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!cs.utexas.edu!geraldo.cc.utexas.edu!not-for-mail From: "Noah Smith" Newsgroups: alt.sys.pdp10 Subject: Re: Emulator Discussion Date: Sat, 12 Feb 2000 10:18:03 -0600 Organization: The University of Texas at Austin; Austin, Texas Lines: 26 Message-ID: <88421k$qef$1@geraldo.cc.utexas.edu> References: <87uls6$dvi$1@nnrp1.deja.com> <883o6d$kl0$1@mushi.colo.neosoft.com> NNTP-Posting-Host: noah.biosci.utexas.edu X-Trace: geraldo.cc.utexas.edu 950373236 27087 146.6.127.120 (12 Feb 2000 16:33:56 GMT) X-Complaints-To: abuse@cc.utexas.edu NNTP-Posting-Date: 12 Feb 2000 16:33:56 GMT X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.5600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.5600 Xref: news3.best.com alt.sys.pdp10:6337 Actually I dont even know what a KLH is... this is bi5.bootstrap.org's telnet prompt: * Welcome to the BI TOPS-20 system, cloned from the late great SRI-NIC. * * It is running on a SUN Sparc using Solaris and the KLH10 virtual DEC-2065. * * For information about it, contact the Bootstrap Institute at 510-713-3550 * BI5, TOPS-20 Monitor 7(21245)-4 @ www.inwap.com/pdp10/models.txt shows the 2065 as: 2065 KL10-E/R(PW), big cache, MOS memory, TOPS-20 using extended addressing same difference? Cheers, Noah Article 6332 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!news.new-york.net!news.kjsl.com!news.usenet2.org!216-164-247-14 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators Date: Sat, 12 Feb 00 12:28:52 GMT Organization: UltraNet Communications, Inc. Lines: 85 Sender: bah@216-164-247-14.s14.tnt3.sbo.ma.dialup.rcn.com Message-ID: <883p09$l91$1@mushi.colo.neosoft.com> References: <389b6dbf$0$226@nntp1.ba.best.com> <20000206093051.23572.00001532@ng-fu1.aol.com> <389dd17f$0$216@nntp1.ba.best.com> NNTP-Posting-Host: 216-164-247-14.s14.tnt3.sbo.ma.dialup.rcn.com X-Trace: mushi.colo.neosoft.com 950363977 21793 216.164.247.14 (12 Feb 2000 13:59:37 GMT) X-Complaints-To: usenet@mushi.colo.neosoft.com NNTP-Posting-Date: 12 Feb 2000 13:59:37 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: news3.best.com alt.sys.pdp10:6332 In article <389dd17f$0$216@nntp1.ba.best.com>, inwap@best.com (Joe Smith) wrote: >In article <20000206093051.23572.00001532@ng-fu1.aol.com>, >JMFBAH wrote: >>>Well, for some, the goal is to be able to load up >>>an unmodified SYSTEM.EXE >>>and have it run. (This is the tack that Systems >>>Concept took.) In this >>>case, you have to emulate the old system, warts and all. >> >>Need I point out that we didn't ship a SYSTEM.EXE? (Don't bother >>me with the boot tape.) And that was because everybody had a >>different hardware configuration. > >The boot tape is what I want to run. Oh, Joe [lamenting emoticon here]. One of the problems we had with you was that you were not practical. The boot tape was _designed_ to literally get a cold system started. That meant that the code included in SYSTEM.EXE was extremely hardware dependent, stripped to the barest essentials so that a real system could get installed. Nobody ran that monitor after their system was established and all the software built to match the hardware and software configurations. > >I want something that emulates a KL or KS with at >least one RP06 disk. But you will never have an RP06 disk. Think for a second. All disk I/O done via the emulator will have to interface with the device driver of the hard disk that is on your system. Now, that means that all of TW's disk drivers will consist of the few entry points with a POPJ P, or a JRST to the resident device driver entry point. You do want to do some useful work, don't you? > >I don't mind having to run MONGEN to build a new SYSTEM.EXE. >But I seriously object to being forced to rewrite VMSER.MAC because >the darned thing is "better" than a real PDP-10. But you are going to have to use the device drivers of the hardware you are running emulator on. > >>Yes. And implementing a simulator is not the way to do it. >>Trying to make software behave like a KS or KL is a simulator >>and just fraught with CTCH22s that even I can't solve (and I >>was good at solving those). > >?! Consider all the ramifications of code dealing with the &$#%& KL cache. Consider that none of the CPUs that you will run this emulator on will be a 36-bit word architecture. Consider that none of the I/O devices will be of DEC PDP-10 manufacture. Consider that none of the network will be DECnet nor ANF10 protocol. > >>So far, we have CONOs that are different from monitor version >>to monitor version (ignore the reason why they're that way). > >What is the reason that they're that way? Hardware in some cases and a version skew of LINK in other cases. The reason we had to live with KI addressing on a KL was we didn't get the LINK that supported the KL instruction set. > >>What were those CONOs used for? > >The monitor takes advantage of new features in >the microcode or it doesn't. Answer my question. Don't give me a red herring. The function of the CONOs was to deal with specific hardware settings. Right? Look, I've never programmed these so give me a specific CONO instruction with a description of what it did...and we'll go from there. /BAH Subtract a hundred and four for e-mail. Article 6338 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators Date: Sat, 12 Feb 2000 13:42:59 -0800 Organization: Networks & Distributed Computing Lines: 76 Message-ID: References: <389b6dbf$0$226@nntp1.ba.best.com> <20000206093051.23572.00001532@ng-fu1.aol.com> <389dd17f$0$216@nntp1.ba.best.com> <883p09$l91$1@mushi.colo.neosoft.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 950391786 20802 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: sra In-Reply-To: <883p09$l91$1@mushi.colo.neosoft.com> Xref: news3.best.com alt.sys.pdp10:6338 On Sat, 12 Feb 2000 jmfbahciv@aol.com wrote: > >The boot tape is what I want to run. > > Oh, Joe [lamenting emoticon here]. One of the problems we > had with you was that you were not practical. Yeah, it is impractical to expect TOPS-10 to have the features that TOPS-20 did, such as reasonable pre-built monitors. Even the boot tape's MONITR.EXE was reasonable to run as a production monitor. Did TOPS-10 ever fix the bogosity that you had to rebuild the monitor every time you added a new disk drive? In TOPS-20, any KL monitor would run on any KL system; ditto any KS monitor on any KS system. The only real difference was that some monitors were build with smaller size limits to give users more memory on tiny systems. [Later, DEC decided to make TCP/IP monitors a separate product, but that was marketing bogosity. TCP/IP monitors worked just fine on a non-TCP system, they were just bigger.] Sorry, Barb. There *were* things that TOPS-20 had that were much better than TOPS-10. And with all due respect to Tony, one of those things was PHYSIO. If it makes you feel any better, TENEX was just as bad as TOPS-10 that way; PHYSIO was a DEC creation. In the last years of the SAIL system (it and TOPS-10 had a common ancestor in PDP-6 days) when SAIL used RP07s, we ended up throwing out all the old disk and tape code and grafting in PHYSIO in its place. Too bad that was never done for TOPS-10. But the PDP-10 groups (both of them) had major Not Invented Here disease, and were busy sniping each other while VMS was slitting both their throats. > the code included in SYSTEM.EXE was extremely > hardware dependent, stripped to the barest essentials so > that a real system could get installed. Nobody ran that > monitor after their system was established and all the software > built to match the hardware and software configurations. It didn't have to be that way. Perhaps in the KA and KI days, it did; but by the time of the KL it was possible to do things right. > All disk I/O done via the emulator will have to interface > with the device driver of the hard disk that is on your > system. Or with the filesystem system calls of the operating system. The KLH10 emulator has a single big file that represents a RP06 disk; it's opened with open(), read with read(), written with write(), and addressed with lseek(). My relatively slow NEC laptop emulates a KS10 system with MASSBUS disks and tapes faster than a real KS10. Faster boxes can emulate an ordinary KL faster than a real KL. So why spend all that time developing new device drivers, etc. when you can run your unmodified binaries from 20 years ago? > Consider all the ramifications of code dealing with the &$#%& > KL cache. Consider that none of the CPUs that you will run > this emulator on will be a 36-bit word architecture. Consider > that none of the I/O devices will be of DEC PDP-10 manufacture. > Consider that none of the network will be DECnet nor ANF10 > protocol. All of these are solved problems. DECnet isn't particularly interesting (nor for that matter anything in today's pathetic remnants of DEC). TCP is what is interesting. I don't think that DEC ever did TCP for TOPS-10, but it was done outside DEC. -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 6336 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulators References: <3893629C.DE234E85@bellatlantic.net> <20000203093548.26317.00001380@ng-cn1.aol.com> <389b6dbf$0$226@nntp1.ba.best.com> <883pp4$l91$3@mushi.colo.neosoft.com> Organization: D Bit, Troy, NY From: wilson@dbit.com (John Wilson) NNTP-Posting-Host: dbit.dbit.com X-Original-NNTP-Posting-Host: dbit.dbit.com Message-ID: <38a58c33_3@news.wizvax.net> Date: 12 Feb 2000 11:37:07 -0500 X-Trace: 12 Feb 2000 11:37:07 -0500, dbit.dbit.com Lines: 41 XPident: wilson X-Original-NNTP-Posting-Host: 199.181.141.3 XPident: Unknown Path: news3.best.com!news1.best.com!nntp.primenet.com!nntp.gctr.net!newspeer1.nac.net!news.mv.net!newspeer.phoen-x.net!news.wizvax.net!dbit.com!wilson Xref: news3.best.com alt.sys.pdp10:6336 In article <883pp4$l91$3@mushi.colo.neosoft.com>, wrote: >>I, personally, would rather have an emulator that can take a TOPS-10 >>fullsave (for either KS or KL) and just run. > >Sure, Joe. Anybody would. But it's just not practical. If you >put this constraint on an emulation project, you'll never do >any useful work under a TOPS10. But that's the whole *point* of software emulators -- to run original software unchanged, including OSes. Emulating complicated seemingly irrelevant stuff like RHes and RP06es when you're really doing I/O on PC hardware, or worse yet, read()/write() calls, may seem inelegant but it works surprisingly well, and anyway doing a faithful hardware emulation and treating the software as a "black box" really has a better chance of succeeding than trying to rewrite *all* of the OS's device drivers and memory management stuff simultaneously, while debugging your instruction set emulation, and hoping that the resulting frankenstein will run. PCs are a horrible design but ya gotta give them credit, they're *fast* (even if they aren't anywhere near as fast as they would be with a decent I/O system). In my PDP-11 emulator I go the whole 9 yards, software RP06es, TU77s, RH11s/RH70s, TU56es, TU60s, RP03s, RS04s, DELUAs, and everything else. In some cases it takes a ridiculous amount of code to make the original device drivers happy (the whole project is up to 135K lines of code now, only about 5K of which are the base instruction set), but it all runs in a flash and the resulting system is *way* faster than the high-end PDP-11s, *especially* in I/O-intensive systems. All the yuckiness is at the endpoints of the I/O transfer, doing the math on the cyl/track/sector numbers and dealing out the ATN interrupts for seek completion and simultaneous rewinds and all that crap, but in the middle you're just blazing away copying the bits in and out of the emulated memory. Plus, a dump of a real system is often all you have to work with, so there can be chicken/egg problems otherwise. I guess I just must have *forgotten* to hack in a complete set of 8-bit-byte-oriented SCSI device drivers and dump them to tape, the last time I had my KS10 running! :-) Seriously I wish I really *had* written a TU80 boot block, keeping a TU45 alive and well is *definitely* not practical. Well, some day. John Wilson D Bit Article 6357 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newshub.northeast.verio.net!verio!feeder.qis.net!remarQ-easT!remarQ.com!supernews.com!news.mindspring.net!firehose.mindspring.com!207.69.128.180!alderson From: alderson@netcom2.netcom.com (Richard M. Alderson III) Newsgroups: alt.sys.pdp10 Subject: Re: Emulator Discussion Date: 18 Feb 2000 01:03:40 GMT Organization: NETCOM On-line services Lines: 42 Message-ID: References: <87uls6$dvi$1@nnrp1.deja.com> <883o6d$kl0$1@mushi.colo.neosoft.com> <88421k$qef$1@geraldo.cc.utexas.edu> Reply-To: alderson@netcom.com NNTP-Posting-Host: c7.b7.09.66 X-Server-Date: 18 Feb 2000 01:07:46 GMT In-reply-to: "Noah Smith"'s message of Sat, 12 Feb 2000 10:18:03 -0600 Xref: news3.best.com alt.sys.pdp10:6357 In article <88421k$qef$1@geraldo.cc.utexas.edu> "Noah Smith" writes: >Actually I dont even know what a KLH is... ^^^ KLH is Ken L. Harrenstein, who wrote a PDP-10 emulator, in two versions. The first was a KS10 emulator; the second was a KL10 emulator, which Digital tried to sell to 36-bit customers as a tool for migrating from the PDP-10 to Alpha systems running VMS. As far as I know, no one was ever that desperate. >this is bi5.bootstrap.org's telnet prompt: >* Welcome to the BI TOPS-20 system, cloned from the late great SRI-NIC. >* >* It is running on a SUN Sparc using Solaris and the KLH10 virtual DEC-2065. >* >* For information about it, contact the Bootstrap Institute at 510-713-3550 >* > BI5, TOPS-20 Monitor 7(21245)-4 >@ Cool. That means that Doug Englebart (inventor of the mouse and so much more) is running the system he created at SRI (once the only place to get an IP address on the Internet) on Ken Harrenstein's emulator on a Sparc. We tried to sell Doug a Toad-1, but the Navy (his funding agency) wouldn't go for it. >www.inwap.com/pdp10/models.txt shows the 2065 as: >2065 KL10-E/R(PW), big cache, MOS memory, TOPS-20 using extended >addressing >same difference? Well, only in that KLH's emulator looks like a 2065. Rich Alderson Last LOTS Tops-20 Systems Programmer, 1984-1991 Current maintainer, MIT TECO EMACS (v. 170) Customer Advocacy, XKL LLC, 1993-now last name @ XKL dot COM Chief systems administrator, XKL LLC, 1998-now Article 6395 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newspeer1.nac.net!news-FFM2.ecrc.net!news.mindspring.net!firehose.mindspring.com!207.69.128.180!alderson From: alderson@netcom2.netcom.com (Richard M. Alderson III) Newsgroups: alt.sys.pdp10 Subject: Re: Emulator Discussion Date: 21 Feb 2000 02:12:29 GMT Organization: NETCOM On-line services Lines: 38 Message-ID: References: <87uls6$dvi$1@nnrp1.deja.com><883o6d$kl0$1@mushi.colo.neosoft.com><88421k$qef$1@geraldo.cc.utexas.edu><88jur3$sj0$1@geraldo.cc.utexas.edu> <88ka0c$7vv$1@geraldo.cc.utexas.edu> Reply-To: alderson@netcom.com NNTP-Posting-Host: c7.b7.09.66 X-Server-Date: 21 Feb 2000 02:12:00 GMT In-reply-to: "Noah Smith"'s message of Fri, 18 Feb 2000 14:11:53 -0600 Xref: news3.best.com alt.sys.pdp10:6395 In article <88ka0c$7vv$1@geraldo.cc.utexas.edu> "Noah Smith" writes: >Thanks Rich, sorry. Just trying to be careful to make sure I understand the >terminology correctly. When you mentioned: >> Doug Englebart (inventor of the mouse and so much more) >> is running the system he created at SRI >the word created made me think about the statement: >> BI TOPS-20 system, cloned from the late great SRI-NIC >and wonder what SRI-NIC was and what cloning it involved. At first glance it >seemed like SRI-NIC was probly a KL and cloning it meant moving its software >to the emulator; sounds like first glance was probly right. OK, I understand, and I apologize for the confusion. SRI-NIC was indeed a KL, running Tops-20. It provided several services to the ARPANET, and to the early Internet, such as hosting the FTP repository for the RFC library, and the official HOSTS.TXT file, and so on. Doug Englebart ended up at SRI after British Telecom sold Tymnet to MCI, continuing to work on the Human Augmentation project, his ideas for using computers to increase the ability of human beings to do whatever work they wished to do. The particular piece of software I had in mind was the AUGMENT system, which ran on TENEX and on Tops-20, and which he put on SRI-NIC when he was there. Since SRI-NIC was decommissioned a while back, I was glad to see that Doug's work is continuing, now on an emulated KL. He's a very nice man, and I enjoyed working with him briefly while I was at Stanford. Rich Alderson Last LOTS Tops-20 Systems Programmer, 1984-1991 Current maintainer, MIT TECO EMACS (v. 170) Customer Advocacy, XKL LLC, 1993-now last name @ XKL dot COM Chief systems administrator, XKL LLC, 1998-now Article 7011 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!feed1.news.rcn.net!rcn!howland.erols.net!newsfeed.skycache.com!portc01.blue.aol.com!uunet!ffx.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader2.news.uu.net!not-for-mail Message-ID: <38CF5F9C.6D375362@trailing-edge.com> Date: Wed, 15 Mar 2000 10:02:04 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 SYSTEM.EXE needed, any version References: <8am8af$8nl$1@nnrp1.deja.com> <38cedf7c$0$202@nntp1.ba.best.com> <8amtto$p20$1@nnrp1.deja.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 62 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader2.news.uu.net 953132525 9314 63.73.218.130 Xref: news3.best.com alt.sys.pdp10:7011 Ric Werme wrote: > > hsnewman@my-deja.com writes: > > >In article <38cedf7c$0$202@nntp1.ba.best.com>, > >inwap@best.com (Joe Smith) wrote: > >> In article <8am8af$8nl$1@nnrp1.deja.com>, > >wrote: > >> >I am looking for a copy of the SYSTEM.EXE file from any version of > >the > >> >TOPS-10 operating system. If you have one, I would really appreciate > >> >your posting it, so I can bring TOPS-10 up on the internet. > >> > >> What are you going to be running it on? > >> Do you have an actual PDP-10 that you will be connecting to the > >Internet? > >> Do you have a PDP-10 emulator? (KL or KS? Which author?) > > >I have a PDP-10 simulator/emulator. The author is unimportant. > > Perhaps the author is unimportant to you or your goals, but I suspect > that over 90% of the subscribers to this newsgroup are interested in the > technical details. It seems odd you want to "bring TOPS-10 up on the > Internet" without sharing the technical details. Actually, it'd be entirely consistent with the history of PDP-10 emulators. I know of at least 4 fairly complete PDP-10 emulators, one of which really boots and runs TOPS-20, yet the authors of these emulators have yet to discuss technical details or release the code. There is another emulator under development, and its author has offered to let me into an "exclusive club" of inner knowledge about development, but only on the condition that I not discuss the details with the world at large. I refused membership in that "club" because I don't like that attitude, though I really do give my best wishes to the developer and attempt to help him out whenever I'm asked questions by him. Similarly, I know of several sets of TOPS-10 and TOPS-20 distribution tapes sitting in a basement or a storage space, in probably rather poor environmental conditions, just rotting away because the current owners don't want to share them. It's a shame that this "my-my-me-me-mine" attitude exists, but what's a guy to do about it? The emulator authors are entitled to do whatever they want with their own code, and the owners of a tape can just let it sit there and rot away, and we can't force them to share if they don't want to. I've tried to change this attitude with the establishment of an archive - http://pdp-10.trailing-edge.com/ - for distribution of tape and disk images for emulator development (under the conditions of the DEC 36-bit hobbyist license) but so far there have been exactly two contributors. Not to belittle the contributions of those two kind folks - we all owe our thanks to the distribution tapes they've donated - but I was kind-of hoping there'd be a more sharing attitude among the folks who posess the materials. Tim. Article 7852 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!nntp.primenet.com!nntp.gctr.net!newsfeed.cwix.com!natasha.rmii.com!abq.news.ans.net!news-w.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!newshost.transarc.com!pat From: Pat Barron Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 System Reference? Date: 31 Mar 2000 16:24:34 GMT Organization: Transarc Corporation Lines: 59 Message-ID: References: NNTP-Posting-Host: smithfield.transarc.ibm.com In-reply-to: "Timothy Stark"'s message of Thu, 30 Mar 2000 19:55:47 -0500 Xref: news3.best.com alt.sys.pdp10:7852 Well, it's certainly nice to see a message about something besides real flamewars, shell scripts, TOPS-20 vs. Unix, and/or annoying MRC ... :-) In article "Timothy Stark" writes: > I am working on my PDP-10 emulator. I implemented some opcodes. Now I am > figuring out how APR and PI system works because my PDP-10 Processor > Reference Manual might not have enough information to hold everything for > all processors (KA10, KI10, KL10, and KS10). Also I learned that PDP-10 > System Reference have more information for devices, etc. > I am looking for it right now. I'm working on a similar project (though I haven't touched it in some time). The only System Reference I can help with is the KS10 reference. Go get ftp://ftp.ai.mit.edu/pub/users/alan/kshack.tgz which is some of the KS10-specific bits from ITS. It is a gzip'ed tar file. Within that archive, you will find a file called "ref.1", which is the KS10 reference. I found that to be extremely valuable in understanding how I/O works on the KS10 (especially stuff like interacting with the CTY, which as far as I can tell, isn't documented anywhere else). There is also the KS10 microcode for ITS, and some other interesting stuff. If you are going to emulate a KS10, it is also helpful to have copies of the PDP-11 Peripherals Handbook, the PDP-11 Terminals and Communications handbook, and possibly the PDP-11/70 processor reference - all of the disk/tape/communications devices on the KS10 are UNIBUS devices (MASSBUS disk and tape via RH11-C UNIBUS/MASSBUS adapter), and you will find hints about them in these books. > I have TOPS10/TOPS20 stuffs from pdp-10.trailing-edge.com but I found out > that they are designed for KS10 processor. Will they work with other > processors? I know that APR and PI are different on each kind of > processors. I would expect that the user mode programs should work just fine across any processor, but the KS10 monitors won't run on the KL10 (or any other processor). I'm trying to track down a set of TOPS-20 V7 tapes for KL10 right now, but I don't know if I will be successful (I also have a line on a complete set of TOPS-20 tapes for KS10 - what's on the trailing-edge archive just a the moment isn't complete). I will turn this material over to Tim Shoppa for archiving if I am successful. I certainly don't want to discourage you from implementing whatever you wish, but you will probably have a much greater chance for success if you work towards the KS10; it's a much simpler machine than the KL10, and the low-level docs you will need are more easily available for the KS10 than for other processors. I would not think that emulating a KI10 or KA10 would be that interesting, as you will probably have difficulty finding software to run on it - emulating a KA10 would actually be really interesting to me, if I could find a hardware description of the BBN Pager somewhere, and if I could find a copy of TENEX somewhere, but I have not been able to locate either of those two items. In the unlikely event that you could find software for it, one might also try emulating the PDP-6, but I've never seen any archived PDP-6 software anywhere ... --Pat. Article 7874 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!hermes.visi.com!news-out.visi.com!newsfeed.wirehub.nl!news.lightning.net!abq.news.ans.net!news-w.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!newshost.transarc.com!pat From: Pat Barron Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 System Reference? Date: 31 Mar 2000 21:16:48 GMT Organization: Transarc Corporation Lines: 15 Distribution: world Message-ID: References: NNTP-Posting-Host: smithfield.transarc.ibm.com In-reply-to: Pat Barron's message of 31 Mar 2000 16:24:34 GMT Xref: news3.best.com alt.sys.pdp10:7874 In article Pat Barron writes: > I'm working on a similar project (though I haven't touched it in some > time). The only System Reference I can help with is the KS10 reference. > Go get ftp://ftp.ai.mit.edu/pub/users/alan/kshack.tgz which is some of > the KS10-specific bits from ITS. Ooops! The correct URL is: ftp://ftp.ai.mit.edu/pub/users/alan/its/kshack.tgz This was originally missing ... ^^^ Sorry! --Pat. Article 8373 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.direct.ca!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: Tue, 2 May 2000 10:46:04 -0700 Organization: Networks & Distributed Computing Lines: 20 Message-ID: References: <8el6ii$dhe$1@spies.com> <8emeb7$jpj$1@bob.news.rcn.net> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 957289566 24162 (None) 140.142.17.35 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: rded To: jmfbahciv@aol.com In-Reply-To: <8emeb7$jpj$1@bob.news.rcn.net> Xref: news3.best.com alt.sys.pdp10:8373 On Tue, 2 May 2000 jmfbahciv@aol.com wrote: > It makes no sense to > have a TOPS10 run on top of Win98 on top of DOS. > Besides, it's almost obscene to do this. I ported an old version (KS10) on KLH's emulator to both Mac and to Win9x/NT/Win2K. On a modestly fast Pentium class machine, it emulates a KS10 faster than a KS10 is a KS10. I contend that it can be considered to be a real PDP-10, on the grounds that if you don't because of the emulation, then you must also consider anything after a KI not to be a "real PDP-10" since the later machines were all microcoded. -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 8379 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!newsfeed.concentric.net!global-news-master From: sjm Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: 03 May 2000 00:30:34 EDT Organization: The Consulate of the Republic of Belaria Lines: 62 Sender: sjm Message-ID: <8eoa1a$7p1@journal.concentric.net> References: NNTP-Posting-Host: 216.112.97.80 X-no-archive: yes X-frog-flavor: extra-crunchy User-Agent: tin/pre-1.4-19990517 ("Psychonaut") (UNIX) (Linux/2.2.12 (i686)) Xref: news3.best.com alt.sys.pdp10:8379 Forget it. There are no PDP-10 emulators. There never have been, and there never will be. From time to time, someone will try to tell you that someone named "KLH" wrote one. It's a myth. He never existed, and there's certainly no emulator. There are at least forty-eight other non-existent emulators that people will, from time to time, try to convince you are real. No one has offered convincing evidence that any of them have ever actually existed. They are figments of fevered, wishful imaginations. Your only hope is that one of the four-hundred-and-eighty ongoing projects to write a PDP-10 emulator will eventually produce something that compiles. At least a third of these projects are also completely imaginary, so don't get your hopes up. Those that aren't will be ready for (limited) public review under strict NDA within eight years if current agressive development schedules are met. TOPS-10 and TOPS-20 software is available online thanks to Tim Shoppa and some anonymous donors. Plenty more would be, but the community has largely, and very wisely, decided that it would be far better if the existing software were allowed to mildew unsaved in basements. This will preserve its purity so it is not touched by infidels. So, there, you see the situation hopeless. The only way to win is not to play. On a serious note, I don't mean to disparage the very real hard work of people like Daniel Seagraves, Megan Gentry, and (apparently) Timothy Stark. But I'm getting tired of hearing about PDP-10 emulators. I've been hearing about KLH's, Stu's and others for what feels like DECADES, but so far absolutely NO ONE has provided any proof that they're real. I have no reason to doubt that they are, but you've all got to know how frustrating it is by now. I've never used a PDP-10, in any of its many forms. I've wanted to for a very long time. I've been interested in finding out what made TOPS-10 so special for so many people, but I don't think I'm going to any time soon, unless one of these MYSTICAL emulators eventually appears. Some of you may be saying to yourself "Jeesh, this guy has no right to complain. Has he ever tried to write an emulator himself? Has he even OFFERED to help Megan or Daniel? What a twit." And you'd be right. I've never done a lick of work to make it happen. But that doesn't change how silly the whole situation is. I mean, come on. Yeah, I still hope something will come of all these emulators. But I'm not holding my breath. I do wish that what sounds like a very interesting and historical technology could be preserved for future use. If we're all lucky, it will be, eventually. -Seth -- "As a general rule, the man in the habit of murdering | Seth Morabito bookbinders, though he performs a distinct service | sethm@loomcom.com to society, only wastes his own time and takes no | personal advantage." -- Kenneth Grahame (1898) | Perth ==> * Article 8389 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!tank.news.pipex.net!pipex!ams.uu.net!zur.uu.net!ffx.uu.net!spool0.news.uu.net!reader3.news.uu.net!not-for-mail Message-ID: <390FEFB6.72AD7D9D@trailing-edge.com> Date: Wed, 03 May 2000 09:21:58 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc Subject: Re: PDP-10 Emulator References: <8eoa1a$7p1@journal.concentric.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 82 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader3.news.uu.net 957360119 7562 63.73.218.130 Xref: news3.best.com alt.sys.pdp10:8389 alt.folklore.computers:156238 comp.org.decus:6101 comp.emulators.misc:62638 sjm wrote: > > > Your only hope is that one of the four-hundred-and-eighty ongoing > projects to write a PDP-10 emulator will eventually produce > something that compiles. At least a third of these projects are also > completely imaginary, so don't get your hopes up. Those that aren't > will be ready for (limited) public review under strict NDA within > eight years if current agressive development schedules are met. A good number of the emulator (simulator?) authors are a bit afraid that releasing their current source code would require that they support it. I'm of the other opinion - if they release their current source code (I don't care if it's GPL or not GPL, I do *not* want to get into that religious war!) then others will support it. What's really encouraging is that several of the authors of the DECUS freeware in the archives have wandered across the collection and told me that they appreciate seeing software that they used or wrote themselves, sometimes 30 or more years ago. > TOPS-10 and TOPS-20 software is available online thanks to Tim Shoppa > and some anonymous donors. You're welcome :-). It's nice to see that at least I - and both the anonymous and not-so-anonymous donors mentioned at the archive home page, http://pdp-10.trailing-edge.com/ - come off as the good guys. We're really trying. I've seen some emulator progress mentioned here as a result of the sources being available and browsable, and that's a Good Thing. So far several hundred people have visited the archive site and downloaded or browsed through the software available, and that's a Good Thing too. > Plenty more would be, but the community > has largely, and very wisely, decided that it would be far better if > the existing software were allowed to mildew unsaved in basements. > This will preserve its purity so it is not touched by infidels. It really breaks my heart and pains me that so much software is out there just rotting away. Some of the current holders of the tapes don't understand that the tapes *are* still readable (sure, not everyone has 9-track and 7-track drives anymore, and not everyone has the ability to read and decode TOPS-10 BACKUP or TOPS-20 DUMPER tapes, but I do!), others don't understand that there is a wide interest in the software. There are some other organizations - such as computer museums - that probably don't understand the value of the tapes or disks they currently have in their collection. Many of these museums have members/volunteers that read this newsgroup, and I hope that those readers pay attemtion to my plea below. Thus I'll repeat my plea: If anyone has any PDP-10 software covered by the DEC 36-bit hobbyist license, please get it to me. I'll pay shipping both ways, and I'm an expert at archiving old data. Your tapes will be properly cared for, carefully read, and returned. If you want, you'll be thanked on the archive page for the stuff you provide, or if you prefer you can remain anonymous. Just don't let the software rot away! The best reference I can give is what I've done so far. Just look at http://pdp-10.trailing-edge.com/ and you'll find hundreds of megabytes of PDP-10 software, preserved in both *exact* tape or disk image form and as browsable, human-readable files. Both commercial software covered by the DEC 36-bit hobbyist license and the DECUS 10- and 20- freeware collections are available there. -- Tim Shoppa Email: shoppa@trailing-edge.com Trailing Edge Technology WWW: http://www.trailing-edge.com/ 7328 Bradley Blvd Voice: 301-767-5917 Bethesda, MD, USA 20817 Fax: 301-767-5927 Article 8398 of alt.sys.pdp10: Path: news3.best.com!nntp1.ba.best.com!not-for-mail Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc Subject: Re: PDP-10 Emulator References: <390FEFB6.72AD7D9D@trailing-edge.com> <8epvur$rep$1@ssauraab-i-1.production.compuserve.com> Organization: a user of Best Internet Communications, Inc. www.best.com From: falconer@best.com (Edward A. Falk) Date: 03 May 2000 20:27:05 GMT Lines: 64 Message-ID: <39108b99$0$206@nntp1.ba.best.com> NNTP-Posting-Host: shell3.ba.best.com X-Trace: nntp1.ba.best.com 957385625 206 falconer@206.184.139.134 Xref: news3.best.com alt.sys.pdp10:8398 alt.folklore.computers:156274 comp.org.decus:6106 comp.emulators.misc:62642 In article <8epvur$rep$1@ssauraab-i-1.production.compuserve.com>, Dave G. wrote: >Sorry for being obtuse, but what are the big problems emulating the PDP-10 >(lack of documentation, transferring the original software, copyrights, or >maybe just the scale of the endeavour)? I missed the beginning of this >thread and this is all a bit before my time. Part of the problem is that it's nearly impossible to get the documentation. But there's more to the equation. Having written an emulator once, I have a couple comments to make. Writing an emulator for a CPU is actually quite easy. But it's only a tiny fraction of the job. Writing emulators for the peripherals also needs to be done, and this can seriously complicate matters. Further, you also need to correctly emulate the bugs and undocumented features. This sounds inane, but it's true. When people write code for a specific computer, they don't debug until the code works according to the specifications of the underlying hardware, they just debug until it works. If it happens that they've stumbled across a hardware bug or undocumented feature, then the software is very likely to break the first time it's run on an emulator that does not share this bug or feature. Here's a real case in point: On the emulator I wrote, we had a third party vendor who had a major application. At one point in the application, the program ran off the end of executable code and into a data area, executing the random data as instructions until it eventually ran into real code again. As luck would have it, no disaster resulted. Otherwise, the vendor would have found out about the bug and fixed it. Our emulator reacted differently to this random data due to the presense of a large number of undefined opcodes. This caused the application to crash. Asking a third-party vendor to fix an application which ran just fine on the real computer was out of the question. Thus, we had to isolate every one of those undefined opcodes, find out what they did on the real hardware, and emulate the behavior correctly. Actually, I won't swear that we got it all correct -- we just debugged it until it worked. Every time we thought the emulator was perfected, we found a new application that tickled some previously-undiscovered undocumented behavior. Timing is another problem. I've heard that there was an early version of a PDP-10 emulator that was failing under Tops-10 because emulated disk I/O was happening too fast. I'm currently working with an emulator for a heavily pipelined architecture. Beilieve me, getting all the timings right is a bear. Bottom line: writing a working emulator is non-trivial and a bastard to support. -- -ed falk, falk@falconer.vip.best.com. See *********************#*************#* http://www.rahul.net/falk/whatToDo.html #**************F******!******!*!!**** and read 12 Simple Things You Can Do ******!***************************#** to Save the Internet **#******#*********!**WW*W**WW**** Article 8401 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!nntp.primenet.com!nntp.gctr.net!newsfeed.gamma.ru!Gamma.RU!hermes.visi.com!news-out.visi.com!uunet!ffx.uu.net!spool0.news.uu.net!reader3.news.uu.net!not-for-mail Message-ID: <39105E1C.6DC3E519@trailing-edge.com> Date: Wed, 03 May 2000 17:13:01 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc Subject: Re: PDP-10 Emulator References: <390FEFB6.72AD7D9D@trailing-edge.com> <8epvur$rep$1@ssauraab-i-1.production.compuserve.com> <39108b99$0$206@nntp1.ba.best.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 35 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader3.news.uu.net 957388382 7565 63.73.218.130 Xref: news3.best.com alt.sys.pdp10:8401 alt.folklore.computers:156281 comp.org.decus:6107 comp.emulators.misc:62643 Edward A. Falk wrote: > Part of the problem is that it's nearly impossible to get the > documentation. Actually, it's not that bad. Eric Smith has several of the relevant manuals online at http://www.36bit.org/dec/manual/ , most notably the "DECsystem-10/DECSYSTEM-20 Processor Reference Manual". This is a big win. The fact that the CPU spec is over 500 pages means that it is a lot harder than many of the microcomputers commonly emulated. > Writing an emulator for a CPU is actually quite easy. It's complicated for at least some of the PDP-10's by the presence of loadable microcode. > Further, you also need to correctly emulate the bugs and undocumented > features. This sounds inane, but it's true. Even more complicated for the PDP-10's with loadable microcode because most of the microcode versions have subtle bugs, some of which are used by the software to identify the CPU type :-). > But it's > only a tiny fraction of the job. Writing emulators for the > peripherals also needs to be done, and this can seriously complicate > matters. Absolutely. If you look at the TOPS-10 and -20 sources you can see that many times the OS writers and the peripheral hardware designers didn't agree as to how things are supposed to work. And these are guys who (at least nominally) were supposed to be working together! Tim. Article 8413 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc Subject: Re: PDP-10 Emulator References: <8epvur$rep$1@ssauraab-i-1.production.compuserve.com> <39108b99$0$206@nntp1.ba.best.com> Organization: D Bit, Troy, NY From: wilson@dbit.com (John Wilson) NNTP-Posting-Host: dbit.dbit.com X-Original-NNTP-Posting-Host: dbit.dbit.com Message-ID: <3910d117_1@news.wizvax.net> Date: 3 May 2000 21:23:35 -0400 X-Trace: 3 May 2000 21:23:35 -0400, dbit.dbit.com Lines: 53 XPident: wilson X-Original-NNTP-Posting-Host: 199.181.141.3 XPident: news Path: news3.best.com!news2.best.com!newsfeed.berkeley.edu!europa.netcrusader.net!194.176.220.129!newsfeed.icl.net!colt.net!diablo.netcom.net.uk!netcom.net.uk!newsfeed.wizvax.net!news.wizvax.net!dbit.com!wilson Xref: news3.best.com alt.sys.pdp10:8413 alt.folklore.computers:156291 comp.org.decus:6109 comp.emulators.misc:62645 In article <39108b99$0$206@nntp1.ba.best.com>, Edward A. Falk wrote: >Timing is another problem. I've heard that there was an early >version of a PDP-10 emulator that was failing under Tops-10 because >emulated disk I/O was happening too fast. IIRC, this was a spot in ITS (in DSKDMP I think?) under KLH's KS10 emulator (I heard about it but didn't see it), where it would start a disk I/O transfer using a buffer which included the code currently being executed, and then it would jump out of the way before the disk had a chance to start DMAing stuff in. But on an emulator, the easiest way to emulate disk I/O is to suspend instruction execution do the entire I/O "instantaneously" between instructions, so it would fill the whole buffer before the next instruction was fetched. Since it worked fine on real KS10s, this bug is the emulator's problem, just like the thing you described with executing random data and living to tell about it. I had similar problems with my PDP-11 emulator, and worked around them by putting in a queueing system so that most operations of most devices can be delayed by a configurable # of instruction fetches before that operation is begun (at which point it still may happen in a flash, if it's to a file or device that's accessed using blocking I/O). It's a big pain but this makes it possible to address the actual problem, that the user code assumes it can always finish the last N instructions before a particular spot, before there's any possibility that the device will have completed (or even begun) the operation. Anyway, although I've never finished a KS10 emulator, I've done plenty of work on one and I really think that the instruction set *is* the main concern. It's freakin' huge!!! The paging and UMR and interrupt systems have to be done right too, but they aren't all *that* complicated, at least on the KS. Some of the devices will be tricky, especially the TU77 since the TM03 documentation is so full of nasty little omissions etc., but generally the device support in the OSes for the KS10 is terrible so there just aren't that many devices to worry about in the first place. Do a TU77 and an RP06 and a DZ and you're practically there! The KS10 OSes don't support many of the other zillions of older PDP-11 disk/tape devices that *would* have worked (the RK611 even has an 18-bit mode, and the RK11D can be ECOed into an RK11E too), so you can skip all of those, and the KS10 didn't live long enough to really overlap with the newer PDP-11/VAX stuff (no MSCP at all, and no DELUA, the DEUNA was around but drew too much current to actually install one). So you're off the hook for all that stuff too. AFAIK no serial mux besides the DZ11 was ever officially supported on the KS10 (which is ironic because the KS10 CPU's backplane is actually made from a DH11 PCB with different wire wrap!), there was the DUP11/KMC11 set that was used for sync serial I/O but you're probably not going to get much call for that on a Unix or Windows based emulator. The DZ is dead easy, so really, serial I/O is no biggie either. John Wilson D Bit Article 8418 of alt.sys.pdp10: From: "Henry W. Miller" Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc References: <390FEFB6.72AD7D9D@trailing-edge.com> <8epvur$rep$1@ssauraab-i-1.production.compuserve.com> <39108b99$0$206@nntp1.ba.best.com> Subject: Re: PDP-10 Emulator Date: Thu, 4 May 2000 00:44:36 -0700 Lines: 103 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4029.2901 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4029.2901 NNTP-Posting-Host: 216.126.149.196 X-Original-NNTP-Posting-Host: 216.126.149.196 Message-ID: <39112a5a@news2.starnetinc.com> X-Trace: 4 May 2000 02:44:26 -0600, 216.126.149.196 Path: news3.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.enteract.com!chicago-news-feed1.bbnplanet.com!news.gtei.net!news2.starnetinc.com!216.126.149.196 Xref: news3.best.com alt.sys.pdp10:8418 alt.folklore.computers:156306 comp.org.decus:6113 comp.emulators.misc:62650 "Edward A. Falk" wrote in message news:39108b99$0$206@nntp1.ba.best.com... > In article <8epvur$rep$1@ssauraab-i-1.production.compuserve.com>, > Dave G. wrote: > >Sorry for being obtuse, but what are the big problems emulating the PDP-10 > >(lack of documentation, transferring the original software, copyrights, or > >maybe just the scale of the endeavour)? I missed the beginning of this > >thread and this is all a bit before my time. > Ed, > Part of the problem is that it's nearly impossible to get the > documentation. But there's more to the equation. > > Having written an emulator once, I have a couple comments to make. > > Writing an emulator for a CPU is actually quite easy. But it's > only a tiny fraction of the job. Writing emulators for the > peripherals also needs to be done, and this can seriously complicate > matters. > > Further, you also need to correctly emulate the bugs and undocumented > features. This sounds inane, but it's true. > I have TONS of old documentation, much to my wife's chagrin. As soon as I get my scanner up and running, this is one of my goals: scan all of that old stuff into a human/machine readable format. This is going to have to be a low priority item, but hopefully it'll happen. When it does, and as soon as I can get a new ANONYMOUS FTP account set up, or possibly WEBify it, I'll make it available. > When people write code for a specific computer, they don't debug > until the code works according to the specifications of the underlying > hardware, they just debug until it works. If it happens that > they've stumbled across a hardware bug or undocumented feature, > then the software is very likely to break the first time it's run > on an emulator that does not share this bug or feature. > > Here's a real case in point: On the emulator I wrote, we had a > third party vendor who had a major application. At one point in > the application, the program ran off the end of executable code > and into a data area, executing the random data as instructions > until it eventually ran into real code again. As luck would have > it, no disaster resulted. Otherwise, the vendor would have found > out about the bug and fixed it. > > Our emulator reacted differently to this random data due to the > presense of a large number of undefined opcodes. This caused the > application to crash. Asking a third-party vendor to fix an > application which ran just fine on the real computer was out of > the question. Thus, we had to isolate every one of those undefined > opcodes, find out what they did on the real hardware, and emulate > the behavior correctly. Actually, I won't swear that we got it > all correct -- we just debugged it until it worked. > Here's another fun point, although it's a software issue: when I was doing the TCP/IP for TENEX close to 20 years or so ago, I found out that many versions of TENEX had different JSYS values for certain calls. Try to get people to get consistent! It made it really difficult to try to put out a unified release for people, without requiring the host system programmer to perform some hacking, other than the usual tweaking of values in STORAGE/STG. > Every time we thought the emulator was perfected, we found a new > application that tickled some previously-undiscovered undocumented > behavior. > > Timing is another problem. I've heard that there was an early > version of a PDP-10 emulator that was failing under Tops-10 because > emulated disk I/O was happening too fast. > As I recall the folklore, the Amdahl 470 had the same problem, as it ran too fast and thus broke some of the carefully defined timing loops in OS/360. > I'm currently working with an emulator for a heavily pipelined > architecture. Beilieve me, getting all the timings right is a > bear. > > Bottom line: writing a working emulator is non-trivial and a > bastard to support. > > -- > -ed falk, falk@falconer.vip.best.com. See *********************#*************#* > http://www.rahul.net/falk/whatToDo.html #**************F******!******!*!!**** > and read 12 Simple Things You Can Do ******!***************************#** > to Save the Internet **#******#*********!**WW*W**WW**** -HWM Article 8411 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!feed1.news.rcn.net!rcn!europa.netcrusader.net!152.163.239.129!portc01.blue.aol.com!portc03.blue.aol.com!nntp2.giganews.com!nntp3.giganews.com!news4.giganews.com.POSTED!not-for-mail From: "Timothy Stark" Newsgroups: alt.sys.pdp10 References: Subject: Re: PDP-10 Emulator Lines: 25 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Message-ID: NNTP-Posting-Date: Wed, 03 May 2000 19:05:58 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-O1ketSx3KOc+F7NeYLYgG+XHaxF6MAdBWn8DVdR1nU9lwVTfBHqOhryGwPgurDxdDKjg/V52sIiV+2K!rV5SnovQdLmr/KrzKIcYS8wmowDPt/8= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 3 May 2000 20:07:52 -0400 Xref: news3.best.com alt.sys.pdp10:8411 Hello Folks: I read all articles about this subject "PDP-10 Emulator". Well, there are existing PDP-10 emulators. I was able to get my emulator to print "BOOT>" prompt with KL paging system mode turned on. My goal is print a dot. I am writing my KS-10 emulator in progress and am implementing RH11/TM02/TM03 interface and testing it. I played with KSBOOT.EXE and it worked so well and did not crash my emulator. :-) I entered '/tm02' at a BOOT> prompt and got result message - "No TM02 or TM03 drive" before I started to implement my RH11 interface. Yes, I will release my KS10 emulator under GPL license soon but I do not know when yet. Then, I will work on it to expand to KL10 emulator then PDP6/KA10/KI10 later. I learned that my monitor sources support KL10 too but was compiled with KS10 default. I got them from Tim's archives. I know that a sin (fear) controls some authors keep their emulators from people's hands. Indeed, my KS10 microcode sources, monitor sources, TOPS-10 sources and other resources do help my KS10 emulator development so much. -- Tim Stark Article 8434 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!newsfeed.mathworks.com!news.voicenet.com!newspeer.monmouth.com!nntp2.deja.com!nnrp1.deja.com!not-for-mail From: hsnewman@my-deja.com Newsgroups: alt.sys.pdp10 Subject: Re: PDP10 cross assembler? Date: Thu, 04 May 2000 20:21:05 GMT Organization: Deja.com - Before you buy. Lines: 27 Message-ID: <8esm2s$orh$1@nnrp1.deja.com> References: <3911aeb5$0$208@nntp1.ba.best.com> <8esepg$g9a$1@nnrp1.deja.com> NNTP-Posting-Host: 207.19.97.134 X-Article-Creation-Date: Thu May 04 20:21:05 2000 GMT X-Http-User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows 98) X-Http-Proxy: 1.0 PROXY:5000, 1.0 x38.deja.com:80 (Squid/1.1.22) for client 207.19.97.134 X-MyDeja-Info: XMYDJUIDhsnewman Xref: news3.best.com alt.sys.pdp10:8434 There are massive KLH remarks within the source, and there is a windos executable out there (kcc.exe) also. -HN In article , Mark Crispin wrote: > On Thu, 4 May 2000 hsnewman@my-deja.com wrote: > > I am not sure about this, but there is a program call kcc which was > > written by KLH. Seems to have something to do with cross compiling.... > > KCC was a C compiler for the PDP-10, and was written by Kok Chen. > > -- Mark -- > > * RCW 19.190 notice: This email address is located in Washington State. * > * Unsolicited commercial email may be billed $500 per message. * > Science does not emerge from voting, party politics, or public debate. > > Sent via Deja.com http://www.deja.com/ Before you buy. Article 8441 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!feed1.news.rcn.net!rcn!not-for-mail From: Michael Ross Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: Fri, 05 May 2000 07:59:36 -0400 Lines: 26 Message-ID: <6jd5hsknofrqq9mkl3un749ojdv0sijl7s@4ax.com> References: <8erlne$eaa$1@bob.news.rcn.net> <8esa4h01gkc@enews1.newsguy.com> <4ls3hsgok9pjjfhm3ngsgvqc59sqj9oh71@4ax.com> <8eu8ij$iej$1@bob.news.rcn.net> <39127997.558035EA@trailing-edge.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: UIPOwG05+s9P3Lke77Uqzeepe7RohfB+ovGhI7Pewmw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 5 May 2000 11:59:41 GMT X-Newsreader: Forte Agent 1.7/32.534 Xref: news3.best.com alt.sys.pdp10:8441 On Fri, 05 May 2000 07:34:47 -0400, Tim Shoppa wrote: >jmfbahciv@aol.com wrote: >> But _none_ of the TOPS-10 sources are GPL, AFAICT. > >The GPL isn't entirely a bed of roses, y'know. snip.. >Those are *really* very good terms. Home hobbyists are allowed to >read, use, and modify the sources and binaries, and I can't think >of any more we could ask for. That was my reading of the license also. And my original points about Digital, contracts, and GPL were in respect of the *emulator* in any case. Patents and employee contracts and TOPS licences are *irrelevant* - nothing to prevent a current or former Digital employee from lending his or her efforts to an open-source/GPled -10 emulator, and if they have aquired a TOPS license through the 'home user' agreement, from running it. Mike (ex-Digital UK) Rangers Catering Corps - 'We boil for the One, we fry for the One' http://www.corestore.org Article 8808 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!news.dra.com!sn-xit-01!supernews.com!207.207.0.27.MISMATCH!nntp2.giganews.com!nntp3.giganews.com!news4.giganews.com.POSTED!not-for-mail From: "Timothy Stark" Newsgroups: alt.sys.pdp10 Subject: PDP-10 Emulator Status... Lines: 26 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Message-ID: NNTP-Posting-Date: Tue, 23 May 2000 17:15:13 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-QKHZpWP/GPonpDueiGBi9ZnU8dhFnili2ZB8CuGlpfxZ8CZ73PBmPrXTBR/ftKTJlSUh/zOkcis0HCW!r78QF3fL/I41eAXb/Ip8Kd8lrsOJlCc= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 23 May 2000 18:17:01 -0400 Xref: news3.best.com alt.sys.pdp10:8808 Folks, Well, someone showed me a snippit of XPCW from Stu's emulator. I fixed XPCW instruction and a problem disappeared quickly... I learned that User/UserIO settings follow caller's flags. I implemented flag checking for User/UserIO flag settings. A problem finally had been resolved. Three problems are left now... However, I am not happy with Stu's emulator because I found out that some readers on this group have it but refuse to share with new readers who are looking for PDP-10 emulators. I asked someone about it. He showed me GPL version on it. I told him that he can distribute it but he refused. Well, I still do not have it and look for it forever.... Ok, I will re-write RH11 interface since I received a copy of some docuements (PDP-11/70 handbook) from someone. Also, I am developing console prompts like DDT functions, attach/detach files, etc... Now I have my own ftp site. I will put GPL'd pdp-0.1 (pre-alpha) package on it soon. I broke files into multi-level directories so that multi-emulator can be developed like PDP-8, PDP-11, VAX, etc. I will let you know that. Thank you for your patience. -- Tim Stark Article 8812 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Status... References: X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 23 May 2000 21:30:41 -0700 Message-ID: Lines: 27 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 23 May 2000 21:37:18 -0700, ruckus.brouhaha.com Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.berkeley.edu!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: news3.best.com alt.sys.pdp10:8812 "Timothy Stark" wrote: > However, I am not happy with Stu's emulator because I found out that some > readers on this group have it but refuse to share with new readers who are > looking for PDP-10 emulators. I asked someone about it. He showed me GPL > version on it. I told him that he can distribute it but he refused. Well, > I still do not have it and look for it forever.... I don't know who "someone" is, but he/she is following Stu's wishes in this regard. mbg@world.std.com (Megan) writes: > GPL says that one *may* redistribute, NOT that they HAVE to. I do > have a copy of it... but I received it with the request of the > author to NOT redistribute it. I am simply honoring his request. > Also, I never saw a reference to GPL... I'll have to check. Stu intended at one time to eventually release it under the GPL, and so some versions of it were given to a few people with GPL notices in the source file. But Stu also requested that the recipients *not* publicly distribute it until he was ready to do an official release. If you get binaries of it from someone, the GPL requires that they provide the sources as well. But Megan is right that the GPL doesn't require any redistribution. I'm sure many people would be happy to see a public release of Stu's code, but it's up to Stu, not a vote of the masses. Article 8758 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc,vmsnet.pdp-11 Subject: Re: Commercial PDP-11 emulators (was: PDP-10 Emulator) References: <8eoa1a$7p1@journal.concentric.net> <390FEFB6.72AD7D9D@trailing-edge.com> Organization: D Bit, Troy, NY From: wilson@dbit.com (John Wilson) NNTP-Posting-Host: dbit.dbit.com X-Original-NNTP-Posting-Host: dbit.dbit.com Message-ID: <39257885_1@news.wizvax.net> Date: 19 May 2000 13:23:17 -0400 X-Trace: 19 May 2000 13:23:17 -0400, dbit.dbit.com Lines: 104 XPident: wilson X-Original-NNTP-Posting-Host: 199.181.141.3 XPident: news Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.cwix.com!news.mv.net!newspeer.phoen-x.net!news.wizvax.net!dbit.com!wilson Xref: news3.best.com alt.sys.pdp10:8758 alt.folklore.computers:157440 comp.org.decus:6236 comp.emulators.misc:62831 vmsnet.pdp-11:9966 In article , Lord Apollyon wrote: >Not only are all of the DECENT ones commercial, they're F*(!@*# >thousands of dollars per copy. One obnoxious software I won't mention >even brags about how cheap they are and then has the gonads to mention >their US$2,000+ price tag. What's the deal here, folks? I suppose you're talking about Ersatz-11 here. I guess the thing to ask is, how much do *you* think a commercial PDP-11 emulator should cost? How long do you think it takes to write one? How much do you think the development process costs? How many copies do you think are sold in a year? How much support does a typical customer need? Do you think the sales volume would be much higher if the price were lower? Are your expectations realistic? >Nostalgia ain't worth $2k, No it sure ain't, whatever the eBay folks may think. But production-quality commercial software is not aimed at nostalgic folks. If all you're looking for is to feel warm fuzzies about the old days, why aren't you satisfied with the demo version of E11? It's free (for non-commercial use), and has all the main features of the full version. Sure there are a number of limitations but it still compares very favorably to the various open-source PDP-11 emulators are available. Speaking of which, why don't you just use one of them? Or a real PDP-11 for that matter? >but what I wanna know is who is PAYING that >kind of price? Simple -- any commercial PDP-11 user who knows the following: - brand new PDP-11/93 used to cost $27K (many older models were even more) - currently available hardware PDP-11 replacements cost ~$10K-$15K - E11's only known commercial software competitor costs ~$11K - E11 costs ~$3K for one copy, much less in quantity, and it's easier to use and faster than any of the above -- *very* substantially faster than any of DEC's original PDP-11 CPUs - just for perspective: most Mentec PDP-11 OSes cost more than E11 E11 is a bargain for serious commercial users. I've had customers tell me that they almost passed E11 over because they figured that at such a low price, it must be a toy. It costs *far* less than any other alternative open to them, the only exception being to keep the old DEC hardware limping along with used parts. Which is good enough for some users (certainly hobbyists), but most commercial users have other concerns, like the *guaranteed* availability of spares, the cost of machine room space and power and climate control for a large machine, not to mention total system speed, and customer support. Many times, when a user buys E11, they can move to a smaller office, terminate the service contract (do you know what those cost?!), and if anything breaks they can get replacement parts at the nearest mall at 9 o'clock at night. Plus everything works faster -- for some applications, a whole order of magnitude faster. And next year they can make it much faster still, by buying whatever new AMD or Intel CPU is hot that week for a few hundred dollars, and popping it in their PC. And, D Bit's customer support is better than what many users are used to (same-day bug fixes are not unusual). What's not to like? The commercial version of E11 is not aimed at hobby users, because it's impossible to pick a price that would be fair to everyone. At its current price, it's out of the reach of almost all hobbyists. It doesn't really matter by how *much* it's out of their reach. If the price were lowered to something everyone could afford, say $50, then even if every active hobby PDP-11 user in the world bought a copy, I'd be lucky to get a couple of thousand dollars, TOTAL, for six years' work. It's a shrinking market, there's no way to create legions of new customers by fiddling with the price. Especially not when many/most old-time PDP-11 hobby users find emulators distasteful in the first place, so they only want to use real DEC hardware, which you can get for little or nothing if you know where to look. >I mean, these greedy dudes don't seem out of business >yet - so SOMEONE is willingly taking the knife in the chest (repeatedly) >for these pieces of software. Greedy, huh. Have you ever tried to run a business? I haven't been paid yet this year, although my tax prep service, lawyer, ISP, hardware vendors, and print ad folks sure have. And I sank a couple of $K into upgrading demo.dbit.com (a fast RSX11M+ system which anyone is free to use) a few months ago, for which I was rewarded by some bozo breaking in to the system and installing back doors. It's taken me over six years to write E11 -- this is not some trivial weekend project, it's over 140,000 lines of code, and it required piles of expensive DEC documentation to write, plus a *lot* of hard work on reverse-engineering the undocumented stuff on real hardware. Which ran into some real money too, for example I paid $3000 for my used KDJ11EB CPU. But I was grateful for that deal because the going rate at the time was closer to $4000. See how that works? What's so special about being a full-time emulator author, that I don't deserve to make a living at it? Do *you* do your day job for free? >I'd love to see a PDP-10 emulator, but it's of personal/nostalgic >interest only. If the author is expecting a quick-buck out of it, he >can find something else. Here's an idea -- why don't you berate existing emulator authors on an international public forum? That way, people with the know-how to write a good PDP-10 emulator for free will come flocking to your door!!! :-) :-) Talk about taking a knife in the chest. Quick buck, yeah. E11 had four years of development before the first copy sold, and that was for half price (the old, lower price, where it turned out I wasn't making money) since it was a beta version. John Wilson D Bit Article 8774 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.direct.ca!brick.direct.ca!quark.idirect.com.POSTED!not-for-mail Message-ID: <3926B6F3.131E35E5@idirect.com> From: Jerome Fine X-Mailer: Mozilla 4.03 [en] (Win95; U) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10,alt.folklore.computers,comp.org.decus,comp.emulators.misc,vmsnet.pdp-11 Subject: Re: Commercial PDP-11 emulatore (was: PDP-10 Emulator) References: <8eoa1a$7p1@journal.concentric.net> <390FEFB6.72AD7D9D@trailing-edge.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 97 Date: Sat, 20 May 2000 16:00:07 GMT NNTP-Posting-Host: 209.161.237.28 X-Complaints-To: abuse@idirect.com X-Trace: quark.idirect.com 958838407 209.161.237.28 (Sat, 20 May 2000 12:00:07 EDT) NNTP-Posting-Date: Sat, 20 May 2000 12:00:07 EDT Organization: Internet Direct - http://www.mydirect.com Xref: news3.best.com alt.sys.pdp10:8774 alt.folklore.computers:157489 comp.org.decus:6255 comp.emulators.misc:62854 vmsnet.pdp-11:9981 >Lord Apollyon wrote: > This may be way off-topic, and a rant out of left field... for that I'm > sorry. > > I was perusing looking for a PDP/11 emulator - and I seem to have > stumbled upon the quest for the Lost City of Gold.... Jerome Fine replies: If you are referring to the Supnik emulator from DEC your are 90% correct since it does a really good job, especially if you are using a UNIX platform. The source is available, so it can be modified. Unfortunately, it does not (from my last information - this might have changed) support MSCP device drivers. If, as has been inferred, you are referring to the Eratz-11 emulator for the PDP-11 by John Wilson, then I do believe that you are 1000% correct. In my opinion, this emulator (BOTH the FREE hobby version and the commercial versions) do a fantastic job. (see: http://www.dbit.com) > Not only are all of the DECENT ones commercial, they're F*(!@*# > thousands of dollars per copy. One obnoxious software I won't mention > even brags about how cheap they are and then has the gonads to mention > their US$2,000+ price tag. What's the deal here, folks? I think that you need to distinguish between the commercial version which would be extremely reasonable at even double the actual price AND the hobby version of Eratz-11 which is free. And since from personal involvement with the use of the Eratz-11 emulator, I know it is a VERY effective product, I do not understand why you are complaining. If you are using Eratz-11 for commercial purposes, you are welcome to try and find real PDP-11 hardware that you will also need to maintain. Not only will it probably cost more to run, but it will be MUCH slower than using Eratz-11 on current PC hardware (up to TEN times slower on real PDP-11/93 hardware as compared to a less expensive Pentium III 800). Since I doubt VERY much if Lord Apollyon will bother to reply to this post, I hope that this will be the end of his rant. And if you are really thinking about making a more reasonable complaint about the hobby version of Eratz-11 which is free, then maybe you should actually target the cost of the various PDP-11 operating systems. Mentec/DEC/Compaq do offer a free hobby license when using RSTS/RSX/RT-11 older versions of these operating systems (from what I have last heard - is it still available?), but ONLY when using the Supnik emulator. If you wish to legally use real PDP-11 hardware or the Eratz-11 hobby emulator, then the cost of the license/distribution for any of three OSs mentioned are far more that that $ US 2,000 Lord Apollyon is complaining about, even though Compaq/DEC is offering a free hobby license for current versions of VMS. Why the discrepancy between VMS and PDP-11 hobby license availability? > Nostalgia ain't worth $2k, but what I wanna know is who is PAYING that > kind of price? I mean, these greedy dudes don't seem out of business > yet - so SOMEONE is willingly taking the knife in the chest (repeatedly) > for these pieces of software. > > How annoying. YES!! Definitely "How annoying.", but I suspect that you have missed your target by a VERY WIDE mark. The hobby version of Eratz-11 is free. It is the actual operating systems on the PDP-11 (RSTS/RSX/RT-11) that still cost those big dollars except for those very old versions (in the case of RT-11, V5.3 from 1985) which can legally be run only under the Supnik emulator as opposed to current VMS versions which Compaq/DEC allow for hobby use on real VAX/Alpha hardware. I have seen some references in the past year that Mentec/Compaq/DEC are contemplating making the hobby license for at least RT-11 (maybe RSTS/RSX as well) less restrictive. It might be worth while to point out that while "Compaq/DEC" does allow current versions of VMS to be used on a hobby basis (well I guess I have already pointed that out a few times), "Mentec/Compaq/DEC" do not allow current PDP-11 RSTS/RSX/RT-11 to be used on a hobby basis. If I look at that aspect, it should be very easy to note the difference between "Compac/DEC" and "Mentec/Compaq/DEC". I suggest that Lord Apollyon try and figure out where the bottleneck to using current versions of RSTS/RSX/RT-11 on real PDP-11 hardware lies, let alone on a hobby version of Eratz-11. Perhaps "Compaq/DEC" do not want to spend any money on lawyers to make significant changes to the present hobby license for PDP-11 OSs, but a trivial change which at least allows even an 8 year old version of RT-11 from 1992 on the Supnik emulator would only need the version number for RT-11 changed from V5.3 to V5.6 and if "Mentec/Compaq/DEC" were doing that much, going all the way to V5.7 which will soon be 2 years old in November of 2000 would be even more reasonable if sales of V5.7 have now fallen to a very low level. In fact, if more hobby users of V5.7 of RT-11 were interested, future commercial sales might occur for a longer period since there might be a larger support pool that would still be interested in helping commercial users. And then again, maybe not since anyone who is still interested in using RT-11 is likely to already be still using RT-11 and will not stop just because they don't have a Y2K compliant version to use. I suspect that by now, almost everyone still using a non-Y2K compliant version of RT-11 will continue doing so rather than spending $ US 1600 for a V5.7 distribution. Does anyone else have any additional comments? Sincerely yours, Jerome Fine Article 8814 of alt.sys.pdp10: From: mc@hack.org Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: 24 May 2000 11:52:42 +0200 Organization: Temple of the Moby Hack Lines: 28 Sender: mc@caerbannogh.area41.hack.org Message-ID: <87u2foz34l.fsf@caerbannogh.area41.hack.org> References: <8eoa1a$7p1@journal.concentric.net> <8eov6j$fdh$3@bob.news.rcn.net> <8epa0q$u02$1@nnrp1.deja.com> <8epvna$qh7@chronicle.concentric.net> Reply-To: mc@hack.org NNTP-Posting-Host: h246n3fls20o974.telia.com Mime-Version: 1.0 (generated by tm-edit 7.106) Content-Type: text/plain; charset=US-ASCII X-Trace: hitler.sno.pp.se 959161971 72038 212.181.166.246 (24 May 2000 09:52:51 GMT) X-Complaints-To: abuse@sno.pp.se NNTP-Posting-Date: 24 May 2000 09:52:51 GMT X-Waved: Chicken (dead). Curse all those damn UCE bots! X-Newsreader: Gnus v5.6.45/Emacs 19.34 Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!howland.erols.net!news-peer-europe.sprintlink.net!news.sprintlink.net!newnews.hk-r.se!news.lth.se!grendel.df.lth.se!snopp!not-for-mail Xref: news3.best.com alt.sys.pdp10:8814 sjm writes: > If KLH would simply pop up and tell the world how much a license for > his emulator costs, I think I'd be a lot less frustrated. But we've > never even heard that. Ken Harrenstein (KLH) poppep up on ITS-LOVERS a few months ago, I believe. He basically said that he's working to get the rights to distribute his emulator. He also said that he has recently become a father. This is most likely the reason we haven't much from him since. > 1. What platform(s) does it run on? I think it was developed on the Alpha. I guess it will run on most Unix platforms. > 2. Are sources available? According to KLH, it will be. > 6. Who currently holds copyright on the software? DEC^H^H^HCompaq, I believe. -- Mikael "MC" Cardell Temple of the Moby Hack !sirE liaH Article 8819 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!ply1.onvoy!upp1.onvoy!msc1.onvoy!onvoy.com!hardy.tc.umn.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: Wed, 24 May 2000 15:38:41 -0700 Organization: Networks & Distributed Computing Lines: 34 Message-ID: References: <8eoa1a$7p1@journal.concentric.net> <8eov6j$fdh$3@bob.news.rcn.net> <8epa0q$u02$1@nnrp1.deja.com> <8epvna$qh7@chronicle.concentric.net> <87u2foz34l.fsf@caerbannogh.area41.hack.org> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 959207924 24172 (None) 140.142.17.35 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: whlam In-Reply-To: <87u2foz34l.fsf@caerbannogh.area41.hack.org> Xref: news3.best.com alt.sys.pdp10:8819 On 24 May 2000 mc@hack.org wrote: > Ken Harrenstein (KLH) poppep up on ITS-LOVERS a few months ago, I > believe. He basically said that he's working to get the rights to > distribute his emulator. He also said that he has recently become a > father. This is most likely the reason we haven't much from him since. I might also add that in the 25 or so years that I've known him, KLH does not get involved with flamefests or flamers. If people start flaming at him, he just tunes them out. He's a hell of a nice guy, but judging from the attitudes of certain people in this newsgroup I'm not surprised that he's kept a low profile. > > 1. What platform(s) does it run on? > I think it was developed on the Alpha. I guess it will run on most > Unix platforms. The original two platforms were SPARC and 68K. I subsequently ported it to MacOS and Windows, but I don't think that KLH used that code; it was a kludgy port anyway (e.g. I do polling on the keyboard). My compressed filesystem code was a nice hack, but it's unnecessary these days. KLH subsequently concentrated on the Alpha because at the time that was the only system that could emulate a KL at KL speeds or better. One thing nice about KLH's emulator is that it works on a variety of platforms. I've been unable to get Stu's emulator to work on non-Intel. -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 8834 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!not-for-mail From: "Alan H. Martin" Newsgroups: alt.sys.pdp10 Subject: Re: PDP-10 Emulator Date: Fri, 26 May 2000 23:41:59 -0400 Lines: 18 Message-ID: <392F4407.C5B6412E@MA.UltraNet.Com> References: <8eoa1a$7p1@journal.concentric.net> <8eov6j$fdh$3@bob.news.rcn.net> <8epa0q$u02$1@nnrp1.deja.com> <8epvna$qh7@chronicle.concentric.net> <87u2foz34l.fsf@caerbannogh.area41.hack.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: PTce0x7r1fy+ffTek6VTjBEy0QcOb4+uujPHZ7j7AM4= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 27 May 2000 03:42:42 GMT X-Accept-Language: en,en-US,en-GB,es X-Mailer: Mozilla 4.7 [en] (Win95; U) Xref: news3.best.com alt.sys.pdp10:8834 Mark Crispin wrote: > One thing nice about KLH's emulator is that it works on a variety of > platforms. I've been unable to get Stu's emulator to work on non-Intel. Worked fine on Alpha/UNIX V3.0 with DEC C V5.something after a few tiny changes: 1. Stop deleting *~ in make clean (strictly optional). 2. setenv ETHER_OBJS ether_bpf.o before configure (pfopen.o comes for free in libc.a). 3. Rename routine pl to pl36 (conflicted with something in libc.a, I think). /AHM -- Alan Howard Martin AMartin@MA.UltraNet.Com Article 8851 of alt.sys.pdp10: Path: news3.best.com!news1.best.com!uninett.no!news.algonet.se!algonet!feed01.nntp.se.dataphone.net!nntp.se.dataphone.net!newsfeedZ.netscum.dQ!netscum.int!oleane.net!oleane!newsfeed.icl.net!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: Re: Daniel Seagrave's e10 Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 40 Date: Tue, 30 May 2000 12:30:13 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 959707148 198.199.189.6 (Tue, 30 May 2000 12:19:08 CDT) NNTP-Posting-Date: Tue, 30 May 2000 12:19:08 CDT Xref: news3.best.com alt.sys.pdp10:8851 On Tue, 30 May 2000, lars brinkhoff wrote: > I haven't seem any news about Daniel Seagrave's e10 PDP-10 emulator > lately. What happened to him and/or it? I'm still here, and still hacking. It's just slow going. I took a break from coding when my stepdad was diagnosed with cancer (He is making a recovery, he's alive and kicking! [swearing and swinging too ^_^]) and I never really recovered from it. I'm currently toying with a rewrite (Yes, again), this time using pthreads and other stuff. The current stumbling block is that I still have no idea how do handle math, and nobody has a decent explanation of it. I have it decoding instructions, and using the console in the normal manner (PI 0 requests, not writing directly into memory). (Basically, I ran into some DEEP DEEP pits, I forgot to allow for the KL accounting timers and the ITS XCTR (XCT Relocated) instruction, and a rewrite currently looks more attractive than the massive surgery a fix would entail. I also made some incorrect assumptions as to how the hardware worked, mostly in the PI system, which would need even more surgery to fix.) I'm trying to beat that KS guy to release but I don't think I'll get it... v_v I also have a copy of kx10, I'm hacking on that too. When I get stuck on one, I switch to the other and hack awhile. (The way they're written is different enough that I don't have to worry about crossing code.) "Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT "You'd be surprised what you can live through..." -- Anonymous "...A man can pass his family and his name down through his sons, but it's his honour that gets passed through his daughters. He can see the best and worst of life in his girls. A daughter is something far too precious, and he'll do anything to protect her." -- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver Article 4282 of alt.sys.pdp10: Path: nuq-read.news.verio.net!iad-artgen.news.verio.net!iad-peer.news.verio.net!news.verio.net!news-peer.gip.net!news.gsl.net!gip.net!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: E10 R5 "Winter Planet #1" is out. :P Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 24 Date: Mon, 5 Jun 2000 13:17:23 -0500 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 960228365 198.199.189.6 (Mon, 05 Jun 2000 13:06:05 CDT) NNTP-Posting-Date: Mon, 05 Jun 2000 13:06:05 CDT Xref: iad-artgen.news.verio.net alt.sys.pdp10:4282 E10-KL10, Revision 5, release name "Winter Planet #1". (I kept forgetting to change the version number, so I changed it to a name. Bonus points to anyone that points out where I'm getting the names from before it gets obvious. :) Anyway, this code will replace the current E10 Rewrite 2 code, once I get it up to where R2 is in functionality. This one doesn't run anything, it's a release mostly so everyone can see how things fit together, and so you all know I'm not dead yet. (No, that's not a Monty Python refrence, but now I guess it is. :) In any case, visit ftp://bony.umtec.com/pub/kl10/revision5 to get a copy. Not that it does much... "Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT "You'd be surprised what you can live through..." -- Anonymous "...A man can pass his family and his name down through his sons, but it's his honour that gets passed through his daughters. He can see the best and worst of life in his girls. A daughter is something far too precious, and he'll do anything to protect her." -- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver Article 4306 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc03.blue.aol.com!nntp2.giganews.com!nntp3.giganews.com!news5.