From brad at anduin.eldar.org Sat May 1 01:38:45 2010 From: brad at anduin.eldar.org (Brad Spencer) Date: Fri, 30 Apr 2010 11:38:45 -0400 (EDT) Subject: [TUHS] UniFlex In-Reply-To: <20100430095823.78e5a8ac@cnb.csic.es> (jrvalverde@cnb.csic.es) References: <130de1e4e4162da466b3dc04bbc53c70.squirrel@webmail.xs4all.nl> <20100425022942.GA15137@dereel.lemis.com> <201004261903.36146.wes.parish@paradise.net.nz> <20100426080700.GD15137@dereel.lemis.com> <1272274418.4bd55df2c23b4@www.paradise.net.nz> <4BD79DE2.2010203@laposte.net> <20100430095823.78e5a8ac@cnb.csic.es> Message-ID: <201004301538.o3UFcjtY002780@anduin.eldar.org> P.S. (sorry for following up to myself) I had a look into the files, and the system comes with commented source code in assembler. FWIW this is a full distribution, including development environment (C, Cobol, Fortran-77,..), editors, kernel, VSAM database, etc... all of it with source code and documentation, From my first cursory look most of it is written in assemble, comes with sample test code and is documented enough to be understandable. The pascal, cobol and fortran 77 compilers are written in pascal (!), the C compiler is written in assembler. It looks like the environment must have been only vaguely UNIX-like but yet I find it mesmerizing enough considering where it ran and when. It adds another dimension to understand the impact UNIX had and how it spun off lookalikes and sprung the imagination of developers of the time expanding its heritage sideways. j -- EMBnet/CNB Scientific Computing Service Solving all your computer needs for Scientific Research. http://bioportal.cnb.csic.es http://www.es.embnet.org _______________________________________________ TUHS mailing list TUHS at minnie.tuhs.org https://minnie.tuhs.org/mailman/listinfo/tuhs Wow, what a blast from the past. I used Flex on the Radio Shack Color Computer in 64MB of memory in the '80s for a while. Yes, the entire thing was assembly and some Fortran. I did not have the source to most of it for my version, only the compiled binaries. The version I used ran off floppy disks, of course, and used a strange sector layout. Something like 17 sectors of 256 bytes, instead of the 18 sectors 256 bytes that the Color Computer usually used. Neat system in some ways. Way more "professional" then is peers, except for OS/9 from Microware. Flex was only very vaguely Unix, however. OS/9, which I used a whole lot on the Color Computer and Color Computer 3, was a lot more Unix like. I understand that the 6809 version is floating around the Net. It was also all 6809 assembly, multitasking and multiuser. Very Unix V4/V5/V6 like in a number of ways. -- Brad Spencer - brad at anduin.eldar.org - KC8VKS http://anduin.eldar.org - & - http://anduin.ipv6.eldar.org [IPv6 only] From dugo at xs4all.nl Wed May 5 04:26:52 2010 From: dugo at xs4all.nl (Jacob Goense) Date: Tue, 4 May 2010 20:26:52 +0200 Subject: [TUHS] 386BSD on Bochs & Qemu... In-Reply-To: <20100426032303.GC15137@dereel.lemis.com> References: <130de1e4e4162da466b3dc04bbc53c70.squirrel@webmail.xs4all.nl> <20100425022942.GA15137@dereel.lemis.com> <5a7c3451f8bbb8efaaa6b9c809214a55.squirrel@webmail.xs4all.nl> <20100426032303.GC15137@dereel.lemis.com> Message-ID: <97c86725a66a6581c02553876124c880.squirrel@webmail.xs4all.nl> On Mon, April 26, 2010 05:23 "Greg 'groggy' Lehey" wrote: > OK, I mounted the CD-ROM and looked for copyright statements. The > only one I found was /COPYRGHT.TXT, a modified 8 paragraph BSD license > with \r\n line delimiters, attached. > > Arguably it only applies to the OS sources, but it's the only license > I can see. Indeed, it can be argued that the statement refers to 386BSD Release 1.0 sec, not the 386BSD Reference CD-ROM as a whole, but it is hard to draw a line, if any. > It probibits commercial distributions, but the important > clause from our point of view is: > > * 5. Non-commercial distribution of the complete source and/or binary > * release at no charge to the user (such as from an official Internet > * archive site) is permitted. > > I was going to take out just the source tree, but another clause > states: > > * 7. Non-commercial