From: "Jan Gray" Newsgroups: comp.arch,alt.folklore.computers References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> Subject: Coherent Lines: 26 Organization: Gray Research LLC X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Message-ID: X-Trace: /K+neUEkhpz7YU6qM2EiNxf0fNx6IVDyAlopBbEJ6ABTwfL5w4i04oWUusudGvM+BeUyVeVSaw4R!sjEURw64bbDHBmK5XAkObKv9sEUyxNmA6c9JahHzcSEQyxFCA3WRPftzkcH98/ZcazksV0Ifub3F!GijA X-Complaints-To: abuse@gte.net X-Abuse-Info: Please be sure to forward a copy of ALL headers X-Abuse-Info: Otherwise we will be unable to process your complaint properly NNTP-Posting-Date: Tue, 07 Nov 2000 05:45:50 GMT Distribution: world Date: Tue, 07 Nov 2000 05:45:50 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!sunqbc.risq.qc.ca!newsfeed.direct.ca!look.ca!paloalto-snh1.gtei.net!news.gtei.net!dfiatx1-snr1.gtei.net.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67734 "R.E.Ballard ( Rex Ballard )" wrote in message news:8u74l6$sch$1@nnrp1.deja.com... > ... Mark Williams Coherant was based > on Minux and came out about the same time as Slackware 1.0. Incorrect -- Coherent predated both Minix and Linux by many years. From an old alt.folklore.computers posting by William G. Lederer: "Coherent was originally written by four very bright programmers from the University of Waterloo (arguably the best undergrad CS department in the hemisphere if not anywhere) to run on the PDP-11. This was done in an astonishingly short time. As it was based on the concepts of V7 Unix, it was designed to be portable, and as a result, it was easily moved to the z8000, and in 1982 to the IBM PC. Keep in mind that this was with no memory management hardware, something that experts kept saying couldn't be done. Similar arguments were offered before Coherent was brought up on the Atari ST, again with no memory management." If I recall correctly, two of those bright programmers were Randall Howard (now MKS CEO) and Johann George (now Sourcelight Technologies CEO). Jan Gray ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: 07 Nov 2000 21:49:29 +0100 Organization: My own Private Self Lines: 54 Message-ID: <6ur94na4o6.fsf@chonsp.franklin.ch> References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 973630173 689 10.0.3.2 (7 Nov 2000 20:49:33 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 7 Nov 2000 20:49:33 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:67746 "Jan Gray" writes: > "R.E.Ballard ( Rex Ballard )" wrote in message > news:8u74l6$sch$1@nnrp1.deja.com... > > ... Mark Williams Coherant was based > > on Minux and came out about the same time as Slackware 1.0. > > Incorrect -- Coherent predated both Minix and Linux by many years. Thats right. > From an old alt.folklore.computers posting by William G. Lederer: > > hemisphere if not anywhere) to run on the PDP-11. This was done in an > the z8000, and in 1982 to the IBM PC. Keep in mind that this was with > no memory management hardware, Here WGL errs: The IBM PC / 8088 has memory management, known as "segments". If you took an SID (separate instruction data) PDP11 program to the PC, you simply set CS = base of instruction segment, DS = ES = SS = base of data segment. Non-SID you set CS = xS = only data segment. And off you were. These two arrangements were both common under MS-DOS. What the PCs 8088 did not have, was memory _protection_, nor an kernel or supervisor mode to prevent user programs from directly accessing IO hardware or segment registers. But so long programs did not try to deliberately do bad thing (and the C compiler simply does not generate code to do that) it will work just like a PDP11/40 and above, even without pointer errors crashing the OS, they may though strike other processes that follow direct after in physical memory! > something that experts kept saying > couldn't be done. Which experts claimmed that? > Similar arguments were offered before Coherent was > brought up on the Atari ST, again with no memory management." Now that one had no memory managment, but with an properly written libc/malloc() it should be no big problem, so long you don't mind pointer errors wiping out the entire system, as was also the case with Unix V1 on PDP11/20! -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: d_lau@my-deja.com Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Tue, 07 Nov 2000 21:31:54 GMT Organization: Deja.com - Before you buy. Lines: 30 Message-ID: <8u9sca$5g4$1@nnrp1.deja.com> References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> NNTP-Posting-Host: 132.233.247.4 X-Article-Creation-Date: Tue Nov 07 21:31:54 2000 GMT X-Http-User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT) X-Http-Proxy: 1.0 proxy1.fm.intel.com:911 (Squid/2.2.STABLE5-hno.20000202), 1.0 x52.deja.com:80 (Squid/1.1.22) for client 143.183.198.57, 132.233.247.4 X-MyDeja-Info: XMYDJUIDd_lau Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.imp.ch!uni-erlangen.de!newsfeeds.belnet.be!news.belnet.be!btnet-peer1!btnet-peer0!btnet-peer!btnet!diablo.netcom.net.uk!netcom.net.uk!cpk-news-hub1.bbnplanet.com!news.gtei.net!nntp2.deja.com!nnrp1.deja.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67826 In article , "Jan Gray" wrote: > "R.E.Ballard ( Rex Ballard )" wrote in message > news:8u74l6$sch$1@nnrp1.deja.com... > > ... Mark Williams Coherant was based > > on Minux and came out about the same time as Slackware 1.0. > > Incorrect -- Coherent predated both Minix and Linux by many years. > If I recall correctly, two of those bright programmers were Randall Howard > (now MKS CEO) and Johann George (now Sourcelight Technologies CEO). > > Jan Gray I remember Coherent from Mark Williams a long time ago (I think the name "Mark Williams" had a story of its own, but I can't remember the details; it was told to me when I visited their HQ in Chicago many years ago). I understand the company is no more. So what ever happened to the intellectual properties of the company? Did they sell off the code base to someone else or will the code ever make its way into the public domain? With the new interest in "open source", it would be useful to have another source of tested *nix clone utilities. My guess is that MKS probably holds the rights to (most of) the source code now. But MW also had a compiler/assembler/loader that MKS does not sell, so is it possible to port that over to Minix (if we can get the source files)? Sent via Deja.com http://www.deja.com/ Before you buy. ###### Message-ID: <3A08B10E.7D9919BF@ev1.net> Date: Tue, 07 Nov 2000 17:49:02 -0800 From: Charles Richmond Reply-To: richmond@ev1.net Organization: Cannine Computer Center X-Mailer: Mozilla 4.7C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.folklore.computers Subject: Re: Coherent References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit NNTP-Posting-Host: taydal-207-55-144-193.ev1.net X-Trace: 7 Nov 2000 17:45:40 -0600, taydal-207-55-144-193.ev1.net Lines: 34 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!uni-erlangen.de!newsfeeds.belnet.be!news.belnet.be!skynet.be!newsfeed.skycache.com!Cidera!newsa.ev1.net!taydal-207-55-144-193.ev1.net Xref: chonsp.franklin.ch alt.folklore.computers:67795 d_lau@my-deja.com wrote: > > [snip...] [snip...] [snip...] > > I remember Coherent from Mark Williams a long time ago (I think the > name "Mark Williams" had a story of its own, but I can't remember the > details; it was told to me when I visited their HQ in Chicago many > years ago). > Mr. Dennis Ritchie posted a few months ago about a trip he made (with someone else, IIRC) to Mark Williams to check that Coherent did *not* include any source code from A.T.