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: ITS-Lovers mailing list? Newsgroups: alt.sys.pdp10 References: <85r99z85x8.fsf@junk.nocrew.org> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 16 Message-ID: NNTP-Posting-Date: Thu, 15 Jun 2000 09:09:15 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-JAzI0Cw9cDUis7D8es4hRMgCh/BOvR8Skfaoe9jvqcG1+rC1/c/0xluIZ1NCE98J6l1ccGMKsPfdcy6!8z7J4CN2ccB7AtBo+YBIxZthDgA= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 15 Jun 2000 14:09:15 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4306 lars brinkhoff wrote: >> Also, I am looking for ITS Reference 1.5 manual. > ftp://publications.ai.mit.edu/ai-publications/0-499/AIM-161A.ps Thank you for information. I will implement ITS instructions into my emulator. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 4307 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!news-peer-europe.sprintlink.net!news.sprintlink.net!newsfeed.bahnhof.se!news.defero.net!junk.nocrew.org!not-for-mail From: lars brinkhoff Newsgroups: alt.sys.pdp10 Subject: Re: ITS-Lovers mailing list? Organization: NoCrew Laboratories Lines: 6 Sender: lars@junk.nocrew.org Message-ID: <85itvb81zx.fsf@junk.nocrew.org> References: <85r99z85x8.fsf@junk.nocrew.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 Date: Thu, 15 Jun 2000 14:18:46 GMT NNTP-Posting-Host: 212.73.17.42 X-Complaints-To: abuse@defero.net X-Trace: news.defero.net 961078726 212.73.17.42 (Thu, 15 Jun 2000 16:18:46 MET DST) NNTP-Posting-Date: Thu, 15 Jun 2000 16:18:46 MET DST Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4307 Timothy Stark writes: > Thank you for information. I will implement ITS instructions into my > emulator. Do you have the ITS source code? ftp://ftp.swiss.ai.mit.edu/pub/its Article 4309 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!logbridge.uoregon.edu!newsfeed.direct.ca!nntp2.giganews.com!nntp3.giganews.com!news4.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Do KS10 support Ethernet? Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 21 Message-ID: <4J925.173295$701.2240811@news4.giganews.com> NNTP-Posting-Date: Thu, 15 Jun 2000 14:00:16 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-3mbjxPbBv0cVKUD1dsPoae/i88EuATFe6cP/opNwZ1YxMgWwkCRnd+jAaEoO1qXYN3cWTtkvElBvRbb!MK88l13l4kw6cGuY1TgFGBJi2YU= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 15 Jun 2000 19:00:16 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4309 Hello folks: I studied TOPS-10 7.04 monitor sources for Ethernet but found out that it supports Ethernet for KL10 only. How about KS10 processor? However, I found KMC-11, DUP-11 and DMR-11 in DNADLL.MAC. I never heard of them before. How do they works? Are they Ethernet? I know DZ-11 for terminals. I never heard of CD11 before. What is that? I studied ITS sources that I got from ftp.swiss.ai.mit.edu. It is under GPL right now. I learned that it supports Ethernet for KS10 with NI0101 interface. I learned that NI0101 have three registers only. Is that true? If not, I am looking for its specs. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 4310 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!znr.news.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!newshost.transarc.com!pat From: Pat Barron Newsgroups: alt.sys.pdp10 Subject: Re: Do KS10 support Ethernet? Date: 15 Jun 2000 19:42:54 GMT Organization: Transarc Corporation Lines: 36 Message-ID: References: <4J925.173295$701.2240811@news4.giganews.com> NNTP-Posting-Host: smithfield.transarc.ibm.com In-reply-to: Timothy Stark's message of Thu, 15 Jun 2000 19:00:16 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4310 In article <4J925.173295$701.2240811@news4.giganews.com> Timothy Stark writes: > I studied TOPS-10 7.04 monitor sources for Ethernet but found out that > it supports Ethernet for KL10 only. How about KS10 processor? However, > I found KMC-11, DUP-11 and DMR-11 in DNADLL.MAC. I never heard of them > before. How do they works? Are they Ethernet? I know DZ-11 for > terminals. I never heard of CD11 before. What is that? None of these are Ethernet. DUP11 is a synchronous serial communications interface. KMC11 is a general-purpose microprocessor (i.e., you download "microcode" to it, and it runs independently of its host processor), and saw a lot of use as a communications processor (with something like a DUP11) to do DDCMP or other protocol processing, to offload that work from the host CPU. The DMR11 is a synchronous serial interface that has its own microprocessor (don't remember if it had downloadable microcode or not) that could do either DDCMP or SDLC, or both? (I forget the exact details....). There was a prepackaged combination of the DUP11 and KMC11 you could get for the KS10, and in that configuration the two boards together were known as a DN20-BA. > I studied ITS sources that I got from ftp.swiss.ai.mit.edu. It is under > GPL right now. I learned that it supports Ethernet for KS10 with NI0101 > interface. I learned that NI0101 have three registers only. Is that true? > If not, I am looking for its specs. As far as I know, the Ethernet support in ITS was never completed and does not work. The best chance (almost certainly the only chance, if you don't want to hack the ITS code itself) you probably have of speaking TCP/IP from an emulated ITS system is to pretend you are an ACC LH/DH interface talking to an IMP .... I'm still trying to find a copy of the manual for the ACC LH/DH interface, though I have the 4.2BSD driver for it, and could probably reverse-engineer an emulation from that (in theory one could do the reverse engineering from the LH/DH driver code in ITS, but the Unix driver written in C is a lot easier to understand, IMHO....). --Pat. Article 4316 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!logbridge.uoregon.edu!news.maxwell.syr.edu!europa.netcrusader.net!199.45.45.8!cyclone1.ba-dsg.net!typhoon1.ba-dsg.net.POSTED!not-for-mail Message-ID: <3949610B.FE57CFF@bellatlantic.net> From: hg/jb Reply-To: shsrms@bellatlantic.net Organization: The Keltic League X-Mailer: Mozilla 4.73 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: Do KS10 support Ethernet? References: <4J925.173295$701.2240811@news4.giganews.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 47 Date: Thu, 15 Jun 2000 23:00:21 GMT NNTP-Posting-Host: 138.88.35.169 X-Complaints-To: newsadmin@bellatlantic.net X-Trace: typhoon1.ba-dsg.net 961110021 138.88.35.169 (Thu, 15 Jun 2000 19:00:21 EDT) NNTP-Posting-Date: Thu, 15 Jun 2000 19:00:21 EDT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4316 "Stephen H. Westin" wrote: > > Pat Barron writes: > > > In article <4J925.173295$701.2240811@news4.giganews.com> Timothy Stark writes: > > > I studied TOPS-10 7.04 monitor sources for Ethernet but found out that > > > it supports Ethernet for KL10 only. How about KS10 processor? However, > > > I found KMC-11, DUP-11 and DMR-11 in DNADLL.MAC. I never heard of them > > > before. How do they works? Are they Ethernet? I know DZ-11 for > > > terminals. I never heard of CD11 before. What is that? > > > > None of these are Ethernet. DUP11 is a synchronous serial communications > > interface. KMC11 is a general-purpose microprocessor (i.e., you download > > "microcode" to it, and it runs independently of its host processor), and > > saw a lot of use as a communications processor (with something like > > a DUP11) to do DDCMP or other protocol processing, to offload that > > work from the host CPU. The DMR11 is a synchronous serial interface > > that has its own microprocessor (don't remember if it had downloadable > > microcode or not) that could do either DDCMP or SDLC, or both? (I forget > > the exact details....). > > The DMC-11 was, as I recall, a version of the KMC-11 packaged to to > high-speed point-to-point communications. The medium was dual coax: > one cable for each direction. The DMR-11 was the "DMC replacement"; > the DMC apparently was troublesome enough (it certainly was for us) > that some reengineering was deemed necessary. The correct order is DMC, then KMC. DMC was about a 300ns proc, with 5 option cards (1-4), 1Mb, 500Kb, 250Kb, and 128 or so Kb boards with built in ethernet like nrzi modulator/demodulators, an (5) the eia rs232/v.35/RS449A interface card. The KMC is a 180ns or so proc, that can drive any of those cards with the same microcode, or as the COMM-IOP from CSS, drive sync/async (with special cards up to 8 lines) OR with the DUP/DZ turn interrupt devices into DMA devices. The KMC is a sped up DMC, an IR was added to hold the instruction for speeding up the ram. Faster ram, we could have gone faster. The IR was my idea. I think Paul Aluisi did the DMR, after my time, so I don't know all the specs on it. bob > > > > -- > -Stephen H. Westin > Any information or opinions in this message are mine: they do not > represent the position of Cornell University or any of its sponsors. Article 4317 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!su-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.direct.ca!hammer.uoregon.edu!news.hawaii.edu!not-for-mail From: Jim Thomas Newsgroups: alt.sys.pdp10 Subject: Re: Do KS10 support Ethernet? Date: 15 Jun 2000 13:09:59 -1000 Organization: Canada France Hawai`i Telescope Lines: 23 Message-ID: References: <4J925.173295$701.2240811@news4.giganews.com> NNTP-Posting-Host: atlas.cfht.hawaii.edu X-Trace: news.hawaii.edu 961110599 6562 128.171.80.135 (15 Jun 2000 23:09:59 GMT) X-Complaints-To: usenet@hawaii.edu NNTP-Posting-Date: 15 Jun 2000 23:09:59 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4317 >>>>> "Stephen" == Stephen H Westin writes: Stephen> Pat Barron writes: .... >> The >> DMR11 is a synchronous serial interface that has its own >> microprocessor (don't remember if it had downloadable microcode or >> not) that could do either DDCMP or SDLC, or both? (I forget the >> exact details....). Stephen> The DMC-11 was, as I recall, a version of the KMC-11 packaged Stephen> to to high-speed point-to-point communications. The medium was Stephen> dual coax: one cable for each direction. The DMR-11 was the Stephen> "DMC replacement"; the DMC apparently was troublesome enough Stephen> (it certainly was for us) that some reengineering was deemed Stephen> necessary. The DMC-11 could do both coax at 56K or 1M (DMC11-AL + DMC11-M[AD]) or serial 232/V.35 (DMC11-AR + DMC11-DA/FA). The DMR added 422/423/449 serial. Both had DDCMP in ROM. (ref: 1980 terminals and communications handbook 1987 PDP Systems and Options Catalog) Article 4311 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!paloalto-snf1.gtei.net!news.gtei.net!forum.apple.com!news.apple.com!haxrus.apple.com!user From: aek@spies.com (Al Kossow) Newsgroups: alt.sys.pdp10 Subject: Re: Do KS10 support Ethernet? Date: Thu, 15 Jun 2000 12:46:22 -0700 Organization: Apple Computer, Inc. Lines: 29 Message-ID: References: <4J925.173295$701.2240811@news4.giganews.com> NNTP-Posting-Host: haxrus.apple.com X-Trace: news.apple.com 961098381 4232 17.205.21.66 (15 Jun 2000 19:46:21 GMT) X-Complaints-To: usenet@news.apple.com NNTP-Posting-Date: 15 Jun 2000 19:46:21 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4311 In article <4J925.173295$701.2240811@news4.giganews.com>, Timothy Stark wrote: > Hello folks: > > I studied TOPS-10 7.04 monitor sources for Ethernet but found out that > it supports Ethernet for KL10 only. How about KS10 processor? However, > I found KMC-11, DUP-11 and DMR-11 in DNADLL.MAC. I never heard of them > before. How do they works? Are they Ethernet? I know DZ-11 for > terminals. I never heard of CD11 before. What is that? > KMC-11's are microcoded processors used as commnications processors. DUP-11's are synchronous serial interfaces. The Unibus Ethernet would be a DEUNA. > I studied ITS sources that I got from ftp.swiss.ai.mit.edu. It is under > GPL right now. I learned that it supports Ethernet for KS10 with NI0101 > interface. I learned that NI0101 have three registers only. Is that true? > If not, I am looking for its specs. > The NI-1010 was a third party Unibus ethernet interface card made by Interlan. It seems unlikely that DEC would have supported that particular interface, given that they sold their own (the DEUNA) -- The eBay Curse: "May you find everything you're looking for.." Article 4314 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!portc03.blue.aol.com!nntp2.giganews.com!nntp3.giganews.com!news5.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: DEUNA specs, etc. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 26 Message-ID: NNTP-Posting-Date: Thu, 15 Jun 2000 16:53:26 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-kQQBhYTKUkcdyOccZAyzkoq37au0ibfGA34frZE1pX2Q/O9OkJI52EiVwWtPGer1hA/9Me5P9VXySCQ!0TwhA5a8muCqpiB3uxt2C2dP45A= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 15 Jun 2000 21:53:26 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4314 Hello Folks: Thank you for information about KMC-11, CD-11, etc. I will implement DEUNA and DZ-11 first soon after I finish to implement RH11 and general data structure for multi-processing units and multi-level device mapping, etc. For DEUNA, I looked into TOPS-10 7.04 sources again and found out that it was missing UNASER.MAC! I am looking for its specs that is compatible with other platforms like VAX and PDP-11. How about KLNI specs? That's why I still need TOPS-20 monitor and exec sources for KS-10 and KL-10 processors because of TCP/IP support, etc. For ITS, I need NI1010 specs. Thank you! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 4319 of alt.sys.pdp10: Path: nuq-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: DEUNA specs, etc. Date: Thu, 15 Jun 2000 16:19:15 -0700 Organization: Networks & Distributed Computing Lines: 42 Message-ID: References: NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 961111158 33642 (None) 140.142.17.35 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: martell To: Timothy Stark In-Reply-To: Xref: dfw-artgen.news.verio.net alt.sys.pdp10:4319 On Thu, 15 Jun 2000, Timothy Stark wrote: > That's why I still need TOPS-20 monitor and exec sources for KS-10 and > KL-10 processors because of TCP/IP support, etc. The EXEC is just a user program (like a shell). You don't need EXEC sources for your emulator, although you probably want some EXEC binary (along with lots of other binaries) to make it useful to run TOPS-20... TOPS-20 never supported TCP/IP on the KS10; nor did it ever support Ethernet. I made DECnet Phase IV run on the KS (1987 or thereabouts), but that was using the KDP (KMC11 plus DUP11). This was as an add on to TOPS-20 release 4.1. About 8 or 9 years ago, I spent some time working on getting TCP/IP to work on the KS10. I actually got a TCP/IP KS10 monitor built, but it crashes at startup with a bughlt in the TCP/IP free storage routines. I made some efforts to track down the cause, and never did. TOPS-20 TCP/IP free storage was always a mess, and TCP/IP free storage bughlts were always the #1 cause of Internet TOPS-20 system crashes. On *my* KL systems, they were the *only* cause of crashes, because I had everything else thoroughly debugged... :-) Those damn TCP/IP free storage routines were great in telling you that the heap was corrupted, but utterly useless in pointing you at what corrupted it. TOPS-20 7.0 had a newer IPFREE.MAC than the one that I used for my KS TCP/IP project, which purports to be a "new version from Stu Grossman". It still crashed. I put in some code at the INTFR1 bughlt to check to see if the clobberaged looked like TCP octets, and if so I sniffed to see if I could find the block trailer and if so recreate the block header. This was sometime in the early 1990s. There were plenty of other bogons with TOPS-20 TCP/IP, including improper handling of Telnet protocol (TCOBN didn't do doubling of 0xff bytes from the application); I fixed this but DEC never adopted the fix. -- Mark -- * RCW 19.190 notice: This email address is located in Washington State. * * Unsolicited commercial email may be billed $500 per message. * Science does not emerge from voting, party politics, or public debate. Article 14 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!su-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.direct.ca!nntp2.giganews.com!nntp3.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: KS10 = no network?! Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 26 Message-ID: NNTP-Posting-Date: Fri, 16 Jun 2000 07:43:12 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-nR3q/6g+sr5PmfKLyyGHOknUQgH9Hoyt7Afx1sZEcOHh2MvhnE4cutTaGZn4VEQpWF1qtIxUjJUp1Th!xYb6XnmMcDkiRPqFe5vlsQysXBU= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 16 Jun 2000 12:43:12 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:14 Hello Folks: I am suprised to hear that. I found out that KS10 does not support network like TCP/IP, etc on both operating system execpt partial support with ITS for NI1010 interface. I would end up the network-less PDP-10 emulator. :-( Some years ago, I was a student at Gallaudet University. When they upgraded TOPS-10 to v7.03 or v7.04. They installed DECnet on DECsystem10 with PDP10-PDP11 interface box. I tried that and it worked fine. However, it have a problem. When I login into PDP10 from VAX directly for 2nd session, entire interface box crashed. Next day, it was back. I tried it again. It resulted the same! Also, I found sources under ANF-10 directory. They looks like PDP-11 assembly to me. What is ANF-10? That is for PDP-10/PDP-11 interface for DECnet, etc? Right? -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 179 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!nntp2.giganews.com!nntp3.giganews.com!news5.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: My PDP10 Emulator released to public Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 43 Message-ID: NNTP-Posting-Date: Tue, 18 Jul 2000 18:13:31 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-ULXBJio8KqVd+w6H85a8yhSamtmiDh8Au6YiGZqM2cAL1aNsZOYXQVfDT6hg9owOv4+DP2IMdkZPcI+!5KB9HpOAZV1/45aBYmeiS3iWJjw= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 18 Jul 2000 23:13:31 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:179 Hello Folks: Sorry for delays because I was so busy and had two classes for Visual Basic recently. I finally released my copy to public after I cleaned up everything. My PDP10 emulator had not print a dot yet at this time. I still am working on it right now. I am developing my backup routines to read and write backup routines. Also disk device routines are not finished yet. I rewrote my TOPS-20 paging system completely but I had a problem with CST routines. I looked into TOPS-10 sources and learned that my implementation is for Microcode v117 but TOPS-10 sources requires least v124. My TOPS-20 paging systems for V117 is not compatible with TOPS-10 v7.04! I found KS10.ULD in bootable tape file and learned that its version was v130 that supports CST on/off routines that v117 does not have. I put if-then statement around CST routines to check CBR is not zero. If CBR is zero, do not do CST routines. That's why I am looking for KS10 v130 source (.MCR or .MCL) file to complete TOPS-20 paging routines, etc. Also, I implemented disassembly routines to display instructions that had been executed so that we can understand them clearly and can find a bug easily. To to enable debugging, look in pdp10/ks10/cpu.c and changed to 1 from 0 on "pdp10_debug" at the beginning of its file. Also, I am figuring out how to discern which address is executive or user at VMA variable in paging system routines. To get my open-sourced PDP10 emulator: ftp://ftp.firesword7.net/dec/pdp10-20000718.tgz If you fix, modify, add routines, you can send me patches to update. Yes, I am learning to be the software developer for first time in my life. :-) -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 195 of alt.sys.pdp10: Newsgroups: alt.sys.pdp10 Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!ptdnetP!newsgate.ptd.net!newshub.northeast.verio.net!verio!news-feeds.jump.net!world!mbg From: mbg@world.std.com (Megan) Subject: Timothy Stark's emulator Message-ID: Date: Fri, 21 Jul 2000 00:42:15 GMT Organization: The World Public Access UNIX, Brookline, MA Lines: 27 Xref: nntp1.ba.best.com alt.sys.pdp10:195 Tim - I'm also very impressed with the work that you've done. I feel I need to comment, however, on the fact that numerous symbols, comments and indeed code itself looks like it was lifted practically verbatim from the emulator I was working on and had made available. I have no problem with use of the code (sincerest form of flattery and all that...) but since some of the code was developed by me, I would appreciate at least an attribution... For those who would like to compare, simply look at the file I have had available for some time in ftp://ftp.std.com/ftp/pub/mbg/pdp10/ Megan Gentry Former RT-11 Developer +--------------------------------+-------------------------------------+ | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | | Unix Support Engineering Group | (home): mbg!world.std.com | | Compaq Computer Corporation | addresses need '@' in place of '!' | | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | | Nashua, NH 03062 | "pdp-11 programmer - some assembler | | (603) 884 1055 | required." - mbg KB1FCA | +--------------------------------+-------------------------------------+ Article 198 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.direct.ca!look.ca!nntp2.giganews.com!nntp3.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Timothy Stark's emulator Newsgroups: alt.sys.pdp10 References: User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 34 Message-ID: NNTP-Posting-Date: Thu, 20 Jul 2000 22:30:32 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-29PXWTTRm1w42lWvHqBsslUpJUg+Xd4GX3WRyW7nAh47/KzTsFwJewzCIUTE4aqYG81v8B1XFs1eSGk!DQpW3Pqw6g7q77X9RMuDUA3UuMU= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 21 Jul 2000 03:30:32 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:198 Megan: Yeah. I learned from your emulator alot. I started to write my emulator from scratch myself. Today I compared mine with your emulator again and indeed noticed that. Yes, you are right. However, I now am able write my own code because I was new to how emulator works. I will put your credit in my code. -- Tim Stark Megan wrote: > Tim - I'm also very impressed with the work that you've done. > I feel I need to comment, however, on the fact that numerous > symbols, comments and indeed code itself looks like it was > lifted practically verbatim from the emulator I was working > on and had made available. I have no problem with use of the > code (sincerest form of flattery and all that...) but since some > of the code was developed by me, I would appreciate at least an > attribution... > For those who would like to compare, simply look at the file I > have had available for some time in > ftp://ftp.std.com/ftp/pub/mbg/pdp10/ -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 886 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feeder.via.net!logbridge.uoregon.edu!newsfeed.direct.ca!look.ca!nntp2.giganews.com!nntp3.giganews.com!news4.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Help! Write problems - logs here Newsgroups: alt.sys.pdp10 References: <5wWn5.495408$MB.7562182@news6.giganews.com> <39A04189.B605292D@prescienttech.com> <39A00C89.1BA34FF1@trailing-edge.com> <8nv0dv$2a3p$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.16 (i686)) Lines: 39 Message-ID: NNTP-Posting-Date: Wed, 23 Aug 2000 18:22:19 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-DMtofyDlu6UFAL/u99zuTfi8tImpFoDyPdN79cbBO5xw4YrZYXxIgmO+vqXKIIrJYha1nyQmm6NaWjx!whuZ3JxdiK1p1CwF21o5XWBQ8tY= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 23 Aug 2000 23:22:19 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:886 Daniel Seagraves wrote: > On Wed, 23 Aug 2000, Timothy Stark wrote: >> My emulator supports both RP04 and RP06. > Good for it. Too bad you're emulating a little toy. ^_^ Well, I am emulating a RP06 right now. I tried TOPS-10 7.03 with it and it was so happy. It complained about unknown system pack mounted. I tried to put 'TOPS-10' at HOM block (173-175). Now it displayed "#$%^# system pack mounted - considered off-line". Yes it was garbage. I had looked into ONCMOD.MAC in TOPS-10 v7.03 (not v7.04) and learned that in VOLIPTR (sp?) rountine. I have change it and will try it later. Yes, I should put "PS",,"TO" on 173, "0",,"-1" on 174, "",,"" on 175 in HOM block. I am working on TOPS-10 paging system tomorrow because FORM Disk Formatter wants TOPS-10 paging, not TOPS-20. That's why I have to run it first before install TOPS-10. RDI routines for bootable devices worked so great! I finally got rid of KSBOOT.EXE in my emulator directory. RDI is not same as RIM. RDI is the small loader within a block at page 001 (001000) and takes care of rest of binary within a file to being loaded. Its information is found in SMFILE.MAC. Now I can boot tape files on my emulator! :-) To create RDI or RIM file, use SMFILE to convert EXE to RDI or RIM binary. I did trying to boot TOPS-20 v4.1 installation tape but it complains about page fails. I will work on it later. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1229 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feeder.via.net!209.249.123.233.MISMATCH!xfer10.netnews.com!netnews.com!feed1.news.rcn.net!rcn!207-172-216-199 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: PDP10 screen log with 'SET W V F'. Date: Tue, 19 Sep 00 08:59:44 GMT Organization: UltraNet Communications, Inc. Lines: 48 Message-ID: <8q7kse$esv$1@bob.news.rcn.net> References: <8q4ok8$pp$2@bob.news.rcn.net> <8q6cel$1im$1@nntp9.atl.mindspring.net> X-Trace: kAEJTtfavB25NHiQdp+VGFxuJFe3y6145eMG6i0DXLA= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 19 Sep 2000 12:06:38 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1229 In article <8q6cel$1im$1@nntp9.atl.mindspring.net>, bugs@netcom5.netcom.com (Mark Hittinger) wrote: >Jim Thomas writes: >>PIP forever :-) The only two possibilities for writing are the SAVE >>commands and generating a core dump, right? And I'm ashamed to admit I >>can't remember how to get a core dump :-( > >There was a dump command which would cause "daemon" to attach to your >tty and then it dumped your core into a core file. It needed to attach >to your tty to rescan the command line to find your options. > >This 'daemon' attaching to your tty was extremely insecure. >I think daemon >used jobpeek uuo to actually get your core dump data - but I may be >mis-remembering that - we can all look at daemon.mac and remember now >thanks to Shoppa and friends :-) DAEMON was referred to as the swappable part of the monitor. > >Daemon was one of the first user level jobs that would start up (it ran >on [1,2]) and it took care of a lot of stuff like the fact files etc in >addition to core dumps. I made sure those FACT files went away. > >I don't think Timothy is far enough along to have daemon running but its >pretty amazing he's this far. Nope. He's not that far at all. > >Speaking of ANF-10 it might be fun to put something into the emulator that >would allow ANF-10 to run on top of a tcp-ip connection >between two or more emulators. I suspect RDH could answer this. I don't see a problem since ANF-10 was yakking at all kinds of machines. For all I know, those guys anticipated TCP/IP and put all the hooks in. They were real good about stuff like that. >Timothy ought to have dibs on host name TWINKY :-) I don't know. That would be up to CDO to give away. /BAH Subtract a hundred and four for e-mail. Article 1388 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: PDP-10 Emulator Status... Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 39 Message-ID: NNTP-Posting-Date: Fri, 29 Sep 2000 14:15:44 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-Be8zYb2oJcwKn4DOtu/llu5sifGGQ6JZNnmKQoJ8j/mcs/wjHT7H9Ih4Sff7XnOLiT81O8i3spAG0DM!la4hFZx3MF4zSLZKYnTntqdgX/Q= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 29 Sep 2000 19:15:50 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1388 Hi folks: I just finished to implement PXCT switches for all instructions that support PXCT according to PDP-10 Reference Processor manual and KS-10 microcode source code. I tested it by running varying programs and operating systems. They worked fine execpt CORE instruction that still needs to being fixed for high-segment core allocation (SMFILE.EXE tried to get 16KW high-seg allocation). About TOPS-10 v7.03, segementation violation problems on UNIX's side had been resolved by made big changes in memory access. I had informed that its system is ready and await for a dot to login as 1,2 but no a dot prompt. I looked into my debug log and noticed series of page fail traps every address! However, TOPS-10 v7.04 is running fine - big difference! About TOPS-20 v4.1, I tried to boot MTBOOT.RDI from its installation tape but still resulted the same. That's why I still need its source code to understand how do it works. However, I noticed that MTBOOT was searching for the end of memory to get memory size forever! Every address, it read and write its content and check 777777 for any results. It ended up an infinite loop. :-( That's why I still need its source code. Due to lack of source codes, TOPS-20 is not supported at this time. About FORM Disk Formattor, it finally worked fine since LUUO routines had been fixed for correctly return back where LUUO had been executed. However, my debug log file showed that formattor tried to issue unibus initialization to reset all devices and controllers on desired unibus adaptor. Oh, I forget to implement that! I will implement that soon. It tried to issue initialization but it happened nothing. RHCS2 should be zero after it wrote dummy unit 7 to RHCS2. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1411 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feeder.via.net!HSNX.atgi.net!sjc-peer.news.verio.net!news.verio.net!sjc-read.POSTED!not-for-mail User-Agent: Microsoft-Outlook-Express-Macintosh-Edition/5.02.2022 Subject: Re: PDP-10 in an FPGA chip (was Re: Timothy Stark's effort) From: David G Conroy Newsgroups: alt.sys.pdp10 Message-ID: References: <39D163D1.C119CCF1@bartek.net> <39D16BC9.F048C9D2@bartek.net> <8qslsk$jom$3@bob.news.rcn.net> <39D1F581.CE63E00C@bartek.net> <8qv92s$es9$6@bob.news.rcn.net> <39D3BD18.639FF1AC@bartek.net> <6u7l7tbhos.fsf_-_@chonsp.franklin.ch> <39D55251.36123656@jetnet.ab.ca> Mime-version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Lines: 32 Date: Sun, 01 Oct 2000 18:48:18 GMT NNTP-Posting-Host: 207.21.131.70 X-Complaints-To: abuse@verio.net X-Trace: sjc-read 970426098 207.21.131.70 (Sun, 01 Oct 2000 18:48:18 GMT) NNTP-Posting-Date: Sun, 01 Oct 2000 18:48:18 GMT Organization: Verio Xref: nntp1.ba.best.com alt.sys.pdp10:1411 >> >> Now that would be nice. Some guy has done it for the PDP-8 (actually >> he implemented a DECmate): >> >> http://surfin.spies.com/~dgc/pdp8x/ >> >> For that he used the 2 smallest Xilinx chips (CPU XCS10 and IO XCS05). >> With some of the bigger ones a full KS-10 including cache should be >> possible. This is actually in my to-do list, after I finish the universal 18-bit PCB, and debug the FPGA code for the PDP-1 (which is done) and the PDP-4/PDP-7 (which is all but done). I have lots of sketches of bits of a PDP-10, most based on an 18-bit data path, driven by fairly vertical microcode. Lots of things are worked out, but there are some ugly parts. There are too many NOPS in some of the key microflows because there is nothing which can be done in the microbranch delay slot (it almost all cases the branch is on the "address translation miss" condition). Multiply and divide are ghastly because it takes 4 microcycles to shift a 72 bit number in an 18 bit datapath. Most problems are a result of trying to make the micromachine very simple, and I think it's too simple. And I actually didn't implement a DECmate; if I had been willing to settle for that, I would not have had to build my own UARTS just to implement non-distructive read from the keyboard buffer correctly. Article 1429 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: BACKUP successful! Newsgroups: alt.sys.pdp10 References: <402A5.27630$A4.849196@news1.giganews.com> <8qsmdm$jom$5@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 22 Message-ID: <%lHC5.5593$wx5.155113@news2.giganews.com> NNTP-Posting-Date: Wed, 04 Oct 2000 09:43:07 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-zuVteIzEusjl8mzouNH1vraQk3v3CUYjk1nJKrKTJnwwS3zat1XK7ZaXrk3+efV1quhvTI0F4pm/TAl!aFcoi0WBVQkK6O53CrFvxQFSR+I= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 04 Oct 2000 14:43:09 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1429 jmfbahciv@aol.com wrote: >>Good news! Backup program finally worked so well (returned back to square >>one). I successfully installed files on my emulated RP06 disk from my >>tape file without any problems. > Fandamntastic! Now you're humpin'. Barb: Thank you! :-) I restored CUSP tapes into my emulated RP06 disk successfully. It took estimated 15 minutes to complete. Middle of restore process asked me to mount next tape and type "GO" to continue. After 15 minutes, it finally prints ""Done". It worked so well! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1433 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: BACKUP successful! Newsgroups: alt.sys.pdp10 References: <402A5.27630$A4.849196@news1.giganews.com> <8qsmdm$jom$5@bob.news.rcn.net> <%lHC5.5593$wx5.155113@news2.giganews.com> <39DB65F6.CD436FBB@bartek.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 43 Message-ID: NNTP-Posting-Date: Wed, 04 Oct 2000 17:18:09 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-U72pT0S3n7skn5MiMhTpk5bZU0424/hSoDYl9YpMjSVtsKVrZXZzvDQ8Fib0MZ0JnjrjiD5qLIxzSgT!rKSRd4FKoTMS9jSbNJ4itS6s73g= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 04 Oct 2000 22:18:09 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1433 Arthur Krewat wrote: > What speed machine are you running this on, and under what operating > system? Just curious, since it would help to guage what speed the > emulator would run on other machines (like mine!).. Yes, I am running my Pentium III 800EB processor on my home-built PC (ASUS CUSL2 motherboard). Also, I am running my emulator under Red Hat Linux 7. On my system, my emulator runs at estimated 450 KIPS or 0.4 MIPS. My emulator is faster than the original KS10 system (0.3 MIPS). Also disk/tape I/O is much faster. My emulator loaded SYSTEM.EXE from a tape file and it took a few seconds to complete. The KS10 emulator runs at 450/500 Mhz Pentium III will equal to the original KS10 system for speed. If you have 500 Mhz Alpha processor, my KS10 emulator will run faster due to 64-bit wide processor. > I can't wait! > This may have been mentioned before, but I've only been reading the > newsgroup religiously for the last few weeks and I may have missed > something. > What are your plans for releasing this emulator? Would you be willing > to email it to individual people that would want to test it? > thanks! and your efforts are GREATLY appreciated! I am finishing my code soon before it reach a beta version. I will finish a few instructons more like FP, DMUL, DDIV, and EXTEND instructions. Next plan, I will extend to KL10 processor but still am looking for RH10/RH20 and DTE20 interface register-level functional specifications. A known bug is still existing - CORE instruction that attempted to assign high-segment allocation while running SMFILE.EXE. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1515 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!feed2.onemain.com!feed1.onemain.com!novia!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news5.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: PDP-10 Emulator Status Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 573 Message-ID: <3ljF5.26901$Ly1.378022@news5.giganews.com> NNTP-Posting-Date: Thu, 12 Oct 2000 08:50:55 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-eIue9J7Lru5aeJa79QdA4YcUuVxwfJyiZNoovXviU05pRLpGwB2Ns+90sOHI90e7LS5rYMyAvQZg3fV!MrqMt7pkLrlTh4y6WX0OxIL9J6uV X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 12 Oct 2000 13:50:55 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1515 Hello folks: Good news! I finally got rid of some major problems with my emulator since I implemented trap routines for TRAP1/TRAP2 flags to get rid of "?Illegal trap instruction" when I tried to copy exe files in NFT program. I added a few functions to my RP routines (OFFSET, RETURN, and RECALIBRATE) to get rid of "HALT 0 at PC 0" messages and returned back to KS10> prompt. I fixed BAE settings in RH controller because BAE settings was destroyed after I/O process so that it resulted "?Transmission Error". (I noticed that it resulted 177000 instead of 377000 in my debug log file and quickly fixed it.) I still am working on the minor problems like SMFILE, backup errors when restore is completed, etc. I unknownly fixed a hidden bug in shift/rotate routines when I found out different results from SMFILE program. That's why I changed shift/rotate routines to speed up instead of one shift each loop. For example (ROT instruction): count %= 36; if (count < 0) count = 36 - count; ac0 &= WORD36_ONES; ac0 = (ac0 << count) | (ac0 >> (36 - count)); I will work on DMUL, DDIV, EXTEND, and FP instructions soon. I noticed that SYSTAT reports big errors for time usage. I found out that clock interrupt uses DDIV to update. At this time, DDIV acts like NOP. Well, here is screen log: KS10 Thursday 12-Oct-100 8:13:29 .LOGIN .ASSIGN MTA0: TAPE: MTA0 assigned .GET TAPE:BACKUP Job setup .SAVE BACKUP BACKUP saved .RUN BACKUP /TAPE TAPE: /SUPERSEDE ALWAYS /INTERCHANGE /FILES /RESTORE SYS:=*.* ! SYSTEM EXE "Done /RESTORE DEC:=*.* ! KSBOOT EXE SMFILE EXE SMFILE HLP SMFILE TXT T10KL RAM T10KL ULD "Done /EXIT .[^E pressed] KS10> attach mta0: ../tops10a.tpc Unit mta0 had been attached with '../tops10a.tpc' file. KS10> continue Running now... .RUN BACKUP /SUPERSEDE ALWAYS /PROTECTION 155 /SSNAME ALL /REWIND ?BKPNTS No tape specified /TAPE TAPE: /REWIND /FILES /RESTORE DSKB: !10,7 DSKB CUSP BWR CUSPS CTL TOOLS CTL BLDCST CTL CUSPS RST [long typing files deleted] PLTUTL MAC LOGIN HLP LOGIN RNH LOGIN EXE LOGIN MAC DSKB:[10,7,LOGIN]LOGIN.MAC (BLOCK=245) [MTA0: Read(C/H/S) = 30385440/0/0] $BKPEOT Reached EOT--mount new tape then type "GO" /[^E pressed] KS10> attach mta0: ../tops10b.tpc Unit mta0 had been attached with '../tops10b.tpc' file. KS10> continue Running now... GO !10,7 LOGIN MAC LOGOUT EXE KJOB HLP REATTA HLP [long typing files deleted] FELOAD P11 DDT11 EXE CUSP DIR CUSP RNO %BKPIBL Incorrect block length reading non-file data %BKPIBL Incorrect block length reading non-file data %BKPNBF Not BACKUP format reading non-file data [long repeating backup errors deleted] %BKPIBL Incorrect block length reading non-file data %BKPIBL Incorrect block length reading non-file data ?BKPRTE Reached tape error maximum . .RUN DSKB:[10,7,NFT]NFT *COPY SYS:=DSKB:[10,7,*,*,*,*,*]*.EXE DSKB:[1,4]KSBOOT.EXE <= DSKB:[10,7]KSBOOT.EXE DSKB:[1,4]SMFILE.EXE <= DSKB:[10,7]SMFILE.EXE DSKB:[1,4]REACT.EXE <= DSKB:[10,7,ACCT]REACT.EXE DSKB:[1,4]ACTDAE.EXE <= DSKB:[10,7,ACCT]ACTDAE.EXE [long copy processing deleted] DSKB:[1,4]TECO.EXE <= DSKB:[10,7,TECO]TECO.EXE DSKB:[1,4]TGHA.EXE <= DSKB:[10,7,TGHA]TGHA.EXE DSKB:[1,4]FELOAD.EXE <= DSKB:[10,7,FELOAD]FELOAD.EXE DSKB:[1,4]DDT11.EXE <= DSKB:[10,7,FELOAD]DDT11.EXE Total of 1871360 words in 14620 blocks in 112 files *^C .DIR %WLDSLE Search list empty .DIR DSKB: BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] .DIR SYS: SAT SYS 89 <355> 12-Oct-00 704(151) DSKB: [1,4] HOME SYS 29 <355> 12-Oct-00 704(151) SWAP SYS 8009 <357> 12-Oct-00 704(151) BADBLK SYS 9 <357> 12-Oct-00 704(151) CRASH EXE 4100 <357> (undated) 704(151) SYSTEM EXE 1236 <057> 7-Oct-88 704(33461) KSBOOT EXE 44 <155> 1-Sep-88 4(76) SMFILE EXE 104 <155> 1-Sep-88 2(7) REACT EXE 320 <155> 1-Sep-88 33B(242) ACTDAE EXE 304 <155> 1-Sep-88 2B(155) BACKUP EXE 128 <155> 1-Sep-88 5A(625) BTDDT EXE 132 <155> 1-Sep-88 44C(701) BOOT EXE 184 <155> 1-Sep-88 4(76) BTONCE EXE 156 <155> 1-Sep-88 4(76) KSONCE EXE 144 <155> 1-Sep-88 4(76) COMPIL EXE 40 <155> 1-Sep-88 22G(605) CREDIR EXE 68 <155> 1-Sep-88 3B(111) CREF EXE 28 <155> 1-Sep-88 53C(101) CRSCPY EXE 68 <155> 1-Sep-88 1B(61) CTHNRT EXE 136 <155> 1-Sep-88 5A(323) DAEMON EXE 48 <155> 1-Sep-88 23(1025) DDT EXE 160 <155> 1-Sep-88 44C(701) FILDDT EXE 64 <155> 1-Sep-88 44C(701) VMDDT EXE 72 <155> 1-Sep-88 44C(701) DDT11 EXE 296 <155> 18-Nov-87 7E(120002)-3 DECLAR EXE 60 <155> 1-Sep-88 2(16) MX EXE 220 <155> 1-Sep-88 1(475) MS EXE 260 <155> 1-Sep-88 11(6041) DELFIL EXE 16 <155> 1-Sep-88 3B(6) DIRECT EXE 136 <155> 1-Sep-88 7A(503) DSKLST EXE 24 <155> 1-Sep-88 33(4) DSKRAT EXE 20 <155> 2-Sep-88 14A(24) DTELDR EXE 56 <155> 1-Sep-88 5A(54) F11 EXE 120 <155> 1-Sep-88 2(102) FAL EXE 328 <155> 1-Sep-88 2A(50) FE EXE 24 <155> 1-Sep-88 2(50) FEFILE EXE 16 <155> 1-Sep-88 1(27) FILCOM EXE 28 <155> 1-Sep-88 22A(121) FILDAE EXE 60 <155> 1-Sep-88 4(44) BATCON EXE 128 <155> 1-Sep-88 105(5027) CATCNV EXE 160 <155> 1-Sep-88 (1) CATLOG EXE 196 <155> 1-Sep-88 2A(36) CDRIVE EXE 76 <155> 1-Sep-88 1(114) GALGEN EXE 228 <155> 23-Aug-88 5(2124) GLXLIB EXE 168 <155> 1-Sep-88 5(1601) LPTSPL EXE 180 <155> 1-Sep-88 105(4103) NEBULA EXE 72 <155> 1-Sep-88 1(20) OPR EXE 204 <155> 1-Sep-88 5(1050) ORION EXE 384 <155> 1-Sep-88 5(1050) PLEASE EXE 192 <155> 1-Sep-88 104(115) PULSAR EXE 228 <155> 1-Sep-88 5(543) QUASAR EXE 416 <155> 1-Sep-88 5(1512) QUEUE EXE 160 <155> 1-Sep-88 106(724) SPRINT EXE 92 <155> 1-Sep-88 105(4410) GLOB EXE 20 <155> 1-Sep-88 5C(141) HELP EXE 16 <155> 1-Sep-88 7(100) INITIA EXE 52 <155> 1-Sep-88 12(343) KDPLDR EXE 68 <155> 1-Sep-88 1(6) LINK EXE 52 <155> 1-Sep-88 6(2420) LNK999 EXE 48 <155> 1-Sep-88 6(2420) LNKERR EXE 128 <155> 1-Sep-88 6(2420) LNKLOD EXE 140 <155> 1-Sep-88 6(2420) LNKMAP EXE 64 <155> 1-Sep-88 6(2420) LNKOV1 EXE 156 <155> 1-Sep-88 6(2420) LNKOV2 EXE 116 <155> 1-Sep-88 6(2420) LNKSCN EXE 88 <155> 1-Sep-88 6(2420) LNKXIT EXE 108 <155> 1-Sep-88 6(2420) LOGIN EXE 132 <155> 1-Sep-88 65(3010) LOGOUT EXE 132 <155> 1-Sep-88 65(3010) LP20 EXE 76 <155> 1-Sep-88 1(4) MACDLX EXE 60 <155> 30-Apr-86 31(1065) MACRO EXE 108 <155> 1-Sep-88 53B(1247) MAKLIB EXE 120 <155> 1-Sep-88 2C(132) MIC EXE 68 <155> 1-Sep-88 11C(1235) NETLDR EXE 60 <155> 1-Sep-88 4(146) NETWOR EXE 80 <155> 1-Sep-88 3C(44) NFT EXE 392 <155> 1-Sep-88 2A(21) OPSER EXE 32 <155> 1-Sep-88 5E(133) PATH EXE 76 <155> 1-Sep-88 3A(30) PIP EXE 52 <155> 1-Sep-88 33D(546) PROJCT EXE 20 <155> 1-Sep-88 1(11) QUOLST EXE 12 <155> 1-Sep-88 5(42) RMS EXE 140 <155> 1-Sep-88 1A(123) RMSCOB EXE 140 <155> 1-Sep-88 RMSUTL EXE 176 <155> 1-Sep-88 1A(123) RSXT10 EXE 24 <155> 1-Sep-88 3(42) RUNOFF EXE 108 <155> 1-Sep-88 12(263) SCDSET EXE 228 <155> 1-Sep-88 5(22) SETSRC EXE 16 <155> 1-Sep-88 12B(47) COMPAR EXE 24 <155> 1-Sep-88 2(20) MERGE EXE 28 <155> 1-Sep-88 2(15) UPDATE EXE 24 <155> 1-Sep-88 2(13) NOTIFY EXE 292 <155> 30-Apr-86 2(1140) SCOMM EXE 84 <155> 2-Sep-88 SPEAR EXE 300 <155> 2-Sep-88 2(1140) SPRCOM EXE 284 <155> 30-Apr-86 2(1140) SPRINS EXE 332 <155> 30-Apr-86 2(1140) SPRKLE EXE 336 <155> 30-Apr-86 2(1140) SPRRET EXE 848 <155> 2-Sep-88 2(1140) SPRSUM EXE 388 <155> 2-Sep-88 2(1140) SYSANS EXE 56 <155> 1-Sep-88 704(653) SYSDLT EXE 56 <155> 1-Sep-88 704(653) SYSDPA EXE 56 <155> 1-Sep-88 704(653) SYSDPB EXE 56 <155> 1-Sep-88 704(653) SYSDPY EXE 56 <155> 1-Sep-88 704(653) SYSHZL EXE 56 <155> 1-Sep-88 704(653) SYSV50 EXE 56 <155> 1-Sep-88 704(653) SYSV52 EXE 56 <155> 1-Sep-88 704(653) SYSV61 EXE 56 <155> 1-Sep-88 704(653) SYSVBX EXE 60 <155> 1-Sep-88 704(653) SYSTAT EXE 60 <155> 1-Sep-88 704(416) TECO EXE 36 <155> 1-Sep-88 24A(235) TGHA EXE 100 <155> 1-Sep-88 4(20) FELOAD EXE 308 <155> 2-May-88 7E(120002)-3 Total of 27536 blocks in 114 files on DSKB: [1,4] .TYPE DSKB:[10,7,DECLAR]DECLAR.HLP [DECLARE.HLP 17-June-1988] The DECLARE program allows a user to manipulate user-defined commands. The syntax to define a command is: NAME=FILESPEC or NAME/UNIQUE:n=FILESPEC where "n" can be any of 1, 2, 3, 4, ALL, or NONE, or a list of any of these values enclosed in parentheses, as in (1,2,4). or NAME/AUTOPUSH=FILESPEC or a combination of the above. When a command has been defined, the monitor will from then on search the user's command table in parallel with the monitor command table. In the case of conflicts, exact definitions are given precedence, and in the case of conflicts within this, the user's commands are given precedence. In the case of conflicts arising from an abbreviation being given, the monitor re-searches its own command table without searching the user's command table, and if it finds a unique match, it will use it. A user may define "uniqueness" for a command, which the monitor will behave as if the command had been given an exact definition for that many characters. Examples: .DECLARE TECO=SYS:STECO .TECO ;Will run SYS:STECO .TEC ;Will run SYS:TECO (Monitor overrides in the case of abb rev.) .TE ;Will run SYS:TECO However, .DECLARE TECO/UNIQUE:2=SYS:STECO .TECO ;Will run SYS:STECO .TEC ;Will run SYS:TECO (Monitor command has /UNIQUENESS:3) .TE ;Will run SYS:STECO .T ;Will say ?T? since even the monitor commands conflict And, .DECLARE TECO/UNIQUE:(2,3)=SYS:STECO .TECO ;Will run SYS:STECO .TEC ;Will run SYS:STECO .TE ;Will run SYS:STECO .T ;Will say ?T? as before Note that most programs will only recognize a very few command names, and should be invoked only with such a command name. For example, to run CRSCPY via a command, the command must be "CRSCPY" or some abbreviation thereof. Other options allowed by this program are: /AUTOPUSH When this switch is given on a command definition, it causes that command to do an auto-push, thus causing that command to leave the previous core image intact. This feature is dependent upon having context service (CTXSER) loaded in your monitor. CTXSER is loaded by default. However, a command which is run via an auto-push cannot change TMPCOR, pathological name definitions, or command definitions for your job. Thus, many programs will not perform as expected if they are run via an auto-push command. /CLEAR Clear all user defined commands. This is a standalone switch. /LIST Lists the command names that the user currently has defined. This is also a standalone switch. /KILL This switch will remove the definition of a command. This switch also requires a COMMAND NAME. /SORT Sorts the commands that the user currently has defined. Sometimes this is necessary due to the way the monitor scans the command tables. This is a standalone switch. DECLARE name Lists the full definition for the command "name". The command name may contain wildcards, in which case any matching user-defined commands will have their definitions listed. .SYSTAT Status of KS10 at 8:31:10 on 12-Oct-100 Uptime 17:41, 539% Null time = 539% Idle + 0% Lost, 0% Overhead 1 Jobs in use out of 35. 1 logged in, 0 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] CTY SYSTAT 22+SPY RN 0 Swapping space used = 19/2000 = 1% Virt. Core used = 22/2000 = 1% 606P Core left Active swapping ratio = 22/628 = .03 Average job size =22/1 = 22.0P+0/1 = .00P Total=22/1 = 22.0P Busy devices: Device Job Why Logical MTA0 1 as TAPE System File Structures: Name Free Mount DSKB 203570 1 Total Free 203570 .R QUOLST User: 1,2 Str used left:(in) (out) (sys) .R DSKLST /H FOR HELP */H TYPE IN NAMES OF FILE STRUCTURES TO BE PRINTED, PLUS SWITCHES TO SELECT THOSE SEGMENTS OF THE LISTING DESIRED. SWITCHES INCLUDE: /U - PRINT UNITS COMPRISING FILE STRUCTURE /V - PRINT VITAL STATISTICS (HOME BLOCK) /F - PRINT FILE INFORMATION /S - PRINT SAT BLOCKS /B - PRINT BAT BLOCKS /E - PRINT ERROR SUMMARY /P - PRINT PERFORMANCE SUMMARY /Q - PRINT QUICK LISTING (SHORT LOOKUPS INSTEAD OF EXTENDED) /N - NARROW PAPER /H - TYPE THIS TEXT IF NO SWITCHES ARE SPECIFIED, ALL SEGMENTS ARE PRINTED. IF NO FILE STRUCTURES ARE SPECIFIED, ALL FILE STRUCTURES WILL BE PRINTED, SO CARRIAGE RETURN SPECIFIES A COMPLETE LISTING. */U [system crashed quietly. (^C did not work) - had exit this emulator.] [^E pressed] KS10> quit Exiting PDP10 Emulator [sword7@tstark pdp10]$ [sword7@tstark pdp10]$ ./ks10 -f ks10.cfg Welcome to KS10 Emulator v0.7 (Alpha) KS10> boot mta0: Booting MTA0... Now running... BOOT V4(76) BOOT>/t [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 12-oct-2000 Time: 8:46 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] %FRCLIN INITIA: Can't read SYS:SYSJOB.INI %%TTY STOMPER - Not starting %FRCLIN INITIA - Logging out ? INTTNF SYS:TTY.INI not found KS10 08:46:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 .LOGIN 1,2 .R OPSER [OPRPAF Processing auto command file] ?OPRALF LOOKUP failure 0 * [CCPWFD Waiting for file daemon to start] ^C .DIR BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] .R SMFILE DECSYSTEM 2020 DIAGNOSTICS FE-FILE PROGRAM VERSION 2.7, TOPS-10, KS10, CPU#=8 [FOR HELP TYPE "HELP"] ?Stopcode IME, type=JOB, on CPU0 at 12-Oct-100 8:46:24 Job 2 on CTY running SMFILE User [1,2] UUO is 047000,,000036 at exec PC 350336 Page fail word = 111000,,436777, page fail PC = 347441 CPU Status Block APRID = 610000,,011610 CONI APR, = 001060,,002001 CONI PI, = 000000,,000377 CONI PAG, = 000000,,060002 DATAI PAG, = 500100,,001321 [Dumping on DSKB:CRASH.EXE[1,4]] [Aborting job] [Continuing system] ?Monitor error at exec PC 350335; UUO at exec PC 350335 . [CCPWFD Waiting for file daemon to start] .SYSTAT Status of KS10 at 8:46:29 on 12-Oct-100 Uptime 28, 20141% Null time = 20141% Idle + 0% Lost, 0% Overhead 20 Jobs in use out of 35. 1 logged in, 2 detached. Job Who Line# What Size(P) State Run Time 1 2,5 DET41 CRSCPY 8+15 SL 0 $ 2 [OPR] CTY SYSTAT 22+SPY RN 0 3 2,5 30 INITIA 5+11 SL 0 4 2,5 31 INITIA 5+11 SL 0 5 2,5 32 INITIA 5+11 SL 0 6 2,5 33 INITIA 5+11 SL 0 7 2,5 DET41 CRSCPY 8+15 SL 0 $ 10 2,5 20 INITIA 5+11 SL 0 11 2,5 21 INITIA 5+11 SL 0 12 2,5 22 INITIA 5+11 SL 0 13 2,5 23 INITIA 5+11 SL 0 14 2,5 24 INITIA 5+11 SL 0 15 2,5 25 INITIA 5+11 SL 0 16 2,5 26 INITIA 5+11 SL 0 17 2,5 27 INITIA 5+11 SL 0 18 2,5 34 INITIA 5+11 SL 0 19 2,5 35 INITIA 5+11 SL 0 20 2,5 36 INITIA 5+11 SL 0 21 2,5 37 INITIA 5+11 SL 0 22 2,5 40 INITIA 5+11 SL 0 $ means Execute Only High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 17 CRSCPY DSKB [SYS] 15 2 Swapping space used = 89/2000 = 4% Virt. Core used = 149/2000 = 7% 479P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 191/(191+149) = 56% Average job size =123/20 = 6.1P+217/20 = 10.8P Total=340/20 = 17.0P No busy devices System File Structures: Name Free Mount DSKB 185920 4 Total Free 185920 . %CCPFDS The file daemon did not start; proceeding [CCPCPY Copied DSKB:CRASH.EXE[1,4] to DSKB:IME001.EXE[10,1]] . .[^E pressed] KS10> quit Exiting PDP10 Emulator [sword7@tstark pdp10]$ End of screen log. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org, sword7@firesword7.net -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1621 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: PDP10 (KS10) Emulator status. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 185 Message-ID: NNTP-Posting-Date: Mon, 23 Oct 2000 21:07:44 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-JEdjyxWcL2/iAGhgm26NHXgt6xLVzxG7PuK1dB9lFoSaBqIEQETGBBhfIvRl6kD5GfIYClB53T69fKd!0YRZqrrVUbv5lI/ewVHchI1WjrYJ X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 24 Oct 2000 02:07:45 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1621 Hello folks: Good news! I resolved the problem with page fault handler after I rewrote my entire page fill routine (pager_Translate) by implementing setjmp/longjmp statements to avoid data corruption during instruction process (messing up acculumators). When a page fail trap occurs, abandon instruction process immediately by long-jump back to main CPU loop routine bypass instruction routines. Now I was able to run TOPS-10 system so well execpt ACTDAE that just crashed. However, I was unable to debug ACTDAE due to too complication because several INITIA programms are running in the same time. :-( With NOINIT option, I was unable to login as 1,2 because it is waiting for DAEMON starting forever. How do I debug ACTDAE program with NOINIT option at "startup-option:" prompt? Look at log file below. Also, I tried to boot TOPS-20 v4.1 tape and finally got "MTBOOT>" prompt! I am trying figure how to boot TOPS-20 v4.1 system... With TOPS-10 v7.03, I finally got a dot and was able LOGIN as 1,2. Now I am looking for MIG 7.03 because MIG 7.04 is very different to 7.03. Well here is my log file: KS10> boot mta0: Booting MTA0... Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 23-oct-2000 Time: 9:29 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 09:29:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . [CCPWFD Waiting for file daemon to start] ?Error in detached job 7[ACTDAE] ? ?HALT at user PC 000017 %%TTY STOMPER - Starting unknown] .LOGIN 1,2 .R OPR ^C ^C .SYSTAT Status of KS10 at 9:29:22 on 23-Oct-100 Uptime 21, -27030% Null time = -27030% Idle + 0% Lost, 0% Overhead 35 Jobs in use out of 35. 7 logged in, 6 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL -1:0-35:0-26 $ 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 5 2,5 35 INITIA 5+11 SL 0 6 2,5 37 INITIA 5+11 SL 0 7 [OPR] DET41 ACTDAE 90+40 ^C 0 8 2,5 36 INITIA 5+11 SL 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 10 2,5 34 INITIA 5+11 SL 0 11 [OPR] DET41 QUASAR 112+40 TO 0 12 2,5 40 INITIA 5+11 SL 0 13 [OPR] 41 4 ^C 0 14 2,5 33 INITIA 5+11 SL 0 15 2,5 6 INITIA 5+11 SL 0 16 2,5 7 INITIA 5+11 SL 0 17 2,5 10 INITIA 5+11 SL 0 18 2,5 11 INITIA 5+11 SL 0 19 2,5 12 INITIA 5+11 SL 0 20 2,5 13 INITIA 5+11 SL 0 21 2,5 14 INITIA 5+11 SL 0 22 2,5 15 INITIA 5+11 SL 0 23 2,5 16 INITIA 5+11 SL 0 24 2,5 17 INITIA 5+11 SL 0 25 2,5 20 INITIA 5+11 SL 0 26 2,5 21 INITIA 5+11 SL 0 27 2,5 22 INITIA 5+11 SL 0 28 2,5 23 INITIA 5+11 SL 0 29 2,5 24 INITIA 5+11 SL 0 30 2,5 25 INITIA 5+11 SL 0 31 2,5 26 INITIA 5+11 SL 0 32 2,5 27 INITIA 5+11 SL 0 33 2,5 30 INITIA 5+11 SL 0 34 2,5 31 INITIA 5+11 SL 0 35 2,5 32 INITIA 5+11 SL 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 28 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 378/2000 = 19% Virt. Core used = 483/2000 = 24% 145P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 337/(337+483) = 41% Average job size =417/35 = 11.9P+403/35 = 11.5P Total=820/35 = 23.4P No busy devices System File Structures: Name Free Mount DSKB 170480 3 Total Free 170480 .SYSTAT Status of KS10 at 9:29:46 on 23-Oct-100 Uptime 45, -12643% Null time = -12643% Idle + 0% Lost, 0% Overhead 7 Jobs in use out of 35. 7 logged in, 5 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 ACTDAE 90+40 ^C 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 11 [OPR] DET41 QUASAR 112+40 TO 0 13 [OPR] 41 4 ^C 0 n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 304/2000 = 15% Virt. Core used = 325/2000 = 16% 303P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 40/(40+325) = 11% Average job size =274/7 = 39.1P+91/7 = 13.0P Total=365/7 = 52.1P No busy devices System File Structures: Name Free Mount DSKB 170480 2 Total Free 170480 .R ACTDAE ? ACTDAE GLXIPC Requested PID belongs to job 7 ? ?HALT at user PC 625331 . KS10> quit Exiting PDP10 Emulator [sword7@sword7-0 pdp10]$ Thank you! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1663 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!cyclone.swbell.net!cyclone-sf.pbi.net!63.208.208.143!feed2.onemain.com!feed1.onemain.com!novia!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 14 Message-ID: <_AKJ5.44084$YX4.1318904@news2.giganews.com> NNTP-Posting-Date: Wed, 25 Oct 2000 19:07:22 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-zUSTilVCo+HVSFdWPE7s6wW/q0WdnfeQ0eYuv1M5rZwx7xaw9ImThu5MXrYr/bwU0aBZQqvIcYeZAOp!+P5KEAuRr37WN0lYCBw5XOEMjIw= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 26 Oct 2000 00:07:22 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1663 Mark Crispin wrote: > Trivia question: what was the slowest machine ever to run TOPS-20? KS-10 Processor - 0.3 MIPS. My KS10 emulator is faster than original KS-10 on Intel 800 Mhz Pentium III system. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1693 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!logbridge.uoregon.edu!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Thu, 26 Oct 2000 10:31:57 -0700 Organization: Networks & Distributed Computing Lines: 31 Message-ID: References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 972581520 30656 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: pchar To: Pat Barron In-Reply-To: Xref: nntp1.ba.best.com alt.sys.pdp10:1693 On 26 Oct 2000, Pat Barron wrote: > > Trivia question: what was the slowest machine ever to run TOPS-20? > Well, I am guessing that this is sort of a trick question, and that > the answer is along the lines of the Apple Powerbook 145 (running KLH's > emulator)..... ;-) Close enough. Yes, this was a trick question. Note that the question specifically said "slowest machine ever to run TOPS-20"; it did not say "slowest PDP-10." The slowest machine ever to run TOPS-20 was my Apple PowerBook 100, which had a 16MHz 68000 processor and 8MB of RAM. It ran TOPS-20 under an early version of the KLH emulator at approximately 1/60th the speed of a KS. [Compare my 100MHz Cyrix (Pentium clone) notebook, which is slightly faster than a KS.] As far as I know, that PowerBook 100 was the first notebook computer to run TOPS-20. Yes, it was intolerably slow. I still have that PowerBook (although it spent a couple years with my step-sister when she was an undergrad), but I don't run TOPS-20 on it any more. This also debunked the myth that you couldn't run a demand-paged virtual memory operating system on a 68000 based system because it had no MMU (and there wasn't a good one until the 68030). -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article 1699 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.mesh.ad.jp!sjc-peer.news.verio.net!sea-feed.news.verio.net!news.verio.net!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Thu, 26 Oct 2000 18:48:39 -0700 Organization: Networks & Distributed Computing Lines: 26 Message-ID: References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 972611322 68462 (None) 140.142.17.37 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: hatk In-Reply-To: Xref: nntp1.ba.best.com alt.sys.pdp10:1699 On Thu, 26 Oct 2000, Timothy Stark wrote: > Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is > 0.3 MIPS. An 800MHz machine, and you're only slightly faster than a KS? I get that class performance on my 100MHz Cyrix clone with the KLH emulator. Stu Grossman's emulator is comparable. I think that after you finish debugging, you should review and find out what's making your emulator 8-10 times slower than it should be. I suspect that it may have to do with how you organize memory. > > This also debunked the myth that you couldn't run a demand-paged virtual > > memory operating system on a 68000 based system because it had no MMU (and > > there wasn't a good one until the 68030). > Yeah. that is right. Pentium class have a demand-paged virtual memory. > :-) Yes, but my point had to do with doing demand-paged virtual memory on what would be roughly the equivalent of a PC-XT. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article 1700 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8t7ks7$3u1$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 26 Message-ID: NNTP-Posting-Date: Thu, 26 Oct 2000 22:15:48 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-Q4SFOshQOu3p2nuUXWuCGo3KUpx6c0lJpHA4B7Wv1fJFTQSXpH+993VWgtJE86I7ZDAJT9lFAKjZTEV!Iklzq7xUarcjS4/jZnABOJy9W28= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 27 Oct 2000 03:15:48 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1700 Mark Crispin wrote: > I think that after you finish debugging, you should review and find out > what's making your emulator 8-10 times slower than it should be. I > suspect that it may have to do with how you organize memory. Hmmmm. I have not implemented cache yet. With cache, will it be much faster? I am trying to bring my emulator faster. Is KLH's emulator written in C language? Without paging system, it is around 0.46 MIPS. With paging on, estimated 0.32 MIPS. Hmmm. With 64-bit variables, I found out that several instructions per single statement. With 32-bit variables, just a few instructions. Big difference. Hmmm. I am using GCC compiler v2.95. Thank you. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1718 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!cyclone2.usenetserver.com!news-out.usenetserver.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 76 Message-ID: NNTP-Posting-Date: Fri, 27 Oct 2000 09:13:06 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-nWvB6UcQNqtqs4FvIvsVmzAtlJddn9uJ141L8NdaAfRXBkbDcSDXR1P8cZT81ucvJZbDk+6dIvOnW9h!7THoH8aSJB7qtvzWlHOiofSYrvY= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 27 Oct 2000 14:13:07 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1718 Alexander Viro wrote: > x86 _is_ register-starved. It's actually worse than screwing the instructions > count, you are putting serious pressure on register allocator and it's _not_ > a good idea. Hmm. I will find out. > And I'm not sure that raw CPU speed is anywhere near the bottleneck - TLB > and dcache footprints may be more serious. BTW, could you post vmstat 1 > output both on the normally loaded system and on a minimal loads? Ideally - > booting with init=/bin/sh, mounting the filesystems and saying > # /usr/bin/open -c 2 > # exec vmstat 1 > (emulator on the second VC, vmstat on the first; for clean shutdown after > that use Alt-SysRq-I, Alt-SysRq-S, Alt-SysRq-U, Alt-SysRq-B). I checked with vmstat while I ran my KS10 emulator and vmstat shows 60% user time and 40% system time. No swap activity. My vmstat show that: procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 2 0 0 3220 3192 45868 33920 0 0 0 0 114 87 63 37 0 1 0 0 3220 3192 45868 33920 0 0 0 0 115 124 67 33 0 1 0 0 3220 3192 45868 33920 0 0 0 0 114 26 65 35 0 1 0 0 3220 3192 45868 33920 0 0 0 6 155 35 59 41 0 1 0 0 3220 3192 45868 33920 0 0 0 0 124 85 62 38 0 1 0 1 3220 3192 45868 33920 0 0 0 0 115 63 63 37 0 1 0 1 3220 3192 45868 33920 0 0 0 0 116 57 65 35 0 1 0 1 3220 3192 45868 33920 0 0 0 0 110 107 57 43 0 1 0 1 3220 3192 45868 33920 0 0 0 0 113 31 62 38 0 1 0 1 3220 3192 45868 33920 0 0 0 0 110 13 70 30 0 1 0 1 3220 3192 45868 33920 0 0 0 0 130 32 64 36 0 1 0 1 3220 3192 45868 33920 0 0 0 0 128 64 66 34 0 1 0 1 3220 3080 45980 33920 0 0 0 0 140 78 63 37 0 1 0 1 3220 2968 46088 33924 0 0 1 3 141 97 60 40 0 1 0 1 3220 3092 45960 33924 0 0 0 0 118 1078 64 36 0 That is all. Well, I am changing my emulated instructions to provide a direct access to AC blocks instead of vRead36/vWrite36 routines. I made modifications on MOVE- and SOJ- instructions as initial test. Then I compiled and ran my emulator. It seems worked fine. My log file show speed improvement. Est. 20 KIPS gained. On 450 Mhz system, it almost matches original KS speed - varying 0.25 to 0.34 MIPS. Also, I am adding KL10 specification to my emulator to provide all procesors per configuration, not just KS10 emulation right now. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1722 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcasq$kc4@weyl.math.psu.edu> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 36 Message-ID: <22jK5.96420$bI6.3525945@news1.giganews.com> NNTP-Posting-Date: Fri, 27 Oct 2000 12:35:58 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-sRsWYlYwacgaoipacudxwncfaACs4dgIbwF9/xXCnQsbBP7MhVSDC4MnkdDDPcbPAeQwj3/f5yyOqH7!qFYhgVtXJjfMgS9/spQEcXzAEdk= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 27 Oct 2000 17:35:58 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1722 Alexander Viro wrote: > IOW, not IO-bound, lots of time spent in the kernel, resulting in bursts of > context switches from time to time (depending on the system load it may > be normal or not, but for a heavy CPU-bound process it seems to be too high). > I have a nasty suspicion that you are misusing poll() or select() - the > picture is very typical for that. strace might be useful here - just > count the system calls and see what sticks out. Hmm. I did not use poll() or select() but my emulator check any key press by using read(0, &ch, 1) after every instruction. I am looking for way to set interrupts when key presses. Every emulated instructions are extremely brief because I tried -pg option but its record shows that all emulated instructions show 0.0 microsecond per call! Yes, DDIV instruction show 0.0 ms/call! Every instruction finished less than 1 microsec... However, pager_Translate is one of the most heavy usage - 20% of my emulator's execution. > Another thing is the TLB footprint - VSZ and RSS of the process might give > some information here, but unfortunately x86 doesn't let you really see > what happens in the bloody thing ;-/ > Keep in mind that context switch => TLB flush - x86 has only one MMU > context and effects are nasty. BTW, which kernel were you using when you > took that vmstat? Linux version is 2.2.16 or Red Hat Linux 7.0 -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1736 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news5.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Whoa! My emulator went into warp drive! Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tcasq$kc4@weyl.math.psu.edu> <22jK5.96420$bI6.3525945@news1.giganews.com> <8tckpf$ko5@weyl.math.psu.edu> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 24 Message-ID: NNTP-Posting-Date: Fri, 27 Oct 2000 16:56:00 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-lMR9rFO9iDs0l6/WTNKtJUyaut/8VculYgmCIDoWjsNOq7ty0FhkTM85v9NyTjivQme247T/8AqvImN!sfWFzBItsKAqxix5AbozpMrqTYQ= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 27 Oct 2000 21:56:00 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1736 >>Hmm. I did not use poll() or select() but my emulator check any key press >>by using read(0, &ch, 1) after every instruction. I am looking for way to >>set interrupts when key presses. Every emulated instructions are Well, I commented out the keyboard poll routines in its function routine. I compiled and ran it. I looked into my log file. Whoa! My emulator went into warp drive! 500 KIPS gained! Now 0.8 MIPS (uncached and TOPS-20 paging on) from 0.3-0.45 MIPS on the 800 Mhz system! Now my emulator slightly outperforms KI10 Processor. Yes, it beat all processors execpt KL10 (1.7 MIPS). Ok, I will implement fork process for I/O bound (includes keyboard poll, etc.) Thank you for your help! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1767 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!feed1.news.rcn.net!rcn!not-for-mail From: "Alan H. Martin" Newsgroups: alt.sys.pdp10 Subject: Re: Whoa! My emulator went into warp drive! Date: Sat, 28 Oct 2000 21:56:14 -0400 Lines: 32 Message-ID: <39FB83BE.BC503E73@MA.UltraNet.Com> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tckpf$ko5@weyl.math.psu.edu> <7xd7glm735.fsf@ruckus.brouhaha.com> <8teu5p$o1o@weyl.math.psu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: o9AbKPRNL9T3aYTtittY3s+TuerqdYCjGvjoWMG10Cc= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 29 Oct 2000 01:56:15 GMT X-Accept-Language: en,en-US,en-GB,es X-Mailer: Mozilla 4.75 [en] (Win95; U) Xref: nntp1.ba.best.com alt.sys.pdp10:1767 Alexander Viro wrote: > > In article <7xd7glm735.fsf@ruckus.brouhaha.com>, > Paul Rubin wrote: > >Timothy Stark writes: > >> Ok, I will implement fork process for I/O bound (includes keyboard poll, > >> etc.) > > > >You should be able to simulate interrupt driven i/o with the F_GETSIG > >flag to the fcntl system call (see my previous note). You don't need > >a separate process. > > What's wrong with a separate process? Compared to use of signals... I'ld > say that it's much cleaner solution. A signal handler that sets a volatile flag and dismisses + A test of the volatile flag at the head of the instruction decoder What's the performance cost of testing a volatile flag, especially compared to your earlier post about testing a buffer emptiness at the same place(s)? (I definitely don't see the advantage of the rest of the suggestion to create a separate thread that drains the O/S input queue only to fill an input queue in the simulator, compared to just doing a vanilla read when the input ready flag is set by the signal routine. This project is a PDP-10 simulator, not a user-mode CTY input queue management system with optional PDP-10 simulation). /AHM -- Alan Howard Martin AMartin@MA.UltraNet.Com Article 1768 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!63.208.208.143!feed2.onemain.com!feed1.onemain.com!newsfeed.utk.edu!washdc3-snf1!news.gtei.net!news.telebeam.net!news.ems.psu.edu!news3.cac.psu.edu!not-for-mail From: viro@weyl.math.psu.edu (Alexander Viro) Newsgroups: alt.sys.pdp10 Subject: Re: Whoa! My emulator went into warp drive! Date: 28 Oct 2000 22:44:37 -0400 Organization: -ENOENT Lines: 36 Message-ID: <8tg2ul$pdq@weyl.math.psu.edu> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <7xd7glm735.fsf@ruckus.brouhaha.com> <8teu5p$o1o@weyl.math.psu.edu> <39FB83BE.BC503E73@MA.UltraNet.Com> NNTP-Posting-Host: weyl.math.psu.edu Xref: nntp1.ba.best.com alt.sys.pdp10:1768 In article <39FB83BE.BC503E73@MA.UltraNet.Com>, Alan H. Martin wrote: >> What's wrong with a separate process? Compared to use of signals... I'ld >> say that it's much cleaner solution. > >A signal handler that sets a volatile flag and dismisses ^^^^^^^^^^^^^^ == ugly mess > + >A test of the volatile flag at the head of the instruction decoder >What's the performance cost of testing a volatile flag, especially >compared to your earlier post about testing a buffer emptiness at >the same place(s)? And what stops the reader from setting the same flag without messing with signals? >(I definitely don't see the advantage of the rest of the suggestion to >create a separate thread that drains the O/S input queue only to fill >an input queue in the simulator, compared to just doing a vanilla read >when the input ready flag is set by the signal routine. This project >is a PDP-10 simulator, not a user-mode CTY input queue management >system with optional PDP-10 simulation). For one thing, SIGIO doesn't work for pipes. Think whatever you want, but redirections are Good Thing(tm) and making the thing dependent on the stdin coming from TTY or socket... not nice. Besides, look at the API used. Kludges all over the place - fcntl()/ioctl() and signal(). IOW, the barf-bag. SIGIO _is_ a kludge that happens to work in some cases. fork() and read() - core API. And yes, processes are cheap, it's UNIX, after all. -- "You're one of those condescending Unix computer users!" "Here's a nickel, kid. Get yourself a better computer" - Dilbert. Article 1849 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!sunqbc.risq.qc.ca!newsxfer3.itd.umich.edu!forum.apple.com!news.apple.com!haxrus.apple.com!user From: aek@spies.com (Al Kossow) Newsgroups: alt.sys.pdp10 Subject: Re: Whoa! My emulator went into warp drive! Date: Wed, 01 Nov 2000 18:01:40 -0800 Organization: Apple Computer, Inc. Lines: 20 Message-ID: References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <7xd7glm735.fsf@ruckus.brouhaha.com> <8teu5p$o1o@weyl.math.psu.edu> <8tqgn4$25v6$1@nntp1.ba.best.com> NNTP-Posting-Host: haxrus.apple.com X-Trace: news.apple.com 973130499 4200 17.205.21.66 (2 Nov 2000 02:01:39 GMT) X-Complaints-To: usenet@news.apple.com NNTP-Posting-Date: 2 Nov 2000 02:01:39 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1849 In article <8tqgn4$25v6$1@nntp1.ba.best.com>, falconer@best.com (Edward A. Falk) wrote: > >What's wrong with a separate process? Compared to use of signals... I'ld > >say that it's much cleaner solution. > > Frankly, I'd spawn a process for every emulated device. Use > shared memory for the emulated bus. Use interprocess signals > for interrupts. > Very unix-centric, aren't we? Tim, if it's possible, please keep any OS-specific details in a separate OS-dependent portion of the emulator, to make porting to non-Unix systems easier. -- The eBay Curse: "May you find everything you're looking for.." Article 1851 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!nntp.frontiernet.net!nntp.gblx.net!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Whoa! My emulator went into warp drive! Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <7xd7glm735.fsf@ruckus.brouhaha.com> <8teu5p$o1o@weyl.math.psu.edu> <8tqgn4$25v6$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 16 Message-ID: <725M5.112701$bI6.4328449@news1.giganews.com> NNTP-Posting-Date: Wed, 01 Nov 2000 21:18:27 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-ZkqDjnTnX9I3OLOIU94x8VMok9fZZ3ivn75DwpObdloWgG6ZJ4stTdUABy44A0FB7FliHcouIK/+yqg!iiV1QYwEFK01y+sjlkqwd1FaDho= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 02 Nov 2000 03:18:30 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1851 Al Kossow wrote: > Tim, if it's possible, please keep any OS-specific details in > a separate OS-dependent portion of the emulator, to make porting > to non-Unix systems easier. Of course, we can port my emulator to other platforms like Windows, OpenVMS, MacOS, etc. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1740 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!portc01.blue.aol.com!newsjunkie.ans.net!news.chips.ibm.com!newsfeed.btv.ibm.com!newshost.transarc.ibm.com!pat From: Pat Barron Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: 27 Oct 2000 22:47:34 GMT Organization: IBM Corporation Lines: 23 Distribution: world Message-ID: References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcasq$kc4@weyl.math.psu.edu> <22jK5.96420$bI6.3525945@news1.giganews.com> NNTP-Posting-Host: smithfield.transarc.ibm.com In-reply-to: Timothy Stark's message of Fri, 27 Oct 2000 17:35:58 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1740 In article <22jK5.96420$bI6.3525945@news1.giganews.com> Timothy Stark writes: > Hmm. I did not use poll() or select() but my emulator check any key press > by using read(0, &ch, 1) after every instruction. Well, that'll certainly slow you down - it costs you a system call per emulated instruction. CTY I/O is very rare (compared to instruction execution) so it costs you plenty (in terms of time spent) to use a system call to poll for it each time you go through the instruction dispatch loop. >I am looking for way to set interrupts when key presses. On FreeBSD and Linux, you can use the F_SETOWN function of fcntl() to get a SIGIO signal sent to your process when I/O is possible on a descriptor. The overhead of catching a signal is high, but it should be important when you consider how much CTY I/O is done. If you don't want to do this, then maybe you can still do polling, but only check for an available keypress once every hundred emulated instructions or so (i.e., don't do it every time - serial I/O is comparatively slow, so there's no need to....). --Pat. Article 1743 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news-hog.berkeley.edu!ucberkeley!nntp.cs.ubc.ca!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcasq$kc4@weyl.math.psu.edu> <22jK5.96420$bI6.3525945@news1.giganews.com> Distribution: world User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 40 Message-ID: NNTP-Posting-Date: Fri, 27 Oct 2000 21:51:52 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-vNp8zXnbPNPgrA2Y7Tvy3iMdc9Hk+Kr47XCtTCSEOG0qBzQHtR2EUcovoAuPwXv1h5NpkZBiSWaQZbb!eppl9S3FlK3cQqyB6sdXqU50GN0= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sat, 28 Oct 2000 02:51:52 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1743 Pat Barron wrote: > Well, that'll certainly slow you down - it costs you a system call > per emulated instruction. CTY I/O is very rare (compared to instruction > execution) so it costs you plenty (in terms of time spent) to use > a system call to poll for it each time you go through the instruction > dispatch loop. Yeah. I discovered that when I commented out keyboard poll routine! Whoa. I changed the front-end routines and removed keyboard poll call from main loop. I placed ks10fe_ctyOutput in apr_io function. Now 0.83 MIPS. Another speed gain - 20 KIPS increased! > On FreeBSD and Linux, you can use the F_SETOWN function of fcntl() to > get a SIGIO signal sent to your process when I/O is possible on a > descriptor. The overhead of catching a signal is high, but it should > be important when you consider how much CTY I/O is done. I tried to implement that but it did not work. It happened nothing when I pressed some keys. :-( My new routine: void ks10fe_Initialize(void) { emu_IOTrap = ks10fe_ctyInput; if (fcntl(fileno(stdin), F_SETOWN) < 0) perror("STDIN 1"); if (fcntl(fileno(stdin), F_SETFL, O_ASYNC) < 0) perror("STDIN 2"); } It was executed successfully. No error reports but it happened nothing! Yes, I put signal(SIGIO, emu_IO) in main.c to call keyboard input routine. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1744 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsxfer.eecs.umich.edu!news.bu.edu!budd From: budd@csa.bu.edu (Phil Budne) Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Followup-To: alt.sys.pdp10 Date: 28 Oct 2000 04:51:43 GMT Organization: Boston University Computer Science Dept. Lines: 80 Message-ID: <8tdm0v$6hp$1@news3.bu.edu> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <22jK5.96420$bI6.3525945@news1.giganews.com> X-Trace: news3.bu.edu 972708703 6713 128.197.12.3 (28 Oct 2000 04:51:43 GMT) X-Complaints-To: news@bu.edu Xref: nntp1.ba.best.com alt.sys.pdp10:1744 In article , Timothy Stark wrote: >Pat Barron wrote: >> On FreeBSD and Linux, you can use the F_SETOWN function of fcntl() to >> get a SIGIO signal sent to your process when I/O is possible on a >> descriptor. The overhead of catching a signal is high, but it should >> be important when you consider how much CTY I/O is done. > >I tried to implement that but it did not work. It happened nothing when >I pressed some keys. :-( My new routine: > >void ks10fe_Initialize(void) >{ > emu_IOTrap = ks10fe_ctyInput; > if (fcntl(fileno(stdin), F_SETOWN) < 0) > perror("STDIN 1"); > if (fcntl(fileno(stdin), F_SETFL, O_ASYNC) < 0) > perror("STDIN 2"); >} Here is "sigio.c", which works on SunOS4, FreeBSD, and Linux (from an 8051 simulator I wrote); #include /* STDIN_FILENO */ #include #include void sigio(int x) { puts("SIGIO"); } main() { int fl; signal(SIGIO, sigio); if (fcntl(STDIN_FILENO, F_SETOWN, getpid()) < 0) { perror("F_SETOWN"); exit(1); } fl = fcntl(STDIN_FILENO, F_GETFL, 0); if (fl == -1) { perror("F_GETFL"); exit(1); } if (fcntl(STDIN_FILENO, F_SETFL, fl|FASYNC) < 0) { perror("F_SETFL"); exit(1); } puts("type return"); pause(); } And here is how you do it on SVR4 STREAMS systems (SunOS5, IRIX); #include /* STDIN_FILENO */ #include #include void sigpoll(int x) { puts("SIGPOLL"); } main() { int fl; signal(SIGPOLL, sigpoll); if (ioctl(STDIN_FILENO, I_SETSIG, S_INPUT) < 0) { perror("I_SETSIG"); exit(1); } puts("type return"); pause(); } Article 1750 of alt.sys.pdp10: Sender: phr2000-nospam@ruckus.brouhaha.com From: Paul Rubin Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcasq$kc4@weyl.math.psu.edu> <22jK5.96420$bI6.3525945@news1.giganews.com> Date: 28 Oct 2000 06:23:42 -0700 Message-ID: <7xem11m74x.fsf@ruckus.brouhaha.com> Organization: Nightsong/Fort GNOX Lines: 16 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 28 Oct 2000 06:26:52 -0700, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com alt.sys.pdp10:1750 Timothy Stark writes: > Hmm. I did not use poll() or select() but my emulator check any key press > by using read(0, &ch, 1) after every instruction. I am looking for way to > set interrupts when key presses. This is probably killing you. I've looked at the source for Stu Grossman's emulator and while it's written fairly carefully, it doesn't go to great lengths to bum cycles. It's written in straightforward C. I bet that with careful tuning and maybe some inline assembly code here and there, it could run twice as fast on an x86 as it currently does. On the other hand, I think it's not nearly as far along as your emulator in simulating the 10's i/o devices etc. See the fcntl(2) man page for info on getting a SIGIO interrupt when a key is hit (use F_GETSIG), or you could look at the GNU Emacs sources to see how it's done. Article 1759 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcasq$kc4@weyl.math.psu.edu> <22jK5.96420$bI6.3525945@news1.giganews.com> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 28 Oct 2000 11:29:19 -0700 Message-ID: Lines: 23 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 28 Oct 2000 11:32:31 -0700, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com alt.sys.pdp10:1759 Timothy Stark writes: > Hmm. I did not use poll() or select() but my emulator check any key press > by using read(0, &ch, 1) after every instruction. System calls are expensive. I'm sure everyone will point out that you need to use asynchronous I/O. But as an intermediate step, you might try doing your read() call only on simulated branch instructions. In a simulator I wrote (for something other than a PDP-10), this proved to be quite effective. Another option is to poll I/O after every "n" simulated instructions, for some reasonable value of n. > However, pager_Translate is one of the most heavy usage - 20% of > my emulator's execution. Try keeping the results of the latest translation, and having pager_Translate first see if it can reuse that. This is effectively a single-entry TLB. If that helps, you might want a few entries. Or you could use a 512-entry TLB, one entry for every virtual page (of a single-section PDP-10). This is more efficient than an intermediate-sized TLB since you can simply take the top nine bits of the address as the TLB index, rather than having to search a fully-associative TLB. Article 1723 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nycmny1-snh1.gtei.net!news.gtei.net!hermes.visi.com!news-out.visi.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03-Where will it end? Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcblb$sn0$1@nnrp1.deja.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 31 Message-ID: NNTP-Posting-Date: Fri, 27 Oct 2000 12:48:55 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-cMLJHwWbYhKIgCPrV4vPiepmsFk8bfsg/sMbDE31nRpmzBuwFRNAIgZ0Z9S7ebcwNhFp1ozvcD6sqOB!lIdMB+LPNaBGFAYhmJh3xyYg9+Q= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 27 Oct 2000 17:48:55 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1723 hsnewman@my-deja.com wrote: > Tim, > I hope you know there are alot of people actively waiting for the next > release of the emulator. I've noticed that you have talked of adding a > Vax emulator to the emulator, as well as other types of pdp10 systems. > My real question is: where will it all end? By posting it, you could > increase the fingers testing the system, and please note that those > fingers could be informed that revisions will be comming out. > Typically, a scope of a release is set mostly in concrete, but without > a deep footer so you can make some changes. > Could you tell me (and all those folks waiting) what the scope of the > next release will be? > -HN Well, thank you for information. KS10 emulation is near finished and will be released soon. Thank you for you patience. I already started to work on KL10 so that save some of my developement time. Only one problem is remaining - Quasar daemon that crashed on my emulator and it is much difficult to debug with running Quasar daemon. That's why I need stand-alone system to test it. If I was unable debug that, I will still release my emulator soon and let others to work on it. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1747 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!192.148.253.68!netnews.com!feed1.news.rcn.net!rcn!207-172-245-143 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03-Where will it end? Date: Sat, 28 Oct 00 07:46:13 GMT Organization: UltraNet Communications, Inc. Lines: 54 Message-ID: <8tebhr$jaf$4@bob.news.rcn.net> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8tb18d$ind@weyl.math.psu.edu> <8tcblb$sn0$1@nnrp1.deja.com> X-Trace: 6Rc6zCF5uV0y0TdAgkU75Q9yJE+/O5GqM4N+fy1SJsg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 28 Oct 2000 10:59:07 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1747 In article , Timothy Stark wrote: >hsnewman@my-deja.com wrote: >> Tim, >> I hope you know there are alot of people actively waiting for the next >> release of the emulator. I've noticed that you have talked of adding a >> Vax emulator to the emulator, as well as other types of pdp10 systems. >> My real question is: where will it all end? By posting it, you could >> increase the fingers testing the system, and please note that those >> fingers could be informed that revisions will be comming out. >> Typically, a scope of a release is set mostly in concrete, but without >> a deep footer so you can make some changes. >> Could you tell me (and all those folks waiting) what the scope of the >> next release will be? >> -HN > >Well, thank you for information. KS10 emulation is near finished and will >be released soon. Thank you for you patience. I already started to work >on KL10 so that save some of my developement time. This is why you need to make clear what your intentions are. I think I've figured out your work pattern, but people like the above poster are assuming that the work pattern is completely different. > >Only one problem is remaining - Quasar daemon that crashed on my emulator >and it is much difficult to debug with running Quasar daemon. That's why >I need stand-alone system to test it. I thought you were using a PC. Is it possible that you misunderstood my reference to stand-alone? > If I was unable debug that, I will >still release my emulator soon and let others to work on it. You might have to do just this. I can't seem to be able to even teach keystrokes without being able to do them. Debugging GALAXY was just as complicated as debugging the monitor. To hear Larry talk about it when they were writing the beast, it was more difficult. Since I can't really see how terminals are set up on your "system", I haven't been able to figure out how, let alone tell you how. Getting symbols in that software is a first step, but you need to be able to build the thing...yeah, yeah, Mark, no need to rub it in :-). That CAI stuff is the SIXBIT 3-character crash code that gets appended to "GLX". Sometimes, I can still dream PDP-10. /BAH Subtract a hundred and four for e-mail. Article 1705 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news-hog.berkeley.edu!newsfeed.berkeley.edu!ucberkeley!sjc1.nntp.concentric.net!newsfeed.concentric.net!global-news-master From: "Seth J. Morabito" Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: 27 Oct 2000 06:57:03 GMT Organization: The Consulate of the Republic of Belaria Lines: 18 Message-ID: <8tb8vv$hs7@dispatch.concentric.net> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8t7ks7$3u1$1@nntp1.ba.best.com> NNTP-Posting-Host: 216.112.97.80 X-no-archive: yes X-frog-flavor: extra-crunchy User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Xref: nntp1.ba.best.com alt.sys.pdp10:1705 Mark Crispin wrote: : On Thu, 26 Oct 2000, Timothy Stark wrote: :> Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is :> 0.3 MIPS. : An 800MHz machine, and you're only slightly faster than a KS? I get that : class performance on my 100MHz Cyrix clone with the KLH emulator. Stu : Grossman's emulator is comparable. It doesn't matter HOW fast the KLH or Grossman emulators are if noone can get them! Thankfully, in just a short time Mr. Stark will be releasing his emulator to the public, and I'd infinitely rather have a slow emulator than no emulator at all. (OK, I promised myself I would stop griping about the mysterious KLH emulator in public. I'll get around to it, I swear.) -Seth Article 1729 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.skycache.com!Cidera!cpk-news-hub1.bbnplanet.com!washdc3-snf1!news.gtei.net!news.telebeam.net!news.ems.psu.edu!news3.cac.psu.edu!not-for-mail From: viro@weyl.math.psu.edu (Alexander Viro) Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: 27 Oct 2000 15:26:49 -0400 Organization: -ENOENT Lines: 15 Message-ID: <8tcktp$kor@weyl.math.psu.edu> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> NNTP-Posting-Host: weyl.math.psu.edu Xref: nntp1.ba.best.com alt.sys.pdp10:1729 In article , Eric Smith wrote: >Timothy Stark writes: >> Hmmmm. I have not implemented cache yet. With cache, will it be much >> faster? I am trying to bring my emulator faster. Is KLH's emulator >> written in C language? > >If you simulate a cache, you will make your simulator *slower*. Not necessary - you may very well end up with smaller active set and unless you get too horrible overhead in icache you can actually win. -- "You're one of those condescending Unix computer users!" "Here's a nickel, kid. Get yourself a better computer" - Dilbert. Article 1707 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.37!feed1.news.rcn.net!rcn!207-172-216-115 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Fri, 27 Oct 00 07:33:21 GMT Organization: UltraNet Communications, Inc. Lines: 33 Message-ID: <8tbmdd$r90$1@bob.news.rcn.net> References: <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> X-Trace: 3gWDgtD1J9IKazTzr6NR+Er7/tBgDvhOIv8ofn2IqoE= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 27 Oct 2000 10:46:05 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1707 In article , Mark Crispin wrote: >On Thu, 26 Oct 2000, Timothy Stark wrote: >> Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is >> 0.3 MIPS. > >An 800MHz machine, and you're only slightly faster than a KS? I get that >class performance on my 100MHz Cyrix clone with the KLH emulator. Stu >Grossman's emulator is comparable. > >I think that after you finish debugging, you should review and find out >what's making your emulator 8-10 times slower than it should be. I >suspect that it may have to do with how you organize memory. JMF used SYSDPY with CDO's changes in them. This stuff was not shipped with any of the 6.03 stuff so you'll have to look for it on the 7-series tapes. And I put all of that stuff on the Customer Supported Tape. Now, IIRC, and my brain is hazy this week, SYSDPY can output data that another program written by Chuck (in FORTRAN?) takes as input and pretties up a display. JMF also seemed to be able to learn a lot just by staring at SYSDPY running. He spent hours staring at the screen. Getting SYSDPY to work on your terminal may not be a trivial task and would be a really good hack project to assign to somebody else. /BAH Subtract a hundred and four for e-mail. Article 1714 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!uninett.no!news.ost.eltele.no!not-for-mail From: blarsen@nonews.net (Bjørn Hell Larsen) Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: 27 Oct 2000 13:30:42 +0200 Organization: Nonews is good news Lines: 48 Message-ID: References: <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> <8tbmdd$r90$1@bob.news.rcn.net> NNTP-Posting-Host: locke.svg.infostream.no Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Trace: news.ost.eltele.no 972646177 1114 213.236.169.78 (27 Oct 2000 11:29:37 GMT) X-Complaints-To: newsmaster@ost.eltele.no NNTP-Posting-Date: 27 Oct 2000 11:29:37 GMT User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.4 Xref: nntp1.ba.best.com alt.sys.pdp10:1714 jmfbahciv@aol.com writes: > In article > , > Mark Crispin wrote: > >On Thu, 26 Oct 2000, Timothy Stark wrote: > >> Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is > >> 0.3 MIPS. > > > >An 800MHz machine, and you're only slightly faster than a KS? I get that > >class performance on my 100MHz Cyrix clone with the KLH emulator. Stu > >Grossman's emulator is comparable. > > > >I think that after you finish debugging, you should review and find out > >what's making your emulator 8-10 times slower than it should be. I > >suspect that it may have to do with how you organize memory. > > > JMF used SYSDPY with CDO's changes in them. This stuff was > not shipped with any of the 6.03 stuff so you'll have to > look for it on the 7-series tapes. And I put all of that > stuff on the Customer Supported Tape. > > Now, IIRC, and my brain is hazy this week, SYSDPY can output > data that another program written by Chuck (in FORTRAN?) takes > as input and pretties up a display. JMF also seemed to be > able to learn a lot just by staring at SYSDPY running. He > spent hours staring at the screen. Getting SYSDPY to work > on your terminal may not be a trivial task and would be a > really good hack project to assign to somebody else. I volunteer, FWIW. I always loved SYSDPY. We had a locally produced Tandberg display terminal running a hacked version of SYSDPY in the operator room at UiO, and I was totally amazed at the level of detail of the information it gave us, and how much it let us know about the wellbeing of the system. I haven't touched DDT since '86, and it is starting to show. :) Bjørn -- DoD#2537 / Senior MC - '90 FZ750 (Fizz) / '99 VT750C2 (Brumm) Article 1716 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsxfer.eecs.umich.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-216-115 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Fri, 27 Oct 00 10:15:57 GMT Organization: UltraNet Communications, Inc. Lines: 60 Message-ID: <8tbvu8$2hk$6@bob.news.rcn.net> References: <8t7ks7$3u1$1@nntp1.ba.best.com> <8tbmdd$r90$1@bob.news.rcn.net> X-Trace: 4EqOySiZSBQuh5dvBC+ix25A5B4gRWIkPPRp/XDKEhM= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 27 Oct 2000 13:28:40 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1716 In article , blarsen@nonews.net (Bjørn Hell Larsen) wrote: > >jmfbahciv@aol.com writes: > >> In article >> , >> Mark Crispin wrote: >> >On Thu, 26 Oct 2000, Timothy Stark wrote: >> >> Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is >> >> 0.3 MIPS. >> > >> >An 800MHz machine, and you're only slightly faster than a KS? I get that >> >class performance on my 100MHz Cyrix clone with the KLH emulator. Stu >> >Grossman's emulator is comparable. >> > >> >I think that after you finish debugging, you should review and find out >> >what's making your emulator 8-10 times slower than it should be. I >> >suspect that it may have to do with how you organize memory. >> >> >> JMF used SYSDPY with CDO's changes in them. This stuff was >> not shipped with any of the 6.03 stuff so you'll have to >> look for it on the 7-series tapes. And I put all of that >> stuff on the Customer Supported Tape. >> >> Now, IIRC, and my brain is hazy this week, SYSDPY can output >> data that another program written by Chuck (in FORTRAN?) takes >> as input and pretties up a display. JMF also seemed to be >> able to learn a lot just by staring at SYSDPY running. He >> spent hours staring at the screen. Getting SYSDPY to work >> on your terminal may not be a trivial task and would be a >> really good hack project to assign to somebody else. > >I volunteer, FWIW. > >I always loved SYSDPY. All of our consoles would finish up with a running SYSDPY. > >We had a locally produced Tandberg display terminal running a hacked >version of SYSDPY in the operator room at UiO, and I was totally >amazed at the level of detail of the information it gave us, and >how much it let us know about the wellbeing of the system. > >I haven't touched DDT since '86, and it is starting to show. :) I'll warn you now. I closed my eyes and tried to envision how I debugged the ACTDAE and couldn't find the line feed key. Remember how SYSDPY had all those magic twanggers for commands that would change the display screen? I don't know if the equivalent character will work. For all I know, which isn't much, two characters get shipped. /BAH Subtract a hundred and four for e-mail. Article 1813 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for CUSP tape files for v6.03 Newsgroups: alt.sys.pdp10 References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8t7ks7$3u1$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 30 Message-ID: NNTP-Posting-Date: Mon, 30 Oct 2000 18:38:36 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-iLhoalA/WkmJlHEUK86wfJUZaXmGZ2sU/SBYqH0Cwh2xga3W1jKG7CpGWjxsoZ0PSeVnM92Pjt3nyce!3Op4Rfs4Zws5vufSMDClhgDbjCk= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 31 Oct 2000 00:38:37 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1813 Mark Crispin wrote: > On Thu, 26 Oct 2000, Timothy Stark wrote: >> Mine is 0.46 MIPS on my 800 Mhz PC system. Original KS-10 is >> 0.3 MIPS. > An 800MHz machine, and you're only slightly faster than a KS? I get that > class performance on my 100MHz Cyrix clone with the KLH emulator. Stu > Grossman's emulator is comparable. > I think that after you finish debugging, you should review and find out > what's making your emulator 8-10 times slower than it should be. I > suspect that it may have to do with how you organize memory. Yeah. I resolved its problem. Now my emulator runs at up to 1.3 MIPS on 800 Mhz system. Now - 3 to 4 times faster than a KS. Also, I am working on them to improve some performance. Its problem was keyboard poll routine. It had been major-changed. Whoa. Old keyboard poll routines stole 2/3 of its performance! Thank you for let me know. No one let me have Stu's or KLH's emulator to compare with my emulator for performance. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1681 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!feed1.news.rcn.net!rcn!207-172-255-18 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Thu, 26 Oct 00 09:15:49 GMT Organization: UltraNet Communications, Inc. Lines: 34 Message-ID: <8t9819$erl$3@bob.news.rcn.net> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> X-Trace: Hb2S6WZ7SUFQ3YhFpUtSZhs0AJPCmgIFQwnG7N0icYg= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Oct 2000 12:28:25 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1681 In article , Timothy Stark wrote: >Joe Smith wrote: >> In article , >> Timothy Stark wrote: >>>What happened to TOPS v7.05? > >> TOPS-10 v7.04 was the last release. > >Ok, I got it. Thank you for information. Except he's confusing you. 7.04 is a monitor version...not a CUSP version. The only reason that the two are tied together is because we found it easier to field test and ship updated CUSPs with a monitor development cycle. And it took us 15-20 years to figure that all out. > >> As I understand it, the software changes needed to get TOPS-10 running >> on the prototype of the "follow-on 36-bit processor" (KC10, aka Jupiter) >> were never completed, so there never was a 7.05 Monitor. > >> But what about the Minnow? The fabled PDP-10 on a desk (smaller and >> cheaper than the KS)? Was it functional enough to run TOPS-10 or >> TOPS-20 before the plug got pulled on that project? > >Hmm. I never heard of "Minnow" before. However, PDP-10 can be on a desk >at today's technology. It could have been on a desk twice in our corporate history. /BAH Subtract a hundred and four for e-mail. Article 1679 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!feed1.news.rcn.net!rcn!207-172-255-18 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Looking for CUSP tape files for v6.03 Date: Thu, 26 Oct 00 09:12:16 GMT Organization: UltraNet Communications, Inc. Lines: 41 Message-ID: <8t97qk$erl$1@bob.news.rcn.net> References: <8hoH5.62649$bI6.2262607@news1.giganews.com> <8smntu$3f1$11@bob.news.rcn.net> <8t7ks7$3u1$1@nntp1.ba.best.com> X-Trace: PnPlHM9H6xY9d4QNw5nGLXNd/P5SKM3zdNzbmkR5zrw= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Oct 2000 12:24:52 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:1679 In article <8t7ks7$3u1$1@nntp1.ba.best.com>, inwap@best.com (Joe Smith) wrote: >In article , >Timothy Stark wrote: >>What happened to TOPS v7.05? > >TOPS-10 v7.04 was the last release. V7.04 was the release I was working on when I got sick. I had heard that there was a final release that the idiots called V7.05 when it really should have been V7.04A since it was only a maintenance release. Now, it certainly wouldn't surprise me to learn that they only submitted this final set of tapes to SDC with no distribution. It would have been typical of their thinking. > >As I understand it, the software changes needed to get TOPS-10 running >on the prototype of the "follow-on 36-bit processor" (KC10, aka Jupiter) >were never completed, so there never was a 7.05 Monitor. Joe, you've got the naming conventions all wrong. If the KC had been shipped, the version of the monitor supporting that piece of hardware might have been 7.03 since we were working on the 7.02 ship when the project got killed. It would have never been 7.05 since that version would have been shipped after 7.04. The date for 7.04 was 1987...or was it 1986...four years after the Jupiter. > >But what about the Minnow? The fabled PDP-10 on a desk (smaller and >cheaper than the KS)? Was it functional enough to run TOPS-10 or >TOPS-20 before the plug got pulled on that project? I doubt it. The -10 software guys weren't working on it at all. /BAH Subtract a hundred and four for e-mail. Article 1814 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!enews.sgi.com!paloalto-snf1.gtei.net!news.gtei.net!forum.apple.com!news.apple.com!haxrus.apple.com!user From: aek@spies.com (Al Kossow) Newsgroups: alt.sys.pdp10 Subject: Re: PDP10 (KS10) Emulator status. Date: Mon, 30 Oct 2000 16:59:06 -0800 Organization: Apple Computer, Inc. Lines: 20 Message-ID: References: <8t3sc8$dak$1@bob.news.rcn.net> <8t3v2j$j87$3@bob.news.rcn.net> <6EhJ5.86142$bI6.2817650@news1.giganews.com> NNTP-Posting-Host: haxrus.apple.com X-Trace: news.apple.com 972953943 20532 17.205.21.66 (31 Oct 2000 00:59:03 GMT) X-Complaints-To: usenet@news.apple.com NNTP-Posting-Date: 31 Oct 2000 00:59:03 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1814 In article , aek@spies.com (Al Kossow) wrote: > In article <6EhJ5.86142$bI6.2817650@news1.giganews.com>, Timothy Stark > wrote: > > > I have not implemented DZ-11 devices yet > > because I need DZ-11 functional specs. > > Eric Smith should have this information for you. I have the > DZ tech an programming manuals coming as well. I've scanned in the programming section of the DZ11 tech ref. You can find it at http://www.spies.com/aek/pdf/dec/DZ11Prog.pdf -- The eBay Curse: "May you find everything you're looking for.." Article 1807 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!logbridge.uoregon.edu!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Yeah. I/O Async worked now. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 31 Message-ID: NNTP-Posting-Date: Mon, 30 Oct 2000 08:01:24 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-5ZrFSNjrRMTZRNJuOOxS96AuI+V83lNAodNF3f6GY+ehbTgn6WZrXmQ9mPJnMA+TGTdyiyuXPQwid06!OSqXszZp1817tBvCLkyElljdyDs= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Mon, 30 Oct 2000 14:01:24 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1807 Hello folks: Thank to Phil for examples about keyboard I/O async. I removed keyboard poll routine from main loop where is executed after every emulated instruction. It reduced its performance greatly. I learned a lesson about that. Now I break a routine into two routines (one for input and one for output). I put a output routine in the emulated WRAPR instruction (within Set selected flags block) and wrote SIGIO interrupt routine to call a input routine during fe_Initialize routine. Following an example from Phil's posting, I implemented three fcntl statements with error checking. Also, for fe_Cleanup, I implemented three "inverted" fcntls. I compiled and run it. It worked greatly and finally accepted my key presses. Yeah. I noticed that TOPS-10 execution is much faster than before! I looked into my debug log file and was so impressed. On 450 Mhz system, it is varying 0.47 to 0.84 MIPS. It was 0.15 to 0.33 MIPS before. I believe that my emulator could break 1 MIPS barrier on 800 Mhz or above system. I believe that pager_Translate reduces its performance 1/3 times. I would implement a cache to test performance, etc. My goal is match KL10 speed or faster. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1812 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Yeah. I/O Async worked now. Newsgroups: alt.sys.pdp10 References: User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 23 Message-ID: NNTP-Posting-Date: Mon, 30 Oct 2000 18:26:53 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-mGXW3zFsn6U2uavWa1oSI/vAMVmcKIJcW/7rHprkja8pz0jcT8TcQsuefS7s6LwquXLSe9UI9A78WJM!+zb28wJvsHjQJ5S1HGQeUraDtYY= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 31 Oct 2000 00:26:53 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1812 Timothy Stark wrote: > I compiled and run it. It worked greatly and finally accepted my key > presses. Yeah. I noticed that TOPS-10 execution is much faster than > before! I looked into my debug log file and was so impressed. On 450 Mhz > system, it is varying 0.47 to 0.84 MIPS. It was 0.15 to 0.33 MIPS before. > I believe that my emulator could break 1 MIPS barrier on 800 Mhz or above > system. I believe that pager_Translate reduces its performance 1/3 times. > I would implement a cache to test performance, etc. My goal is match KL10 > speed or faster. Yeah. On my 800 Mhz system, it was 0.60 to 1.3 MIPS. My emulator did break 1 MIPS barrier. Yes. my emulator now is 2 to 4 times faster than a original KS10 machine but almost match KL10 speed. To beat KL10 speed, you might need least 1 Ghz Pentium. :-) -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1843 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Re: Yeah. I/O Async worked now. References: <39fec660@News.Destek.net> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 01 Nov 2000 09:40:03 -0800 Message-ID: Lines: 24 X-Newsreader: Gnus v5.7/Emacs 20.7 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 1 Nov 2000 09:44:43 -0800, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com alt.sys.pdp10:1843 cherkus@unimaster.com (Dave Cherkus) writes: > But caching the results of virtual-to-physical translations (i.e. > implementing a TLB in software) within the emulator should be a big > win. The only rub is you need to know when to invalidate the contents > of that cache. I don't know squat about PDP10s so I can't help you > there. The KL10 uses a DATAO instruction to the "PAG" device to invalidate the TLB. On a KL10 with an MCA25, they've added a K ("Keep") bit to the section and map pointers, and it's possible to selectively invaliate only the TLB entries not marked "Keep". The motivation of the Keep bit is presumably to avoid invalidating translations of resident monitor pages. The KS10 probably works similarly, although I don't think it has the Keep bit. The documentation for both processors is present in the DECsystem-10/ DECSYSTEM-20 Processor Reference Manual, available in a 36.2 megabyte PDF file from my web site: http://www.36bit.org/dec/manual/ I don't think the PRM documents the Keep bit though. I only in the last few weeks found a manual with an appendix documenting the MCA25. Article 1836 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Well, tuning done.... Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 28 Message-ID: NNTP-Posting-Date: Tue, 31 Oct 2000 22:52:44 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-F16iLPCFLc52mrPBUK4kkkWuyaN2AB20i9C162lx+etVuhf/JfKThZVeKeao2Kk1bKOJ4TkMgpKrm2X!oNtGHAIsa5bvrzJVA1FDFCnRwug= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 01 Nov 2000 04:52:44 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1836 Hello folks: Well, my tuning for performance was just finished. On 800 Mhz, my emulator reached 3.74 MIPS(!). When I ran TOPS-10 on my emulator. Whoa! TYPE command is too fast for me to read. Entire document file displayed within a second. When I pressed CR at "BOOT>", entire TOPS-10 was almost instantly loaded within a second. When I typed SYSTAT, information was displayed and scrolled almost instantly. Also, I fixed my new I/O keyboard async routine when I discovered that some characters were lost while TOPS-10 asked my terminal for indentification. It cause the CTY derailement and misalignment. After it was fixed, TOPS-10 finally recongized my terminal as VT102. Well, I will finish up my rest of instructions for switch to direct-acces to AC blocks and EXTEND instructions before release to public. Then I will work on rest of PDP-6, KA, KI, and KI then VAX and PDP-11, etc.... Whew. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1668 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone2.usenetserver.com!news-out.usenetserver.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Found a bug in my emulator. Newsgroups: alt.sys.pdp10 References: <8pcir3$2qb6$1@nntp1.ba.best.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 16 Message-ID: NNTP-Posting-Date: Wed, 25 Oct 2000 21:36:26 CDT Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-vii8WsKeRTjZwYblyZpC1G/Uqav9kyRIfuejlBqUctiH3x3paMiKVuUPv6SbTi3LOJwy/YL8b6VGLsh!mTEsUg6CkT+VIBgMd5uzVAyOACU= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 26 Oct 2000 02:36:27 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1668 Joe Smith wrote: > UMOVE, UMOVEM, and PXCT are some of the most difficult instructions to > get right. Congratulations for clearing this hurdle. Joe: Yeah. I resolved all problems and everthing worked fine. Whew. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1876 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp-out.monmouth.com!newspeer.monmouth.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: PDP10 Emulator Status (Almost done) Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 134 Message-ID: NNTP-Posting-Date: Thu, 02 Nov 2000 22:24:12 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-c9qYM0khekQKCOwdz8oCxJ4CHWbemD6lGxtE66fxygu7x9v40vWrLQh5ZygTyiFoQgSsWLEumzaGb73!L5WkUJVzDMXDrgiq6yg6+Sigc4ep X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 03 Nov 2000 04:24:12 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1876 Hello folks: I fixed some math problems in my emulator because they produced incorrect values. That's why you noticed weird values in time, etc. when I posted my screen log files. Well, I ended up negated time but mentioned correct values. It reported only 2 seconds of CPU time usage when I executed 'kjob'. Also, I still have some problems with Quasar daemon. It randomly complained about errors - "Illegal UUO in PC 000000", "No such PID", and "Unknown node name" each boot time. I still am working on it and wrong-sign values. On 800 Mhz, MIPS meter was up to 3.8 (with debug facility), and up to 4.5 (without debug facility). I tested it without running X-Windows but it resulted the same. My emulator now is estimated 8 to 10 times faster than my previous one with time/poll every instruction. It destroyed 9/10 of its performance! -- Tim Stark My screen log here: KS10> boot mta0: Booting MTA0... Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 2-nov-2000 Time: 20:47 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 20:47:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] ?Error in detached job 7[ACTDAE] ? ?HALT at user PC 000017 %%TTY STOMPER - Starting ^C ^C .systat Status of KS10 at 20:47:29 on 02-Nov-100 Uptime 28, 61% Null time = 61% Idle + 0% Lost, 1% Overhead 11 Jobs in use out of 35. 7 logged in, 6 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 ACTDAE 90+40 ^C 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 10 2,5 33 INITIA 5+11 SL 0 11 [OPR] DET41 QUASAR 112+40 TO 0 12 2,5 36 INITIA 5+11 SL 0 13 [OPR] 41 4 ^C 0 14 2,5 40 INITIA 5+11 SL 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 4 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 330/2000 = 17% Virt. Core used = 363/2000 = 18% 265P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 73/(73+363) = 17% Average job size =297/11 = 27.0P+139/11 = 12.6P Total=436/11 = 39.6P No busy devices System File Structures: Name Free Mount DSKB 170470 3 Total Free 170470 .set opr cty .att 11 From job 3 .^C .cont ? Stopcode - ILM - Illegal memory reference at PC in .JBTPC Program QUASAR 5(1512) + GLXLIB 5(1601) error at PC 652064 in module GLXINT Last GLXLIB error at PC 000000 was 21; No such pid Crash block begins at 674000 [Stopping program] .att 3 From job 11 .kjob [LGNJSP Other jobs same PPN] Job 3 User OPR [1,2] Logged-off CTY at -20:-48:-45 on 2-Nov-100 Runtime: 0:00:02, KCS:27, Connect time: 0:-1:-45 Disk Reads:588, Writes:2, Blocks saved:0 . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1897 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: PDP10 Emulator Status (Almost done) Newsgroups: alt.sys.pdp10 References: <8tuggs$slj$5@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 27 Message-ID: <3wNM5.69058$YX4.2158647@news2.giganews.com> NNTP-Posting-Date: Fri, 03 Nov 2000 23:54:07 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-sJBMspUAhUdTlRfBsU/e45MrrCVYngjzemRenr9sLfTQHDe4xE/BPvmVxfKy1EWLj9z16038A2eIPyb!tzgzL8AGInG+S9dR5QcW4qKovRk= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sat, 04 Nov 2000 05:54:08 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1897 Timothy Stark wrote: > Well, I found a bug in my emaulated MUL instruction that cause negated > time from both postive numbers. I will fix it by rewrite my emulated > MUL instaurction routine. MUL instruction is located in CNTDT routine (at > end of that routine) in SCAN.MAC. I put "data watch" in my emulated > MUL routine and noticed that it is frequent in my log file. MUL instruction had been fixed. KJOB now reports correct time, etc... (No more negated time!) .kjob [LGNJSP Other jobs same PPN] Job 3 User OPR [1,2] Logged-off CTY at 0:50:03 on 4-Nov-100 Runtime: 0:00:00, KCS:16, Connect time: 0:01:01 Disk Reads:399, Writes:0, Blocks saved:0 Remaining one thing is fix Quasar problems right now... -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1876 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp-out.monmouth.com!newspeer.monmouth.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: PDP10 Emulator Status (Almost done) Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 134 Message-ID: NNTP-Posting-Date: Thu, 02 Nov 2000 22:24:12 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-c9qYM0khekQKCOwdz8oCxJ4CHWbemD6lGxtE66fxygu7x9v40vWrLQh5ZygTyiFoQgSsWLEumzaGb73!L5WkUJVzDMXDrgiq6yg6+Sigc4ep X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 03 Nov 2000 04:24:12 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1876 Hello folks: I fixed some math problems in my emulator because they produced incorrect values. That's why you noticed weird values in time, etc. when I posted my screen log files. Well, I ended up negated time but mentioned correct values. It reported only 2 seconds of CPU time usage when I executed 'kjob'. Also, I still have some problems with Quasar daemon. It randomly complained about errors - "Illegal UUO in PC 000000", "No such PID", and "Unknown node name" each boot time. I still am working on it and wrong-sign values. On 800 Mhz, MIPS meter was up to 3.8 (with debug facility), and up to 4.5 (without debug facility). I tested it without running X-Windows but it resulted the same. My emulator now is estimated 8 to 10 times faster than my previous one with time/poll every instruction. It destroyed 9/10 of its performance! -- Tim Stark My screen log here: KS10> boot mta0: Booting MTA0... Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 2-nov-2000 Time: 20:47 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 20:47:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] ?Error in detached job 7[ACTDAE] ? ?HALT at user PC 000017 %%TTY STOMPER - Starting ^C ^C .systat Status of KS10 at 20:47:29 on 02-Nov-100 Uptime 28, 61% Null time = 61% Idle + 0% Lost, 1% Overhead 11 Jobs in use out of 35. 7 logged in, 6 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 ACTDAE 90+40 ^C 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 10 2,5 33 INITIA 5+11 SL 0 11 [OPR] DET41 QUASAR 112+40 TO 0 12 2,5 36 INITIA 5+11 SL 0 13 [OPR] 41 4 ^C 0 14 2,5 40 INITIA 5+11 SL 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 4 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 330/2000 = 17% Virt. Core used = 363/2000 = 18% 265P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 73/(73+363) = 17% Average job size =297/11 = 27.0P+139/11 = 12.6P Total=436/11 = 39.6P No busy devices System File Structures: Name Free Mount DSKB 170470 3 Total Free 170470 .set opr cty .att 11 From job 3 .^C .cont ? Stopcode - ILM - Illegal memory reference at PC in .JBTPC Program QUASAR 5(1512) + GLXLIB 5(1601) error at PC 652064 in module GLXINT Last GLXLIB error at PC 000000 was 21; No such pid Crash block begins at 674000 [Stopping program] .att 3 From job 11 .kjob [LGNJSP Other jobs same PPN] Job 3 User OPR [1,2] Logged-off CTY at -20:-48:-45 on 2-Nov-100 Runtime: 0:00:02, KCS:27, Connect time: 0:-1:-45 Disk Reads:588, Writes:2, Blocks saved:0 . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 1897 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: PDP10 Emulator Status (Almost done) Newsgroups: alt.sys.pdp10 References: <8tuggs$slj$5@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 27 Message-ID: <3wNM5.69058$YX4.2158647@news2.giganews.com> NNTP-Posting-Date: Fri, 03 Nov 2000 23:54:07 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-sJBMspUAhUdTlRfBsU/e45MrrCVYngjzemRenr9sLfTQHDe4xE/BPvmVxfKy1EWLj9z16038A2eIPyb!tzgzL8AGInG+S9dR5QcW4qKovRk= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sat, 04 Nov 2000 05:54:08 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:1897 Timothy Stark wrote: > Well, I found a bug in my emaulated MUL instruction that cause negated > time from both postive numbers. I will fix it by rewrite my emulated > MUL instaurction routine. MUL instruction is located in CNTDT routine (at > end of that routine) in SCAN.MAC. I put "data watch" in my emulated > MUL routine and noticed that it is frequent in my log file. MUL instruction had been fixed. KJOB now reports correct time, etc... (No more negated time!) .kjob [LGNJSP Other jobs same PPN] Job 3 User OPR [1,2] Logged-off CTY at 0:50:03 on 4-Nov-100 Runtime: 0:00:00, KCS:16, Connect time: 0:01:01 Disk Reads:399, Writes:0, Blocks saved:0 Remaining one thing is fix Quasar problems right now... -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2013 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!feed2.news.rcn.net!rcn!nntp.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: PDP10 Emulator Status Newsgroups: alt.sys.pdp10 References: <3OFQ5.7402$%j3.61457@news6.giganews.com> <3A1A2E97.6E1AEDA4@softstar.it> <3A1AF1B3.D90C9758@bartek.net> <8vhcf9$84l@weyl.math.psu.edu> <3A1C4FBE.652A6945@bartek.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 22 Message-ID: NNTP-Posting-Date: Sun, 26 Nov 2000 15:06:14 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-U79U4L276lcnW0A8gtE5NUVyHBRTzPd+fH3FqVAuKz4B9+X68Dno+amMIIZJ4Ec9cVveiccjI9pPx6u!nErA3AWLVn5GMNHuNB/+PSeVaCI= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sun, 26 Nov 2000 21:06:14 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2013 Arthur Krewat wrote: > This one I like the most. I've been working with a company who's > credo seems to be "use Microsoft - we don't have to buy it". > Too bad they pay for it every day in terms of headaches for users > and the admins. I returned home from my thanksgiving vacation. Well, my employer is strictly Microsoft customer but I am using Red Hat Linux on my second desktop PC to develop my KS10 emulator. Using UNIX operating system is good for software developement if you just know C/C++ language. It is more difficult to develop software on Microsoft products. Soon I will resume to continue developing my KS10 emulator to complete. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2014 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!feed2.news.rcn.net!rcn!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Tim Stark's emulator, Lars GCC port and ... Newsgroups: alt.sys.pdp10 References: <3A213AEA.D783B415@bellatlantic.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 37 Message-ID: NNTP-Posting-Date: Sun, 26 Nov 2000 15:21:56 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-qe2ZJNiabXTVHBQljqsd0lYl9+Qe1CT0UwwiHGJjAAUD+YIUol+jr1VaGD3BIUovlqAGH5pZ+DLKBiv!yRB9CxgslQlj4mfxO8g2ueSk/uc= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sun, 26 Nov 2000 21:21:56 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2014 StrangeBrew wrote: > When Tim does release his emulator, what are the chances of starting > a distributed net group to run RC5 cracks on the emulated 10? Real soon but I do not know when yet. I returned home from my vacation. I will resume to develop my KS10 emulator to complete. EXTEND, DMUL, and FP instructions are not implemented yet. A bug in my emulator that cause galaxy daemons to crash is like finding a pin in the entire earth. :-( On DDT, when I pressed '.', it instantly displayed "KA10 floating instruction"... That means FP instructions not implemented yet. I tested my emulator on my Pentium II/400 Mhz laptop. It was up to 1.8 MIPS. On Pentium III/800E, it was up to 4.5 MIPS. > My plan is to run the emulator on one of my alphas, running netbsd, > so I am up for this!! > I just think it would be slick to have a "team pdp-10" running > distributed net code... Yeah. Good idea. If you run my emulator on Alpha, it will be faster because of 64-bit arch Alpha CPU. I need to find ethernet func specs for unibus (designed for VAX and PDP-11 computers). > I just wonder if it would be possible to run the 10 emulator on > a cray, running rc5 cracks.. I already ran it on my Cray-like PC -- Pentium III/800E system. :-) -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2039 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Tim Stark's emulator, Lars GCC port and ... Newsgroups: alt.sys.pdp10 References: <3A213AEA.D783B415@bellatlantic.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 17 Message-ID: NNTP-Posting-Date: Tue, 28 Nov 2000 07:10:37 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-D8Qw5lJomN7EMHybBibyZsHgPvXVgkyN51DkZQUfUVzE4FudxwyfA0CJcwNdHCPBpIo72ZNDski4S5s!kcOoKhHkVjK8y8hZqT7wj5BzmoI= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 28 Nov 2000 13:10:37 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2039 Timothy Stark wrote: > Real soon but I do not know when yet. I returned home from my vacation. > I will resume to develop my KS10 emulator to complete. EXTEND, DMUL, and > FP instructions are not implemented yet. A bug in my emulator that cause > galaxy daemons to crash is like finding a pin in the entire earth. :-( DMUL instruction had been implemented. Boot routine for RP/RM disk drive is being implemented according to SMFILE.TXT. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2177 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!nntp.frontiernet.net!nntp.gblx.net!nntp.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: TECO - compiled succesfully :-) Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 281 Message-ID: <4yEX5.43392$%j3.459109@news6.giganews.com> NNTP-Posting-Date: Wed, 06 Dec 2000 22:41:36 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-G8e5cySpZpTxN6fHAP1SMOPoJ1QAjmeKQ3+YMsDlvellbmj5RxhsDW1GtARWZXiIsMqRcDZ0Uxc2Ytv!qRe9x5n0n78tgoMBfns2p4WoYGPt X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 07 Dec 2000 04:41:36 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2177 Hello folks: Well, I tried compile teco.mac into teco.exe file. It was so successful but LINK attempted to exit but "Illegal Memory Reference" displayed instead. I believe that it happened during context switch or so. It contains BLT 17,17 instruction according to my log file. Well, I am trying figure this out later. I tried two different methods for compilation. First method is using COMPILE command. Second method is using R MACRO command. You can usei either. I included both methods for compilation here. It produced CREF file and resulted a neat listing. :-) -- Tim Stark Screen log here: KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 272 T 6 S 1] [Reading Monitor Pre-boot at C 272 T 8 S 13] Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 06-dec-2000 Time: 22:53 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 22:53:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] ?Error in detached job 7[ACTDAE] ? ?HALT at user PC 000017 %%TTY STOMPER - Starting ^C ^C .declare setsrc=sys:setsrc .setsrc *cp [10,7,teco] *^Z EXIT .dir TECO ERR 44 <155> 30-Apr-86 DSKB: [10,7,TECO] TECO EXE 36 <155> 1-Sep-88 24A(235) TECO MAC 273 <155> 30-Apr-86 Total of 353 blocks in 3 files on DSKB: [10,7,TECO] .copy [1,2]=teco.mac .setsrc *cp [1,2] *^Z EXIT .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TEST TXT 1 <057> 30-Nov-00 TECO MAC 273 <057> 30-Apr-86 Total of 402 blocks in 3 files on DSKB: [1,2] .del test.txt Files deleted: DSKB:TEST.TXT 01 Blocks freed .compile /cref /comp teco.mac MACRO: TECO EXIT .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 6-Dec-00 TECO CRF 647 <057> 6-Dec-00 Total of 1099 blocks in 4 files on DSKB: [1,2] .r macro *teco=teco.mac NO ERRORS DETECTED HI-SEG. BREAK IS 407002 PROGRAM BREAK IS 002157 CPU TIME USED 00:01.368 48P CORE USED *^Z .r link *teco/ssave=teco/go ? ?Illegal memory reference at user PC 000046 (EXE file was created successfully but link attempted to exit here.) .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 6-Dec-00 TECO CRF 647 <057> 6-Dec-00 TECO EXE 36 <057> 6-Dec-00 24A(235) Total of 1135 blocks in 5 files on DSKB: [1,2] .run teco *ex$$ .run teco *iThis is a first line... This is a second line... this is a third line... $$ *j$$ *t$$ This is a first line... *ht$$ This is a first line... This is a first line... This is a second line... this is a third line... *ex$$ ?NFO No File for Output *ewtext.txt$$ *ex$$ .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 6-Dec-00 TECO CRF 647 <057> 6-Dec-00 TECO EXE 36 <057> 6-Dec-00 24A(235) TEXT TXT 1 <057> 6-Dec-00 Total of 1136 blocks in 6 files on DSKB: [1,2] .type text.txt This is a first line... This is a second line... this is a third line... .r cref *teco=teco.crf [CRFXKC 14K core] *^Z .type teco.lst TECO VERSION 24A(235) MACRO %53B(1247) 22:54 6-Dec-:0 Page 1 TECO MAC 30-Apr-86 17:18 TEXT EDITOR AND CORRECTOR RC CLEME NTS/PMH/CAM/EAR/DML/JNG/BGS/DCE/MHK/CGN 1 TITLE TECO VERSION 24A(235) 2 SUBTTL TEXT EDITOR AND CORRECTOR RC CLEMENTS/PMH/CAM/EAR/DML/JNG/BGS/DCE/MHK/CGN 3 4 000235 EDIT==235 5 002401 000235 VERSION==XWD 2401,EDIT 6 (Rest of this listings were deleted) .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 6-Dec-00 TECO EXE 36 <057> 6-Dec-00 24A(235) TEXT TXT 1 <057> 6-Dec-00 TECO LST 656 <057> 6-Dec-00 Total of 1145 blocks in 6 files on DSKB: [1,2] .systat Status of KS10 at 22:58:35 on 06-Dec-100 Uptime 5:34, 95% Null time = 95% Idle + 0% Lost, 0% Overhead 7 Jobs in use out of 35. 7 logged in, 5 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 3 [OPR] CTY SYSTAT 22+SPY RN 7 4 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 ACTDAE 90+40 ^C 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 11 [OPR] DET41 QUASAR 112+40 TO 0 13 [OPR] 41 4 ^C 0 n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 304/2000 = 15% Virt. Core used = 325/2000 = 16% 303P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 40/(40+325) = 11% Average job size =274/7 = 39.1P+91/7 = 13.0P Total=365/7 = 52.1P No busy devices System File Structures: Name Free Mount DSKB 169370 2 Total Free 169370 .kjob [LGNJSP Other jobs same PPN] Job 3 User OPR [1,2] Logged-off CTY at 22:58:41 on 6-Dec-100 Runtime: 0:00:07, KCS:175, Connect time: 0:05:40 Disk Reads:3957, Writes:1782, Blocks saved:770 . KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2243 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Interesting article Newsgroups: alt.sys.pdp10 References: <90qqll$g55$1@nnrp1.deja.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 23 Message-ID: NNTP-Posting-Date: Fri, 08 Dec 2000 09:16:27 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-PGjzzflupLCqIXP4HP4hA4buGwaU4y+NgKtAvkehWVfXdATurgyrWVNxAeaWo6HQaP2FkNcYDLY7bBw!QddVG2Odm1ZyK6W9nDFIqL/SRvM= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 08 Dec 2000 15:16:27 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2243 Harris S. Newman wrote: > I just read an article on John Fletcher at: > http://www.ornl.gov/ORNL/SC/john_fletcher_1.html > In it it mentions that he wrote, with Mike Nemanic, an package that > emulates the pdp-10 on a Cray? > Anyone know about this? Anyone have access to it to post? I was so impressed to read it. Yes. my emulator runs much faster than a real PDP-10 on Pentium III 800EB system. If you run my emulator on Pentium IV 1.4 or 1.5 Ghz or Itanium (sp?), my emulator runs faster more than 10 times than a real PDP-10. On 800 Mhz system my emulator runs 10 times faster (4 MIPS) than a real KS. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2266 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!feed2.news.rcn.net!rcn!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: BLT 17,17 bug had been fixed but.. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 198 Message-ID: <6TiY5.46657$%j3.530564@news6.giganews.com> NNTP-Posting-Date: Fri, 08 Dec 2000 22:51:14 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-KzOqo273SV3qIdBeDaivyYL4lyMInEJg44YFjZsqGFMN8TOhomgGQxqTDU51wppRixnX79ZSd9eqMW8!0MX5KvmzkYCG6lGNWZ0pD3mlsJcu X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sat, 09 Dec 2000 04:51:14 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2266 Hello folks, Good news! I found a BLT 17,17 bug in my emulator that was responsible for GALAXY crashes, etc... But I now have another GALAXY-related problem. Now ACTDAE is complaining about stopcode WFO messages. After a BLT 17,17 bug had been fixed, some bugs disappeared within GALAXY daemons. QUASAR sometimes crashed with "?Illegal UUO at user PC 000000". This screen log showed that QUASAR did not crash yet. Some times, QUASAR did. I discovered that a BLT 17,17 always overwrote stack pointer when a block transfer was complete. When I attempted to boot TOPS-20 v4.1, it constantly displayed STKOVF fatal error messages. I decided to look into my debug log file and discovered that! Also, LINK was attempted to exit but it crashed at location 45. It was a BLT 17,17 instruction too. When it was fixed, now LINK normally exits. MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] Does anyone know about stopcode WFO message? Where did MOVEKEY come from? I tried to run ACTDAE again but it displayed the same message. ACTDAE always crashed with that same message. Thank you! -- Tim Stark Well, my screen log here: KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 272 T 6 S 1] [Reading Monitor Pre-boot at C 272 T 8 S 13] Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 8-dec-2000 Time: 22:56 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 22:56:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] %%TTY STOMPER - Starting ^C ^C .systat Status of KS10 at 22:56:48 on 08-Dec-100 Uptime 48, 83% Null time = 83% Idle + 0% Lost, 1% Overhead 7 Jobs in use out of 35. 7 logged in, 5 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 ACTDAE 132+40 TO 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 11 [OPR] DET41 QUASAR 112+40 SL 0 13 [OPR] 41 4 ^C 0 n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 346/2000 = 17% Virt. Core used = 367/2000 = 18% 261P Core left Active swapping ratio = 22/628 = .03 Virt. Core saved by sharing = 40/(40+367) = 10% Average job size =316/7 = 45.1P+91/7 = 13.0P Total=407/7 = 58.1P No busy devices System File Structures: Name Free Mount DSKB 169390 2 Total Free 169390 .set opr cty .att 7 From job 3 .^C .cont MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] .r actdae MOVEKEY ^C ^C .r actdae MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] .att 3 From job 7 .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 6-Dec-00 TECO EXE 36 <057> 6-Dec-00 24A(235) TEXT TXT 1 <057> 6-Dec-00 TECO LST 656 <057> 6-Dec-00 Total of 1145 blocks in 6 files on DSKB: [1,2] .compile /cref /comp teco MACRO: TECO EXIT .r link * Monitor KS10 System uptime :03:59 Current date/time Friday 8-Dec-100 23:00:00 teco/ssave=teco/go EXIT .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 8-Dec-00 TECO EXE 36 <057> 8-Dec-00 24A(235) TEXT TXT 1 <057> 6-Dec-00 TECO LST 656 <057> 6-Dec-00 TECO CRF 647 <057> 8-Dec-00 Total of 1792 blocks in 7 files on DSKB: [1,2] .run teco *ex$$ . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2290 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!news.gtei.net!europa.netcrusader.net!152.163.239.129!portc01.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: BLT 17,17 bug had been fixed but.. Newsgroups: alt.sys.pdp10 References: <6TiY5.46657$%j3.530564@news6.giganews.com> <90teal$10b$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Lines: 411 Message-ID: NNTP-Posting-Date: Sat, 09 Dec 2000 18:31:37 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-0ddOajQKj2tO8jCJczlu0SCP4zjsTQgd3cb0r4nXX0X9DA71E75k8Qg4jmPe7YGAv8C3713f3cbpYMU!0OuWEyQTjxLgZB8nqdvqUaKIyPrf X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sun, 10 Dec 2000 00:31:38 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2290 Barb: Yeah. You are right about sharable high-segement. I discovered that all INITIA programs used same shared high segement too. I tried to get and save it. It resulted the same. Oh, I forget to SET W V F. At end of screen log, I did SET W V F to trace ACTDAE's access to files. Hmm. SCDMAP.SYS is not existing that cause ACTDAE to crash possibly. No, QUASAR did not crash on my 800 Mhz system. If 450 Mhz system, QUASAR sometimes crashes. Thank you! -- Tim Stark That is screen log here: KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 272 T 6 S 1] [Reading Monitor Pre-boot at C 272 T 8 S 13] Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 9-dec-2000 Time: 18:51 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 18:51:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] %%TTY STOMPER - Starting ^C ^C .systat Status of KS10 at 18:51:15 on 09-Dec-100 Uptime 14, 49% Null time = 49% Idle + 0% Lost, 2% Overhead 35 Jobs in use out of 35. 7 logged in, 6 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 5 2,5 0 INITIA 5+11 SL 0 6 2,5 1 INITIA 5+11 SL 0 7 [OPR] DET41 ACTDAE 132+40 TO 0 8 2,5 2 INITIA 5+11 SL 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 10 2,5 3 INITIA 5+11 SL 0 11 [OPR] DET41 QUASAR 112+40 SL 0 12 2,5 4 INITIA 5+11 SL 0 13 [OPR] 41 4 ^C 0 14 2,5 5 INITIA 5+11 SL 0 15 2,5 6 INITIA 5+11 SL 0 16 2,5 7 INITIA 5+11 SL 0 17 2,5 10 INITIA 5+11 SL 0 18 2,5 11 INITIA 5+11 SL 0 19 2,5 12 INITIA 5+11 SL 0 20 2,5 13 INITIA 5+11 SL 0 21 2,5 14 INITIA 5+11 SL 0 22 2,5 15 INITIA 5+11 SL 0 23 2,5 16 INITIA 5+11 SL 0 24 2,5 17 INITIA 5+11 SL 0 25 2,5 20 INITIA 5+11 SL 0 26 2,5 21 INITIA 5+11 SL 0 27 2,5 22 INITIA 5+11 SL 0 28 2,5 23 INITIA 5+11 SL 0 29 2,5 24 INITIA 5+11 SL 0 30 2,5 25 INITIA 5+11 RN 0 31 2,5 26 INITIA 5+11 RN 0 32 2,5 27 INITIA 5+11 SL 0 33 2,5 30 INITIA 5+11 SL 0 34 2,5 31 INITIA 5+11 SL 0 35 2,5 32 INITIA 5+11 SL 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 28 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 420/2000 = 21% Virt. Core used = 525/2000 = 26% 103P Core left Active swapping ratio = 43/628 = .06 Virt. Core saved by sharing = 337/(337+525) = 39% Average job size =459/35 = 13.1P+403/35 = 11.5P Total=862/35 = 24.6P No busy devices System File Structures: Name Free Mount DSKB 169070 3 Total Free 169070 .set opr cty .att 7 From job 3 .^C .cont MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] .ddt VMDDT prmadr? GLXINT prmadr/ BLKI 100,ACTDA1#+1 RETADR#/ BLKI 100,ACTINI#+15 THSPRM#/ ACTDA2# S1%S2#/ 11 prmadr^ THSPRM#/ ACTDA2# ^ RETADR#/ BLKI 100,ACTINI#+15 ^ PRMADR#/ BLKI 100,ACTDA1#+1 ^ INTRPC/ 0 ^ BASINT/ ACTPSI# ^C .get sys:actdae.exe Job setup .save dskb[1,2]actdae.exe ?save dskb[1,2]a? .save dskb:p[1,2]actdae.exe ACTDAE saved .rename actdae.exe actdae.hld ?CMLCME Command error detected before: .hld .rename actdae.hld=actdae.exe Files renamed: ? No file named DSK:ACTDAE.EXE .att 3 From job 7 .ren actdae.hld=sys:actdae.exe Files renamed: DSKB:ACTDAE.EXE .copy sys:actdae.exe=actdae.exe .att 7 From job 3 .r actdae MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] . KS10> quit Exiting PDP10 Emulator [sword7@sword7-0 pdp10]$ ./ks10 -f ks10.cfg Welcome to KS10 Emulator v0.8 (Alpha) KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 272 T 6 S 1] [Reading Monitor Pre-boot at C 272 T 8 S 13] Now running... BOOT V4(76) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 07-Oct-88 Why reload: new Date: 9-dec-2000 Time: 18:56 Startup option: go [Rebuilding the system search list from the HOM blocks] [Rebuilding the active swapping list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] KS10 18:56:01 CTY system 5000 Connected to Node CENTRA(0) Line # 42 [Checking terminal type . . . found VT100] .LOGIN 1,2 .R OPR [CCPWFD Waiting for file daemon to start] %%TTY STOMPER - Starting ^C ^C .systat Status of KS10 at 18:56:15 on 09-Dec-100 Uptime 14, 52% Null time = 52% Idle + 0% Lost, 2% Overhead 35 Jobs in use out of 35. 7 logged in, 6 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+11 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 22+SPY RN 0 4 [OPR] DET41 FILDAE 17 HB 0 5 2,5 35 INITIA 3 RN 0 6 2,5 1 INITIA 5+11 SL 0 7 [OPR] DET41 ACTDAE 132+40 TO 0 8 2,5 2 INITIA 5+11 SL 0 9 [OPR] DET41 DAEMON 20+SPY SL 0 10 2,5 3 INITIA 5+11 SL 0 11 [OPR] DET41 QUASAR 112+40 RN 0 12 2,5 4 INITIA 5+11 SL 0 13 [OPR] 41 4 ^C 0 14 2,5 5 INITIA 5+11 SL 0 15 2,5 6 INITIA 5+11 SL 0 16 2,5 7 INITIA 5+11 SL 0 17 2,5 10 INITIA 5+11 SL 0 18 2,5 11 INITIA 5+11 SL 0 19 2,5 12 INITIA 5+11 SL 0 20 2,5 13 INITIA 5+11 SL 0 21 2,5 14 INITIA 5+11 SL 0 22 2,5 15 INITIA 5+11 SL 0 23 2,5 16 INITIA 5+11 SL 0 24 2,5 17 INITIA 5+11 SL 0 25 2,5 20 INITIA 5+11 SL 0 26 2,5 21 INITIA 5+11 SL 0 27 2,5 22 INITIA 5+11 SL 0 28 2,5 23 INITIA 5+11 SL 0 29 2,5 24 INITIA 5+11 SL 0 30 2,5 25 INITIA 5+11 SL 0 31 2,5 26 INITIA 5+11 SL 0 32 2,5 27 INITIA 5+11 SL 0 33 2,5 30 INITIA 5+11 SL 0 34 2,5 31 INITIA 5+11 SL 0 35 2,5 32 INITIA 5+11 SL 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 11 27 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 40 2 Swapping space used = 419/2000 = 21% Virt. Core used = 523/2000 = 26% 105P Core left Active swapping ratio = 177/628 = .28 Virt. Core saved by sharing = 326/(326+523) = 38% Average job size =457/35 = 13.0P+392/35 = 11.2P Total=849/35 = 24.2P No busy devices System File Structures: Name Free Mount DSKB 168760 3 Total Free 168760 .set opr cty .att 7 From job 3 .^C .cont MOVEKEY ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] .dir sys:actdae.exe ACTDAE EXE 304 <057> 9-Dec-00 2B(155) DSKB: [1,4] .att 3 From job 7 .del actdae.exe Files deleted: DSKB:ACTDAE.EXE 304 Blocks freed .dir BACKUP EXE 128 <057> 12-Oct-00 5A(625) DSKB: [1,2] TECO MAC 273 <057> 30-Apr-86 TECO REL 51 <057> 8-Dec-00 TECO EXE 36 <057> 8-Dec-00 24A(235) TEXT TXT 1 <057> 6-Dec-00 TECO LST 656 <057> 8-Dec-00 Total of 1145 blocks in 6 files on DSKB: [1,2] .dir actdae.* %WLDNSF No such files as actdae.* .dir sys:actdae.* ACTDAE HLD 304 <055> 1-Sep-88 DSKB: [1,4] ACTDAE SYS 36 <157> 24-Jun-88 ACTDAE EXE 304 <057> 9-Dec-00 2B(155) Total of 644 blocks in 3 files on DSKB: [1,4] .del sys:actdae.exe Files deleted: DSKB:ACTDAE.EXE 304 Blocks freed .ren actdae.exe=sys:actdae.hld Files renamed: DSKB:ACTDAE.HLD .dir sys:actdae.* ACTDAE EXE 304 <055> 1-Sep-88 2B(155) DSKB: [1,4] ACTDAE SYS 36 <157> 24-Jun-88 Total of 340 blocks in 2 files on DSKB: [1,4] .set w f v .r actdae [S:ACTDAE 2B(155)] [ + GLXLIB 5(1601)] [FL20: DSKB0:USAGE.OUT[1,7]] [FL20: DSKB0:USAGE.OUT[1,7]] [FE20: DSKB0:USAGE.OUT[1,7]] [FL21: DSKB0:FAILUR.LOG[1,7]] [FE21: DSKB0:FAILUR.LOG[1,7]] [FL22: DSKB0:USEJOB.BIN[1,7]] [FE22: DSKB0:USEJOB.BIN[1,7]] [FL23: DSKB0:ACTDAE.SYS[1,4]] [FL23: DSKB0:ACTDAE.SYS[1,4]] [FE23: DSKB0:ACTDAE.SYS[1,4]] MOVEKEY [FL23: DSKB0:ACTDAE.SYS[1,4]] [R23: DSKB0:ACTDAE.SYS[1,4]] [FL23: DSKB0:ACTDAE.SYS[1,4]] [FL23: DSKB0:ACTDAE.SYS[1,4]] [FE23: DSKB0:ACTDAE.SYS[1,4]] [FL24: SYS:SCDMAP.SYS, error 0] [FL22: DSKB0:USEJOB.BIN[1,7]] [FE22: DSKB0:USEJOB.BIN[1,7]] ? Stopcode - WFO - WTO Function in S1 out of range at address PRMADR Program ACTDAE 2B(155) + GLXLIB 5(1601) error at PC 653015 in module GLXINT Last GLXLIB error at PC 632245 was 11; File not found Crash block begins at 674000 [Stopping program] . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2404 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news-hog.berkeley.edu!ucberkeley!nntp.cs.ubc.ca!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Help! TOPS-20 Installation Newsgroups: alt.sys.pdp10 References: <3LTY5.118666$DG3.2359238@news2.giganews.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 130 Message-ID: NNTP-Posting-Date: Mon, 11 Dec 2000 08:16:00 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-DAL0XYKGCERRDpS6QjJapStg0DCjFfy3nnNNmqlNh0V7kKalNcJ4L2xsqjPeTbVEmKxyDt6d9BdNGDg!hRvemkq0beC5nicESEnq/th0YTIQ X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Mon, 11 Dec 2000 14:16:00 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2404 Mark Crispin wrote: > So you got to EXEC0. Excellent. This means that job 0 fork 0 has been > created, and it's trying to do the system initialization stuff that has to > run at process level. > If it did, in fact, output the CRLF, then it should be calling routine > CHKHOM which will go and read the home blocks for every physical disk on > the system. If it gets to CHKHOM, let me know and I'll tell you what to > do next. Mark: Yes, CHKHOM had been executed on my emulator. With debug ioregs and iodata turned on, it looks like that CHKHOM tried to access my emulated RP06 disk. It checked registers first but quietly gave up without reading blocks from my disk image. Also, an interrupts occured every second to check disk drives for status. My debug log showed that. Look at three log files below for more information. -- Tim Stark My screen log file: KS10> boot mta0: Booting MTA0... Now running... MTBOOT>/l MTBOOT>/g140 EDDT dbugsw/ 0 1 eddtf/ 0 1 fsiini$b chkhom$b 147$g $1B>>FSIINI/ JSP CX,SAVP $p $2B>>CHKHOM#/ JSP CX,SAVPQ $p ^@^@^@^@^@^@ [invisible Ctrl-@ characters printed but they are visible in vi editor.] KS10> quit My debug log file: CPU: 545322 instructions per second. CTY: 147$g PAGER: *** Page Fail Trap: 201203,,760540 at PC 005513 PAGER: *** Page Fail Trap: 201203,,760010 at PC 005513 PAGER: *** Page Fail Trap: 011000,,247000 at PC 014235 PAGER: *** Page Fail Trap: 011000,,250000 at PC 014235 CPU: 623955 instructions per second. CPU: 420243 instructions per second. CPU: 405703 instructions per second. CTY: $1B>>FSIINI/ JSP CX,SAVP $p PAGER: *** Page Fail Trap: 001000,,751000 at PC 025061 CPU: 408766 instructions per second. CPU: 398913 instructions per second. CTY: $2B>>CHKHOM#/ JSP CX,SAVPQ $p CPU: 552983 instructions per second. CTY: CPU: 1607547 instructions per second. \ CPU: 1641816 instructions per second. > a null job is running in AC3. CPU: 1637163 instructions per second. / My debug log file #2: (with debug ioregs and iodata turned on) CPU: 452636 instructions per second. CTY: $1B>>FSIINI/ JSP CX,SAVP $p PAGER: *** Page Fail Trap: 001000,,751000 at PC 025061 CPU: 462019 instructions per second. CPU: 450256 instructions per second. CPU: 443522 instructions per second. CTY: $2B>>CHKHOM#/ JSP CX,SAVPQ $p UBA: (W) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) <= 000000 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (R) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) => 000000 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (W) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) <= 000000 UBA: (W) Controller: 1 Register Address: 776700 RH11: RHCS1 (00) <= 000100 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (R) Controller: 1 Register Address: 776714 RP06: RPER1 (14) => 000000 CTY: CPU: 1619605 instructions per second. UBA: (W) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) <= 000000 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (R) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) => 000000 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (W) Controller: 1 Register Address: 776710 RH11: RHCS2 (10) <= 000000 UBA: (W) Controller: 1 Register Address: 776700 RH11: RHCS1 (00) <= 000100 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (R) Controller: 1 Register Address: 776714 RP06: RPER1 (14) => 000000 UBA: (R) Controller: 1 Register Address: 776700 RP06: RPCS1 (00) => 000000 RH11: RHCS1 (00) => 000300 UBA: (R) Controller: 3 Register Address: 772440 TM02: MTCS1 (00) => 004000 RH11: RHCS1 (00) => 004300 CPU: 1748183 instructions per second. [repeating messages beyond here...] -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2493 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: TOPS-20 tape hunt: status report. Newsgroups: alt.sys.pdp10 References: <9163q4$fe@dispatch.concentric.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 27 Message-ID: NNTP-Posting-Date: Tue, 12 Dec 2000 20:40:51 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-32Q/gtLUaUzsIap/0G7zklELRCglynAoS7jPPrbaG6C4OuU5yh+8UX6B3apDl4xRwRl4pj0+NiHIjLc!kIWa29Lymm5EBBa/4MvbjCT7qug= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 13 Dec 2000 02:40:51 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2493 Erno Palonheimo wrote: >> Assuming that you can dig up more tapes over the next few weeks, and >> can get them all successfully read, I'm sure they'll make a great >> addition to the PDP-10 archives. I'm looking forward to combining >> them with Tim Stark's emulator. > Well, I think I'll revive our TOPS-20 installation in the emulator as > soon as it starts to emulate KL10. Well, my TS-10 emulator will support all PDP and VAX processors, not just KS10 processor. I will release initial KS10 emulator when I was able to run TOPS-10 v7.04 and TOPS-20 v4.1 operating system without any problems. Also, I already worked on KL10 implementation as well. I still need KS diagnostics package to test all instructions first before I release my KS-10 emulator. I have it but I have a problem with it because it wants running TOPS-20 operating system (did not recongize TOPS-10 filesystem). -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2497 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!news.maxwell.syr.edu!wn4feed!wn2feed!worldnet.att.net!207.207.0.27!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: TOPS-20 tape hunt: status report. Newsgroups: alt.sys.pdp10 References: <9163q4$fe@dispatch.concentric.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 35 Message-ID: NNTP-Posting-Date: Tue, 12 Dec 2000 22:08:58 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-7H3d6UkFKQV0Cy63M4hwOi1WnUeqWitwPBkVggLsCeMlrG/zM65JJa8ZSdJD1LxqupsSpKo84ewjcMe!QyHc6VO3B+6Z7Cf5qgWF/jC1r4s= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 13 Dec 2000 04:08:58 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2497 Eric Smith wrote: > I strongly suggest doing the VAX stuff and the PDPs other than the -6 and > -10 as separate projects, since there's so little in common. That's not > to say that you won't be able to reuse some of your code, but it's not > clear that there's any benefit to mashing it all together in one big > lump. Yeah. I am aware of that. I am using -DKS10, -DKL10, etc. to compile separately. >> I will release initial KS10 emulator when I was able >> to run TOPS-10 v7.04 and TOPS-20 v4.1 operating system without any >> problems. > I suppose we'd better be very patient, then, since TOPS-10 v7.04 and > TOPS-20 v4.1 never ran "without any problems" on *anything*. Hmmm. A few instructions are left to complete KS-10 implementation. Double-precision floating points and some EXTEND instructions are to go. All signle-precision floating instructions had been implemented recently. TOPS-10 v7.04 was running fine execpt Galaxy daemons. Also, I found and fixed a few bugs in PI and RP06 that blocked TOPS-20 from being booted. Also, today I rewrote XMOVEI and XHLLI instructions for both KL10 and KS10. I separated old code (SETMI and HLLI) from that for KA10 and KI10 by using -DKx10 suboptions. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2496 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!logbridge.uoregon.edu!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Good news! TOPS-20 v4.1 booted successfully Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 52 Message-ID: NNTP-Posting-Date: Tue, 12 Dec 2000 21:56:59 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-7Il/cZdSgF0P8dLVeE00byHFJCjbfQfzrrArdrgl6J4gFZqlYraDDpQE7NMdXc2uUprMPTow26Rgy5r!noLuXEgXB8wYJPXOkIgIUmjic7c= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 13 Dec 2000 03:56:59 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2496 Hello folks: Good news! I found three bugs in my emulator blocked TOPS-20 v4.1 from being booted. First bug is in my emulated RP06 disk routines. I forget to put RPCS1_DVA in RPCS1 registers when disk drive routines had been initialized or reset. I used EDDT to set UDSKIO breakpoint and turned trace on in my emulator then pressed esc-p to continue. Then I looked into my debug log file carefully. I noticed that it checked if DVA bit is on. If not, it did nothing with disk drives. If set, it will read blocks from RP06 disk. That bug had been fixed. When I compiled and ran my emulator again but it returned back to KS10> prompt without any error report. I turned debug interrupts on and ran it. I found out that it attempted to issue vector interrupt to KS10 processor. It ended up wrong location that contains complete zero word. It cause HALT for illegal interrupt instruction. I fixed first bug to report error correctly. I compiled and tested it. Now it reports illegal instruction (where and its content). Now I fixed another bug in the same routine quickly. I changed from pRead36 to vRead36 with executive mode to fetch interrupt instruction correctly. Another two bugs had been fixed. I compiled it again and run it. TOPS-20 v4.1 finally printed a message! Whew! Mark, what next? I need instructions to how to install TOPS-20 v4.1 system. Thank you!! -- Tim Stark That is my screen log file: KS10> boot mta0: Booting MTA0... Now running... MTBOOT> PS NOT FOUND. WHAT IS THE NAME OF THE PUBLIC STRUCTURE? KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2507 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Good news! TOPS-20 v4.1 booted successfully Newsgroups: alt.sys.pdp10 References: User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 89 Message-ID: NNTP-Posting-Date: Wed, 13 Dec 2000 08:56:21 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-jmgvmgA+0pfQH5bwBqlkrA1tyO5ct8W3L1V30hA90Rm91vt2W9E1vmCxfDFCM+oXrstX73TJDI3RRoZ!HicLXEB5T7lornmUEATGxoUhRfk= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 13 Dec 2000 14:56:21 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2507 Mark Crispin wrote: > OK, good. This time, start at location 143, which will lead you through > the dialog to build the public structure. The default answers are OK. > After you do all that, you'll eventually get to "DDMP RUNNING". Now do > ^C. If all goes well, you'll get an "MX>" prompt. > Let me know if you get that far. Hello folks (and Mark): Ok. When I tried to answer to first question but TOPS-10 only received first character and ignored rest of my data. Yes, I quickly found a bug in my FE routines. I found out that TOPS-20 did not clear location CTYIWD completely but just turned off valid bit. That cause my FE routines to locked up forever but did not affect rest of my emulator. My FE routine checked if CTYIWD is zero word, then put character from keyboard into that. I fixed a bug by changing routine to check if valid bit is reset before go ahead to put character on it. Otherwise, put character in queue that is waiting for next available space in CTYIWD. Now TOPS-20 was able accept all of my data right now... Also, I ended up another problem when my FE routines stalled middle of questions. I found out that character interrupt failed. I fixed a bug quickly by re-send character interrupt if TOPS-20 failed to clear valid bit in CTYIWD after countdown (in my FE queue routines). A problem went away. Well, I ended up BUGHLT "BTBCRT". Well, I have BUGSTRINGS.TXT that was found in that same TOPS-20 installation tape. It means "Can't initialize bit table for public structure." Mark, do you know about that? What is solution? My debug log file showed that TOPS-20 loaded MONITOR.EXE from a tape and put it into my emulated RP06 disk file then updated BAT blocks. Then BUGHLT message disaplayed! Thank you! -- Tim Stark My screen log file: KS10> boot mta0: Booting MTA0... Now running... MTBOOT>/l MTBOOT>/g143 [FOR ADDITIONAL INFORMATION TYPE "?" TO ANY OF THE FOLLOWING QUESTIONS.] DO YOU WANT TO REPLACE THE FILE SYSTEM ON THE PUBLIC STRUCTURE? YES DO YOU WANT TO DEFINE THE PUBLIC STRUCTURE? YES HOW MANY PACKS ARE IN THIS STRUCTURE: 1 ON WHICH "CHANNEL,UNIT" IS LOGICAL PACK # 0 MOUNTED: 0,0 DO YOU WANT THE DEFAULT SWAPPING SPACE? YES DO YOU WANT THE DEFAULT SIZE FRONT END FILE SYSTEM? YES DO YOU WANT THE DEFAULT SIZE BOOTSTRAP AREA? YES [STRUCTURE "PS" SUCCESSFULLY DEFINED] [PS MOUNTED] ?PS UNIT 0 HAS NO BAT BLOCKS. DO YOU WANT TO WRITE A SET OF PROTOTYPE BAT BLOCKS? YES ********** *BUGHLT "BTBCRT" ********** CPU: HALT (Reason Code: 005037) at PC 005037 KS10> exit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2521 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!uninett.no!newsfeed1.enitel.no!masternews.telia.net!news-sto.telia.net!news.defero.net!junk.nocrew.org!not-for-mail From: Lars Brinkhoff Newsgroups: alt.sys.pdp10 Subject: Re: Good news! TOPS-20 v4.1 booted successfully Organization: nocrew Lines: 9 Sender: lars@junk.nocrew.org Message-ID: <853dfrpl0z.fsf@junk.nocrew.org> References: <9184lg$1utj$1@citadel.in.taronga.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.7 Date: Thu, 14 Dec 2000 06:38:52 GMT NNTP-Posting-Host: 212.73.17.42 X-Complaints-To: abuse@defero.net X-Trace: news.defero.net 976775932 212.73.17.42 (Thu, 14 Dec 2000 07:38:52 MET) NNTP-Posting-Date: Thu, 14 Dec 2000 07:38:52 MET Xref: nntp1.ba.best.com alt.sys.pdp10:2521 "Don Chiasson" writes: > I have a copy of Gorin's excellent book on assembly language > programming. It's also available in PostScript format by anonymous FTP from toad.xkl.com. -- http://lars.nocrew.org/ Article 2611 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.direct.ca!look.ca!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-20: MX> Prompt finally printed. Date: Sat, 16 Dec 2000 21:55:17 -0800 Organization: Networks & Distributed Computing Lines: 55 Message-ID: References: NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 977032520 206878 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: lampin To: Timothy Stark In-Reply-To: Xref: nntp1.ba.best.com alt.sys.pdp10:2611 On Sun, 17 Dec 2000, Timothy Stark wrote: > Now I finally got a MX> prompt. What is about "INTERRUPT AT 0" for? > Is it normal or another bug? I did not press ^C to interrupt that. Excellent! All is well!! INTERRUPT AT 0 basically means that the top-level process for the job got an error at PC 0. Since the top-level process for the job at this point is totally empty with no initial PC and no memory, that's what would happen. In other words, at this particular point, that's what is supposed to happen. MX> is the prompt for the mini-EXEC. TOPS-20 has a miniature command decoder built into the kernel. Unlike TOPS-10, where the entire command decoder is in the kernel, the TOPS-20 mini-EXEC only has a few commands and is only accessible by privileged users. Assuming that you have the installation tape mounted, here is what you do after a 143 start, filesystem create, and entering the mini-EXEC. Note that means the return key. 1) Type "GMTA0:". "GET FILE MTA0:" is echoed followed by another MX> prompt. 2) Type "S". "START" is echoed. You should now be in the TOPS-20 command decoder (EXEC). So now you know two mini-EXEC commands, G and S. There are three or four others which we can discuss later (you don't need to know them for now). 3) At the "@" prompt, type "ENABLE". You should now have a "$" prompt. You need to have your privileges enabled for the next step. 4) Type "MTA0:". This should give you the "DLUSER>" prompt. 5) Type "LOAD MTA0:". Wait for it to finish and give you another "DLUSER>" prompt. 6) Type "EXIT". You should be at the "$" prompt. 7) Type "MTA0:". This should give you the "DUMPER>" prompt. 8) Type "TAPE MTA0:". You should get another "DUMPER>" prompt 9) Type "RESTORE". It should say something like "LOADING FILES INTO PS:" (or some other directory). If it does not say "LOADING FILES", let me know what it does say. 10) Repeat step 9 until you reach end of tape. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article 2637 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: KLAD contents? Date: 17 Dec 2000 20:51:06 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 20 Sender: alderson+news@panix6.panix.com Message-ID: References: <3A3CFE93.C23F9CF3@bartek.net> NNTP-Posting-Host: panix6.panix.com X-Trace: news.panix.com 977104266 25371 166.84.0.231 (18 Dec 2000 01:51:06 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 18 Dec 2000 01:51:06 GMT X-Newsreader: Gnus v5.7/Emacs 20.6 Xref: nntp1.ba.best.com alt.sys.pdp10:2637 Timothy Stark writes: > I am trying figure how to run KLAD diagnostics at this time. When a Field Circus engineer had to run diags on a KL (for example), he would run a special task under RSX-20F, called DBOOT. This would in turn read commands from the console, or take them from .CMD files with informative names like B.CMD and BT.CMD. If you look at those, you will find that the former runs a string of 11-based diagnostics to check out the physical state of the KL, while the latter runs the instruction set diags if the KL passes the B string. The .CMD files of course refer to .A10 and .A11 files, but you can substitute the .SAV files and run each one individually, in the order they occur in the BT string. Under Tops-10. -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless Article 2638 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!feed2.news.rcn.net!rcn!nntp.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: TOPS-10 v7.03 installation successfully... Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 219 Message-ID: NNTP-Posting-Date: Sun, 17 Dec 2000 21:51:48 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-of6pFe9/tECs7dduyZHdZ/XNmJi2CxH5urizwnMZiTWCAeJB5uN5tXtD8PoOL64KOpUeT0xuRLPIMyy!Xs1le0HPlLih1XMNvVfjgyVOzz/9 X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Mon, 18 Dec 2000 03:51:48 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2638 Barb: I re-installed TOPS-10 v7.03 software. I corrected file protections on SYS:. I changed to <055> from <057> on all SYS: files. However, I did entered 'PROTECTION 055' before 'RESTORE SYS:=*.EXE,...'. Then I made DSKB: bootable by running SMFILE. I rebooted system from RPA0: directly without any problems. Good news! What guess? GALAXY daemons are running fine on TOPS-10 v7.03. But not on v7.04. Still problem with ACTDAE. I checked with it but just "MOVEKEY" but no fatal errors. Only I have a problem with OPR and REACT programs. When I entered commands on OPR> and REACT> prompt, I received no response - just sat forever. I had to press ^C to get rid of it. Ok, I will send you a copy of ACTDAE.MAC that you request soon. Ok, I looked into my debug.log that my emulator produced. Hmmm. No page fault trap messages! That's why GALAXY daemons did not crash. I tried to assign MTA0: on BACKUP when GALAXY daemons are running, they rejected my request! I was told that MTA0: is not available. How do I make MTA0: accessible and available? Now, I will work on KLAD diagnostics to test my emulated instructions. -- Tim Stark Here is screen log file: KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 216 T 9 S 11] [Reading Monitor Pre-boot at C 216 T 12 S 3] Now running... BOOT V3(47) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 03-19-86 Why reload: new Date: 17-dec-2000 Time: 22:13 Startup option: go [Rebuilding the system search list from the HOMe blocks] [Rebuilding the active swapping list from the HOMe blocks] [Rebuilding the system dump list from the HOMe blocks] KS10 22:13:01 CTY system 5000 Connected to Node (0) Line # 42 .LOGIN 1,2 .R OPSER [OPRPAF Processing auto command file] ?OPRALF LOOKUP failure 0 * [CCPWFD Waiting for file daemon to start] *SYSTAT !22:13:17(0) Status of KS10 at 22:13:18 on 17-Dec-100 Uptime 16, 80% Null time = 80% Idle + 0% Lost, 4% Overhead 14 Jobs in use out of 35. 12 logged in, 11 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET STOMPR 9+8 SL 0 01 2 2,5 DET CRSCPY 8+15 SL 0 $ 3 [OPR] CTY OPSER 5+6 RN 1 4 [OPR] DET ACTDAE 137+39 TO 1 5 [OPR] DET DAEMON 20+SPY SL 0 6 [OPR] DET FILDAE 17 HB 0 7 [OPR] DET QUASAR 109+39 SL 1 8 [OPR] DET PULSAR 65+39 HB 0 9 [OPR] DET CATLOG 30+39 HB 0 10 2,5 P0J3 SYSTAT 21+SPY RN 0 $ 11 [OPR] DET ORION 94+39 SL 1 12 [OPR] DET BATCON 38+39 SL 0 13 [OPR] 41 4 ^C 0 14 [OPR] DET LPTSPL 59+39 HB 0 $ means Execute Only n means job runs in HPQ n Jnn is the controlling job, Pnn corresponds to TTY43+nn High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 CRSCPY DSKB [SYS] 15 1 OPSER DSKB [SYS] 6 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 642/2000 = 32% Virt. Core used = 684/2000 = 34% 91P Core left Active swapping ratio = 32/775 = .04 Virt. Core saved by sharing = 234/(234+684) = 25% Average job size =616/14 = 44.0P+302/14 = 21.5P Total=918/14 = 65.5P No busy devices System File Structures: Name Free Mount DSKB 184610 3 Total Free 184610 .KJOB . *^C .R OPR OPR> ? Invalid OPR command specified: "OPR" 22:13:10 -- Begin auto take file -- File: SYS:SYSTEM.CMD[3,3] 22:13:10 -- Auto Take error on line 3 -- Unrecognized switch or keyword: "IGNORE" 22:13:10 -- End auto take file -- 22:13:10 -- [SYSTEM]CATALOG error -- Input error reading DSKB:STRLST.SYS[1,4]; End of file 22:13:10 -- [SYSTEM]CATALOG error -- Input error reading DSKB:MTALST.SYS[1,4]; End of file : "OPR" OPR>EXIT .SET TTY VT100 .systat Status of KS10 at 22:14:15 on 17-Dec-100 Uptime 1:14, 106% Null time = 106% Idle + 0% Lost, 2% Overhead 12 Jobs in use out of 35. 12 logged in, 10 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 0 01 3 [OPR] CTY SYSTAT 21+SPY RN 4 4 [OPR] DET41 ACTDAE 137+39 TO 1 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 RN 1 8 [OPR] DET41 PULSAR 65+39 HB 0 9 [OPR] DET41 CATLOG 30+39 HB 0 11 [OPR] DET41 ORION 90+39 RN 1 12 [OPR] DET41 BATCON 38+39 SL 0 13 [OPR] 41 4 ^C 0 14 [OPR] DET41 LPTSPL 59+39 HB 0 n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 610/2000 = 31% Virt. Core used = 646/2000 = 32% 129P Core left Active swapping ratio = 259/775 = .33 Virt. Core saved by sharing = 234/(234+646) = 27% Average job size =599/12 = 49.9P+281/12 = 23.4P Total=880/12 = 73.3P No busy devices System File Structures: Name Free Mount DSKB 184610 2 Total Free 184610 .r opr OPR> ? Invalid OPR command specified: "opr" OPR>exit .dir act: USAGE OUT 10 <177> 17-Dec-00 DSKB: [1,7] FAILUR LOG 0 <177> 17-Dec-00 USEJOB BIN 36 <057> 17-Dec-00 Total of 46 blocks in 3 files on DSKB: [1,7] .dir sys:*.sys SAT SYS 89 <355> 17-Dec-00 DSKB: [1,4] HOME SYS 29 <355> 17-Dec-00 SWAP SYS 8009 <357> 17-Dec-00 BADBLK SYS 9 <357> 17-Dec-00 CRASH SYS 1 <055> 17-Dec-00 ACTDAE SYS 36 <157> 5-Mar-86 DTALST SYS 0 <055> 1-May-85 MTALST SYS 0 <055> 1-May-85 PROJCT SYS 4 <055> 1-May-85 STRLST SYS 0 <055> 1-May-85 AVAIL SYS 6 <155> 17-Dec-00 ERROR SYS 2 <155> 17-Dec-00 Total of 8185 blocks in 12 files on DSKB: [1,4] .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPSER [1,2] Logged-off CTY at 22:15:00 on 17-Dec0 Runtime: 0:00:05, KCS:70, Connect time: 0:01:59 Disk Reads:707, Writes:0, Blocks saved:0 .KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2703 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!wn4feed!wn1feed!worldnet.att.net!207.207.0.27!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: my first KLAD tests - passed. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 59 Message-ID: <_yS%5.110540$IP1.3695733@news1.giganews.com> NNTP-Posting-Date: Tue, 19 Dec 2000 17:54:34 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-bJe7BLyJwdIefpFUPpPW4OlEMTX0V7jGiWSHfEeEr9PGb00zxRqBJURHf6j1Ub4Qyq49sLizUY3deb8!l50xpy6cm00R3Q82IQukuQdTaoU= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 19 Dec 2000 23:54:34 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2703 Hello folks: I was trying to run simple SMBC2 to see if it worked or now. Yes, it worked so well! :-) First time, I tried to run it and got HALT message. It said "?HALT instruction at user PC 21503". I looked into SMBC2.SEQ file and found out that SOS instruction failed to set CRY0/CRY1 flags. I fixed my emulated SOS instruction in my emulator by changed from 'if (*ac0 == -1)' to 'if (*ac0 != -1)'. I recompiled it and tried it again. SMBC2 normally exit without any HALT messages. That means some basic instruction test had been passed. I will work on other stuffs to test more instructions... That was tested on TOPS-10 v7.03. -- Tim Stark Here is my screen log file: .dir SMBC2 CCL 1 <155> 22-Feb-89 DSKB: [6,10] SMBC2 CTL 2 <155> 22-Feb-89 SMBC2 EXE 136 <155> 22-Feb-89 (1) SMBC2 HST 1 <155> 22-Feb-89 SMBC2 INP 3 <155> 22-Feb-89 SMBC2 SAV 125 <155> 22-Feb-89 (1) SMBC2 SEQ 5160 <155> 22-Feb-89 SMBC2 TXT 11 <155> 22-Feb-89 SMBC2A MAC 58 <155> 22-Feb-89 SMBC2B MAC 88 <155> 22-Feb-89 SMBC2C MAC 135 <155> 22-Feb-89 SMBC2D MAC 103 <155> 22-Feb-89 SMBC2E MAC 142 <155> 22-Feb-89 SMBC2F MAC 175 <155> 22-Feb-89 SMBC2G MAC 66 <155> 22-Feb-89 SMBC2H MAC 48 <155> 22-Feb-89 SMBC2S MAC 1 <155> 22-Feb-89 SMBC2T MAC 19 <155> 22-Feb-89 Total of 6274 blocks in 18 files on DSKB: [6,10] .run smbc2 DECSYSTEM-2020 BOOT CHECK 2 DIAGNOSTIC [SMBC2] EXIT . KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2789 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.tele.dk!195.158.233.21!news1.ebone.net!news.ebone.net!news.net.uni-c.dk!newsfeed.sunet.se!news01.sunet.se!news-sto.telia.net!news.defero.net!junk.nocrew.org!not-for-mail From: Lars Brinkhoff Newsgroups: alt.sys.pdp10 Subject: Re: Good news! TOPS-20 v4.1 booted successfully Organization: nocrew Lines: 20 Sender: lars@junk.nocrew.org Message-ID: <858zp9d53g.fsf@junk.nocrew.org> References: <3A42327D.E1D1CFDD@bartek.dontspamme.net> <91te36$9kh$1@spies.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.7 Date: Thu, 21 Dec 2000 18:05:47 GMT NNTP-Posting-Host: 212.73.17.42 X-Complaints-To: abuse@defero.net X-Trace: news.defero.net 977421947 212.73.17.42 (Thu, 21 Dec 2000 19:05:47 MET) NNTP-Posting-Date: Thu, 21 Dec 2000 19:05:47 MET Xref: nntp1.ba.best.com alt.sys.pdp10:2789 aek@spies.com (Al Kossow) writes: > I guess none of the PARC-oids hang out here. Other than the IEEE > Computer article on it, and having seen the clock module from MAXC 1 > in my boss's office, I know very little about the machines. [From "Origins and Development of TOPS-20" by Dan Murphy] Stanford wasn't the only other place where a PDP-10 architecture machine was being designed and built. The newly opened Xerox PARC (Palo Alto Research Center) included a number of people from BBN and other sites familiar with TENEX, and they were quite interested in pursuing their work on that system. Since Xerox was then in the computer business, PARC couldn't quite manage to just buy a PDP-10 from DEC. Consequently, they built their own. It was a single-processor machine compatible with the KA10 and BBN paging and was designed and built in a relatively short time. It ran TENEX for a number of years. -- http://lars.nocrew.org/ Article 2701 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!feed2.news.rcn.net!rcn!nntp.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: TOPS-20: MX> prompt problems. Newsgroups: alt.sys.pdp10 References: <2Sf%5.72464$%j3.886637@news6.giganews.com> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 109 Message-ID: NNTP-Posting-Date: Tue, 19 Dec 2000 17:52:16 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-IVjXo5je+0Y9VhprrX1VQV9dqjPqbSaOJLRfbwJnc2H5C1YVf8fvGPhA2BYLzWnVR6AcdFtIJ8Uq7so!FUB/k+mcz/9J9VcDlaLeXIbTd3Ip X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 19 Dec 2000 23:52:16 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2701 Mark Crispin wrote: > Try doing "R" (RESET) then repeat with "GMTA0:". Mark: Good news! I finally got a @ prompt much like a dot on TOPS-10 operating system. Well, I found another bug in my emulator that cause executation derailment during PXCT instruction's presence and page fault traps. I fixed in page fault trap handler in my emulator by adding 'pager_PC' to save PC before execute instructions because PC might be incremented sometimes. If so, it cause executable derailment by missing a re-trying instruction when page fault trap occurs. However, I ended up another bug that I kept repeating same BUGCHK messages 4 to 5 times each second. I fixed it in prority interrupt routines. Since I already implemented CTY input delay (100 instruction countdown), no longer lagged etc... After I removed a patch that re-send PI request for channel 1, I no longer received annoying repeating BUGCHK messages that complaint about unexecpted interrupt on channel 1. Whew. Also, another BUGINF messages showed up about MTA0: problems. I will work on it right now. -- Tim Stark Here is my screen log file: KS10> boot mta0: Booting MTA0... Now running... MTBOOT>/l MTBOOT>/g143 [FOR ADDITIONAL INFORMATION TYPE "?" TO ANY OF THE FOLLOWING QUESTIONS.] DO YOU WANT TO REPLACE THE FILE SYSTEM ON THE PUBLIC STRUCTURE? Y DO YOU WANT TO DEFINE THE PUBLIC STRUCTURE? Y HOW MANY PACKS ARE IN THIS STRUCTURE: 1 ON WHICH "CHANNEL,UNIT" IS LOGICAL PACK # 0 MOUNTED: 0,0 DO YOU WANT THE DEFAULT SWAPPING SPACE? Y DO YOU WANT THE DEFAULT SIZE FRONT END FILE SYSTEM? Y DO YOU WANT THE DEFAULT SIZE BOOTSTRAP AREA? Y [STRUCTURE "PS" SUCCESSFULLY DEFINED] [PS MOUNTED] %%NO SETSPD System restarting, wait... ENTER CURRENT DATE AND TIME: 19-DEC-2000 18:20 YOU HAVE ENTERED TUESDAY, 19-DECEMBER-2000 6:20PM, IS THIS CORRECT (Y,N) Y WHY RELOAD? NEW ACCOUNTS-TABLE.BIN NOT FOUND - ACCOUNT VALIDATION IS DISABLED RUNNING DDMP NO SYSJOB ^C NO EXEC MX>GET FILE MTA0: INTERRUPT AT 0 MX>RESET MX>GET FILE MTA0: ******************** *BUGINF "OVRDTA" at 19-Dec-2000 18:20:41 *PHYSIO - OVERDUE TRANSFER ABORTED *Additional data: 1, 0, 0 ******************** ******************** *BUGINF "OVRDTA" at 19-Dec-2000 18:20:59 *PHYSIO - OVERDUE TRANSFER ABORTED *Additional data: 1, 0, 0 ******************** MX>START TOPS-20 Command processor 5.1(1354) @DIRECTORY ^C @HELP ^C @ KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2670 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail From: "Alan H. Martin" Newsgroups: alt.sys.pdp10 Subject: Re: BLT 17,17 bug had been fixed but.. Date: Tue, 19 Dec 2000 00:48:57 -0500 Lines: 35 Message-ID: <3A3EF6C9.914EE31B@MA.UltraNet.Com> References: <6TiY5.46657$%j3.530564@news6.giganews.com> <90teal$10b$1@bob.news.rcn.net> <9105q9$8d3$1@bob.news.rcn.net> <9154bi$8am$1@bob.news.rcn.net> <917jrq$2s6$2@bob.news.rcn.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: UmFuZG9tSVZamziBEQXPxXoRQTsxLtpekypnXuGvd7T9i8ROiL/FHnmrx/4jp9hi X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 19 Dec 2000 05:49:36 GMT X-Mailer: Mozilla 4.75 [en] (Win95; U) X-Accept-Language: en,en-US,en-GB,es Xref: nntp1.ba.best.com alt.sys.pdp10:2670 jmfbahciv@aol.com wrote: > > In article , > Timothy Stark wrote: ... > >Well, I found where did MOVEKEY come from..... RMS-10 interface! > >I used DDT to narrowed down to OPNCOM routine in ACTRMS.MAC (now > >ACTLIB.MAC). $CREATE 0(F) called to RMS$10+12 to execute Bliss-10 > >routines to open RMS-10 based keyed file. ... > > Ohmygawd!!!! What did those people do? ... " ;112) Rip out the ACCTIO interface and replace with ACTRMS. /TL " In particular, as Timothy alluded, SYS:ACTDAE.SYS is an RMS ISAM file. > Is there an RMS person lurking around here? Is that MOVEKEY > actually part of the EXE; so that whatever is getting picked > up via a LDB and then BLT is off by some number of words? > > The only thing that makes sense is that an indirection is > screwing up. The last I read, the simulator doesn't implement EXTEND instructions. Maybe a MOVSLJ took a left at Albuquerque. However, MOVEKEY is the name of an RMS routine; I just don't know what file it's defined in (too many RMS files to guess easily). How that string appears though, I just don't know. /AHM -- Alan Howard Martin AMartin@MA.UltraNet.Com Article 2686 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.44!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-95 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: BLT 17,17 bug had been fixed but.. Date: Tue, 19 Dec 00 13:06:31 GMT Organization: UltraNet Communications, Inc. Lines: 28 Message-ID: <91nqev$43p$3@autumn.news.rcn.net> References: <6TiY5.46657$%j3.530564@news6.giganews.com> <90teal$10b$1@bob.news.rcn.net> <9105q9$8d3$1@bob.news.rcn.net> <9154bi$8am$1@bob.news.rcn.net> <3A3EF415.A440E5C0@MA.UltraNet.Com> X-Trace: UmFuZG9tSVa7Yfkh80+s0nYtT+TbTO7c9F0Z9OYmcoDlddjAZJyvC0Qui5dgxtvA X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 19 Dec 2000 14:13:51 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:2686 In article <3A3EF415.A440E5C0@MA.UltraNet.Com>, "Alan H. Martin" wrote: >Timothy Stark wrote: >> >> Well, I found where did MOVEKEY come from..... RMS-10 interface! >> I used DDT to narrowed down to OPNCOM routine in ACTRMS.MAC (now >> ACTLIB.MAC). $CREATE 0(F) called to RMS$10+12 to execute Bliss-10 >> routines to open RMS-10 based keyed file. RMS-10 interface was written >> in BLISS-10 language. > >The RMS$10 dispatch table is in the MACRO-10 file: >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSDSP. MAC >but the code executed for $CREATE is in the BLISS-36 file: >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSOPN. B36 And I assure you that RMS did not use the BLISS that was corporately designated as BLISS-36. We were not allowed to ship that EXE except on the DECnet tapes so we could soak customers for big bucks. RMS would have not got on the CUSP tape with my also shipping its building tools. /BAH Subtract a hundred and four for e-mail. Article 2709 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.44!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail From: "Alan H. Martin" Newsgroups: alt.sys.pdp10 Subject: Re: BLT 17,17 bug had been fixed but.. Date: Tue, 19 Dec 2000 21:02:44 -0500 Lines: 27 Message-ID: <3A401344.F50169B6@MA.UltraNet.Com> References: <6TiY5.46657$%j3.530564@news6.giganews.com> <90teal$10b$1@bob.news.rcn.net> <9105q9$8d3$1@bob.news.rcn.net> <9154bi$8am$1@bob.news.rcn.net> <3A3EF415.A440E5C0@MA.UltraNet.Com> <91nqev$43p$3@autumn.news.rcn.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: UmFuZG9tSVaVBh30bq8UEo4I11an8R7wI+9lzZ7IsmVLvbw5GcBEc/ofvvzQgKwz X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 20 Dec 2000 02:03:25 GMT X-Mailer: Mozilla 4.75 [en] (Win95; U) X-Accept-Language: en,en-US,en-GB,es Xref: nntp1.ba.best.com alt.sys.pdp10:2709 jmfbahciv@aol.com wrote: > > In article <3A3EF415.A440E5C0@MA.UltraNet.Com>, > "Alan H. Martin" wrote: > > > >The RMS$10 dispatch table is in the MACRO-10 file: > >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSDSP.MAC > >but the code executed for $CREATE is in the BLISS-36 file: > >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSOPN.B36 > > And I assure you that RMS did not use the BLISS that was > corporately designated as BLISS-36. We were not allowed to > ship that EXE except on the DECnet tapes so we could soak > customers for big bucks. > > RMS would have not got on the CUSP tape with my also shipping > its building tools. Despite RMSIDX having a BLISS-36 extension, containg BLISS-36 code, and being compiled in: http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMS10.CTL by invoking the BLISS-36 compiler... /AHM -- Alan Howard Martin AMartin@MA.UltraNet.Com Article 2721 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-122 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: BLT 17,17 bug had been fixed but.. Date: Wed, 20 Dec 00 10:23:27 GMT Organization: UltraNet Communications, Inc. Lines: 58 Message-ID: <91q59j$9h2$2@bob.news.rcn.net> References: <6TiY5.46657$%j3.530564@news6.giganews.com> <90teal$10b$1@bob.news.rcn.net> <9105q9$8d3$1@bob.news.rcn.net> <9154bi$8am$1@bob.news.rcn.net> <3A3EF415.A440E5C0@MA.UltraNet.Com> <91nqev$43p$3@autumn.news.rcn.net> <3A401344.F50169B6@MA.UltraNet.Com> X-Trace: UmFuZG9tSVbXu4Sz4TCDwaYcIpiImQfBNPyoz/Z7teRnDr5HmuWULqwSvt5QgPZw X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 20 Dec 2000 11:30:59 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:2721 In article <3A401344.F50169B6@MA.UltraNet.Com>, "Alan H. Martin" wrote: >jmfbahciv@aol.com wrote: >> >> In article <3A3EF415.A440E5C0@MA.UltraNet.Com>, >> "Alan H. Martin" wrote: >> > >> >The RMS$10 dispatch table is in the MACRO-10 file: >> >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSDSP. MAC >> >but the code executed for $CREATE is in the BLISS-36 file: >> >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMSOPN. B36 >> >> And I assure you that RMS did not use the BLISS that was >> corporately designated as BLISS-36. We were not allowed to >> ship that EXE except on the DECnet tapes so we could soak >> customers for big bucks. >> >> RMS would have not got on the CUSP tape with my also shipping >> its building tools. > >Despite RMSIDX having a BLISS-36 extension, containg BLISS-36 code, >and being compiled in: > >http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/RMS10.C TL > >by invoking the BLISS-36 compiler... And I'm telling you that RMS-10 was written when there wasn't a BLISS-36 mess around. We did not get BLISS-36 until we started to fool with DECnet. That was way after RMS-10 got shipped. If there was such a file yakking about BLISS-36, I have no idea who did it because there wasn't any developement done on RMS...only bug fixes. Period. The only thing I can think of is that somebody messed with RMS when the ACTDAE was done, and that's even worse because, not only did BLISS-10 and BLISS-36 generate differing code, but interim edit updates of BLISS-36 would generate different code. Mixing these all up is going to cause a headache so big that all work will stop. The only way we (TOPS-10) monitor managed to get DECnet Phase IV out was by doing our own freeze of the crap and ignoring the north. I may be hazy on programming the -10, and hazy on debugging but I am an expert on building and packaging the software that was shipped out of the TOPS-10 monitor group. The language group beat to a completely different drummer. And the TOPS-20 monitor group has a mess until they had a software engineer working full time on packaging. /BAH Subtract a hundred and four for e-mail. Article 2616 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!newsfeed.skycache.com!Cidera!xfer10.netnews.com!netnews.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-16 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Need for KS10 diagnostics Date: Sun, 17 Dec 00 14:14:05 GMT Organization: UltraNet Communications, Inc. Lines: 86 Message-ID: <91ill4$r7g$1@bob.news.rcn.net> References: <3A3A338C.8DE7FFA5@bartek.net> <91fpoi$li9$1@bob.news.rcn.net> <3A3B94DE.4FE198AA@bartek.net> X-Trace: UmFuZG9tSVZ40QKh2O5to4QURWHoZL+8J4sf9mG+rJ6WJenRa/F+3FTGxwGdSbS7 X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 17 Dec 2000 15:21:08 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:2616 In article <3A3B94DE.4FE198AA@bartek.net>, Arthur Krewat wrote: >jmfbahciv@aol.com wrote: >> >> I didn't understand the dependency either. If it's because >> the tape was written with DUMPER, BACKUP can read that tape >> just fine if an /INTERCHAGE switch is given. He looses any >> directory information but that's not interesting anyway. > >I have a sneaking suspicion those diags on the "red" pack >are coded to use TOPS-20 monitor calls or depend on some other >TOPS-20'ish thingy... Hardware diagnostics? Unless that departement drastically changed their business plan, diags shouldn't be depenedent on us (monitor types). > ...otherwise, my site would have made it >possible to run them under TOPS-10. The boot code itself >is TOPS-20, isn't it? Or is it the same between -10 and -20? Which boot code? Not the microcode. Diags had to run on a cold machine. Or else field service would have had to do it the hard way and use scopes. Diags were also a maintenance tool for field service. >> > Integer and floating point math needs to be >> >checked - can someone come up with some extensive >> >tests of this? >> >> The MACRO-10 code can be written in ASCII. How do you >> build the damn thing without a -10? I've been gnawing >> on this problem for weeks now. > >Since Tim has been able to compile TECO in his emulator, and >it came up with the same code as the original .EXE I see one >possibility: Go into TECO in his emulator, and cut-and-paste >the MACRO code into it and compile it under his emulator. You are assuming that the checksum code also worked :-). All we proved was consistency. Hint: The restore could have equivalent problems; all we proved with that checksum is that, if there are problems with generating code and writing the disk, they're all behaving alike. And we've only half exercised MACRO.EXE, let alone the monitor functionality. > >If any bugs arise, use DDT to make sure the generated code >is correct, then figure out why it coughed... The problem is that he doesn't have a really good idea how to do and look at a BLT from the PDP-10 side. He knows how to look at it from his emulator side. Now, we know how to look at it from the -10 side but we need a -10 to do the test . We are in a great CTCH22. I am now dreaming teaching techniques to Tim. So I'm getting into the swing...now if I could only remember without a gazillion parity errors.... /BAH > >> Speaking of playing, I am going to have to have a Unix >> system to play, aren't I? This is a really really >> underhanded way to get me transferred to Unix, guys. > >I am perfectly willing to put up a Linux box on the Internet, >where you can run this thing. It may be only a 400Mhz Celeron, >but it would certainly beat the original KS10 for speed! > >better yet would be to get it running under Solaris X86... > >> And what kind of a hardware system should I buy? For various >> reasons I am going to need that idiotic port that trackballs >> have converted to (serial ports are apparently passe'). > >USB (Universal Serial Bus) ??? What a waste of engineering talent >(or lack of).... why didn't they just use 10/100Base-T? It's serial >too! > >art k. Subtract a hundred and four for e-mail. Article 2736 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!wn4feed!wn2feed!worldnet.att.net!207.207.0.27!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Need for KS10 diagnostics Newsgroups: alt.sys.pdp10 References: <3A3A338C.8DE7FFA5@bartek.net> <91fpoi$li9$1@bob.news.rcn.net> <877l4xogc6.fsf@cartman.azz.net> <3A3F9F11.48B8D682@trailing-edge.com> <91qack$jmn$10@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 20 Message-ID: <%h306.114404$IP1.3801183@news1.giganews.com> NNTP-Posting-Date: Wed, 20 Dec 2000 08:23:55 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-oNIXjfxjMQCoqVbQY/iocW0LPas5yjRxw8ke3vydtHQxHmjmvgwczSfJH7/QMheo7XyzeXGBFPG1Up/!I/d9HBCeiDUQs2kiR8vEPc33FhI= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 20 Dec 2000 14:23:55 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2736 jmfbahciv@aol.com wrote: > Tim, Daniel is confused. Yeah. He is. I know inner workings of bootable tape so well. There are 512-byte blocks with a few tape marks before BACKUP-based blocks. First blocks are for FE loader. I used my own routines in my emulator to load RDI loader into my emulator's memory then transfer its control to the loader to load rest of stuffs to complete transfer. That's why I hold a B.A. degree in computer science from my university (Gallaudet University). -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2604 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nntp-out.monmouth.com!newspeer.monmouth.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news6.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Looking for MIG 7.03 Newsgroups: alt.sys.pdp10 References: <91dar9$n8j$8@bob.news.rcn.net> <91fr3c$hp$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 44 Message-ID: NNTP-Posting-Date: Sat, 16 Dec 2000 17:25:15 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-uZe+dBoNldSyW4LhI/rYxk7UDDnA2xfusQeCtypArXZKzs3TnX4VqZU4YgeRroz9HXtKGGZ2m6glRZL!B2MTwhbfj/G2wl7shdxLNSZfrow= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Sat, 16 Dec 2000 23:25:15 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:2604 jmfbahciv@aol.com wrote: > The 7.03 MIG would have said: > R BACKUP > TAPE MTA0: > INTERCHANGE > RESTORE SYS:=*.EXE > EXIT > You don't have to start from scratch. Just mount that tape > and do the above restore and you'll get the result you need. > Likewise, for every copy command in the 7.04 MIG, you will > need a similar commmand. > For instance, > RESTORE REL:=*.REL Barb: I successfully completed TOPS-10 v7.03 installation. I can do multi-restore once: RESTORE SYS:=*.EXE,*.VFU,*.RAM,TECO.ERR,HLP:=*.HLP,DOC:=*.DOC,..... It worked so well for me. I checked SYS:, HLP:, etc... They looks like so good! Then I made DSKB bootable by running SMFILE.EXE. When it is done. I rebooted system and was able boot DSKB: directly. After I entered "GO" on Startup prompt, I ended up "Lookup error 2" then a dot. I tried entered "LOGIN", etc.. I kept told same message - Lookup error 2. What is that for? Yes, I made a backup copy first before I tried. Thank you! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 2864 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.slurp.net!bony.umtec.com!root Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: E10 "Just Communication" release notice Message-ID: Approved: Why bother? MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 17 Date: Wed, 27 Dec 2000 12:51:12 -0600 NNTP-Posting-Host: 198.199.189.6 X-Trace: newsfeed.slurp.net 977939185 198.199.189.6 (Wed, 27 Dec 2000 11:46:25 CDT) NNTP-Posting-Date: Wed, 27 Dec 2000 11:46:25 CDT Xref: nntp1.ba.best.com alt.sys.pdp10:2864 After a year, updated stuff. Doesn't make anything new work, just an update to give everyone else something to play with. Now, I'm off to go play with this klad stuff and see how far I get. Out of date details and the current source (etc etc) are at http://bony.umtec.com/pdp10.html, or ftp to bony.umtec.com, in /pub/kl10. "Confuse, annoy, and DEE-STROY!" -- Jet Wolf | "Nothing Happens." -- ADVENT "You'd be surprised what you can live through..." -- Anonymous "...A man can pass his family and his name down through his sons, but it's his honour that gets passed through his daughters. He can see the best and worst of life in his girls. A daughter is something far too precious, and he'll do anything to protect her." -- Reichsfuehrer Siegfried Koenig, _Matrose_Mond_, David Oliver Article 3108 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!EU.net!Norway.EU.net!not-for-mail From: mrr@foo.eunet.no (Morten Reistad) Newsgroups: alt.sys.pdp10 Subject: Re: E10 is down for the moment... Date: 8 Jan 2001 22:11:14 GMT Organization: EUnet Lines: 67 Message-ID: <93ddu2$816$1@oslo-nntp.eunet.no> References: <3A58E08E.2E94298@digiweb.com> NNTP-Posting-Host: www.reistad.priv.no X-Trace: oslo-nntp.eunet.no 978991874 8230 193.71.26.162 (8 Jan 2001 22:11:14 GMT) X-Complaints-To: newsmaster@eunet.no NNTP-Posting-Date: 8 Jan 2001 22:11:14 GMT X-Newsreader: trn 4.0-test69 (20 September 1998) Xref: nntp1.ba.best.com alt.sys.pdp10:3108 In article <3A58E08E.2E94298@digiweb.com>, Hans B Pufal wrote: >Daniel Seagraves wrote: > >Sorry to hear of your troubles, > >> As far as E10 is concerned, I snagged the KLAD-reading stuff posted here >> (but I had to have it mailed to me, I lost the original post! Can whoever >> wrote this please drop me a line so I can credit you?) and put it in, read >> in DFKAA, and it barfed. Math wants rewritten (again). Sigh. It's >> blowing on the test instructions. So get I get to argue with that >> again... > >It was Morton Reistad who posted the original seqparse code. I'll be happy to repost it, but if any of you have done modifications I should integrate those first. Otherwise it seems we are duplicating work here. The E10 can get through DAKAA almost right off the last distribution. I also got down to DAKAB, found two instruction blopers, and some bit alignment problems I could work around with an OR during memory storage. Midway through DAKAC there are some logic/alu bit shift/tests that bring out large problems with the whole math implementation. So the status seems to be : == Instruction semantic correctness seems almost there. == Integer/ALU and flag handling needs an overhaul. Are there any 10s left where we could test exact flag behaviour with DDT? == Floating point also seems to work, but I suspect some bombs there as well; and we can't do those tests before DAKAC stops bombing. We could use the host hardware for FP; and just do translations of the FP format on load/store if this becomes a problem. > >My modoifed vrsion of your emulator gets a long way down DKFAA but halts >trying to do >IO on address 177 - is that used for anything? > >I can also pass DKFAB! > >I am doing the port to Linux now and should do a preliminary release in >the next day or two. Ports? I run the emulator on FreeBSD 4.0-STABLE, 2.7-STABLE, RedHat 6.2 and Mandrake 7.0 with no problems at all. CPUs vary from 486 to PIII@800Mhz. I just wrote 'make' on all of them. Yes, I do have some portfolio of machines at home and at work. > >Regards > > -- Hans B Pufal Article 3174 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news-hog.berkeley.edu!ucberkeley!nntp.cs.ubc.ca!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Barb, "MOVEKEY" finally disappeared in ACTDAE!! Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 247 Message-ID: NNTP-Posting-Date: Thu, 11 Jan 2001 21:23:16 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-U4j0XAH0gSbtBrrg2yxj1XvK3hpVIeuh6ZyUOHIefFf9IH3xgarWjxFqREyeTi3RP/oK2z8GWNV2X63!gmBW708hon0XrVK5sH/BhNQaZzIq X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 12 Jan 2001 03:23:18 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3174 Hello folks: Good news! I finally got rid of "MOVEKEY" in ACTDAE daemon after I fixed CMPS string instruction. When I reviewed my Opcode_EXTEND routine, I was shocked to notice that 'DO_JUMP(PC)' at case EXT_SKIP within switch statement! I changed to 'DO_SKIP'. I recompiled it and noticed big difference in ACTDAE's behavior! Yes, "MOVEKEY" finally disappeared from my screen!! Also, I got stopcode AFF message - Accounting Initialization Failure. Then I fixed some bugs in CMPS series instruction. ACTDAE finally runs smoothly without any problems. Also look at GALAXY daemons no more crashes anymore after I fixed PUSH*/POP* instruction. They ended up a big mess during page fault traps. Yeah. Thank to KLAD sources package because I was able to fix many bugs by using KLAD. My emulator is becoming reliable more and more... However, I have some problems with REACT and LOGIN. When I attempted to log into system as [1,2], I was told that invalid login entry. .login 1,2 Job 2 KS10 CTY ?LGNIET Invalid entry # Does anyone know solution about that by fixing that? Also, I was unable to access OPR> prompt because of ACTDAE or so. Thank you! -- Tim Stark Here is my screen log: (TOPS-10 7.03) KS10> boot rpa0: Booting RPA0... [Reading first HOM block...] [Reading FE-FILE Page 0 at C 216 T 9 S 11] [Reading Monitor Pre-boot at C 216 T 12 S 3] Now running... BOOT V3(47) BOOT> [Loading from DSKB:SYSTEM.EXE[1,4]] KS10 03-19-86 Why reload: new Date: 11-jan-2001 Time: 21:38 Startup option: go [Rebuilding the system search list from the HOMe blocks] [Rebuilding the active swapping list from the HOMe blocks] [Rebuilding the system dump list from the HOMe blocks] KS10 21:38:01 CTY system 5000 Connected to Node (0) Line # 42 .LOGIN 1,2 .R OPSER [OPRPAF Processing auto command file] ?OPRALF LOOKUP failure 0 * [CCPWFD Waiting for file daemon to start] *%%TTY STOMPER - Starting ^C .SET TTY VT100 .systat Status of KS10 at 21:38:18 on 11-Jan-101 Uptime 17, 85% Null time = 84% Idle + 1% Lost, 1% Overhead 13 Jobs in use out of 35. 12 logged in, 11 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 0 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 21+SPY RN 0 4 [OPR] DET41 ACTDAE 180+39 SL 1 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 0 9 [OPR] DET41 CATLOG 30+39 HB 0 10 [OPR] DET41 ORION 94+39 SL 1 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] 41 4 ^C 0 16 [OPR] DET41 LPTSPL 59+39 HB 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 677/2000 = 34% Virt. Core used = 716/2000 = 36% 59P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+716) = 25% Average job size =654/13 = 50.3P+296/13 = 22.7P Total=950/13 = 73.0P No busy devices System File Structures: Name Free Mount DSKB 178170 3 Total Free 178170 .r react REACT> %REACME Command error; Unrecognized switch or keyword: "react" REACT>help ^C .declare react=sys:react .react REACT>help ^C .declare opr=sys:opr .opr OPR> 21:38:09 -- Message from the Accounting System -- Account validation is not required 21:38:10 -- Begin auto take file -- File: SYS:SYSTEM.CMD[3,3] 21:38:10 -- Auto Take error on line 3 -- Unrecognized switch or keyword: "IGNORE" 21:38:10 -- End auto take file -- 21:38:10 -- [SYSTEM]CATALOG error -- Input error reading DSKB:STRLST.SYS[1,4]; End of file 21:38:10 -- [SYSTEM]CATALOG error -- Input error reading DSKB:MTALST.SYS[1,4]; End of file 21:38:10 -- [SYSTEM]CATALOG error -- Input error reading DSKB:DTALST.SYS[1,4]; End of file 21:38:11 Printer 0 -- Not available right now -- OPR> OPR> OPR>help ^C .^ .dir dskb:[1,7] USAGE OUT 33 <177> 11-Jan-01 DSKB: [1,7] FAILUR LOG 0 <177> 17-Dec-00 USEJOB BIN 36 <057> 11-Jan-01 SYSDEV BIN 1 <057> 11-Jan-01 Total of 70 blocks in 4 files on DSKB: [1,7] .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPSER [1,2] Logged-off CTY at 21:40:12 on 11-Jan-1 Runtime: 0:00:05, KCS:120, Connect time: 0:02:10 Disk Reads:1356, Writes:1, Blocks saved:0 .systat Status of KS10 at 21:40:15 on 11-Jan-101 Uptime 2:14, 111% Null time = 111% Idle + 0% Lost, 2% Overhead 12 Jobs in use out of 35. 11 logged in, 10 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 0 01 2 2,5 CTY SYSTAT 21+SPY RN 0 $ 4 [OPR] DET41 ACTDAE 180+39 SL 1 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 0 9 [OPR] DET41 CATLOG 30+39 HB 0 10 [OPR] DET41 ORION 91+39 SL 1 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] 41 4 ^C SW 0 16 [OPR] DET41 LPTSPL 59+39 HB 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 654/2000 = 33% Virt. Core used = 690/2000 = 35% 85P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+690) = 25% Average job size =643/12 = 53.5P+281/12 = 23.4P Total=924/12 = 77.0P No busy devices System File Structures: Name Free Mount DSKB 178170 1 Total Free 178170 .KJOB .login 1,2 Job 2 KS10 CTY ?LGNIET Invalid entry #3,5 ?LGNIET Invalid entry #103011,4207 ?LGNIET Invalid entry #^C .KJOB .^E KS10> quit Exiting PDP10 Emulator -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3172 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!wn4feed!worldnet.att.net!207.207.0.27!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: KLAD sources package - successful Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 133 Message-ID: NNTP-Posting-Date: Thu, 11 Jan 2001 21:25:27 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-TpJe3NuXBFoZ2BtS73edb9i4QXnUMwBt+j6sUVOszjeHQ0aqUqe48Y/u/Qlr3DAFThYDrxF4s9YQgag!q0KSoQPikV7b4H8iv3IHMJBaXDu6 X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 12 Jan 2001 03:25:27 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3172 Hello folks: Yeah. KLAD sources package are so useful for debugging the PDP-10 emulator. Yes, I still am working on my emulated EXTEND instructions like CMPSx, CVTDBx, CVTBDx, EDIT, and MOVSx instructions. Thank you! -- Tim Stark My screen log file is: Unit mta0 had been attached with '../ksdiag2.tap' file. KS10> boot mta0: Booting MTA0... Now running... * SMMAG [DSQDE] - DECSYSTEM 2020 DIAGNOSTIC MAGTAPE MONITOR - VER 0.3 * SMMAG CMD - dskcg DECSYSTEM 2020 ADVANCED INSTRUCTION DIAGNOSTIC (7) [DSKCG] VERSION 0.2, SV=0.3, CPU#=5000, MCV=130, MCO=610, HO=0, KASW=000000 000000 TTY SWITCH CONTROL ? - 0,S OR Y - 0 SWITCHES = 000000 000000 PC= 031175 SWITCHES = 000000 000000 ERROR IN COMPARE STRING INSTRUCTION TEST INSTRUCTION: CMPSL CASE # 4 SOURCE STRING: 1 DESTIN STRING: 4400036020 E0+1: 0 E0+2: 0 AC BLOCKS: ---- FINAL ---- ORIGINAL CORRECT ACTUAL 000000 000001 000000 000000 000000 000000 440500 035774 370500 035774 370500 035774 000000 000000 000000 000000 000000 000000 000000 000001 000000 000000 000000 000000 444400 036020 004400 036020 004400 036020 INSTRUCTION MADE AN INCORRECT N-SKIP RETURN PC= 034740 SWITCHES = 000000 000000 ERROR IN MUUO OPERATION - CORRECT=MUUO ACTUAL=MUUO PC CORRECT: 123400 035564 ACTUAL: 700000 031416 DISCREP: 623400 004172 UNEXPECTED MUUO OCCURED ? FATAL PROGRAM ERROR AT 034741 EDDT SMMAG CMD - My smtape.c source code to create KS diagnostic tape from a scratch to test my emulator: #include #include #include #include #include // To make a listing of files to being appended to a tape file, // put them in smtape.inp. // To compile: gcc smtape.c -o smtape FILE *inpFile; // Input file int inFile; // Input data file int outFile; // Output data file char cBuffer[512*5]; // 512-words tape block int main(int argc, char **argv) { char filename[80]; int cntBlocks; int lenBlock; int tm = 0; int recSize = 512*5; inpFile = fopen("smtape.inp", "r"); outFile = open("ksdiag.tap", O_WRONLY|O_CREAT, 0700); while (!feof(inpFile)) { if (fscanf(inpFile, "%s", filename) != EOF) { printf("File: %s - ", filename); cntBlocks = 0; if ((inFile = open(filename, O_RDONLY)) < 0) printf("Error: %s\n", strerror(errno)); else { while ((lenBlock = read(inFile, cBuffer, 512*5)) > 0) { cntBlocks++; write(outFile, &recSize, sizeof(int)); write(outFile, cBuffer, 512*5); write(outFile, &recSize, sizeof(int)); } write(outFile, &tm, sizeof(int)); printf("%d blocks - Ok\n", cntBlocks); close(inFile); } } } write(outFile, &tm, sizeof(int)); close (outFile); fclose (inpFile); } -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3233 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!hermes.visi.com!news-out.visi.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: A problem on OPR and REACT programs.. Newsgroups: alt.sys.pdp10 References: <6ZC96.20807$lV5.338311@news2.giganews.com> <949hin$c51$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 353 Message-ID: <1k0a6.248281$IP1.8278091@news1.giganews.com> NNTP-Posting-Date: Fri, 19 Jan 2001 13:11:57 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-DKg9rxPvKq1thujlgOYyXCEw/peeuNvVQCSfedUVrxGKuLt8ULxUMsIo+wGpCov3d1CG+rNLGKh6e2E!LjIL7jH1KhOSz3hNu8abtmrKgT6n X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Fri, 19 Jan 2001 19:11:58 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3233 Barb, Yes, you are right about terminal I/O routines. I just tested how terminal I/O routines behave by using OPR and REACT programs. I discovered that that problems are clearly reproductive. I played my keyboard by pressing every key. I found out that ^X, SPACE, and @ characters are "break signal" to wake them up to process my commands. I believe that terminal I/O routines misinterpret SPACE and @ characters as break signal instead of ^Z and CR characters. When I pressed CR keys three times then SPACE key, I got three OPR> prompts printed. When I tried four CR keys then SPACE key, I got four OPR> prompts too. Also, I tried four ^Z keys then ^X key at text input prompt, I got OPR>, OPR exit, and a two dot prompts. Then "? ^X" message. I looked into OPR source code and traced to K%TXTI routine in GLXKBD.MAC source in GLXLIB directory. It uses ECHO and SETMSK routines. I set watch in IDIVI and LSH instruction in my PDP-10 emulator but... My debug log file showed that K%TXTI correctly interpret ^Z key as break signal, etc. I can't find any problem with them! :-( I tried DDT but I was told that ECHO, etc are unknown symbols so that I can't debug it without symbols. Well, however, I noticed that long execution after each SPACE key pressed but it does not show SPACE key as break signal mask during ECHO or so routine. I believe that terminal I/O buffer is not in sync with OPR's input buffer. That bothers me so much. No, I do not use the same buffer for both input and output. I use separated stream buffer for input and output each. -- Tim Stark That is my another screen log: KS10 12:58:01 CTY system 5000 Connected to Node (0) Line # 42 .LOGIN 1,2 .R OPSER [OPRPAF Processing auto command file] ?OPRALF LOOKUP failure 0 * [CCPWFD Waiting for file daemon to start] ^C .SET TTY VT100 .%%TTY STOMPER - Starting .systat Status of KS10 at 12:58:15 on 19-Jan-101 Uptime 14, 43% Null time = 43% Idle + 0% Lost, 4% Overhead 13 Jobs in use out of 35. 12 logged in, 11 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 1 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 21+SPY RN 1 4 [OPR] DET41 ACTDAE 180+39 SL 2 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 0 9 [OPR] DET41 CATLOG 30+39 HB 1 10 [OPR] DET41 ORION 99+39 SL 1 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] DET41 LPTSPL 59+39 HB 1 13 [OPR] 41 4 ^C 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 682/2000 = 34% Virt. Core used = 721/2000 = 36% 54P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+721) = 25% Average job size =659/13 = 50.6P+296/13 = 22.7P Total=955/13 = 73.4P No busy devices System File Structures: Name Free Mount DSKB 184460 3 Total Free 184460 .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPSER [1,2] Logged-off CTY at 12:58:19 on 19-Jan-1 Runtime: 0:00:02, KCS:23, Connect time: 0:00:17 Disk Reads:179, Writes:0, Blocks saved:0 .login operator Job 3 KS10 CTY [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 12:58 19-Jan-1 Friday .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPERATOR [1,2] Logged-off CTY at 12:58:28 on 19-Jan-1 Runtime: 0:00:00, KCS:10, Connect time: 0:00:06 Disk Reads:126, Writes:0, Blocks saved:0 .login Job 3 KS10 CTY #operator [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 12:58 19-Jan-1 Friday // Hmmm. With using usernames, login worked fine. PPN numbers did // not work and resulted "invalid entry" error messages. However, // LOGIN program did not update latest access in ACTDAE.SYS. .r react REACT> %REACME Command error; Unrecognized switch or keyword: "react" REACT>show PPN: [1,2] User name: OPERATOR Profile default: -none- Personal name: -none- Distribution location: -none- Mailing address: -none- Expiration date: never LOGIN times: Weekdays 0:23 Weekends 0:23 Access types: Local, Subjob of batch, Batch Requirements for LOGIN: Account and remark strings are not required Name is not required Password under timesharing and batch Password change not required Minimum password length: -none- Password change interval: -none- Password changes are prohibited Schedular type: 0 Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 511 IPCF quotas: Send 511, Receive 511, PIDs 511 Privileges: Disk priority: 3, HPQ: 15, System operator Administrative, ENQ-DEQ, IPCF, LOCK, METER, POKE, RTTRP, CPU, SPY-all-core, SPY-monitor, TRPSET, Unspooling Spooled device bits: CDP, CDR, LPT, PLT, PTP Watch bits: MTA Structure quotas: Structure Quota in Quota out Reserved Status --------- ----------- ---------- ---------- ---------- DSKA 100000 100000 0 DSKB 100000 100000 0 DSKC 100000 100000 0 Administrative data: Profile last changed by [1,2] at 5-Mar-86 1:59:09 Last access succeeded on 4-Mar-86 17:48:52 Last password change at 2-Dec-85 5:56:37 // Hmmm. Login did not update latest access successful... ^C .r opr OPR> ? Invalid OPR command specified: "opr" 12:58:12 -- Message from the Accounting System -- Account validation is not required 12:58:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:STRLST.SYS[1,4]; End of file 12:58:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:MTALST.SYS[1,4]; End of file 12:58:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:DTALST.SYS[1,4]; End of file 12:58:13 -- Begin auto take file -- File: SYS:SYSTEM.CMD[3,3] 12:58:13 -- Auto Take error on line 3 -- Unrecognized switch or keyword: "IGNORE" 12:58:13 -- End auto take file -- 12:58:14 Printer 0 -- Not available right now -- ? : "opr" OPR>showstatusdisk/all OPR>showstatustape/all 12:59:17 -- Disk Drive Status -- Drive Type Status AVR STR Volume Unit# ------ ---- ----------- --- ----- ------ ----- RPA0 RP06 Mounted Yes DSKB ONC134 0 RPA1 RP06 Free Yes OPR> Monitor KS10 System uptime :01:59 Current date/time 19-Jan-101 13:00:00 12:59:37 -- Tape Drive Status -- Drive Trk Status AVR Density ------ --- ----------- --- ------------- MTA000 9 Free No 800/1600 MTA001 9 Free No 800/1600 OPR>settapemta0:unavailable Enter text and terminate with ^Z Ok, fine., ^Z ^Z OPR> // I had pressed ^Z then ^X to get out of text input. For more information, // look for my remarks below. 13:00:33 Device MTA000 -- Unavailable for use -- OPR>showstatustapemta0: OPR> 13:01:22 -- Tape Drive Status -- Drive Trk Status AVR Density ------ --- ----------- --- ------------- MTA000 9 Unavailable No 800/1600 OPR>settapemta0:available OPR> 13:02:20 Device MTA000 -- Available for use -- OPR>settapemta0:unavailable Enter text and terminate with ^Z Ok Ok. ^Z ^Z ^Z ^Z OPR>^Z .. ?^X? // I just tested how terminal I/O routines behave. I found out that their // problems are reproductive. When I pressed ^Z three times then pressed // ^X. I played with my keyboard by pressing every ctrl keys. I discovered // that ^X acts like "^Z" key to exit a text input. Other ^Z's in // terminal still were in buffer to being processed. When I pressed // ^X, then previous ^Z's keys were processed. I believe that terminal // I/O routines misinterpret ^X as break signal instead of ^Z. At // this time, I have press ^Z then ^X to exit an text input until I // find and fix a bug in my emulator. . . .systat Status of KS10 at 13:03:30 on 19-Jan-101 Uptime 5:29, 112% Null time = 112% Idle + 0% Lost, 2% Overhead 12 Jobs in use out of 35. 12 logged in, 10 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 1 01 3 [OPR] CTY SYSTAT 21+SPY RN 10 4 [OPR] DET41 ACTDAE 180+39 SL 3 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 108+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 0 9 [OPR] DET41 CATLOG 30+39 HB 1 10 [OPR] DET41 ORION 100+39 SL 2 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] DET41 LPTSPL 59+39 HB 1 13 [OPR] 41 4 ^C SW 0 n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 662/2000 = 33% Virt. Core used = 698/2000 = 35% 77P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+698) = 25% Average job size =651/12 = 54.2P+281/12 = 23.4P Total=932/12 = 77.6P No busy devices System File Structures: Name Free Mount DSKB 184450 2 Total Free 184450 .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPERATOR [1,2] Logged-off CTY at 13:03:33 on 19-Jan-1 Runtime: 0:00:11, KCS:615, Connect time: 0:05:03 Disk Reads:583, Writes:0, Blocks saved:0 .login operator Job 2 KS10 CTY [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 13:03 19-Jan-1 Friday .kjob [LGTOUL Other users logged-in under [1,2]] Job 2 User OPERATOR [1,2] Logged-off CTY at 13:03:39 on 19-Jan-1 Runtime: 0:00:00, KCS:7, Connect time: 0:00:02 Disk Reads:1, Writes:0, Blocks saved:0 . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3250 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!wn4feed!worldnet.att.net!207.207.0.27!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: A problem on OPR and REACT programs.. Newsgroups: alt.sys.pdp10 References: <6ZC96.20807$lV5.338311@news2.giganews.com> <949hin$c51$1@bob.news.rcn.net> <1k0a6.248281$IP1.8278091@news1.giganews.com> <94c920$25n$1@bob.news.rcn.net> <94eoqp$bk7$2@bob.news.rcn.net> <3A6AF7E3.BE9C3248@bartek.dontspamme.net> <94h48i$15e$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 476 Message-ID: NNTP-Posting-Date: Mon, 22 Jan 2001 14:45:42 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-p6N8EMG9HzzpVaW1QDf9LXGnsdnNxkwQScolRU7cUuZgSzdM7vQ6k7Dv8c8NJqdMnGo6xRKEJnl6KCm!ZlImg/xIoPBkbuT4/alOnrtUSM5p X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Mon, 22 Jan 2001 20:45:42 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3250 jmfbahciv@aol.com wrote: > Sounds like our bug. I've been waiting for you to trip over > one of those :-) but I was fearful that it would be a monitor > bug. Now, the question is--how in the hell did that one > get out? Try entering the PPN as 1/2 instead of 1,2 and > see if that one works. I don't know anybody who used a comma. Barb, I used my emulator to produce results from EXTEND instructions. Also, I found a possible bug near or around EXTEND instruction. AC #1 had a bad byte pointer that produced a corrupted string (for username field). I still am looking for a specific bug near MOVSLJ instruction or so. Also, I just tested REACT by created a few accounts. They were created successfully but they had corrupted username so that LOGIN can't find accounts in ACTDAE.SYS execpt PPN 1,2. About log files, I added spaces between words on REACT command because it did not echo SPACE characters back. Also, I tried 1/2 on Login but it resulted the same... I am working on it right now... -- Tim Stark My screen log file: .copy sys:foo.exe=sys:react.exe .r foo REACT> %REACME Command error; Unrecognized switch or keyword: "foo" REACT>exit .r react REACT> %REACME Command error; Unrecognized switch or keyword: "r" // Hmmm. Non-standard terminal I/O problems. I still am working on // it. REACT>insert New user PPN: [200,200] %REANDF No default project 200 profile found USER>name "user1" USER>personal "Timothy Stark" USER>core USER>ipcf USER>context USER>enq USER>priv PRIVILEGES>admin PRIVILEGES>done USER>show PPN: [200,200] User name: user1 Profile default: [200,%] or [%,%] Personal name: Timothy Stark * Distribution location: -none- * Mailing address: -none- * Expiration date: never * LOGIN times: Weekdays -none- Weekends -none- * Access types: -none- * Requirements for LOGIN: Account and remark strings are not required Name is not required Password is not required Password change at next LOGIN Minimum password length: -none- Password change interval: -none- Password changes are allowed * Schedular type: 0 * Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 100 IPCF quotas: Send 2, Receive 5, PIDs 2 Privileges: Administrative * Spooled device bits: -none- * Watch bits: -none- * Structure quotas: -none- Administrative data: Profile last changed by (PPN unknown) at 262130-3 0:00:00 Last access -never- Last password change -none- USER>password test USER>done User [200,200] user1 inserted with password TEST A total of one user inserted // That account was created successfully but... REACT>exit .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPSER [1,2] Logged-off CTY at 13:08:14 on 22-Jan-1 Runtime: 0:00:09, KCS:407, Connect time: 0:02:12 Disk Reads:1621, Writes:342, Blocks saved:0 .login user1 Job 2 KS10 CTY ?LGNIET Invalid entry #operator [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 13:08 22-Jan-1 Monday // New account did not work with LOGIN. Only Operator account // is accepted. I checked my log file and found out that first // four characters of username were corrupted. Look at my debug // log file below. .r react REACT> %REACME Command error; Unrecognized switch or keyword: "react" REACT>show operator PPN: [1,2] User name: OPERATOR Profile default: -none- Personal name: -none- Distribution location: -none- Mailing address: -none- Expiration date: never LOGIN times: Weekdays 0:23 Weekends 0:23 Access types: Local, Subjob of batch, Batch Requirements for LOGIN: Account and remark strings are not required Name is not required Password under timesharing and batch Password change not required Minimum password length: -none- Password change interval: -none- Password changes are prohibited Schedular type: 0 Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 511 IPCF quotas: Send 511, Receive 511, PIDs 511 Privileges: Disk priority: 3, HPQ: 15, System operator Administrative, ENQ-DEQ, IPCF, LOCK, METER, POKE, RTTRP, CPU, SPY-all-core, SPY-monitor, TRPSET, Unspooling Spooled device bits: CDP, CDR, LPT, PLT, PTP Watch bits: MTA Structure quotas: Structure Quota in Quota out Reserved Status --------- ----------- ---------- ---------- ---------- DSKA 100000 100000 0 DSKB 100000 100000 0 DSKC 100000 100000 0 Administrative data: Profile last changed by [1,2] at 5-Mar-86 1:59:09 Last access succeeded on 4-Mar-86 17:48:52 Last password change at 2-Dec-85 5:56:37 A total of one user listed // Hmmm. With using username on show command worked fine. Without // any username or PPN, show command caught an infinite loop of // the Operator account's profile information. REACT>show user1 ?REANSU No such user user1 // Account user1 (200,200) mysteriously disappeared from ACTDAE.SYS file! REACT>exit .kjob [LGTOUL Other users logged-in under [1,2]] Job 2 User OPERATOR [1 ,2] Logged-off CTY at 13:09:09 on 22-Jan-1 Runtime: 0:00:03, KCS:191, Connect time: 0:00:46 Disk Reads:462, Writes:0, Blocks saved:0 . My summarized debug log file: CPU: 2107385 instructions per second. \ CPU: 2102292 instructions per second. > est. 2.1 MIPS on PIII/450. :-) CPU: 2094094 instructions per second. / CTY: USER>done : : MOVS: Instruction (Opcode 016) at PC 071271 // MOVSLJ Instruction MOVS: AC0 = +000000000047 AC1 = +341100030011 <- Bad byte pointer MOVS: AC3 = +000000000047 AC4 = -441100032337 Should be 441100030011. MOVS: F1 = 000000000000 STR: 030011/ 125123105122 => 451 (Pos = 19 Size = 9) = * STR: 032337/ 451000000000 <= 451 (Pos = 27 Size = 9) STR: 030011/ 125123105122 => 442 (Pos = 10 Size = 9) = * STR: 032337/ 451442000000 <= 442 (Pos = 18 Size = 9) STR: 030011/ 125123105122 => 451 (Pos = 1 Size = 9) = * STR: 032337/ 451442451000 <= 451 (Pos = 9 Size = 9) STR: 030012/ 061000000000 => 61 (Pos = 27 Size = 9) = 1 STR: 032337/ 451442451061 <= 61 (Pos = 0 Size = 9) STR: 030012/ 061000000000 => 0 (Pos = 18 Size = 9) STR: 032340/ 000000000000 <= 0 (Pos = 27 Size = 9) STR: 030012/ 061000000000 => 0 (Pos = 9 Size = 9) STR: 032340/ 000000000000 <= 0 (Pos = 18 Size = 9) STR: 030012/ 061000000000 => 0 (Pos = 0 Size = 9) STR: 032340/ 000000000000 <= 0 (Pos = 9 Size = 9) STR: 030013/ 000000000000 => 0 (Pos = 27 Size = 9) : : STR: 030022/ 000000000000 => 0 (Pos = 9 Size = 9) STR: 032350/ 000000000000 <= 0 (Pos = 18 Size = 9) STR: 030022/ 000000000000 => 0 (Pos = 0 Size = 9) STR: 032350/ 000000000000 <= 0 (Pos = 9 Size = 9) // This instruction producted corrupted username (a first few characters). // and REACT put an affected string into ACTDAE.SYS for LOGIN or SHOW // command. : : CTY: .login user1 CTY: Job 2 KS10 CTY EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000011 AC1 = -444400027762 CMPS: AC3 = +000000000011 AC4 = -444400216004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 125123105122 => 125123105122 (Pos = 0 Size = 36) STR: 216004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 125123105122 B2 = 777777777777 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223005 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 125123105122 => 125 (Pos = 27 Size = 9) STR: 223005/ 044446431000 => 44 (Pos = 27 Size = 9) STR: B1 = 000000000125 B2 = 000000000044 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223022 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 125123105122 => 125 (Pos = 27 Size = 9) STR: 223022/ 044446451124 => 44 (Pos = 27 Size = 9) STR: B1 = 000000000125 B2 = 000000000044 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223037 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 125123105122 => 125 (Pos = 27 Size = 9) STR: 223037/ 117120105122 => 117 (Pos = 27 Size = 9) STR: B1 = 000000000125 B2 = 000000000117 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223054 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 125123105122 => 125 (Pos = 27 Size = 9) STR: 223054/ 441041442105 => 441 (Pos = 27 Size = 9) STR: B1 = 000000000125 B2 = 000000000441 // Yes, LOGIN can't find "USER1" in ACTDAE.SYS because of corrupted // characters in first word that were created by REACT program. // Note: ACTDAE.SYS has four accounts. I created three accounts to test. CTY: ?LGNIET Invalid entry CTY: #operator EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000011 AC1 = -444400027762 CMPS: AC3 = +000000000011 AC4 = -444400216004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 117120105122 => 117120105122 (Pos = 0 Size = 36) STR: 216004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 117120105122 B2 = 777777777777 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223005 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 117120105122 => 117 (Pos = 27 Size = 9) STR: 223005/ 044446431000 => 44 (Pos = 27 Size = 9) STR: B1 = 000000000117 B2 = 000000000044 EXTEND: 003 at PC 071357 CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223022 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 117120105122 => 117 (Pos = 27 Size = 9) STR: 223022/ 044446451124 => 44 (Pos = 27 Size = 9) STR: B1 = 000000000117 B2 = 000000000044 EXTEND: 003 at PC 071357 // CMPSLE instruction CMPS: Instruction (Opcode 003) at PC 071356 CMPS: AC0 = +000000000047 AC1 = -441100027762 CMPS: AC3 = +000000000047 AC4 = -441100223037 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 117120105122 => 117 (Pos = 27 Size = 9) = O STR: 223037/ 117120105122 => 117 (Pos = 27 Size = 9) STR: 027762/ 117120105122 => 120 (Pos = 18 Size = 9) = P STR: 223037/ 117120105122 => 120 (Pos = 18 Size = 9) STR: 027762/ 117120105122 => 105 (Pos = 9 Size = 9) = E STR: 223037/ 117120105122 => 105 (Pos = 9 Size = 9) STR: 027762/ 117120105122 => 122 (Pos = 0 Size = 9) = R STR: 223037/ 117120105122 => 122 (Pos = 0 Size = 9) STR: 027763/ 101124117122 => 101 (Pos = 27 Size = 9) = A STR: 223040/ 101124117122 => 101 (Pos = 27 Size = 9) STR: 027763/ 101124117122 => 124 (Pos = 18 Size = 9) = T STR: 223040/ 101124117122 => 124 (Pos = 18 Size = 9) STR: 027763/ 101124117122 => 117 (Pos = 9 Size = 9) = O STR: 223040/ 101124117122 => 117 (Pos = 9 Size = 9) STR: 027763/ 101124117122 => 122 (Pos = 0 Size = 9) = R STR: 223040/ 101124117122 => 122 (Pos = 0 Size = 9) STR: 027764/ 000000000000 => 0 (Pos = 27 Size = 9) STR: 223041/ 000000000000 => 0 (Pos = 27 Size = 9) STR: 027764/ 000000000000 => 0 (Pos = 18 Size = 9) STR: 223041/ 000000000000 => 0 (Pos = 18 Size = 9) STR: 027764/ 000000000000 => 0 (Pos = 9 Size = 9) : : STR: 027773/ 000000000000 => 0 (Pos = 18 Size = 9) STR: 223050/ 000000000000 => 0 (Pos = 18 Size = 9) STR: 027773/ 000000000000 => 0 (Pos = 9 Size = 9) STR: 223050/ 000000000000 => 0 (Pos = 9 Size = 9) STR: B1 = 000000000000 B2 = 000000000000 // Yes, LOGIN found "OPERATOR" in ACTDAE.SYS but... EXTEND: 016 at PC 071272 // MOVSLJ Instruction MOVS: Instruction (Opcode 016) at PC 071271 MOVS: AC0 = +000000000047 AC1 = +341100211115 <= AC1 has a corrupted MOVS: AC3 = +000000000047 AC4 = -441100125516 byte pointer. MOVS: F1 = 000000000000 Should be 441100211115. STR: 211115/ 117120105122 => 450 (Pos = 19 Size = 9) = * \ STR: 125516/ 450000000000 <= 450 (Pos = 27 Size = 9) \ Corrupted STR: 211115/ 117120105122 => 42 (Pos = 10 Size = 9) = * > characters STR: 125516/ 450042000000 <= 42 (Pos = 18 Size = 9) / due to a bug STR: 211115/ 117120105122 => 451 (Pos = 1 Size = 9) = * / somewhere. STR: 125516/ 450042451000 <= 451 (Pos = 9 Size = 9) STR: 211116/ 101124117122 => 101 (Pos = 27 Size = 9) = A STR: 125516/ 450042451101 <= 101 (Pos = 0 Size = 9) STR: 211116/ 101124117122 => 124 (Pos = 18 Size = 9) = T STR: 125517/ 124000000000 <= 124 (Pos = 27 Size = 9) STR: 211116/ 101124117122 => 117 (Pos = 9 Size = 9) = O STR: 125517/ 124117000000 <= 117 (Pos = 18 Size = 9) STR: 211116/ 101124117122 => 122 (Pos = 0 Size = 9) = R STR: 125517/ 124117122000 <= 122 (Pos = 9 Size = 9) STR: 211117/ 000000000000 => 0 (Pos = 27 Size = 9) STR: 125517/ 124117122000 <= 0 (Pos = 0 Size = 9) STR: 211117/ 000000000000 => 0 (Pos = 18 Size = 9) STR: 125520/ 000000000000 <= 0 (Pos = 27 Size = 9) STR: 211117/ 000000000000 => 0 (Pos = 9 Size = 9) : : STR: 211126/ 000000000000 => 0 (Pos = 9 Size = 9) STR: 125527/ 000000000000 <= 0 (Pos = 18 Size = 9) STR: 211126/ 000000000000 => 0 (Pos = 0 Size = 9) STR: 125527/ 000000000000 <= 0 (Pos = 9 Size = 9) CTY: [LGNJSP Other jobs same PPN] CTY: [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] CTY: 13:08 22-Jan-1 Monday CTY: CTY: .r react CTY: CTY: REACT> CTY: %REACME Command error; Unrecognized switch or keyword: "react" CTY: REACT>show operator // Same results from "LOGIN OPERATOR". CTY: PPN: [1,2] CTY: User name: OPERATOR CTY: Profile default: -none- CTY: Personal name: -none- CTY: Distribution location: -none- CTY: Mailing address: -none- CTY: CTY: Expiration date: never CTY: LOGIN times: Weekdays 0:23 Weekends 0:23 CTY: Access types: Local, Subjob of batch, Batch CTY: Requirements for LOGIN: CTY: Account and remark strings are not required CTY: Name is not required CTY: Password under timesharing and batch CTY: Password change not required CTY: Minimum password length: -none- CTY: Password change interval: -none- CTY: Password changes are prohibited CTY: Schedular type: 0 CTY: Program to run: -none- CTY: CTY: Context-quotas: Contexts 4, Total pages 1000 CTY: Core Limits: Physical 512, Virtual 512 CTY: ENQ/DEQ quota: 511 CTY: IPCF quotas: Send 511, Receive 511, PIDs 511 CTY: Privileges: Disk priority: 3, HPQ: 15, System operator CTY: Administrative, ENQ-DEQ, IPCF, LOCK, METER, POKE, RTTRP, CPU, SPY-all-co CTY: re, SPY-monitor, TRPSET, Unspooling CPU: 457706 instructions per second. CTY: Spooled device bits: CDP, CDR, LPT, PLT, PTP CTY: Watch bits: MTA CTY: CTY: Structure quotas: CTY: Structure Quota in Quota out Reserved Status CTY: --------- ----------- ---------- ---------- ---------- CTY: DSKA 100000 100000 0 CTY: DSKB 100000 100000 0 CTY: DSKC 100000 100000 0 CTY: CTY: Administrative data: CTY: Profile last changed by [1,2] at 5-Mar-86 1:59:09 CTY: Last access succeeded on 4-Mar-86 17:48:52 CTY: Last password change at 2-Dec-85 5:56:37 CTY: CTY: CTY: A total of one user listed CTY: REACT>show user1 // Same results when attempt log into system as "USER1" above. CTY: ?REANSU No such user user1 CTY: REACT>exit CTY: -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3258 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!portc03.blue.aol.com!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: A problem on OPR and REACT programs.. Newsgroups: alt.sys.pdp10 References: <6ZC96.20807$lV5.338311@news2.giganews.com> <949hin$c51$1@bob.news.rcn.net> <1k0a6.248281$IP1.8278091@news1.giganews.com> <94c920$25n$1@bob.news.rcn.net> <94eoqp$bk7$2@bob.news.rcn.net> <3A6AF7E3.BE9C3248@bartek.dontspamme.net> <94h48i$15e$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 33 Message-ID: NNTP-Posting-Date: Tue, 23 Jan 2001 09:09:06 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-0XeJ1VaMwEeF3W9hwTfLIRS8eC+H34J9ftrdeHl3doXf5pM+P0Nqz5nTKvWfV4nZJRjnU3KD4r9hhQL!+aVWGBXa++gy7AIVHJFbk7lzbw8= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 23 Jan 2001 15:09:06 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3258 jmfbahciv@aol.com wrote: > Sounds like our bug. I've been waiting for you to trip over > one of those :-) but I was fearful that it would be a monitor > bug. Now, the question is--how in the hell did that one > get out? Try entering the PPN as 1/2 instead of 1,2 and > see if that one works. I don't know anybody who used a comma. Barb, No, I found a bug in my emulator. That was ADJBP instruction. It sometimes produce an incorrect byte pointer. Yes, it gave -36 instead of 36 in position field. When I fixed a bug in ADJBP, All problems disappeared at once!! Now terminal I/O worked properly with OPR and REACT. They no longer complaint about weird something within command line. Also they now echo spaces! When I pressed CR, it woke up and processed my commands properly. On REACT, no more infinite loops when I just executed 'show'. A few problems are remaining right now. I tried to enter PPN 1,2 but it did not work properly. Only usernames worked. I still am working on it right now... Whew! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3259 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: A problem on OPR and REACT programs.. Newsgroups: alt.sys.pdp10 References: <6ZC96.20807$lV5.338311@news2.giganews.com> <949hin$c51$1@bob.news.rcn.net> <1k0a6.248281$IP1.8278091@news1.giganews.com> <94c920$25n$1@bob.news.rcn.net> <94eoqp$bk7$2@bob.news.rcn.net> <3A6AF7E3.BE9C3248@bartek.dontspamme.net> <94h48i$15e$1@bob.news.rcn.net> User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 25 Message-ID: NNTP-Posting-Date: Tue, 23 Jan 2001 10:17:22 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-sWA3o2jPvZayVX45D5qnY3tKisfD9fQs1E//fKxyDu0oFZ8o2FzIISvv3nNWT+KlPR7+baiRSCfrkAX!4HolGjOFfLpNwGrJvI5rAFHaX78= X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Tue, 23 Jan 2001 16:17:22 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3259 Timothy Stark wrote: > A few problems are remaining right now. I tried to enter PPN 1,2 > but it did not work properly. Only usernames worked. Well, I finally recongized that ACTDAE.SYS was messed up because a bug in EXTEND instructions put some accounts in wrong ordering. I tested some different PPNs like 100,102, 200,200, 103011,4207, 100,500, etc... I found out that some PPNs worked fine. Yes, I was able logged into system by using 'login 100,500'. I was unable to log into system like 'login 100,100' or 'login 1,2'. I will restore ACTDAE.SYS from a tape soon to resolve login problems. Yes, a bug in EXTEND instructions already was fixed before to produce wrong SKIP return that I assume. I will make more tests after ACTDAE.SYS restored from a tape... Keep you informed later. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3268 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!hermes.visi.com!news-out.visi.com!nntp.flash.net!nntp.giganews.com!nntp3.aus1.giganews.com!news2.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: TS-10 emulator status. Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 586 Message-ID: NNTP-Posting-Date: Wed, 24 Jan 2001 08:42:43 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-vYCd105LPyf7AxO4fSgYocvQm5AODV73gEYSonXG0kIYeaHjS/PRdDHFnaQgpWSMDRKTZbRNjL5BGZd!sgvmi0qI+a50bZynruKrSwTVs3la X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Wed, 24 Jan 2001 14:42:43 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3268 Barb: Yes, I already restored entire disk image from one of my backup files. When I finished my installation, I saved it for backup archives... Good news! I finally resolved all problems with my emulator by using TOPS-10 7.03 operating system. I tried every command and did not encounter any problems. That was a nasty ADJBP bug cause that terminal I/O messed up while using OPR and REACT, login did not work with using PPNs, etc.... I can't believe that buggy ADJBP passed a test on DSKAM! The case had been closed... About TOPS-10 7.04 operating system, I resolved all problems execpt one thing. I was unable use OPR and REACT programs because they stalled forever when they attempted to communicate with GALAXY daemons. I looked into SYSTAT and they looks like fine. About TOPS-20 4.1 operating system, I was able to get the @ prompt on my emulator. When I pressed '?', I got a listing of commands. That's why I need a copy of installation guide to complete TOPS-20 installation. Only a problem is stalling terminal I/O routines. Now I have TOPS-20 sources, I looked into APRSRV.MAC and finally understand it. I quoted a partial PIAPR procedure from APRSRV.MAC: CONSZ APR,APINK ;CTY INTERRUPT? JRST [ ISB DLSCHN ;REQUEST TERMINAL INTERRUPT CONO APR,APFCLR+APINK+APRCHN ;CLEAR INTERRUPTING FLAG XJEN PIAPRX] ;EXIT Now I found out that this instruction stalled my CTY routines in my emulator because my CTY routine instantly cleared CTYOWD and sent an interrupt back to KS10 Processor but later instruction destroy that requested interrupt before dismiss that instruction! ! I have press any key to release stalled process to continue. I will fix that soon. However, I still am not finishing my TS-10 emulator project because LP20, DZ11, and other few devices had not implemented yet. A few instructions are not implemented yet. I implemented CMPS, MOVS, and CVTDB/CVTBD instructions. EDIT instruction had not implemented yet. FP, CVTxx, and other few instructions did not pass required tests with using KLAD package. When all instructions passed tests, the emulator will be released to public. -- Tim Stark Here is screen log file: KS10 08:22:01 CTY system 5000 Connected to Node (0) Line # 42 .LOGIN 1,2 .R OPSER [OPRPAF Processing auto command file] ?OPRALF LOOKUP failure 0 * [CCPWFD Waiting for file daemon to start] %%TTY STOMPER - Starting ^C .SET TTY VT100 .systat Status of KS10 at 8:22:25 on 24-Jan-101 Uptime 25, 74% Null time = 74% Idle + 0% Lost, 2% Overhead 13 Jobs in use out of 35. 12 logged in, 11 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 1 01 2 2,5 DET41 CRSCPY 8+15 SL 0 $ 3 [OPR] CTY SYSTAT 21+SPY RN 0 4 [OPR] DET41 ACTDAE 180+39 SL 2 5 [OPR] DET41 DAEMON 20+SPY SL 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 1 9 [OPR] DET41 CATLOG 30+39 HB 1 10 [OPR] DET41 ORION 99+39 SL 1 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] DET41 LPTSPL 59+39 HB 1 13 [OPR] 41 4 ^C 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 CRSCPY DSKB [SYS] 15 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 682/2000 = 34% Virt. Core used = 721/2000 = 36% 54P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+721) = 25% Average job size =659/13 = 50.6P+296/13 = 22.7P Total=955/13 = 73.4P No busy devices System File Structures: Name Free Mount DSKB 184600 3 Total Free 184600 .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPSER [1,2] Logged-off CTY at 8:22:30 on 24-Jan-1 Runtime: 0:00:01, KCS:15, Connect time: 0:00:28 Disk Reads:179, Writes:0, Blocks saved:0 .login 1,2 Job 3 KS10 CTY [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 08:22 24-Jan-1 Wednesday // Yeah. Finally login with using PPNs worked properly now. // The ADJBP bug cause the invalid entry login errors when // attempt to log in system with using PPNs. .kjob [LGTOUL Other users logged-in under [1,2]] Job 3 User OPERATOR [1,2] Logged-off CTY at 8:22:36 on 24-Jan-1 Runtime: 0:00:00, KCS:9, Connect time: 0:00:03 Disk Reads:126, Writes:0, Blocks saved:0 .login operator Job 2 KS10 CTY [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 08:22 24-Jan-1 Wednesday .r react REACT>insert New user PPN: [103011,4207] %REANDF No default project 103011 profile found USER>name "Sword7" USER>personal "Timothy Stark" USER>password sword7 USER>enq USER>core USER>ipcf USER>context USER>show PPN: [103011,4207] User name: Sword7 Profile default: [103011,%] or [%,%] Personal name: Timothy Stark * Distribution location: -none- * Mailing address: -none- * Expiration date: never * LOGIN times: Weekdays -none- Weekends -none- * Access types: -none- * Requirements for LOGIN: Account and remark strings are not required Name is not required Password is not required Password change at next LOGIN Minimum password length: -none- Password change interval: -none- Password changes are allowed * Schedular type: 0 * Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 100 IPCF quotas: Send 2, Receive 5, PIDs 2 * Privileges: -none- * Spooled device bits: -none- * Watch bits: -none- * Structure quotas: -none- Administrative data: Profile last changed by (PPN unknown) at 262130-3 0:00:00 Last access -never- Last password change -none- USER>login LOGIN-TIMES>all LOGIN-TIMES>done USER>priv PRIVILEGES>admin PRIVILEGES>done USER>done User [103011,4207] Sword7 inserted with password SWORD7 A total of one user inserted REACT>exit // That account was created successfully without any problems. .kjob [LGTOUL Other users logged-in under [1,2]] Job 2 User OPERATOR [1,2] Logged-off CTY at 8:23:55 on 24-Jan-1 Runtime: 0:00:02, KCS:140, Connect time: 0:01:15 Disk Reads:338, Writes:0, Blocks saved:0 .login 103011,4207 Job 2 KS10 CTY %LGNSLE Search list is empty 08:24 24-Jan-1 Wednesday .systat Status of KS10 at 8:24:03 on 24-Jan-101 Uptime 2:02, 105% Null time = 105% Idle + 0% Lost, 2% Overhead 12 Jobs in use out of 35. 