TUHS

tuhs@tuhs.org

6420 discussions

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

> From: John Cowan <cowan(a)mercury.ccil.org> > Well, provided the compiler is honest, contra [Ken]. A thought on this: The C compiler actually produces assembler, which can be (fairly easily) visually audited; yes, yes, I know about disassembly, but trust me, having done some recently (the RL bootstrap(s)), disassembled code is a lot harder to grok! So, really, to find the Thompson hack, we'd have to edit the binaries of the assembler! For real grins, we could write a program to convert .s format assembler to .mac syntax, run the results through Macro-11, and link it with the other linker... :-) Also, I found on what's going on here: > What was wierd was that in the new one, the routine csv is one word > shorter (and so is csv.o). So now I don't understand what made them the > same sizes!? The new ones should have been one word shorter!? Still > poking into this... The C compiler is linked with the -n flag, which produces pure code. What the linker documentation doesn't say (and I never realized this 'back in the day') is that when this option is used, it rounds up the size of the text segment to the nearest click (0100). So, in c2 (which is what I was looking at), the last instruction is at 015446, _etext is at 015450, but if you look at the executable header, it lists a text size of 015500 - i.e. 030 more bytes. And indeed there are 014 words of '0' in the executable file before the data starts. And if you link c2 _without_ the -n flag, it shows 015450 in the header as the text size. So that's why the two versions of all the C compiler phases were the same size (as files); it rounded up to the same place in both, hiding the one-word difference in text size. Noel

10 years, 8 months

short file names...

by scj＠yaccman.com

Erg! Discussion of file name length brought back some chilling memories of very early Unix, when file names were at most 6 characters long. Longer names were accepted but truncated at 6 characters. So you could edit ABCDE.c, store it, read it and edit it again, but the file system knew it as "ABCDE." So when you compiled the program, the compiler produced ABCDE.o, which overwrote the source code!

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

> To boot up the root pack (I don't think I did this at any point; I've > always mounted it as a subsidiary drive) > ... > The disk has a working RL bootstrap in block 0, it should boot OK. So I recently had to reboot my machine, and I took the opportunity to try this; it worked right off, booted 'unix' OK. (I didn't try any of the other Unixes in the root directory.) I had only that pack mounted on DL0, nothing else. > So, just for grins, because I was curious (after your question), I did > try recompiling the C compiler, to see what I'd get. > What I got were three files (c0, c1 and c2) which were _the exact same > size_ (down to the byte) as the binaries on the V6 Research distro, but > had a number of differences when compared with 'cmp -l'. Odd! > ... > I'll take a gander tomorrow and try and work it out. So, this turned out to be because I had replaced the csv.o in libc.a with a new one, because the standard V6 one doesn't work with long returns (which use R1 as well as R0, and the V6 cret bashed R1). I put the old csv.o back, and re-linked them, and this time c? all turned out identical. So the source in the distro really is the source for the running compiler on it. What was wierd was that in the new one, the routine csv is one word shorter (and so is csv.o). So now I don't understand what made them the same sizes!? The new ones should have been one word shorter!? Still poking into this... I understand most of the differences between the versions of c? with the old and new csv.o; in all the jumps to cret, the indirect word in the instruction was off by two (because cret was one word lower because csv was one word shorter); that, along with different contents in csv.o, created most of the differences. Why one word shorter? Because in csv: tst -(sp) / creates a temporary on top of the stack jmp (r0) had been replaced with: jsr pc,(r0) (saving one instruction, and making it one word smaller). Noel