&T. Unix. IIRC, he said that Mark Williams was originally a paint company, but that the son of the owner was taking it into computer software. I believe that Coherent was found clean of A.T.&T. code... > > I understand the company is no more. So what ever > happened to the intellectual properties of the company? Did they sell > off the code base to someone else or will the code ever make its way > into the public domain? With the new interest in "open source", it > would be useful to have another source of tested *nix clone utilities. > My guess is that MKS probably holds the rights to (most of) the source > code now. But MW also had a compiler/assembler/loader that MKS does > not sell, so is it possible to port that over to Minix (if we can get > the source files)? > I do *not* know about the fate of the Mark Williams company...they did have a C compiler that ran on the PC and also had an Atari ST version of that compiler. -- +-------------------------------------------------------------+ | Charles and Francis Richmond | +-------------------------------------------------------------+ ###### From: zaitcev@yahoo.com (Pete Zaitcev) Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> Message-ID: User-Agent: slrn/0.9.6.2 (Linux) Lines: 21 Date: Wed, 08 Nov 2000 06:41:47 GMT NNTP-Posting-Host: 24.19.219.48 X-Complaints-To: abuse@home.net X-Trace: news1.frmt1.sfba.home.com 973665707 24.19.219.48 (Tue, 07 Nov 2000 22:41:47 PST) NNTP-Posting-Date: Tue, 07 Nov 2000 22:41:47 PST Organization: @Home Network Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!newshub2.rdc1.sfba.home.com!news.home.com!news1.frmt1.sfba.home.com.POSTED!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67779 > > Similar arguments were offered before Coherent was > > brought up on the Atari ST, again with no memory management." > > Now that one had no memory managment, but with an properly written > libc/malloc() it should be no big problem, so long you don't mind > pointer errors wiping out the entire system, as was also the case with > Unix V1 on PDP11/20! What did you do on fork(2)? Actually, I am thinking about porting ucLinux to SPARClite (lame Fujitsu SPARC without MMU). UcLinux runs on embedded 68ks without MMU. I was told that ucLinux uses A5 based data and duplicates them. As long as children are careful not to attempt to unwind their duplicated stack and do exec() ASAP, it works better than swapping whole parent out. I pretty much gave up on running Linux binaries as long as source compatibility can be maintained. *sigh* --Pete ###### From: Chris Baird Newsgroups: alt.folklore.computers Subject: Re: Coherent Organization: World Wide Weasels Lines: 16 Message-ID: References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> X-No-Archive: yes X-Newsreader: Gnus v5.7/Emacs 20.7 NNTP-Posting-Host: bombadil.apana.org.au X-Trace: ozemail.com.au 973703381 203.14.156.3 (Thu, 09 Nov 2000 04:09:41 EST) NNTP-Posting-Date: Thu, 09 Nov 2000 04:09:41 EST Distribution: world Date: 08 Nov 2000 21:35:04 +1100 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!machtgarnix.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news-fra.pop.de!newsfeed.germany.net!news.tele.dk!205.231.236.10!newspeer.monmouth.com!cpk-news-hub1.bbnplanet.com!news.gtei.net!newsfeed.skycache.com!Cidera!xfer10.netnews.com!netnews.com!newspeer.cwnet.com!sjc1.nntp.concentric.net!newsfeed.concentric.net!newsfeed.ozemail.com.au!ozemail.com.au!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67781 > I understand the company is no more. So what ever > happened to the intellectual properties of the company? Linux Press/Cheapbytes did buy rights for the Coherent Lexicon recently (~16 months ago). However they immediately went around hassling all the remaining FTP archives citing "copyright violation". (So much for a company claiming to support "open source"..) > Did they sell off the code base to someone else or will the code > ever make its way into the public domain? Numerous attempts have been made, but Robert Schwartz is very much against the idea, unfortunately. -- Chris,, ###### From: "Philip 'Yes, that's my address' Newton" Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Wed, 08 Nov 2000 12:45:39 +0100 Organization: very little Lines: 50 Message-ID: References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> NNTP-Posting-Host: internet-router.datenrevision.de (195.222.209.7) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: fu-berlin.de 973683942 1420324 195.222.209.7 (16 [11583]) X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!fu-berlin.de!uni-berlin.de!internet-router.datenrevision.DE!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67805 On 07 Nov 2000 21:49:29 +0100, Neil Franklin wrote: > Here WGL errs: The IBM PC / 8088 has memory management, known as > "segments". > > If you took an SID (separate instruction data) PDP11 program to the > PC, you simply set CS = base of instruction segment, DS = ES = SS = > base of data segment. Turbo C called this the "small" memory model. > Non-SID you set CS = xS = only data segment. And this was "tiny". > And off you were. These two arrangements were both common under MS-DOS. Yep. Turbo C also offered "medium", "compact", "large", and "huge" memory models. These used 32-bit (actually, effectively only 20-bit) addresses for code (compact), data (medium), or both (large). (I may have compact and medium the wrong way round.) "Huge" model was for 32-bit code, 32-bit data where an individual data item (an array, for example) could be larger than 64 KB (one segment), so pointer arithmetic was always done normalised (0 <= offset <= 0xf) so that the offset didn't overflow and wrap around to the beginning of the segment. > What the PCs 8088 did not have, was memory _protection_, nor an kernel > or supervisor mode to prevent user programs from directly accessing IO > hardware or segment registers. But so long programs did not try to > deliberately do bad thing (and the C compiler simply does not generate > code to do that) it will work just like a PDP11/40 and above, even > without pointer errors crashing the OS, they may though strike other > processes that follow direct after in physical memory! In small mode yes, but in tiny mode, CS == DS and you could overwrite code if your pointers were off. I believe that the Turbo C runtime (possibly also other compilers' runtimes, too) squirreled the first few bytes of the segment off to a safe place before main() was called, and on exit, compared the first few bytes to the saved copy. If they had been changed, it was a fairly safe sign that someone had assigned through a (16-bit, offset-only) NULL pointer, and an appropriate message was emitted. Better late than never :) Cheers, Philip -- Philip Newton If you're not part of the solution, you're part of the precipitate ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: 08 Nov 2000 21:28:09 +0100 Organization: My own Private Self Lines: 60 Message-ID: <6uwveetdie.fsf@chonsp.franklin.ch> References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 973715292 1449 10.0.3.2 (8 Nov 2000 20:28:12 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 8 Nov 2000 20:28:12 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:67809 "Philip 'Yes, that's my address' Newton" writes: > On 07 Nov 2000 21:49:29 +0100, Neil Franklin > wrote: > > > Here WGL errs: The IBM PC / 8088 has memory management, known as > > "segments". > > > > If you took an SID (separate instruction data) PDP11 program to the > > PC, you simply set CS = base of instruction segment, DS = ES = SS = > > base of data segment. > > Turbo C called this the "small" memory model. Yes, MS also. That name I still remembered. > > Non-SID you set CS = xS = only data segment. > > And this was "tiny". Thats the name I was grappling for, so I left the names away. > > And off you were. These two arrangements were both common under MS-DOS. > > Yep. Turbo C also offered "medium", "compact", "large", and "huge" > memory models. These used 32-bit (actually, effectively only 20-bit) > addresses for code (compact), data (medium), or both (large). (I may > have compact and medium the wrong way round.) Yes. Medium is multi-code-one-data. Uses far jmp/jsr, but still no segment modifying by users program. > > without pointer errors crashing the OS, they may though strike other > > processes that follow direct after in physical memory! > > In small mode yes, but in tiny mode, CS == DS and you could overwrite > code if your pointers were off. User program code, yes. But the OS was usually below where any segment was pointing at the time user code had the processor. > I believe that the Turbo C runtime > (possibly also other compilers' runtimes, too) squirreled the first > few bytes of the segment off to a safe place before main() was called, > and on exit, compared the first few bytes to the saved copy. If they > had been changed, it was a fairly safe sign that someone had assigned > through a (16-bit, offset-only) NULL pointer, and an appropriate > message was emitted. Better late than never :) Cool technique. I had never heard of that one. I used MS tools under MS-DOS :-). -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### Path: chonsp.franklin.ch!not-for-mail From: Neil Franklin Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: 08 Nov 2000 21:28:38 +0100 Organization: My own Private Self Lines: 59 Message-ID: <6uu29itdhl.fsf@chonsp.franklin.ch> References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> NNTP-Posting-Host: chonsp.franklin.ch X-Trace: chonsp.franklin.ch 973715320 1449 10.0.3.2 (8 Nov 2000 20:28:40 GMT) X-Complaints-To: news@chonsp.franklin.ch NNTP-Posting-Date: 8 Nov 2000 20:28:40 GMT X-Newsreader: Gnus v5.7/Emacs 20.4 Xref: chonsp.franklin.ch alt.folklore.computers:67810 zaitcev@yahoo.com (Pete Zaitcev) writes: > > > Similar arguments were offered before Coherent was > > > brought up on the Atari ST, again with no memory management." > > > > Now that one had no memory managment, but with an properly written > > libc/malloc() it should be no big problem, so long you don't mind > > pointer errors wiping out the entire system, as was also the case with > > Unix V1 on PDP11/20! > > What did you do on fork(2)? Create an new data segment, i.e. allocate space for it, set the new heap-base pointer (points to bottom), copy parent processes data segment to it, and set new stack pointer (points to child_base + parent_stack - parent-base). All memory accesses are PC relative (in code segment), stack relative or base relative, with any offset going outside the segments space being a (deadly) bug. > Actually, I am thinking about porting ucLinux to SPARClite > (lame Fujitsu SPARC without MMU). UcLinux runs on embedded > 68ks without MMU. I was told that ucLinux uses A5 > based data and duplicates them. Most likely. That would make A5 the heap-base (A7 is allways the stack pointer). Using A5 + 16 bit offset addressing (signed, IIRC) will on 68k limit you to 32k data segment or force you to point A5 to base+32k. > As long as children are careful > not to attempt to unwind their duplicated stack Not unwind the stack _too_much_, which would overrun the stack end (= process data end) and damage possible other processes further up in memory. But that would be just a special case of pointer error wiping out the system, that I alluded to in my previous post. > and do exec() > ASAP, it works better than swapping whole parent out. No need to swap out the parent, as long as you have enouch RAM. Just give the child its own sandbox. > I pretty much gave up on running Linux binaries as long as > source compatibility can be maintained. *sigh* And x86 binaries do run badly on 68k and Sparc. And you can Unix on any processor with enough address space, the original PDP11/20 at AT&T having 24k RAM, IIRC, and no MMU. People often forget these humble origins. -- Neil Franklin, neil@franklin.ch.remove http://neil.franklin.ch/ Nerd, Geek, Hacker, Unix Guru, Sysadmin, Roleplayer, LARPer, Mystic ###### From: pete@fenelon.com Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Wed, 08 Nov 2000 20:16:10 -0000 Organization: Posted via Supernews, http://www.supernews.com Message-ID: Sender: Pete Fenelon References: <39DAA461.9C4A4CD8@olg.com> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> User-Agent: tin/1.5.6-20000803 ("Dust") (UNIX) (FreeBSD/4.1.1-STABLE (i386)) X-Complaints-To: newsabuse@supernews.com Lines: 13 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!machtgarnix.switch.ch!news-ge.switch.ch!news-fra1.dfn.de!news.tele.dk!194.176.220.129!newsfeed.icl.net!ldn-newsfeed.speedport.net!newsfeed.speedport.net!newspeer1.nac.net!news.he.net!sn-xit-03!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67898 d_lau@my-deja.com wrote: > I remember Coherent from Mark Williams a long time ago (I think the > name "Mark Williams" had a story of its own, but I can't remember the > details; it was told to me when I visited their HQ in Chicago many > years ago). I understand the company is no more. dmr once posted something here about a visit to Mark Williams to determine whether or not they'd infringed AT&T copyright. He decided they hadn't, and that Coherent was clearly a "new" implementation, although one done by people who knew AT&T code very well. I'm sure the article is there on deja, if deja can still be bothered archiving anything these days... pete ###### From: Joe Morris Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Wed, 8 Nov 2000 23:10:04 +0000 (UTC) Organization: Bozo Central in Atlanta Lines: 15 Message-ID: <8ucmgc$i2f$1@og1.olagrande.net> References: <39DAA461.9C4A4CD8@olg.com> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> NNTP-Posting-Host: og1.olagrande.net X-Trace: og1.olagrande.net 973725004 18511 63.98.209.2 (8 Nov 2000 23:10:04 GMT) X-Complaints-To: abuse@olagrande.net NNTP-Posting-Date: Wed, 8 Nov 2000 23:10:04 +0000 (UTC) User-Agent: tin/1.4.4-20000803 ("Vet for the Insane") (UNIX) (SunOS/5.8 (sun4u)) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!hammer.uoregon.edu!logbridge.uoregon.edu!uunet!sea.uu.net!sac.uu.net!lax.uu.net!dfw.uu.net!news.olagrande.net!