and/or commercial distribution of an incomplete, > * altered, or otherwise modified source and/or binary release is not > * permitted. Same issue here, would excluding eg. .book & .articles make the Release incomplete or not? > Since this is the only copyright statement, I assume that this means I > can put up the entire CD image, but not just part of it, so that's > what I've done. It's at http://www.lemis.com/grog/src/386BSD-1.0.bz2 . The only thing I could find that makes some sort of demarcation is the installer, which leaves some bits behind when doing a full install. Without vehement arguing from stakeholders or copyright lawyers to the contrary I assume the same and thank you for putting it up there. /Jacob From dugo at xs4all.nl Wed May 5 05:37:15 2010 From: dugo at xs4all.nl (Jacob Goense) Date: Tue, 4 May 2010 21:37:15 +0200 Subject: [TUHS] 386BSD on Bochs & Qemu... In-Reply-To: References: <130de1e4e4162da466b3dc04bbc53c70.squirrel@webmail.xs4all.nl> <20100425022942.GA15137@dereel.lemis.com> <5a7c3451f8bbb8efaaa6b9c809214a55.squirrel@webmail.xs4all.nl> <20100426032303.GC15137@dereel.lemis.com> Message-ID: <69d171f5fd5a71aebb54fec721835024.squirrel@webmail.xs4all.nl> On Mon, April 26, 2010 07:22 "Jason Stevens" wrote: > Well I've been able to find this much out... > > The CD has some kind of weird 'live' CD filesystem to it... It would > seem that 386BSD 1.0 demanded you have an Adaptec 1542 controller > hooked up, and with special roms & whatnot it could 'boot' from the > CD... > Needless to say, this predates anything like IDE CDROM's or or what > most emulators will emulate. That "Bootable CD" button on the CD cover is just a marketing fact AFAICT. I don't have the foggiest how that was done in the pre eltorito days on an x86. > That being said, they did include the 'boot' program which is touched > on in the magazine series, as a MS-DOS bootloader. Mock code aside, no source code for boot.exe. No source code for the install program and no Tiny 386BSD Release 1.0 boot floppy image on the CD. I can't get my head around why they bothered to document to the point where you can almost port BSD to a Commodore 64 and leave these bits out. > So I've just slapped together a MS-DOS floppy, with the boot & 386bsd > kernel and tried it on on Qemu, to an early kernel panic. After numerous attect vectors I took the MS-DOS5 floppy route w/ Qemu as well, but nothing could get it to boot cdrom. Then in a recalcitrant mood I stuck the image to -hdb, an empty 1G drive to -hda, booted flop and gave a BOOT.EXE 386BSD.DDB wd1d at the DOS prompt and booted into The 386BSD SAMPLER. The installer ran fine as long as it didn't have to deal with swap space. Not that it left me with a bootable system, but it's a start. Will document on gunkies.org if I can get things stable. > I'll have to test later if it can 'mount' an ISO image that's been > 'dd''d to a hard disk..... Skip the dd'ing, Qemu eats the image raw ;) /Jacob From claunia at claunia.com Wed May 5 05:41:22 2010 From: claunia at claunia.com (Natalia Portillo) Date: Tue, 4 May 2010 20:41:22 +0100 Subject: [TUHS] 386BSD on Bochs & Qemu... In-Reply-To: <69d171f5fd5a71aebb54fec721835024.squirrel@webmail.xs4all.nl> References: <130de1e4e4162da466b3dc04bbc53c70.squirrel@webmail.xs4all.nl> <20100425022942.GA15137@dereel.lemis.com> <5a7c3451f8bbb8efaaa6b9c809214a55.squirrel@webmail.xs4all.nl> <20100426032303.GC15137@dereel.lemis.com> <69d171f5fd5a71aebb54fec721835024.squirrel@webmail.xs4all.nl> Message-ID: <3144E4A4-EF81-4DD5-8F44-C66367616EC0@claunia.com> El 04/05/2010, a las 20:37, Jacob Goense escribió: > On Mon, April 26, 2010 07:22 "Jason Stevens" wrote: >> Well I've been able to find this much out... >> >> The CD has some kind of weird 'live' CD filesystem to it... It would >> seem that 386BSD 1.0 demanded you have an Adaptec 1542 controller >> hooked up, and with special roms & whatnot it could 'boot' from the >> CD... >> Needless to say, this predates anything like IDE CDROM's or or what >> most emulators will emulate. > > That "Bootable CD" button on the CD cover is just a marketing fact > AFAICT. I don't have the foggiest