12 logged in, 10 detached. Job Who Line# What Size(P) State Run Time 1 [OPR] DET41 STOMPR 9+8 SL 1 01 2 103011,4207 CTY SYSTAT 21+SPY RN 0 $ 4 [OPR] DET41 ACTDAE 180+39 SL 3 5 [OPR] DET41 DAEMON 20+SPY SL SW 0 6 [OPR] DET41 FILDAE 17 HB 0 7 [OPR] DET41 QUASAR 109+39 SL 1 8 [OPR] DET41 PULSAR 65+39 HB 1 9 [OPR] DET41 CATLOG 30+39 HB 1 10 [OPR] DET41 ORION 99+39 SL 1 11 [OPR] DET41 BATCON 38+39 SL 0 12 [OPR] DET41 LPTSPL 59+39 HB 1 13 [OPR] 41 4 ^C SW 0 $ means Execute Only n means job runs in HPQ n High Segments: Program Device Owner High(P) Users INITIA DSKB [SYS] 8 1 GLXLIB DSKB [SYS] 39 7 Swapping space used = 662/2000 = 33% Virt. Core used = 698/2000 = 35% 77P Core left Active swapping ratio = 21/775 = .02 Virt. Core saved by sharing = 234/(234+698) = 25% Average job size =651/12 = 54.2P+281/12 = 23.4P Total=932/12 = 77.6P No busy devices System File Structures: Name Free Mount DSKB 184600 1 Total Free 184600 // Yeah. User account 103011,4207 was created sucessfully. Login // with using PPNs worked fine now. .kjob Job 2 User SWORD7 [103011,4207] Logged-off CTY at 8:24:07 on 24-Jan-1 Runtime: 0:00:01, KCS:13, Connect time: 0:00:06 Disk Reads:56, Writes:0, Blocks saved:0 .login 1,2 Job 2 KS10 CTY [LGNJSP Other jobs same PPN] [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] 08:24 24-Jan-1 Wednesday .r opr OPR> 8:22:12 -- Message from the Accounting System -- Account validation is not required 8:22:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:STRLST.SYS[1,4]; End of file 8:22:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:MTALST.SYS[1,4]; End of file 8:22:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:DTALST.SYS[1,4]; End of file 8:22:14 -- Begin auto take file -- File: SYS:SYSTEM.CMD[3,3] 8:22:14 -- Auto Take error on line 3 -- Unrecognized switch or keyword: "IGNORE" 8:22:14 -- End auto take file -- 8:22:14 Printer 0 -- Not available right now -- OPR>show status tape /all OPR> 8:24:29 -- Tape Drive Status -- Drive Trk Status AVR Density ------ --- ----------- --- ------------- MTA000 9 Free No 800/1600 MTA001 9 Free No 800/1600 OPR>show status disk /all OPR> 8:24:33 -- Disk Drive Status -- Drive Type Status AVR STR Volume Unit# ------ ---- ----------- --- ----- ------ ----- RPA0 RP06 Mounted Yes DSKB ONC134 0 RPA1 RP06 Free Yes OPR>show queues OPR> 8:24:36 -- The queues are empty -- OPR>exit // Yeah! Termial I/O routines finally worked properly since the bug in // ADJBP instruction had been fixed. .r dsklst /H FOR HELP */u EXIT .show queues Printer Queue: Job Name Req# Limit User -------- ------ ------- ------------------------ DSKLST 7 1 OPERATOR [1,2] There is 1 job in the queue (none in progress); 1 pages // Yes, DSKLST submit its listing to print spooling directly. .r opr OPR>cancel printer 7 OPR> 8:24:56 -- 1 job canceled -- OPR>show queue OPR> 8:24:59 -- The queues are empty -- OPR>exit // Yeah. Print spooling worked so well... I have not implement // LP20 device routines yet. I just cancelled print job 7 successfully. .please Enter text, terminate with CTRL/Z to wait for response, or ESCape to send message and exit Help! Help! Help! Please fix a nasty bug in ADJBP instruction!!! Thank you! ^Z [PLSOPN Operator at LOCAL has been notified at 8:25:22] ^C .r opr OPR> 8:25:23 <1> -- Message from timesharing user -- job 2 [1,2] OPERATOR at TTY42 Help! Help! Help! Please fix a nasty bug in ADJBP instruction!!! Thank you! OPR>exit // Yeah. PLEASE command worked so well. It sent messages via GALAXY daemons. .r opr OPR>set tape mta0: unavailable Enter text and terminate with ^Z Be be right back.... ^Z OPR> 8:26:14 Device MTA000 -- Unavailable for use -- OPR>exit .r backup /tape mta0: /exit .r opr OPR>set tape mta0: available OPR> 8:26:32 Device MTA000 -- Available for use -- OPR>exit .r backup /tape mta0: ?BKPCOM Can't OPEN mag tape /exit .kjob [LGTOUL Other users logged-in under [1,2]] Job 2 User OPERATOR [1,2] Logged-off CTY at 8:26:58 on 24-Jan-1 Runtime: 0:00:07, KCS:228, Connect time: 0:02:48 Disk Reads:1661, Writes:13, Blocks saved:0 . Here is the summarized debug log file: CTY: .login 1,2 EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000001 AC1 = -444400027762 CMPS: AC3 = +000000000001 AC4 = -444400204004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 000001000002 => 1000002 (Pos = 0 Size = 36) STR: 204004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 000001000002 B2 = 777777777777 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211006 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 000001000002 => 0 (Pos = 27 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 27 Size = 9) STR: 027762/ 000001000002 => 1 (Pos = 18 Size = 9) STR: 211007/ 000001000002 => 1 (Pos = 18 Size = 9) STR: 027762/ 000001000002 => 0 (Pos = 9 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 9 Size = 9) STR: 027762/ 000001000002 => 2 (Pos = 0 Size = 9) STR: 211007/ 000001000002 => 2 (Pos = 0 Size = 9) STR: B1 = 000000000002 B2 = 000000000002 // Yes, LOGIN found the [1,2] account and started to complete // login procedure... EXTEND: 016 at PC 071272 MOVS: Instruction (Opcode 016) at PC 071271 MOVS: AC0 = +000000000004 AC1 = +001100211006 MOVS: AC3 = +000000000004 AC4 = -441100125516 MOVS: F1 = 000000000000 STR: 211007/ 000001000002 => 0 (Pos = 27 Size = 9) STR: 125516/ 000000000000 <= 0 (Pos = 27 Size = 9) STR: 211007/ 000001000002 => 1 (Pos = 18 Size = 9) STR: 125516/ 000001000000 <= 1 (Pos = 18 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 9 Size = 9) STR: 125516/ 000001000000 <= 0 (Pos = 9 Size = 9) STR: 211007/ 000001000002 => 2 (Pos = 0 Size = 9) STR: 125516/ 000001000002 <= 2 (Pos = 0 Size = 9) CTY: [LGNJSP Other jobs same PPN] CTY: [LGNLAS Last access to [1,2] succeeded on 4-Mar-86:17:48:53] CTY: 08:22 24-Jan-1 Wednesday CTY: CTY: . CTY: .login 103011,4207 CTY: Job 2 KS10 CTY EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000001 AC1 = -444400027762 CMPS: AC3 = +000000000001 AC4 = -444400204004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011004207 => 103011004207 (Pos = 0 Size = 36) STR: 204004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 103011004207 B2 = 777777777777 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211006 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011004207 => 103 (Pos = 27 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 27 Size = 9) STR: B1 = 000000000103 B2 = 000000000000 // LOGIN tried to match that account with the [1,2] account. // No, It did not match but continuing to search for another // accounts... EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211126 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011004207 => 103 (Pos = 27 Size = 9) STR: 211127/ 103011004207 => 103 (Pos = 27 Size = 9) STR: 027762/ 103011004207 => 11 (Pos = 18 Size = 9) STR: 211127/ 103011004207 => 11 (Pos = 18 Size = 9) STR: 027762/ 103011004207 => 4 (Pos = 9 Size = 9) STR: 211127/ 103011004207 => 4 (Pos = 9 Size = 9) STR: 027762/ 103011004207 => 207 (Pos = 0 Size = 9) STR: 211127/ 103011004207 => 207 (Pos = 0 Size = 9) STR: B1 = 000000000207 B2 = 000000000207 // Yes, LOGIN found the [103011,4207] account in ACTDAE.SYS... // Started to process login procedure to complete... EXTEND: 016 at PC 071272 MOVS: Instruction (Opcode 016) at PC 071271 MOVS: AC0 = +000000000004 AC1 = +001100211126 <- Finally a good byte MOVS: AC3 = +000000000004 AC4 = -441100125516 pointer due to the MOVS: F1 = 000000000000 ADJBP bug fix. STR: 211127/ 103011004207 => 103 (Pos = 27 Size = 9) STR: 125516/ 103000000000 <= 103 (Pos = 27 Size = 9) STR: 211127/ 103011004207 => 11 (Pos = 18 Size = 9) STR: 125516/ 103011000000 <= 11 (Pos = 18 Size = 9) STR: 211127/ 103011004207 => 4 (Pos = 9 Size = 9) STR: 125516/ 103011004000 <= 4 (Pos = 9 Size = 9) STR: 211127/ 103011004207 => 207 (Pos = 0 Size = 9) STR: 125516/ 103011004207 <= 207 (Pos = 0 Size = 9) EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000001 AC1 = -444400027762 CMPS: AC3 = +000000000001 AC4 = -444400204004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011777777 => 103011777777 (Pos = 0 Size = 36) STR: 204004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 103011777777 B2 = 777777777777 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211006 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011777777 => 103 (Pos = 27 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 27 Size = 9) STR: B1 = 000000000103 B2 = 000000000000 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211126 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 103011777777 => 103 (Pos = 27 Size = 9) STR: 211127/ 103011004207 => 103 (Pos = 27 Size = 9) STR: 027762/ 103011777777 => 11 (Pos = 18 Size = 9) STR: 211127/ 103011004207 => 11 (Pos = 18 Size = 9) STR: 027762/ 103011777777 => 777 (Pos = 9 Size = 9) STR: 211127/ 103011004207 => 4 (Pos = 9 Size = 9) STR: B1 = 000000000777 B2 = 000000000004 EXTEND: 003 at PC 057304 CMPS: Instruction (Opcode 003) at PC 057303 CMPS: AC0 = +000000000001 AC1 = -444400027762 CMPS: AC3 = +000000000001 AC4 = -444400204004 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: 204004/ 777777777777 => 777777777777 (Pos = 0 Size = 36) STR: B1 = 777777777777 B2 = 777777777777 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211006 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 777777777777 => 777 (Pos = 27 Size = 9) STR: 211007/ 000001000002 => 0 (Pos = 27 Size = 9) STR: B1 = 000000000777 B2 = 000000000000 EXTEND: 003 at PC 071463 CMPS: Instruction (Opcode 003) at PC 071462 CMPS: AC0 = +000000000004 AC1 = -441100027762 CMPS: AC3 = +000000000004 AC4 = +001100211126 CMPS: F1 = 000000000000, F2 = 000000000000 STR: 027762/ 777777777777 => 777 (Pos = 27 Size = 9) STR: 211127/ 103011004207 => 103 (Pos = 27 Size = 9) STR: B1 = 000000000777 B2 = 000000000103 CTY: %LGNSLE Search list is empty CTY: 08:24 24-Jan-1 Wednesday CTY: CTY: . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3270 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!sjc-peer.news.verio.net!news.verio.net!ord-read.news.verio.net.POSTED!not-for-mail Newsgroups: alt.sys.pdp10 Subject: Re: TS-10 emulator status. From: jeverett@wwa.DEFEAT.UCE.BOTS.com (John Everett) Organization: Everett Associates X-Newsreader: WinVN 0.99.8 (x86 32bit) References: MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII Lines: 29 Message-ID: <3xDb6.1405$ws2.91181@ord-read.news.verio.net> Date: Wed, 24 Jan 2001 16:37:19 GMT NNTP-Posting-Host: 157.238.71.25 X-Complaints-To: abuse@verio.net X-Trace: ord-read.news.verio.net 980354239 157.238.71.25 (Wed, 24 Jan 2001 16:37:19 GMT) NNTP-Posting-Date: Wed, 24 Jan 2001 16:37:19 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3270 In article , sword7@grace.speakeasy.org says... > >However, I still am not finishing my TS-10 emulator project because >LP20, DZ11, and other few devices had not implemented yet. Tim, are you just trying to get LPT output routed to a printer, or are you committed to emulating devices as originally configured by DEC? I'm asking, because it seems to me emulating an LP11 would be easier than an LP20. While DEC never released a KS-10 with an LP11, while I was with ADP all of our LPT support was via LP11s. The LP11 device service routine, while not on any DEC release tapes, may still be available from sources I am still in touch with. BTW, even while I was porting TOPS-10 to the KS, using an LP20 never made any sense to me. Because of the "NUXI problem" one had to manipulate the output stream a byte at a time to get it into an order that made sense to the UBA for a DMA transfer to the LP20. I seemed to me that if you had to handle the data a byte at a time anyway it was just as efficient (more efficient?) to simply ship it to a "byte at a time" device like the LP11. Besides, the LP11 was a lot cheaper than an LP20. For this reason, the LP20 service routine was one of the few pieces of the KS-10 implentation I wasn't responsible for. Doug Detroy implemented that module. -- jeverettwwacom (John Everett) http://www.wwa.com/~jeverett Article 3273 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!xfer13.netnews.com!netnews.com!howland.erols.net!panix!news.panix.com!not-for-mail From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: TS-10 emulator status. Date: 24 Jan 2001 14:49:02 -0500 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 98 Sender: alderson+news@panix2.panix.com Message-ID: References: NNTP-Posting-Host: panix2.panix.com X-Trace: news.panix.com 980365742 785 166.84.0.227 (24 Jan 2001 19:49:02 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 24 Jan 2001 19:49:02 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Xref: nntp1.ba.best.com alt.sys.pdp10:3273 Tim, I don't have a 4.1 installation for Tops-20, but the contents did not change much if at all between 4.1 and 7.0 on the KL-10; I imagine that they were the same or similar on the KS-10. The Installation Tape contains 6 files: MONITR.EXE EXEC.EXE DLUSER.EXE DLUSER save set DUMPER.EXE DUMPER save sets (one long file, multiple sets internally) There was a bug in DLUSER that actually made the installation manual wrong at one point, so here's how to install a Tops-20 system once you are at the EXEC prompt ("@"). The indented comments are for your information; obviously, you should not type them. @enable turns on OPERATOR privileges--you are logged in as OPERATOR at this point--and changes prompt to "$" $rewind mta0: to make your life easier should you need to re-boot $copy mta0: monitr.exe $copy mta0: exec.exe now you can re-boot from disk and get an EXEC started $copy mta0: dluser.exe NOTE THE DIRECTORY! $copy mta0: USERS.TXT this is the DLUSER data file's default name, which makes life easier $copy mta0: dumper.exe let's get all the copying out of the way now, and start the installa- tion process $dluser DLUSER> structure ps: default, but let's do everything by the book, as it were DLUSER> load users.txt defines the directory layout of the public structure DLUSER> exit The Installation Tape contains , , and in DUMPER save sets. Repeat the RESTORE command below to load all sets. $dumper [Using MTA-DUMPER:] DUMPER> tape mta0: must specify a tape, and MTA-DUMPER: is not (yet) defined DUMPER> restore ps:<*>*.*.* This will get the files I'm assuming that you have named the public structure "PS:", but if you called it something else (e.g., "FOO:") the command should read restore ps:<*>*.*.* foo:<*>*.*.* DUMPER> restore ps:<*>*.*.* This gets the initial files DUMPER> restore ps:<*>*.*.* This gets the GALAXY files DUMPER> restore ps:<*>*.*.* This gets the files DUMPER> exit $unload mta0: $information disk-usage ps:<*> This will tell you how much disk space is taken up by each directory The next step is optional: If you want to catch the memory image of the system in case of monitor crashes, you need to create a repository file called DUMP.EXE of the memory size on your system. For example, if you have 4MW of memory, specify "4096" below. $run makdmp.exe MAKDMP> create ps:dump.exe 4096 MAKDMP> exit Once you've gotten this far, we can explore the next things you need to do to set up a Tops-20 system. And maybe in the mean time someone will find a KS-10 Installation Guide. It won't be me, since I've never worked with the little guys (except as a tourist at MIT). -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless Article 3368 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.44!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail From: Spider Boardman Newsgroups: alt.sys.pdp10 Subject: e: TS-10 Emulator Status... Date: Sat, 03 Feb 2001 02:21:31 -0500 Organization: Duchy of Wabesylvan Obspauk Lines: 40 Message-ID: <3A7BB17B.8FF08582@Orb.Nashua.NH.US> References: <95ekh7$88r$2@bob.news.rcn.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: UmFuZG9tSVaZgyW5nQXpSBfqAppIWgFl1G3uAmPDMh5IzRfDMzJZf8bzTljt3jmT X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 3 Feb 2001 07:21:35 GMT X-Mailer: Mozilla 4.7 [en] (X11; I; OSF1 V4.0 alpha) X-Accept-Language: en Xref: nntp1.ba.best.com alt.sys.pdp10:3368 jmfbahciv@aol.com wrote: > In article , > Timothy Stark wrote: > > I am > >figuring how to set user directories for PPN accounts. I struggled > >to add PPN's directories but I was told that they are not cataloged! > > The only person who might be able to tell you is Spider. I had > thought that everything was done with one PPN insert operation. > It used to be that there were separate steps to set up AUXACC, > QUOLST, and STRLST. The first two were PPN-specific but the > latter was disk structure specific. The old AUXACC functions are definitely part of what's managed by REACT talking with ACTDAE. I forget just where the split wound up with regards to quotas, although I do remember that we can establish the login structure quotas in REACT. The old STRLST functions, though, are now handled by CATLOG talking with CATDAE. The "not cataloged" error Tom's getting comes from REACT's own checks with CATDAE to validate the structures. > I can give you a work around to create UFDs (use CREDIR) but > I can't help you with doing it the "right" way since I didn't > play operator after 7.02. Of course, that won't get him a JSL when he logs in. Tom, you'll have to look at what it takes to get your disk structure(s) understood through CATLOG before you can successfully set them up in REACT. Unfortunately, CATDAE and that part of REACT were DPM's, not mine, so I can't help much. I'd be going by CATLOG's on-line help, too. -- Spider Boardman (at home) spider@Orb.Nashua.NH.US The management (my cats) made me say this. http://www.ultranet.com/~spiderb PGP public key fingerprint: 96 72 D2 C6 E0 92 32 89 F6 B2 C2 A0 1C AB 1F DC Article 3371 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!xfer13.netnews.com!netnews.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail From: Spider Boardman Newsgroups: alt.sys.pdp10 Subject: Re: e: TS-10 Emulator Status... Date: Sat, 03 Feb 2001 20:18:31 -0500 Organization: Duchy of Wabesylvan Obspauk Lines: 26 Message-ID: <3A7CADE7.ECE1B56@Orb.Nashua.NH.US> References: <95ekh7$88r$2@bob.news.rcn.net> <3A7BB17B.8FF08582@Orb.Nashua.NH.US> <95gv5b$676$1@bob.news.rcn.net> <3A7C943C.E67F57F8@Orb.Nashua.NH.US> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: UmFuZG9tSVbNiqG7VVEHAtH+DvxIS6CBN4ehs5IOxZuXoj29OROrhjJuxofx15M/ X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 4 Feb 2001 01:18:30 GMT X-Mailer: Mozilla 4.7 [en] (X11; I; OSF1 V4.0 alpha) X-Accept-Language: en Xref: nntp1.ba.best.com alt.sys.pdp10:3371 Spider Boardman wrote: > jmfbahciv@aol.com wrote: > > > In article <3A7BB17B.8FF08582@Orb.Nashua.NH.US>, > > Spider Boardman wrote: > > > > What CATDAE? I don't remember one running in Tim's SYSTATs. That > > may be the problem. > > It may be been temporary. By the time we shipped 7.04, we'd made a lot of the > Galaxy components start-on-demand, for example. I *think* CATDAE remained as > a component, but I won't swear that DPM didn't wind up merging it into QUASAR > when I wasn't looking. OK, I misremembered badly. CATLOG was the daemon, and OPR was the user interface. I got confused by how the catalogue tables get stuffed into OPR and ORION when they're built--it happens by hidden magic inside GALCNF when it writes out GLXCNF.MAC. C'est la guerre. -- Spider Boardman (at home) spider@Orb.Nashua.NH.US The management (my cats) made me say this. http://www.ultranet.com/~spiderb PGP public key fingerprint: 96 72 D2 C6 E0 92 32 89 F6 B2 C2 A0 1C AB 1F DC Article 3408 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!upp1.onvoy!msc1.onvoy!ply1.onvoy!onvoy.com!newsfeed.direct.ca!look.ca!nntp2.aus1.giganews.com!nntp3.aus1.giganews.com!news4.aus1.giganews.com.POSTED!not-for-mail From: Timothy Stark Subject: Re: Barb, DSKB is not in catalog! Newsgroups: alt.sys.pdp10 References: User-Agent: tin/1.4.2-20000205 ("Possession") (UNIX) (Linux/2.2.17 (i686)) Lines: 300 Message-ID: NNTP-Posting-Date: Thu, 08 Feb 2001 09:49:55 CST Organization: Giganews.Com - Premium News Outsourcing X-Trace: sv2-05msUj17fC1zlZWexAw0bTDanjyqG3kUCpLby6cHHff26TK+3Y3F0pEI30koUokdyleyunAVczrPF8M!rGNBzrBmcj4JZtpPwYG5HrgyRiKt X-Complaints-To: abuse@GigaNews.Com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly Date: Thu, 08 Feb 2001 15:49:56 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3408 Barb, Good news! I finally figured them out and successfully added DSKB disk structure to a list in STRLST.SYS file. Then I tried that on REACT and successfully added disk quota to [103011,4207] and was able login into system as [103011,4207] with correct user directory now. Also, I ran QUOLST to check disk quota and it correctly display disk quota information. Whew!! However, I have a problem with "IGNORE" keyword. -- Tim Stark Here is my log file: .dir sys:???lst.sys DTALST SYS 0 <157> 1-May-85 DSKB: [1,4] MTALST SYS 0 <157> 1-May-85 STRLST SYS 0 <157> 1-May-85 Total of 0 blocks in 3 files on DSKB: [1,4] .type sys:system.cmd SET PRINTER 0 PAGE-LIMIT 2000 SET CARD-PUNCH 0 OUTPUT-LIMIT 1000 SET CARD-PUNCH 0 LIMIT-EXCEEDED-ACTION IGNORE <--- It failed on OPR. Hmmm. SET BATCH-STREAM 0 TIME-LIMIT 0:10 ;TIME: 0 TO 10 MINUTES SET BATCH-STREAM 1 TIME-LIMIT 0:60 ;TIME: 0 TO 1 HOUR SET BATCH-STREAM 2 TIME-LIMIT 10:100000 ; FOR LONG MOTHERS START BATCH-STREAM 0:2 START PRINTER 0 START CARD-PUNCH 0 START READER 0 DISABLE OUTPUT-DISPLAY ALL-MESSAGES /JOB ENABLE VOLUME-RECOGNITION TAPE-DRIVES .set w v f .r opr [S:OPR 5(760) from DSKB0:OPR.EXE[1,4]] [ + GLXLIB 5(1522)] OPR> 10:10:10 -- Message from the Accounting System -- Account validation is not required 10:10:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:STRLST.SYS[1,4]; End of file 10:10:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:MTALST.SYS[1,4]; End of file 10:10:13 -- [SYSTEM]CATALOG error -- Input error reading DSKB:DTALST.SYS[1,4]; End of file 10:10:13 -- Begin auto take file -- File: SYS:SYSTEM.CMD[3,3] OPR>enter catalog CATALOG>show structure * ? Not confirmed: "" 10:10:13 -- Auto Take error on line 3 -- Unrecognized switch or keyword: "IGNORE" 10:10:13 -- End auto take file -- 10:10:14 Printer 0 -- Not available right now -- CATALOG>show structure all CATALOG> 10:12:06 -- Structure ALL is not in the catalog -- CATALOG>exit [snip - checked HELP CATLOG for more information....] .r opr [S:OPR 5(760) from DSKB0:OPR.EXE[1,4]] [ + GLXLIB 5(1522)] OPR>enter catalog CATALOG>show structure * ? Not confirmed: "" CATALOG>show structure all CATALOG> 10:19:34 -- Structure ALL is not in the catalog -- // Barb, no, wildcards did not work... CATALOG>insert structure dskb CATALOG>>add dskb0 rp06 1 CATALOG>>done CATALOG> 10:19:51 -- Structure DSKB inserted -- CATALOG>show structure dskb CATALOG> 10:19:55 -- Structure catalog -- Structure DSKB Owned by no one Unit Type Class ------ ------ ----- DSKB0 RP06 1 // I finally had DSKB in structure list (STRLST.SYS). Whew! CATALOG>exit .r react [S:REACT 33(233) from DSKB0:REACT.EXE[1,4]] [ + GLXLIB 5(1522)] REACT>show [103011,4207] PPN: [103011,4207] User name: SWORD7 Profile default: [103011,%] or [%,%] Personal name: Timothy Stark * Distribution location: -none- * Mailing address: -none- * Expiration date: never LOGIN times: Weekdays 0:23 Weekends 0:23 * Access types: -none- Requirements for LOGIN: Account and remark strings are not required Name is not required Password under timesharing Password change not required Minimum password length: -none- Password change interval: -none- Password changes are allowed * Schedular type: 0 * Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 100 IPCF quotas: Send 2, Receive 5, PIDs 2 Privileges: Administrative * Spooled device bits: -none- * Watch bits: -none- * Structure quotas: -none- Administrative data: Profile last changed by [1,2] at 26-Jan-1 13:47:46 Last access failed on 26-Jan-1 13:50:27 Last password change at 26-Jan-1 13:45:23 A total of one user listed REACT>change [103011,4207] USER>? one of the following: ACCESS-TYPES CONTEXT-QUOTAS CORE-LIMITS DISTRIBUTION-LOCATION ENQ-DEQ-QUOTA EXPIRATION-DATE IPCF-QUOTAS LOGIN-TIMES MAILING-ADDRESS NAME PASSWORD PERSONAL-NAME PRIVILEGES PROFILE-DEFAULT PROGRAM-TO-RUN REQUIREMENTS SCHEDULAR-TYPE SPOOLED-DEVICES STRUCTURE-QUOTAS WATCH-BITS or one of the following: DEFAULT DONE HELP QUIT RESTORE SHOW USER>structure STRUCTURE-QUOTAS>? one of the following: ADD REMOVE or one of the following: DEFAULT DONE HELP NONE RESTORE SHOW or confirm with carriage return STRUCTURE-QUOTAS>show -none- STRUCTURE-QUOTAS>add ? structure name STRUCTURE-QUOTAS>add dskb STRUCTURE-QUOTAS>show Structure Quota in Quota out Reserved Status --------- ----------- ---------- ---------- ---------- DSKB 1000 1000 0 STRUCTURE-QUOTAS>done USER>done User [103011,4207] SWORD7 changed A total of one user changed // Yeah! disk quota had been added successfully because DSKB had be in // structure list (STRLST.SYS). REACT>show [103011,4207] PPN: [103011,4207] User name: SWORD7 Profile default: [103011,%] or [%,%] Personal name: Timothy Stark * Distribution location: -none- * Mailing address: -none- * Expiration date: never LOGIN times: Weekdays 0:23 Weekends 0:23 * Access types: -none- Requirements for LOGIN: Account and remark strings are not required Name is not required Password under timesharing Password change not required Minimum password length: -none- Password change interval: -none- Password changes are allowed * Schedular type: 0 * Program to run: -none- Context-quotas: Contexts 4, Total pages 1000 Core Limits: Physical 512, Virtual 512 ENQ/DEQ quota: 100 IPCF quotas: Send 2, Receive 5, PIDs 2 Privileges: Administrative * Spooled device bits: -none- * Watch bits: -none- Structure quotas: Structure Quota in Quota out Reserved Status --------- ----------- ---------- ---------- ---------- DSKB 1000 1000 0 Administrative data: Profile last changed by [1,2] at 8-Feb-1 10:21:56 Last access failed on 26-Jan-1 13:50:27 Last password change at 26-Jan-1 13:45:23 // Yeah. It finally worked.. A total of one user listed REACT>exit .r opr [S:OPR 5(760) from DSKB0:OPR.EXE[1,4]] [ + GLXLIB 5(1522)] OPR>enter catalog CATALOG>show structure dskb CATALOG> 10:22:34 -- Structure catalog -- Structure DSKB Owned by no one Unit Type Class ------ ------ ----- DSKB0 RP06 1 // It looks like good... CATALOG>exit .kjob [S:LOGOUT 103(2112) + from DSKB0:LOGOUT.EXE[1,4]] [L0: DSK:SWITCH.INI[1,2], error 0] [LGTOUL Other users logged-in under [1,2]] [L0: DSKB0:[1,2].UFD[1,1]] [L0: DSKB0:[1,2].UFD[1,1]] [L1: DSKB0:[1,2].UFD[1,1]] [FL20: DSKB0:[1,2].UFD[1,1]] [FL20: DSKB0:[1,2].UFD[1,1]] Job 3 User OPSER [1,2] Logged-off CTY at 10:22:45 on 8-Feb-1 Runtime: 0:00:39, KCS:954, Connect time: 0:12:43 Disk Reads:1430, Writes:14, Blocks saved:0 .login 103011,4207 Job 2 KS10 CTY %LGNLAF Last access to [103011,4207] failed on 26-Jan-1:13:50:27 10:22 8-Feb-1 Thursday .dir %WLDDEM Directory is empty // It finally worked because DSKB now is in catalog database. .r quolst User: 103011,4207 Str used left:(in) (out) (sys) DSKB: 0 1000 1000 184460 // Yes, it worked fine.... .kjob Job 2 User SWORD7 [103011,4207] Logged-off CTY at 10:23:29 on 8-Feb-1 Runtime: 0:00:01, KCS:13, Connect time: 0:00:39 Disk Reads:143, Writes:12, Blocks saved:0 . -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 3415 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!europa.netcrusader.net!207.172.3.44!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-76 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Barb, DSKB is not in catalog! Date: Fri, 09 Feb 01 10:05:38 GMT Organization: UltraNet Communications, Inc. Lines: 56 Message-ID: <960jql$fh3$1@bob.news.rcn.net> References: X-Trace: UmFuZG9tSVabdIQ9/ETzWGyoCXWn9Ap7NIAlc/AYPTryyr/EoCIyeEgqrWZK4bYK X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 9 Feb 2001 11:20:53 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:3415 In article , Timothy Stark wrote: >Barb, > >Good news! I finally figured them out and successfully added DSKB disk >structure to a list in STRLST.SYS file. Then I tried that on REACT and >successfully added disk quota to [103011,4207] and was able login into >system as [103011,4207] with correct user directory now. Also, >I ran QUOLST to check disk quota and it correctly display disk >quota information. Whew!! Good for you!!! I don't see how you stick with it without a set of notebooks lying in wait above your head. :-) >..type sys:system.cmd >SET PRINTER 0 PAGE-LIMIT 2000 >SET CARD-PUNCH 0 OUTPUT-LIMIT 1000 >SET CARD-PUNCH 0 LIMIT-EXCEEDED-ACTION IGNORE <--- It failed on OPR. Hmmm. Try doing this command to OPR by hand. I have no idea what that IGNORE is for or if it should even be there. Actually I have not idea why in the hell we have card punches in the script. > Structure quotas: > Structure Quota in Quota out Reserved Status > --------- ----------- ---------- ---------- ---------- > DSKB 1000 1000 0 OK, now just as a warning for young and foolish ;-). TOPS-10 will forcibly enforce that quota. You will only be allowed to have 1000 blocks on the disk. If you try to log out with more (and there were ways to get over quota without getting blown up), LOGOUT will cheerfully delete files until the quota is below 1000. One of the things I always did was set my Quota In to be many blocks less than the Quota Out. That way my jobs got their hand slapped before I reached logged out quota but none of the files ever got deleted. It was very useful for the PPNs I used to run batch jobs when I did the software builds for packaging the distribution tapes. And 1000 blocks isn't near enough for doing much development work :-). Just don't make the figure larger than the RP06 can handle. /BAH Subtract a hundred and four for e-mail. Article 3455 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feeder.via.net!newshub2.rdc1.sfba.home.com!news.home.com!news-out.usenetserver.com!news-out.usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: DZ11 docs Organization: Kilonet.net Lines: 31 Message-ID: <3A888166.7C1FCC7C@bartek.dontspamme.net> References: <3A88030B.E3E62919@bartek.dontspamme.net> <969uqh$s2s$1@spies.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Tue, 13 Feb 2001 00:40:48 GMT NNTP-Posting-Host: 167.206.68.16 X-Trace: news02.optonline.net 982024848 167.206.68.16 (Mon, 12 Feb 2001 19:40:48 EST) NNTP-Posting-Date: Mon, 12 Feb 2001 19:40:48 EST Xref: nntp1.ba.best.com alt.sys.pdp10:3455 Al Kossow wrote: > > From article <3A88030B.E3E62919@bartek.dontspamme.net>, by Arthur Krewat : > > > > Everytime someone starts talking "DZ11", I salivate! > > > > That wouldn't be the reaction of anyone that had to write a > driver for one.. Kinda like Pavlov's dog. I know DZ11's personally from KS10's. I know Tim Stark is almost finished with his KS10 emulator - DING! What was so bad about that poor DZ? While not having written a driver from scratch, I did dig into the one in 6.03 for it (or whatever the hell hacked version we had - probably thanks to ADP). I got it to do 19200 baud with no problems. While working on a "network thang", I looked into driving the DZ11 directly to do serial I/O between all five of our KS10's. I started out OPENing TTYnnn, which was really a virtual port in the DCA crossed-over to another port in the other KS10's. I decided to take a look at driving the existing DZ11's that were not used at all. Oh well... when the TS-10 is done, I plan on diving into the DZ11 again to get incoming telnet's to work. Unless Tim in his typically lightning-speed fashion incorporates it himself :) art k. Article 3461 of alt.sys.pdp10: Sender: prep@k9 Newsgroups: alt.sys.pdp10 Subject: Re: DZ11 docs References: <960jql$fh3$1@bob.news.rcn.net> <9610cv$77p$1@bob.news.rcn.net> <3A85706B.B4A96DC3@nospam.nospam> <3A85A0BE.42749562@bartek.dontspamme.net> <3A87128E.D97698A8@digiweb.com> <3A87609A.2945DC32@bartek.dontspamme.net> <3A87776E.D31F6B57@digiweb.com> <3A88030B.E3E62919@bartek.dontspamme.net> From: Paul Repacholi Date: 13 Feb 2001 21:26:08 +0800 Message-ID: <87n1bqhfyn.fsf@prep.synonet.com> Lines: 16 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: 149.d01.pe.iqnet.net.au X-Trace: 13 Feb 2001 20:51:00 +0800, 149.d01.pe.iqnet.net.au Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!news.maxwell.syr.edu!newsfeed.icl.net!skynet.be!newsfeed.iinet.net.au!news.waia.asn.au!usenet.per.paradox.net.au!127.0.0.1!nobody Xref: nntp1.ba.best.com alt.sys.pdp10:3461 Arthur Krewat writes: > Sorry about that. My mistake. > > Everytime someone starts talking "DZ11", I salivate! Strange salia you have... The DZ is the pits. Fries its neibours, is a pain to drive, and has its own quircks as well. All the overhead of a Mux and a single line interface in one. Did save $$ ( vs a DH ) or space ( vs DLs ). -- Paul Repacholi 1 Crescent Rd., +61 (08) 9257-1001 Kalamunda. West Australia 6076 Raw, Cooked or Well-done, it's all half baked. Article 3598 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!newsfeed.direct.ca!look.ca!newshub2.rdc1.sfba.home.com!news.home.com!news-sjo.usenetserver.com!news-out.usenetserver.com!e3500-chi1.usenetserver.com.POSTED!not-for-mail From: Timothy Stark Subject: TS10 Emulator Status - Prepared for release now... Newsgroups: alt.sys.pdp10 User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.16-22 (i686)) Lines: 57 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Thu, 22 Feb 2001 21:57:25 EST Organization: WebUseNet Corp http://www.usenetserver.com - Home of the fastest NNTP servers on the Net. Date: Fri, 23 Feb 2001 02:57:25 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:3598 Hello folks: Good news! I finally made it by almost finished my KS10 emulator. Floating instructions are not finished yet but... TOPS-10 v7.03 operating system run reliably on my KS10 emulator. I ran it on my KS10 emulator more than one hour without any problems. Whew. However, I have some problems with TOPS-10 v7.04 and TOPS-20 v4.1 operating systems. For TOPS-10 v7.03, it worked fine. No problems yet. For TOPS-10 v7.04, I was unable to communicate with galaxy daemons through OPR program. Also LOGIN did not work due the same problem. For TOPS-10 v4.1, I had some problems with interrupt handling. Some interrupt handling are stalling frequently. For example, TOPS-20 v4.1 always stopped during printing characters. They are reproductive because they always result the same. I apologized to Megan because I accidently borrowed her code to complete my emulator last summer. I completely removed her code from it. I rewrote my all instructions to access previous and current AC blocks directly not through memory routines, etc. Also I thank to Robert Supnik for providing his KS10 codes to me. I studied his KS10 code and corrected my some codes. My emulator now runs more reliably than before. It just happened. When I followed KS10 microcode source by wrote my ADJBP instruction. When I looked his KS10 code, I was shocked to notice that my code is very similar to his! Well, I assume that he did followed same KS10 microcode by writing his. :-) Now I ran KLAD tests and fixed many bugs in my some instructions. I was forced to fix some instructions that ended up more similar like his code.. That made KLAD tests happy. Yes, their tests are so tough because I had to follow KS10 microcode documentation, his KS10 code, etc. That's why they require to follow alogrithms on KS10 microcode and his KS10 code in order to work properly. I am trying work on them looks different from other codes. I apologizes for that. Well, I now decide to release my code to public and let developers to fix some instructions, etc. to resolve problems with TOPS-10 v7.04 and TOPS-20 v4.1 operating systems. I read my book about CVS. Its title is "Open Source Developement with CVS". I learned that I can use 'cvs import' to upload them into SourceForge. I will release my code to public night tomorrow. Any segguestions? Thank you! -- Tim Stark Article 3550 of alt.sys.pdp10: Sender: phr2001-nospam@ruckus.brouhaha.com From: Paul Rubin Newsgroups: comp.sys.dec,comp.os.vms,alt.sys.pdp10 Subject: Re: Looking for KA630 CPU Module User's Guide. References: <3A8B15EE.E8B34742@bellatlantic.net> <3A8C24A6.1C0D4F88@bartek.dontspamme.net> <3A8C551E.1FD0CF81@softstar.it> Date: 16 Feb 2001 09:59:58 -0800 Message-ID: <7x7l2qik4h.fsf@ruckus.brouhaha.com> Organization: Nightsong/Fort GNOX Lines: 16 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 16 Feb 2001 10:00:17 -0800, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.kjsl.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com comp.sys.dec:3271 comp.os.vms:22826 alt.sys.pdp10:3550 "Timothy Stark" writes: > Also, I still am not finished my emulator yet. However, I now am > considering to release my unfinished > emulator sources and let developers to finish my emulator because I now > realized that finishing my emulator is taking forever... I do not have my > own ftp site at this time. I am looking for new webhosting provider that > offer both virtual web and ftp server. Tim, I do hope you release emulator source and now is as good a time as any. But I want to add that I think you've done an amazing job of developing it to this point. The only other people who have gotten as far were very experienced PDP-10 hackers like KLH, and corporate funded multi-person projects consisting of same (i.e. XKL). My hat is off to you. Paul Article 3662 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: CVS instructions for creating repository (was Re: TS-10 Release - Slightly delayed.) References: X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 26 Feb 2001 16:10:11 -0800 Message-ID: Lines: 185 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 26 Feb 2001 16:11:37 -0800, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!cyclone.swbell.net!cyclone-sf.pbi.net!64.42.15.2!HSNX.atgi.net!news.kjsl.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com alt.sys.pdp10:3662 Timothy Stark writes: > Help! I tried to create my local repository but have some problems with > 'cvs import' command. [...] > $ cvs import -m "Initial TS10 Developement" ts10 >> > When I attempt to execute 'cvs import ...', I was kept getting > usage information instead! :-( Don't use cvs import. I emailed you detailed instructions on what to do. In case you didn't get my email, here's a copy: Date: 26 Feb 2001 00:13:03 -0000 From: Eric Smith To: Timothy Stark Subject: CVS repository creation Timothy, Here's some step-by-step instructions to help you with CVS. It is easiest if you put your source directory into a nested directory (which need not contain anything else). I'm assuming you'll have it in the directory ~/work/ts10 but you can easily substitute anywhere else. I also assume that you are using the bash shell; if you use tcsh you'll have to use setenv instead of export. I've shown the shell prompt as '% '. 1. Set the CVSROOT environment variable to point to where you want the repository to live: % export CVSROOT=~/ts10-cvs For tcsh, the equivalent would be: % setenv CVSROOT ~/ts10-cvs (Note the space instead of the equals sign.) 2. Create the repository directory: % mkdir $CVSROOT 3. Tell CVS to initialize the repository: % cvs init 4. Create the subdirectory in the repository for your source files: % mkdir $CVSROOT/ts10 % ls $CVSROOT CVSROOT ts10 % If your working source tree has subdirectories, create a similar subdirectory structure here. For instance, if you have ts10/foo, ts10/bar, and ts10/bar/baz: % mkdir $CVSROOT/ts10/foo $CVSROOT/ts10/bar $CVSROOT/ts10/bar/baz 5. cd to the parent of the source directory: % cd ~/work (Note that your ts10 source code is in the ts10 directory *below* this one. If you do an "ls" at this point, it should just show the ts10 subdirectory.) 6. Do a cvs checkout. This sets up the CVS subdirectories in the source tree: % cvs checkout ts10 This will produce output with one line for each file in ts10, with a question mark at the left. This indicates that cvs has found those files but they are not yet part of the repository. 7. cd into the source directory: % cd ts10 8. do a cvs add of all files you want to go into the repository, this tells CVS that you are going to check them in: % cvs add README Makefile *.c *.h (Include any other files you want in source control. If you want to leave some source or header files out that aren't actually part of your release, you can list the source files explicitly rather than using a wildcard.) 9. If you have subdirectories, cd into them and do a cvs add on them as well. cd back to the top (ts10) directory when you're done. 10. commit the files, this actually puts them into the repository: % cvs commit -m "initial version" The -m and string give the initial log message, use anything you feel appopriate. cvs commit works recursively, and will commit all the files you've added, even in subdirectories. If you ever need to commit files individually, you can give the files on the command line. For instance, if you wanted foo.c and bar.c to have a different initial log message than the rest, then before doing a commit of everything (as above), you can do this: % cvs commit -m "a different log message" foo.c bar.c 11. verify that everything looks good: % cvs status | more This will give details about every file in the repository. Each file should have a working revision and a repository revision of 1.1, and Status of Up-to-date. Sometimes I like a more concise summary with just the Status field: % cvs status | grep Status (note the upper case S on the Status in the grep) For extra verification, check out a copy into a new directory and do a build. This will make sure you haven't accidentally omitted any files: 1. Make sure CVSROOT is set as in step 1 of the setup above: % export CVSROOT=~/ts10-cvs 2. Make a new directory to check out into: % mkdir ~/work2 3. cd into that directory % cd ~/work2 4. check out the code: % cvs checkout ts10 5. cd into ts10: % cd ts10 6. build the code as usual: % make Once you've done that, you can blow away work2 if you like: % cd ~ % rm -rf ~/work2 Package up the repository: 1. cd into the repository top-level directory: % cd $CVSROOT 2. tar it up: % tar -cvzf ~/ts10-cvs.tar.gz ./ts10 3. cd out: % cd ~ 4. If you like, email me the repository and I'll test it on my system to make sure there are no problems. However, if you've followed the steps above, I think it should be fine. 5. scp the file to SourceForge: scp ts10-cvs.tar.gz sword7@ts10.sourceforge.net:/ 6. Submit a request to the Sourceforge CVS people to import the repository - for directions see http://sourceforge.net/docman/display_doc.php?docid=768&group_id=1 Let me know if you run into any trouble. Best regards, Eric Article 3676 of alt.sys.pdp10: Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.sys.pdp10 Subject: Re: CVS instructions for creating repository; a few questions References: <3A9B1D14.CB1BE3B9@mail.bcpl.net> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 26 Feb 2001 22:29:00 -0800 Message-ID: Lines: 55 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 26 Feb 2001 22:30:29 -0800, ruckus.brouhaha.com Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!arclight.uoregon.edu!logbridge.uoregon.edu!HSNX.atgi.net!news.kjsl.com!news.spies.com!ruckus.brouhaha.com Xref: nntp1.ba.best.com alt.sys.pdp10:3676 Ken McMonigal writes: > 1) Mr. Stark will have a login name as "Project Admin"? > 2) The rest of us would be listed as developers? Timothy is the project admin and developer. I have admin privs only because I registered the SourceForge project. I'm willing to help with admin tasks, but Timothy can drop my admin privs if he wants. Developer access is only for those who are approved to directly commit changes to the source code in CVS. Timothy can decide who he wants to approve for this. Usually it's only a few people at most. Often this is determined by who in the community becomes most active in improving the software. Other people can certainly be developers, just as thousands of people are developers of the Linux kernel, but only a few are "official" developers. That's not intended to denigrate anyone else's work; it's just to keep some centralized control over things to avoid complete chaos. > 3) The "cvs import" site is only, in our case, a common site for > the repository of the TS-10 & does not relate to the SourceForge > Project? The project home page will probably eventually be: http://www.ts10.org/ It doesn't work yet, don't bother trying. The SourceForge project summary page will be: http://sourceforge.net/projects/ts10/ > 4) Just as one has to set bits in Unix to ensure read-only status > for a website, did the previous post include security commands > to ensure that the software is read-only? The CVS repository on SourceForge will only be writable by the developers that Timothy designates. > 5) What is the mechanism to fix (minor) bugs to Mr. Stark's > software, as well as agreed upon basic enhancements? Anyone will be able to submit bug reports and patches via SourceForge. It is up to the officially designated developers to determine what to do about them. It is my intention to try to help Timothy with the routine admin tasks so he doesn't have to waste time on them. This is why I registed the SourceForge project and the domain name for him. I'll probably also build RPM packages for those who want easy imstallation on Red Hat Linux. If Timothy wants me to, I'm willing to handle the general release process. Best regards, Eric Article 3718 of alt.sys.pdp10: Sender: prep@k9 Newsgroups: alt.sys.pdp10 Subject: Re: TS10 Emulator Status... References: From: Paul Repacholi Date: 04 Mar 2001 00:19:08 +0800 Message-ID: <87g0gu6d1f.fsf@prep.synonet.com> Lines: 17 User-Agent: Gnus/5.0807 (Gnus v5.8.7) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: 202.61.201.213 X-Trace: 3 Mar 2001 23:37:18 +0800, 202.61.201.213 Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsfeed.direct.ca!look.ca!newshub2.rdc1.sfba.home.com!news.home.com!sjc1.nntp.concentric.net!newsfeed.concentric.net!newsfeed.ozemail.com.au!news1.optus.net.au!optus!news.mel.connect.com.au!news.per.connect.com.au!newsfeed.iinet.net.au!news.waia.asn.au!usenet.per.paradox.net.au!127.0.0.1!nobody Xref: nntp1.ba.best.com alt.sys.pdp10:3718 Timothy Stark writes: > Since I uploaded my TS10 sources into SourceForge and submited my request > to SourceForge to import it. Today I checked my TS10 project web site > and it did not show up yet. I still am waiting for that. When > my sources finally is imported, I will inform you that as soon as possible. > Also, I will update my sources because I implemented some VAX > instructions, etc. It is there now ( 1600UT ) but nothing in the CVS area. -- Paul Repacholi 1 Crescent Rd., +61 (08) 9257-1001 Kalamunda. West Australia 6076 Raw, Cooked or Well-done, it's all half baked. Article 3724 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!sn-xit-04!supernews.com!cabal10.airnews.net!news.airnews.net!cabal11.airnews.net!news.airnews.net!uunet!ash.uu.net!spool1.news.uu.net!spool0.news.uu.net!reader3.news.uu.net!not-for-mail Message-ID: <3AA1FE60.608965A4@trailing-edge.com> Date: Sun, 04 Mar 2001 08:35:44 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: New and Improved access to PDP-10 software archives Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 23 NNTP-Posting-Host: 63.73.218.130 X-Trace: reader3.news.uu.net 983712944 15152 63.73.218.130 Xref: nntp1.ba.best.com alt.sys.pdp10:3724 In anticipation of release of Timothy Stark's PDP-10 emulator, I've made gzipped ".tap" style tape archive images available through the archive site at http://pdp-10.trailing-edge.com/ or, if you prefer direct ftp access, at ftp://ftp.trailing-edge.com/pub/pdp10/ I will still keep the ".tpc" style tape archive images around, as all my tape image tools work from them, and of course anyone can convert a ".tpc" image to a ".tap" image by simply running the mtcvtv23 program that comes with Bob Supnik's simulator package. But I do realize the very real advantages of directly distributing John Wilson's ".tap" format image for emulator users, and will continue to do what I can to make life easy for those users, including the possibility of distributing disk images containing already- installed TOPS-10 and TOPS-20 systems when such images become available. Tim. Article 3725 of alt.sys.pdp10: From: Neil Franklin Newsgroups: alt.sys.pdp10 Subject: Re: O/S for PDP-10 plus PDP-10 FPGA box prototype Date: 04 Mar 2001 15:08:56 +0100 Organization: My own Private Self Lines: 67 Message-ID: <6uk8658w3r.fsf@chonsp.franklin.ch> References: <3aa0266f.4659287@news.m.iinet.net.au> <3ad0at8ae9q71gtgbali4u2ttt5jiurqr2@4ax.com> <3AA0844E.A2E68512@mail.bcpl.net> <3aa17069.4085815@news.m.iinet.net.au> X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 4 Mar 2001 14:08:56 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 NNTP-Posting-Host: galapagos.ethz.ch X-Trace: 4 Mar 2001 15:09:42 +0100, galapagos.ethz.ch Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!enews.sgi.com!news-zh.switch.ch!pfaff.ethz.ch!chonsp.franklin.ch!not-for-mail Xref: nntp1.ba.best.com alt.sys.pdp10:3725 berd_kalamunda@techemail.com (Rolie Baldock) writes: > We have enough info from Tom Stockebrand and others in this NG to put > DECTAPEs on a KI/KA ten. Neat. Usually they only were on KL-10s. I like DECTAPEs since I first saw this elegantly simple device running on an PDP-8/A. > Moreover we could make the controller much > simpler using currently available microcontrollers. From the "Small Computer Handbook" 1973: ------------- TD8-E DECTAPE CONTROL The TD8-E is contained on a single quad Flip-Chip module which plugs directly into the OMNIBUS of the PDP-8/E. ------------- That looks like a simple circuit. Direct in FPGA stuff. No microprocessor, not even trap&emulate. An real old device connected. Actually given the size, the KI-10 will just about fit inside the TU-56 drive. > KA/KI ten on a FPGA chip will hit the scene soon I expect 1-2 years, depending on how much you want to wait for, before starting DECTAPEing. > and hopefully > I can build one from his design details which he promised to make > available. It will be available. Actually what already exists [1] of it is already available [2]. [1] 32 words of memory, instruction pointer and incrementer, an data bus multiplexer, 4xx binary logic instruction unit. No interconnects yet (I was going to try them today [3]), no instruction decoder or timing control yet. [2] As usual on the project web site: http://neil.franklin.ch/Projects/PDP-10/ [3] Sister just phoned and invited me for evening meal, housewarming her new flat. > VIVE ERIC !!!!!!!!!!!!!! > My humble apologies to Mr Franklin re the KA/KI on an FPGA chip. Just > too many names to remember. :-) When I first read Eric (without Kelly) I first interpreted it as Eric Smith, who I know to also be in FPGAs. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Hacker, Unix Guru, El Eng FH/BSc, Sysadmin, Roleplayer, LARPer, Mystic Article 3855 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!opentransit.net!wanadoo.fr!not-for-mail From: Hans B Pufal Newsgroups: alt.sys.pdp10 Subject: Re: Web URL for Daniel Seagrave's emulator? Date: Sun, 11 Mar 2001 06:56:01 +0100 Organization: Wanadoo, l'internet avec France Telecom Lines: 18 Message-ID: <3AAB1371.59D57628@digiweb.com> References: <3AAA38EC.77A4141A@trailing-edge.com> NNTP-Posting-Host: agrenoble-101-2-1-165.abo.wanadoo.fr Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: wanadoo.fr 984290186 19980 193.251.69.165 (11 Mar 2001 05:56:26 GMT) X-Complaints-To: abuse@wanadoo.fr NNTP-Posting-Date: 11 Mar 2001 05:56:26 GMT To: Daniel Seagraves X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U) X-Accept-Language: en Xref: nntp1.ba.best.com alt.sys.pdp10:3855 Daniel Seagraves wrote: > As for E10 I don't have it on the internet by myself anymore - Nobody > wants it, it doesn't work, and Hans Pufal is working on a better version > of it anyway. I take it that you have lost interest in this now Daniel? I did take Daniels code, restructured it a bit and am slowly working my way through fixing problems thrown up by the KLAD tests. Its on the web at , I'll put up my latest version later on today. It compiles and runs on Linux and Windoze 2000 (and presumably Win98). See readme etc. I thnk it uses only standard C except in a handlng console input which is factored out as a separate file., so hopefully it should port easily. -- Hans B Pufal Article 4545 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!enews.sgi.com!newshub2.rdc1.sfba.home.com!news.home.com!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: simh V26 beta 7 Date: Tue, 17 Apr 2001 13:03:24 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <4htodtssvi76n4be08ugdsh0749as9034t@4ax.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 29 Xref: nntp1.ba.best.com alt.sys.pdp10:4545 I'll be travelling for the rest of this week, so I'm posting the latest updates now. http://www.tiac.net/users/mps/retro/sources/simh26beta7.zip Check the 0README, there have been some changes in the UI. For the -10, this version will completely build 7.03. My first try at 7.04 is a complete washout: whenever I try to start the boostrap monitor, after DESTROYing, DEFINEing, and REFRESHing DSKB, I get the following dialog: Startup option: noinitia [Rebuilding the system search list from the HOM blocks] % No structures in system search list [Rebuilding the active swapping list from the HOM blocks] % No units in active swapping list [Rebuilding the system dump list from the HOM blocks] Startup option: Tim Stark's sources describe a 'bug' in 7.04 AUTCON; is this the symptom that is seen? Since the memory layout between TS10 and ismh is different (256KW vs 1024KW), I am not sure whether the AUTCON code would be at the place in memory. Any other pointers? /Bob Supnik Article 4685 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!out.nntp.be!propagator-dallas!news-in-dallas.newsfeeds.com!in.nntp.be!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!news-peer1.tiac.net!posterchild2.tiac.net!news@tiac.net From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp11 Subject: simh v2.6 released Date: Sat, 28 Apr 2001 22:12:19 -0400 Organization: www.us.inter.net Lines: 27 Message-ID: NNTP-Posting-Host: ip150.bedford17.ma.pub-ip.psi.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Newsreader: Forte Agent 1.8/32.548 Xref: nntp1.ba.best.com alt.sys.pdp10:4685 alt.sys.pdp11:2161 Simh V2.6 has been released. Sources can be downloaded from the simh web site at: http://www.tiac.net/users/mps/retro/index.html New features in simh V2.6: - PDP-10 simulator - PDP-11 TS11/TSV05, TC11 - PDP-8 DF32, TC08 - PDP-9/15 DECtape - New SCP commands: enable/disable device; show devices - Real time clock autocalibration - Bug fixes and tweaks The PDP-10 simulator successfully runs TOPS-10 7.03 and has the same issues as TS10 with TOPS-10 7.04. It has not been tested with TOPS-20. The TS11/TSV05 is a 22-bit addressing magtape device for Ultrix-11. It has been tested under RT-11. The PDP-8 devices have been tested under OS/8. The 18b DECtapes have been tested with diagnostics. /Bob Supnik Article 4717 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!uunet!dca.uu.net!ash.uu.net!spool0.news.uu.net!reader0.news.uu.net!not-for-mail From: shoppa@mudd.trailing-edge.com Newsgroups: alt.sys.pdp10 Subject: ts10 Tarball Updated Date: 3 May 2001 19:50:07 -0400 Organization: A poorly-installed InterNetNews site Message-ID: <9csqrf$bdm$1@mudd.trailing-edge.com> References: <3AA9E445.4448B2F@trailing-edge.com> <3AF10BC3.948AD8DA@inwap.com> <9crmsq01c85@drn.newsguy.com> Lines: 15 NNTP-Posting-Host: 63.73.218.151 X-Trace: reader0.news.uu.net 988933826 16237 shoppa@63.73.218.151 Xref: nntp1.ba.best.com alt.sys.pdp10:4717 In article <9crmsq01c85@drn.newsguy.com>, Tim Shoppa wrote: >In article <3AF10BC3.948AD8DA@inwap.com>, Joe says... >>> In "ftp://ftp.trailing-edge.com/pub/ts10.tar.gz" you will find >>> a Gzipped tarball that was current as of Friday night @8PM. >> >>There's a problem with that file. The tarball has been updated, and I've added a file in /pub called MD5SUMS containing checksums in case there is any doubt about your file transfer. ts10.tar.gz is now a soft link pointing to the most recent tarball. Tim. Article 4573 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.slurp.net!sakura.lunar-tokyo.net!dseagrav Newsgroups: alt.sys.pdp10 From: Daniel Seagraves Subject: DZ11 code updated again... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Lines: 11 Date: Wed, 18 Apr 2001 17:17:23 -0500 NNTP-Posting-Host: 209.251.101.202 X-Trace: newsfeed.slurp.net 987631879 209.251.101.202 (Wed, 18 Apr 2001 17:11:19 CDT) NNTP-Posting-Date: Wed, 18 Apr 2001 17:11:19 CDT Xref: nntp1.ba.best.com alt.sys.pdp10:4573 ftp to sakura.lunar-tokyo.net, pub/ts10-dz11.04182001.tgz is the new code. This one is a stand-alone tarball - It's the current version of ts10, in it's entirety, with the DZ11 stuff installed. Said code is running on Asuka.umtec.com as well. And on that note, I'm going to leave off playing with TOPS10 for awhile, to focus on getting ITS up and running under ts10. I'm working at building a monitor on Toad, with moderate success. Article 4591 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!enews.sgi.com!news.xtra.co.nz!news.mel.connect.com.au!news1.optus.net.au!optus!bunyip.cc.uq.edu.au!not-for-mail From: Wilber Williams Newsgroups: alt.sys.pdp10 Subject: Re: Some questions Date: Fri, 20 Apr 2001 13:52:20 +1000 Organization: The University of Queensland Lines: 23 Message-ID: <3ADFB274.260FEA10@its.uq.edu.au> References: <9bnec0$msq$1@news.netmar.com> <7xg0f4odi3.fsf@ruckus.brouhaha.com> <7xbspsodea.fsf@ruckus.brouhaha.com> NNTP-Posting-Host: chook.cc.uq.edu.au Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: bunyip.cc.uq.edu.au 987738968 18958 130.102.128.100 (20 Apr 2001 03:56:08 GMT) X-Complaints-To: news@uq.edu.au NNTP-Posting-Date: 20 Apr 2001 03:56:08 GMT X-Mailer: Mozilla 4.61 [en] (Win98; I) X-Accept-Language: en Xref: nntp1.ba.best.com alt.sys.pdp10:4591 From memory, KA "average" instruction time was 2.9 microseconds (varying from about 2 for simple instructions like JFCL to about 18 for a floating divide) and the KL did between about 400-500 nanoseconds. Wilber Paul Rubin wrote: > > Oops, I didn't mean Foonly F1. I'm thinking of a different model which > was supposed to be faster than a KL. I'm not that up on these things. > > Btw, I think a KL was around 3 mips, a KI around 1 mips, and a KA > maybe 0.5 mips. A PDP-6 may have been about like a KA, but I don't know. -- Wilber Associate Director, Networks And Design Services +----------------------------------------------------------------------+ |Wilber WILLIAMS w.williams@its.uq.edu.au | | Telephone : +61 7 3365 4232 | |ITS The University of Queensland Facsimile : +61 7 3365 7539 | |Brisbane Qld 4072 AUSTRALIA Mobile : +61 412 210 063 | +----------------------------------------------------------------------+ Article 4593 of alt.sys.pdp10: Message-ID: <3AE0173F.C3A56251@Empire.Net> Date: Fri, 20 Apr 2001 07:02:23 -0400 From: John Sauter Organization: System Eyes Computer Store X-Mailer: Mozilla 4.73 [en]C-CCK-MCD NSCPCD47 (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: Some questions References: <9bnec0$msq$1@news.netmar.com> <7xg0f4odi3.fsf@ruckus.brouhaha.com> <7xbspsodea.fsf@ruckus.brouhaha.com> <3ADFB274.260FEA10@its.uq.edu.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: 198.144.140.16 X-Trace: News.Destek.net 987764534 198.144.140.16 (20 Apr 2001 07:02:14 -0500) Lines: 31 Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!news.maxwell.syr.edu!news-out.nuthinbutnews.com!propagator!feed2.newsfeeds.com!newsfeeds.com!News.Destek.net Xref: nntp1.ba.best.com alt.sys.pdp10:4593 The fastest I ever saw a KA10 run was 1 MHz. That was the null job: AOJA 0,1 in register 1. By not having to go to the memory bus it was able to count register 0 about 1 million times per second. John Sauter (J_Sauter@Empire.Net) Wilber Williams wrote: > > From memory, KA "average" instruction time was 2.9 microseconds (varying > from about 2 for simple instructions like JFCL to about 18 for a > floating divide) and the KL did between about 400-500 nanoseconds. > > Wilber > > Paul Rubin wrote: > > > > Oops, I didn't mean Foonly F1. I'm thinking of a different model which > > was supposed to be faster than a KL. I'm not that up on these things. > > > > Btw, I think a KL was around 3 mips, a KI around 1 mips, and a KA > > maybe 0.5 mips. A PDP-6 may have been about like a KA, but I don't know. > > -- > Wilber > Associate Director, Networks And Design Services > +----------------------------------------------------------------------+ > |Wilber WILLIAMS w.williams@its.uq.edu.au | > | Telephone : +61 7 3365 4232 | > |ITS The University of Queensland Facsimile : +61 7 3365 7539 | > |Brisbane Qld 4072 AUSTRALIA Mobile : +61 412 210 063 | > +----------------------------------------------------------------------+ Article 4599 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!enews.sgi.com!newshub2.rdc1.sfba.home.com!news.home.com!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: ITS [Was Re: Some questions] Date: Fri, 20 Apr 2001 11:03:10 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <9bnec0$msq$1@news.netmar.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 9 Xref: nntp1.ba.best.com alt.sys.pdp10:4599 simh has most of the hooks for running ITS, including the unique I/O instructions and paging. I still have to finish the 1PROC flag (a single stepping flag) and also the PC sampling feature. All the changes are documented in the KS10 microcode for ITS (KSHACK). The key problem, as Tim noted, is a bootable disk image. /Bob Article 4607 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!opentransit.net!proxad.net!wanadoo.fr!not-for-mail From: Hans B Pufal Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 installation/setup instructions? Date: Sat, 21 Apr 2001 09:05:02 +0200 Organization: Wanadoo, l'internet avec France Telecom Lines: 31 Message-ID: <3AE1311E.AF51C6FE@digiweb.com> References: NNTP-Posting-Host: agrenoble-101-2-1-165.abo.wanadoo.fr Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: wanadoo.fr 987836753 28300 193.251.69.165 (21 Apr 2001 07:05:53 GMT) X-Complaints-To: abuse@wanadoo.fr NNTP-Posting-Date: 21 Apr 2001 07:05:53 GMT X-Mailer: Mozilla 4.75 [en] (Windows NT 5.0; U) X-Accept-Language: en Xref: nntp1.ba.best.com alt.sys.pdp10:4607 Megan wrote: > Someone posted a set of them sometime a week or so ago, but > they scrolled off the list before I could snag a copy... > Can someone please repost them or send a pointer to where > they can be found? If you mean the MIGs for TOPS10 7.03 and 7.04 they are at The 7.03 version is pure ascii text, the 7.04 is a zip of the individual TIFF page scans. There is also a PDF version of the 7.04 scans. -- HBP > > Thanks... > > Megan Gentry > Former RT-11 Developer > > +--------------------------------+-------------------------------------+ > | Megan Gentry, EMT/B, PP-ASEL | Internet (work): gentry!zk3.dec.com | > | Unix Support Engineering Group | (home): mbg!world.std.com | > | Compaq Computer Corporation | addresses need '@' in place of '!' | > | 110 Spitbrook Rd. ZK03-2/T43 | URL: http://world.std.com/~mbg/ | > | Nashua, NH 03062 | "pdp-11 programmer - some assembler | > | (603) 884 1055 | required." - mbg KB1FCA | > +--------------------------------+-------------------------------------+ Article 4619 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!typhoon.aracnet.com!news.he.net!sn-xit-03!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 installation/setup instructions? Date: Sat, 21 Apr 2001 22:31:21 -0000 Message-ID: References: <3AE1311E.AF51C6FE@digiweb.com> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 24 Xref: nntp1.ba.best.com alt.sys.pdp10:4619 Megan wrote: > Now, the problem is that I've restored DSKB but when I go > to run CREDIR, it isn't found (nor, for that matter, is > DIRECT, so I can't even see where things are). Do you run TOPS-10 v7.03? If so, you have to restore tapes twice. .R BACKUP /TAPE MTA0: /RESTORE DSKB: /EXIT .R BACKUP /@[10,7,BUILD]CUSP.RST /EXIT -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 4693 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!216-164-247-119 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: tops20 or 7.04 success anyone? Date: Tue, 01 May 01 10:55:26 GMT Organization: UltraNet Communications, Inc. Lines: 67 Message-ID: <9cmdcq$ab$1@bob.news.rcn.net> References: <9ciql5$pb6$1@slb4.atl.mindspring.net> <3AE8F968.92E2A06E@jetnet.ab.ca> <9ck4h70o5i@drn.newsguy.com> X-Trace: UmFuZG9tSVb+sdCuKZzOH7PtlgSJIxOWLNBrS2/gH4GMxgSZUZcu87e2KC/p/QCG X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 1 May 2001 13:23:38 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:4693 In article <9ck4h70o5i@drn.newsguy.com>, Tim Shoppa wrote: >In article <3AE8F968.92E2A06E@jetnet.ab.ca>, Ben says... >> >>Mark Hittinger wrote: >>> >>> ?hung device at pc x when its not >>> ?illegal uuo at pc x when its a valid instruction >>> ?illegal memory reference at pc x when its not >>> ?io to unassigned channel at pc x when its not >> >>Sounds like timing problems. A emulator speed test may be useful. > >While I do agree that "timing" (as in race conditions) is certainly a >possibility, I don't think that a "speed test" will help. Race conditions >may be hidden on faster or slower hardware, but hiding them is never >a solution to the fundamental problem. > >The race condition may be in TOPS-20 or TOPS-10 7.04 itself >and this race was always hidden on real hardware. I wouldn't say that it was "hidden". It may very well have been put into the code to solve a race condition on the real hardware. This has been one of my concerns with the emulation. An emulation is not a simulation. > I've seen this happen >in other emulators where I/O always completes instantly and the emulated >CPU sees the interrupt or DMA take place exactly at the point that the >request is issued instead of a little later. In these cases the problem >is at the intersection of emulator and OS driver design, not anything to >do with emulation speed. Right. That's because the design had to "wait" for the hardware to catch up or complete or dismiss or whatever. Part of an OS efficiency is to do other stuff waiting for "slower" hardware. Everytime a "faster" piece of gear was put on the system, JMF and TW had a lot of little tweaks that they had to do just to keep the operating system efficient. For instance, putting KA peripherals on a KI, made jobs I/O bound. When those peripherals were replaced with faster gear to solve the I/O bound problems, a system would become CPU bound. It was a never-ending circular tweaking effort. In the case of these emulators, both CPU and I/O speeds have increased. I had no idea what kinds of problems would show up, but I certainly expected them :-). What I can't explain is the 7.03 vs. 7.04 differences. I really didn't think that disk I/O had been tweaked since TW was long gone. But then there were all youngsters in the group who hadn't learned about not touching things that work. IF, and notice the big IF, I had the energy to expend, I'd probably try to approach the problem by running the 7.04 monitor with 7.03 CUSPs. If the disk geometry remains in tact, we can then assume that the problems are in how somebody does I/O in the 7.04 CUSP ship. Then I would try to run the 7.04 CUSPs under 7.03 to see if something stupid happens. But this may be fraught with problems if the CUSPs use some new feature of 7.04. I just don't remember the details to say if such restrictions exist. /BAH Subtract a hundred and four for e-mail. Article 4703 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!wn3feed!worldnet.att.net!24.0.0.38!newshub2.rdc1.sfba.home.com!news.home.com!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator Updates Date: Thu, 03 May 2001 02:59:59 -0000 Message-ID: Sender: Timothy Stark User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.16-22 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 34 Xref: nntp1.ba.best.com alt.sys.pdp10:4703 Hello folks: Today latest updates had been uploaded to SourceForge. You may go ahead to get them. KLINIK port (2nd console device) had been added. Due to security reason, password is required to enter KLINIK device. Look into 'fe.c' for defines. You can change your KLINIK password, port, etc for your own. Its default port for CTY and KLINIK are 5000 and 5001. On TOPS-20 v4.1, I attempted to log into as KLINIK but quickly was told that all remote logins not allowed. New call/carrier loss detection worked fine with TOPS-20 v4.1 but.. TOPS-10 series normally does not detect them. Also, bugs in IMUL and floating instructions had been fixed for better accuracy, etc. They now passed all DSKDA tests with random values. In floating routines, all 32-bit fractions had changed to 64-bit fractions for better accuracy according to DSKDA tests. A bug in TM02/TM03 routines had been fixed. Now you can backspace tape record (space reverse function). For more information, read ChangeLog in pdp10 directory. For next updates, I will work on overdue problems in TOPS-20 v4.1 to resolve. I debugged TOPS-20 v4.1 by running DUMPER to restore files. I learned that only happened to tape operations. Disk operations in TOPS-20 worked fine. I was able to copy DLUSER, DUMPER, MONITR, and EXEC into TOPS-20 system without any problems but. I ended up overdue problems during DUMPER restores. -- Tim Stark Article 4766 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!EU.net!npeer.kpnqwest.net!newsfeeds.belnet.be!news.belnet.be!grolier!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator Status Date: Thu, 10 May 2001 03:47:39 -0000 Message-ID: Sender: Timothy Stark User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.16-22 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 31 Xref: nntp1.ba.best.com alt.sys.pdp10:4766 Hello folks: Finally first KL10 execution on my TS10 emulator!... It emulates KL10-A processor (extended addressing not implemented yet). Yes, I was able load and run BOOTM.EXE from Linux filesystem at TS10> prompt without any problem. Also, I fixed some bugs in disassembly routines because they disassembled KS10-only I/O instructions. After bugs were fixed, it now disassemble any Kx10 I/O instructions. I added some configraton flags, etc to indicate which PDP-10 processor. According to my debug log file, I noticed that first mysteriously codes (starting at location 332) were executed that are not found in BOOTM.MAC. Where did they come from? It transfers mysterious codes to device code 220 (not implemented yet - acts like null device). I looked into BOOTM.MAC and can't find any in it. However, I found device code 220 -- PDC. It looks like PDP-8 interface to me. Now I remember that that is for DA-10 interface? right? It finally reached location 1000 (BOOTM entry point). It correctly detected a processor as KL10 Processor, etc.. Later, it is trying print something and waiting forever for done flag. Good. I will work on device DTE interface soon. Also I have to implement APR, PI, and PAG to complete KL10-A emulation. Yes, it is KS10-compatible but entire different I/O interface. I have a question for you. I never heard of device code 220 (PDC) before. What is PDC for? How does it work? Is it a DA-10 (PDP-8 based) interface? -- Tim Stark Article 4783 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!feed2.news.rcn.net!rcn!netnews.com!newsfeed.skycache.com!Cidera!cyclone-sjo1.usenetserver.com!news-out-sjo.usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: TS10 Emulator Updates [Solaris] Organization: Kilonet.net Lines: 33 Message-ID: <3AFC1050.ADA578D5@bartek.dontspamme.net> References: <3AFB941F.E2A9D80A@inwap.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 11 May 2001 16:21:19 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 989598079 24.186.100.134 (Fri, 11 May 2001 12:21:19 EDT) NNTP-Posting-Date: Fri, 11 May 2001 12:21:19 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:4783 Timothy Stark wrote: > > Joe Smith wrote: > > What does it take to get ts10 to run on Solaris-7 or Solaris-8? > > I do not have Solaris 7/8 operating system to port it. Arthur or someone > can port it to Solaris completely. I downloaded today's CVS, and compiled it - I had to make the following changes: emu/socket.c - add #include to define FASYNC pdp10/cpu_main.c - I like to use -DIDLE so I had to change the WORD36_SXT to a SXT36 Also had to add "-lnsl -lsocket" to final link in main Makefile Other than that, I had no problems whatsoever, except for an annoying blocking on the "panel". I can startup with "ts10 -f ./ks10.cfg" and get the TS10 prompt. From there, typing doesn't do anything, although if I type "boot rpa0" (no echo), and then telnet into the console port 5000, it comes up and runs correctly. I had fixed this in an earlier version, and so far, can't remember what I did - I'll look into it a little more. The above was done with Solaris 8 X86 and Gcc-2.8.1. I was able to boot an existing RP06 with my MONGEN'd monitor. I also have access to Solaris 2.6 and 7 Sparc, and will try this there also, later today. art k. Article 4842 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news-hog.berkeley.edu!ucberkeley!nntp-relay.ihug.net!ihug.co.nz!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Update to simh Date: Tue, 15 May 2001 09:29:48 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <1nb2gt87bit2t89e30vv6augi8c694udlq@4ax.com> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 7 Xref: nntp1.ba.best.com alt.sys.pdp10:4842 Tim Litt has found a couple of small problems in the PDP-10, which are fixed as of last night. In addition, there are updates to the PDP-15 and a copy of an Advanced Software System DECtape that boots and runs. http://www.tiac.net/users/mps/retro/index.html /Bob Article 4857 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!panix!news.panix.com!not-for-mail From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: H316/516 (was Re: Update to simh) Date: 16 May 2001 14:02:48 -0400 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 14 Sender: alderson+news@panix6.panix.com Message-ID: References: <1nb2gt87bit2t89e30vv6augi8c694udlq@4ax.com> <9du15g$fp0$1@bob.news.rcn.net> NNTP-Posting-Host: panix6.panix.com X-Trace: news.panix.com 990036158 2604 166.84.0.231 (16 May 2001 18:02:38 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: 16 May 2001 18:02:38 GMT X-Newsreader: Gnus v5.7/Emacs 20.7 Xref: nntp1.ba.best.com alt.sys.pdp10:4857 jmfbahciv@aol.com writes: > Are you talking about RIM, RIM10, or RIM10B formats? There's a desciption in > the MACRO section of the phone book. It doesn't tell what the program data > is. No, he's talking about non-PDP-10, non-DEC gear at this point. The H316 and H516 were Honeywell boxes, used among other things as ARPANET IMPs (way in the way back when...). Bob's simh is a true wonder, with more personalities than Sibyl. -- Rich Alderson alderson+news@panix.com "You get what anybody gets. You get a lifetime." --Death, of the Endless Article 4888 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newsxfer.eecs.umich.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feeder.qis.net!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: Update to simh Date: Fri, 18 May 2001 11:20:50 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <1nb2gt87bit2t89e30vv6augi8c694udlq@4ax.com> <3B02E062.55C1016D@bartek.dontspamme.net> <9e0cic$42h$3@bob.news.rcn.net> <3B03CA44.6A8C8956@bartek.dontspamme.net> <9e0ilk$ro0$2@bob.news.rcn.net> <3B03DA08.88EE986B@bartek.dontspamme.net> <9e0nbl$dlg$1@bob.news.rcn.net> <3B03E648.29DFC9B4@bartek.dontspamme.net> <9e312u$pbo$2@bob.news.rcn.net> <3B0516E7.9573B388@bartek.dontspamme.net> <3B051AD1.6F52DA2F@bartek.dontspamme.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 37 Xref: nntp1.ba.best.com alt.sys.pdp10:4888 NP, Art. The clock logic compensates for processor speed by checking the amount of real time elapsed every simulated second, and speeding up or slowing down the clock as required. On a different thread, about getting sys000.exe into the 'right format'. I hacked up one of the public domain backup programs to do extraction of binary files as well as ASCII files. The binary format is 36b in 64b, assuming that the on tape format was PDP-10 core dump. I used this successfully to get the diagnostic .EXE's and SAV's off the KLAD tape. simh can load a SAV or EXE in this format into memory via the LOAD command. The hacked sources were on the dbit ftp site; I will also post them on the simh site over the weekend. /Bob On Fri, 18 May 2001 12:56:24 GMT, Arthur Krewat wrote: >Bob, > I received the DZ11 code, but when replying, I was >looked upon as a spammer - I use my own mail server on a cable >modem, which means I am NOT legit :) > > Anyway, I noticed a few more things: > >1) TOPS-10 DAYTIM command reports that the time is NOT slipping. >TS10 does, but as much as a few hours overnight. Tim? > >2) SYSTAT reports 99% idle time after being left running overnight. >TS10 reports 119%. > >thanks! > >art k. Article 4892 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!news-peer1.tiac.net!posterchild2.tiac.net!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: TOPS-20 installation on SIMH, part 1 Date: Sat, 19 May 2001 14:57:19 -0400 Organization: WWW.US.INTER.NET Lines: 242 Message-ID: NNTP-Posting-Host: ip122.bedford.ma.pub-ip.psi.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: news-central.tiac.net 990298784 14803 38.32.9.122 (19 May 2001 18:59:44 GMT) X-Complaints-To: abuse@us.inter.net NNTP-Posting-Date: Sat, 19 May 2001 18:59:44 +0000 (UTC) X-Newsreader: Forte Agent 1.8/32.548 Xref: nntp1.ba.best.com alt.sys.pdp10:4892 I've gotten all the way through a TOPS-20 V4.1 installation on simh, but not without some caveats. To start with, this log is produced with the next version of the simulator, which includes logging. It also includes a shortcut for dealing with the TOPS-20 V4.1 boot problem: sim> set cpu tops20v41 catches the bad WRCSTM in the boot process. To do this by hand, sim> d break 7 sim> BOOT TU or RP Breakpoint at PC: 000007 sim> d 40127 770000000000 sim> c The first part of the installation (up to the reboot) works just as stated in the manuals. After the reboot, things aren't so clean: ------ sim> log t20.txt Logging to file "t20.txt" sim> d wru 006 sim> att tu c:\temp\tops20_v41_i.tap sim> set rp rp06 sim> set cpu t20V41 sim> att rp t20.dsk sim> boot tu MTBOOT>/L MTBOOT>/G143 [FOR ADDITIONAL INFORMATION TYPE "?" TO ANY OF THE FOLLOWING QUESTIONS.] DO YOU WANT TO REPLACE THE FILE SYSTEM ON THE PUBLIC STRUCTURE? Y DO YOU WANT TO DEFINE THE PUBLIC STRUCTURE? Y HOW MANY PACKS ARE IN THIS STRUCTURE: 1 ON WHICH "CHANNEL,UNIT" IS LOGICAL PACK # 0 MOUNTED: 0,0 DO YOU WANT THE DEFAULT SWAPPING SPACE? Y DO YOU WANT THE DEFAULT SIZE FRONT END FILE SYSTEM? Y DO YOU WANT THE DEFAULT SIZE BOOTSTRAP AREA? Y [STRUCTURE "PS" SUCCESSFULLY DEFINED] [PS MOUNTED] ?PS UNIT 0 HAS NO BAT BLOCKS. DO YOU WANT TO WRITE A SET OF PROTOTYPE BAT BLOCKS? Y %%NO SETSPD System restarting, wait... ENTER CURRENT DATE AND TIME: 19-MAY-2001 11:13 YOU HAVE ENTERED SATURDAY, 19-MAY-2001 11:13AM, IS THIS CORRECT (Y,N) Y WHY RELOAD? INSTALLATION ACCOUNTS-TABLE.BIN NOT FOUND - ACCOUNT VALIDATION IS DISABLED RUNNING DDMP NO SYSJOB NO EXEC MX>GET FILE MTA0: INTERRUPT AT 0 MX>GET FILE MTA0: MX>START TOPS-20 Command processor 5.1(1354) @ENABLE (CAPABILITIES) $RUN (PROGRAM) MTA0: DLUSER>LOAD (FROM FILE) MTA0: DONE. DLUSER>EXIT $RUN (PROGRAM) MTA0: DUMPER>TAPE (DEVICE) MTA0: DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS:*.*.* DUMPER tape # 1, "SYSTEM files for TOPS-20 V4.1", Thursday, 7-Apr-83 1721 End of saveset DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS:*.*.* DUMPER tape # 1, "SUBSYS files for TOPS-20 V4.1", Thursday, 7-Apr-83 1723 End of saveset DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS:*.*.* DUMPER tape # 1, "UETP files for TOPS-20 V4.1", Thursday, 7-Apr-83 1726 End of saveset DUMPER>EXIT $UNLOAD (DEVICE) MTA0: $INFORMATION (ABOUT) DISK-USAGE (OF DIRECTORY) PS:<*> PS: 59 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 4 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 0 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 0 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 0 Pages assigned 250 Working pages, 250 Permanent pages allowed PS: 0 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 2595 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 2685 Pages assigned +INF Working pages, +INF Permanent pages allowed PS: 0 Pages assigned 100000 Working pages, 100000 Permanent pages allowed PS: 1712 Pages assigned 4000 Working pages, 4000 Permanent pages allowed PS: 0 Pages assigned 500 Working pages, 500 Permanent pages allowed Total of 7055 Pages assigned, in 11 directories. 