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67878 pete@fenelon.com wrote: > dmr once posted something here about a visit to Mark Williams to determine > whether or not they'd infringed AT&T copyright. He decided they hadn't, > and that Coherent was clearly a "new" implementation, although one done > by people who knew AT&T code very well. I'm sure the article is there > on deja, if deja can still be bothered archiving anything these days... Sigh, this is just like the usenet of old. The only history is the repost. So basically we had about 4 years with "hot" archives and now we're back to the old ways. And to think I almost deleted 300MB of old news posts! -- Joe Morris, SysAdmin and Not Insane Atlanta stories: http://www.olagrande.net/users/jolomo/atlanta.html ###### Mime-Version: 1.0 X-Newsreader: knews 1.0b.1 References: <39DAA461.9C4A4CD8@olg.com> <8rf81v$ol8$1@nnrp1.deja.com> <%E7E5.342$D81.117141@nnrp2.sbc.net> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> From: wkt@henry.cs.adfa.edu.au (Warren Toomey) Subject: Re: Coherent Newsgroups: alt.folklore.computers Content-Type: text/plain; charset=us-ascii X-Original-NNTP-Posting-Host: henry.cs.adfa.edu.au Message-ID: <3a09d96f@news.adfa.edu.au> Date: 9 Nov 2000 09:53:35 +1000 Lines: 13 NNTP-Posting-Host: 131.236.1.1 X-Original-NNTP-Posting-Host: 131.236.1.1 X-Trace: 9 Nov 2000 09:53:31 +1000, 131.236.1.1 Organization: Australian National University Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!intgwpad.nntp.telstra.net!news1.optus.net.au!optus!clarion.carno.net.au!news.adfa.edu.au!henry.cs.adfa.edu.au Xref: chonsp.franklin.ch alt.folklore.computers:67862 In article , zaitcev@yahoo.com (Pete Zaitcev) writes: >> Now that one had no memory managment, but with an properly written >> libc/malloc() it should be no big problem, so long you don't mind >> pointer errors wiping out the entire system, as was also the case with >> Unix V1 on PDP11/20! > > What did you do on fork(2)? Keep all non-running processes on disk, and swap them in/out as required. Don't forget, processes were around 8K in size or less :-) Warren ###### From: Dennis Ritchie Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Thu, 09 Nov 2000 03:14:21 +0000 Organization: Bell Labs / Lucent Technologies Lines: 85 Message-ID: <3A0A168D.F5C2053D@bell-labs.com> References: <39DAA461.9C4A4CD8@olg.com> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> NNTP-Posting-Host: trux.cs.bell-labs.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 4.61 [en] (WinNT; U) X-Accept-Language: en Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!news.tesion.net!news.belwue.de!news.uni-stuttgart.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!news.netcologne.de!fr.clara.net!heighliner.fr.clara.net!xfer10.netnews.com!netnews.com!europa.netcrusader.net!204.127.161.3!wn3feed!worldnet.att.net!207.24.196.41!nntphub.cb.lucent.com!news Xref: chonsp.franklin.ch alt.folklore.computers:67865 pete@fenelon.com wrote: > dmr once posted something here about a visit to Mark Williams to determine > whether or not they'd infringed AT&T copyright. He decided they hadn't, > and that Coherent was clearly a "new" implementation, although one done > by people who knew AT&T code very well. I'm sure the article is there > on deja, if deja can still be bothered archiving anything these days... I managed to dredge up a local copy of one of the pieces I wrote about the visit to MWC. The file date is 10 Apr 1998, trustable only as an upper bound. I wish I'd saved the original notes from the visit. (I couldn't find it on Deja or elsewhere either). Herewith: An anecdote: sometime fairly early after the Mark Williams company started offering their Coherent system (a Unix clone), some AT&T legal people asked me to visit Mark Williams for purposes of determining whether what they were offering was a rip-off (i.e. essentially a copy) of the currently licensed Unix done by us. I find it hard to reconstruct the date this happened, but it was a long time ago; probably early 1980s. I went to Chicago with Otis Wilson, who was then involved in Unix licensing. It was a rather strange experience. The Mark Williams company was a paint producer, and I was given to understand that the subsidiary that was doing Coherent was, approximately, a corporation arranged by a father who, approaching retirement, had more or less shut down the older business and was using the corporate name and legal setup to help his son in a new venture. Otis and I visited the offices of Mark Williams on the outskirts of Chicago and were received with courtesy and some deference. We talked to the father and the son (Bob Swartz, i.e. the guy behind Coherent). There had been communication before, and from their point of view we were like the IRS auditors coming in. From my point of view, I felt the same, except that playing that role was a new, and not particularly welcome, experience. The locale of the company was in an industrial section and it definitely retained the flavor of a the offices of a paint company being recycled. What I actually did was to play around with Coherent and look for peculiarities, bugs, etc. that I knew about in the Unix distributions of the time. Whatever legal stuff had been talked about in the letters between MWC and AT&T didn't allow us to look at their source. I'd made some notes about things to look for. I concluded two things: First, that it was very hard to believe that Coherent and its basic applications were not created without considerable study of the OS code and details of its applications. Second, that looking at various corners convinced me that I couldn't find anything that was copied. It might have been that some parts were written with our source nearby, but at least the effort had been made to rewrite. If it came to it, I could never honestly testify that my opinion was that what they generated was irreproducible from the manual. I wrote up a detailed description of this. I can't find it, probably because at the time I was advised that it was privileged lawyer/client material. Partly at the time, partly thereafter, I learned that a variety of Unix enthusiasts (several from U. Toronto) had spent time there. In the event, "we" (=AT&T) backed off, possibly after other thinking and investigation that I'd wasn't involved in. So far as I know, after that MWC and Coherent were free to offer their system and allow it to succeed or fail in the market. I suppose there's a second story about the suit by USL against BSDI and then UCB, but my own involvement was far tinier and didn't get me a trip to Falls Church or Berkeley to snoop. What advice I offered in this situation was exactly in line with that about MWC/Coherent, and as it turned out the resolution (though more costly for all) was pretty much the same. (As a capper, Bob Swartz came by Bell Labs a week or so ago, and we had a pleasant social visit.) Dennis ###### From: Eric