how that was done in the pre eltorito > days on an x86. SCSI HBAs with integrated boot firmware in BIOS compatible way (that is, trapping INT 13h) can be used to boot ANYTHING in SCSI that behaves like a random access block device. That means, floppies, LS-120, ZIP, hard disks, CD-ROMs. From reed at reedmedia.net Thu May 6 06:55:24 2010 From: reed at reedmedia.net (Jeremy C. Reed) Date: Wed, 5 May 2010 15:55:24 -0500 (CDT) Subject: [TUHS] wrong HISTORY in 4.4BSD manual pages? Message-ID: I was looking at several NetBSD manual pages and saw that some HISTORY sections had wrong .Bx or BSD reference like: HISTORY The xstr command appeared in 3.0BSD. I looked at a few and saw this was in 4.4BSD manual pages. By the way, when were these history sections added? (They aren't in 4.2BSD manual pages. I should look at 4.3 before asking ...) I didn't see any that refered to original Berkeley UNIX Software Tape nor 2BSD. But from looking at the 1BSD and 2BSD, I see: apropos was in 2bsd colcrt was in 1bsd but not in 2bsd even though 2BSD iul, soelim, and ssp manuals referenced it (why missing from 2BSD?) colrm was in 1bsd but not in 2bsd csh was in 2bsd (even 1bsd referenced the upcoming "csh") ctags was in 2bsd, but as a shell script using ed expand was in 1bsd and 2bsd finger was in 2bsd fmt was in 2bsd from was in 2bsd head was in 1bsd and 2bsd lock was in 2bsd last was in 1bsd and 2bsd mkstr was in 1bsd and 2bsd msgs was in 2bsd printenv was in 2bsd soelim was in first BSD and 2BSD tset was in 1bsd and 2bsd w was in 2bsd as finger -sf whatis was in 2bsd whereis was in 2bsd xstr was in 2bsd lastlog file format was in 1bsd and 2bsd (?? maybe different format??) Any comments on the above? Or is this simply because "2BSD" is not a operating system release per se, so "3.0BSD" is correct? But this makes me wonder if my 2BSD versions are newer than first 2BSD, so really 3BSD is correct for some of this. I was going to ask a NetBSD list about this to fix these histories, but decided to consult TUHS instead. Okay to change history to fix history? :) From reed at reedmedia.net Thu May 6 11:40:12 2010 From: reed at reedmedia.net (Jeremy C. Reed) Date: Wed, 5 May 2010 20:40:12 -0500 (CDT) Subject: [TUHS] wrong HISTORY in 4.4BSD manual pages? In-Reply-To: References: Message-ID: > Or is this simply because "2BSD" is not a operating system release per > se, so "3.0BSD" is correct? > > But this makes me wonder if my 2BSD versions are newer than first > 2BSD, so really 3BSD is correct for some of this. I was asked off list what version of 2BSD as it may have been the 4.x stuff backported. So it is the 2bsd.tar.gz from the minnie.tuhs.org archive (which is near identical to the spencer_2bsd.tar.gz). Its TAPE file says: May 10, 1979. The READ_ME says: Thu Apr 19 23:25:43 PST 1979. And don't see any date stamps included in any files newer than May 1979. (So not a backport since is 8 months older than 3BSD and 17 months older than 4BSD.) From lm at bitmover.com Thu May 6 13:02:47 2010 From: lm at bitmover.com (Larry McVoy) Date: Wed, 5 May 2010 20:02:47 -0700 Subject: [TUHS] wrong HISTORY in 4.4BSD manual pages? In-Reply-To: References: Message-ID: <20100506030247.GI23511@bitmover.com> 2.9BSD, as I recall, was the last release for PDP-11's. Again, as I recall, it did a pile of work to take advantage of the larger address space (I think there was 64KI and 64KD). I don't recall every seeing a 3BSD release. On Wed, May 05, 2010 at 08:40:12PM -0500, Jeremy C. Reed wrote: > > Or is this simply because "2BSD" is not a operating system release per > > se, so "3.0BSD" is correct? > > > > But this makes me wonder if my 2BSD versions are newer than first > > 2BSD, so really 3BSD is correct for some of this. > > I was asked off list what version of 2BSD as it may have been the 4.x > stuff backported. So it is the 2bsd.tar.gz from the minnie.tuhs.org > archive (which is near identical to the spencer_2bsd.tar.gz). Its TAPE > file says: May 10, 1979. The READ_ME says: Thu Apr 19 23:25:43 PST 1979. > And don't see any date stamps included in any files newer than May 1979. > (So not a backport since is 8 months older