Modified Supnik emulator for the 11

Hi folks. Having had absolutely no luck getting the Begemot emulator to work under FreeBSD 4.0-CURRENT, I've been modifying the Supnik 2.3d emulator for the pdp-11. So far, I've replaced the KL11 code with a DL11 driver that handles four lines. Additionally, I stole the networked tty_net driver from the begemot which now provides telnet access to all four ports. Additionally, I'm working on a DZ-11 driver as-we-speak, and will do the DEQNA next. These four ports work great on the RSTS/E and 2.11 images I have. As well, I've tweaked the clock timing to significantly improve timekeeping for my machine. Also, I have been modifying an ANSI magtape util package (ansir/ansiw/survey) to deal with the mt images that the supnik package produces. Makes for easy exchange into RSTS, etc. Is anyone else out the hacking it up and interested in sharing any work? And if anyone has managed to get the Begemot emulator working on recent FreeBSD-4.0-CURRENT versions, I'd be grateful if you could share the information and changes with me. Thanks and regards, -skots -- Scott G. Akmentins-Taylor InterNet: staylor(a)mrynet.com MRY Systems staylor(a)mrynet.lv (Skots Gregorijs Akmentins-Teilors -- just call me "Skots") ----- Labak miris neka sarkans ----- Received: (from major@localhost) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id HAA33694 for pups-liszt; Thu, 26 Aug 1999 07:55:06 +1000 (EST) Received: from shelbyville.oai.com (IDENT:root@[204.57.59.144]) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id HAA33684 for &lt;pups(a)minnie.cs.adfa.edu.au&gt;; Thu, 26 Aug 1999 07:54:57 +1000 (EST) Received: (from mirian@localhost) by shelbyville.oai.com (8.8.8/8.8.8) id RAA08490; Wed, 25 Aug 1999 17:52:23 -0400 (EDT) From: Mirian Crzig Lennox &lt;lennox(a)alcita.com&gt; To: pups(a)minnie.cs.adfa.edu.au Subject: Re: Modified Supnik emulator for the 11 References: &lt;199908252113.OAA15059(a)mrynet.com&gt; Original-Sender: lennox(a)alcita.com Organization: Alcita Technologies, Inc. Date: 25 Aug 1999 17:52:22 -0400 In-Reply-To: staylor(a)mrynet.com&#x27;s message of "Wed, 25 Aug 1999 14:13:40 +0000" Message-ID: &lt;m3iu63plyh.fsf(a)shelbyville.oai.com&gt; Sender: owner-pups(a)minnie.cs.adfa.edu.au Precedence: bulk staylor(a)mrynet.com (S. Akmentins-Teilors) writes: Really! So far I have had no luck getting Supnik 2.3 to work with the elfje rl02 images on the PUPS archive. I've always had to use the significantly-hacked-up 2.2 emulator instead. What did you change and/or what disk images are you using? I would love to find an arrangement that makes it possible to run an emulated 2.11bsd system with large-capacity RP06 images.. that would allow one to have around a quarter-gig of disk space. :) -- Mirian Crzig Lennox Systems Anarchist Invest in America -- buy a Congressman! Received: (from major@localhost) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id IAA33801 for pups-liszt; Thu, 26 Aug 1999 08:12:04 +1000 (EST) Received: from moe.2bsd.com (0(a)MOE.2BSD.COM [206.139.202.200]) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id IAA33797 for &lt;pups(a)minnie.cs.adfa.edu.au&gt;; Thu, 26 Aug 1999 08:11:55 +1000 (EST) Received: (from sms@localhost) by moe.2bsd.com (8.9.0/8.9.0) id PAA18433 for pups(a)minnie.cs.adfa.edu.au; Wed, 25 Aug 1999 15:08:53 -0700 (PDT) Date: Wed, 25 Aug 1999 15:08:53 -0700 (PDT) From: "Steven M. Schultz" &lt;sms(a)moe.2bsd.com&gt; Message-Id: &lt;199908252208.PAA18433(a)moe.2bsd.com&gt; To: pups(a)minnie.cs.adfa.edu.au Subject: Re: Modified Supnik emulator for the 11 Sender: owner-pups(a)minnie.cs.adfa.edu.au Precedence: bulk Scott - Howdy! Hmmm, the Begemot emulator is difficult to set up due to an inscrutable configfile format but it runs here under BSD/OS 4.0.1 after making a couple 'tweeks'. I wonder if the problems you're having are due to FreeBSD switching to ELF. At one time BSD/OS used a.out also and "P11" built/ran just fine - the the OS switched to ELF and P11 would no longer build. I had thought simply editing the instab.s would be enough but after doing that P11 wouldn't run right at all. What I did was add a "-u" option to 'geni' and then regenerate the instab.s file ("geni -u ...") _without_ the underscore characters present. The compiler no longer generates leading '_' characters so having them in the instab.s file causes problems. Regenerating and assembling instab.s cleared up all the problems I was having. Below are the changes I've made to P11 - some are specific to getting the various IOprogs to run under BSD/OS but the changes to geni.c are OS independent. THe other change I had to make was to 'devices.c' to speed up the clock - it's still not right for a PPro-200 but is better than it was (the clock was running far too slow, now it's just ~10% too slow). Not FreeBSD but if you're getting bit by the same thing I did earlier under another BSD that switched from a.out to ELF the changes below may be useful to you. Steven Schultz sms(a)moe.2bsd.com *** ./Utils/geni.c.old Sat Oct 11 14:01:39 1997 --- ./Utils/geni.c Thu Aug 19 21:07:56 1999 *************** *** 49,54 **** --- 49,55 ---- int code; /* current instruction code */ int ccc; /* current microinstruction count */ int coo = -1; /* what output to generate */ + int no_ul = 0; /* Don't generate leading _ */ int profiler_output; char *ul; /* the undeline character, if needed */ char *ofile; /* output file name */ *************** *** 123,131 **** int opt; set_argv0(argv[0]); ! while((opt = getopt(argc, argv, "vmpo:")) != EOF) switch(opt) { case 'v': verbose++; break; --- 124,135 ---- int opt; set_argv0(argv[0]); ! while((opt = getopt(argc, argv, "uvmpo:")) != EOF) switch(opt) { + case 'u': + no_ul++; + break; case 'v': verbose++; break; *************** *** 275,284 **** tab_out_i386as() { printf("\t.file\t\"%s\"\n", ifile); ! printf("\t.globl\t_instab\n"); printf(".text\n"); printf("\t.align\t2\n"); ! printf("_instab:\n"); for(code = 0; code < 0x10000; code++) { printf("\t.long\t"); for(ccc = 0; ccc < 4; ccc++) --- 279,288 ---- tab_out_i386as() { printf("\t.file\t\"%s\"\n", ifile); ! printf("\t.globl\t%s\n", no_ul ? "instab" : "_instab"); printf(".text\n"); printf("\t.align\t2\n"); ! printf("%s:\n", no_ul ? "instab" : "_instab"); for(code = 0; code < 0x10000; code++) { printf("\t.long\t"); for(ccc = 0; ccc < 4; ccc++) *************** *** 632,637 **** --- 636,646 ---- switch(coo) { case COO_i386as: + if (no_ul) + { + ul = ""; + break; + } case COO_sun_as: case COO_i386_aout: ul = "_"; *** ./IOProgs/epp_bpf.c.old Sat Oct 11 14:02:28 1997 --- ./IOProgs/epp_bpf.c Wed Jun 17 22:16:50 1998 *************** *** 341,347 **** panic("read(bpf): %s", strerror(errno)); bpf_ptr = bpf_buf; bpf_end = bpf_buf + ret; ! INFO("read_input: bpf_read = %d.\n", ret); } /* --- 341,347 ---- panic("read(bpf): %s", strerror(errno)); bpf_ptr = bpf_buf; bpf_end = bpf_buf + ret; ! info("read_input: bpf_read = %d.\n", ret); } /* *************** *** 351,357 **** bpf_ptr = bpf_ptr + BPF_WORDALIGN(h->bh_hdrlen + h->bh_caplen); if(h->bh_caplen < h->bh_datalen) { ! INFO("caplen(%lu) < datalen(%lu) ??? - packet dropped.\n", h->bh_caplen, h->bh_datalen); ret = 0; } else { *pbuf = (u_char *)h + h->bh_hdrlen; --- 351,357 ---- bpf_ptr = bpf_ptr + BPF_WORDALIGN(h->bh_hdrlen + h->bh_caplen); if(h->bh_caplen < h->bh_datalen) { ! info("caplen(%lu) < datalen(%lu) ??? - packet dropped.