Fischer Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: 9 Nov 2000 06:37:54 GMT Lines: 10 Message-ID: <8udgo2$8re$1@bob.news.rcn.net> References: <39DAA461.9C4A4CD8@olg.com> <8u9sca$5g4$1@nnrp1.deja.com> <3A0A168D.F5C2053D@bell-labs.com> X-Trace: XTugpp/g81k36S3JpvHC7GFBmyiHn4LCGtgL/jX5x4I= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 9 Nov 2000 06:37:54 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Originator: enf@enteract.com (Eric Fischer) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:67873 Dennis Ritchie wrote: > I managed to dredge up a local copy of one of the pieces I wrote > about the visit to MWC. The file date is 10 Apr 1998, trustable > only as an upper bound. That was the date that appeared on your article. (If anybody wants a copy with headers intact, I can repost them.) eric ###### From: Peter Lund Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Thu, 9 Nov 2000 13:21:38 +0100 Organization: Department of Computer Science, University of Copenhagen Lines: 24 Message-ID: References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> <6uwveetdie.fsf@chonsp.franklin.ch> NNTP-Posting-Host: ask.diku.dk Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: munin.diku.dk 973772499 29403 130.225.96.225 (9 Nov 2000 12:21:39 GMT) X-Complaints-To: news@diku.dk NNTP-Posting-Date: 9 Nov 2000 12:21:39 GMT In-Reply-To: <6uwveetdie.fsf@chonsp.franklin.ch> Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!machtgarnix.switch.ch!news-ge.switch.ch!news.maxwell.syr.edu!howland.erols.net!news.net.uni-c.dk!munin.diku.dk!ask.diku.dk!firefly Xref: chonsp.franklin.ch alt.folklore.computers:67928 On 8 Nov 2000, Neil Franklin wrote: > > I believe that the Turbo C runtime > > (possibly also other compilers' runtimes, too) squirreled the first > > few bytes of the segment off to a safe place before main() was called, > > and on exit, compared the first few bytes to the saved copy. If they > > had been changed, it was a fairly safe sign that someone had assigned > > through a (16-bit, offset-only) NULL pointer, and an appropriate > > message was emitted. Better late than never :) > > Cool technique. I had never heard of that one. I used MS tools under > MS-DOS :-). That's not quite how they did it. The segment started with some static stuff: a copyright message and some startup code. At exit the area was checked again to see if the copyright message and (some of the) code was still the same. I think they just checksummed it. Nothing was squirreled away. -Peter And Visual Basic programmers should be paid minimum wage :) -- Jeffrey Straszheim (on comp.lang.functional) ###### From: aek@spies.com (Al Kossow) Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Thu, 09 Nov 2000 07:25:51 -0800 Organization: Apple Computer, Inc. Lines: 118 Message-ID: References: <39DAA461.9C4A4CD8@olg.com> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> <3A0A168D.F5C2053D@bell-labs.com> NNTP-Posting-Host: haxrus.apple.com X-Trace: news.apple.com 973783551 24010 17.205.21.66 (9 Nov 2000 15:25:51 GMT) X-Complaints-To: usenet@news.apple.com NNTP-Posting-Date: 9 Nov 2000 15:25:51 GMT Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsxfer.eecs.umich.edu!newsxfer3.itd.umich.edu!forum.apple.com!news.apple.com!haxrus.apple.com!user Xref: chonsp.franklin.ch alt.folklore.computers:68054 In article <3A0A168D.F5C2053D@bell-labs.com>, Dennis Ritchie wrote: > > It was a rather strange experience. The Mark Williams company > was a paint producer, and I was given to understand that > the subsidiary that was doing Coherent was, approximately, > a corporation arranged by a father who, approaching > retirement, had more or less shut down the older business > and was using the corporate name and legal setup to help > his son in a new venture. > here is a posting I found on MWC's history, from 1995 --- From: wgl@MCS.COM (William G. Lederer) Newsgroups: alt.folklore.computers Subject: Sad news on the passing of MWC (Mark Williams Company) Date: 4 Apr 1995 16:39:55 -0500 Organization: Another MCSNet Subscriber, Chicago's First Public-Access Internet! Lines: 86 Message-ID: <3lsebb$slh@Mercury.mcs.com> NNTP-Posting-Host: mercury.mcs.com Summary: some notes on early days Keywords: coherent Let's C Reactions to the Sad News It was with mixed emotions that I read of the "sad news" of the closing of Mark Williams Company on January 31, 1995. As an alumnus of Mark Williams, I have followed the discussion since then here in comp.os.coherent, particularly with respect to whether or not the news of this transition was sad, expected, life threatening, or accidental. Having been away from Mark Williams for a few years, I cannot speak to the events leading up to the closing of the doors, nor the reasons for the closing. However, I would like to offer some memories about life on the inside of Mark Williams and some of the events and people that made it a special place to work during a very exciting time in the industry. If you love doing software, working in a software product environment is the place to be. You likely are building tools that you will yourself use (or eating your own dogfood, as some would say). You know how the tools work; there is no noisy channel between you and the people who know how it should work. You don't need to spend weeks or years gathering requirements, making buy vs build decisions. Maybe you spend some time convincing management that is what to do, but by and large you spend time building and refining products that are useful. Coherent was originally written by four very bright programmers from the University of Waterloo (arguably the best undergrad CS department in the hemisphere if not anywhere) to run on the PDP-11. This was done in an astonishingly short time. As it was based on the concepts of V7 Unix, it was designed to be portable, and as a result, it was easily moved to the z8000, and in 1982 to the IBM PC. Keep in mind that this was with no memory management hardware, something that experts kept saying couldn't be done. Similar arguments were offered before Coherent was brought up on the Atari ST, again with no memory management. The C compiler technology became of economical interest in the mid 80s, and MWC was early to market with compilers for the IBM PC, clones, and the DEC Rainbow. As Borland was doing very well with Turbo Pascal, MWC anticipated their entry into the C compiler market by two full years with Let's C (one of the better puns in the C compiler market). Coherent was also ported to a machine built by Commodore based on the Z8000-2. Unfortunatly, this machine never saw the market, as Commodore felt it had to decide between the Z8000 machine and the Amiga. The last prototype of this machine sat on a lab table, parts exposed. Well after Commodore pulled the plug, we left the machine running, and occasionally fired off massive compiles of parts of the system. When it was finally turned off, the 'uptime' command reported that Coherent had been up for 72 days. But in some sense, the people at Mark Williams made it a high-energy place to be. At that time, there were never more than about 20 people ath MWC. Yet there were 14 languages fluently spoken, including