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by Norman Wilson

The web page mentions files-11 which is ODS-1. Technically (it is all coming back to me now): FILES-11 is a family of file systems that started in RSX-11 (or perhaps before but that's the oldest instance I know). ODS-1 is really FILES-11 ODS-1; ODS is `on-disk [something].' RSX-11 used ODS-1. VMS used ODS-2. I'm not sure of all the differences offhand, but they were substantial enough that we ended up writing two different programs to fetch files to UNIX from RSX and VMS volumes (we had the latter to deal with too). Certainly the directory entries were different between the two: ODS-1 used RADIX-50-encoded file names with at most six characters plus an at-most-three- character `extension' (a term which newbies sometimes improperly import into UNIX as well); I forget the exact filename rules in VMS, but filenames certainly could be longer than six characters. I've found the early-1980s programs I remembered. There were two, getrsx.c and getvms.c; two programs, one for each file-system format. They are surely full of ancient sloppiness that won't compile or won't work right under a modern C compiler, and they make assumptions about byte order. I'll spend some time in the next few days going over them and see if I can quickly get something workable. A footnote as to their origin: in the world where we wrote these programs, we had not only multiple systems, but shared disk drives. The disk drives themselves were dual-ported; the controllers we used could connect to multiple hosts as well. Each system had its own dedicated disk drives, but the UNIX systems could also see the drives belonging to the RSX and VMS systems; hence the file-fetching programs, since this was well before the sort of networking we take for granted these days. On the other hand, we had several UNIX systems which spoke uucp to one another, and that was occasionally used for large file transfers. To speed that up, I taught uucico a new protocol, whereby control information still went over a serial line, but data blocks were transferred over a chunk of raw shared disk (with appropriate locks, of course). It was a simpler world back then, but that made it a lot more fun. Norman Wilson Toronto ON

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by Norman Wilson

This is also an RSX pack (I think), but when I tried to boot it, it said "THIS VOLUME DOES NOT CONTAIN A HARDWARE BOOTABLE SYSTEM", and since I don't know how to mount disks under RSX-11 I left it at that. There exists somewhere a UNIX program that reads an ODS-1 file-system image and produces directory listings and extracts named files. I know it exists because it was written by a friend (and probably hacked around a little by me) almost 35 years ago, when we both worked in a place that had some UNIX and some RSX-11. That means `somewhere' probably includes some place in my old files. I'll see if I can dig it out, upgrade it to work cleanly with modern C (it's just possible I have an ODS-1 file system image lying around somewher too), and post it either somewhere on the web or just to the list. Norman Wilson Toronto ON

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

> From: Larry McVoy <lm(a)bitmover.com> > have you gotten to a point where you can rebuild the world and install > your newly built stuff? Well, I haven't tried to do that (it's not something that I'm that interested in), but it _should_ be possible, since the 'vanilla' V6 distribution does include the source for pretty much everything (including the C compiler, assembler, loader, etc). (This does not include the stuff from the Shoppa disk, like the new C compiler, where I don't have the source. [The PWB distribution, which includes a C compiler from 1977, is probably pretty close. Looking into the PWB stuff is one of my next projects; we have 17 different versions of that stuff, and I'd like to see what the differences among them are, and maybe create a 'canonical' PWB.] Also, per the 'Improvements' page, I have source for the Standard I/O Library, but I'm using the binary library from the Shoppa disk, which may or may not correspond to that source.) Noel

10 years, 8 months

An amusing error message

by jnc＠mercury.lcs.mit.edu

Found in: http://minnie.tuhs.org/cgi-bin/utree.pl?file=1BSD/s6/man.c this minor gem: execl("/bin/ssp", "ssp", 0); execl("/usr/bin/ssp", "ssp", 0); printf("Can't find ssp!\n"); execl("/bin/cat", "cat", 0); printf("or cat - gott in himmel!\n"); exit(1); Not as good as "hodie natus" (Google it if you don't know of it - it's a classic), but mildly amusing. Noel

