12 Apr
2017
12 Apr
'17
11:18 p.m.
Hello all...
I have a BSD 4.2 distribution that I imaged a few years back. It differs
from the one in Warren's archives in subtle but significant ways. It
appears to be a few months earlier than the one at:
http://www.tuhs.org/Archive/Distributions/UCB/4.2BSD/
Case in point, from srcsys.tar:
diff -r ./GENERIC/vers.c
/home/krewat/archive/unix/tuhs/4BSD/Distributions/4.2BSD/bin/srcsys/GENERIC/vers.c
1c1
< char version[] = "4.2 BSD UNIX #8: Sun Oct 2 12:03:09 PDT 1983\n";
---
char version[] = "4.2 BSD UNIX #9: Wed Nov 2
16:00:29 PST 1983\n";
or from src.tar:
diff -r ./bin/mail.c
/home/krewat/archive/unix/tuhs/4BSD/Distributions/4.2BSD/bin/src/bin/mail.c
2c2
< static char sccsid[] = "(a)(#)mail.c 4.18 (Berkeley) 9/9/83";
---
static char sccsid[] = "(a)(#)mail.c 4.21
(Berkeley) 11/1/83";
449,454d448
< if (strcmp(my_name, "root") &&
< strcmp(my_name, "daemon") &&
< strcmp(my_name, "network")) {
< usage();
< done();
< }
(was doing a diff -r of the entire tree)
stand (first file on tape) is the same down to the checksum. But the
checksum for every other file (tape files, not individual files) on the
tape is different.
Also weird is that on my distribution, these exist:
< -rwxrwxr-x 0/10 0 Jul 26 16:37 1983 ./ucb/lisp/doc/
< -r--r--r-- 0/10 3522 Jul 26 03:08 1983 ./ucb/lisp/doc/lmacs
< -r--r--r-- 0/10 5196 Sep 25 21:01 1983 ./ucb/lisp/doc/ch0.n
< -r--r--r-- 0/10 13713 Sep 25 21:01 1983 ./ucb/lisp/doc/ch1.n
< -r--r--r-- 0/10 42144 Sep 25 21:01 1983 ./ucb/lisp/doc/ch2.n
< -r--r--r-- 0/10 9278 Jul 26 03:09 1983 ./ucb/lisp/doc/ch3.n
< -r--r--r-- 0/10 34324 Sep 25 21:01 1983 ./ucb/lisp/doc/ch4.n
< -r--r--r-- 0/10 18913 Jul 26 03:10 1983 ./ucb/lisp/doc/ch5.n
< -r--r--r-- 0/10 21543 Sep 25 21:01 1983 ./ucb/lisp/doc/ch6.n
< -r--r--r-- 0/10 162 Jul 26 03:10 1983 ./ucb/lisp/doc/ch61.n
< -r--r--r-- 0/10 19093 Jul 26 03:10 1983 ./ucb/lisp/doc/ch7.n
< -r--r--r-- 0/10 26713 Sep 25 21:01 1983 ./ucb/lisp/doc/ch8.n
< -r--r--r-- 0/10 12914 Jul 26 03:11 1983 ./ucb/lisp/doc/ch9.n
< -r--r--r-- 0/10 8107 Jul 26 03:11 1983 ./ucb/lisp/doc/ch10.n
< -r--r--r-- 0/10 8694 Sep 25 21:01 1983 ./ucb/lisp/doc/ch11.n
< -r--r--r-- 0/10 22170 Jul 26 03:12 1983 ./ucb/lisp/doc/ch12.n
< -r--r--r-- 0/10 11453 Jul 26 03:12 1983 ./ucb/lisp/doc/ch13.n
< -r--r--r-- 0/10 6678 Jul 26 03:12 1983 ./ucb/lisp/doc/ch14.n
< -r--r--r-- 0/10 6368 Jul 26 03:12 1983 ./ucb/lisp/doc/ch15.n
< -r--r--r-- 0/10 36465 Jul 26 03:13 1983 ./ucb/lisp/doc/ch16.n
< -r--r--r-- 0/10 6301 Jul 26 03:13 1983 ./ucb/lisp/doc/chb.n
< -r--r--r-- 0/10 4364 Jul 26 03:13 1983 ./ucb/lisp/doc/chc.n
< -r--r--r-- 0/10 5700 Jul 26 03:13 1983 ./ucb/lisp/doc/Makefile
< -r--r--r-- 0/10 12 Jul 26 03:13 1983 ./ucb/lisp/doc/indexsed
< -r--r--r-- 0/10 98 Jul 26 03:13 1983 ./ucb/lisp/doc/mantags
< -r--r--r-- 0/10 166 Jul 26 03:13 1983 ./ucb/lisp/doc/extrnames.awk
< -r--r--r-- 0/10 94 Jul 26 03:13 1983 ./ucb/lisp/doc/fixmks.sed
< -r--r--r-- 0/10 57142 Jul 26 03:14 1983 ./ucb/lisp/doc/franz.n
But in the TUHS version, ./usb/lisp/doc is linked to /usr/doc/lisp -
which doesn't seem to exist.