Hebrew, Arabic, Greek, Ukranian, and Polish. Three people were members of PhiBeta Keppa, one PhD student (who got the degree before leaving Mark Williams) in divinity studies, an M.D., (and another one in recent times), a classically educated scholar, a former encyclopedia salesman, daughter of a high-tech entrepreneur, a student of Donald Knuth, an aspiring actor, a former IBM salesman, a real estate developer whose spouse was a carpenter, the son of an advertising display manufacturer, a political activist working for nuclear disarmament, a chess master who played simultaneous blindfold chess games, a part-time locksmith, and a pilot or two. All have moved on to other endeavours. Many have started successful businesses of their own, others have moved onto larger ventures. In this sense, MWC, for many of us then and I suspect in recent days as well, was a very high-energy learning experience, which is why sometimes the term 'alumnus' is used. In another sense this is sad, but in the sense of leaving behind a fondly-remembered place on a journey, or moving out of a familiar house to go off into the world. Should we worry about those MWC folk that had to turn out the lights and unplug the phones? I don't think so -- they are all quite talented and will land on their feet. Unless you are inclined to worry about friends and acquaintances heading off into this increasingly re-engineered world. -- ------------------------------------------------------------------------------- William G. Lederer wgl@mcs.com CIEX, Incorporated wgl@wglu.chi.il.us -- The eBay Curse: "May you find everything you're looking for.." ###### From: pete@fenelon.com Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Thu, 09 Nov 2000 19:27:52 -0000 Organization: Posted via Supernews, http://www.supernews.com Message-ID: Sender: Pete Fenelon References: <39DAA461.9C4A4CD8@olg.com> <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <8u9sca$5g4$1@nnrp1.deja.com> <3A0A168D.F5C2053D@bell-labs.com> User-Agent: tin/1.5.6-20000803 ("Dust") (UNIX) (FreeBSD/4.1.1-STABLE (i386)) X-Complaints-To: newsabuse@supernews.com Lines: 14 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!sunqbc.risq.qc.ca!news-peer.gip.net!news.gsl.net!gip.net!newshub2.rdc1.sfba.home.com!news.home.com!newsfeed.mesh.ad.jp!nntp-relay.ihug.net!ihug.co.nz!sn-xit-02!sn-post-01!supernews.com!corp.supernews.com!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68095 Al Kossow wrote: (quoting) > > The C compiler technology became of economical interest in the mid 80s, > and MWC was early to market with compilers for the IBM PC, clones, and > the DEC Rainbow. As Borland was doing very well with Turbo Pascal, MWC > anticipated their entry into the C compiler market by two full years > with Let's C (one of the better puns in the C compiler market). A very nice compiler it was too - it surfaced on (at least) the Atari ST as well, and was pretty much my favourite C compiler on there, until the rather good Sozobon freeware one arrived. pete ###### From: Brian Inglis Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Thu, 09 Nov 2000 21:32:30 -0700 Organization: Systematic Software Lines: 59 Message-ID: <7trm0toh6s2jq9ls38tqlu6gdvh0tivqa8@4ax.com> References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> Reply-To: Brian.dot.Inglis@SystematicSw.ab.ca NNTP-Posting-Host: h-207-148-135-146.dial.cadvision.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: news3.cadvision.com 973830751 11908 207.148.135.146 (10 Nov 2000 04:32:31 GMT) X-Complaints-To: news@cadvision.com NNTP-Posting-Date: Fri, 10 Nov 2000 04:32:31 +0000 (UTC) X-Newsreader: Forte Agent 1.7/32.534 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.direct.ca!look.ca!hekyl.ab.tac.net!nntp.cadvision.com!207.228.64.17.MISMATCH!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68149 On Wed, 08 Nov 2000 12:45:39 +0100, "Philip 'Yes, that's my address' Newton" wrote: >On 07 Nov 2000 21:49:29 +0100, Neil Franklin >wrote: > >> Here WGL errs: The IBM PC / 8088 has memory management, known as >> "segments". >> >> If you took an SID (separate instruction data) PDP11 program to the >> PC, you simply set CS = base of instruction segment, DS = ES = SS = >> base of data segment. > >Turbo C called this the "small" memory model. > >> Non-SID you set CS = xS = only data segment. > >And this was "tiny". > >> And off you were. These two arrangements were both common under MS-DOS. > >Yep. Turbo C also offered "medium", "compact", "large", and "huge" >memory models. These used 32-bit (actually, effectively only 20-bit) >addresses for code (compact), data (medium), or both (large). (I may >have compact and medium the wrong way round.) "Huge" model was for >32-bit code, 32-bit data where an individual data item (an array, for >example) could be larger than 64 KB (one segment), so pointer >arithmetic was always done normalised (0 <= offset <= 0xf) so that the >offset didn't overflow and wrap around to the beginning of the >segment. > >> What the PCs 8088 did not have, was memory _protection_, nor an kernel >> or supervisor mode to prevent user programs from directly accessing IO >> hardware or segment registers. But so long programs did not try to >> deliberately do bad thing (and the C compiler simply does not generate >> code to do that) it will work just like a PDP11/40 and above, even >> without pointer errors crashing the OS, they may though strike other >> processes that follow direct after in physical memory! > >In small mode yes, but in tiny mode, CS == DS and you could overwrite >code if your pointers were off. I believe that the Turbo C runtime >(possibly also other compilers' runtimes, too) squirreled the first >few bytes of the segment off to a safe place before main() was called, The first 0x100/256. bytes were the Program Segment Prefix and had various bits of fixed content, starting with an int 32 in the first two bytes. >and on exit, compared the first few bytes to the saved copy. If they >had been changed, it was a fairly safe sign that someone had assigned >through a (16-bit, offset-only) NULL pointer, and an appropriate >message was emitted. Better late than never :) > >Cheers, >Philip Thanks. Take care, Brian Inglis Calgary, Alberta, Canada -- Brian_Inglis@CSi.com (Brian dot Inglis at SystematicSw dot ab dot ca) use address above to reply ###### From: "Philip 'Yes, that's my address' Newton" Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Nov 2000 08:20:53 +0100 Organization: very little Lines: 24 Message-ID: References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> <6uwveetdie.fsf@chonsp.franklin.ch> NNTP-Posting-Host: internet-router.datenrevision.de (195.222.209.7) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: fu-berlin.de 973840857 1877510 195.222.209.7 (16 [11583]) X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!fu-berlin.de!uni-berlin.de!internet-router.datenrevision.DE!