Article 2429 of vmsnet.pdp-11: Path: cs.utk.edu!emory!europa.eng.gtefsd.com!library.ucla.edu!ihnp4.ucsd.edu!mvb.saic.com!info-pdp11 From: Jan Brittenson Newsgroups: vmsnet.pdp-11 Subject: Re: Instruction execution frequencies Message-ID: Date: Thu, 10 Mar 94 10:05 EST Organization: nope X-Gateway-Source-Info: Mailing List Lines: 29 Date: Wed, 9 Mar 1994 22:34:04 -0500 (EST) From: Pat_Barron@transarc.com Does anyone have a table of how frequently each instruction in the PDP-11 instruction set gets executed in real-life code? I'm toying with the idea of building a (yet another) PDP-11 simulator, and I want to be able to decode the most frequently used instructions first.... What my simulator does is, I have an instruction set description in a separate file, that I run through a preprocessor to generate all the decoding logic. It outputs a jump table with 64k entries and simply does a ``goto instruction_dispatch[*((unsigned short *) pc)++];'' to dispatch an instruction. INSTRUCTION_DISPATCH is an array of pointers to labels. In fact, each instruction ends with a dispatch of the next one. The current mode PC is a global register declared as void *. Obviously, it relies on GCC for efficiency, but that's hardly a problem since GCC will run on anything of even marginal interest. BTW, I also just srbk() to get 4M of memory right at the start, plus allocate an additional 256k map entries for the I/O page -- one for each word. Each I/O page map entry points to a device switch for device emulators, or a pointer to an "invalid" device whose sole purpose is to generate exceptions. Decvice emulators run as separate processes and talk to the main emulator via datagrams and signals. This is roughly where I have been for quite a while since I've had far too much to do on other projects -- particularly those that pay bills. -- Jan Brittenson bson@ai.mit.edu Article 3599 of vmsnet.pdp-11: Path: cs.utk.edu!gatech!europa.eng.gtefsd.com!news.umbc.edu!cs.umd.edu!highpower.cs.umd.edu!fms From: fms@highpower.cs.umd.edu (Marat Fayzullin) Newsgroups: comp.emulators.misc,vmsnet.pdp-11 Subject: Re: PDP11 Emulator Date: 10 Nov 1994 01:53:43 GMT Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742 Lines: 21 Message-ID: <39ruf7$4om@mimsy.cs.umd.edu> References: <1994Nov8.115838.8332@zippy.dct.ac.uk> <1994Nov8.115908.8333@zippy.dct.ac.uk> <39qc4r$2eh@mimsy.cs.umd.edu> NNTP-Posting-Host: highpower.cs.umd.edu X-Newsreader: TIN [version 1.2 PL2] Xref: cs.utk.edu comp.emulators.misc:1668 vmsnet.pdp-11:3599 Marat Fayzullin (fms@highpower.cs.umd.edu) wrote: : Anybody wanna write down a list of PDP11 emulators with their : sources on the Net? I need one for the Emulation WWW Page : [http://www.cs.umd.edu/users/fms/comp/] I'd like to thank all people who send me the information. Check out the DEC part of WWW page again - there are links to several new PDP11 emulators. TO Megan Gentry [gentry@rusure.zko.dec.D5NET.dec.com]: Our mailer can't deliver mail to your address, so here is the reply to your mail: So-called "russian" emulator is for IBM PC [MSDOS]. Try running it on 486 or something similar. It actually emulates DVK-2 [and maybe DVK-3/4] - a Russian desktop PDP11 clone. It even has cyrillic letters :) Marat Article 3600 of vmsnet.pdp-11: Newsgroups: vmsnet.pdp-11 Path: cs.utk.edu!gatech!howland.reston.ans.net!pipex!uunet!world!mbg From: mbg@world.std.com (Megan) Subject: Re: PDP11 Emulator Message-ID: Organization: The World Public Access UNIX, Brookline, MA References: <1994Nov8.115838.8332@zippy.dct.ac.uk> Date: Thu, 10 Nov 1994 03:00:30 GMT Lines: 16 mcscs2mm@zippy.dct.ac.uk (Malcolm MacArthur) writes: >My question is: does anyone have a version of this emulator that'll compile >on VMS, or know of any PD emulators for an '11 that'll run under VMS? There is an emulator for pdp-11 (and pdp-8) which compiles on Ultrix, DEC OSF/1 and VMS. It is available from: ftp.std.com:/ftp/pub/mbg/pdp_8_11_emulators.tar.Z There are .COM files which will build it on VMS as well as a Makefile to build on U*x. Megan Gentry Former RT-11 Developer Article 3623 of vmsnet.pdp-11: Newsgroups: comp.emulators.misc,vmsnet.pdp-11,alt.sys.pdp-11 Path: cs.utk.edu!martha.utk.edu!darwin.sura.net!haven.umd.edu!purdue!lerc.nasa.gov!magnus.acs.ohio-state.edu!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!scsing.switch.ch!news.dfn.de!gs.dfn.de!zib-berlin.de!ceres.fokus.gmd.de!jbm From: micheel@fokus.gmd.de (Joerg Micheel) Subject: PDP-11 emulator available Message-ID: <1994Nov14.125443.11228@fokus.gmd.de> Keywords: emulator,pdp-11 Sender: jbm@fokus.gmd.de (Joerg Micheel) Organization: GMD-Fokus (Berlin) Date: Mon, 14 Nov 1994 12:54:43 GMT Lines: 45 Xref: cs.utk.edu comp.emulators.misc:1714 vmsnet.pdp-11:3623 Version 2.0 of p11 - the Begemot PDP-11 emulator - is available on ftp.fokus.gmd.de:/pub/nthp/11. This release supports the following platforms: BSD/386 1.1 x86 SunOS 4.1.3 SPARC Solaris 2.x SPARC DEC/OSF1 T3.2 ALPHA The distribution consists of both source and binary kits. In addition, for those of you not being fortunate to have bootable RL02 disk images, a demo version of the emulator is run on the ftp server. Try one of the following telnet ftp.fokus.gmd.de 10000 telnet ftp.fokus.gmd.de 10001 telnet ftp.fokus.gmd.de 10002 telnet ftp.fokus.gmd.de 10003 to get connected to one of the four available terminals. If telnet connects, but this terminal is currently in use, you'll get a short message. You should then try connecting to another port. Once you are online, log into RSX-11M with hel guest/ (no password necessary). Don't forget to say "bye", before disconnecting with telnet. The current implementation emulates a KDJ11A with 1MByte of RAM, one or more RLV12 disk controllers, a boot ROM, one or more KL11A controllers and a standard line printer interface. While emulation speed has not been fully evaluated yet, the integer performance on high-end workstations seems comparable to the real KDJ11A. DMA performace peeks at nearly twice the speed of the real machine. Enjoy, Begemot Willi Harti Brandt Joerg Micheel -- GMD German National Research Corporation for Mathematics and FOKUS Data Processing - Research Institute for Open Communication Systems Berlin, Germany Article 4927 of vmsnet.pdp-11: Path: cs.utk.edu!nntp.memphis.edu!netnews.wku.edu!mvb.saic.com!news.mathworks.com!newshost.marcam.com!usc!howland.reston.ans.net!news-e1a.megaweb.com!newstf01.news.aol.com!uunet!in1.uu.net!library.erc.clarkson.edu!rpi!wilsonj Newsgroups: vmsnet.pdp-11 Subject: Re: Virtual Disk Message-ID: <3u0rkp$cfg@usenet.rpi.edu> From: wilsonj@alum01.its.rpi.edu (John Wilson) Date: 12 Jul 1995 16:01:29 GMT References: <3u0bdv$hi4@atusks02.aut.alcatel.at> Organization: Rensselaer Polytechnic Institute NNTP-Posting-Host: alum01.its.rpi.edu Lines: 27 In article <3u0bdv$hi4@atusks02.aut.alcatel.at>, Brunner Guenther wrote: >How can i create a virtual Disk for the PDP-11 Emulator for DOS ? >(ERSATZ-11 from John Wilson) > >I have a PDP-11/84 running