62563 Pages free on PS:, 13437 pages used. $CONNECT (TO DIRECTORY) PS: $COPY (FROM) 2020-MONMED.EXE.1 (TO) MONITR.EXE 2020-MONMED.EXE.1 => MONITR.EXE.1 [OK] $TERMINAL (MODE IS) NO RAISE $copy (FROM) tty: (TO) monnam.txt TTY: => MONNAM.TXT.1 Simh TOPS-20 V4.1 ^Z $copy (FROM) tty: (TO) TAPNAM.TXT TTY: => TAPNAM.TXT.1 SIMH-V41 ^Z $terminal (MODE IS) raise $CREATE (FILE) 4-1-CONFIG.CMD Input: 4-1-CONFIG.CMD.1 00100 !TERMINAL SPEEDS 00200 !CURRENTLY, DZ11'S ARE NOT IMPLEMENTED 00300 TERMINAL 1-40 SPEED 0 00400 DEFINE NEW: PS:,SYS: 00500 DEFINE OLD: PS:,SYS: 00600 DEFINE HLP: SYS: 00700 PRINTER 0 LOWERCASE VFU SYS:NORMAL.VFU 00800 PRINTER 0 LOWERCASE RAM SYS:LP96.RAM 00900 TIMEZONE 5 01000 BIAS 8 01100 ENABLE TAPE-DRIVE-ALLOCATION 01200 $ *EU [4-1-CONFIG.CMD.1] $TYPE (FILE) 4-1-CONFIG.CMD !TERMINAL SPEEDS !CURRENTLY, DZ11'S ARE NOT IMPLEMENTED TERMINAL 1-32 SPEED 0 DEFINE NEW: PS:,SYS: DEFINE OLD: PS:,SYS: DEFINE HLP: SYS: PRINTER 0 LOWERCASE VFU SYS:NORMAL.VFU PRINTER 0 LOWERCASE RAM SYS:LP96.RAM TIMEZONE 5 BIAS 8 ENABLE TAPE-DRIVE-ALLOCATION $^ECREATE (DIRECTORY NAME) PS: [Old] $$PASSWORD DEC-20 $$USER-GROUP (NUMBER) 100 $$IPCF $$ $^ECREATE (DIRECTORY NAME) PS: [New] $$ $CONNECT (TO DIRECTORY) PS: $CREATE (FILE) LPFORM.INI Input: LPFORM.INI.1 00100 NORMAL/BANNER:1/HEADER:1/TRAILER:1 00200 NARROW/BANNER:1/HEADER:1/TRAILER:1/WIDTH:80 00300 $ *EU [LPFORM.INI.1] $CONNECT (TO DIRECTORY) PS: $RUN (PROGRAM) SMFILE DECSYSTEM 2020 DIAGNOSTICS FE-FILE PROGRAM VERSION 0.3, TOPS-20, KS10, CPU#=4097 [FOR HELP TYPE "HELP"] SMFILE>WRITE SETUP PS:BOOTSTRAP.BIN SMFILE>WRITE RESET SMFILE>READ KS10.ULD SMFILE>SERIAL 4097 SMFILE>WRITE CRAM SMFILE>WRITE BOOT SMBOOT.EXE SMFILE>WRITE DONE [HOME BLOCKS SET] SMFILE>OUTPUT CRAM PS:KS10.RAM SMFILE>OUTPUT MTBOOT SMMTBT.EXE PS:MTBOOT.RDI SMFILE>EXIT $ Simulation stopped, PC: 000003 (SOJG 2,3) sim> set fe stop sim> c **HALTED** HALT instruction, PC: 010754 (JSR 0,5335) sim> exit Goodbye Log file closed Article 4893 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!news-peer1.tiac.net!posterchild2.tiac.net!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Installing TOPS20 on simh, part 2 Date: Sat, 19 May 2001 15:02:35 -0400 Organization: WWW.US.INTER.NET Lines: 280 Message-ID: NNTP-Posting-Host: ip122.bedford.ma.pub-ip.psi.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: news-central.tiac.net 990299099 14951 38.32.9.122 (19 May 2001 19:04:59 GMT) X-Complaints-To: abuse@us.inter.net NNTP-Posting-Date: Sat, 19 May 2001 19:04:59 +0000 (UTC) X-Newsreader: Forte Agent 1.8/32.548 Xref: nntp1.ba.best.com alt.sys.pdp10:4893 After the reboot, there are a couple of glitches: 1. The 4-1-config.cmd file seems to have an improper command in it, or I have the wrong format for setting speed. 2. There's a bug check as the system comes up, stating that the checkpoint area needs to be rebuilt. Apparently, it does get rebuilt properly. 3. Most critically, there are errors in UETP. The Verify program fails with "improper generations" on the CRAM and BOOT files. In a subsequent run, I tried the RANCBL and RANFOR programs. The RANCBL runs to completion, but RANFOR gives an error message, but no details, every time. I suspect there are still bugs in the floating point code. Does anyone know how to debug from UETP programs? Thanks, /Bob ----- Logging to file "t20p2.txt" sim> set rp rp06 sim> set cpu tops20v41 sim> att rp t20.dsk sim> boot rp BOOT> [PS MOUNTED] SETSPD: TERMINAL 1-40 SPEED 0 ?INVALID CHARACTER IN NUMBER System restarting, wait... ENTER CURRENT DATE AND TIME: 19-MAY-2001 14:05 YOU HAVE ENTERED SATURDAY, 19-MAY-2001 2:05PM, IS THIS CORRECT (Y,N) Y WHY RELOAD? TS ACCOUNTS-TABLE.BIN NOT FOUND - ACCOUNT VALIDATION IS DISABLED RUN CHECKD? N ********************* BUGINF "UXXFIT" AT 19-MAY-2001 14:05:06 *CHECKPOINT FILE NOT IN CORRECT FORMAT FOR THIS SYSTEM, REBUILDING... *JOB: 0, USER: OPERATOR ******************** RUNNING DDMP SYSJOB 5(20) STARTED AT 19-MAY-2001 1405 RUN SYS:ORION RUN SYS:QUASAR RUN SYS:MOUNTR RUN SYS:INFO RUN SYS:MAILER RUN SYS:MAPPER RUN SYS:LPTSPL RUN SYS:LPTSPL RUN SYS:CDRIVE RUN SYS:SPRINT JOB 0 /LOG OPERATOR XX OPERATOR ENA ^ESET LOGIN PSEUDO ^ESET LOGIN CONSOLE ^ESET OPERATOR PTYCON GET SYSTEM:PTYCON.ATO / JOB 1 /LOG OPERATOR XX OPERATOR ENA RUN SYS:BATCON / SJ 1: @LOG OPERATOR OPERATOR SJ 0: @LOG OPERATOR OPERATOR SJ 0: JOB 1 ON TTY43 19-MAY-2001 14:05:14 SJ 0: @ENA SJ 0: $^ESET LOGIN PSEUDO SJ 0: $^ESET LOGIN CONSOLE SJ 0: $^ESET OPERATOR SJ 0: $PTYCON SJ 0: PTYCON> GET SYSTEM:PTYCON.ATO SJ 0: PTYCON> SILENCE SJ 1: JOB 2 ON TTY44 19-MAY-2001 14:05:15 SJ 1: @ENA SJ 1: $RUN SYS:BATCON [From OPERATOR on line 45 to all: SYSTEM IN OPERATION] SJ 0: PTYCON.LOG.1 SJ 0: PTYCON> W ALL SJ 0: OPR(0) 3 OPERATOR OPR TI 0:0:0 SJ 0: PTYCON> CONN OPR SJ 0: [CONNECTED TO SUBJOB OPR(0)] Simh TOPS-20 V4.1, TOPS-20 Monitor 4.1(5471) @SYSTAT OPERATOR 0 DET SYSJOB OPERATOR 1 43 PTYCON OPERATOR 2 44 BATCON OPERATOR 3 45 OPR OPERATOR @ATTACH (USER) OPERATOR (JOB #) 1 [Attached to TTY43, confirm] Password: OPR>SHOW STATUS TAPE-DRIVE /ALL OPR> 14:06:12 -- Tape Drive Status -- DRIVE STATE VOLID REQ# JOB# USER ----- ----- ------ ---- ---- ---- MTA0: Unloaded AVR: Yes MTA1: Unloaded AVR: Yes MTA2: Unloaded AVR: Yes MTA3: Unloaded AVR: Yes MTA4: Unloaded AVR: Yes MTA5: Unloaded AVR: Yes MTA6: Unloaded AVR: Yes MTA7: Unloaded AVR: Yes OPR>SET TAPE-DRIVE MTA0: UNAVAILABLE Enter text and terminate with ^Z INSTALLING SOFTWARE^Z OPR> 14:06:46 -- Tape Drive Set Unavailable -- MTA0: set unavailable for assignment by MOUNTR OPR>PUSH $ASSIGN MTA0: $DUMPER DUMPER>TAPE (DEVICE) MTA0: DUMPER> Simulation stopped, PC: 000003 (SOJG 2,3) sim> att tu c:\temp\tops20_v41_d.tap sim> c DUMPER>REWIND DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS: DUMPER tape # 1, "DOCUMENTATION files for TOPS-20 V4.1", Thursday, 7-Apr-83 1245 Loading file(s) into PS: End of saveset DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS: DUMPER tape # 1, "SYSTEM files for TOPS-20 V4.1", Thursday, 7-Apr-83 1246 End of saveset DUMPER>RESTORE (TAPE FILES) PS:<*>*.*.* (TO) PS: DUMPER tape # 1, "SUBSYS files for TOPS-20 V4.1", Thursday, 7-Apr-83 1248 End of saveset DUMPER>EXIT $UNLOAD (DEVICE) MTA0: $DEASSIGN (DEVICE) MTA0: $TAKE (COMMANDS FROM) PS:SET-UP.CMD [New] [New] [New] [New] [New] [New] [New] [New] [New] [New] [New] [New] End of SET-UP.CMD.1 $CONNECT (TO DIRECTORY) PS: PS: Over working storage allocation by 510 page(s). $RUN (PROGRAM) UETP.EXE [19-May-1 14:10:47 User Environment Test Package ] UETP>TAKE (commands from file) VERIFY.CMD UETP>ENABLE VERIFY/CYCLE:1 ! Verify that the correct software [14:11:45 ENABLE COMPLETED] UETP>BEGIN ! is installed [14:11:46 BEGIN COMPLETED] UETP>DEFAULT/CYCLE:00:15 [14:11:47 DEFAULT COMPLETED] UETP>ENABLE RANCBL [14:11:48 ENABLE COMPLETED] UETP>ENABLE RANFOR [14:11:48 ENABLE COMPLETED] UETP>ENABLE MTA0 [14:11:48 ENABLE COMPLETED] UETP>STATUS [19-May-1 14:11:48] Test Depth Status Cycle Times Error Error Start name run count limit time ====== ===== ======= ===== ===== ===== ===== ===== VERIFY VER Queued 1 0 0 0 19-May-1 14:11:45 RANCBL VER Enabled 0:15 0 0 0 RANFOR VER Enabled 0:15 0 0 0 MTA0 VER Enabled 0:15 0 0 0 START VERIFY 14:11:48 UETP> ** ERROR VERIFY 1 14:12:01 ? Verification error: Error "No such generation number" occurred in line: SYSTEM:KS10.RAM.1 612701 P ** ERROR VERIFY 1 14:12:01 ? Verification error: Error "No such generation number" occurred in line: SYSTEM:MTBOOT.RDI.1 724337 P END VERIFY 14:12:31 [All tests complete on processor # 4097] [19-May-1 14:12:49] Test Depth Status Cycle Times Error Error Start name run count limit time ====== ===== ======= ===== ===== ===== ===== ===== VERIFY VER Ended 1 1 2 0 19-May-1 14:11:45 RANCBL VER Enabled 0:15 0 0 0 RANFOR VER Enabled 0:15 0 0 0 MTA0 VER Enabled 0:15 0 0 0 UETP>EXIT $TAKE (COMMANDS FROM) CLEANUPP\U\-UP.CMD [Old] [Old] [Old] [Old] [Old] [Old] [Old] [Old] [Old] [Old] [Old] [Old] End of CLEAN-UP.CMD.1 $POP OPR>SET TAPE-DRIVE MTA0: AVAILABLE (for use by system) ? OPR GLXINT Send to ORION failed $OPR OPR> 14:11:46 Batch-Stream 0 -- Started -- 14:11:46 Batch-Stream 1 -- Started -- 14:11:46 Batch-Stream 2 -- Started -- 14:11:48 Batch-Stream 0 -- Begin -- Job VERIFY Req #1 for OPERATOR 14:12:31 Batch-Stream 0 -- End -- Job VERIFY Req #1 for OPERATOR OPR>SET TAPE-DRIVE MTA0: AVAILABLE OPR> 14:14:33 -- Tape Drive Set Available -- MTA0: set available for assignment by MOUNTR OPR> Simulation stopped, PC: 000003 (SOJG 2,3) sim> set fe stop sim> c **HALTED** HALT instruction, PC: 010754 (JSR 0,5335) sim> exit Goodbye Log file closed Article 4900 of alt.sys.pdp10: Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.sys.pdp10 References: <3b0748b5.1950474@news.orng1.occa.home.com> Subject: Re: Can't get ts10 to boot using ksbootable Date: Sun, 20 May 2001 09:43:45 -0400 Lines: 30 Organization: The Estopinal Group X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.3018.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.3018.1300 NNTP-Posting-Host: mailgate.theestopinalgroup.com X-Original-NNTP-Posting-Host: mailgate.theestopinalgroup.com Message-ID: <3b07ca13_2@news.iglou.com> X-Trace: news.iglou.com 990366227 mailgate.theestopinalgroup.com (20 May 2001 09:43:47 -0400) X-Authenticated-User: teg X-Original-NNTP-Posting-Host: 204.255.235.119 Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!nntp.abs.net!uunet!dca.uu.net!ash.uu.net!news.iglou.com Xref: nntp1.ba.best.com alt.sys.pdp10:4900 "Roger McCarty" wrote in message news:3b0748b5.1950474@news.orng1.occa.home.com... > Hi! > > I've just started poking around with my old friend the PDP-10. > But, I am having a bit of trouble with ts10. > As you can see, I never get the following: > > BOOTM V6(37) > > BTM>/tm02 > ^^^^^ > Could someone please help by getting me past this little obstacle? Assuming you're running the version from May 5th, the session in which you start up the emulator no longer becomes the operator's CTY console. Instead, while you leave the emulator running in the first session, you need to start another session, this one with TELNET, and direct it to port 5000. You won't be prompted by the host OS to log in; instead, you'll get a KS10 Emulator banner. Press the space bar a couple of times, and you should get the BOOT> prompt. From there, the posted instructions should work just fine. hth, -doug quebbeman Article 4922 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!uio.no!news.kth.se!chicken.stacken.kth.se!thordn From: Thord Nilson Newsgroups: alt.sys.pdp10 Subject: Re: Update to simh Date: Fri, 25 May 2001 01:07:50 +0200 Lines: 23 Message-ID: References: <9e0ilk$ro0$2@bob.news.rcn.net> <3B03DA08.88EE986B@bartek.dontspamme.net> <9e0nbl$dlg$1@bob.news.rcn.net> <3B03E648.29DFC9B4@bartek.dontspamme.net> <9e312u$pbo$2@bob.news.rcn.net> <3B0516E7.9573B388@bartek.dontspamme.net> <3B051AD1.6F52DA2F@bartek.dontspamme.net> <3B0C4527.66E8EDF@bartek.dontspamme.net> <3B0C6CDB.7250AD99@bartek.dontspamme.net> <3B0D8756.536CD933@bartek.dontspamme.net> NNTP-Posting-Host: chicken.stacken.kth.se Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <3B0D8756.536CD933@bartek.dontspamme.net> Xref: nntp1.ba.best.com alt.sys.pdp10:4922 On Thu, 24 May 2001, Arthur Krewat wrote: > Arthur Krewat wrote: > > > > Good ole gatekeeper.dec.com ... /pub/digital/sim/sources/sim_2.3b.tar.Z > > > > I have been unable to find the hacked version of sim_2.3b with DZ11 support. > Apparently, there are hooks placed in the rest of the code... > > any ideas? > > aak > The latest hacked version (sim_2.3d) with DZ11 support can be found on http://elvira.stacken.kth.se /Thord. -------------------------------------------- ! Thord Nilson ! thordn@stacken.kth.se ! -------------------------------------------- Article 4927 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!news-peer1.tiac.net!posterchild2.tiac.net!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: Can't Write to Tape Drive in TS10 Emulator? Date: Sat, 26 May 2001 10:56:41 -0400 Organization: WWW.US.INTER.NET Lines: 36 Message-ID: References: <3b0edc43$1_3@news.iglou.com> <3b0edfc1_3@news.iglou.com> NNTP-Posting-Host: ip100.bedford16.ma.pub-ip.psi.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: news-central.tiac.net 990889166 26474 38.32.90.100 (26 May 2001 14:59:26 GMT) X-Complaints-To: abuse@us.inter.net NNTP-Posting-Date: Sat, 26 May 2001 14:59:26 +0000 (UTC) X-Newsreader: Forte Agent 1.8/32.548 Xref: nntp1.ba.best.com alt.sys.pdp10:4927 The tape implementation in TS10 is read only. Try simh if you need to write tapes. /Bob On Fri, 25 May 2001 18:42:05 -0400, "Douglas H. Quebbeman" wrote: >"Douglas H. Quebbeman" wrote in message >news:3b0edc43$1_3@news.iglou.com... >> Hello, >> >> I don't seem to be able to write to MTA1: though I can >> read from it just fine... is this s limitation of TS10 >> version 0.8.0 / KS10 version 0.8.5 ? >> >> My attach directive is: >> >> attach mta1: /stage10/monopoly.tap > >I failed to mention I'm running TOPS-10 7.03, and the >specific error message I'm getting from BACKUP is > >$BKPTWL tape write locked--add write ring then type "GO" > >I've got a stack of write-enable rings at home, but I >can't figure out how to get them shoved into the emulator... > >;-) > >tia, >-dq > > Article 4928 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!sn-xit-04!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: Re: Can't Write to Tape Drive in TS10 Emulator? Date: Sat, 26 May 2001 15:28:59 -0000 Message-ID: References: <3b0edc43$1_3@news.iglou.com> <3b0edfc1_3@news.iglou.com> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 26 Xref: nntp1.ba.best.com alt.sys.pdp10:4928 Bob Supnik wrote: > The tape implementation in TS10 is read only. > Try simh if you need to write tapes. > /Bob Yeah. That is right. I thought that it was. I will fix it soon. Also, I found another suspect that cause illegal UUO message when it executed RANDOMIZE in BASIC program. I am looking for a bug to fix in progress.. When IMUL was executed, it was trapped due to overflow bit set. It was attempted to execute trap instruction. It feteched 0,,5 and tried to execute. It cause UUO trap. I am looking for a bug that put 0,,5 in trap address or so.. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 4954 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!news.maxwell.syr.edu!news.stealth.net!news-east.rr.com!news.rr.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: simh_2.6 w/DZ11 Organization: Kilonet.net Lines: 47 Message-ID: <3B16CF6D.FE34BF8E@bartek.dontspamme.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Thu, 31 May 2001 23:11:34 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 991350694 24.186.100.134 (Thu, 31 May 2001 19:11:34 EDT) NNTP-Posting-Date: Thu, 31 May 2001 19:11:34 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:4954 Hello all, I put up simh_2.6-pdp10.tar.gz as http://pdp10.kilonet.org/files/sim_2.6-pdp10.tar.gz This includes the DZ11 code from Thorn Nilson. Do an "attach dz0 " where portno is the TCP/IP port. My config file: kilowatt% more ks10.cfg attach rp0 tops10-rpa0.simh.dsk attach tu0 /home/krewat/ts10/tops10_703_ksbootable_bb-x138b-bb.tap attach tu1 /home/krewat/tmp/scratch-tape.tap attach dz0 12000 set tim y2k Do a "make pdp10" and it should build. Makefile assumes you have gcc, and depending on your OS, you may have to remove the "-lnsl -lsocket" from the build line. I did this under Solaris 8 with GCC 2.8.1 Then do a: pdp10 ks10.cfg And, if booting from tape, do a "boot tu0" or from disk "boot rp0" Then, do "telnet localhost 12000" and away we go. I know of no bugs or problems. The system time is perfect. I have stress tested it with up to 8 terminals all running twenty SYS commands and no problems - on a dual Celeron 533. I will put up bootable disk images of TOPS-10 7.03 for RP06 as time permits. The only lingering problem is if you kill the simulator with a ^E and quit, you can't reattach dz0 to port 12000 for a while. I know what it is, but have yet to really investigate. Good ole TIME_WAIT. aak -- Arthur Krewat VP of Engineering, Bartek, Inc. krewat@bartek.dontspamme.net Article 4966 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: simh_2.6 w/DZ11 Date: Fri, 01 Jun 2001 09:22:32 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <3B16CF6D.FE34BF8E@bartek.dontspamme.net> X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 22 Xref: nntp1.ba.best.com alt.sys.pdp10:4966 Tim is right, alas. All simh disks are stored in natural word length, in the next larger power-of-2 byte container, in little endian format. So the PDP-8 stores 12b in 16b, the PDP-11 16b in 16b, the PDP-15 18b in 32b, the VAX 32b in 32b, and the PDP-10 36b in 64b. Simh has 'endian independent' binary I/O routines (fxread and fxwrite) for dealing with endian-ness, but I don't have a big-endian system for consistent testing. On Fri, 01 Jun 2001 12:27:30 -0000, Timothy Stark wrote: >Zane H. Healy wrote: >> Are RP06 images compatible between TS10 and simh 2.6? > >No, RP06 images are not compatible between TS10 and simh 2.6. My TS10's >RP06 routine uses 5-byte format for 36-bit words like same format in >tape files. Simh 2.6 uses 64-bit values as 36-bit words so that I >believe. > >-- Tim Stark Article 5155 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp8,alt.sys.pdp11 Subject: Simh big-endian bug - fixed in 12-Jun-2001 release Date: Wed, 13 Jun 2001 09:35:52 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 10 Xref: nntp1.ba.best.com alt.sys.pdp10:5155 alt.sys.pdp8:847 alt.sys.pdp11:2346 Dave Conroy found a bug in SIMH's big-endian I/O routines, that cause them to malfunction on large block transfers. This impacts, primarily, the simulation of DECtapes, floppy disks, and fixed head disks, which buffer their data in memory. (Impacted simulators: PDP-8, PDP-11, PDP-9/15, Nova.) Please get the latest source release from http://www.tiac.net/users/mps/retro/index.html. Thanks /Bob Article 5236 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.skycache.com!Cidera!portc03.blue.aol.com!news.stealth.net!news-east.rr.com!news.rr.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Disk image for 7.03 TOPS-10 Organization: Kilonet.net Lines: 54 Message-ID: <3B2A2D7D.E4A4A9D3@bartek.dontspamme.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 15 Jun 2001 15:46:43 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 992620003 24.186.100.134 (Fri, 15 Jun 2001 11:46:43 EDT) NNTP-Posting-Date: Fri, 15 Jun 2001 11:46:43 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:5236 All who have downloaded the TOPS-10 7.03 disk image: The system name is set to "LIRICS Timesharing Gold". If you want to change this, do this: .COPY SYS:SYSNEW.EXE=SYS:SYSTEM.EXE .R FILDDT File: SYS:SYSNEW.EXE/P $$t CONFIG/ LIRIC "/LIRIC/<^J> CNFTBL+1/ S Tim "/S Tim/<^J> CNFTBL+2/ eshar "/eshar/<^J> CNFTBL+3/ ing G "/ing G/<^J> CNFTBL+4/ old "/old/<^J> <^Z> EXIT . $ is an (escape) $$t forces ASCII printout. <^J> is a Control-J (line feed) <^Z> is a Control-Z - EXIT! You need to type the first "CONFIG/", the rest will be done automatically with the ^J Text is entered inside the "/ and the ending / CONFIG and CNFTBL symbols are identical in 7.03, while in 6.03, I think it's just CONFIG. After doing a "boot rp0" and getting the BOOT> prompt, you can type "SYSNEW" and it will boot the new SYSTEM.EXE If it works do this: .REN SYS:SYSOLD.EXE=SYS:SYSTEM.EXE .REN SYS:SYSTEM.EXE=SYS:SYSNEW.EXE This will move the altered monitor in place of the original and save the original as SYSOLD.EXE, which you can always boot from if you need to. aak ps: LIRICS is Long Island Regional Instructional Computer Society It was originally a PDP-8, then a KA10, then five KS10's, serving high-schools and guidance departments across Suffolk County. Gold was one of the last KS10's purchased, and had a Megaword of memory. One college (South Hampton) used LIRICS, and possibly some Nassau schools. Many high-schoolers got their feet wet on mainframes this way - thank god it was DEC equipment :) Article 5348 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feeder.via.net!cyclone-sf.pbi.net!209.10.34.151!newsfeed.sjc.globix.net!news.sjc.globix.net!not-for-mail Message-ID: <3B33CCC0.BC257B8A@transmeta.com> From: "William H. Maddox III" X-Mailer: Mozilla 4.75 [en] (X11; U; Linux 2.2.17 i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New ... GNU/Linux ITS project References: <6k74jt8g121a0h0cr21v2lad5chb4e5ils@4ax.com> <3B32D8EB.8499BDB3@texas.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Organization: Transmeta Corporation Cache-Post-Path: palladium.transmeta.com!unknown@maddox-linux.transmeta.com X-Cache: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) Lines: 23 Date: Fri, 22 Jun 2001 15:54:56 -0700 NNTP-Posting-Host: 209.10.217.66 X-Complaints-To: news@globix.net X-Trace: news.sjc.globix.net 993250497 209.10.217.66 (Fri, 22 Jun 2001 15:54:57 PDT) NNTP-Posting-Date: Fri, 22 Jun 2001 15:54:57 PDT Xref: nntp1.ba.best.com alt.sys.pdp10:5348 Daniel Seagraves wrote: > > On Fri, 22 Jun 2001, Joseph F. Young wrote: > > > Are there tape images for ITS available that we can pull the system+utilities > > from? > > No. The ITS no longer exists in binary form, and even if it did, it's > legal status is shaky and we cannot give it out to anyone. It's the > property of MIT, the United States Department of Defense, The people who > wrote it, or some amalgamation of the three. I'm not sure. All I know > is, it's not in the public domain and I (or any of the other people who > have copies) cannot give it to anyone (legally, at least). The sources to ITS are available from http://www.swiss.ai.mit.edu/ftpdir/its/ . They are GPL'd, copyright MIT. I apologize if this this is not new info. I did not see Joseph's original message or any preceding discussion. Bill Maddox Article 5360 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.cwix.com!news.dpc.net!sakura.lunar-tokyo.net!dseagrav From: Daniel Seagraves Newsgroups: alt.sys.pdp10 Subject: Re: New ... GNU/Linux ITS project Date: Sat, 23 Jun 2001 15:07:47 -0500 Organization: Alliance Internet Technologies, LLC Lines: 11 Message-ID: References: <6k74jt8g121a0h0cr21v2lad5chb4e5ils@4ax.com> <3B32D8EB.8499BDB3@texas.net> <3B33CCC0.BC257B8A@transmeta.com> NNTP-Posting-Host: sakura.lunar-tokyo.net Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII In-Reply-To: <3B33CCC0.BC257B8A@transmeta.com> Xref: nntp1.ba.best.com alt.sys.pdp10:5360 On Fri, 22 Jun 2001, William H. Maddox III wrote: > The sources to ITS are available from > http://www.swiss.ai.mit.edu/ftpdir/its/ . > They are GPL'd, copyright MIT. Not exactly; That's only SYSTEM; and SYSDOC; - Essentially, the monitor (kernel) and docs. Useless without the rest of the system. Article 5361 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!news.maxwell.syr.edu!btnet-peer1!btnet-feed5!btnet!news.enterprise.net!not-for-mail Message-ID: <3B329012.AF10C143@enterprise.net> From: Donald Fisk Organization: The Rosen Association X-Mailer: Mozilla 4.7 [en] (X11; I; Linux 2.2.13 i586) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: New ... GNU/Linux ITS project References: <6k74jt8g121a0h0cr21v2lad5chb4e5ils@4ax.com> <3B32D8EB.8499BDB3@texas.net> <3B33CCC0.BC257B8A@transmeta.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 33 Date: Fri, 22 Jun 2001 01:23:46 +0100 NNTP-Posting-Host: 212.161.115.27 X-Complaints-To: news@enterprise.net X-Trace: news.enterprise.net 993334451 212.161.115.27 (Sat, 23 Jun 2001 23:14:11 BST) NNTP-Posting-Date: Sat, 23 Jun 2001 23:14:11 BST Xref: nntp1.ba.best.com alt.sys.pdp10:5361 Daniel Seagraves wrote: > > On Fri, 22 Jun 2001, William H. Maddox III wrote: > > > The sources to ITS are available from > > http://www.swiss.ai.mit.edu/ftpdir/its/ . > > They are GPL'd, copyright MIT. > > Not exactly; That's only SYSTEM; and SYSDOC; - Essentially, the monitor > (kernel) and docs. Useless without the rest of the system. Alan Bawden's FTP directory (ftp://ftp.ai.mit.edu/people/alan/its/) contains the source for Teco, and the Teco for Emacs as well as the ITS source and documentation, some of which makes poignant reading. Still no sign of either a Midas assembler or Maclisp or Macsyma, but I'm sure something is out there somewhere. The Midas assembler is the most important -- without that, nothing else will work. I had no direct contact with ITS at the time it was still being used, but I *have* used Genera, its immediate successor which bears more than a passing similarity to ITS. There are four things that can be done at this point: leave ITS in cryogenics, base an entirely new operating system on it (like Genera), try to bring it back as a museum curio by getting it to run on a PDP-10 emulator, or bringing it back as a living operating system by porting it to C (in which case there's some catching up to do). -- Le Hibou "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." -- G.B. Shaw Article 5320 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: "Michael L. Umbricht" Newsgroups: alt.sys.pdp10 Subject: RCS/RI DECsystems Date: Wed, 20 Jun 2001 19:08:31 -0400 Organization: The Ocean State Free-Net Message-ID: <3B312CEF.ABCB92D3@osfn.org> Reply-To: mikeu@shrimp.osfn.org X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4m) X-Accept-Language: en MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 17 Xref: nntp1.ba.best.com alt.sys.pdp10:5320 I've started to organize the directory structure where RCS stores our "work in progress" (note that our main web site is at http://www.osfn.org/rcs/ this other http://starfish.rcsri.org/rcs/ site is only for notes and info relating to our restoration projects, and these files are mostly lists of modules etc.) http://starfish.rcsri.org/rcs/DECsystem/ with subdirs for 2020 and 2065 We hold a public Open House on the third Saturday of each month. Or drop me a note if you are going to be in the Providence area, and we can arrange for a visit. Within the next couple of weeks we will have a work session to continue the restoration of the pdp-11/40 front-end in the KL. -mikeu Retro-Computing Society of RI Article 5364 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator Status - Almost a dot on KL! Date: Sun, 24 Jun 2001 03:39:37 -0000 Message-ID: Summary: User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 327 Xref: nntp1.ba.best.com alt.sys.pdp10:5364 Hello folks: Since I implemented RH20/MBA/RPx routines into my TS10 emulator, I sucessfully initialized/formatted the RP06 disk emulation file. It took a very few seconds to complete refresh processing. It is faster than my RH11 routines under my KS10 emulation! That's why I implemented transfer routines with a array of 18-bit words (block) each write/read access. Also, I fixed two instructions (CAIx and byte instructions) that were not compatible with extended addressing that cause EWB stopcodes (a bug in byte instruction during SAT initialization) and NPJ stopcodes (a bug in CAIx instruction). Now it finally print 'To automatically log-in under [1,2], type LOGIN' then EUE stopcode. Yes, it now is trying to print a dot! :-) I am working on it. EUE stopcode is occuring during CTX routines. It crashed at 1,,415557. I am figuring out why... My screen log is here: telnet localhost 5000 Trying 127.0.0.1... Connected to localhost.localdomain (127.0.0.1). Escape character is '^]'. Welcome to KL10 Emulator KL10 03-19-86 Why reload: new Date: jun-23-2001 Time: 22:22 %Memory from 4000000 to 5777777 is OFF-LINE Do you want it to be 1) ON-LINE, 2) OFF-LINE, or 3) Down? (Type #) 2 Startup option: go RPA0 first HOM block consistency error RPA0 second HOM block consistency error Do you want to initialize the HOME blocks on this unit? yes ?No structures in system search list ?No units in active swapping list %No structures in system dump list ?No structures in system search list ?No units in active swapping list %No structures in system dump list RPA0 first BAT block consistency error RPA0 second BAT block consistency error Do you want a list of bad regions? yes First block # Blocks 13064 1 Both BAT blocks are bad. Do you want to initialize the BAT blocks on this unit? yes Not normally done, are you sure? yes Initializing BAT blocks Type physical unit name to list number of bad regions ( if none, ALL if all) all RPA0() # Bad regions = 0 Disk file structures: Units not in a file structure: RPA0() Units in active swapping list: Structures in system search list: Structures in system dump list: Type physical unit name to list its parameters ( if none, ALL if all) rpa0 Parameters that require refreshing: Number of SAT blocks on unit = 73 Therefore clusters per SAT = 45656 Therefore words per SAT = 358 Number of K for swapping on unit = 0 Parameters that do not require refreshing: Zero unit ID - New ID needed Number of SAT blocks in core = 0 Type physical unit name to list its parameters ( if none, ALL if all) rpa0 Parameters that require refreshing: Number of SAT blocks on unit = 73 Therefore clusters per SAT = 45656 Therefore words per SAT = 358 Number of K for swapping on unit = 0 Parameters that do not require refreshing: Zero unit ID - New ID needed Number of SAT blocks in core = 0 Type physical unit name to list its parameters ( if none, ALL if all) Before HOME blocks are rewritten, Do you want to change any disk parameters? yes Type structure name to be defined ( if none) dskb Type names of physical units in structure (ALL if all, extra when done) rpa0 Type structure name to change its parameters ( if none, ALL if all) dskb After each printing of current value, type new value or Parameters that do not require refreshing: Number of consecutive blocks tried for on output = 30 Min = 1 Max = 262143 Number of blocks allowed for overdraw per user = 500 Min = 1 Max = 262143 Is this a private structure? (Type Y if private, N if no access restrictions) y Owner PPN is not set Do you want to change the owner PPN? ( if no) Parameters that require refreshing: Number of K for CRASH.EXE = 1536 Min = 0 Max = 4096 Blocks per cluster = 10 Min = 1 Max = 511 Therefore bits per cluster adr. = 15 Therefore blocks per super-cluster = 10 Therefore super-clusters per unit = 30780 Bits per cluster count = 12 Min = 1 Max = 18 Therefore bits per checksum = 9 Type structure name to change its parameters ( if none, ALL if all) Type physical unit name to change its parameters ( if none, ALL if all) rpa0 After each printing of current value, type new value or Parameters that require refreshing: Number of SAT blocks on unit = 8 Min = 8 Max = 81 Therefore clusters per SAT = 3848 Therefore words per SAT = 107 Number of K for swapping on unit = 0 Min = 0 Max = 4807 4807 Computed first logical block for swapping = 134672 The first logical block for swapping = 134672 Min = 12610 Max = 269332 Parameters that do not require refreshing: Zero unit ID - New ID needed TIM Number of SAT blocks in core = 1 Min = 1 Max = 8 Type physical unit name to change its parameters ( if none, ALL if all) Units in active swapping list: Do you want to change the active swapping list? yes For each class, type physical unit names (Extra when done) Class 0 rpa0 Class 1 Structures in system search list: Do you want to change the system search list? yes Type structure names for system search list (Extra when done) dskb Structures in system dump list: Do you want to change the system dump list? yes Type structure names for system dump list (Extra when done) dskb Before HOME blocks are written Type structure name for a list of its parameters ( if none, ALL if all) Type physical unit name to list its parameters ( if none, ALL if all) Do you want to change any disk parameters? ( if no) Type physical units on which to write HOME blocks (Extra when through) ( if none, ALL if all; "ALL" is normal case) all HOME blocks written Disk file structures: Private: Need refreshing: DSKB:RPA0(TIM) Units in active swapping list: RPA0(0) Structures in system search list: DSKB Structures in system dump list: DSKB Type structure name for a list of its parameters ( if none, ALL if all) Type physical unit name to list its parameters ( if none, ALL if all) Do you want to change any disk parameters? ( if no) %Need refreshing: DSKB Type structure name to be refreshed ( if none) dskb Type structure name to be refreshed ( if none) To automatically log-in under [1,2] type "LOGIN" [bells] ?Stopcode EUE, type=DEBUG, on CPU0 at 23-Jun-101 22:22:00 MUUO Flags, Opcode, AC = 004000,,000001 MUUO PC = 1,,415557 MUUO EA = 1,,513471 CPU Status Block APRID = 000000,,003720 ERA = 000001,,000001 CONI APR, = 007760,,000001 CONI PI, = 000000,,000136 CONI PAG, = 000000,,660001 DATAI PAG, = 700107,,000004 AR ARX Data Word = 000000,,000000 IO Page Fail Word = 000000,,000000 SBUS Diags: CNTRLR FNC 0 FNC 1 000000 000002,,000002 000000,,000000 000001 020002,,000002 000000,,000000 000002 040002,,000002 000000,,000000 000003 060002,,000002 000000,,000000 000004 100002,,000002 000000,,000000 000005 120002,,000002 000000,,000000 000006 140002,,000002 000000,,000000 000007 160002,,000002 000000,,000000 000010 200002,,000002 000000,,000000 000011 220002,,000002 000000,,000000 000012 240002,,000002 000000,,000000 000013 260002,,000002 000000,,000000 000014 300002,,000002 000000,,000000 000000 000000,,000000 000000 000000,,000000 000000 000000,,000000 [Continuing system] Connection closed by foreign host. [sword7@sword7-0 sword7]$ exit -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 11186 of alt.sys.pdp10: Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!newsfeed.mathworks.com!feeder.qis.net!sn-xit-02!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Timothy Stark Newsgroups: alt.sys.pdp10 Subject: Re: TS10 Emulator Status - Almost a dot on KL! Date: Tue, 26 Jun 2001 00:54:08 -0000 Message-ID: References: User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 28 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:11186 Hello folks: Well, I found a problem that cause EUE stopcode in my KL10 emulator. I traced through my 40 megs debug log file from JSR @ONCE location to the crash. It was very long execution. I learned how well does TOPS-10 prepare for the timeshare. 1. Set up EDV entry on location 112. 2. Move DDB blocks, part of SYSINI routines, etc. 3. Turn all priority interrupts for devices, etc on. 4. Start I/O devices for timeshare. 5. Turn timer on. 6. Set up the null job preparation. : It reached SETCSH routine in KLSER.MAC. It attempted to print "%% Cache is disabled" on my screen but crashed into EUE stopcode instead of. I am investiaging on the problem in the progress. Yes. it is very close to the dot. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 5403 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-02!supernews.com!postnews1.google.com!not-for-mail From: hsnewman@blkbox.com (HarrisNewman) Newsgroups: alt.sys.pdp10 Subject: Re: Simh2.6 with dz11 status Date: 26 Jun 2001 14:07:59 -0700 Organization: http://groups.google.com/ Lines: 17 Message-ID: References: <3B37B95A.E96921D8@bartek.dontspamme.net> <3B389BDB.ED1EB0F3@bartek.dontspamme.net> <3B38BBEC.B4B18F23@bartek.dontspamme.net> NNTP-Posting-Host: 207.19.97.139 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 993589679 4733 127.0.0.1 (26 Jun 2001 21:07:59 GMT) X-Complaints-To: groups-support@google.com NNTP-Posting-Date: 26 Jun 2001 21:07:59 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5403 Arthur Krewat wrote in message news:<3B38BBEC.B4B18F23@bartek.dontspamme.net>... > Cool, I'll have to check it out - any available PPN's to > login to? > Yes, log on at: telnet newman.hn.org 2020 with user id of decwar Please note it will take you into the game, but you can control c or quit to get out of it. That will take you to the command line. I haven't set up a search list (don't know how). Anyone up to a game of decwar? How bout 7:00 pm CST tonite? -Harris Article 5419 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!netnews.com!newsfeed.nyc.globix.net!news.stealth.net!news-east.rr.com!news.rr.com!cyclone.kc.rr.com!news.kc.rr.com!cyclone.austin.rr.com!cyclone2.austin.rr.com!typhoon.austin.rr.com.POSTED!not-for-mail From: "Harris S. Newman" Newsgroups: alt.sys.pdp10 Subject: Now adventure lives too Lines: 8 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Message-ID: Date: Thu, 28 Jun 2001 15:57:29 GMT NNTP-Posting-Host: 66.25.134.87 X-Complaints-To: abuse@rr.com X-Trace: typhoon.austin.rr.com 993743849 66.25.134.87 (Thu, 28 Jun 2001 10:57:29 CDT) NNTP-Posting-Date: Thu, 28 Jun 2001 10:57:29 CDT Organization: Road Runner - Texas Xref: nntp1.ba.best.com alt.sys.pdp10:5419 Well, I've added adventure to my collection. It also runs at: telnet newman.hn.org 2020 drop out of decwars and r gam:advent -HN Article 5430 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!sn-xit-02!supernews.com!postnews1.google.com!not-for-mail From: hsnewman@blkbox.com (HarrisNewman) Newsgroups: alt.sys.pdp10 Subject: Re: LOGO.MAC Date: 28 Jun 2001 18:48:06 -0700 Organization: http://groups.google.com/ Lines: 12 Message-ID: References: <3B3B67C8.58A7E88E@bartek.dontspamme.net> <9hg1l4$3uj$1@abbenay.CS.Berkeley.EDU> <3B3B96AE.4EFC6790@bartek.dontspamme.net> NNTP-Posting-Host: 207.19.97.139 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 993779286 14515 127.0.0.1 (29 Jun 2001 01:48:06 GMT) X-Complaints-To: groups-support@google.com NNTP-Posting-Date: 29 Jun 2001 01:48:06 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5430 "Zane H. Healy" wrote in message news:... > Arthur Krewat wrote: > > I put it up as > > http://pdp10.kilonet.org/files/logo.mac > > Stupid question. Any good suggestions on how to suck this into simh? > Somehow I don't want to cut and paste chunks of it into SED like I did > with the CARDIAC simulator :^) > > Zane I use the utility backwr to create .tap files.... Article 5574 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!news.maxwell.syr.edu!newsfeed.frii.net!easynews!e420r-sjo4.usenetserver.com!newsfeed.usenetserver.com!e420r-sjo2.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: TOPS 7.03 Installation Procedure for TS10 Newsgroups: alt.sys.pdp10 References: Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 10 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Mon, 09 Jul 2001 18:37:35 EDT Date: Mon, 09 Jul 2001 22:37:35 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5574 Michael Thompson wrote: > I updated Zane's TOPS 7.03 Installation Procedure for the Tim Stark's TS10 > PDP-10 emulator. > What should I do with the updated procedure? Well, if you want to send me a copy I'll replace the copy I've got up on the web at: http://www.aracnet.com/~healyzh/tops-10_install.txt Zane Article 5564 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.novia.net.MISMATCH!novia!news.dpc.net!sakura.lunar-tokyo.net!dseagrav From: Daniel Seagraves Newsgroups: alt.sys.pdp10 Subject: New ts10/DZ11 tarballs... Date: Mon, 9 Jul 2001 09:37:16 -0500 Organization: Alliance Internet Technologies, LLC Lines: 13 Message-ID: NNTP-Posting-Host: sakura.lunar-tokyo.net Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Xref: nntp1.ba.best.com alt.sys.pdp10:5564 Asuka is back up again, and I made new ts10/dz11 tarballs. ftp://sakura.lunar-tokyo.net/pub/ts10.tgz is ts10 from CVS as of this morning. Get that first. Don't compile. ftp://sakura.lunar-tokyo.net/pub/ts10-dz11-070901.tgz is the dz11 code. Unpack in the same directory as the above. Then compile. The only known bug present is that Linux clobbers XON/XOFF for some reason. I haven't had time to fix this yet. Article 5580 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: m_thompson@ids.net (Michael Thompson) Newsgroups: alt.sys.pdp10 Subject: Re: New ts10/DZ11 tarballs... Date: Tue, 10 Jul 2001 12:08:29 -0000 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit) References: MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII X-Complaints-To: newsabuse@supernews.com Lines: 25 Xref: nntp1.ba.best.com alt.sys.pdp10:5580 I tried to compile TS10 with the new version of dz11. The files /dev/dzclient.c and /dev/dzclear.c both use the include . is not in the NetBSD includes and is not part of your package. These programs compile fine without this include. /pdp11/cpu_main.c has an external reference to dz_main() in p10_Go. The linker can't resolve this reference. In article , dseagrav@sakura.lunar-tokyo.net says... > >Asuka is back up again, and I made new ts10/dz11 tarballs. > >ftp://sakura.lunar-tokyo.net/pub/ts10.tgz is ts10 from CVS as of this >morning. Get that first. Don't compile. > >ftp://sakura.lunar-tokyo.net/pub/ts10-dz11-070901.tgz is the dz11 code. >Unpack in the same directory as the above. Then compile. > >The only known bug present is that Linux clobbers XON/XOFF for some >reason. I haven't had time to fix this yet. > > > Article 5582 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed00.sul.t-online.de!newsmm00.sul.t-online.com!t-online.de!news.t-online.com!not-for-mail From: hb-usenet@city-cars-halver.de (Heiko Berges) Newsgroups: alt.sys.pdp10 Subject: Re: New ts10/DZ11 tarballs... Date: Tue, 10 Jul 2001 16:57:55 +0200 Organization: T-Online Lines: 15 Message-ID: <9if55j$6n3$05$1@news.t-online.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: news.t-online.com 994777075 05 6883 nzcgSjISSrAlSO 010710 14:57:55 X-Complaints-To: abuse@t-online.com X-Sender: 520020948171-0001@t-dialin.net User-Agent: slrn/0.9.6.2 (Linux) Xref: nntp1.ba.best.com alt.sys.pdp10:5582 On Tue, 10 Jul 2001 12:08:29 -0000, Michael Thompson wrote: >/pdp11/cpu_main.c has an external reference to dz_main() in p10_Go. > The linker can't resolve >this reference. Add dev/dz.o to the TS10_OBJS - Section in the Makefile: 27c27,28 < dev/uba.o --- > dev/uba.o \ > dev/dz.o Heiko Article 5601 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!cyclone.bc.net!cyclone-sjo1.usenetserver.com!news-out-sjo.usenetserver.com!e420r-sjo4.usenetserver.com!newsfeed.usenetserver.com!e3500-chi1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: TOPS-10 7.03 and Search Lists. Newsgroups: alt.sys.pdp10 References: Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 26 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Thu, 12 Jul 2001 14:06:29 EDT Date: Thu, 12 Jul 2001 18:06:29 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5601 Daniel Seagraves wrote: > On Wed, 11 Jul 2001, Zane H. Healy wrote: >> STRUCTURE-QUOTAS>add dskb >> %REASNC Structure dskb is not cataloged >> STRUCTURE-QUOTAS> >> > .R OPR > OPR> ENTER CATALOG > (Wait a bit) > CATALOG> INSERT STRUCTURE DSKB > ADD (unit) DSKB (type) RP06 (class) 1 > DONE > CATALOG> > That makes a DSKB structure and puts DSKB: in it. > That's how 42,2 on Rainbow Brite has a search list. That does the trick, thanks Daniel! I was really getting tired of running SETSRC every time :^) Once I did that I was able to add structure-quotas with no problem. Zane Article 5603 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!news.maxwell.syr.edu!newsfeed.frii.net!easynews!e420r-sjo4.usenetserver.com!newsfeed.usenetserver.com!e3500-chi1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: simh & TOPS-10 Question Newsgroups: alt.sys.pdp10 Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 9 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Fri, 13 Jul 2001 00:42:27 EDT Date: Fri, 13 Jul 2001 04:42:27 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5603 I've been running TOPS-10 on simh (on genuine DEC hardware) for a month or two now. While I've still got plenty of space on the RP06 I've got mounted, I'm wondering about adding a second RP06 (what can I say DSKB looks lonely). Anyway, can simh handle multiple RP06's (it appears to be able to), and how do I set the disk up under TOPS-10 (I know how to create it under simh). Zane Article 5611 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc01.blue.aol.com!news.maxwell.syr.edu!news-peer1.tiac.net!posterchild2.tiac.net!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: SIMH 2.6b Date: Fri, 13 Jul 2001 18:50:50 -0400 Organization: WWW.US.INTER.NET Lines: 13 Message-ID: NNTP-Posting-Host: ip111.bedford9.ma.pub-ip.psi.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: news-central.tiac.net 995064060 23849 38.32.79.111 (13 Jul 2001 22:41:00 GMT) X-Complaints-To: abuse@us.inter.net NNTP-Posting-Date: Fri, 13 Jul 2001 22:41:00 +0000 (UTC) X-Newsreader: Forte Agent 1.8/32.548 Xref: nntp1.ba.best.com alt.sys.pdp10:5611 There is a serious bug in the PDP-10/PDP-11 RP for BIG ENDIAN systems only. There is a direct call to fread instead of fxread (the endian-switching I/O routine). As a result, operation of the RP drives on big endian systems was either inconsistent or impossible. If you have been trying to get SIMH for the PDP-10 running on a big endian machine, please get the latest version of the simulator from http://www.tiac.net/users/mps/retro/simh26b.zip This bug was found by Peter Schorn; thank you, Peter. /Bob Supnik Article 5618 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!howland.erols.net!news-out.worldnet.att.net.MISMATCH!wn3feed!worldnet.att.net!135.173.83.71!wnfilter1!worldnet-localpost!bgtnsc06-news.ops.worldnet.att.net.POSTED!not-for-mail Reply-To: "Henry Miller" From: "Henry Miller" Newsgroups: alt.sys.pdp10 References: <3B504946.1A44AD03@aconit.org> Subject: Re: SIMH 2.6b Lines: 28 Organization: What? You mean someone organized this? X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Message-ID: Date: Sat, 14 Jul 2001 16:55:20 GMT NNTP-Posting-Host: 12.81.77.95 X-Complaints-To: abuse@worldnet.att.net X-Trace: bgtnsc06-news.ops.worldnet.att.net 995129720 12.81.77.95 (Sat, 14 Jul 2001 16:55:20 GMT) NNTP-Posting-Date: Sat, 14 Jul 2001 16:55:20 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5618 "Hans B Pufal" wrote in message news:3B504946.1A44AD03@aconit.org... > Henry Miller wrote: > > > "Bob Supnik" wrote in message > > > > If you have been trying to get SIMH for the PDP-10 running on a big > > > endian machine, please get the latest version of the simulator from > > > > > > http://www.tiac.net/users/mps/retro/simh26b.zip > > > Getting a "404 - file not found" error when referencing this page. > Hans, > Perhasp http://www.tiac.net/users/mps/retro/sources/simh26b.zip ? > Yeah, probably. Thanks, > -- HBP -HWM Article 5628 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!feed2.news.rcn.net!rcn!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: SIMH 2.6b Date: Mon, 16 Jul 2001 09:02:12 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <0cp5ltkt8542vnmjqkg4dkn6mk1p2onq5u@4ax.com> References: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 27 Xref: nntp1.ba.best.com alt.sys.pdp10:5628 As a number of people pointed out, it is http://www.tiac.net/users/mps/retro/sources/simh26b.zip I updated the code on Sunday to include the beginning of support for the Macintosh (under CodeWarrior), courtesy Louis Chretien. I still need to add the clock calibration code. /Bob Supnik On Fri, 13 Jul 2001 18:50:50 -0400, Bob Supnik wrote: >There is a serious bug in the PDP-10/PDP-11 RP for BIG ENDIAN systems >only. There is a direct call to fread instead of fxread (the >endian-switching I/O routine). As a result, operation of the RP >drives on big endian systems was either inconsistent or impossible. > >If you have been trying to get SIMH for the PDP-10 running on a big >endian machine, please get the latest version of the simulator from > >http://www.tiac.net/users/mps/retro/simh26b.zip > >This bug was found by Peter Schorn; thank you, Peter. > >/Bob Supnik Article 5640 of alt.sys.pdp10: From: Neil Franklin Newsgroups: alt.sys.pdp10 Subject: Re: DECUS tape image archives/simulator refs Date: 18 Jul 2001 23:34:48 +0200 Organization: My own Private Self Lines: 138 Message-ID: <6uwv55ly4n.fsf@chonsp.franklin.ch> References: <3B54C549.36EC4F56@trailing-edge.com> <9j40n60vpe@drn.newsguy.com> X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 18 Jul 2001 21:34:49 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 NNTP-Posting-Host: ascension.ethz.ch X-Trace: 18 Jul 2001 23:34:21 +0200, ascension.ethz.ch Path: nntp1.ba.best.com!news2.best.com!news-hog.berkeley.edu!ucberkeley!enews.sgi.com!news-zh.switch.ch!pfaff.ethz.ch!chonsp.franklin.ch!not-for-mail Xref: nntp1.ba.best.com alt.sys.pdp10:5640 Tim Shoppa writes: > >Thanks for getting the DECUS images up, that'll be a big help! > > I still have to either fix up the web server so that it knows how to serve > 36-bit files, or simply put the binaries on the FTP machine. (It *is* handy > that several search engines have already indexed the ASCII DECUS files.) > > And I know I ask this every couple of months, but maybe this time someone Never noticed it up to now. Good you today just picked a day when I have a bit of time free. > who knows will listen: *Is* there a MIME type for 36-bit binaries? No there is not. I just went and looked at the official standards document (thats what the time got used for), RFC2045, at http://www.faqs.org/rfcs/rfc2045.html This sniplet says it all: ------------- 6. Content-Transfer-Encoding Header Field Many media types which could be usefully transported via email are represented, in their "natural" format, as 8bit character or binary data. .. It is necessary, therefore, to define a standard mechanism for encoding such data into a 7bit short line format. .. mechanism := "7bit" / "8bit" / "binary" / "quoted-printable" / "base64" / ietf-token / x-token ------------- So MIME seems to have been invented as an n*8bit world only. But we do have an possibility, as they left the back door open: ------------- 6.3. New Content-Transfer-Encodings Implementors may, if necessary, define private Content-Transfer- Encoding values, but must use an x-token, which is a name prefixed by "X-", to indicate its non-standard status, e.g., "Content-Transfer- Encoding: x-my-new-encoding". ------------- So what about making such an extension, like when formats for storing tape images in 8bit files got defined? Multiple possible implementations come to mind: 1) send 36bit on 8bit channels as 2*36bit in 9*8bit. If there is a single 36bit left over pad it with 4 or 36 zero bits. This seems to be what FTP does in Image mode, if I get RFC354 right. Call this version x-36bit (or x-6bit analog to 8bit for n*8bit data). For mails this poses a problem because it will need to be additionally base64-ed, making it one encoding in an other. 2) send 12/18/24/36bit data as n times 6bits each packed in one ASCII char. Call this version x-6base64. This is more universal and already mail compatible. But it is inefficient for HTTP/Web. As coding for the latter one could use that used in base64, which first does 3*8->4*6bit and then stuffs each 6bit into one ASCII. The table for this from RFC2045: ------------- Table 1: The Base64 Alphabet Value Encoding Value Encoding Value Encoding Value Encoding 0 A 17 R 34 i 51 z 1 B 18 S 35 j 52 0 2 C 19 T 36 k 53 1 3 D 20 U 37 l 54 2 4 E 21 V 38 m 55 3 5 F 22 W 39 n 56 4 6 G 23 X 40 o 57 5 7 H 24 Y 41 p 58 6 8 I 25 Z 42 q 59 7 9 J 26 a 43 r 60 8 10 K 27 b 44 s 61 9 11 L 28 c 45 t 62 + 12 M 29 d 46 u 63 / 13 N 30 e 47 v 14 O 31 f 48 w (pad) = 15 P 32 g 49 x 16 Q 33 h 50 y The encoded output stream must be represented in lines of no more than 76 characters each. ------------- 4) define both 1) and 2) for HTTP/Web and Mail respectively, just like there exist 8bit and base64 for n*8bit machines in these 2 circumstances. 3) no new encoding at all. Simply make an standard 36bit in 8bit file format, like with (derived from?) the 36bit in 8bit tape container formats and then use an normal 8bit web server. This method would just use the file ending and resulting MIME Content-Type: for controlling the whole process. Note that in addition to an Content-Transfer-Encoding: we will any way need additional Content-Type: entries for the typical files. Looking at RFC2046 we will have to use application/* types. I suggest: application/x-sixbit for SIXBIT text text/* is not allowed for this, as that hast to be ASCII derived application/x-sextet-stream for non-specified n*6bit binaries analog to the official octet-stream for n*8bit application/x-tops10-exe for TOPS-10 .EXE, TOPS-20 analog same also for all other file types, object files, ... this analog to some of the x- types on the SGI machine I looked at There is also an RFC2048 which describes the procedure for having types added as official extensions to MIME. So if we design a standard we could even have it officially blessed. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Hacker, Unix Guru, El Eng HTL/BSc, Sysadmin, Archer, Roleplayer - Intellectual Property is Intellectual Robbery Article 5653 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!feed2.news.rcn.net!rcn!news.maxwell.syr.edu!news.stealth.net!news-east.rr.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: Simh/DZ11 Question Organization: Kilonet.net Lines: 27 Message-ID: <3B57B458.57EA9FD3@bartek.dontspamme.net> References: <3B565540.D465E7BC@bartek.dontspamme.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 20 Jul 2001 04:37:12 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 995603832 24.186.100.134 (Fri, 20 Jul 2001 00:37:12 EDT) NNTP-Posting-Date: Fri, 20 Jul 2001 00:37:12 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:5653 "Zane H. Healy" wrote: > > I've already got it going to console, however, having it go to a file would > be nice. I mean the console of simh, maybe you can redirect stderr/stdout to a file? > Is the DZ11 support still separate from the base simh package, or is it in > simh 2.6b? I don't know, Bob can tell you. I put it into 2.6a, as on pdp10.kilonet.org I have 2.6b, and wanted to hack it into place, but have yet to do so. > > If you want, I can supply you with a SYSTEM.EXE for TOPS-10 7.03 with all > > the necessary stuff already turned on, a 128 job limit, etc. etc. > > That would be cool to have a tape image of, I've not gotten brave enough to > attempt a MONGEN yet, need to find time to do some reading up on it first > anyway. I will make a .TAP image tomorrow for you ... MONGEN'ing is no problem, if you follow the MIG, but it may take more than an hour to read it and actually do the build. Of course, you will get the Y2K'd version :) aak Article 5665 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!ord2-feed1.news.digex.net!dca6-feed2.news.digex.net!intermedia!newsfeed1.cidera.com!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Message-ID: <3B5A3C93.99E0607E@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: on the edge of off topic - simulators on alphas running linux Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 22 Date: Sun, 22 Jul 2001 02:24:03 GMT NNTP-Posting-Host: 138.88.74.230 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 995768643 138.88.74.230 (Sat, 21 Jul 2001 22:24:03 EDT) NNTP-Posting-Date: Sat, 21 Jul 2001 22:24:03 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:5665 hello folks, I have been busy with life but I have had a chance to get linux installed on one of my alphas. I have compiled ts10, and run it today. Under NetBSD, I got a peak of 270Kops reported in the log. With linux on the same beast, same compiler (GCC) I am getting about the same (have not hit the same peak but the numbers are just about the same). I have snagged the DEC, err Q, alpha C compiler (and C++) and am installing thoes - usual deal, I am using a linux distro that does not support the compiler directly but I am close. Ok, so why this drivel? I am curious as to whether or not any one else has compiled and run TS10 on a similar system (meaning alpha, with Linux) and what kind of perf have you gotten? My intent is to compile the simh and ts10 stuff under linux, and see if I can get the binaries to run on NetBSD on the alphas. I just want these alphas to get numbers higher than the Intel PCs... thanks bob Article 5669 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!cyclone1.gnilink.net!spamfinder.gnilink.net!typhoon1.gnilink.net.POSTED!not-for-mail Message-ID: <3B5AC918.C405F89@bellatlantic.net> From: bad bob X-Mailer: Mozilla 4.75 [en]C-CCK-MCD BA45DSL (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10 Subject: Re: on the edge of off topic - simulators on alphas running linux References: <3B5A3C93.99E0607E@bellatlantic.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 27 Date: Sun, 22 Jul 2001 12:23:31 GMT NNTP-Posting-Host: 138.88.74.230 X-Complaints-To: business-support@verizon.com X-Trace: typhoon1.gnilink.net 995804611 138.88.74.230 (Sun, 22 Jul 2001 08:23:31 EDT) NNTP-Posting-Date: Sun, 22 Jul 2001 08:23:31 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:5669 "Zane H. Healy" wrote: > > bad bob wrote: > > My intent is to compile the simh and ts10 stuff under linux, and > > see if I can get the binaries to run on NetBSD on the alphas. > > > I just want these alphas to get numbers higher than the Intel PCs... > > thanks > > Oops, I forgot one thing, if you find a way to figure out how fast simh is > going, I'd love to know. I want to know just how fast it emulates a PDP-10 > on my DEC AlphaStation 200 4/233 :^) Also, you should be able to get simh > running on NetBSD as I'm running it on OpenBSD. Zane, I do have it running on NetBSD - but the gcc for alpha does not take advantage of alpha the way the Q compilers do - at least that is what I am tring to prove. On NetBSD, I get between 210Kops/s and 270Kops/s (according to the debug log), and my numbers so far, on the same machine with Linux, are 210 to 220. I guess this is a test of the compiler efficiency. Thanks!! bob > > Zane Article 5670 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!newspeer.cts.com!easynews!e420r-sjo4.usenetserver.com!usenetserver.com!newshub2.rdc1.sfba.home.com!news.home.com!news3.rdc1.on.home.com.POSTED!not-for-mail From: "Don Chiasson" Newsgroups: alt.sys.pdp10 References: <3B5A3C93.99E0607E@bellatlantic.net> <3B5AC918.C405F89@bellatlantic.net> Subject: Re: on the edge of off topic - simulators on alphas running linux Lines: 23 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Message-ID: Date: Sun, 22 Jul 2001 12:36:46 GMT NNTP-Posting-Host: 24.42.241.65 X-Complaints-To: abuse@home.net X-Trace: news3.rdc1.on.home.com 995805406 24.42.241.65 (Sun, 22 Jul 2001 05:36:46 PDT) NNTP-Posting-Date: Sun, 22 Jul 2001 05:36:46 PDT Organization: Excite@Home - The Leader in Broadband http://home.com/faster Xref: nntp1.ba.best.com alt.sys.pdp10:5670 "bad bob" wrote in message news:3B5AC918.C405F89@bellatlantic.net... > <> > I do have it running on NetBSD - but the gcc for alpha does not take > advantage of alpha the way the Q compilers do - at least that is what > I am tring to prove. On NetBSD, I get between 210Kops/s and 270Kops/s > (according to the debug log), and my numbers so far, on the same > machine with Linux, are 210 to 220. I guess this is a test of the > compiler efficiency. May or may not be relevant, but Kernel Korner in the current (August) issue of Linux Journal has an article on adjusting assembly language routines in the kernel to give higher performance from Alpha processors. The article talks of 40% gains, *but* that is in terms of system overhead (kernel time) for some workloads. Unfortunately, this article is not available on line - have to go to a magazine store. Don e-mail: it's not not, it's hot. Article 5686 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!news.maxwell.syr.edu!newsfeed.wirehub.nl!cyclone-sjo1.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!e3500-chi1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: TOPS-10 7.03 and 7.04 Media Lists Newsgroups: alt.sys.pdp10 Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 114 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Wed, 25 Jul 2001 00:02:23 EDT Date: Wed, 25 Jul 2001 04:02:23 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5686 I finally got tired of trying to keep track of which tapes are which and what is needed for an install, and just used the 703.BWR and 704.BWR files and poking around on http://pdp-10.trailing-edge.com/ to figure out which tapes go in which set, here is what I found, hopefully someone besides just me will find it useful :^) Zane TOPS-10 7.03 The following Information is taken from the 703.BWR file with the specific tape information added by me. 2.0 LIST OF FILES INCLUDED IN 7.03 The 7.03 release consists of the following media: 1. The distribution tape with 2 save sets: A. Monitor files B. ANF10 network files *BB-X140B-SB TOPS-10 7.03 Distribution* 2. MPE - (CPNSER) - an unbundled product *Not Found* 3. DECnet-10 - an unbundled product *Not Found* 4. The appropriate bootable tape, floppies, or DECtapes for your configuration *BB-X138B-BB TOPS-10 7.03 KS Bootable tape* 5. The full DEC-supported CUSP distribution (2 tapes) *BB-X128B-SB TOPS-10 CUSP BIN and SRC, March 1986, tape 1 of 2* *BB-FP63A-SB TOPS-10 CUSP BIN and SRC, March 1986, tape 2 of 2* 6. The customer-supported tape with 2 save sets A. Customer-supported utilities B. Customer-supported monitor files *BB-X130B-SB TOPS-10 Customer Supported CUSPs March 1986* 7. Tools tape *BB-FP64A-SB TOPS-10 Tools* TOPS-10 7.04 The following Information is taken from the 704.BWR file with the specific tape information added by me. 2.0 LIST OF DISTRIBUTED FILES This release of the TOPS-10 operating system consists of the following media: 1. The distribution tape with 2 save sets: A. Monitor files B. ANF-10 network files *BB-X140C-SB TOPS-10 7.04 Montor ANF* 2. MPE - (CPNSER) - an unbundled product *TOPS-10 7.04 MPE 10-OCT-1988* 3. DECnet-10 - an unbundled product *BB-X116C-BB TOPS-10 DECNET-10 V4.0* 4. Digital-supported CUSP distribution (2 tapes) *BB-X128C-SB TOPS-10 CUSP BIN and SRC, tape 1 of 2* *BB-FP63B-SB TOPS-10 CUSP BIN and SRC, tape 2 of 2* 5. Customer-supported tape with 2 save sets A. Customer-supported utilities B. Customer-supported monitor files *BB-X130C-SB TOPS-10 Customer Supported CUSPs* 6. Customer-Supported Tools tape *BB-FP64B-SB TOPS-10 Tools* 7. RSX20F - DECtapes (KL10-D sites) *Not Found* 8. RSX20F - Floppy diskettes (KL10-E/R sites) *Not Found* Article 5687 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!wn3feed!worldnet.att.net!209.155.233.17!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!drn From: Tim Shoppa Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 7.03 and 7.04 Media Lists Date: 25 Jul 2001 05:57:44 -0700 Organization: Newsguy News Service [http://newsguy.com] Lines: 28 Message-ID: <9jmfo806gm@drn.newsguy.com> References: NNTP-Posting-Host: p-078.newsdawg.com X-Newsreader: Direct Read News v2.80 Xref: nntp1.ba.best.com alt.sys.pdp10:5687 In article , "Zane says... > >I finally got tired of trying to keep track of which tapes are which and >what is needed for an install, and just used the 703.BWR and 704.BWR files >and poking around on http://pdp-10.trailing-edge.com/ to figure out which >tapes go in which set, here is what I found, hopefully someone besides just >me will find it useful :^) Thank you, it is interesting to see what is supposed to be there and what actually is available! > 8. RSX20F - Floppy diskettes (KL10-E/R sites) > > *Not Found* John Wilson has three RSX20F V16.00 floppies available from ftp://ftp.dbit.com/pub/pdp10/kl10/rsx20f/ (Tim Stark just alerted me to these last week, though they've been there for a couple of months now.) I'm not sure what the revision history of RSX20F is, or whether RSX20F V16.00 will work with 7.03 or 7.04. Clearly this is all important to folks with KL10's! Tim. Article 5688 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!howland.erols.net!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 7.03 and 7.04 Media Lists Organization: Kilonet.net Lines: 22 Message-ID: <3B5F28CE.F3502365@bartek.dontspamme.net> References: <9jmfo806gm@drn.newsguy.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Wed, 25 Jul 2001 20:17:17 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 996092237 24.186.100.134 (Wed, 25 Jul 2001 16:17:17 EDT) NNTP-Posting-Date: Wed, 25 Jul 2001 16:17:17 EDT Xref: nntp1.ba.best.com alt.sys.pdp10:5688 "Zane H. Healy" wrote: > > One interesting thing to note is that I didn't see any mention of RSX20F > media in the 7.03 doc's, however, I did in the 7.04. Who knows maybe I was > looking at a KS10 only 703.BWR file? Perhaps it used an older version that > the sites were expected to already have and 7.04 used a new version? Just a > WAG. I'll keep my eye out while digging through doc's to see if I spot > anything. Excerpted from 703.mco: 5.3.1 Dropping Back To 7.02 - The major operational difference between 7.02 and 7.03 on a KL10 processor is the fact that 7.03 requires a new microcode and bootstrap (BOOT). 7.02 will run with 7.03 KL microcode, bootstrap, and V15-50 RSX-20F provided Autopatch tape #9 has been applied to 7.02. Which leads me to believe that 7.03 needs at least V15-50 of RSX-20F? I would imagine that V16 supports TOPS-20, and maybe that even V16 would work with TOPS-10??? aak Article 5690 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!out.nntp.be!propagator-dallas!news-in-dallas.newsfeeds.com!in.nntp.be!easynews!e420r-sjo4.usenetserver.com!usenetserver.com!e3500-chi1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: TOPS-10 7.03 and 7.04 Media Lists Newsgroups: alt.sys.pdp10 References: <9jmfo806gm@drn.newsguy.com> <3B5F28CE.F3502365@bartek.dontspamme.net> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 77 Message-ID: <88K77.7573$zN6.5044368@e3500-chi1.usenetserver.com> X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Wed, 25 Jul 2001 21:12:36 EDT Date: Thu, 26 Jul 2001 01:12:36 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:5690 Arthur Krewat wrote: > Excerpted from 703.mco: > 5.3.1 Dropping Back To 7.02 - The major operational difference > between 7.02 and 7.03 on a KL10 processor is the fact that 7.03 > requires a new microcode and bootstrap (BOOT). 7.02 will run with > 7.03 KL microcode, bootstrap, and V15-50 RSX-20F provided Autopatch > tape #9 has been applied to 7.02. > Which leads me to believe that 7.03 needs at least V15-50 of RSX-20F? I would > imagine that V16 supports TOPS-20, and maybe that even V16 would work > with TOPS-10??? OK, I just dug out the following information on 7.04: http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/FELOAD.P11 ;3 Extend table of .STDTB values to include the one for version ; 16-00 of RSX-20F. ; 10-Feb-88 /JJF This would seem to indicate to me that V16, which can be found on John Wilson's site, is the one that is needed for 7.04. Hmm, looking a little more in the file finds the following: ;*** NOTE THAT NEWER VERSIONS OF 20F WILL NEED TO HAVE ENTRIES IN THE DTBX MACRO ;*** SIMPLY LOOK AT THE NEW RSX20F MAP FOR THE VALUE OF .STDTB. .MACRO DTBX ;;LIST OF .STDTB VALUES XP 1506,6350 ;VERSION 1506 XP 1550,6372 ;VERSION 1550 XP 1600,6372 ;VERSION 1600 Based on this I'd guess that 7.04 will will work with V15-06, V15-50, and V16-00. Here is an even stronger indicator that V16-00 is the correct one for 7.04: http://pdp-10.trailing-edge.com/pdp-10/CUSPBINSRC_1OF2_BB-X128C-SB/CUSP.BWR "7.0 RSX20F The CUSP tape contains the TOPS-10 Version 7.04 release of the RSX20F front-end software for both 1090 (DECtape) and 1091/1095 (Floppy) based systems. This software, comprising Version VA16-00 (for 1090's) and VE16-00 (for 1091/1095's) includes the following fixes and enhancements: 1. All known PCOs to RSX20F on TOPS-20 have been applied and tested on TOPS-10. 2. A problem with reading and processing the last card in a deck with CD20 card readers has been fixed. 3. A problem where in some cases the name of the microcode recorded in KLI's configuration file became corrupted has been fixed. 4. The serial numbers of all disk drives visible to the front end's RH11 disk controller are now passed to the Monitor at system startup. In addition, this version of RSX20F comes with a new, simplified, and automated installation procedure. This new procedure, called RSXINS, allows you to install RSX20F on your system in five minutes or less. To support RSXINS, two tools are included: F11 and FELOAD." Unfortunatly I can't find any info for 7.03, and the MIG is highly against providing any RSX20F version info. My guess is that it would want either 15-06 or 15-50, but not the 16-00 (which unfortunatly looks to be all that is available). Zane Article 5695 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.voicenet.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-227 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 7.03 and 7.04 Media Lists Date: Thu, 26 Jul 01 07:05:37 GMT Organization: UltraNet Communications, Inc. Lines: 63 Message-ID: <9joouj$1je$1@bob.news.rcn.net> References: <9jmfo806gm@drn.newsguy.com> X-Trace: UmFuZG9tSVZ0oFPPYBbolDKA3W+o0YNzbkmXlC0Ixye73eL/DgeajFKdCzF1OWc2 X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 26 Jul 2001 09:46:59 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:5695 In article , "Zane H. Healy" wrote: >Tim Shoppa wrote: >> Thank you, it is interesting to see what is supposed to be there and >> what actually is available! > >I figured you might be one of the people that would be interested :^) > >>> 8. RSX20F - Floppy diskettes (KL10-E/R sites) >>> >>> *Not Found* > >> John Wilson has three RSX20F V16.00 floppies available from > >> ftp://ftp.dbit.com/pub/pdp10/kl10/rsx20f/ > >> (Tim Stark just alerted me to these last week, though they've been there >> for a couple of months now.) > >> I'm not sure what the revision history of RSX20F is, or whether RSX20F >V16.00 >> will work with 7.03 or 7.04. Clearly this is all important to folks with >> KL10's! > >One interesting thing to note is that I didn't see any mention of RSX20F >media in the 7.03 doc's, That's because -20F was treated as a different product. > ...however, I did in the 7.04. One of the tasks that I never completed (but somebody told me he did do it) was to ship -20F on the magtapes; this would have been done with 7.04. > .... Who knows maybe I was >looking at a KS10 only 703.BWR file? NOpe. If there had been, it would have been called KS10.BWR. All of the commentary in the 703.BWR file would have applied to the full 7.03 ship. We tried to keep things down to a minimum of BWR files based on the package that shipped. For instance, if CPNSER (SMP) had had a separate submission to SDC, there would have been a CPNSER.BWR on the SMP tape. When the SMP tape was submitted as a part of a full monitor ship, the bewaring would have been included in the 7nn.BWR file on the monitor tape. > ... Perhaps it used an older version that >the sites were expected to already have and 7.04 used a >new version? This was the case with -20F for 7.04 submission. >Just a >WAG. I'll keep my eye out while digging through doc's to see if I spot >anything. /BAH Subtract a hundred and four for e-mail. Article 5702 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-245-95 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: TOPS-10 7.03 and 7.04 Media Lists Date: Fri, 27 Jul 01 08:04:27 GMT Organization: UltraNet Communications, Inc. Lines: 16 Message-ID: <9jrgp3$3s0$3@bob.news.rcn.net> References: <9jmfo806gm@drn.newsguy.com> <9joouj$1je$1@bob.news.rcn.net> X-Trace: UmFuZG9tSVaLb0Hi+LZ6XtTegYHoTqyKNW/STusqMxMYCZYLpXYjyeBQwQ/GLb3T X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 27 Jul 2001 10:45:55 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: nntp1.ba.best.com alt.sys.pdp10:5702 In article , "Zane H. Healy" wrote: >jmfbahciv@aol.com wrote: >> One of the tasks that I never completed (but somebody told me he >> did do it) was to ship -20F on the magtapes; this would have >> been done with 7.04. > >As near as I could figure out yesterday RSX-20F 16-00 is on one of the >magtapes. Good. Then read the BWR file for that tape, and you should have some clue about what the funny extensions are for. /BAH Subtract a hundred and four for e-mail. Article 6331 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!feeder.qis.net!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator - Porting to Cygwin/Win32 Date: Thu, 11 Oct 2001 15:00:51 -0000 Message-ID: Sender: Timothy Stark User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 20 Xref: nntp1.ba.best.com alt.sys.pdp10:6331 Hello folks: I ported my TS10 emulator to Cygwin/Win32. They were compiled successfully (No errors!). No, I did not modify any coding in TS10 emulator before attempting to compile it on Cygwin/Win32. I was able to run ts10 but it was not responsed to my keyboard! I believe that SIGIO might does not work with Cygwin/Win32. Its version is v1.1.8. I am upgrading to Cygwin v1.3.3 today and will find out... Someday, I have plans to port my TS10 emulator to C++ language so that Visual C++ will be able compiling my TS10 emulator with Winsock2 API, etc. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 6333 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!btnet-peer0!btnet-peer1!btnet!isdnet!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: TS10 Emulator - Porting to Cygwin/Solaris. Date: Thu, 11 Oct 2001 19:08:37 -0000 Message-ID: Sender: Timothy Stark References: <3BC5B731.AC7A556@bartek.dontspamme.net> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 37 Xref: nntp1.ba.best.com alt.sys.pdp10:6333 Arthur Krewat wrote: > I downloaded your recent "stable" release and found the same thing under > Solaris - I believe this is the same problem you are having with Windows. > I had to hack the signal handling (and something to do with non-blocking I/O) > on your first couple of releases, but I haven't had the chance to debug > your latest version. > I have a feeling that if you get it working under Windows, it will then > work under Solaris also. Arthur: Yeah. I debugged it and found a problem. Cygwin does not like F_SETOWN argument at fcntl statement in socket.c file. It resulted "Invalid Argument". That's why it did not work with my keyboard. Ok, I will post a problem in solaris newsgroup, etc. I was looking for cygwin newsgroups but can't found them. I checked help (man) facility on Linux and learned that F_SETOWN and F_GETOWN are not supported in SVR4 system because they are BSDism. That's why my TS10 emulator only works on two UNIX major platforms - Linux and BSD systems. When I resolved problems with F_SETOWN/F_GETOWN arguments, TS10 emulator should be work on all UNIX and Cygwins (under all Win32 systems) systems. That's why I am porting my TS10 emulator (stable and developement) to Solaris/Cygwin systems. Thank you! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 6344 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!newshub.northeast.verio.net!verio!feeder.qis.net!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: TS10 Emulator - Porting to Cygwin/Solaris. Date: Fri, 12 Oct 2001 03:40:51 -0000 Message-ID: Sender: Timothy Stark References: <3BC5B731.AC7A556@bartek.dontspamme.net> <3BC652D7.920673C3@bartek.dontspamme.net> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 28 Xref: nntp1.ba.best.com alt.sys.pdp10:6344 Arthur Krewat wrote: > Yes, this is very familiar, it was one of the things I had to change in > the original release of your code. I hope you can get to it soon, I'd love > to try it again. Yeah! I got it. I looked into Tim's archive and found my older copies of TS10. I downloaded all and put them into my ftp site (/pub/ts10/attic) directory. Also I saved them as historical archives.... In verson 3/17/2001, I found out that I_SETSIG should be work on Cygwin and Solaris. I will try that tomorrow. If it works, I will update develop and stable version and let you know that. I know that I_SETSIG does not work on Linux platforms. For Linux and BSD, I have to use F_SETOWN. For Solaris, Cygwin, etc, I have to use I_SETSIG on ioctl function. Also, I plan to add Winsock2 API to my TS10 emulator for Visual C++ compilers for async sockets. My TS10 emulator is interrupt-driven emulation system. Someday, I would get Intel C/C++ compilers to boost its performance... -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 6110 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!sunqbc.risq.qc.ca!newsfeed.telusplanet.net!news0.telusplanet.net.POSTED!not-for-mail Sender: Robert Komar From: Rob Komar Subject: Re: SIMH V2.7 released Newsgroups: alt.sys.pdp10,alt.sys.pdp11 References: <8bh9qt4186iegfvguh1tru1ml14c6pkkh2@4ax.com> User-Agent: tin/1.5.8-20010221 ("Blue Water") (UNIX) (Linux/2.4.8 (i686)) Message-ID: Lines: 11 Date: Wed, 19 Sep 2001 04:06:07 GMT NNTP-Posting-Host: 207.194.19.99 X-Trace: news0.telusplanet.net 1000872367 207.194.19.99 (Tue, 18 Sep 2001 22:06:07 MDT) NNTP-Posting-Date: Tue, 18 Sep 2001 22:06:07 MDT Xref: nntp1.ba.best.com alt.sys.pdp10:6110 alt.sys.pdp11:2554 In alt.sys.pdp10 Bob Supnik wrote: > I've released SIMH V2.7 at http://www.tiac.net/users/mps/retro/ For the past couple of nights, I haven't been able to get the web pages. www.tiac.net responds to pings, but not to http requests (at least not before I run out of patience). Is the server that busy, or does the problem lie elsewhere? Anyone else having problems getting a repsonse? Cheers, Rob Komar Article 6192 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!out.nntp.be!propagator-SanJose!news-in-sanjose!in.nntp.be!newspeer.cwnet.com!sjc1.nntp.concentric.net!newsfeed.concentric.net!global-news-master From: Seth Newsgroups: alt.sys.pdp10 Subject: Re: Info for unzipping SIMH v2.7 Date: 25 Sep 2001 19:59:10 GMT Organization: Concentric Internet Services Lines: 30 Message-ID: <9oqnme$j5r@dispatch.concentric.net> References: <3BAFD9C0.8C471F81@srv.net> <%F_r7.35$Kp2.191767@typhoon.austin.rr.com> NNTP-Posting-Host: 205.158.23.172 User-Agent: tin/pre-1.4-19990517 ("Psychonaut") (UNIX) (Linux/2.4.2 (i686)) Xref: nntp1.ba.best.com alt.sys.pdp10:6192 Harris Newman wrote: > Well, I talked too soon. I was able to get a clean compile, and it does run. > I can get the console to work just fine, but the telnet sessions just hang. > This is shown on my site: > Telnet newman.hn.org 2020 > it comes up with Welcome to the pdp10 simulator, but nothing else happens > (ctl-c, enter etc hit). > Anyway, HELP!!!!!!!1 No, that's fine :) Nothing will appear until you enable the OS to talk to your virtual DZ11 ttys. I had the same problem, I puzzled over it for quite some time until finally finding an answer buried on the web or Usenet somewhere. Assuming that you're running TOPS-10 7.03, you'll need to edit SYS:TTY.INI and add the line: TTY0-31: NAME NOTICE TEXT SPEED:9600 TYPE:VT100 This enables TTY0 through TTY31 to listen at 9600 baud with a VT100 terminal type. Good luck! > -HN -Seth Article 6210 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!howland.erols.net!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: Re: Help with SIMH Lines: 23 Message-ID: <3BB1EF1B.60D9F9B6@bartek.dontspamme.net> References: <8nds7.40085$QK.29319034@news1.sttln1.wa.home.