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68124 On 08 Nov 2000 21:28:09 +0100, Neil Franklin wrote: > "Philip 'Yes, that's my address' Newton" writes: > > > On 07 Nov 2000 21:49:29 +0100, Neil Franklin > > wrote: > > > > > without pointer errors crashing the OS, they may though strike other > > > processes that follow direct after in physical memory! > > > > In small mode yes, but in tiny mode, CS == DS and you could overwrite > > code if your pointers were off. > > User program code, yes. But the OS was usually below where any segment > was pointing at the time user code had the processor. Ah yes. True. Cheers, Philip -- Philip Newton If you're not part of the solution, you're part of the precipitate ###### From: "Philip 'Yes, that's my address' Newton" Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Nov 2000 08:26:41 +0100 Organization: very little Lines: 35 Message-ID: References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> <6uwveetdie.fsf@chonsp.franklin.ch> NNTP-Posting-Host: internet-router.datenrevision.de (195.222.209.7) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: fu-berlin.de 973841205 1824732 195.222.209.7 (16 [11583]) X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!fu-berlin.de!uni-berlin.de!internet-router.datenrevision.DE!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68144 On Thu, 9 Nov 2000 13:21:38 +0100, Peter Lund wrote: > That's not quite how they did it. > The segment started with some static stuff: a copyright message and some > startup code. At exit the area was checked again to see if the copyright > message and (some of the) code was still the same. I think they just > checksummed it. Nothing was squirreled away. But before that, there was other stuff. IIRC, to ensure CP/M compatibility, the first two bytes of a segment were CD 20 -- in assembly, INT 20h; and a word 0000 was pushed on the stack when the program started. This meant that instead of terminating a program "new style" (with INT 21h, with AH = 4Ch and AL = exit code), you could either call INT 20h, or CALL 0 (a la CP/M) or even just RET (which would pop the 0000 off the stack, branch to address zero, and then call INT 20h which would terminate the process). I don't remember any copyright message at the beginning of the segment, actually; the first 80h bytes I believe contained stuff such as space for two file name buffers (11 bytes each) which would be initialised with the first two command-line arguments if they looked like file names, and other things; the next 80h bytes were the command line arguments. It's been too long since I did CP/M, but I think the PSP (program segment prefix) looked just like the first 100h bytes of CP/M address space. They might even have had an INT 21h or something like that at address 5 for compatibility with CP/M programs which accessed BDOS through CALL 5. Ah - fond memories of programming a CP/M machine in CP/M assembly (which was assembled by a program written in BASIC that I typed in from a magazine :). Cheers, Philip -- Philip Newton If you're not part of the solution, you're part of the precipitate ###### From: "Philip 'Yes, that's my address' Newton" Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Nov 2000 08:29:33 +0100 Organization: very little Lines: 18 Message-ID: References: <39DAA461.9C4A4CD8@olg.com> <8u9sca$5g4$1@nnrp1.deja.com> <3A0A168D.F5C2053D@bell-labs.com> <8udgo2$8re$1@bob.news.rcn.net> NNTP-Posting-Host: internet-router.datenrevision.de (195.222.209.7) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: fu-berlin.de 973841377 1824732 195.222.209.7 (16 [11583]) X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!machtgarnix.switch.ch!news-ge.switch.ch!news.imp.ch!fu-berlin.de!uni-berlin.de!internet-router.datenrevision.DE!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68130 On 9 Nov 2000 06:37:54 GMT, Eric Fischer wrote: > Dennis Ritchie wrote: > > > I managed to dredge up a local copy of one of the pieces I wrote > > about the visit to MWC. The file date is 10 Apr 1998, trustable > > only as an upper bound. > > That was the date that appeared on your article. (If anybody wants > a copy with headers intact, I can repost them.) I'd like it. Cheers, Philip -- Philip Newton If you're not part of the solution, you're part of the precipitate ###### From: Peter Lund Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Nov 2000 09:47:42 +0100 Organization: Department of Computer Science, University of Copenhagen Lines: 74 Message-ID: References: <8rtrpm$1r6$1@nnrp1.deja.com> <8s0ai9$28h$1@nnrp1.deja.com> <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> <6uwveetdie.fsf@chonsp.franklin.ch> NNTP-Posting-Host: ask.diku.dk Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Trace: munin.diku.dk 973846064 17806 130.225.96.225 (10 Nov 2000 08:47:44 GMT) X-Complaints-To: news@diku.dk NNTP-Posting-Date: 10 Nov 2000 08:47:44 GMT In-Reply-To: Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!psinet-eu-nl!newsfeeds.belnet.be!news.belnet.be!news.tele.dk!129.240.148.23!uio.no!130.227.3.87.MISMATCH!neel.uni2.net!news.net.uni-c.dk!munin.diku.dk!ask.diku.dk!firefly Xref: chonsp.franklin.ch alt.folklore.computers:68099 On Fri, 10 Nov 2000, Philip 'Yes, that's my address' Newton wrote: > On Thu, 9 Nov 2000 13:21:38 +0100, Peter Lund wrote: > > > That's not quite how they did it. > > The segment started with some static stuff: a copyright message and some > > startup code. At exit the area was checked again to see if the copyright > > message and (some of the) code was still the same. I think they just > > checksummed it. Nothing was squirreled away. > > But before that, there was other stuff. IIRC, to ensure CP/M > compatibility, the first two bytes of a segment were CD 20 -- in > assembly, INT 20h; and a word 0000 was pushed on the stack when the Yep. > program started. This meant that instead of terminating a program "new > style" (with INT 21h, with AH = 4Ch and AL = exit code), you could > either call INT 20h, or CALL 0 (a la CP/M) or even just RET (which > would pop the 0000 off the stack, branch to address zero, and then > call INT 20h which would terminate the process). Yep. > I don't remember any copyright message at the beginning of the > segment, actually; the first 80h bytes I believe contained stuff such It wasn't in the PSP :) It was in the crt0 module. In the two-segment memory model ("small", I think) CS:0 pointed at the PSP but DS:0 pointed to the start of the static data - which contained the copyright message. > as space for two file name buffers (11 bytes each) which would be > initialised with the first two command-line arguments if they looked Yep. > like file names, and other things; Such as the segment of the environment list for the process, a pointer to the file descriptor table for the process - and the table itself it was small enough to fit. And a bunch of other stuff. > the next 80h bytes were the command > line arguments. It's been too long since I did CP/M, but I think the Yep. > PSP (program segment prefix) looked just like the first 100h bytes of > CP/M address space. They might even have had an INT 21h or something Yep. > like that at address 5 for compatibility with CP/M programs which > accessed BDOS through CALL 5. Ah - fond memories of programming a CP/M Yep, they did. > machine in CP/M assembly (which was assembled by a program written in > BASIC that I typed in from a magazine :). > > Cheers, > Philip > -- > Philip Newton > If you're not part of the solution, you're part of the precipitate > -Peter Code is much like elephant dung. The more code, the more bugs. You have to let it rest a while to make sure that the bugs are gone (and large heaps rot slowly). ###### From: "Philip 'Yes, that's my address' Newton" Newsgroups: comp.arch,alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Nov 2000 10:19:38 +0100 Organization: very little Lines: 49 Message-ID: References: <8u74l6$sch$1@nnrp1.deja.com> <6ur94na4o6.fsf@chonsp.franklin.ch> <6uwveetdie.fsf@chonsp.franklin.ch> NNTP-Posting-Host: internet-router.datenrevision.de (195.222.209.7) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: fu-berlin.de 973847989 1516704 195.222.209.7 (16 [11583]) X-Newsreader: Forte Agent 1.8/32.548 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.ifi.unizh.ch!news.imp.ch!fu-berlin.de!uni-berlin.de!internet-router.datenrevision.DE!