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

> From: Gregg Levine <gregg.drwho8(a)gmail.com> > By Shoppa disk do you mean, Shoppe disk The root pack (linked to from my page): http://www.tuhs.org/Archive/PDP-11/Distributions/other/Tim_Shoppa_v6/unix_v… It contains lots of V6 goodies - but not, alas, source for most of them. (E.g. I had to disassemble the RL bootstrap(s) from it.) > I've been trying to figure out how to attach them to the E11 one to > bring the whole thing up. The others are junk (see below). To boot up the root pack (I don't think I did this at any point; I've always mounted it as a subsidiary drive) you'd need to say: mount dl0: unix_v6.rl02 /RL02 and then: boot dl0: The disk has a working RL bootstrap in block 0, it should boot OK. Then you get to take your pick of which unix to boot! There are 7 to chose from: 77 -rwxrwxr-x 1 root 38598 Aug 22 1984 oldunix 77 -rwxrwxr-x 1 root 38504 Jul 19 1984 oldunix.25.7 77 -rwxrwxr-x 1 root 38568 Feb 20 1985 unix 74 -rwxrwxr-x 1 4 36956 Mar 9 1983 unix.jones 69 -rwxrwxr-x 1 root 34408 Aug 16 1983 unix.mlab 76 -rwxrwxr-x 1 4 38316 Sep 3 1982 unix.rxrl 68 -rwxrwxr-x 1 root 33958 Jun 6 1983 unix.tmp and I have no idea how they all differ - or what each one expects to use for a root device and swap device. Looking at 'unix', it gives both rootdev and swapdev as '01000' (which is probably the RL, I'm too lazy to grovel around in bdevsw and make sure). The super block reports 19000 as the size of the file system, and sure enough, swplo is reported as 045070 (19000), and nswap as 02710 (1480). So it's probably set up to run and swap on RL/0. > and any of the set of disks that are in the distributions portion of > the site with his name on them? All of the other disks in the V6 folder with his name on it: http://www.tuhs.org/Archive/PDP-11/Distributions/other/Tim_Shoppa_v6 are junk; here's a brief rundown on each one: copy_num1_user.rl02: user0_backup.rl02: user_backup2.rl02: All very similar to the 'unlabeled' disk (below) - lots of random user files for biology stuff, little if anything of any use/interest. There is a file which isn't listed in the README: unlabeled.rl02: It looks like another /user disk, full of random biology stuff; nothing interesting except a copy of a nice-looking Kermit written in C (by someone at Columbia, IIRC). junk: An RL01 pack (the others are all RL02's); it has a boot block with PDP-11 code in it; I mounted it on a simulator and booted it, and it says it's an RSX-11M V3.2 disk. user01.rl02: This is also an RSX pack (I think), but when I tried to boot it, it said "THIS VOLUME DOES NOT CONTAIN A HARDWARE BOOTABLE SYSTEM", and since I don't know how to mount disks under RSX-11 I left it at that. scratch_disk_1123.rl02: This does indeed seem to be something that was used for disk diagnostics: the boot block contains gubble, including (in the first RL11 block) lots of words with all ones, and a lot of 52525's. Noel

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

>> From: Larry McVoy <lm(a)bitmover.com> >> have you gotten to a point where you can rebuild the world and install >> your newly built stuff? > Well, I haven't tried to do that (it's not something that I'm that > interested in), but it _should_ be possible, since the 'vanilla' V6 > distribution does include the source for pretty much everything > (including the C compiler, assembler, loader, etc). So, just for grins, because I was curious (after your question), I did try recompiling the C compiler, to see what I'd get. What I got were three files (c0, c1 and c2) which were _the exact same size_ (down to the byte) as the binaries on the V6 Research distro, but had a number of differences when compared with 'cmp -l'. Odd! I don't know what the differences result from (and it's too late now to dig into why, I'm fading). I'll take a gander tomorrow and try and work it out. Too bad the binaries in the Research distro have had their symbol tables stripped! That would have made it much easier... My guess is something like 'libraries in different order, so two library routines are swapped around in the linked binary', or something like that (given that the size is an exact match). But I'll need to dig a bit... Noel

10 years, 8 months

Re: [TUHS] Work I've done with a PDP-11 simulator

by jnc＠mercury.lcs.mit.edu

> OK, I have whipped up some material on how to bring V6 up under > Ersatz-11. See here: > http://ana-3.lcs.mit.edu/~jnc/tech/unix/V6Unix.html New location (although the old one still works): http://www.chiappa.net/~jnc/tech/V6Unix.html So it turns out there was a bogon on the old version of this page: it claimed the Windoze version of 'tar' needed Unix system mods (which is incorrect, it was the version for V6 which needed the mods - cut and paste error). Fixed now (it's useful to have a Windoze 'tar' because some of the old TAR files in the archive can't be read by modern tools), and a certain amount of new material added to the page overall. > There is more content/pages coming: the start of an 'advanced things > you can do to improve your V6 Unix' page I've upgraded that somewhat to a complete page (although I'll probably add more at some point in the future, e.g. I have a version of 'ps' which shows sleep channels symbolically, text slots as ordinals, etc but I need to tweak it a bit before I bring it out). The page is here: http://www.chiappa.net/~jnc/tech/ImprovingV6.html now, although the main page links to it too. And I have another page coming, e.g. a 'what to look out for when you're porting stuff to and from V6' guide. Etc, etc. Noel

10 years, 8 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

TUHS