than 3BSD and 17 months older > than 4BSD.) > _______________________________________________ > TUHS mailing list > TUHS at minnie.tuhs.org > https://minnie.tuhs.org/mailman/listinfo/tuhs -- --- Larry McVoy lm at bitmover.com http://www.bitkeeper.com From cowan at ccil.org Thu May 6 15:56:59 2010 From: cowan at ccil.org (John Cowan) Date: Thu, 6 May 2010 01:56:59 -0400 Subject: [TUHS] wrong HISTORY in 4.4BSD manual pages? In-Reply-To: <20100506030247.GI23511@bitmover.com> References: <20100506030247.GI23511@bitmover.com> Message-ID: <20100506055659.GA1342@mercury.ccil.org> Larry McVoy scripsit: > 2.9BSD, as I recall, was the last release for PDP-11's. 2.9BSD was the first 2.x release to have a kernel; it contained a backport of 4.1BSD. Earlier 2.x releases, like 1BSD, were just overlays on AT&T Unix. The current PDP-11 release, however, is 2.11BSD. It's still being worked on: 2.11.447 was released on New Year's Day, 2009. > Again, as I recall, it did a pile of work to take advantage of the > larger address space (I think there was 64KI and 64KD). Correct. That, plus a lot of in-memory overlays, was why the backport was possible at all. > I don't recall every seeing a 3BSD release. 3BSD was released for the VAX at the end of 1979, containing the first Berkeley 32-bit kernel and ports of 2.xBSD userland. It was superseded by 4BSD (later called 4.0BSD) in October 1980. The term 5BSD was avoided by 4.1BSD and later releases to avoid confusion with System V. -- On the Semantic Web, it's too hard to prove John Cowan cowan at ccil.org you're not a dog. --Bill de hOra http://www.ccil.org/~cowan From milov at uwlax.edu Thu May 6 23:07:49 2010 From: milov at uwlax.edu (Milo Velimirovic) Date: Thu, 6 May 2010 08:07:49 -0500 Subject: [TUHS] wrong HISTORY in 4.4BSD manual pages? In-Reply-To: <20100506030247.GI23511@bitmover.com> References: <20100506030247.GI23511@bitmover.com> Message-ID: 2.11BSD is available from the Archives in various forms. (From memory) the 11/44 and up, with the exception of the 11/60, had split I/D address spaces of 64KBytes each. [mutters to self... really must resurrect the 11/44 or 11/84.] On May 5, 2010, at 10:02 PM, Larry McVoy wrote: > 2.9BSD, as I recall, was the last release for PDP-11's. Again, as I > recall, it did a pile of work to take advantage of the larger address > space (I think there was 64KI and 64KD). > > I don't recall every seeing a 3BSD release. > > On Wed, May 05, 2010 at 08:40:12PM -0500, Jeremy C. Reed wrote: >>> Or is this simply because "2BSD" is not a operating system release >>> per >>> se, so "3.0BSD" is correct? >>> >>> But this makes me wonder if my 2BSD versions are newer than first >>> 2BSD, so really 3BSD is correct for some of this. >> >> I was asked off list what version of 2BSD as it may have been the 4.x >> stuff backported. So it is the 2bsd.tar.gz from the minnie.tuhs.org >> archive (which is near identical to the spencer_2bsd.tar.gz). Its >> TAPE >> file says: May 10, 1979. The READ_ME says: Thu Apr 19 23:25:43 PST >> 1979. >> And don't see any date stamps included in any files newer than May >> 1979. >> (So not a backport since is 8 months older than 3BSD and 17 months >> older >> than 4BSD.) >> _______________________________________________ >> TUHS mailing list >> TUHS at minnie.tuhs.org >> https://minnie.tuhs.org/mailman/listinfo/tuhs > > -- > --- > Larry McVoy lm at bitmover.com http://www.bitkeeper.com > _______________________________________________ > TUHS mailing list > TUHS at minnie.tuhs.org > https://minnie.tuhs.org/mailman/listinfo/tuhs > -- Milo Velimirović Unix Network Administrator - ITS Network Services 608.785.6618 Office - 608.386.2817 Cell University of Wisconsin - La Crosse La Crosse, Wisconsin 54601 USA 43 48 48 N 91 13 53 W From BHuntsman at mail2.cu-portland.edu Thu May 13 03:03:02 2010 From: BHuntsman at mail2.cu-portland.edu (Benjamin Huntsman) Date: Wed, 12 May 2010 10:03:02 -0700 Subject: [TUHS] [tuhs] Tenth Edition's Appendix A Message-ID: <621112A569DAE948AD25CCDCF1C07533299927@dolly.ntdom.cupdx> By some chance, does anyone have Appendix A from the man pages/docs from Tenth Edition? Thanks!!