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Wed, 26 Sep 2001 15:12:45 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1001517165 24.186.100.134 (Wed, 26 Sep 2001 11:12:45 EDT) NNTP-Posting-Date: Wed, 26 Sep 2001 11:12:45 EDT Organization: Optimum Online Xref: nntp1.ba.best.com alt.sys.pdp10:6210 Tom Grubbs wrote: > > Have Linux SIMH 2.7 running using the pdp10.kilonet.org tops10.gold 7.03. > Am looking for a way to move files into the simulator. Are there > instructions for setting up kermit or a utility for creating a BACKUP TAP > type file under Linux? There is a utility called "backwr" that writes .TAP images containing the files you supply. Do this: backwr -f a.tap file1 file2 file3 I put up backwr.tar.gz on my web site pdp10.kilonet.org at: http://pdp10.kilonet.org/files/backwr.tar.gz Backwr writes BACKUP savesets. You can then attach this .tap image in SIMH or TS10 and run BACKUP. have fun! aak Article 6215 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!uio.no!news.kth.se!not-for-mail From: Johnny Eriksson Newsgroups: alt.sys.pdp10 Subject: Re: Help with SIMH Date: 27 Sep 2001 22:20:09 +0200 Lines: 26 Sender: bygg@hector-lector.stacken.kth.se Message-ID: References: <8nds7.40085$QK.29319034@news1.sttln1.wa.home.com> <3BB1EF1B.60D9F9B6@bartek.dontspamme.net> NNTP-Posting-Host: hector-lector.stacken.kth.se Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 Xref: nntp1.ba.best.com alt.sys.pdp10:6215 "Tom Grubbs" writes: > Much closer - but still need help > > Good news the "backwr" utility produces .TAP files that I have been able to > use, the bad news is I don't seem to be able to move binary (executable > files) into the simulator. I have some binaries and have also tried ftp to > copy single files from trailing-edge as test cases. Am I missing a step? backwr should write 36 bits to each pdp10 word, from five bytes of input file, taking the low order seven bits of each, packing them left-to-right. The MSB of the fifth byte is then stuffed into the last bit. If your binary files are in some other format (some other 36 <==> 8 conversion) it might not work... I have a version of backwr somewhere, that can expand newlines to CRLFs, will put that up on the net soon. If there is a standard way (different from above) of how to store 36- bit binary files as a stream of bytes, tell me how and I will try to teach backwr how to handle them. > Thanks, > Tom --Johnny Article 6239 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news.voicenet.com!yellow.newsread.com!bad-news.newsread.com!netaxs.com!newsread.com!panix!newsmaster.cc.columbia.edu!watsun.cc.columbia.edu!fdc From: fdc@watsun.cc.columbia.edu (Frank da Cruz) Newsgroups: alt.sys.pdp10 Subject: Re: Using a REAL VT100 with simh Date: 1 Oct 2001 19:00:27 GMT Organization: Columbia University Lines: 33 Message-ID: <9paegb$kqu$1@newsmaster.cc.columbia.edu> References: <%hLt7.34042$QJ3.1202209@sjcpnn01.usenetserver.com> <9p80it$5eg$1@newsmaster.cc.columbia.edu> <3BB885F8.7D4DB51E@bartek.dontspamme.net> NNTP-Posting-Host: watsun.cc.columbia.edu X-Trace: newsmaster.cc.columbia.edu 1001962827 21342 128.59.39.2 (1 Oct 2001 19:00:27 GMT) X-Complaints-To: postmaster@columbia.edu NNTP-Posting-Date: 1 Oct 2001 19:00:27 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:6239 In article <3BB885F8.7D4DB51E@bartek.dontspamme.net>, Arthur Krewat wrote: : "Zane H. Healy" wrote: : > : > Frank da Cruz wrote: : > > A real VT100 depends heavily on super-effective, super-responsive : > > Xon/Xoff flow control with its partner. ESPECIALLY if you're : > > smooth-scrolling (but jump scrolling still needs it). : > : > The Linux box is set to do Xon/Xoff on the serial port, and it works great : > from linux. It seems to work just fine when at the simh 'console', it's : > just when telnet'd in to one of the DZ11 ports that I'm seeing problems. : > : > I'm not sure if I'm doing jump or smooth scrolling, I think it's actually : > doing smooth, I'll have to look into that. I'm still trying to get used : > to the terminal :^) : : I lose the XON/XOFF's when going from a LA100 into a Solaris box and then : telnetting into the DZ11. : : Otherwise, it works fine with Solaris/Intel, doing XON/XOFF properly. : : I somehow think that TELNET drops the ^S/^Q ... : Use C-Kermit as your Unix Telnet client: http://www.columbia.edu/kermit/ckermit.html Tell it to "set flow none" (actually you shouldn't have to tell it that; it's the default). Then it will treat ^S and ^Q as data characters and pass them right through. - Frank Article 6229 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!news.tele.dk!small.news.tele.dk!195.54.122.107!newsfeed1.bredband.com!bredband!uio.no!news.kth.se!not-for-mail From: Johnny Eriksson Newsgroups: alt.sys.pdp10 Subject: Announcement: Date: 01 Oct 2001 11:54:48 +0200 Lines: 31 Sender: bygg@hector-lector.stacken.kth.se Message-ID: NNTP-Posting-Host: hector-lector.stacken.kth.se Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/20.6 Xref: nntp1.ba.best.com alt.sys.pdp10:6229 First, there is a new version of backwr. This new version supports input files in ANSI-ASCII, with or without newline -> CRLF conversion, and also files in CORE-DUMP format. This last format may solve the problem people have had with transferring binary files. See ftp://ftp.stacken.kth.se/pub/pdp10/backwr/ I have also made a slight modification to simh: This implements, for the PDP10, a DEUNA (ethernet) card on the unibus. The mod consists of the following files: pdp10_deuna.c - interface to the pdp10 emulator. dec_deuna.c - the actual card emulating code. eth_dummy.c - ethernet backend, as in /dev/null. eth_bpf.c - ethernet backend that talks bpf. See ftp://ftp.stacken.kth.se/pub/pdp10/simh-deuna/ I would right now label this code as early beta, but is has transferred packets with some success, as in "the pdp10 has answerd arp queries". There are many things still missing from the deuna code. It should be an easy task to write pdp11_deuna.c, anyone interested? The IP stack for Tops-10 is not yet in such a shape that I would label it fit for human consumption. --Johnny Article 6234 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!feed.textport.net!newsfeed.stanford.edu!headwall.stanford.edu!HSNX.atgi.net!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!atlpnn01.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: SIMH code drop 30-Sep-2001 Newsgroups: alt.sys.pdp10,alt.sys.pdp11 References: <3BB886C2.FB2A6F8C@bartek.dontspamme.net> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 12 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Mon, 01 Oct 2001 12:45:32 EDT Date: Mon, 01 Oct 2001 16:45:32 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:6234 alt.sys.pdp11:2600 In alt.sys.pdp10 Arthur Krewat wrote: > I used to get this with simh on Sparc, during boot, it would hang, > looking like a lost interrupt. This was with TOPS-10 7.03 on Sparc/Solaris 8 > and of course, the pdp-10 version of the simulator. > I have not tried simh lately on Sparc, so I am going to try that later > today, once I get my lab back into shape. I used, I think, Beta4 on my Sparc 20/712 with Solaris 8 to setup ITS, it didn't like using an RP06 and I had to go with a RM03. Zane Article 6246 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newsfeed.mathworks.com!portc03.blue.aol.com!newsfeed.skycache.com.MISMATCH!newsfeed1.cidera.com!Cidera!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10,alt.sys.pdp11 Subject: Re: SIMH code drop 30-Sep-2001 Lines: 26 Message-ID: <3BB92941.1DB2A07A@bartek.dontspamme.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 i86pc) X-Accept-Language: en Date: Tue, 02 Oct 2001 02:42:46 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1001990566 24.186.100.134 (Mon, 01 Oct 2001 22:42:46 EDT) NNTP-Posting-Date: Mon, 01 Oct 2001 22:42:46 EDT Organization: Optimum Online Xref: nntp1.ba.best.com alt.sys.pdp10:6246 alt.sys.pdp11:2602 Bob Supnik wrote: > > I'm working with a tester on problems with the RP simulators running > PDP-11 Ultrix and other UNIX derivatives. The RP just hangs up, > apparently dropping an interrupt. > > I reworked the PDP-11 and PDP-10 RP to simulate more accurately the > behavior of multiple, concurrently operating drives. (Previously, the > simulator did not correctly provide separate copies of RPCS1<5:0> for > each drive.) The PDP-10 TU was also reworked to simulate more > accurately the error behavior of non-existent formatters. > > I've run multiple disk operations under RSTS/E, and a complete TOPS-10 > 7.03 SYSGEN, with the new code. I would appreciate help with testing > multi-drive operation under TOPS-10 and TOPS-20. Latest version of 2.7 works fine on Solaris 8/Intel, with a four-disk RP06 structure under TOPS-10 7.03. Also, compiled it under Solaris 8/Sparc (on a Sparcstation-10/612) and it also works fine there too! Same disk images as the Intel box, boots right up, even with the four-disk structure. DZ11 works fine under Solaris 8/Sparc and Intel. aak Article 6255 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator Status Date: Wed, 03 Oct 2001 14:45:11 -0000 Message-ID: Sender: Timothy Stark Summary: User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 50 Xref: nntp1.ba.best.com alt.sys.pdp10:6255 Hello folks: Stable Version: Well, I got a good idea when I realize that my major developement takes too long. I divided into stable and developement versions. I took a latest version of TS10 from Tim's archives and added many bug fixes and write operations to TS10 emulator. Also I removed unfinished VAX emulator from stable version so that it now is in developement version. I tested it and it worked fine. Then I released a stable version back to public. I now have ftp account. You can get a stable version from my new ftp site. That is KS10 emulator only at this time. Known Bugs: TOPS-20 v4.1 is working fine but... When I attempts to log off, I got PAGLCK BUGHLT message. Does anyone resolve that problem by trying that stable version? ftp://ftp.firesword7.net/pub/ts10/stable/ts10-fixed.tgz Developement Version: I still am working on device mapping and MSCP disk controller. I had rewrite entire device mapping structure and elimanated DEVICE and UNIT structures by replacing new MAP_DEVICE, individually xxx_DEVICE, etc.. It almost was finished. It took long time to complete. I intially implemented MSCP disk controller (RQDX3) with initialization sequence. I ran MicroVAX II to test it. It seems working fine. It sumbitted GO bit to execute MSCP packets according to my debug log file. I learned something about Q22 Memory Space. I learned that VMB set ringbase address as location 520 (0x150). I looked into main memory but... I tried 30000150 by dump command and found MSCP packets. There are one response slot and one command slot. I looked to its opcode. It was "Set Controller Characteristics Command" opcode. VMB am waiting for complete process from RQDX3. I was so amazed about that because Q22 Memory Space is mirror-effect when I dumped contents by VAX address (starting 30000000)! :-) Soon I will put latest version (unfinished yet) into developement directory... -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 6288 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!news-hog.berkeley.edu!ucberkeley!cyclone.bc.net!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: TECO Manual Newsgroups: alt.sys.pdp10 Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 14 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Sat, 06 Oct 2001 03:24:25 EDT Date: Sat, 06 Oct 2001 07:24:25 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:6288 I figure I'm not the only one that's been wanting some TECO doc's. Thanks to someone on the CLASSICCMP mailing list pointing out that the TECO manual available at the Sunsite PDP-11 Archive is for the PDP-10, PDP-8, PDP-11, and VAX-11, I've now got it available on the PDP-10 Emulation webpage under Applications Documentation. I've got an Adobe Acrobat copy for easy printing as well as the original text copy. http://www.aracnet.com/~healyzh/pdp10emu.html I've also got the TECO Pocket Guide for the PDP-8, PDP-10 and PDP-11. I'll be scanning it in the next time I have my scanner hooked up. Zane Article 6304 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!newshub.sdsu.edu!newspeer.cts.com!130.94.89.10.MISMATCH!news-out.spamkiller.net!propagator-la!news-in-la.newsfeeds.com!news-in.superfeed.net!tethys.csu.net!nntp.csufresno.edu!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: Write to Tape on TS10 Date: Mon, 08 Oct 2001 01:31:55 -0000 Message-ID: Sender: Timothy Stark References: User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 16 Xref: nntp1.ba.best.com alt.sys.pdp10:6304 Good news! latest version of TS10 emulator is now available in my ftp site - ftp://ftp.firesword7.net/pub/ts10/stable/ts10-fixed.tgz. It now support write operations for tape units. I decided to split my ts10 emulators into stable and development stages and released stable version with bug fixes. Thank you! -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article 6314 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!nntp.primenet.com!nntp.gblx.net!news.maxwell.syr.edu!feed2.onemain.com!feed1.onemain.com!uunet!dca.uu.net!ash.uu.net!spool0900.news.uu.net!reader0901.news.uu.net!not-for-mail Message-ID: <3BC33B1D.43111B1@trailing-edge.com> Date: Tue, 09 Oct 2001 17:59:57 -0400 From: Tim Shoppa Organization: Trailing Edge Technology X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.2 AlphaServer 1200 5/533 4MB) MIME-Version: 1.0 Newsgroups: alt.sys.pdp10,alt.sys.pdp11,vmsnet.pdp-11,comp.os.rsts Subject: Re: Copying 9-track tapes on a Vax References: <091020011746240349%lchretien@mac.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 87 NNTP-Posting-Host: 63.73.218.130 X-Trace: 1002664799 reader1.ash.ops.us.uu.net 8900 63.73.218.130 Xref: nntp1.ba.best.com alt.sys.pdp10:6314 alt.sys.pdp11:2623 vmsnet.pdp-11:3090 comp.os.rsts:119 Louis Chretien wrote: > I know VMSTPCE works, because I recreated a tape and had it read on the > PDP-11 with success. > > So my question is: how do I get from this .TPC format to a .TAP format, > that SIMH can read? Is there a VAX program already written to do this? > Because I tried to bring the file to a Mac and a PC in zip format, but > i lose the block sizes, rendering the file unuseable. I use four steps: 1. Use the Infozip "ZIP" distribution (see ftp://ftp.process.com/vms-freeware/fileserv/zip.zip and install on your VAX) with the "-V" (you have to put the option in quotes because it's upper case) to Zip the .TPC file up *with* RMS record length info included. For example: $ ZIP "-V" MYTAPE.ZIP MYTAPE.TPC 2. FTP the resulting .ZIP file to a Unix box. 3. Unzip to get the .TPC on the Unix box. The Unix box doesn't have RMS, obviously, but the resulting file will have embedded 16-bit-wide record lengths before each record. 4. Now use Bob Supnik's "mtcvtv23" utility to make a .TAP, which has 32-bit-wide record lengths before and after each tape record (unless there was a tape mark, in which case it's just a zero.) In case you don't have mtcvtv23.c, just cut and paste from below: [shoppa@mudd shoppa]$ more mtcvtv23.c /* This program converts a pre V2.3 msimulated magtape to a V2.3 magtape */ #include #include #include #include #define FLPSIZ 65536 int main (int argc, char *argv[]) { int i, k, wc; unsigned char bc[4] = { 0 }; unsigned char buf[FLPSIZ]; char *ppos, oname[256]; FILE *ifile, *ofile; if ((argc < 2) || (argv[0] == NULL)) { printf ("Usage is: verb file [file...]\n"); exit (0); } for (i = 1; i < argc; i++) { strcpy (oname, argv[i]); if (ppos = strrchr (oname, '.')) strcpy (ppos, ".tap"); else strcat (oname, ".new"); ifile = fopen (argv[i], "rb"); if (ifile == NULL) { printf ("Error opening file: %s\n", argv[i]); exit (0); } ofile = fopen (oname, "wb"); if (ofile == NULL) { printf ("Error opening file: %s\n", oname); exit (0); } printf ("Processing file %s\n", argv[i]); for (;;) { k = fread (bc, sizeof (char), 2, ifile); if (k == 0) break; wc = ((unsigned int) bc[1] << 8) | (unsigned int) bc[0]; wc = (wc + 1) & ~1; fwrite (bc, sizeof (char), 4, ofile); if (wc) { k = fread (buf, sizeof (char), wc, ifile); for ( ; k < wc; k++) buf[k] =0; fwrite (buf, sizeof (char), wc, ofile); fwrite (bc, sizeof (char), 4, ofile); } } fclose (ifile); fclose (ofile); } exit (0); } Tim. Article 12727 of alt.sys.pdp10: Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!newspump.sol.net!chcgil2-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed1.cidera.com!Cidera!sjcppf01.usenetserver.com!e420r-sjo4.usenetserver.com!usenetserver.com!sjcpnn01.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: Higher prices for Alpha processors Newsgroups: comp.unix.tru64,comp.os.vms,alt.sys.pdp10,comp.sys.dec References: <3bc2f4b4@siufuxsun02.unifr.ch> <3BC5AEA5.B544BB8C@gtech.com> <3BC5FE2C.396F328A@videotron.ca> <3BC6F8E7.339CFD92@gtech.com> <4liqstoobpn8cn82ab7r6vumh7a3jlrchb@4ax.com> <3BCDE359.FE81D17B@virgin.net> Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 16 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Wed, 17 Oct 2001 22:00:56 EDT Date: Thu, 18 Oct 2001 02:00:56 GMT Xref: dfw-artgen.news.verio.net comp.unix.tru64:6364 comp.os.vms:111204 alt.sys.pdp10:12727 comp.sys.dec:21255 In alt.sys.pdp10 sword7@speakeasy.org wrote: > My development version of TS10 emulator has complete KL10 emulation. KL10 > TOPS-10 system successfully was loaded and run on my TS10 emulator. DTE20 > communications are not completely yet because I have to develop PDP-11/40 > emulator to support KL10 emulator cocurrently. Ah, now this is the news I've been waiting for. Have you tried running TOPS-20 V7.0 on it? Personally I'm not to worried, at this time, about running TOPS-10 under KL10 emulation since we can do that on a KS10 emulator. However, TOPS-20 is a different matter, and the V7.0 tapes look to be the most complete. What devices are you currently supporting on the KL10 emulation, and what devices do you plan on supporting? Zane Article 6390 of alt.sys.pdp10: Path: nntp1.ba.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.cwix.com!sjc-peer.news.verio.net!sea-feed.news.verio.net!news.verio.net!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!Tomobiki-Cho.CAC.Washington.EDU!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Higher prices for Alpha processors Date: Thu, 18 Oct 2001 17:03:13 -0700 Organization: Networks & Distributed Computing Lines: 16 Message-ID: References: <3bc2f4b4@siufuxsun02.unifr.ch> <3BC6F8E7.339CFD92@gtech.com> <4liqstoobpn8cn82ab7r6vumh7a3jlrchb@4ax.com> <3BCDE359.FE81D17B@virgin.net> <3BCE4844.15CEFB55@bartek.dontspamme.net> <0Bsz7.194177$Rb.6653170@sjcpnn01.usenetserver.com> <3bcec9c9$1_2@news.iglou.com> NNTP-Posting-Host: tomobiki-cho.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1003449797 10688 (None) 140.142.17.35 X-Complaints-To: help@cac.washington.edu NNTP-Posting-User: metka In-Reply-To: Xref: nntp1.ba.best.com alt.sys.pdp10:6390 On Thu, 18 Oct 2001, Zane H. Healy wrote: > Douglas H. Quebbeman wrote: > > "Zane H. Healy" wrote in message > >> course I'm still hoping for good news on the KLH10 emulator as was hinted > > Is that a typo, or are you referring to Ken (L.) Harrington's emulator? > It's no typo. Unfortunatly I don't have any details at all. There is good news about KLH10. Watch this newsgroup later this month. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article 6472 of alt.sys.pdp10: Path: nntp1.ba.best.com!news1.best.com!news2.best.com!news.maxwell.syr.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail From: spedraja@ono.com (Sergio Pedraja) Newsgroups: alt.sys.pdp10,alt.sys.pdp8 Subject: Re: Local Connection Between one PDP-8 and one PDP-10 and TSS/8 Date: 28 Oct 2001 23:41:19 -0800 Organization: http://groups.google.com/ Lines: 50 Message-ID: References: <9rbjpb$1p5$2@bob.news.rcn.net> <9ri73l$55t$1@slb2.atl.mindspring.net> NNTP-Posting-Host: 195.53.189.245 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1004341279 30831 127.0.0.1 (29 Oct 2001 07:41:19 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 29 Oct 2001 07:41:19 GMT Xref: nntp1.ba.best.com alt.sys.pdp10:6472 alt.sys.pdp8:1001 Hello, David: > > PT08's. That is the interface for the ASR33 teletype. Right ? > Yes. It also had a RS-232 option. > Seems to have same IOT as standard console port. Normally they were > Line IOT > 1 40/41 ; 2 42/43 ; 3 44/45 ; 4 46/47; 5 11/12 > > Special model for 8/S that went inside an ASR-33 (PT08A) used > 03/04 (normal console IOT). I have in mind the SIMH emulation of the PDP-8/E. It emulates one KL8E and four KL8JA, with KSR-33 everyone. If I understand good, don't worry to have for my purposes KL8JA's or PT08's. The unique difference appears to be that one supports KSR33, and the PT08 the ASR33. Right ? Could be needed to emulate the PT08, KL8JA, KSR33 and ASR33 in different ways ? Is it sufficient with the KL8JA emulation in functional terms ? > They were about a half row double high in a 19" rack mount. Attached to > the external negative bus on original 8, 8/S and 8/I. I have one but > have not tried it yet. How many you can have is the external bus limit > on number of devices/cable length. What was the usual limit ? TSS/8 appears capable to use 32 of them. Was there a limit in IOT's that could be used ? Was this applied to the KL8JA too ? > I have print set but it does not have programming information other than > table above. It uses a W706/W707 which I assume is functionally the > same as the M706/M707 in an 8/I but negative bus with RTL IC's. > From my small computer handbook. > The 685 is simply a switch which allows the 681 to be connected to any one > of 64 teletype lines. It has the IOT's for it, a little long to type > right now, email if you need it. This is very interesting. It appears to be more easy to implement than the PT08's... (?) Now I only need a little explain about how managed TSS/8 the PT08 or 681 connection. In the manual says something about direct support for this last device. > If you think the print set will be of use I can put it on the scan pile. It can be interesting to learn something more about this matter, of course. Can you do it ? Thanks a lot ! Greetings Sergio Article 12877 of alt.sys.pdp10: Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!dca6-feed1.news.digex.net!intermedia!cyclone2.usenetserver.com!usenetserver.com!news01.optonline.net!news02.optonline.net.POSTED!not-for-mail From: Arthur Krewat Newsgroups: alt.sys.pdp10 Subject: TS10 success on Solaris Organization: Kilonet.net Lines: 24 Message-ID: <3BD9C562.35525D6B@bartek.dontspamme.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.78 [en] (X11; U; SunOS 5.8 i86pc) X-Accept-Language: en Date: Fri, 26 Oct 2001 20:22:51 GMT NNTP-Posting-Host: 24.186.100.134 X-Trace: news02.optonline.net 1004127771 24.186.100.134 (Fri, 26 Oct 2001 16:22:51 EDT) NNTP-Posting-Date: Fri, 26 Oct 2001 16:22:51 EDT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:12877 Tim, I got TS10 working on Solaris Intel by doing the following: Removing the check for bytes=0 when reading from STDIN. For whatever reason, read was returning 0 for STDIN and would cause TS10 to exit. Removing the setup for ~(ICANON || ECHO) and removing the tcgetattr and tcsetattr setups. I also removed the echo of each character read from STDIN. Therefor, the net result is that TS10 gets an entire line at once. This is no big deal, and works quite well, since you don't have to pick up backspace/delete yourself. The only remaining thing is after doing a "boot mta0:", then telnetting into port 5000, I have to hit a few keys before output is sent. Such that, I have to key in "/tm02" and after hitting the second or third character, finally, I get the BTM> prompt. great! aak Article 13197 of alt.sys.pdp10: Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!news.maxwell.syr.edu!hub1.nntpserver.com!nntp-relay.ihug.net!ihug.co.nz!telocity-west!TELOCITY!sn-xit-03!sn-xit-01!sn-xit-02!supernews.com!postnews1.google.com!not-for-mail From: jdelgado@saxondesigns.com (John Delgado) Newsgroups: alt.sys.pdp10 Subject: KLH10 diffs for Solaris 8 & SUN C compiler Date: 15 Nov 2001 08:22:57 -0800 Organization: http://groups.google.com/ Lines: 453 Message-ID: <89ab8ed9.0111150822.74f28050@posting.google.com> NNTP-Posting-Host: 171.78.153.104 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1005841377 15073 127.0.0.1 (15 Nov 2001 16:22:57 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 15 Nov 2001 16:22:57 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:13197 These changes are not fully tested, so use at your own risk; these changes were made solely to ensure compilation. No guarantees or accuracies are expressed or implied :-) The following files were updated: Mk-solsparc.mk dpni20.c osdnet.c osdnet.h Changes were made to support the SUN C compiler, multiple cast changes, struct references, removing the "#warning", and #if'ed out calls to include sys/ethernet.h and netinet/arp.h Many thanks to Ken and others for making this available. It sure brought back lots of memories of a past life. ------------------------------ CUT HERE --------------------------------- diff -rc src/Mk-solsparc.mk src.orig/Mk-solsparc.mk *** src/Mk-solsparc.mk Tue Nov 13 19:59:53 2001 --- src.orig/Mk-solsparc.mk Sat Nov 10 16:29:00 2001 *************** *** 22,30 **** # Build definitions # Force use of GCC; Solaris CC still had bugs as of last check. # LIBS are needed only for dpni20, sigh. ! #CC=gcc ! #CFLAGS = -c -g -O -O2 ! #LIBS = -lsocket -lnsl SRC = ../../src CFLAGS_LINT = -ansi -pedantic -Wall -Wshadow \ -Wstrict-prototypes -Wmissing-prototypes \ --- 22,30 ---- # Build definitions # Force use of GCC; Solaris CC still had bugs as of last check. # LIBS are needed only for dpni20, sigh. ! CC=gcc ! CFLAGS = -c -g -O -O2 ! LIBS = -lsocket -lnsl SRC = ../../src CFLAGS_LINT = -ansi -pedantic -Wall -Wshadow \ -Wstrict-prototypes -Wmissing-prototypes \ *************** *** 31,44 **** -Wmissing-declarations -Wredundant-decls ## Alternative definitions using SUN's own compiler ! CC=/opt/SUNWspro/bin/cc ! CFLAGS = -c -fast ! LIBS = -lsocket -lnsl -lrt ! CONFFLAGS_AUX=-DWORD10_USEHUN=1 # Source definitions ! CENVFLAGS = -DCENV_CPU_SPARC=1 -DCENV_SYS_SOLARIS=1 -DNETIF_HAS_SALEN=0 ! #CONFFLAGS_AUX= -DWORD10_USEGCCSPARC=1 # Targets --- 31,44 ---- -Wmissing-declarations -Wredundant-decls ## Alternative definitions using SUN's own compiler ! #CC=/opt/SUNWspro/bin/cc ! #CFLAGS = -c -g -O ! #LIBS = -lsocket -lnsl ! #CONFFLAGS_AUX=-DWORD10_USEHUN=1 # Source definitions ! CENVFLAGS = -DCENV_CPU_SPARC=1 -DCENV_SYS_SOLARIS=1 ! CONFFLAGS_AUX= -DWORD10_USEGCCSPARC=1 # Targets diff -rc src/dpni20.c src.orig/dpni20.c *** src/dpni20.c Tue Nov 13 14:13:30 2001 --- src.orig/dpni20.c Sat Nov 10 16:29:01 2001 *************** *** 1291,1297 **** /* Found it! ife now points to matching entry */ if (!ife->ife_gotea) { ! if (!osn_ifeaget(-1, ife->ife_name, ife->ife_ea, (unsigned char *)NULL)) { error("ARP MyReply failed, no E/N addr for %s", ife->ife_name); return FALSE; } --- 1291,1297 ---- /* Found it! ife now points to matching entry */ if (!ife->ife_gotea) { ! if (!osn_ifeaget(-1, ife->ife_name, ife->ife_ea, (char *)NULL)) { error("ARP MyReply failed, no E/N addr for %s", ife->ife_name); return FALSE; } *************** *** 1325,1335 **** arp.arp_op = htons(ARPOP_REPLY); /* Type REPLY */ /* Sender hdw addr and IP addr for host platform */ ! memcpy((char *)&arp.arp_sha, ife->ife_ea, ETHER_ADRSIZ); memcpy((char *)arp.arp_spa, ife->ife_ipchr, IP_ADRSIZ); /* Target hdw addr and IP addr for emulated 20 */ ! memcpy((char *)&arp.arp_tha, (char *)&ihost_ea, ETHER_ADRSIZ); memcpy((char *)arp.arp_tpa, (char *)&ehost_ip, IP_ADRSIZ); /* Now build raw packet. Do it this way to avoid potential --- 1325,1335 ---- arp.arp_op = htons(ARPOP_REPLY); /* Type REPLY */ /* Sender hdw addr and IP addr for host platform */ ! memcpy((char *)arp.arp_sha, ife->ife_ea, ETHER_ADRSIZ); memcpy((char *)arp.arp_spa, ife->ife_ipchr, IP_ADRSIZ); /* Target hdw addr and IP addr for emulated 20 */ ! memcpy((char *)arp.arp_tha, (char *)&ihost_ea, ETHER_ADRSIZ); memcpy((char *)arp.arp_tpa, (char *)&ehost_ip, IP_ADRSIZ); /* Now build raw packet. Do it this way to avoid potential diff -rc src/osdnet.c src.orig/osdnet.c *** src/osdnet.c Thu Nov 15 09:30:52 2001 --- src.orig/osdnet.c Sat Nov 10 16:29:10 2001 *************** *** 572,578 **** } return FALSE; #else ! return osn_ifeaget(-1, ifnam, eap, (unsigned char *)NULL); #endif } ^L --- 572,578 ---- } return FALSE; #else ! return osn_ifeaget(-1, ifnam, eap, (char *)NULL); #endif } ^L *************** *** 840,846 **** } /* Have IP address, now do ARP lookup hackery */ ! if (!osn_arp_look((struct in_addr *)ipchr, (struct ether_addr *)eap)) { syserr(errno,"Can't find EN addr for \"%s\" %s using ARP", ifnam, ip_adrsprint(ipstr, ipchr)); return FALSE; --- 840,846 ---- } /* Have IP address, now do ARP lookup hackery */ ! if (!osn_arp_look(ipchr, eap)) { syserr(errno,"Can't find EN addr for \"%s\" %s using ARP", ifnam, ip_adrsprint(ipstr, ipchr)); return FALSE; *************** *** 1177,1183 **** return TRUE; #else ! error("\"%s\" could not set ether addr - unimplemented OS code", ifnam); return FALSE; #endif --- 1177,1183 ---- return TRUE; #else ! # warning "Unimplemented OS routine osn_ifeaset()" error("\"%s\" could not set ether addr - unimplemented OS code", ifnam); return FALSE; #endif *************** *** 1230,1236 **** return TRUE; #else ! error("\"%s\" could not %s multicast addr - unimplemented OS code", ifnam, (delf ? "delete" : "add")); return FALSE; --- 1230,1236 ---- return TRUE; #else ! # warning "Unimplemented OS routine osn_ifmcset()" error("\"%s\" could not %s multicast addr - unimplemented OS code", ifnam, (delf ? "delete" : "add")); return FALSE; *************** *** 1944,1950 **** /* Attach to specific unit */ if (dlattachreq(&dc, (long)ppa) || dlokack(&dc)) ! efatal(1, dc.dc_ebuf); /* Bind */ #if OSN_USE_IPONLY /* Note using IP SAP */ --- 1944,1950 ---- /* Attach to specific unit */ if (dlattachreq(&dc, (long)ppa) || dlokack(&dc)) ! efatal(1, dc->dc_ebuf); /* Bind */ #if OSN_USE_IPONLY /* Note using IP SAP */ *************** *** 1953,1959 **** if (dlbindreq(&dc, FAKESAP, 0, DL_CLDLS, 0) #endif || dlbindack(&dc)) ! efatal(1, dc.dc_ebuf); #if OSN_USE_IPONLY /* Apparently only needed for this */ /* Do stuff for "fastpath" which may be needed to allow header to --- 1953,1959 ---- if (dlbindreq(&dc, FAKESAP, 0, DL_CLDLS, 0) #endif || dlbindack(&dc)) ! efatal(1, dc->dc_ebuf); #if OSN_USE_IPONLY /* Apparently only needed for this */ /* Do stuff for "fastpath" which may be needed to allow header to *************** *** 1960,1969 **** be included in data buffer rather than separate control struct. */ if (dlfastpathon(&dc, 0) < 0) ! efatal(1, dc.dc_ebuf); #elif 0 /* !OSN_USE_IPONLY */ /* Apparently not needed */ if (dlfastpathon(&dc, FAKESAP) < 0) ! efatal(1, dc.dc_ebuf); #endif /* Set up various mode & flag stuff */ --- 1960,1969 ---- be included in data buffer rather than separate control struct. */ if (dlfastpathon(&dc, 0) < 0) ! efatal(1, dc->dc_ebuf); #elif 0 /* !OSN_USE_IPONLY */ /* Apparently not needed */ if (dlfastpathon(&dc, FAKESAP) < 0) ! efatal(1, dc->dc_ebuf); #endif /* Set up various mode & flag stuff */ *************** *** 1989,2006 **** */ if ((dlpromisconreq(&dc, DL_PROMISC_PHYS) < 0) || dlokack(&dc) < 0) ! efatal(1, dc.dc_ebuf); # endif /* Evidently must explicitly ask for promiscuous SAPs */ if (dlpromisconreq(&dc, DL_PROMISC_SAP) < 0 || dlokack(&dc) < 0) ! efatal(1, dc.dc_ebuf); /* And multicast too!? To quote tcpdump, ** "you would have thought promiscuous would be sufficient" */ if (dlpromisconreq(&dc, DL_PROMISC_MULTI) < 0 || dlokack(&dc) < 0) ! efatal(1, dc.dc_ebuf); #endif /* !OSN_USE_IPONLY */ /* Find the physical ethernet address of the interface we got. --- 1989,2006 ---- */ if ((dlpromisconreq(&dc, DL_PROMISC_PHYS) < 0) || dlokack(&dc) < 0) ! efatal(1, dc->dc_ebuf); # endif /* Evidently must explicitly ask for promiscuous SAPs */ if (dlpromisconreq(&dc, DL_PROMISC_SAP) < 0 || dlokack(&dc) < 0) ! efatal(1, dc->dc_ebuf); /* And multicast too!? To quote tcpdump, ** "you would have thought promiscuous would be sufficient" */ if (dlpromisconreq(&dc, DL_PROMISC_MULTI) < 0 || dlokack(&dc) < 0) ! efatal(1, dc->dc_ebuf); #endif /* !OSN_USE_IPONLY */ /* Find the physical ethernet address of the interface we got. *************** *** 2008,2015 **** ** correct packet filter (sigh). */ if (dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0 ! || dladdrack(&dc, (unsigned char *)curea) < 0) ! efatal(1, dc.dc_ebuf); /* HACK HACK -- see if ethernet addr already given, and if so, ** try to set it if different. --- 2008,2015 ---- ** correct packet filter (sigh). */ if (dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0 ! || dladdrack(&dc, (char *)curea) < 0) ! efatal(1, dc->dc_ebuf); /* HACK HACK -- see if ethernet addr already given, and if so, ** try to set it if different. *************** *** 2020,2036 **** char new[OSN_EASTRSIZ]; /* Attempt to set our EN addr */ ! eth_adrsprint(old, (unsigned char *)curea); ! eth_adrsprint(new, (unsigned char *)&osnpf->osnpf_ea); ! if (dlsetaddrreq(&dc, (unsigned char *)&osnpf->osnpf_ea) < 0 || dlokack(&dc) < 0) ! efatal(1, dc.dc_ebuf); /* Double-check by fetching new addr again and using it */ if (dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0 ! || dladdrack(&dc, (unsigned char *)curea) < 0) ! efatal(1, dc.dc_ebuf); if (ea_cmp(&osnpf->osnpf_ea, curea) == 0) { dbprintln("\"%s\" E/N addr changed: Old=%s New=%s", --- 2020,2036 ---- char new[OSN_EASTRSIZ]; /* Attempt to set our EN addr */ ! eth_adrsprint(old, (char *)curea); ! eth_adrsprint(new, (char *)&osnpf->osnpf_ea); ! if (dlsetaddrreq(&dc, (char *)&osnpf->osnpf_ea) < 0 || dlokack(&dc) < 0) ! efatal(1, dc->dc_ebuf); /* Double-check by fetching new addr again and using it */ if (dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0 ! || dladdrack(&dc, (char *)curea) < 0) ! efatal(1, dc->dc_ebuf); if (ea_cmp(&osnpf->osnpf_ea, curea) == 0) { dbprintln("\"%s\" E/N addr changed: Old=%s New=%s", *************** *** 2070,2076 **** /* Filter on our IP addr */ pf = pfbuild(arg, &osnpf->osnpf_ip.ia_addr); ! if (strioctl(dc.dc_fd, PFIOCSETF, sizeof(*pf), (char *)pf) < 0) esfatal(1, "PFIOCSETF failed"); } --- 2070,2076 ---- /* Filter on our IP addr */ pf = pfbuild(arg, &osnpf->osnpf_ip.ia_addr); ! if (strioctl(dc->dc_fd, PFIOCSETF, sizeof(*pf), (char *)pf) < 0) esfatal(1, "PFIOCSETF failed"); } *************** *** 2092,2099 **** dc.dc_fd = fd; if ((dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0) ! || (dladdrack(&dc, (unsigned char *)eap) < 0)) { ! efatal(1, dc.dc_ebuf); ea_clr(eap); return FALSE; } --- 2092,2099 ---- dc.dc_fd = fd; if ((dladdrreq(&dc, DL_CURR_PHYS_ADDR) < 0) ! || (dladdrack(&dc, (char *)eap) < 0)) { ! error(1, dc->dc_ebuf); ea_clr(eap); return FALSE; } *************** *** 2194,2206 **** dc->dc_ctl.maxlen = MAXDLBUF; dc->dc_ctl.len = 0; ! dc->dc_ctl.buf = (char *)dc->dc_buf; if (flagsp) *flagsp = 0; #if DL_MAXWAIT signal(SIGALRM, sigalrm); ! if (alarm(DL_MAXWAIT) < 0) { esfatal(1, "%s: alarm set", caller); } #endif --- 2194,2206 ---- dc->dc_ctl.maxlen = MAXDLBUF; dc->dc_ctl.len = 0; ! dc->dc_ctl.buf = dc->dc_buf; if (flagsp) *flagsp = 0; #if DL_MAXWAIT signal(SIGALRM, sigalrm); ! if (alarm(MAXWAIT) < 0) { esfatal(1, "%s: alarm set", caller); } #endif *************** *** 2356,2362 **** static int dladdrack(struct dlpictx *dc, unsigned char *addr) { - int len; union DL_primitives *dlp; unsigned char *ucp; --- 2356,2361 ---- diff -rc src/osdnet.h src.orig/osdnet.h *** src/osdnet.h Thu Nov 15 09:20:03 2001 --- src.orig/osdnet.h Sat Nov 10 16:29:10 2001 *************** *** 118,130 **** # include # include # include - #if !CENV_SYS_SOLARIS # include - #endif # include /* For packet filtering */ - #if !CENV_SYS_SOLARIS # include /* For ARP stuff */ - #endif # include #elif KLH10_NET_PFLT --- 118,126 ---- Article 13096 of alt.sys.pdp10: Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!newsfeed.mathworks.com!portc01.blue.aol.com!newsfeed.skycache.com.MISMATCH!newsfeed1.cidera.com!Cidera!sjcppf01.usenetserver.com!usenetserver.com!sn-xit-04!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: dumb question - TS10 Date: Tue, 13 Nov 2001 13:28:48 -0000 Message-ID: Sender: Timothy Stark References: <3BF07857.44AE83B2@bellatlantic.net> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 16 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:13096 bad bob wrote: > I am apparently booting from the tops10 7.03 disk image from Tim > Shoppa's site. I think that is where I got it....telnet'd to 5000 > and 5001 asks for a password. Look at in pdp10/fe.c for password. You need change KLINK password in fe.c and re-compile it. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article: 15293 of alt.sys.pdp10 Path: sea-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!nntp.abs.net!feeder.qis.net!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator w/Intel C++ 5.0 compiler Date: Fri, 18 Jan 2002 02:46:43 -0000 Message-ID: Sender: Timothy Stark User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.20 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 32 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15293 Hello folks: I have a copy of 30-day eval Intel C++ 5.0 compiler today. I tried it with my TS10 emulator. It resulted much faster than using GCC compiler! With stable version, it complaint about inline that cause mass undefined symbols during link. I had removed all 'inline' from them and successfully compiled it into 'ts10'. I ran it and looked into debug.log. I was so impressed. Now it is 9.8 MIPS on 1.9 GHz P4. I studied -help information and tried with -tpo option. (Multi-file optimization). I successfully compiled it by using -tpo. It cause big boost exceeding 10 MIPS! Now it is 12 MIPS! Tested with TOPS-10 7.04 on TS10 emulator (stable version): Compilation options Result ----------------------------- gcc -O3 8.3 MIPS icc -O3 9.8 MIPS icc -O3 -tpo 12 MIPS Now I am satisfied with Intel C++ compiler for better optimization, etc. Some day, I will buy Intel C++ compiler after 30 day eval. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article: 15271 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!news.maxwell.syr.edu!sn-xit-03!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: Re: CFV: alt.sys.pdp10.notvax Date: Wed, 16 Jan 2002 13:16:39 -0000 Message-ID: Sender: Timothy Stark References: <3c4570b0$1_3@news.iglou.com> User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.20 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 27 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15271 Douglas H. Quebbeman wrote: > "Rich Alderson" wrote in message > news:mddr8ot4oq8.fsf@panix2.panix.com... >> >> Or Tim will stop developing the PDP-10 (that is, the useful) part of TS10. > I surely hope this does *not* come to pass... Well, I still am developing the PDP-10 emulator but I have some problems with my VAX emulator (two serious bugs had been found recently). After bugs had been fixed, NetBSD/vax now boots normally and OpenVMS still crashes the same. :-( (Two bugs in memory access and recovery queues during fault) There is a copy of stable version of PDP10 emulator on my FTP site. (ftp://ftp.firesword7.net/pub/ts10/stable) Thank you for your patience. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article: 16523 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newsfeed.cwix.com!chnws02.ne.ipsvc.net!cyclone.ne.ipsvc.net!24.128.8.70!typhoon.ne.ipsvc.net.POSTED!not-for-mail From: francini1026@mac.com (John J Francini) Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Unpacking my 15-year old office boxes generates memory refreshes Message-ID: References: <1bpu2mlv35.fsf@cs.nmsu.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-transfer-encoding: 8bit X-Newsreader: Yet Another NewsWatcher 2.4.0 Lines: 68 Date: Sat, 09 Mar 2002 04:46:05 GMT NNTP-Posting-Host: 66.31.66.54 X-Complaints-To: abuse@mediaone.net X-Trace: typhoon.ne.ipsvc.net 1015649165 66.31.66.54 (Fri, 08 Mar 2002 23:46:05 EST) NNTP-Posting-Date: Fri, 08 Mar 2002 23:46:05 EST Organization: ATT Broadband Xref: dfw-artgen alt.sys.pdp10:16523 alt.folklore.computers:83683 In article , Rich Alderson wrote: >Brian Inglis writes: > >> On 4 Mar 2002 22:55:05 GMT, "Zane H. Healy" >> wrote: > >>> As of a year ago I think *ALL* of the Macintosh OCR software was several >>> years out of date. I was very surprised to find a MacOS X version of a OCR >>> software package when I was at a computer store over the weekend. I'm >>> wondering how well it might work (unfortunatly I don't remember which >>> product it was). > >> Check the copyright date on commercial software to avoid paying full price >> for stuff that should be in a remainder bin. > >Funny, I got e-mail from Aladdin Systems/Digital River a couple of weeks back >announcing a special deal on Omnipage Pro X, a brand spanking new release of >the old Caere product for Mac OS X, with new features added to a pretty good >piece of OCR software. > >I've been very happy with the previous version, which was not "several years >out of date" when I bought the upgrade for XKL a couple of years back. > >What's the problem? Think nothing interesting gets written for the Mac? > To answer the specific question asked some 15 or so replies back: Macs have no problem displaying or printing monospaced text files. They have even less problems now that MacOS X is out, since it's UNIX under the hood. You can either look at a file in TextEdit (a GUI-based text file viewer) or just open a Terminal window and use standard Unix tools to deal with the file. As to the "nothing interesting" question: Apple and the Mac have had quite a resurgence in the past few years. And while there's still a lot of software that's consigned to the "runs in Classic ghetto (old-style MacOS running as a process under MacOS X in a virtual machine style emulator)", there's more and more stuff coming out every day for MacOS X. If you're a Unix fan, there's a project that's been porting literally hundreds of Open Source programs to Darwin (the underlying FreeBSD/Mach UNIX underneath MacOS X), along with thousands of traditional Mac titles that now run on MacOS X. And back to PDP-10 land: with some help from a former TOPS-10 developer I work with, the KLH-10 emulator now runs, in single-process form, under MacOS X. (Multi-process form will have to wait either until Darwin fully supports SysV shared memory or I or someone else converts KLH to use POSIX shared memory.) john -- John Francini Subtract Ten Twenty-Six for e-mail. +------------------------------------------------------------------------------+ | "I have come to the conclusion that one useless man is called a disgrace; | | that two or more are called a law firm; and that three or more become | | a Congress. And by God I have had _this_ Congress!" | | -- John Adams | +------------------------------------------------------------------------------+ Article: 15293 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!nntp.abs.net!feeder.qis.net!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: sword7@speakeasy.org Newsgroups: alt.sys.pdp10 Subject: TS10 Emulator w/Intel C++ 5.0 compiler Date: Fri, 18 Jan 2002 02:46:43 -0000 Message-ID: Sender: Timothy Stark User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.20 (i686)) X-Complaints-To: newsabuse@supernews.com Lines: 32 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15293 Hello folks: I have a copy of 30-day eval Intel C++ 5.0 compiler today. I tried it with my TS10 emulator. It resulted much faster than using GCC compiler! With stable version, it complaint about inline that cause mass undefined symbols during link. I had removed all 'inline' from them and successfully compiled it into 'ts10'. I ran it and looked into debug.log. I was so impressed. Now it is 9.8 MIPS on 1.9 GHz P4. I studied -help information and tried with -tpo option. (Multi-file optimization). I successfully compiled it by using -tpo. It cause big boost exceeding 10 MIPS! Now it is 12 MIPS! Tested with TOPS-10 7.04 on TS10 emulator (stable version): Compilation options Result ----------------------------- gcc -O3 8.3 MIPS icc -O3 9.8 MIPS icc -O3 -tpo 12 MIPS Now I am satisfied with Intel C++ compiler for better optimization, etc. Some day, I will buy Intel C++ compiler after 30 day eval. -- Tim Stark -- Timothy Stark <>< Inet: sword7@speakeasy.org -------------------------------------------------------------------------- "For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. Amen." -- John 3:16 (King James Version Bible) Article: 15364 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!feed.news.qwest.net!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!btnet-peer1!btnet-peer0!btnet-peer!btnet!xara.net!gxn.net!news.gradwell.net!newsfeed.media.kyoto-u.ac.jp!logbridge.uoregon.edu!panix!yellow.newsread.com!netaxs.com!newsread.com!newsfeeds-atl2!e420r-atl1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: Re: Hercules under the pdp10/20 Newsgroups: alt.sys.pdp10 References: Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 15 Message-ID: <51N18.567$h31.146654@e420r-atl1.usenetserver.com> X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Thu, 17 Jan 2002 22:55:45 EST Date: Fri, 18 Jan 2002 03:55:45 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15364 HarrisNewman wrote: > Yeah, now I've gone too far. I got a new job, this time in a IBM > mainframe shop. Well, being a nerd, and a disaster recovery > coordinator, AND loving emulators, I decided to build Hercules. > Having problems I looked into which systems ./configure looks for to > build to. Both tops10 and tops20 are in there. Does that mean I > could run a mainframe emulator under tops10/20? > ;) > -HN That's wierd! BTW, you might want to look at the following page: http://www.bsp-gmbh.com/hercules/herc_mvs2.html It has an ISO image for a CD that sets up a prebuilt Hercules setup. Zane Article: 15303 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!newsfeed.mathworks.com!portc03.blue.aol.com!newsfeed.skycache.com.MISMATCH!newsfeed1.cidera.com!Cidera!peer1-sjc1.usenetserver.com!usenetserver.com!e420r-atl1.usenetserver.com.POSTED!not-for-mail From: "Zane H. Healy" Subject: META S Manual online Newsgroups: alt.sys.pdp10 Organization: Aracnet User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Lines: 8 Message-ID: X-Complaints-To: abuse@usenetserver.com X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly. NNTP-Posting-Date: Fri, 18 Jan 2002 11:22:30 EST Date: Fri, 18 Jan 2002 16:22:30 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15303 Once again thanks to the efforts of Ronald Tatum there is a manual for METAS online at the PDP-10 Emulation Webpage. http://www.aracnet.com/~healyzh/pdp10emu.html While this manual looks to be for a IBM 360/44, the manual is there as META S was used in implementing ALGOL-W on the PDP-10. Zane Article: 15446 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!newspeer.monmouth.com!news.monmouth.com!not-for-mail From: pechter@i4got.pechter.dyndns.org (Bill Pechter) Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: PDP-10 Archive migration plan Date: 24 Jan 2002 15:38:51 -0500 Organization: Unknown Lines: 25 Message-ID: References: <3C4A7DF8.2AEC4BD7@trailing-edge.com> NNTP-Posting-Host: bg-tc-ppp1151.monmouth.com Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15446 alt.folklore.computers:79580 In article , JD wrote: > >Unless someone does something 'special', FreeBSD will often run Linux >binaries as well or better. So, it does seem that Linux would be a good > choice, but so would FreeBSD. They used to have tools to create a custom > floppy boot also. > >John Yup... but the linux emulation may be a bit large to pack on the floppy with ts10. A bootable CDROM with FreeBSD, ts10 and an emulated writeprotected master TOPS10 and TOPS20 pack would be slick. Mount an old writeable 500mb IDE as a couple of RP06's... automagically copy the images over if the structures are noth there and run. I'll take a look at building a custom 4.5 with ts10, though. Bill -- -- d|i|g|i|t|a|l had it THEN. Don't you wish you could still buy it now! bpechter@shell.monmouth.com|pechter@ureach.com Article: 15322 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!newsfeed.mathworks.com!nntp.abs.net!news.maxwell.syr.edu!sn-xit-03!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp.11,alt.sys.pdp8 Subject: SIMH 2.9 beta release Date: Mon, 21 Jan 2002 09:39:48 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 48 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15322 alt.sys.pdp8:5214 Yes, time for V2.9 beta, on http://simh.trailing-edge.com. The VAX/SDS test code is also V2.9 based. Because of API changes, V2.8 and V2.9 sources cannot be intermixed. New in V2.9: something for everyone (comedy tonight). 1. PDP11/PDP10/VAX: multiboard DZ emulation. The DZ simulator can now be configured with up to 32 lines, representing 4 separate DZ's. (It can probably go to 64, but enough for now.) 2. PC queues: for debugging, simulators now keep a queue of the last 64 PC changes, rather than just one. 3. Consistent device configuration. The ENABLE/DISABLE and ADD/REMOVE commands have been replaced by: SET ENABLED/DISABLED SET ONLINE/OFFLINE Where appropriate, device addresses or numbers can be changed; this is simulator specific. Note that the write enable command is now SET WRITEENABLED 4. Multiple calibrated timers. Instead of just the real-time clock, a simulator can implement up to 8 calibrated timers. 5. All terminal multiplexors implement new commands: SHOW STATISTICS SET DISCONNECT=line The latter is for booting individual users off. 6. HP2100. A bunch of new devices implemented and bugs fixed (thanks to help from Bill McDermith). 7. Altair Z80. Compilation problems on Windoze fixed (thanks to Peter Schorn). 8. The usual minor bug fixes and cosmetic improvements. The VAX is still stuck trying to bring up VMS following successful installation of the hobbyist image. The Interdata 16/32 are written but not tested, ditto the SDS 940, and the GRI 909 is waiting for more documentation. /Bob Article: 15367 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!cambridge1-snf1.gtei.net!news.gtei.net!bloom-beacon.mit.edu!howland.erols.net!enews.sgi.com!newspeer2.tds.net!sn-xit-01!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp.11,alt.sys.pdp8 Subject: Re: SIMH 2.9 beta release Date: Tue, 22 Jan 2002 17:47:21 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 24 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15367 alt.sys.pdp8:5223 There is a working tape of Wollongong UNIX (V7) for the Interdata 32b systems. There may also be images of Interdata's own RTOS's. There are sources to Project Genie, the original 940 timesharing system. It is a very difficult exercise to figure out how to reconstruct a running system from that. Nothing for the GRI-909 except the crystal growing program I wrote for it back in 1972. /Bob On Mon, 21 Jan 2002 22:11:51 GMT, "Zane H. Healy" wrote: >In alt.sys.pdp10 Bob Supnik wrote: >> installation of the hobbyist image. The Interdata 16/32 are written >> but not tested, ditto the SDS 940, and the GRI 909 is waiting for more >> documentation. > >Is there any OS's available for these systems? > > Zane Article: 15368 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!dfw-peer.news.verio.net!news.verio.net!crtntx1-snh1.gtei.net!lsanca1-snf1!news.gtei.net!newsfeed2.earthlink.net!newsfeed.earthlink.net!netnews.com!nntp.abs.net!feeder.qis.net!sn-xit-02!sn-post-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp.11,alt.sys.pdp8 Subject: Re: SIMH 2.9 beta release Date: Tue, 22 Jan 2002 17:51:31 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <60rr4ugmk9ubv7mfnk4tuiu85li24u4j0k@4ax.com> References: X-Newsreader: Forte Agent 1.8/32.548 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: newsabuse@supernews.com Lines: 47 Xref: dfw-artgen.news.verio.net alt.sys.pdp10:15368 alt.sys.pdp8:5224 On 22 Jan 2002 04:43:44 -0800, spedraja@ono.com (Sergio Pedraja) >> 5. All terminal multiplexors implement new commands: >> >> SHOW STATISTICS >> SET DISCONNECT=line >> >> The latter is for booting individual users off. > >Aha. I have one question, just for the way: Can I stay during undetermined >time in the SIMH "simulator control console", having the OS (Unix, by example) >running under ? > Being in the simulator console is equivalent to halting the system. Nothing is running on the target (simulated) system. Time stops, the clock loses interrupts, terminals don't respond, etc. The "unobtrusive" way to kick off a user is to do so using the facilities of the simulated operating system. >Another matter: I've hacked a little the PDP8 simulator to allow more >simulated terminals, in the line we spoke some time ago (TSS/8, the >possible 32 PT08 terminals that could be simulated, etc). I should like >to continue the work but it's possible I could need some help. Some number of additional terminals won't be a problem, but the PDP-8 only has 64 device numbers, each discrete terminal interface takes 2, so 32 individual interfaces aren't possible. I thought the maximum number supported was 19. > >> 6. HP2100. A bunch of new devices implemented and bugs fixed (thanks >> to help from Bill McDermith). > Looking to bring up some of the early disk operating systems, if they can be found. RTE would require floating point and memory manglement, not yet implemented. > >> 7. Altair Z80. Compilation problems on Windoze fixed (thanks to Peter >> Schorn). > >I had some problems with this compiling below MINGW. In appeareance >it does reference to some wrong/changed library or so. > Try the -DNO_INLINE function. Peter made some key routines inline, and not all compilers recognize the option. With this definition, it will compile on VC++. /Bob Article: 16074 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen.news.verio.net!iad-peer.news.verio.net!news.verio.net!newsfeed.cwix.com!wn2feed!worldnet.att.net!204.127.198.203!attbi_feed3!attbi.com!rwcrnsc51.ops.asp.att.net.POSTED!not-for-mail Message-ID: <3C64D1F9.D853CF6E@ev1.net> From: Charles Richmond Reply-To: richmond@ev1.net Organization: Cannine Computer Center X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: PDP-10 Archive migration plan References: <3C4A7DF8.2AEC4BD7@trailing-edge.com> <3C59C6E0.FED8AD60@jetnet.ab.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 25 NNTP-Posting-Host: 12.237.69.87 X-Complaints-To: abuse@attbi.com X-Trace: rwcrnsc51.ops.asp.att.net 1013233094 12.237.69.87 (Sat, 09 Feb 2002 05:38:14 GMT) NNTP-Posting-Date: Sat, 09 Feb 2002 05:38:14 GMT Date: Sat, 09 Feb 2002 05:38:14 GMT Xref: dfw-artgen.news.verio.net alt.sys.pdp10:16074 alt.folklore.computers:81303 "Zane H. Healy" wrote: > > In alt.sys.pdp10 Dowe Keller wrote: > > Definitely, one of the really neat things about UN*X is that its > > command line utilities are so powerful. Just about anything that > > "require" some bloated GUI app, can be done better in UN*X via the > > command line, easier, faster and infinitely more automatable. > > > I suppose the TOPS [12]0 folks are going to chime in now and say how > > infinitely better those systems were than UN*X ;-) > > For a distro like I have in mind about the only thing you'd want X-Windows > for is for xterm's, and emulators that need X-Windows for some reasons (I > know one of the PDP-8 emulators is supposed to have a very nice front panel > implemented via X-Windows. > Doug Jones has a PDP-8 emulator that puts up a nice looking front panel in X-Windows. IIRC, there is a way to open another xterm and type at it like it was the teletype connected to the PDP-8. -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ Article: 17789 of alt.sys.pdp10 From: Neil Franklin Newsgroups: alt.sys.pdp10 Subject: Re: Other OS's for the PDP-10? Date: 12 Sep 2002 23:58:59 +0200 Organization: My own Private Self Lines: 34 Message-ID: <6uadmmyin0.fsf@chonsp.franklin.ch> References: X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 12 Sep 2002 21:58:59 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 NNTP-Posting-Host: galapagos.ethz.ch X-Trace: pfaff.ethz.ch 1031867967 galapagos.ethz.ch (12 Sep 2002 23:59:27 +0200) Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!news-ge.switch.ch!news-zh.switch.ch!pfaff.ethz.ch!chonsp.franklin.ch!not-for-mail Xref: dfw-artgen alt.sys.pdp10:17789 "WarlockD" writes: > How about NetBSD? Seems to be happening. > Though I know its running at 36 bits, is it possible to have some kind of 32 > bit microcode for more compatibility, or would performance suffer? a) Not all PDP-10s were microcoded (KA-10 and KI-10) b) Exchanging microcode for an other instruction set amounts to de-10-ing the 10. It is the specific 36bit instruction set that makes it a 10. Reusing the microengine for something else (AFAIK there was an IBM 360 done for the KL-10 microengine) is just an recycled 10, not an real 10 any more. OTOH taking an other microengine (say an Intel x86) and writing an microcode for the specific 36bit instruction set makes the result an 10, allthough not an genuine DEC one. > There a gcc C compiler for it? That is what NetBSD is using. Was originally ported for under TOPS-20. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Hacker, Unix Guru, El Eng HTL/BSc, Programmer, Archer, Roleplayer - Make your code truely free: put it into the public domain Article: 17794 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!solaris.cc.vt.edu!news.vt.edu!canoe.uoregon.edu!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva1.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Other OS's for the PDP-10? Date: Mon, 16 Sep 2002 13:00:26 -0700 Organization: Networks and Distributed Computing Lines: 28 Message-ID: References: NNTP-Posting-Host: shiva1.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1032206429 36714 (None) 140.142.17.37 X-Complaints-To: help@cac.washington.edu In-Reply-To: Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:17794 Personally, I see little use in running UNIX on a PDP-10. I recognize that it has hack value. However, most PDP-10s these days are microcoded on platforms that can run UNIX much more effectively in native mode. Even using klh10, an x86 CPU will run software 30+ times faster in native mode than as a PDP-10 (and that's damn good, since 100x is more typical). What is far more interesting is running software that can't be run on anything other than a PDP-10. Which brings me to my main point. It seems that TOPS-20 and ITS have been relatively well-preserved. At least a dozen TOPS-20 systems are in 24/7 production on the net today. The barrier against production ITS systems is not as much technical as political; the Internet is a much more hostile place today than it was then ITS systems roamed the each. But what about the other PDP-10 operating systems? We can say that Tenex effectively lives on TOPS-20. However, all we seem to have from TOPS-10 are the DEC distributions. The Internet version of TOPS-20 (CMU, Harvard, LLL, WPAFB) seems to have vanished from the earth. Even sadder, repeated inquiries about WAITS have turned up nothing. That was a truly unique and special environment. Most production work on ITS came to an end after the 1970s, but WAITS was actively used until 1991. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 17815 of alt.sys.pdp10 Reply-To: "Douglas H. Quebbeman" From: "Douglas H. Quebbeman" Newsgroups: alt.sys.pdp10,alt.folklore.computers References: <7x8z1s5176.fsf@ruckus.brouhaha.com> <85sn006fdo.fsf@junk.nocrew.org> Subject: Re: Smallest PDP-10? Date: Mon, 23 Sep 2002 08:59:39 -0400 Lines: 22 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4910.0300 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3d8f103d_1@news.iglou.com> X-Trace: news.iglou.com 1032785981 204.250.0.238 (23 Sep 2002 08:59:41 -0400) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!logbridge.uoregon.edu!uunet!sea.uu.net!ash.uu.net!news.iglou.com!not-for-mail Xref: dfw-artgen alt.sys.pdp10:17815 alt.folklore.computers:99442 "Lars Brinkhoff" wrote in message = news:85sn006fdo.fsf@junk.nocrew.org... > Paul Rubin writes: > > shoppa@trailing-edge.com (Tim Shoppa) writes: > > > What's the smallest hardware configuration capable of running > > > PDP-10 software? (Presumably under an emulator, but I'm not > > > ruling out other implementations!) > > I'm sure I could easily run any of the 10 emulators on my Sharp > > Zaurus handheld with very little porting effort. >=20 > Jacob Nelson did that, and had a page about it here: > http://www.jfet.net/ten/ > Unfortunately, the web server is down. I'll take these down if Jacob asks, but until then: http://members.iglou.com/dougq/dec/systat2.jpg http://members.iglou.com/dougq/dec/Systat-20_screen.jpg -dq Article: 17829 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!logbridge.uoregon.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: Introduction? Date: 25 Sep 2002 02:59:55 GMT Organization: Aracnet Lines: 18 Message-ID: References: NNTP-Posting-Host: p-217.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.19 (i686)) Xref: dfw-artgen alt.sys.pdp10:17829 Jim Haynes wrote: > I just discovered this news group, and am a long time admirer of the PDP10 > but never got to use one in real life. Is there an introduction to what's > going on here? I gather there are one or more machine emulators? that > run under Unix? And maybe some hardware implementations? And files of > software that can be fed in to get a running system? For a good rundown on emulation, as well as pointers to pretty much any information that is available about the PDP-10 you can try my website. http://www.aracnet.com/~healyzh/decemu.html I think the basic rundown is 3 publically available emulators, one commercially available system, and two projects to implement the PDP-10 in FPGA. For OS's TOPS-10, TOPS-20, and ITS are available, sadly the rest appear to be lost in the mists of time. Zane Article: 17818 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!newsfeed.stanford.edu!canoe.uoregon.edu!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva0.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10,alt.folklore.computers Subject: Re: Smallest PDP-10? Date: Mon, 23 Sep 2002 10:14:26 -0700 Organization: Networks and Distributed Computing Lines: 21 Message-ID: References: NNTP-Posting-Host: shiva0.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1032801270 33040 (None) 140.142.17.38 X-Complaints-To: help@cac.washington.edu In-Reply-To: Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:17818 alt.folklore.computers:99485 On Mon, 23 Sep 2002, Tim Shoppa wrote: > What's the smallest hardware configuration capable of running PDP-10 > software? As other people have pointed out, klh10 has been run on a Sharp Zaurus. I think that there is such a thing as too small. I can't imagine doing useful PDP-10 work on a Zaurus, while I have done so on a Sony VAIO Picturebook. 5 times a KL10, 3 times an XKL-1. I think that what is more interesting is price/performance. I think that lingling.panda.com, built for $700 of commodity hardware and benchmarked at 16 times a KL10 (10 times an XKL-1) is the current leader. I'm sure that if I built Lingling today, I would get an even faster system (I used an Athlon 1700). Nevertheless, Lingling is impressive. It builds the TOPS-20 monitor from sources in 12.5 minutes! -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 18548 of alt.sys.pdp10 From: "Douglas H. Quebbeman" Newsgroups: alt.sys.pdp10 References: <3DEFE7D5.7070306@bellatlantic.net> <2_xI9.185$703.28762452@newssvr13.news.prodigy.com> Subject: Re: in 4 days, it will be dec 10. Date: Sun, 8 Dec 2002 12:00:02 -0500 Lines: 41 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 NNTP-Posting-Host: 204.250.0.238 X-Original-NNTP-Posting-Host: 204.250.0.238 Message-ID: <3df37a93_1@news.iglou.com> X-Trace: news.iglou.com 1039366803 204.250.0.238 (8 Dec 2002 12:00:03 -0500) X-Authenticated-User: dougq X-Original-NNTP-Posting-Host: 204.250.0.238 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!newsfeed.mathworks.com!wn13feed!wn14feed!worldnet.att.net!198.6.0.123!uunet!sac.uu.net!news.iglou.com!not-for-mail Xref: dfw-artgen alt.sys.pdp10:18548 "Edward A. Falk" wrote in message news:2_xI9.185$703.28762452@newssvr13.news.prodigy.com... > In article , > TLH 858 wrote: > >bob seems to have said: > > > >Seems to happen every year about this time. > > It only comes once a year, so we always celebrate it. > > > Hey all; I've been away from this newsgroup for a year or so; can > someone summarize the state of the emulation projects? Last I heard, > Tim Stark was in the process of polishing one, but I get the impression > that it was supplanted by another. I'm not sure what the current status of Tim's TS10 is; last I recall, he was working on VAX simulation for the package. Also available: SIMH, and Ken Harrentstein's KLH-10. > Is there a package out there where I can simply download it, install it > and have a pdp-10 on my laptop? Software? Documentation? HOWTOs? For SIMH, go to http://simh.trailing-edge.com and for KLH10, go to http://klh10.trailing-edge.com For KLH10, ready-build versions of TOPS-10 7.04 and ITS are available. For SIMH, I don't see a ready-to-run filesystem, but TOPS-10 7.03 has proven to be pretty easy to install, but I haven't done an install in almost a year now... -dq Article: 18550 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!news-peer.gip.net!news.gsl.net!gip.net!newsfeed.mathworks.com!cyclone.swbell.net!cyclone-sf.pbi.net!129.250.175.17!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: in 4 days, it will be dec 10. Date: 9 Dec 2002 21:52:02 GMT Organization: Aracnet Lines: 20 Message-ID: References: <3DEFE7D5.7070306@bellatlantic.net> <2_xI9.185$703.28762452@newssvr13.news.prodigy.com> NNTP-Posting-Host: p-837.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:18550 Edward A. Falk wrote: > Is there a package out there where I can simply download it, install it > and have a pdp-10 on my laptop? Software? Documentation? HOWTOs? My DEC Emulation website has most of the info you'll need. Be forwarned, I think I've got a bunch of stale links. Unfortuantly I've not had the time to go through and check them like I need to. As a result you'll probably need to go directly to Tim Shoppa's site to grab the tape images. My site can be found at: http://www.aracnet.com/~healyzh/pdp10emu.html I've got pages on all emulated DEC architectures, however, the PDP-10 is the most complete (and it's also the original portion). For an emulator I'd recommend SIMH for TOPS-10 and KLH10 for TOPS-20. Though you should probably take a look at the current state of Tim Starks work, as I've not keep up with it (I think he's mainly been concentrating on his VAX emulator of late). Zane Article: 18551 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newsfeed.wirehub.nl!newsfeed.news2me.com!border1.nntp.aus1.giganews.com!nntp.giganews.com!nntp3.aus1.giganews.com!nntp.speakeasy.net!news.speakeasy.net.POSTED!not-for-mail NNTP-Posting-Date: Mon, 09 Dec 2002 18:16:44 -0600 From: "Timothy Stark" Newsgroups: alt.sys.pdp10 Subject: ts10 emulator Date: Mon, 9 Dec 2002 19:16:43 -0500 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Message-ID: Lines: 56 NNTP-Posting-Host: 66.92.145.31 X-Trace: sv3-S9pfZXo2tO3ft6d58va5HLizvG9PP0LIJTcwKBT29XZChrMU9qJG0AQlJBohKeoRvEG9SmUPv5Z/NE1!0gA+4HmArtzTIBglncXoe3FLJXLU2OXoM79WBVvddcZdjo8iGtevtuUMIj8pjMAg9ivogslx00QI!pFsiSCOCgQPHU/AD X-Complaints-To: abuse@speakeasy.net X-DMCA-Complaints-To: abuse@speakeasy.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:18551 Hello folks: I noticed that on '4 days until Dec 10' articles. Yes, I still am working on my ts10 emulator. Latest snapshots are at ftp.firesword7.net/pub/ts10/develop. I plan to update it on sourceforge soon. I ported my latest version of ts10 emulator to Intel C/C++ 7.0 compiler and successfully compiled it into an exe file. Comparsion with GCC v3.2 on Red Hat Linux 8.0 (on 1.9 Ghz P4 system with 512 MB PC800 RIMM modules). My ts10 emulator has three complete working emulators, pdp10, pdp11, and vax. Section GCC v3.2 Intel C/C++ v7.0 with -ipo option PDP10 8.5 MIPS 10 MIPS PDP11 5.7 MIPS 8.5 MIPS VAX 5.0 MIPS 5.0 MIPS (Peak MIPS is 7.1 MIPS) I finished my VAX emulator and am now working on TMSCP tape drives for VAX and PDP11 in progress. DELQA emulation worked fine. NetBSD and OpenVMS operating system worked fine on it. Also, I implemented TSV05 emulation and was able restore my old tapes that I saved on the University's real VAX systems. In PDP11 emulator, ts10 supports boot controller (BDV11) for KDF11 processor but I still am looking for MRV11 specs because KDJ11 rom image is not compatible with BDV11 boot controller. There still is a bug somewhere in KDJ11 emulator that cause unexpected odd address trap. However, KDF11 emulator worked fine with running XXDP 2.5 operating system. Also, I am working on IBM 370/390 emulator and others. I was able to boot ZZSA editor on IBM 370 emulator with using x3270 terminal software. Well, I am re-implementing my new PDP-10 emulator from scratch to replace my old PDP-10 emulator because it consumes my time too much. That's why I had to re-write it to implement many new macros to save my working time alot. For example, I implemented the main function for several similar instructions (SKIP, SKIPL, SKIPE, SKIPLE, SKIPA, SKIPGE, SKIPN, and SKIPG) by using DEF_SKIP(name, opskip) macro. Also, I now use a pair of halfword instead of whole 64-bit integer for better performance because alot of shifts eats CPU cycles alot. There are many halfmove instructions, etc., I can take advantage of a pair of halfwords easily. For example for simple HLRO instruction: DEF_INST(kx10, HLRO) { w10 w = ReadV(ea); w.rh = w.lh; w.lh = H10_MASK; WriteAC(ac, w); } Thank you! Tim Stark Article: 18550 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!news-peer.gip.net!news.gsl.net!gip.net!newsfeed.mathworks.com!cyclone.swbell.net!cyclone-sf.pbi.net!129.250.175.17!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: in 4 days, it will be dec 10. Date: 9 Dec 2002 21:52:02 GMT Organization: Aracnet Lines: 20 Message-ID: References: <3DEFE7D5.7070306@bellatlantic.net> <2_xI9.185$703.28762452@newssvr13.news.prodigy.com> NNTP-Posting-Host: p-837.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:18550 Edward A. Falk wrote: > Is there a package out there where I can simply download it, install it > and have a pdp-10 on my laptop? Software? Documentation? HOWTOs? My DEC Emulation website has most of the info you'll need. Be forwarned, I think I've got a bunch of stale links. Unfortuantly I've not had the time to go through and check them like I need to. As a result you'll probably need to go directly to Tim Shoppa's site to grab the tape images. My site can be found at: http://www.aracnet.com/~healyzh/pdp10emu.html I've got pages on all emulated DEC architectures, however, the PDP-10 is the most complete (and it's also the original portion). For an emulator I'd recommend SIMH for TOPS-10 and KLH10 for TOPS-20. Though you should probably take a look at the current state of Tim Starks work, as I've not keep up with it (I think he's mainly been concentrating on his VAX emulator of late). Zane Article: 18555 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp.gblx.net!newspump.sol.net!newsfeed.news2me.com!sn-xit-02!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: SIMH Interdata simulator, and new documentation Date: Tue, 10 Dec 2002 13:21:29 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.91/32.564 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 13 Xref: dfw-artgen alt.sys.pdp10:18555 alt.sys.pdp11:7081 alt.sys.pdp8:5669 I've put an update to the Interdata simulators on http://simh.trailing-edge.com. The simulator now runs many of the Interdata diagnostics, both 16b and 32b, and is beta ready. My next step is to try the UNIX V6 and V7 ports. I've also posted some new manuals on SIMH to the web site documentation section. These include: - an update to SIMH internals manual - a new manual on IO structure of the most popular simulators - a new manual on magtape representation and error handling /Bob Article: 18630 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newsfeed.cwix.com!logbridge.uoregon.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Site Update Date: 31 Dec 2002 06:20:16 GMT Organization: Aracnet Lines: 18 Message-ID: NNTP-Posting-Host: p-439.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:18630 I don't want anyone passing out from shock, but I've just done a semi-decent job of going through my PDP-10 Emulation webpage, and have gotten most of the dead links updated. http://www.aracnet.com/~healyzh/pdp10emu.html I've already posted about Arthur Krewat's site being missing, unfortunatly it looks like http://pdp10.nocrew.org/ might be missing as well (hopefully it's just down temporarily). Tim, you used to have a PDF copy of the "TOPS-10 Commands Manual V7.01" manual available at mudd.trailing-edge.com, is it available anywhere? I'll see about getting the page updated to reflect all the excellent documentation that Al has made available. Zane Article: 18653 of alt.sys.pdp10 From: Neil Franklin Newsgroups: alt.folklore.computers,alt.sys.pdp10 Subject: PDP10 and RISC (was Re: vax6k.openecs.org rebirth) Date: 03 Jan 2003 22:00:21 +0100 Organization: My own Private Self Lines: 155 Message-ID: <6uvg16uexm.fsf_-_@chonsp.franklin.ch> References: <3E05FE9D.B3A12286@vax6k.openecs.org> <9dn1va.bu7.ln@via.reistad.priv.no> X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 3 Jan 2003 21:00:24 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 NNTP-Posting-Host: galapagos.ethz.ch X-Trace: pfaff.ethz.ch 1041627688 galapagos.ethz.ch (3 Jan 2003 22:01:28 +0200) Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newspeer1.nwr.nac.net!solnet.ch!solnet.ch!news-ge.switch.ch!irazu.switch.ch!news-zh.switch.ch!pfaff.ethz.ch!chonsp.franklin.ch!not-for-mail Xref: dfw-artgen alt.folklore.computers:108555 alt.sys.pdp10:18653 dropped the prime stuff, added pdp10 jmfbahciv@aol.com writes: > In article <9dn1va.bu7.ln@via.reistad.priv.no>, > Morten Reistad wrote: > >According to : > >>Yup. We do have different definitions. I had forgotten that my > >>idea of PDP-10 architecture having RISC-flavor was not politically > >>correct. :-) > > > >The PDP1/6/10 instruction set style had lots of things that > >anticipated RISC; Actually it simply predates the CISC "detour". The extended PDP10 (KL-10) made in midst of the CISC days had a few strongly CISCy additions (all the string stuff). > >but it fell through the RISC tests on three > >important counts: Effective address calculation ; sheer number > >of instructions and hardware interfaces. > > All right. Now I'm feeling uneducated :-). How does effective > address calculation affect the category? Gadzooks! RISC ist mainly defined as any architecture where instructions are included or excluded on whether they are "reduced complex", i.e. they can be implemented without getting in the way of the rest of the instructions. So stuff that speeds one specific (sub-)job 10 times, but adds 10% time to all other (sub-)jobs get throun out, if it is not used often enough to speed up by 10% to offset that. Main "show stoppers" are things that stall the pipeline by needing data that has not been prefetched (usually because it is only known to be needed too late) or lengthen the basic clock cycle (usually by requiring complex decoders). The PDP10 EA calculation hits on both these places. The culprint is the indirect bit (bit 13, whose number which is very appropriate :-)). a) it forces an memory read of the indirect address, whose address of that address it is only known after the memory access direct before it. So the pipeline stalls for the full time until memory answers. A "slow" 100MHz (=10ns per cycle) running from 100ns memory (that is full memory access cycle) falls to 10% speed if that happens constantly. A 200MHZ processor is already down to 5% speed with same memory. b) even after reading the word coming back, it may have I bits in it (for double/multiple indirect), so the data can not be immediately used but needs first looking at. So this is an case of waiting for decoder. This only applies to using I. The X indexed stuff is only fast memory indirect (so no access time trouble), and only max one indirection (so can be rolled out into hardware (an X=0 driven multiplexer that takes Y or Y+F(X)). So it is even a case of: AND 2,3(0) is fast and good, while AND 2,@3 is slow and bad! And yes, trying to minimise the impact of such indirect addressing is one of the headaches any fast PDP10 maker (such as FPGS cloners) has got to address, if maximal speed is wanted. This means trying to detect the difference of AND 2,@3 (which can be speeded up) and AND 2,@333 (which can not) and speeding up the just the later. But such detection (a 14bit comparison) slows down basic instruction decoding, thus we have again the "how often" trouble. Does speeding AND 2,@3 up by 3-10% (is even dependant on memory speed!) save more time than slowing all non-@ instructions by about 1%? > I've written > this question three times and it's still very clumsy. It is something from deep within your "guy thing" category. > With hardware interfaces...are you talking about peripherals > or memory? I don't know either what he meant by this. Peripherals usually are not relevant for CPU instruction speed. Nor for CPU classification. > I'll leave the number of instructions to somebody else. I sure > didn't think we had many; quite a few of them were just there > to complete the symmetry. They were a lot easier to document. It is not about reduced count, but reduced complexity. This missunderstanding is actually quite widely spread. > >Interestingly enough, this is where the emulator writers struggle > >with performance. Or rather the host processor running the emulator program. It goes into pipeline stalls, even if it has an pipeline friendly architecture. Software can kill any pipeline. Processors can only offer software (usually the compiler) the chance of avoiding this trouble. An emulator for an non-pipelinable processor will allways stall its host processor. As seen in above AND examples. PDP10s can be programmed pipeline friendly or pipeline stalling. But lots of software prefers I over X usage, because it is simpler (and surely faster on the simple microcoded KS-10, possibly the complex microcoded KL-10 had some "dirty trick" to make it same speed) and more universal (anywhere in memory, not just addresses 01-17). So fast PDP10 makers have to try to optimise at least some common cases, at cost of chip space, and clock speed. I can only hope that the gcc compiler port uses X over I as often as possible (or can be changed to in future optimise its code in this way). But then, X is slower than I on old some traditional microcoded processors, but X is faster than I on FPGAs, and I don't know what KI-10 did here (could be both ways). So this needs to be an target processor model dependant selectable option to the compiler! That would be -mtune=... option to gcc/pdp10, for those that want to know. Thanks to Lars for putting that option in, it will be needed one day in the future. > After I learn more about the three, perhaps we can get to this. > First, what is the emulator emulating? What ever the emulated target architecture does. > >Otherwise the large number of regular registers, orthogonal > >instruction set and flat memory space were RISCy features. > Yes. PDP10 is quite good on many aspects. But the central RISC stuff is not one of them. That "I" bit is the worst anti-RISC thing one can do. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Hacker, Unix Guru, El Eng HTL/BSc, Programmer, Archer, Blacksmith - hardware runs the world, software controls the hardware code generates the software, have you coded today? Article: 18684 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newsfeed.cwix.com!newsfeed.mathworks.com!cyclone.swbell.net!newsfeed1.easynews.com!easynews.com!easynews!newsfeed.news2me.com!border1.nntp.aus1.giganews.com!nntp.giganews.com!nntp3.aus1.giganews.com!nntp.speakeasy.net!news.speakeasy.net.POSTED!not-for-mail NNTP-Posting-Date: Sun, 05 Jan 2003 13:57:08 -0600 From: "Timothy Stark" Newsgroups: alt.sys.pdp10 Subject: ts10 emulator status - dynamic links? Date: Sun, 5 Jan 2003 14:57:08 -0500 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Message-ID: Lines: 29 NNTP-Posting-Host: 66.92.145.31 X-Trace: sv3-cBbfYk8DTMO140ey52UdCkFEHlGGW1OVmXVVmUSqrkQnys2qEAtgoUmZ9oT4hMJB5c+1Xat4+t/05q7!x8JNTlIPXaLJ6CAUSzHc0jhX08ghilaiVij2hhw7I+rnHq/lSanMe5isXOrunECGWBQeUHBzd/ZQ!aGeExp3tTvXmYDgu X-Complaints-To: abuse@speakeasy.net X-DMCA-Complaints-To: abuse@speakeasy.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:18684 Hello all, The ts10 emulator now is multi-system emulator system that have three individual working emulaors: PDP-10 (KS10/KL10), PDP-11 (KDF11/KDJ11), and VAX (KA630/KA655). I am considering to implement module support routines to load module files into ts10's memory space by using dlopen function call. What are advantages and disadvantages about modules and performance because dynamic links are new to me? About new version of PDP-10 emulator, I implemented most instructions execpt mul/div, EXTEND, floating, and byte pointer instructions that are not finished yet. I currently am working on pager systems. I successfully ran ksboot.exe on it. You know that I now am using a pair (two left/right halfwords of 36-bit word). I created new data types: a10 (30-bit PDP-10 address), h10 (18-bit word), w10 (36-bit word), and op10 (36-bit operators). About multiply and divide operators, I am figuring out how to implement multiply and divide operators by using a pair (36-bit word). With w10 data type, it have data structure: "typedef struct { h10 lh, rh; } w10;". Because of w10 data type, I have to create op10 inline macros like op10_and, op10_xor, op10_or, op10_add, etc. Also, I am using multi-instruction inline macros much like matrix to save my development and debugging time alot. Thank you! Tim Stark Article: 18750 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newspeer1.nwr.nac.net!newspeer.monmouth.com!news.monmouth.com!shell.monmouth.com!not-for-mail From: pechter@shell.monmouth.com (Bill/Carolyn Pechter) Newsgroups: alt.sys.pdp10 Subject: Re: Injury has kept me out of loop a while - is there a working x86-based simulator yet???? Date: 19 Jan 2003 21:22:02 -0500 Organization: Lakewood MicroSystems Lines: 47 Message-ID: References: <4usg2vkua0aq2pel6a69scu8icvr1g76h9@4ax.com> NNTP-Posting-Host: shell.monmouth.com Xref: dfw-artgen alt.sys.pdp10:18750 In article <4usg2vkua0aq2pel6a69scu8icvr1g76h9@4ax.com>, AtT wrote: > >Sorry for the DUMB QUESTION, but I've been attempting to recover from >on-the-job injury from May 1999 (was a print reporter - was hit over >the head by a flying TV cameraman w/170 lbs gear - Worker's Comp >thinks keeping me on all the dope I want for pain is sufficient >"treatment" after surgery failed) ... anyway, I've been away from the >group for a while. > > dmr > ICC 1970/LIRICS '70-74(officially) Welcome back... They've got 3 TOPS-10 emulators running now... TS10, KLH10 and the SIMH emulator. The last one works out pretty well under Windows and the others work ok under FreeBSD/Solaris/Linux... The worst injury I had while doing my old reporter job involved a whack or two by someone against my long lens while shooting pix for the local weekly rag here in Central NJ. Sorry to hear about the pain. http://simh.trailing-edge.com http://klh10.trailing-edge.com http://pdp-10.trailing-edge.com http://www.pdp10.com TS10 emulator 07-Mar-01: Tim Stark's PDP-10 emulator software available via CVS at www.sourceforge.net/projects/ts10 and a members-only mailing list at groups.yahoo.com/group/ts10emu . ftp://ftp.trailing-edge.com/pub/ts10.tar.gz Bill -- +---------------------------------------------------------------------------+ | Bill and/or Carolyn Pechter | pechter@shell.monmouth.com | | Bill Gates is a Persian cat and a monocle away from being a villain in | | a James Bond movie -- Dennis Miller | +---------------------------------------------------------------------------+ Article: 18753 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!logbridge.uoregon.edu!pln-w!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: Injury has kept me out of loop a while - is there a working x86-based simulator yet???? Date: 20 Jan 2003 21:40:42 GMT Organization: Aracnet Lines: 18 Message-ID: References: <4usg2vkua0aq2pel6a69scu8icvr1g76h9@4ax.com> NNTP-Posting-Host: p-364.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:18753 AtT wrote: > So, is there a useable (runs TOPS-10) package out there yet? A cold > question, I know since I could not program such a magnificent work to > save my life and would just like to get ahold of someone else's to > revisit some old software You'll probably want to check out my website, as it has pointers to everything you'll need for PDP-10 emulation. http://www.aracnet.com/~healyzh/pdp10emu.html > dmr > ICC 1970/LIRICS '70-74(officially) You might also want to check out the following website, though it sounds like you predate the period I think it covers. http://www.lirics.org/ Zane Article: 18751 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!sn-xit-03!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: SIMH 2.10-2 Date: Mon, 20 Jan 2003 09:33:16 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> X-Newsreader: Forte Agent 1.91/32.564 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 13 Xref: dfw-artgen alt.sys.pdp10:18751 alt.sys.pdp11:7218 alt.sys.pdp8:5707 SIMH 2.10-2 was released this weekend, source kit at http://simh.trailing-edge.com. - DELQA fix, VAX simulator now runs Local Area VAXclusters - Interdata 16b and 32b simulators first release - still needs more work to get UNIX running - SDS 940 first release - Stub Unibus Ethernet support for PDP-10, -11, DEUNA/DELUA will be released separately when ready. - HP 2100 Access interprocessor link. /Bob Supnik Article: 18773 of alt.sys.pdp10 Sender: phr2003-nospam@ruckus.brouhaha.com From: Paul Rubin Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> <73dff34c.0301211959.2ed9703e@posting.google.com> Date: 23 Jan 2003 05:33:42 -0800 Message-ID: <7xwukwc7nt.fsf@ruckus.brouhaha.com> Organization: Nightsong/Fort GNOX Lines: 28 User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii NNTP-Posting-Host: 209.66.107.17 X-Trace: 23 Jan 2003 06:13:21 -0800, 209.66.107.17 Path: iad-read.news.verio.net!dfw-artgen!dfw-peer!news.verio.net!crtntx1-snh1.gtei.net!mtvwca1-snh1.ops.genuity.net!paloalto-snf1.gtei.net!news.gtei.net!enews.sgi.com!news.spies.com!209.66.107.17 Xref: dfw-artgen alt.sys.pdp10:18773 alt.sys.pdp11:7248 alt.sys.pdp8:5721 peter@taronga.com (Peter da Silva) writes: > >Much easier to believe than you think. I recently (well, actually over > >a year ago, but I have so little time to work on pet projects it seems > >recent when I look at how little I haveaccomplished) started looking at > >something called The Software Tools Virtual Operating System. In light > >of the success of a certain Virtual Machine (the P-System warmed over) > >I thought maybe there might be a place for a Virtual OS again as well. > > I suspect you'd be disappointed. The only "Software Tools Virtual > OS" I've ever heard of is a bunch of UNIX utilities and libraries > implemented in Fortran and Ratfor. I should have a copy of the tape > somewhere (either on 9-track or a CD copy). It was a bit more than that. I sort of remember it. It included a fair amount of Macro-10 code and was able to simulate a Unix shell including pipelines (I think it used temp files for that), and it simulated the Unix file system (including directory structure) by using one big file on the host OS as a virtual "disk drive" that it did its own file and block allocation etc. inside of. It also included asm code for other machines/OS's to make the Ratfor applications portable to some other systems including maybe VM/370. It was quite an interesting hack, though I don't know if it was really practical. I wrote a lot of Ratfor code under Compuserve's mutant strain of Tops-10, though not using the Virtual OS layer. The Ratfor stuff was really pretty cool it really did feel a lot like Unix/C programming on a PDP-11. Whether that's good or bad is of course a matter of taste. Article: 18777 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!newsfeed.stanford.edu!news-proxy.abbnm.com!citadel.in.taronga.com!not-for-mail From: peter@taronga.com (Peter da Silva) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 Date: 23 Jan 2003 17:28:50 GMT Organization: TSS Inc. Lines: 34 Message-ID: References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> <7xwukwc7nt.fsf@ruckus.brouhaha.com> NNTP-Posting-Host: citadel.in.taronga.com X-Trace: citadel.in.taronga.com 1043342930 10786 10.0.0.43 (23 Jan 2003 17:28:50 GMT) X-Complaints-To: usenet@taronga.com NNTP-Posting-Date: 23 Jan 2003 17:28:50 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Xref: dfw-artgen alt.sys.pdp10:18777 alt.sys.pdp11:7253 alt.sys.pdp8:5722 In article <7xwukwc7nt.fsf@ruckus.brouhaha.com>, Paul Rubin wrote: >peter@taronga.com (Peter da Silva) writes: >> >Much easier to believe than you think. I recently (well, actually over >> >a year ago, but I have so little time to work on pet projects it seems >> >recent when I look at how little I haveaccomplished) started looking at >> >something called The Software Tools Virtual Operating System. In light >> >of the success of a certain Virtual Machine (the P-System warmed over) >> >I thought maybe there might be a place for a Virtual OS again as well. >> I suspect you'd be disappointed. The only "Software Tools Virtual >> OS" I've ever heard of is a bunch of UNIX utilities and libraries >> implemented in Fortran and Ratfor. I should have a copy of the tape >> somewhere (either on 9-track or a CD copy). >It was a bit more than that. I sort of remember it. It included a >fair amount of Macro-10 code and was able to simulate a Unix shell >including pipelines (I think it used temp files for that), and it >simulated the Unix file system (including directory structure) by >using one big file on the host OS as a virtual "disk drive" that it >did its own file and block allocation etc. inside of. Sounds like someone took the Software Tools code and went a little nuts with it on the -10. The version I had had a shell, supporting pipelines with temp files, but very little of it was in assembly and most was in as portable FORTRAN as the authors could manage. It was *definitely* practical. I used it as my shell in RSX-11 for years. -- Rev. Peter da Silva, ULC. 29.6852N 95.5770W WWFD? "Be conservative in what you generate, and liberal in what you accept" -- Matthew 10:16 (l.trans) Article: 18792 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!fu-berlin.de!uni-berlin.de!pa-scranton2b-39.sctnpa.adelphia.NET!not-for-mail From: bill@cs.uofs.edu (Bill Gunshannon) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 Date: 24 Jan 2003 15:43:02 GMT Organization: Computing Sciences Dept., University of Scranton Lines: 71 Sender: bill@gw5.cs.uofs.edu Message-ID: References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> <7xwukwc7nt.fsf@ruckus.brouhaha.com> <7xptqnvkfs.fsf@ruckus.brouhaha.com> <7x65sfzjg4.fsf@ruckus.brouhaha.com> Reply-To: bill@cs.uofs.edu NNTP-Posting-Host: pa-scranton2b-39.sctnpa.adelphia.net (24.49.73.39) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: fu-berlin.de 1043422982 30228041 24.49.73.39 (16 [135708]) X-Newsreader: knews 1.0b.1 Xref: dfw-artgen alt.sys.pdp10:18792 alt.sys.pdp11:7274 alt.sys.pdp8:5731 In article <7x65sfzjg4.fsf@ruckus.brouhaha.com>, Paul Rubin writes: > peter@taronga.com (Peter da Silva) writes: >> >Were you actually using the Virtual OS stuff, that simulated a whole >> >Unix tree-structured file system inside a single RSX-11 file? >> >> Did that capability actually exist anywhere other than the -10? Because >> I've never heard of it, and what I *did* use was christened with the same >> name. > > Yes, I think VM/CMS was also supported, and maybe some other systems > as well. But my memory is hazy. If you still have a tape somewhere, > it would be fun to see that code again. "maybe some other systems"? Appendix D. Machines and Systems The following summarizes the machines and systems used by members of the software tools user group. Most support at least the RATFOR preprocessor and the I/O primitives. Burroughs B 1700 -- local CDC 1784 -- local CDC 6000s, Cybers -- KRONOS, UT-2D, local, DUAL-MACE, SCOPE3, NOS CDC MP-32 -- MPX/OS CDC 7600 -- LTSS, SCOPE II, local Cray -- CPSS DataGeneral Eclipse -- AOS, RDOS (C & S series) ROLM 1602 -- RDOS GEC 4070 -- OS 4000 Honeywell 6000S -- GCOS-3 Honeywell Level 6 -- MOD 6 OS DataGeneral Nova -- RDOS DataGeneral MP-100 -- MP/OS PDP 11s -- RSX-IIM, RSX-IIS, RSX-IID, 1AS, RT-11, RSTS, Unix, DOS, S ACOS 700 -- GCOS AN/UYK-20 -- Level 2 HP 1000, 3000 -- RTE-IVB, MPE-III HP 21MX -- RTE III, RTE IV PDP 15 -- XVM/RSX PDP 20 --TOPS20 VAX -- VMS LSI 11 -- UCSD Pascal, RT-11, DOS-2 IBM S/360, S/370, 303x -- OS/MVT, VM.CMS, MVS, TSO, Wilbur IBM 1130 -- DM2 FACOM M-200, M-190 -- OS IV/F4 HITAC 8700, 8800 -- OS7 M170 -- VOS3 Prime -- PRIMOS SEL 32/77 -- MPX SIEMANS 4004 -- TST TELEFUNKEN TR440 -- BS19 Intel 8080 -- ISIS Intel 8086 -- UCSD Pascal Interdata 70 -- DOS Interdata 8/32 -- OS/32MT Modcomp -- MAX Univac 1100 -- EXEC 8 Univac 90/70 -- VS/9 Xerox Sigma -- RBM, CP-V Zilog Z80 CP/M, Oasis -- CP/M, Oasis bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include Article: 18795 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!newsfeed.icl.net!newsfeed.fjserv.net!newsfeed.arcor-online.net!fu-berlin.de!uni-berlin.de!pa-scranton2b-39.sctnpa.adelphia.NET!not-for-mail From: bill@cs.uofs.edu (Bill Gunshannon) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 Date: 24 Jan 2003 16:30:31 GMT Organization: Computing Sciences Dept., University of Scranton Lines: 32 Sender: bill@gw5.cs.uofs.edu Message-ID: References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> <7xwukwc7nt.fsf@ruckus.brouhaha.com> <7xptqnvkfs.fsf@ruckus.brouhaha.com> <7x65sfzjg4.fsf@ruckus.brouhaha.com> <7xr8b2bl1v.fsf@ruckus.brouhaha.com> Reply-To: bill@cs.uofs.edu NNTP-Posting-Host: pa-scranton2b-39.sctnpa.adelphia.net (24.49.73.39) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: fu-berlin.de 1043425831 30228041 24.49.73.39 (16 [135708]) X-Newsreader: knews 1.0b.1 Xref: dfw-artgen alt.sys.pdp10:18795 alt.sys.pdp11:7277 alt.sys.pdp8:5734 In article <7xr8b2bl1v.fsf@ruckus.brouhaha.com>, Paul Rubin writes: > bill@cs.uofs.edu (Bill Gunshannon) writes: >> > Yes, I think VM/CMS was also supported, and maybe some other systems >> > as well. But my memory is hazy. If you still have a tape somewhere, >> > it would be fun to see that code again. >> >> "maybe some other systems"? >> >> Appendix D. Machines and Systems >> The following summarizes the machines and systems used by >> members of the software tools user group. Most support at least the >> RATFOR preprocessor and the I/O primitives. > > I'm not talking about running just the preprocessor and i/o. > I'm talking about the rather fancy Virtual OS package. That list was as of 1980. Things didn't stop then. It stated that "most supported at least the RATFOR preprocessor and the I/O primitives" not that they only supported that part. All of the systems I used were complete environments by 1983. For another perspective try looking at: http://www.lbl.gov/Science-Articles/Archive/software-award.html bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include Article: 18805 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!newsfeed.stanford.edu!news-proxy.abbnm.com!citadel.in.taronga.com!not-for-mail From: peter@taronga.com (Peter da Silva) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 Date: 24 Jan 2003 21:30:45 GMT Organization: TSS Inc. Lines: 138 Message-ID: References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> NNTP-Posting-Host: citadel.in.taronga.com X-Trace: citadel.in.taronga.com 1043443845 64989 10.0.0.43 (24 Jan 2003 21:30:45 GMT) X-Complaints-To: usenet@taronga.com NNTP-Posting-Date: 24 Jan 2003 21:30:45 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Xref: dfw-artgen alt.sys.pdp10:18805 alt.sys.pdp11:7287 alt.sys.pdp8:5744 In article , Bill Gunshannon wrote: >Somewhere between 1982 and 1986 or so I used the system on: >Univac-1100 -- Exec 8. Complete, done by U Wisconsin I think. >Prime -- PRIMOS. Complete, done by Ga. Tech. >VAX -- VMS. Coplete. Don't know who did it. >I expect that most systems were complete and either used the encapsulated >file system or mapped the Unix style filesystem onto the native filesystem. Fair enough. I used it in various incarnations but never later than 1983, and all of them used native files. The encapsulated file system does seem a little impractical. >My whole point is that all this work was done. All this work was lost. I think that's far too strong a statement. The Software Tools VOS was developed by people at Bell Labs and later Berkeley to provide a UNIX "Software Tools" environment on non-UNIX systems. These people were also working on both AT&T and BSD UNIX. The origin was the book "Software Tools" by Kernighan and Plaugher, and I put my first Software Tools environment together by typing in and porting the listings in that book and getting them to work on RSX. I had no idea, at first, that there was a tape available. :) The second Software Tools book was in Pascal, and unfortunately missing a lot of the content of the first book. The third "Software Tools Book" isn't called "Software Tools". Instead, it's titled "The UNIX Programming Environment". Which tells you what eventually happened with the Software Tools VOS. It's not "lost". It's simply merged back into the UNIX family. What are the tools? ar, cat, cmp, crypt, date, dc, diff, echo, ed, expand, man, mv, pr rev, rm, sort, spell, split, tail, tee, tr, tsort, uniq, wc, and sh These are all versions of the UNIX tools. tcs SCCS ch a subset of sed, a superset of gres. cpress compress (pack, gzip, ...) crt a primitive "more" detab,entab Similar to the unix utility "col". ed,edin member of the ed/ex/... family find,fb something between grep and sed field join/split/... format nroff ratfor ratfor kwic This went into the BSD tapes. lam This is an interesting tool for joining files, and the first part of the system that doesn't have a close cousin on the BSD tapes. Kind of trivial to implement with other tools these days. ll awk 'BEGIN { longest=0; shortest=0 } shortest == 0 || len($0) < shortest { shortest = len($0) } len($0) > longest { longest = len($0) } END { print shortest, longest }' macro m4 mcol another subset of awk os col pl a tiny subset of awk and/or sed sedit sed show make a file printable, nothing exactly equivalent but it's close Later versions had more tools, but all of them were of a similar character: providing a UNIX environment for non-UNIX systems. >And all this work has been re-invented by people who think they are >doing something inovative. Just like The UCSD P-machine and the JAVA VM. The people who "re-invented" it were often the people who invented it in the first place, and most of them invented it on UNIX first. -- Rev. Peter da Silva, ULC. 29.6852N 95.5770W WWFD? "Be conservative in what you generate, and liberal in what you accept" -- Matthew 10:16 (l.trans) Article: 18757 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!elk.ncren.net!news.bu.edu!budd From: budd@csa.bu.edu (Phil Budne) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: Re: SIMH 2.10-2 Followup-To: alt.sys.pdp10 Date: 22 Jan 2003 04:32:55 GMT Organization: Boston University Computer Science Dept. Lines: 32 Message-ID: References: <012o2v47nu0gri6frofbe11scdufftsq1c@4ax.com> <244ba93d.0301211244.6b0ee57f@posting.google.com> X-Trace: news3.bu.edu 1043209975 16121 128.197.12.3 (22 Jan 2003 04:32:55 GMT) X-Complaints-To: news@bu.edu Xref: dfw-artgen alt.sys.pdp10:18757 alt.sys.pdp11:7228 alt.sys.pdp8:5713 In article , Al Kossow wrote: >In article <244ba93d.0301211244.6b0ee57f@posting.google.com>, >spedraja@navegalia.com (Sergio Pedraja) wrote: >> Would it be possible to obtain software for the PDP-7 >> simulator, like DECsys or the primary versions of Unix ? And, What >> about the MIT Time sharing system that ran under the PDP-1 ? > >No copies of this software are known to exist. A number of years ago Bob Supnik (then still a VP at Digital (or whatever it was called at the time)) gave a talk about the SIMH project (and the latest generation Alpha chips) at the Tech Square "Playroom". I was waiting to talk to him; I had a DECtape in a green box that was labled as PDP-15 system software(*) (at the time none was known to have survived). The person before me had a 2400' 1/2" tape they said was a 7-track containing dumps of PDP-1 DECtapes. Bob seemed less than interested when it was confirmed that the software depended on MIT hardware modifications (i.e. 2's complement arithmetic). I've also been told someone at MIT may have a 7-track with CTSS on it. Last time I asked him, Tim Shoppa was willing to read such tapes if the results could be made publicly available. If you have one of the above tapes (or know someone who does), please think about getting the tape read!! (*) The tape in the box turned out to be in PDP-11 format, and to contain some RSX-11 related stuff. Article: 18833 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!cyclone.bc.net!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!Shimo-Tomobiki.Panda.COM!MRC From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Warning compiling simh vax emulator Date: Tue, 28 Jan 2003 11:22:42 -0800 Organization: Networks & Distributed Computing Lines: 17 Sender: mrc@ndcms.cac.washington.edu Message-ID: References: <3E3302AA.1807F93@tiscalinet.it> <3E34F3EE.8CC4AFFF@tiscalinet.it> NNTP-Posting-Host: nntp1.u.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1043781749 40518 (None) 140.142.17.37 X-Complaints-To: help@cac.washington.edu In-Reply-To: Xref: dfw-artgen alt.sys.pdp10:18833 On Tue, 28 Jan 2003, Peter da Silva wrote: > That's pretty bad. The 11/780 was less than 1 MIPS (leading to the question > of whether you were measuting "real MIPS" or "VAX MIPS"). That means the > interpreter is executing hundreds of instructions per VAX instruction. The KLH10 emulator seems to have a mere 40x performance hit compared to the native hardware when doing Dhrystones. Roughly, my experience is that you get a KL10 equivalent in performance for each 100MHz of native CPU, so KLH10 running on a dedicated 700 MHz PIII should be about 7 times a KL10. Of course, the PDP-10 architecture is much easier to implement quickly than the VAX. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 18844 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!dfw-peer!news.verio.net!newsfeed.mathworks.com!arclight.uoregon.edu!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva1.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Warning compiling simh vax emulator Date: Wed, 29 Jan 2003 12:32:28 -0800 Organization: Networks and Distributed Computing Lines: 28 Message-ID: References: <3E3302AA.1807F93@tiscalinet.it> <7xn0lkd882.fsf@ruckus.brouhaha.com> NNTP-Posting-Host: shiva1.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1043872350 41322 (None) 140.142.17.38 X-Complaints-To: help@cac.washington.edu In-Reply-To: Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:18844 On Wed, 29 Jan 2003, Geoff Biscuit wrote: > With all this said, is there a way to slow down SimH, KLH10 etc such > that they resemble the speed of the various machines they are > simulating. > > Obviously serves no purpose other than more faithfully simulating the > whole 'experience' of old iron! In the days of old iron we were always trying to get faster processors. None of us would thought of deliberately slowing things down. 16x KL speed isn't really that great a performance jump. It basically brings us up to 1990 vintage performance; Lingling is comparable to a PMAX running Ultrix. And indeed, by 1990 the last few KLs still running were looking pretty sad. We had many processors of differing speeds along the way. I see Lingling as simply another of the tradition, somewhat behind the curve but much further along than the XKL or SC machines. Now, we do have certain advantages; we've also missed the past 20 years of bloatware. Lingling finally represents a PDP-10 processor that is competant to run Lisp (which for a KL was bloatware). -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 18843 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!cyclone.bc.net!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva1.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Warning compiling simh vax emulator Date: Wed, 29 Jan 2003 12:19:32 -0800 Organization: Networks and Distributed Computing Lines: 12 Message-ID: References: <3E3302AA.1807F93@tiscalinet.it> <7xn0lkd882.fsf@ruckus.brouhaha.com> NNTP-Posting-Host: shiva1.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1043871574 26814 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu In-Reply-To: Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:18843 On Wed, 29 Jan 2003, Magnus Olsson wrote: > >What's more interesting is that building and linking all of the TOPS-20 > >release 7 monitor, including all the TCP and DECnet stuff, takes 12.5 > >minutes under KLH10 on an Athlon 1700+. > How long does it take on real hardware? Hours. Something like 3 hours on a KL, overnight on a KS. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 18854 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva1.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Warning compiling simh vax emulator Date: Thu, 30 Jan 2003 09:26:31 -0800 Organization: Networks and Distributed Computing Lines: 17 Message-ID: References: <3E3302AA.1807F93@tiscalinet.it> <7xn0lkd882.fsf@ruckus.brouhaha.com> NNTP-Posting-Host: shiva1.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1043947593 4692 (None) 140.142.17.35 X-Complaints-To: help@cac.washington.edu In-Reply-To: Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:18854 On Thu, 30 Jan 2003 jmfbahciv@aol.com wrote: > >> >What's more interesting is that building and linking all of the TOPS-20 > >> >release 7 monitor, including all the TCP and DECnet stuff, takes 12.5 > >> >minutes under KLH10 on an Athlon 1700+. > >> How long does it take on real hardware? > >Hours. Something like 3 hours on a KL, overnight on a KS. > Yep. Too bad you couldn't use a faster OS. Actually, the files that were slow to compile were all the DECnet-36 stuff with its intensive use of macros. That junk came from the TOPS-10 losers as I recall; it's cluelessness about TOPS-20 sure looks that way (e.g. the NMXTIM routine). -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 18865 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!207-172-216-9 From: jmfbahciv@aol.com Newsgroups: alt.sys.pdp10 Subject: Re: Warning compiling simh vax emulator Date: Fri, 31 Jan 03 09:39:21 GMT Organization: UltraNet Communications, Inc. Lines: 28 Message-ID: References: <3E3302AA.1807F93@tiscalinet.it> <7xn0lkd882.fsf@ruckus.brouhaha.com> X-Trace: UmFuZG9tSVZzUEZCiAzEFATtjzJBTD84Zryvur/2CBaidQkLkR5oEeM6hS/BKNis X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 31 Jan 2003 10:18:00 GMT X-Newsreader: News Xpress Version 1.0 Beta #4 Xref: dfw-artgen alt.sys.pdp10:18865 In article , Mark Crispin wrote: >On Thu, 30 Jan 2003 jmfbahciv@aol.com wrote: >> >> >What's more interesting is that building and linking all of the TOPS-20 >> >> >release 7 monitor, including all the TCP and DECnet stuff, takes 12.5 >> >> >minutes under KLH10 on an Athlon 1700+. >> >> How long does it take on real hardware? >> >Hours. Something like 3 hours on a KL, overnight on a KS. >> Yep. Too bad you couldn't use a faster OS. > >Actually, the files that were slow to compile were all the DECnet-36 stuff >with its intensive use of macros. That junk came from the TOPS-10 losers >as I recall; it's cluelessness about TOPS-20 sure looks that way (e.g. the >NMXTIM routine). I don't remember who did DECnet in the -20 monitor. IIRC, DECnet per se didn't go into the -20 monitor until phase IV with the ethernet implementation. The previous releases were MCB-based DECnet. The MCB got done first on the -20, then it came over to the -10. We hired our own to do ethernet and they didn't work on the -20. /BAH Subtract a hundred and four for e-mail. Article: 18848 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!nntp1.roc.gblx.net!nntp.gblx.net!nntp1.phx1.gblx.net!nntp.gblx.net!nntp.gblx.net!newsfeed.news2me.com!newsfeed2.earthlink.net!newsfeed.earthlink.net!stamper.news.pas.earthlink.net!newsread1.prod.itd.earthlink.net.POSTED!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp11,alt.sys.pdp10,alt.sys.pdp8 Subject: Precompiled Windows versions of SIMH Message-ID: X-Newsreader: Forte Agent 1.91/32.564 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 5 Date: Thu, 30 Jan 2003 01:51:55 GMT NNTP-Posting-Host: 63.214.89.146 X-Complaints-To: abuse@earthlink.net X-Trace: newsread1.prod.itd.earthlink.net 1043891515 63.214.89.146 (Wed, 29 Jan 2003 17:51:55 PST) NNTP-Posting-Date: Wed, 29 Jan 2003 17:51:55 PST Organization: EarthLink Inc. -- http://www.EarthLink.net Xref: dfw-artgen alt.sys.pdp11:7344 alt.sys.pdp10:18848 alt.sys.pdp8:5765 In response to numerous requests, precompiled Windows executables of SIMH are available on http://simh.trailing-edge.com. The VAX and PDP11 are compiled without Ethernet support. /.Bob Supnik Article: 18921 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail From: baby_p_nut@yahoo.com (Baby Peanut) Newsgroups: alt.sys.pdp11,alt.sys.pdp10,alt.sys.pdp8 Subject: Re: Precompiled Windows versions of SIMH Date: 7 Feb 2003 16:00:21 -0800 Organization: http://groups.google.com/ Lines: 57 Message-ID: References: NNTP-Posting-Host: 198.80.171.28 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1044662421 1796 127.0.0.1 (8 Feb 2003 00:00:21 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 8 Feb 2003 00:00:21 GMT Xref: dfw-artgen alt.sys.pdp11:7438 alt.sys.pdp10:18921 alt.sys.pdp8:5788 Bob Supnik wrote in message news:... > In response to numerous requests, precompiled Windows executables of > SIMH are available on http://simh.trailing-edge.com. The VAX and > PDP11 are compiled without Ethernet support. > > /.Bob Supnik SIMH/VAX 2.10-2 Win32 binaries *with* Ethernet: http://www.tubas.net/~kstailey/simh/ Requires winpcap. http://winpcap.polito.it/install/bin/WinPcap_3_0_a4.exe from http://winpcap.polito.it/install/default.htm since the packet capture drives are not bundled. For network documentation get the SIMH source code from http://simh.trailing-edge.com/ and look through it for readme's and *.txt. To test if winpcap is running OK get WinDump from http://windump.polito.it/ and run "WinDump -D". It should print the name of any network cards you have similar to this: 1.\Device\NPF_El90x1 (3Com 3C90x Ethernet Adapter) If you don't see that you might need to replace the Packet.dll that is in my ZIP file. Look for replacements at the bottom of this page: http://winpcap.polito.it/contact.htm Once you have that working you should use name from "WinDump -D" in a SIMH "attach" command to tell SIMH to map your network card to the virtual DELQA. Given the name from the previous example the attach command would be: ATTACH XQ \Device\NPF_El90x1 It is not well documented but the author of SIMH gave up trying to shove a VT-100 emulator in the Windows binary and put a command to use telnet to gain control of the console. Use set telnet After that command, telnet into the emulator console and then type "boot cpu" from the non-telnet session and the system will boot with console output going to the telnet one. Article: 18937 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!sn-xit-03!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8 Subject: SIMH 2.10-3 and DOS-15 Date: Fri, 14 Feb 2003 09:42:51 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.91/32.564 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 20 Xref: dfw-artgen alt.sys.pdp10:18937 alt.sys.pdp11:7460 alt.sys.pdp8:5790 There is a minor update to SIMH on http://simh.trailing-edge.com. The principal differences: 1. Fixes to the PDP-15 to run DOS-15. 2. Additional breakpoint capabilities. If you are not planning on running DOS-15, there are no functional differences from 2.10-2. Hans Pufal has completed the restoration of DOS-15, including fixing a 30+ year old operating system bug that prevented use of a maximum RF15/RS09 configuration. For details, please see Hans' web page at http://www.aconit.org/hbp/PDP9/DOS The distribution kit for DOS-15 is on the SIMH web site, under software kits. /Bob Supnik Article: 19188 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!cyclone.bc.net!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!shiva1.cac.washington.edu!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: help needed on CPU tests Date: Sat, 8 Mar 2003 15:34:08 -0800 Organization: Networks and Distributed Computing Lines: 103 Message-ID: NNTP-Posting-Host: shiva1.cac.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1047166450 4892 (None) 140.142.17.37 X-Complaints-To: help@cac.washington.edu Content-Length: 314159 (believe this at your own risk) Xref: dfw-artgen alt.sys.pdp10:19188 Here is my current understanding of the tests to determine between various forms of PDP-10 processor. I am lacking code to test for: Foonly F1, F3, F4 Tymshare 26KL Systems Concepts SC20, SC30, SC40 KX10 emulator SIMH emulator Can anyone help fill in some of the missing tests? TITLE GETCPU Get CPU type T1==:1 T2==:2 P==:17 ; 1-10 for DEC (R.I.P.) $CP166==:1 ; PDP-6 $CPKA==:2 ; KA10 $CPKI==:3 ; KI10 $CPKL==:4 ; KL10 $CPKS==:5 ; KS10 ; 10-19 for Foonly (R.I.P.) $CPFO2==:11 ; Foonly F2 ; 20-29 for Systems Concepts ; 30-39 for XKL $CPXK1==:30 ; XKL TOAD-1 ; 40-49 for KLH10 $CPKHL==:40 ; KLH KN10-KL ; Get CPU type ; PUSHJ P,GETCPU ; T1/ CPU type code ; This routine is designed to be easy to modify/extend at the cost of some ; code size. GETCPU::PUSH P,T2 PUSHJ P,GETCP0 POP P,T2 POPJ P, GETCP0: JFCL 17,.+1 ; Clear flags JRST .+1 ; Change PC JFCL 1,GTC166 ; PDP-6 has PC Change flag SETO T1, ; AOBJN of -1 AOBJN T1,.+1 ; KA10 and Foonly F2 carries to left half JUMPE T1,GETCP1 HRLOI T1,010700 ; IBP of EA = -1 IBP T1 TLNE T1,17 ; KA10 overflows JRST GTCKA JRST GTCFO2 ;;; assume Foonly F2 ; Here with T1/0, always GETCP1: BLT T1,0 ; T1 must not be 0 JUMPE T1,GTCKI MOVSI T1,400000 ; ADJBP of normalized pointer by SETZ ADJBP T1,[430100,,0] CAMN T1,[430100,,0] ; KL10 leaves unchanged pointer JRST GTCKL MOVSI T1,450000 ; IBP of OWGBP IBP T1 CAMN T1,[450000,,0] ; KS10 doesn't do OWGBPs JRST GTCKS MOVSI T1,400000 ; SETZ divided by -1 SETO T2, IDIVM T1,T2 ; KLH-KL gets error and don't modify T2 AOJE T2,GTCKHL JRST GTCXK1 ;;; assume XKL-1 GTC166: MOVEI T1,$CP166 POPJ P, GTCKA: MOVEI T1,$CPKA POPJ P, GTCKI: MOVEI T1,$CPKI POPJ P, GTCKL: MOVEI T1,$CPKL POPJ P, GTCKS: MOVEI T1,$CPKS POPJ P, GTCFO2: MOVEI T1,$CPFO2 POPJ P, GTCKHL: MOVEI T1,$CPKHL POPJ P, GTCXK1: MOVEI T1,$CPXK1 POPJ P, END -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 19155 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!sn-xit-03!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10,alt.sys.pdp11,alt.sys.pdp8,comp.os.vms Subject: SIMH 2.10-4 released: major bug fixes to PDP-8, PDP-11, VAX, PDP-15, Interdata Date: Tue, 04 Mar 2003 09:59:30 -0500 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: Forte Agent 1.91/32.564 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 37 Xref: dfw-artgen alt.sys.pdp10:19155 alt.sys.pdp11:7532 alt.sys.pdp8:5823 comp.os.vms:175248 SIMH V2.10-4 was posted on http://simh.trailing-edge.com last night. This release includes significant bug fixes, in particular: PDP-11/VAX - fixed problem booting simulated tape images on the TQK50. Both standalone BACKUP and Ultrix booting have been tested. In detail: - Fixed queue ordering problem in TQK50 and RQDX3 - Fixed modifier checking problem in TQK50 - Fixed vector calculation problem (VMS only) in TQK50 and RQDX3 - Added user-defined drive types for TQK50 and RQDX3 - Fixed autosize algorithm in RX211 PDP-15 - fixed problems related to DOS-15. In detail: - Fixed three EAE bugs - Fixed autosize algorithm in RF15 - Fixed interrupt, BOT handling in TC59 PDP-8: - Fixed autosize algorithm in RF08, DF32, RX28 - Fixed BOT handling in TM8E, checked tape simulation with OS/8 Interdata - Unix V6 and V7 kits debugged and released. In detail: - Fixed precision clock calibration algorithm - Fixed handling of disk cylinder overflow errors on reads - Fixed tape handling of tape marks on writes SCP: - added .ini file capability - added multiple actions per breakpoint - fixed bug in multiword deposit to file - added magtape emulation library Misc: all line clocks can be set to 50Hz or 60Hz. /Bob Supnik Article: 19231 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!c03.atl99!rip!news.webusenet.com!prodigy.com!newsmst01.news.prodigy.com!prodigy.com!postmaster.news.prodigy.com!newssvr21.news.prodigy.com.POSTED!3a2e9f9f!not-for-mail User-Agent: Microsoft-Outlook-Express-Macintosh-Edition/5.0.5 Subject: Re: Any news on David Conroy's PDP-10 clone? From: "David G. Conroy" Newsgroups: alt.sys.pdp10 Message-ID: References: <3e6e155c@post.newsfeed.com> Mime-version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Lines: 25 NNTP-Posting-Host: 66.123.169.88 X-Complaints-To: abuse@prodigy.net X-Trace: newssvr21.news.prodigy.com 1047441794 ST000 66.123.169.88 (Tue, 11 Mar 2003 23:03:14 EST) NNTP-Posting-Date: Tue, 11 Mar 2003 23:03:14 EST Organization: Prodigy Internet http://www.prodigy.com X-UserInfo1: OHZACY_DVZVS@^LYMRKNOPDA[X_LPO@FDY^L\UQHWIWDUWYADNVOPCKZBL\NX_KHV^GY[KVMG^ZPNHSCZNS[^UXFJVWYXVXKBH[XRWWBBDTN@AX\JSBVH]_@T\EKJHBMZ\_WZJFNRY]YWKSPED_U^NC\HSZ\WS[KEAYI@DO@\K@BP\LD[\GTMPLDFVU]ASJM Date: Wed, 12 Mar 2003 04:03:14 GMT Xref: dfw-artgen alt.sys.pdp10:19231 Last time I posted, I was running ITS, with new drivers for modern (16c450 console, ATA disk) emulated peripherals, on the high-level model (micromachine register-level emulation, emulating the real microcode). I've been very busy with other things (both work related and non-work related), but I'm pretty much at this state on a low-level model (registers and control equations) as well. Most of the work getting to this state was redesigning the micropipeline, taking time from the ALU and giving it to the AC RAM address calculation and read access (the former is pretty complicated since it needs to deal with PXCT various +N operations). There are still three microinstructions in flight at the same time, but I did need to add a bypass for the micro-condition-codes so that a branch right after an ALU operation got to use the *new* codes (a similar case already existed for the EA<20 check to avoid needed a hazard-avoiding NOP in the main indirect address calculation flow). Recently I been thinking about packaging, and looking at some routing and signal integrity issues, mainly on the 32Kx8 SRAMS which make up the control store. dgc Article: 19480 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: How To ??? Date: 11 Apr 2003 21:00:07 GMT Organization: Aracnet Lines: 20 Message-ID: References: NNTP-Posting-Host: p-972.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:19480 John R. Strohm wrote: > Does anyone have IN ONE PLACE a list of the pieces I have to download and a > procedure for putting them together to end up with a running ITS on a > simulated KS10 or KLH10? > Same question for TOPS-10 (or TWENEX). > I already have the emulators. I need the OS and applications and an > "assembly procedure". I've got a webpage that should point you to everything you'll need. http://www.aracnet.com/~healyzh/pdp10emu.html The page has pointers to emulators, software, and documentation (actually the page has pointers to just about every piece of PDP-10 related info on the web). BTW, TWENEX isn't available, but TOPS-20 is. TOPS-20 V7 on KLH10 is quite nice! Zane Article: 19973 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!dfw-peer!news.verio.net!newsfeed!arclight.uoregon.edu!logbridge.uoregon.edu!news.u.washington.edu!140.142.17.34.MISMATCH!news.u.washington.edu!Ikkoku-Kan.Panda.COM!mrc From: Mark Crispin Newsgroups: alt.sys.pdp10 Subject: Re: Hardware required for general KLH10 use Date: Wed, 21 May 2003 23:18:45 -0700 Organization: University of Washington Lines: 13 Sender: mrc@Ikkoku-Kan.Panda.COM Message-ID: References: NNTP-Posting-Host: nntp2.u.washington.edu Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: nntp1.u.washington.edu 1053584334 37494 (None) 140.142.17.39 X-Complaints-To: help@cac.washington.edu In-Reply-To: Xref: dfw-artgen alt.sys.pdp10:19973 On Wed, 22 May 2003, Zane H. Healy wrote: > He's stated that for every 200Mhz you get roughly 1x the > speed of a KL10, so a 1Ghz Pentium III would be roughly 5x the speed of a > KL10. I don't know how accurate this is. This seems conservative to me. I get closer to 100MHz to equal a KL10, so 1GHz is about 10x KL10 speed. -- Mark -- http://staff.washington.edu/mrc Science does not emerge from voting, party politics, or public debate. Article: 20147 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!dfw-peer!news.verio.net!newsfeed!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: Getting started guide? Date: 27 May 2003 22:26:28 GMT Organization: Aracnet Lines: 22 Message-ID: References: NNTP-Posting-Host: p-973.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:20147 Edward A. Falk wrote: > I haven't touched a DEC-10 in 20 years, but I think I'd like to set > one up for old time's sake. Is there a guide? Which emulator is > the emulator of choice for beginners? Where do I get OS's and how do > I install them? How do I administer them? > I think this would make a good subject for an FAQ. Simply go to the following webpage: http://www.aracnet.com/~healyzh/pdp10emu.html There are links to everything you need. I recommend using the Monitor Install Guides rather than the cheat sheets that I and others have written up. As far as I know, all the links are up to date, I've not had time to check them lately. Oh, as for the best emulator, I prefer SIMH for TOPS-10 and KLH10 for TOPS-20 (I have yet to find the time to figure out ITS). Zane Article: 20205 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!dfw-peer!news.verio.net!newsfeed!pln-e!spln!dex!extra.newsguy.com!newsp.newsguy.com!not-for-mail From: "Zane H. Healy" Newsgroups: alt.sys.pdp10 Subject: Re: Getting started guide? Date: 31 May 2003 01:17:53 GMT Organization: Aracnet Lines: 13 Message-ID: References: <06e6bb.1g2.ln@escape.shannon.net> NNTP-Posting-Host: p-841.newsdawg.com User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (Linux/2.2.22 (i686)) Xref: dfw-artgen alt.sys.pdp10:20205 Charles Shannon Hendrix wrote: > Were there no good books on TOPS20? > I remember we had a decent one in college, but it was done by some other > school. > I'd give a lot to have it right now, because they had good guides on > TOPS-10, VMS, UNIX, and some IBM stuff too. Any chance this was the TOPS-20 portion? http://www.36bit.org/dec/third_party_doc/dundee_pi16.txt Zane Article: 21293 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!pln-e!extra.newsguy.com!lotsanews.com!news.maxwell.syr.edu!feed2.news.rcn.net!rcn!elnk-atl-nf1!newsfeed.earthlink.net!newsfeeds-atl2!news.webusenet.com!cyclone01.bloor.is.net.cable.rogers.com!news01.bloor.is.net.cable.rogers.com.POSTED!not-for-mail From: "Bill" Newsgroups: alt.sys.pdp10 References: Subject: Re: TOPS-10 V7.03 pre-built - no job search list Lines: 46 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1106 X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Message-ID: Date: Tue, 12 Aug 2003 19:30:43 GMT NNTP-Posting-Host: 24.156.110.159 X-Complaints-To: abuse@rogers.com X-Trace: news01.bloor.is.net.cable.rogers.com 1060716643 24.156.110.159 (Tue, 12 Aug 2003 15:30:43 EDT) NNTP-Posting-Date: Tue, 12 Aug 2003 15:30:43 EDT Xref: dfw-artgen alt.sys.pdp10:21293 Many thanks!!!!!!!! I had missed the fact that STRUCTURE-QUOTAS was not defined in these new users' profiles and that that was the key to getting the search list set up too. BTW, you would think this stuff would be in Operator's Command Manual as you suggest, but in fact it's hidden within the Monitor Installation Guide (MIG)... Turned out it was a bit more complex than just adding DSKB to the STRUCTURE-QUOTAS as I also had to struggle through figuring out how to CATALOG the structure first with OPR ("%REASNC, Structure not cataloged"). But now all is well ... learning more about ancient operating systems than I ever needed to know :) Thanks again for pointing me in the right direction ! wrote in message news:bhag37$h2t$2@bob.news.rcn.net... > In article , > "Bill" wrote: > >I wonder if someone can help with a minor TOPS-10 problem ... this is the > >prebuilt 7.03 that has been kicking around the net (originally from Paul > >Allen???) ... any user except [1,2] logging in does not get a job search > >list automatically set up ... DSK: seems to point nowhere ... no error is > >indicated on login except "%WLDEM Search list is empty"; it is necessary > to > >manually run SETSRC after each login thusly: > >..R SETSRC > >*A DSKB > > > >I have looked thru the docs and am at a loss. Any thoughts? > > It doesn't have anything to do with the monitor. hmm...running > 7.03...look at the Operator's Command Manual for entering PPN > search lists in the accounting files running REACT. In pre-7.03 > days, the file where this info is stored would AUXACC.SYS. I > can't recall the name of the file that replaced all of them > when the ACTDAE started doing the file changes. > > /BAH > > Subtract a hundred and four for e-mail. Article: 21336 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!newsfeed.media.kyoto-u.ac.jp!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!sn-xit-03!sn-xit-01!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: m_thompson-NOSPAM@ids.net (Michael Thompson) Newsgroups: alt.sys.pdp10 Subject: Re: newbie q: Installing TOPS-10 onto SIMH Date: Mon, 25 Aug 2003 12:13:17 -0000 Organization: Posted via Supernews, http://www.supernews.com Message-ID: X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit) References: <3f494475$0$45999$65c69314@mercury.nildram.net> <3f49f41b$0$46013$65c69314@mercury.nildram.net> MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII X-Complaints-To: abuse@supernews.com Lines: 142 Xref: dfw-artgen alt.sys.pdp10:21336 In article <3f49f41b$0$46013$65c69314@mercury.nildram.net>, withheld@not-for-mail.example.com says... > >Timothy Stark wrote: > >> >> "Martin" wrote in message >> news:3f494475$0$45999$65c69314@mercury.nildram.net... >>> I did find something at: >>> http://www.inwap.com/pdp10/emulators/ts10-703-install.txt >>> >>> This is aimed at the TS10 emulator (couldn't get that to boot off the >>> "magtape") which is fairly close; and with some guesswork I got SIMH to >>> boot the first magtape and copy some CUSP files from the next two tapes. >>> But the instructions don't seem quite right, and SIMH just hung when I >>> tried to boot from disk after that. >> >> Martin, >> >> Which version of ts10 emulator did you try boot off the "magtape"? I will >> look into that and fix any bugs. Try stable version instead of >> developement version. Also, I currently am working on new entire ts10 >> core >> for hybrid C/C++ interface to support class statements. With C++ version, >> VAX emulator is faster than C version. >> >> Thank you! >> Tim Stark > >Tim, > >I took the latest TS10 from Sourceforge.net (ts10-021004.tgz from the FILES >area, not from CVS) and it built straight off on Red Hat 9, no problems. > >I edited the ks10.cfg provided, so that RPA0 was in the local directory, not >/opt, and added RPA1 for good measure. > >Then I fired it up: > ./ts10 -f ks10.cfg > >It starts up OK but I can't assign MTA0: to the boot tape image from >http://pdp-10.trailing-edge.com - see output pasted below ("Reason: 6"). >I'm probably just doing something silly. (I notice the two files used for >RPA0 and RPA1 do get created, but are zero-length!) > >I was amazed to find that there are at least 3 emulators for this old beast: >TS10, SIMH and KLH10. In the end I got KLH10 to work with the "twonky" >disk image, then managed to install BASIC from one of the tape images on >trailing-edge. > >So after a gap of about 20 years, I am able once again to run things like >"SYSTAT", and "R BASIC" at the dot prompt! I also had a quick go with >RUNOFF, though since I can't remember the dot-commands it didn't produce >much output. But I guess if I search the web enough, I can probably find >the RUNOFF manual, and then annoy people at work by using RUNOFF to produce >my reports. It would certainly be more stable than MS Word. But the >laser-printer output might not be as pretty. > >Amazing how things progress. The standard was 110 baud dialup then, now >it's 57000 baud analog or 64000 baud ISDN, and we still consider it >annoying that ADSL isn't available here. > >Does anyone have some favourite applications to suggest from the DECUS tape >images? It would be great to see what this beast can do, after all these >years. > >Anyone else here ever use the DECsystem 10 (1092?) at Hatfield Polytechnic >(UK)? It was replaced by a DECsystem 2020 that we never used, as we were >moved onto a PDP-11/70 running RSTS/E v7.0. I later had a summer job at a >DEC reseller, and installed RSTS/E on a few PDP-11 and LSI-II boxes. I >guess it might be fun to do that again in emulation sometime - though the >PDP11 lacks the vintage charm of 36 bits! > >I really admire the skill and dedication of Timothy and others who've >managed to make these emulators run in such an impressive fashion. It's >amazing that this is possible at all. Funny how a mere laptop can run >emulated code faster than a huge room full of the original hardware. > >All the best, > >- Martin Johnson. > >======================================================== > >[root@speedy ts10]# ./ts10 -f ks10.cfg >TS10 Emulator System v0.8.7 (Late Alpha) > >Created PDP10 on SYS0: - DECsystem-10/20 Emulator v0.8.7 (Alpha) >CPU Flags = 00000010 >Activating console on TCP port 5000. >Activating KLINIK on TCP port 5001. >Created KS10 on CPU0: - KS10 Emulator v0.8.9 (Late Alpha) >Created UBA on UBA: - Unibus Interface v0.1 (Pre-Alpha) >Configured UBA on UBA1: - Unibus Interface v0.1 (Pre-Alpha) >Configured UBA on UBA3: - Unibus Interface v0.1 (Pre-Alpha) >Selected UBA1: (UBA) on SYS0: - Unibus Interface v0.1 (Pre-Alpha) >RHA0: CSR = 776700, MASK = 77, VEC = 254, IPL = 8 >Created RH11 on RHA0: - Disk/Tape Massbus Controller v0.8 (Alpha) >Selected UBA3: (UBA) on SYS0: - Unibus Interface v0.1 (Pre-Alpha) >RHA1: CSR = 772440, MASK = 37, VEC = 224, IPL = 9 >Created RH11 on RHA1: - Disk/Tape Massbus Controller v0.8 (Alpha) >Created TCU150 on CLK: - Timing Control Unit v0.7 (Alpha) >Selected RHA0: (RH11) on SYS0: - Disk/Tape Massbus Controller v0.8 (Alpha) >RPA0: Device RP06 (RP) - Created. >Configured RP06 on RPA0: - RP06 - 176MB Disk Pack Drive >RPA1: Device RP06 (RP) - Created. >Configured RP06 on RPA1: - RP06 - 176MB Disk Pack Drive >Selected RHA1: (RH11) on SYS0: - Disk/Tape Massbus Controller v0.8 (Alpha) >MTA0: Device TM03 (TM) - Created. >Configured TM03 on MTA0: - TM03 - Tape Controller >MTA1: Device TM03 (TM) - Created. >Configured TM03 on MTA1: - TM03 - Tape Controller >Debug Console had been turned on. > >Device Type Description >------ ---- ----------- >CLK: TCU150 Timing Control Unit v0.7 (Alpha) >CPU0: KS10 KS10 Emulator v0.8.9 (Late Alpha) >MTA0: TM03 TM03 - Tape Controller >MTA1: TM03 TM03 - Tape Controller >RHA0: RH11 Disk/Tape Massbus Controller v0.8 (Alpha) >RHA1: RH11 Disk/Tape Massbus Controller v0.8 (Alpha) >RPA0: RP06 RP06 - 176MB Disk Pack Drive >RPA1: RP06 RP06 - 176MB Disk Pack Drive >UBA: UBA Unibus Interface v0.1 (Pre-Alpha) >UBA1: UBA Unibus Interface v0.1 (Pre-Alpha) >UBA3: UBA Unibus Interface v0.1 (Pre-Alpha) > >Total 11 devices. > >TS10> attach mta0: tops10_703_ksbootable_bb-x138b-bb.tap >Reason: 6 >TS10> boot mta0: >Not Bootable >TS10> The machines at Hatfield that are in the Serial Number List are: 143 TOPS-10 Hatfield Polytechnic Hatfield, Herts, UK KA10 2337 TOPS-10 Hatfield Polytechnic Hatfield, Herts, UK KL1091 4249 TOPS-10 Hatfield Polytechnic Hatfield, Herts, UK KS10 Article: 21342 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!HSNX.atgi.net!newsfeed.media.kyoto-u.ac.jp!in.100proofnews.com!in.100proofnews.com!cycny01.gnilink.net!cyclone1.gnilink.net!small1.nntp.aus1.giganews.com!border1.nntp.aus1.giganews.com!intern1.nntp.aus1.giganews.com!nntp.giganews.com!nntp.speakeasy.net!news.speakeasy.net.POSTED!not-for-mail NNTP-Posting-Date: Mon, 25 Aug 2003 22:40:11 -0500 From: "Timothy Stark" Newsgroups: alt.sys.pdp10 References: <3f494475$0$45999$65c69314@mercury.nildram.net> <3f49f41b$0$46013$65c69314@mercury.nildram.net> Subject: Re: newbie q: Installing TOPS-10 onto SIMH Date: Mon, 25 Aug 2003 23:40:10 -0400 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 Message-ID: Lines: 29 NNTP-Posting-Host: 66.92.145.31 X-Trace: sv3-hBSWWDNtlOp9UWkdc9URxh5wMxyZAIrGRK3m3ohR0OxQXMExYU6u5E27gbSIDyayUzN4vbt9HRnXkBF!qRfSbP+j+bXGaE+0CS4Dkk3KR6Olim9eNuss0rL5LInRFDcSE9th83jAYzCoPgJfR/geoYYFnTch!jGmtmkPeCVFRjec7 X-Complaints-To: abuse@speakeasy.net X-DMCA-Complaints-To: abuse@speakeasy.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:21342 "Martin" wrote in message news:3f49f41b$0$46013$65c69314@mercury.nildram.net... > Timothy Stark wrote: > I took the latest TS10 from Sourceforge.net (ts10-021004.tgz from the FILES > area, not from CVS) and it built straight off on Red Hat 9, no problems. > > I edited the ks10.cfg provided, so that RPA0 was in the local directory, not > /opt, and added RPA1 for good measure. > > Then I fired it up: > ./ts10 -f ks10.cfg > > It starts up OK but I can't assign MTA0: to the boot tape image from > http://pdp-10.trailing-edge.com - see output pasted below ("Reason: 6"). > I'm probably just doing something silly. (I notice the two files used for > RPA0 and RPA1 do get created, but are zero-length!) Ok, I will look into that and will fix bugs. Thank you for let me know about new bugs in MTA emulation. Also, latest snapshots for both devel and stable are available at ftp.firesword7.net/pub/ts10. That version is developement stage. Try stable version and let me know what happens. Thank you. Tim Stark Article: 21337 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!sn-xit-03!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: newbie q: Installing TOPS-10 onto SIMH Date: Mon, 25 Aug 2003 10:27:01 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <3f494475$0$45999$65c69314@mercury.nildram.net> X-Newsreader: Forte Agent 1.93/32.576 English (American) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 40 Xref: dfw-artgen alt.sys.pdp10:21337 The TOPS-10 Monitor Installation Guide for 7.03 can be found at Zane Healy's site: http://www.aracnet.com/~healyzh/pdp10emu.html There is also a pointer to the MIG for 7.04. Both are accurate enough to install a monitor from the trailing-edge.com archive tapes. Note that there is an install bug in 7.04 and a work-around for it; see the SIMH PDP-10 documentation. /Bob Supnik On Mon, 25 Aug 2003 00:04:12 +0100, Martin wrote: >Under RedHat 9.0, I've just built SIMH from : > http://simh.trailing-edge.com/ >and found plenty of TOPS tape images at: > http://pdp-10.trailing-edge.com/ > >Does anyone have a worked example of how to get TOPS installed? > >I did find something at: >http://www.inwap.com/pdp10/emulators/ts10-703-install.txt > >This is aimed at the TS10 emulator (couldn't get that to boot off the >"magtape") which is fairly close; and with some guesswork I got SIMH to >boot the first magtape and copy some CUSP files from the next two tapes. >But the instructions don't seem quite right, and SIMH just hung when I >tried to boot from disk after that. > >Not sure what I'd use this for of course, but the DEC-10 was the first >machine I ever used, back in the early 80's with a 110 baud ASR33 teletype! >Kids today, with their fancy bit-mapped displays... Don't know they're >born. > >Cheers > >- Martin. Article: 21343 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!in.100proofnews.com!in.100proofnews.com!elnk-atl-nf1!elnk-nf2-pas!elnk-pas-nf1!newsfeed.earthlink.net!sn-xit-02!sn-xit-06!sn-post-01!supernews.com!corp.supernews.com!not-for-mail From: Bob Supnik Newsgroups: alt.sys.pdp10 Subject: Re: newbie q: Installing TOPS-10 onto SIMH Date: Tue, 26 Aug 2003 09:08:14 -0400 Organization: Posted via Supernews, http://www.supernews.com Message-ID: References: <3f494475$0$45999$65c69314@mercury.nildram.net> X-Newsreader: Forte Agent 1.93/32.576 English (American) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Complaints-To: abuse@supernews.com Lines: 26 Xref: dfw-artgen alt.sys.pdp10:21343 Sorry... wrong context. There is a bug in 7.04 installation. In counting the number of magtape units, the code overflows the CSR field into the next bit, which is the inhibit increment bit. This impacted Tim's debugging of TS10. SIMH, in one of its rare simplifications, didn't implement the inhibit increment bit, allowing 7.04 to work "properly". /Bob On 25 Aug 2003 13:47:02 -0400, Rich Alderson wrote: >Bob Supnik writes: > >> There is also a pointer to the MIG for 7.04. Both are accurate enough >> to install a monitor from the trailing-edge.com archive tapes. Note >> that there is an install bug in 7.04 and a work-around for it; see the >> SIMH PDP-10 documentation. > >Bob, I just looked through the SIMH docs again, and I can't find any reference >to a 7.04 install bug. What am I missing? > >BTW, I've run all of ITS, Tops-20 v4.1 and Tops-10 v7.04 on your simulator now. >Thanks so much! Article: 21346 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!newsfeed.berkeley.edu!ucberkeley!cyclone-sf.pbi.net!216.218.192.242!news.he.net!newsfeed1.easynews.com!easynews.com!easynews!small1.nntp.aus1.giganews.com!border1.nntp.aus1.giganews.com!intern1.nntp.aus1.giganews.com!nntp.giganews.com!nntp.speakeasy.net!news.speakeasy.net.POSTED!not-for-mail NNTP-Posting-Date: Tue, 26 Aug 2003 18:32:24 -0500 From: "Timothy Stark" Newsgroups: alt.sys.pdp10 References: <3f494475$0$45999$65c69314@mercury.nildram.net> Subject: Re: newbie q: Installing TOPS-10 onto SIMH Date: Tue, 26 Aug 2003 19:32:22 -0400 X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 Message-ID: Lines: 23 NNTP-Posting-Host: 66.92.145.31 X-Trace: sv3-t2u1jg0rG0Ng5+kl1cez5iqseoNIyXWT4e6otM4Hyt617STJlG6BJAXAPIt5uU/9bqCwasUv5+e5+M3!GAJHbmxNotM157rG0DzO7cDAqpCUQb7NkNH557RxV0LmQyCqCOasq08hQ2wxeMlxhqaFHTVEFxoQ!wwnm1kYE7BUgSzDo X-Complaints-To: abuse@speakeasy.net X-DMCA-Complaints-To: abuse@speakeasy.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:21346 "Bob Supnik" wrote in message news:opmmkvo46c2igjhvur4iftttpk8d1kjas9@4ax.com... > Sorry... wrong context. > > There is a bug in 7.04 installation. In counting the number of > magtape units, the code overflows the CSR field into the next bit, > which is the inhibit increment bit. This impacted Tim's debugging of > TS10. SIMH, in one of its rare simplifications, didn't implement the > inhibit increment bit, allowing 7.04 to work "properly". > > /Bob Bob, Yeah. I am aware of that because I discovered a bug in TOPS-10 7.04 monitor code. I already fixed it by remove inhibit increment bit in TU emulation code in MBA package before. All problems disappeared. Now TOPS-10 7.04 is working so well on TS10 emulator right now. Tim Article: 21476 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!in.100proofnews.com!in.100proofnews.com!news.sprintnetops.net!not-for-mail From: "Geoffrey G. Rochat" <777geoff777@777pkworks777.777com777> Newsgroups: alt.sys.pdp10 Subject: Re: Imlac PDS-1 Date: Fri, 26 Sep 2003 12:54:56 -0400 Organization: Sprint Advanced Network Services Lines: 7 Message-ID: References: Reply-To: "Geoffrey G. Rochat" <777geoff777@777pkworks777.777com777> NNTP-Posting-Host: 208.244.111.136 X-Trace: news.utelfla.com 1064594641 19020 208.244.111.136 (26 Sep 2003 16:44:01 GMT) X-Complaints-To: usenet@news.utelfla.com NNTP-Posting-Date: Fri, 26 Sep 2003 16:44:01 +0000 (UTC) X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Xref: dfw-artgen alt.sys.pdp10:21476 > I would like to see an emulator for it. Heck, I'd even like to see a scan > of the manuals. http://www.spies.com/~aek/pdf/imlac/ Article: 21471 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!newsfeed.berkeley.edu!ucberkeley!priapus.visi.com!news-out.visi.com!petbe.visi.com!news.maxwell.syr.edu!newsfeed.stanford.edu!postnews1.google.com!not-for-mail From: shoppa@trailing-edge.com (Tim Shoppa) Newsgroups: alt.sys.pdp10,alt.sys.pdp11,vmsnet.pdp-11,alt.folklore.computers Subject: Re: Hurricane takes SIMH, PDP-11, PDP-10 archives offline Date: 25 Sep 2003 09:26:14 -0700 Organization: http://groups.google.com/ Lines: 14 Message-ID: References: NNTP-Posting-Host: 170.121.15.35 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Trace: posting.google.com 1064507177 21921 127.0.0.1 (25 Sep 2003 16:26:17 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: 25 Sep 2003 16:26:17 GMT Xref: dfw-artgen alt.sys.pdp10:21471 alt.sys.pdp11:8667 vmsnet.pdp-11:30007 alt.folklore.computers:137313 shoppa@trailing-edge.com (Tim Shoppa) wrote in message news:... >... And nearly a week later, most everything is back up. Within seconds of the mail service being enabled I was flooded with literally hundreds of connections from spam and virus-infected Windows boxes, so mail is temporariliy off until I can figure out how to throttle that junk, in preference of keeping the various archives and Bob's SIMH stuff available. There are still some major intersections without traffic lights near my house, and several side roads still have down trees across them, but things are getting cleared up. Tim. Article: 21501 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!pln-e!extra.newsguy.com!lotsanews.com!skynet.be!skynet.be!newsgate.cistron.nl!news.cambrium.nl!news.cambrium.nl!news.cambrium.nl!news2.euro.net!fi.sn.net!newsfeed1.fi.sn.net!uio.no!news.kth.se!not-for-mail From: Johnny Eriksson Newsgroups: alt.sys.pdp10 Subject: Re: SIMH CPU patch Date: 30 Sep 2003 18:00:10 GMT Organization: Royal Institute of Technology, Sweden Lines: 179 Sender: bygg@gustavskorv.stacken.kth.se Message-ID: References: <3f79b202$0$17180$afc38c87@news.easynet.co.uk> NNTP-Posting-Host: gustavskorv.stacken.kth.se Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii User-Agent: Gnus/5.0803 (Gnus v5.8.3) Emacs/21.2 Xref: dfw-artgen alt.sys.pdp10:21501 lawrie@riffraff.plig.net (Michael Lawrie) writes: > Ok... I give up, I am resorting to posting to here to ask. I have found > a couple of previous threads that have discussed this but there don't > seem to have been any conclusions. > > I am running SIMH which seems to be my best bet for OpenBSD and FreeBSD > and so far, I like it - Buuuut, now I have had the simulator thrown off > three machines for being antisocial. > > The problem is, even at nice 15, it still uses every teensy bit of CPU > time it can possibly grab. On a machine that doesn't do that much it > will sit using 98% of the CPU regardless of the priority it is on and > even though the machine being simulated is completely idle. On a > dedicated machine this is fine but on a timesharing system where the > systems manager actually does CPU accounting, this is at the least > hard to explain and at worse, "expensive". > > I have seen a couple of discussions about putting a microsecond sleep > instruction in every 100 CPU ticks or so, but there don't seem to be > any actual patches and I could delve into the code, but I don't really > want to screw something up fairly fundamentally if someone else has > already investigated this properly and come up with a solution. You did not tell us what OS you run on your '10. I use Tops-10, and have conjured up the following: (yes, this is for simh version 2.9). *** pdp10_cpu.c Sun Feb 24 09:35:09 2002 --- dist/PDP10/pdp10_cpu.c Thu Jan 3 19:33:44 2002 *************** *** 182,191 **** extern int32 sim_interval; extern int32 sim_brk_types, sim_brk_dflt, sim_brk_summ; /* breakpoint info */ extern UNIT tim_unit; - - int32 sleep_interval = 0; - int32 idlefast = 10; - t_bool idleflag = TRUE; /* Forward and external declarations */ --- 182,187 ---- *************** *** 369,376 **** { FLDATA (T20V41, cpu_unit.flags, UNIT_V_T20V41), REG_HRO }, { ORDATA (WRU, sim_int_char, 8) }, { FLDATA (STOP_ILL, stop_op0, 0) }, - { FLDATA (IDLEFLAG, idleflag, 0) }, - { DRDATA (IDLEFAST, idlefast, 16) }, { BRDATA (REG, acs, 8, 36, AC_NUM * AC_NBLK) }, { NULL } }; --- 365,370 ---- *************** *** 583,657 **** #define POPF if (LRZ (AC(ac)) == RMASK) SETF (F_T2) #define DMOVNF if (rs[1] == 0) { MOVNF (rs[0]); } - - /* Check for a standard null job at PC. */ - - t_bool nuljob(a10 PC) - { - /* typical null job: - - 1: sojg 6,1 ;367300,,1 - 2: movei 6,n ;201300,,n - 3: tdne 7,addr ;612340,,addr - 4: wake ;47000,,73 - 5: aoja 1 ;344000,,1 - - ; or - - 1: sojg 6,1 ;367300,,1 - 2: movei 6,n ;201300,,n - 3: aoja 1 ;344000,,1 - */ - - if (PC != 2) return FALSE; - if (ac_cur[1] != XWD(0367300, 1)) return FALSE; - if ((ac_cur[2] >> 18) != 0201300) return FALSE; - if (ac_cur[3] == XWD(0344000, 1)) return TRUE; - if ((ac_cur[3] >> 18) != 0612340) return FALSE; - if (ac_cur[4] != XWD(047000, 073)) return FALSE; - if (ac_cur[5] == XWD(0344000, 1)) return TRUE; - - return FALSE; - } - - #include - - extern int32 tmxr_poll; /* calibrated delay */ - - int32 sys_idle(int32 cycles) - { - struct timeval starttime, stoptime; - struct timespec zzz; - - uint32 nspi; /* Nano-seconds per instruction. */ - uint32 ips; /* Instructions per second. */ - uint32 sleeptime; /* Time slept, in usec. */ - - if (cycles <= 0) return 1; /* Silly request? */ - - ips = tmxr_poll * 50; /* UGH! */ - nspi = 1000000000 / ips; - - if (nspi <= 0) return 1; /* Fast hardware? */ - - zzz.tv_sec = 0; - zzz.tv_nsec = nspi * cycles; - /* zzz.tv_nsec = 1; */ - - (void) gettimeofday(&starttime, NULL); - (void) nanosleep(&zzz, NULL); - (void) gettimeofday(&stoptime, NULL); - - sleeptime = (stoptime.tv_sec - starttime.tv_sec) * 1000000; - sleeptime += stoptime.tv_usec; - sleeptime -= starttime.tv_usec; - sleeptime /= 1000; /* Convert to milliseconds. */ - if (sleeptime == 0) return 1; /* Short sleep. */ - - return ((ips * sleeptime) / 1000); - } - - t_stat sim_instr (void) { a10 PC; --- 577,582 ---- *************** *** 723,739 **** pager_tc = FALSE; /* not in trap cycle */ pflgs = 0; /* not in PXCT */ xct_cnt = 0; /* count XCT's */ - - if (sleep_interval > 0) { - if (sleep_interval > idlefast) { - sim_interval -= idlefast; - sleep_interval -= idlefast; - } else { - sim_interval -= sleep_interval; - sleep_interval = 0; - } - } - if (sim_interval <= 0) { /* check clock queue */ if (i = sim_process_event ()) ABORT (i); /* error? stop sim */ pi_eval (); } /* eval pi system */ --- 648,653 ---- *************** *** 1086,1096 **** case 0364: SOJ; JUMP(ea); break; /* SOJA */ case 0365: SOJ; if (TGE (AC(ac))) JUMP (ea); break;/* SOJGE */ case 0366: SOJ; if (TN (AC(ac))) JUMP (ea); break; /* SOJN */ ! case 0367: if (idleflag && (sleep_interval == 0) && nuljob(PC)) { ! sleep_interval = sys_idle(sim_interval); ! if (sleep_interval == 0) idleflag = FALSE; ! } ! SOJ; if (TG (AC(ac))) JUMP (ea); break; /* SOJG */ case 0370: SOS; break; /* SOS */ case 0371: SOS; if (TL (mb)) INCPC; break; /* SOSL */ case 0372: SOS; if (TE (mb)) INCPC; break; /* SOSE */ --- 1000,1006 ---- case 0364: SOJ; JUMP(ea); break; /* SOJA */ case 0365: SOJ; if (TGE (AC(ac))) JUMP (ea); break;/* SOJGE */ case 0366: SOJ; if (TN (AC(ac))) JUMP (ea); break; /* SOJN */ ! case 0367: SOJ; if (TG (AC(ac))) JUMP (ea); break; /* SOJG */ case 0370: SOS; break; /* SOS */ case 0371: SOS; if (TL (mb)) INCPC; break; /* SOSL */ case 0372: SOS; if (TE (mb)) INCPC; break; /* SOSE */ > Michael. --Johnny Article: 21504 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!hammer.uoregon.edu!news.algonet.se!algonet!news.tele.dk!news.tele.dk!small.news.tele.dk!oleane.net!oleane!easynet-quince!easynet.net!easynet-post1!not-for-mail Newsgroups: alt.sys.pdp10 Subject: Re: SIMH CPU patch References: <3f79b202$0$17180$afc38c87@news.easynet.co.uk> Organization: PLiG X-Newsreader: trn 4.0-test76 (Apr 2, 2001) From: lawrie@riffraff.plig.net (Michael Lawrie) Originator: lawrie@riffraff.plig.net (Michael Lawrie) Date: 30 Sep 2003 21:05:33 GMT Lines: 20 Message-ID: <3f79f01d$0$17179$afc38c87@news.easynet.co.uk> NNTP-Posting-Host: 195.40.6.40 X-Trace: DXC=?G?U[PYIX;Oi0NlgV@<_, Johnny Eriksson wrote: >You did not tell us what OS you run on your '10. > >I use Tops-10, and have conjured up the following: > >(yes, this is for simh version 2.9). > >*** pdp10_cpu.c Sun Feb 24 09:35:09 2002 >--- dist/PDP10/pdp10_cpu.c Thu Jan 3 19:33:44 2002 I use TOPS-10 as well - And applying this manually to V3.0-1 it so far works beautifully. I even saw it go down from 2% CPU time to 0.7% so, looks good. It never went below 98% before. Thanks! Saved my sanity for a while. >--Johnny Michael. Article: 21523 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!sjc-peer.news.verio.net!news.verio.net!newsfeed.berkeley.edu!ucberkeley!priapus.visi.com!news-out.visi.com!hermes.visi.com!feed2.news.rcn.net!rcn!news.maxwell.syr.edu!newsswitch.lcs.mit.edu!bloom-beacon.mit.edu!panix!not-for-mail From: Rich Alderson Newsgroups: alt.sys.pdp10 Subject: Re: Installing TOPS-20 on simh Date: 02 Oct 2003 13:14:05 -0400 Organization: Systems Administration, XKL LLC, Redmond WA 98052 Lines: 56 Sender: alderson+news@panix5.panix.com Message-ID: References: NNTP-Posting-Host: panix5.panix.com X-Trace: reader2.panix.com 1065114845 18666 166.84.1.5 (2 Oct 2003 17:14:05 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: Thu, 2 Oct 2003 17:14:05 +0000 (UTC) X-Newsreader: Gnus v5.7/Emacs 20.7 Xref: dfw-artgen alt.sys.pdp10:21523 Andreas Davour writes: > I have no great knowledge of pdp10 disks and devices and I have > forgotten all about what the klh10 config file means. > Anyone care to refreshen my memory on what disks and controllers I > should set, attach and otherwise fiddle with? > Are there any ini file "out there" I could use, even? I use the following. RP06.init assumes no disk image file (t20.dsk in mine), sets up two tape drives with the installation tapes, and goes from there. Afterwards, use RP06.boot to run with 32 terminals. I change the interrupt character to ^\ (the standard on the KL-10, KS-10, SC-30M, and Toad-1) from the default ^E (a bad choice for Tops-20, because there is a set of privileged commands prefixed with ^E). Have fun! RP06.init --------- set cpu tops20v41 d wru 034 at tu0 bb-d867e-bm_tops20_v41_2020_instl.tap at tu1 bb-d868e-bm_tops20_v41_2020_dist_1of2.tap set rp0 rp06 at rp0 t20.dsk b tu0 ^Z RP06.boot --------- set cpu tops20v41 d wru 034 set dz enable set dz lines=32 attach -am dz 5200 set rp0 rp06 at rp0 t20.dsk b rp0 ^Z -- Rich Alderson | /"\ ASCII ribbon | news@alderson.users.panix.com | \ / campaign against | "You get what anybody gets. You get a lifetime." | x HTML mail and | --Death, of the Endless | / \ postings | Article: 21561 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!washdc3-snh1.gtei.net!crtntx1-snh1.gtei.net!news.gtei.net!newsfeed1.easynews.com!easynews.com!easynews!small1.nntp.aus1.giganews.com!border3.nntp.aus1.giganews.com!intern1.nntp.aus1.giganews.com!nntp.giganews.com!nntp.comcast.com!news.comcast.com.POSTED!not-for-mail NNTP-Posting-Date: Wed, 22 Oct 2003 22:54:51 -0500 Date: Wed, 22 Oct 2003 21:54:56 -0600 From: e-mail@end_of.msg (Mike Cesari) Newsgroups: alt.sys.pdp10 Subject: Multi-disk boot structure for TOPS-10 on SIMH Message-ID: User-Agent: NewsWatcher-X 2.2.3b2 Lines: 187 NNTP-Posting-Host: 67.161.150.18 X-Trace: sv3-PuV7Hw5TbrGUwiFwiC+SUEx+mSEEP+LSnTFvckwodMTspeN6GFU4AZnHSbGdT+/dS8f8GvLoKepCffv!q+HM+3yPeuT7OdYHUoD3lD7tMrJrBlaxAxdp/n+rtoiD+cUg03ND8p6MB5c= X-Complaints-To: abuse@comcast.net X-DMCA-Complaints-To: dmca@comcast.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:21561 Hi all, I discovered how to get a multi-disk boot structure working (TOPS-10 v7.04 on simh). In all my googling I never saw a solution posted, so here is mine. These are snippits from an installation log file. (The actual log is rather big.) My notes start with "=== ". === (In this example, logical unit 0 [DSKB0:] is RPA0:) === The trick is where you tell SMFILE to put the boot stuff. === (See Section 6.2 (page 6-6) of the Monitor Installation Guide.) === Where the MIG says to use the structure name (dskb:) === use the name of logical unit 0 (dskb0:) instead: .r credir Create directory: dskb0:[6,2020]/exit Created DSKB0:[6,2020].UFD/PROTECTION:775 .run dec:smfile DECSYSTEM 2020 DIAGNOSTICS FE-FILE PROGRAM VERSION 2.7, TOPS-10, KS10, CPU#=4097 [FOR HELP TYPE "HELP"] SMFILE>write setup dskb0: SMFILE>write reset SMFILE>read dec:t10kl.ram SMFILE>serial 4097 SMFILE>write cram SMFILE>write boot dec:ksboot.exe SMFILE>write done SMFILE>exit EXIT . === Once you have finished the installation and rebooted, look at what === systat tells you: .systat Status of KS10 at 20:55:24 on 13-Oct-103 === System File Structures: Name Free Mount DSKB 1893740 2 Total Free 1893740 === You have lots of disk, but not a lot of core. I haven't seen the solution === to using all of simh's available memory posted here (and I've googled === this subject to death, too) so I'll summarize what I found: === 0) Start Chapter 9 in the MIG. === 1) Answer MONGEN dialog question #58 (pg. 9-8 in the MIG) with 1024 === 2) Complete Chapter 9 and reboot. === 3) From the OPR> prompt... OPR>enter ? one of the following: CATALOG CONFIG LCP NCP QUOTA OPR>enter config CONFIG>show hardware 20:57:45 CONFIG -- SHOW HARDWARE-CONFIGURATION -- CPU Configuration KS10 serial number 4097, microcode version 130 Cache status: Disabled UNIBUS adapters: UBA 1: High PIA: 3 Low PIA: 0 UBA 3: High PIA: 2 Low PIA: 2 Monitor Memory Configuration Monitor configured for 2048P (1024K) of physical memory of which 1024P (512K) is off-line Status Range (P) Range (K) ------- ----------- ----------- Online 0 - 1023 0 - 511 Offline 1024 - 2047 512 - 1023 === === OK, now put the rest of memory online... CONFIG>add memory 512k 1024k 20:58:50 CONFIG -- ADD MEMORY -- Memory from address 512K to 1024K is now online (Words from 2000000 to 4000000) === type "show hardware" again, and you'll see: 20:59:02 CONFIG -- SHOW HARDWARE-CONFIGURATION -- === Monitor Memory Configuration Monitor configured for 2048P (1024K) of physical memory Disk Configuration Drive Type Str Volume ------ ---- ---- ------ RPA0 RP06 DSKB DSKB0 RPA1 RP06 DSKB DSKB1 RPA2 RP06 DSKB DSKB2 RPA3 RP06 DSKB DSKB3 RPA4 RP06 DSKB DSKB4 RPA5 RP06 DSKB DSKB5 RPA6 RP06 DSKB DSKB6 === etc. === When you return to the monitor, edit the file sys:system.cmd and === add the line: === CONFIG ADD MEMORY 512K 1024K === but we're not finished yet. CATALOG needs to know about the rest of the === disks in the structure. You see above that I have 7 disks. They are all === part of DSKB:. But now look at what CATALOG tells us: CONFIG>return OPR>enter catalog CATALOG> Monitor Doofus System uptime :03:24 Current date/time Monday 13-Oct-103 21:00:00 CATALOG>? one of the following: ENTER EXIT PUSH RETURN TAKE WAIT or one of the following: DELETE DISABLE ENABLE HELP INSERT LIST MODIFY SHOW CATALOG> CATALOG>show struct dskb CATALOG> 21:01:01 -- Structure catalog -- Structure DSKB Owned by [1,2] Unit Type Class ------ ------ ----- DSKB0 RP06 0 === Hmmm., only one disk? Add the rest of the disks to the structure: CATALOG>mod str dskb CATALOG>>add dskb1 rp06 0 CATALOG>>add dskb2 rp06 0 CATALOG>>add dskb3 rp06 0 CATALOG>>add dskb4 rp06 0 CATALOG>>add dskb5 rp06 0 CATALOG>>add dskb6 rp06 0 CATALOG>>done CATALOG> 21:02:29 -- Structure DSKB modified -- CATALOG>sho str dskb CATALOG> 21:02:37 -- Structure catalog -- Structure DSKB Owned by [1,2] Unit Type Class ------ ------ ----- DSKB0 RP06 0 DSKB1 RP06 0 DSKB2 RP06 0 DSKB3 RP06 0 DSKB4 RP06 0 DSKB5 RP06 0 DSKB6 RP06 0 CATALOG>exit === Now we're finished. Many thanks to those who've created the simulators and those who've made the software and documentation available. I never had the chance to use or play with either TOPS-10 or TOPS-20. This is cool stuff. Mike -- mcesari (at) comcast(dot)net Article: 21562 of alt.sys.pdp10 Path: iad-read.news.verio.net!dfw-artgen!iad-peer.news.verio.net!news.verio.net!news.maxwell.syr.edu!small1.nntp.aus1.giganews.com!border1.nntp.aus1.giganews.com!intern1.nntp.aus1.giganews.com!nntp.giganews.com!nntp.comcast.com!news.comcast.com.POSTED!not-for-mail NNTP-Posting-Date: Thu, 23 Oct 2003 00:23:47 -0500 Date: Wed, 22 Oct 2003 23:23:53 -0600 From: e-mail@end_of.msg (Mike Cesari) Newsgroups: alt.sys.pdp10 Subject: Re: Multi-disk boot structure for TOPS-10 on SIMH Message-ID: References: User-Agent: NewsWatcher-X 2.2.3b2 Lines: 40 NNTP-Posting-Host: 67.161.150.18 X-Trace: sv3-Vt6g6rR2HDY48ddJxRwh30i6xidrLWr2rbGHdL86YTfNDKUWdDredFjOKDs9CyCSfzYagStv8dV7C4R!7Vy/nJwVCRqef+OmpvVlQzvKV3Lo/pcLEEibyGxDTk0hbIQGTuKZfDwc2IU= X-Complaints-To: abuse@comcast.net X-DMCA-Complaints-To: dmca@comcast.net X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly X-Postfilter: 1.1 Xref: dfw-artgen alt.sys.pdp10:21562 Hi, again, I forgot one of those little details that when they bite, they bite _hard_: After the structure has been defined (during ONCE) several status messages are printed. Among these is this: % No units in active swapping list Typing "show asl" gives: Startup option: show asl Swapping Units: Unit UnitID Class Space (K) ------ ------ ----- --------- RPA0 DSKB0 0 4807 (Memory size = 512K) Huh. Well, changing the asl makes ONCE happy: Startup option: change asl Enter names of units to appear in the active swapping list, one per line. Terminate list with a blank line. rpa0 [Rebuilding the system search list from the HOM blocks] [Rebuilding the system dump list from the HOM blocks] Sorry about the omission, Mike -- mcesari (at) comcast(dot)net