RSX11M+ 3.0 with two RA60. Ersatz-11 as yet does not have MSCP emulation (nor does any other emulator that I know of, it's extremely complicated and the docs blow chunks), so you'll need to create a volume with some other kind of system device. E11 also doesn't have 22-bit addressing yet so shoe-horning M+ will be no fun. It runs 11M fine, and Eduard Vopicka is working on an 11M driver for the hypothetical disk device that E11 supports, which allows huge disks of variable sizes and is intended to be a stopgap replacement for MSCP. He's not done because I've been too busy to finish some changes to E11 he requested. Anyway I have no idea what's involved in genning an 18-bit version of M+, or whether it's possible to build an image file of, say, an RK07, on a system that has only RA60s. I have a program called PUTR.COM which handles OS/8 and RT-11 filesystems (and can read RSTS ones), which among other things knows how to build RT-11 boot volumes of any size, given the monitor and driver files. But it doesn't know Files-11 yet. It's about to come out in a new version which supports (my understanding of) RX23, RX26, RX33, and double-sided RX50 floppies, in addition to the stuff the previous version supported. John Wilson Article 6415 of vmsnet.pdp-11: Path: cs.utk.edu!news.msfc.nasa.gov!newsfeed.internetmci.com!usenet.logical.net!news.wizvax.net!wilson From: wilson@dbit.dbit.com (John Wilson) Newsgroups: vmsnet.pdp-11,comp.sys.dec Subject: Re: PC MACRO-11 cross assembler? Date: 25 May 1996 06:41:31 GMT Organization: D Bit, Troy, NY Lines: 36 Message-ID: <4o6a2r$lge@news.wizvax.net> References: <1996May20.173314.5903@friend.kastle.com> <31A31F0C.1840@ENGSYS.MICRO.HONEYWELL.COM> NNTP-Posting-Host: dbit.dbit.com Xref: cs.utk.edu vmsnet.pdp-11:6415 comp.sys.dec:40513 In article <31A31F0C.1840@ENGSYS.MICRO.HONEYWELL.COM>, GRAY wrote: >There is a good emulator at > >FTP://TATS.WIZVAX.NET/pub/e11/ Note: the hostname of this machine has been changed to FTP.DBIT.COM, I incorporated but the NYS Div. of Corporations said the name TATS was already taken so now it's D Bit. The old name still works as an alias but who knows how long that will last, so please update your mirror Wayne! I'm sorry there hasn't been an update to E11 in quite a while, I'm working hard on getting the V2.0 beta release done as soon as possible. It includes emulation of several flavors of magtape, several new disks, and adds 11/24, /44, /70, and /94 CPU types (using a limited 384KB 22-bit MMU), plus the dual video display feature now works correctly. There are a bunch of loose ends to tie up before it's ready though (major parts have been rewritten), and I've got some gizmos planned that I'd like to add if it doesn't take too long. Performance on a midrange Pentium is roughly equal to the 11/94, a bit more or a bit less depending on what you're doing. As soon as the V2.0 beta is done I'll be working on cranking out the full-blown commercial version ASAP, >2 years of "beta" is just too long. It will have 4MB memory, big disks, serial muxes, and support for some kind of bus adapter for real PDP-11 hardware, either the Bit3 PCI-to-Q22 board set or something custom (I can't find any emulation-capable Unibus adapters, and adding the cost of an Able Microverter to the already high Bit3 price is way too much, besides being kludgey). Something like the V2.0 beta (or better) will still be available indefinitely for free home use and as a demo for the "big" version. John Wilson D BIT