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68150 On Fri, 10 Nov 2000 09:47:42 +0100, Peter Lund wrote: > On Fri, 10 Nov 2000, Philip 'Yes, that's my address' Newton wrote: > > > I don't remember any copyright message at the beginning of the > > segment, actually; the first 80h bytes I believe contained stuff such > > It wasn't in the PSP :) > It was in the crt0 module. In the two-segment memory model ("small", I > think) CS:0 pointed at the PSP but DS:0 pointed to the start of the static > data - which contained the copyright message. Ah, small memory model. I was thinking of tiny, where DS == CS, and writing into NULL would overwrite the CD 20 etc. > > as space for two file name buffers (11 bytes each) which would be > > initialised with the first two command-line arguments if they looked > > Yep. I think they were called FCBs, weren't they? And originally used for all sorts of stuff such as opening, renaming (which is why you needed two), and so on. > > like file names, and other things; > > Such as the segment of the environment list for the process, a pointer to > the file descriptor table for the process - and the table itself it was > small enough to fit. And a bunch of other stuff. Ah yes. I thought there was something about the environment in there. But it's been too long. > > They might even have had an INT 21h or something > > like that at address 5 for compatibility with CP/M programs which > > accessed BDOS through CALL 5. Ah - fond memories of programming a CP/M > > Yep, they did. :) Fun. Again -- ah, the memories. Cheers, Philip -- Philip Newton If you're not part of the solution, you're part of the precipitate ###### From: Eric Fischer Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: 11 Nov 2000 19:23:08 GMT Lines: 96 Message-ID: <8uk6as$7jc$1@bob.news.rcn.net> References: <39DAA461.9C4A4CD8@olg.com> <3A0A168D.F5C2053D@bell-labs.com> <8udgo2$8re$1@bob.news.rcn.net> X-Trace: lfM85kg75j4Birjs7MRfPaG+QM+khXDcb9HDPaSf0lI= X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 11 Nov 2000 19:23:08 GMT X-Newsreader: trn 4.0-test72 (19 April 1999) Originator: enf@enteract.com (Eric Fischer) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news.nextra.ch!news1.sunrise.ch!news.imp.ch!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail Xref: chonsp.franklin.ch alt.folklore.computers:68175 Philip 'Yes, that's my address' Newton quotes me: > > (If anybody wants a copy with headers intact, I can repost them.) > > I'd like it. OK, then, here it is: From: Dennis Ritchie Newsgroups: alt.folklore.computers Subject: Re: Coherent Date: Fri, 10 Apr 1998 08:05:27 +0100 Organization: Bell Labs, Lucent Technologies Lines: 70 Message-ID: <352DC4B7.3030@bell-labs.com> References: <3528ff87.19065567@news.nabi.net> <6gc5mt$hg4$1@news.sas.ab.ca> <6gjmu6$e32$1@roch.zetnet.co.uk> Reply-To: dmr@bell-labs.com NNTP-Posting-Host: cebu.cs.bell-labs.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 3.01 (Win95; U) An anecdote: sometime fairly early after the Mark Williams company started offering their Coherent system (a Unix clone), some AT&T legal people asked me to visit Mark Williams for purposes of determining whether what they were offering was a rip-off (i.e. essentially a copy) of the currently licensed Unix done by us. I find it hard to reconstruct the date this happened, but it was a long time ago; probably early 1980s. I went to Chicago with Otis Wilson, who was then involved in Unix licensing. It was a rather strange experience. The Mark Williams company was a paint producer, and I was given to understand that the subsidiary that was doing Coherent was, approximately, a corporation arranged by a father who, approaching retirement, had more or less shut down the older business and was using the corporate name and legal setup to help his son in a new venture. Otis and I visited the offices of Mark Williams on the outskirts of Chicago and were received with courtesy and some deference. We talked to the father and the son (Bob Swartz, i.e. the guy behind Coherent). There had been communication before, and from their point of view we were like the IRS auditors coming in. From my point of view, I felt the same, except that playing that role was a new, and not particularly welcome, experience. The locale of the company was in an industrial section and it definitely retained the flavor of a the offices of a paint company being recycled. What I actually did was to play around with Coherent and look for peculiarities, bugs, etc. that I knew about in the Unix distributions of the time. Whatever legal stuff had been talked about in the letters between MWC and AT&T didn't allow us to look at their source. I'd made some notes about things to look for. I concluded two things: First, that it was very hard to believe that Coherent and its basic applications were not created without considerable study of the OS code and details of its applications. Second, that looking at various corners convinced me that I couldn't find anything that was copied. It might have been that some parts were written with our source nearby, but at least the effort had been made to rewrite. If it came to it, I could never honestly testify that my opinion was that what they generated was irreproducible from the manual. I wrote up a detailed description of this. I can't find it, probably because at the time I was advised that it was privileged lawyer/client material. Partly at the time, partly thereafter, I learned that a variety of Unix enthusiasts (several from U. Toronto) had spent time there. In the event, "we" (=AT&T) backed off, possibly after other thinking and investigation that I'd wasn't involved in. So far as I know, after that MWC and Coherent were free to offer their system and allow it to succeed or fail in the market. I suppose there's a second story about the suit by USL against BSDI and then UCB, but my own involvement was far tinier and didn't get me a trip to Falls Church or Berkeley to snoop. What advice I offered in this situation was exactly in line with that about MWC/Coherent, and as it turned out the resolution (though more costly for all) was pretty much the same. (As a capper, Bob Swartz came by Bell Labs a week or so ago, and we had a pleasant social visit.) Dennis