\n", h->bh_caplen, h->bh_datalen); ret = 0; } else { *pbuf = (u_char *)h + h->bh_hdrlen; *************** *** 360,366 **** *more = bpf_ptr < bpf_end; ! INFO("read_input: %d. (more=%d)\n", ret, *more); return ret; } --- 360,366 ---- *more = bpf_ptr < bpf_end; ! info("read_input: %d. (more=%d)\n", ret, *more); return ret; } *** ./IOProgs/epp_tun.c.old Sat Jan 31 02:52:26 1998 --- ./IOProgs/epp_tun.c Tue Aug 17 19:47:37 1999 *************** *** 13,19 **** --- 13,21 ---- # include <sys/ioctl.h> # include <sys/select.h> # include <net/if.h> + #ifndef __bsdi__ # include <net/if_var.h> + #endif # include <net/if_tun.h> # include "epp.h" # include "../libutil/util.h" *************** *** 44,50 **** argv += optind; if(argc != 3) ! panic("need one arg"); parse_ether(my_ether, argv[1]); parse_ether(other_ether, argv[2]); --- 46,52 ---- argv += optind; if(argc != 3) ! panic("need two args"); parse_ether(my_ether, argv[1]); parse_ether(other_ether, argv[2]); *** ./Config/M-i386-bsdi.old Sun Oct 12 07:10:03 1997 --- ./Config/M-i386-bsdi Wed Jun 17 20:50:19 1998 *************** *** 27,33 **** * define the cookie for the geni program (look into Utils/geni.c) * If you want geni output an object file (see later) this cookie * is used only for the profiler output */ ! /* # define MAKE_GENIS */ # define MAKE_GENIE_COOKIE "i386-as" /* define command to set data limit to K kilobytes, if you need it */ --- 27,33 ---- * define the cookie for the geni program (look into Utils/geni.c) * If you want geni output an object file (see later) this cookie * is used only for the profiler output */ ! # define MAKE_GENIS # define MAKE_GENIE_COOKIE "i386-as" /* define command to set data limit to K kilobytes, if you need it */ *************** *** 43,49 **** /* if you have the gnu libbfd and liberty you can geni have to output * object code instead of C or assembler. You must define the following: */ ! # define MAKE_HAVE_LIBBFD /* if you have it, you may have to set up the right paths. */ # define MAKE_CC_BFD_INCL -I/usr/gnu/include --- 43,49 ---- /* if you have the gnu libbfd and liberty you can geni have to output * object code instead of C or assembler. You must define the following: */ ! /* # define MAKE_HAVE_LIBBFD */ /* if you have it, you may have to set up the right paths. */ # define MAKE_CC_BFD_INCL -I/usr/gnu/include *** ./device.c.old Sat Oct 11 14:17:24 1997 --- ./device.c Thu Aug 19 23:05:53 1999 *************** *** 7,14 **** * generic device support */ ! # define TINTERVAL 20 /* msecs between clock ticks */ ! typedef struct Async Async; typedef struct Timer Timer; --- 7,14 ---- * generic device support */ ! # define TINTERVAL 16 /* msecs between clock ticks */ ! /* Should be 16.666666 for US 60hz */ typedef struct Async Async; typedef struct Timer Timer; Received: (from major@localhost) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id IAA33844 for pups-liszt; Thu, 26 Aug 1999 08:18:19 +1000 (EST) Received: from moe.2bsd.com (0(a)MOE.2BSD.COM [206.139.202.200]) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id IAA33839 for &lt;pups(a)minnie.cs.adfa.edu.au&gt;; Thu, 26 Aug 1999 08:18:11 +1000 (EST) Received: (from sms@localhost) by moe.2bsd.com (8.9.0/8.9.0) id PAA18523 for pups(a)minnie.cs.adfa.edu.au; Wed, 25 Aug 1999 15:17:56 -0700 (PDT) Date: Wed, 25 Aug 1999 15:17:56 -0700 (PDT) From: "Steven M. Schultz" &lt;sms(a)moe.2bsd.com&gt; Message-Id: &lt;199908252217.PAA18523(a)moe.2bsd.com&gt; To: pups(a)minnie.cs.adfa.edu.au Subject: Re: Modified Supnik emulator for the 11 Sender: owner-pups(a)minnie.cs.adfa.edu.au Precedence: bulk Hi - Hmm, I've been using Supnik's 2.3 emulator (on and off - I prefer running the real 11/73 though most of the time) and now that "vi" works (there was a bug in the "div" instruction which Bob fixed not all that long ago for 2.3) the emulator's more useful than it was. You might try using the 2.11 images from the PUPS CD instead. Create a "tape file" (the instructions are in the 2.11 distribution directory) and then use a "toggle in" bootstrap for the "mt" device. The config file I use for this is: set cpu 22B set cpu 2048K set rp0 rp06 set rl0 rl02 set rl1 rl02 set rl2 rl02 set rl3 rl02 set tm0 locked at rp0 rp0 at rl0 root.rl02 at rl1 usr1.rl02 at rl2 usr2.rl02 at rl3 usr3.rl02 at rk0 junk0.rk05 at rk1 junk1.rk05 at rk2 junk2.rk05 at rk3 junk3.rk05 at rk4 junk4.rk05 at rk5 junk5.rk05 at rk6 junk6.rk05 at rk7 junk7.rk05 at tm0 mt0 at tm1 mt1 # at tm1 /zip/mt0 Place your "2.11 boot tape file" (the 'makesimtape' program which is also available in the archive and on the CD is used to create Supnik emulator tape files) in to the file "mt0" and then follow the instructions in the setup/install documentation on how to boot a tape if you don't have tape bootroms (it's less than a dozen instructions you need to toggle in the octal for). Oh - and since the "RP06" disk is just an image to the host computer (to the PDP-11 it is a RP06 ;)) the image IS interchangeable between emulators - I've used the same RP06 image under both (obviously not at the same time) the Supnik and Begemot emulators. Works fine. The biggest problem with the Supnik emulator is that the clock runs far far too fast (at least with a PPro-200 running the emulator) and after running for an extended period of time the PDP-11 system ends up several hours in the future. Steven Schultz sms(a)wlv.iipo.gtegsc.com Received: (from major@localhost) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id IAA33945 for pups-liszt; Thu, 26 Aug 1999 08:25:13 +1000 (EST) Received: from begemot.org (negara.cs.waikato.ac.nz [130.217.248.112]) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id IAA33939 for &lt;pups(a)minnie.cs.adfa.edu.au&gt;; Thu, 26 Aug 1999 08:25:02 +1000 (EST) Received: (from joerg@localhost) by begemot.org (8.9.2/8.9.2) id KAA11280; Thu, 26 Aug 1999 10:25:27 +1200 (NZST) (envelope-from joerg) Date: Thu, 26 Aug 1999 10:25:27 +1200 From: "Joerg B. Micheel" &lt;joerg(a)begemot.org&gt; To: "Steven M. Schultz" &lt;sms(a)moe.2bsd.com&gt; Cc: pups(a)minnie.cs.adfa.edu.au, joerg(a)begemot.org Subject: Re: Modified Supnik emulator for the 11 Message-ID: &lt;19990826102527.A11262(a)begemot.org&gt; References: &lt;199908252208.PAA18433(a)moe.2bsd.com&gt; Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: &lt;199908252208.PAA18433(a)moe.2bsd.com&gt;; from Steven M. Schultz on Wed, Aug 25, 1999 at 03:08:53PM -0700 Organization: Begemot Computer Associates Operating-System: ... powered by FreeBSD Sender: owner-pups(a)minnie.cs.adfa.edu.au Precedence: bulk On Wed, Aug 25, 1999 at 03:08:53PM -0700, Steven M. Schultz wrote: That's correct. I believe there are also problems with p11 missing a couple of timer interrupts. All of the complaints are entirely appropriate, the whole thing needs major cleanup. I hope one of us will finally get around doing some serious work on it again, soon. Joerg -- Joerg B. Micheel Email: &lt;joerg(a)begemot.org&gt; Begemot Computer Associates Phone: +64 7 8562148 6 Kakanui Avenue, Hillcrest Fax: +64 7 8562148 Hamilton, New Zealand Pager: +64 868 38222 Received: (from major@localhost) by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id IAA34224 for pups-liszt; Thu, 26 Aug 1999 08:57:38 +1000 (EST)