These are just examples of what I see and not the only differences.
13 Apr
13 Apr
12:37 a.m.
On Wed, 12 Apr 2017, Arthur Krewat wrote:
diff -r ./GENERIC/vers.c
/home/krewat/archive/unix/tuhs/4BSD/Distributions/4.2BSD/bin/srcsys/GENERIC/vers.c
1c1
< char version[] = "4.2 BSD UNIX #8: Sun Oct 2 12:03:09 PDT 1983\n";
---
By the way, the CSRG archives CD set's 4.1c.2 and 4.1c.1 version is:
static char SccsId[] = "(a)(#)mail.c 4.13 2/9/83";
char version[] = "4.2 BSD UNIX #9: Wed Nov 2
16:00:29 PST 1983\n";
or from src.tar:
diff -r ./bin/mail.c
/home/krewat/archive/unix/tuhs/4BSD/Distributions/4.2BSD/bin/src/bin/mail.c
2c2
< static char sccsid[] = "(a)(#)mail.c 4.18 (Berkeley) 9/9/83";
---
> static char sccsid[] = "(a)(#)mail.c 4.21 (Berkeley) 11/1/83"; But in the TUHS version, ./usb/lisp/doc is linked to
/usr/doc/lisp -
which doesn't seem to exist.
From the quick look of the copies I have including from
the CSRG
archives I purchased, I assume the TUHS version is the same as the CSRG
archives.
Will you make your copy of the files available?
1:10 a.m.
On 4/12/2017 2:37 PM, Jeremy C. Reed wrote:
From the quick look of the copies I have including from the CSRG
archives I purchased, I assume the TUHS version is the same as the CSRG
archives.
Will you make your copy of the files available?
Absolutely! As soon as I determine without a doubt that it contains
nothing proprietary.
8 p.m.
Question: In BSD-land, how do you "version" something like this? Two
different "snapshots" of BSD 4.2 - what file/version #/whatever is used
to definitely put a "version number" on this?
FYI, the distribution I have, vfont.tar is truncated, it was the last
file on the first tape, and at the time may have had a read error.
4,372,480 bytes of it were read, while the TUHS version is 5,888,000
However, the part that I have which is about 80% of the tar matches the
first 74% of the TUHS 4.2BSD distribution after extracting the files,
even though the checksum of that first 74% doesn't match.
So when I cobble this together, I'll just use the TUHS 4.2BSD vfont.tar
and make a note of it in the index.
On 4/12/2017 3:10 PM, Arthur Krewat wrote:
On 4/12/2017 2:37 PM, Jeremy C. Reed wrote:
From the quick look of the copies I have including from the CSRG
archives I purchased, I assume the TUHS version is the same as the CSRG
archives.
Will you make your copy of the files available?
Absolutely! As soon as I determine without a doubt that it contains
nothing proprietary.
8:10 p.m.
This should read:
However, the part that I have which is about 74% of the tar matches the
same 74% of the TUHS 4.2BSD distribution after extracting the files,
even though the checksum of that first 74% doesn't match.
On 4/13/2017 10:00 AM, Arthur Krewat wrote:
However, the part that I have which is about 80% of
the tar matches
the first 74% of the TUHS 4.2BSD distribution after extracting the
files, even though the checksum of that first 74% doesn't match.
2592
days inactive
2593
days old
4 comments
2 participants
participants (2)
-
Arthur Krewat -
Jeremy C. Reed