On Tue, 30 Aug 2011, Lyndon Nerenberg wrote:
>>From the start if the SCCS history (April 9,
1980 ) through May 17,
>>machdep.c identified the system as version 3.1. Delta 3.6 (May 18)
>>changed the version string to be the SCCS delta of machdep.c, thus
>>the version number jumped from 3.1 to 3.6. The version appears to
>>have tracked the machdep.c delta until Nov 10 when it was hardwired
>>to '4.1'. (I say appears because I didn't take the time to examine
>>all 27 deltas between 3.6 and 4.1.)
Yes. I saw the same, such as 3.6 to 4.1 (nothing between):
-char version[] = "VM/UNIX (Berkeley Version %I%) %H% \n";
+char version[] = "VM/UNIX (Berkeley Version %I%) %G% \n";
3.6 char version[] = "VM/UNIX (Berkeley Version 3.34) 08/31/11 \n";
(That is today's date per %H%.)
4.1 char version[] = "VM/UNIX (Berkeley Version 4.1) 11/10/80 \n";
D 4.1 80/11/10 15:25:31 bill 42 35 00033/00011/00386
D 3.34 80/10/22 09:34:05 bill 35 34 00001/00001/00396
After a cursory search I can't find any SCCS log
references to a 4.0
release.
But search for "stamp for 4bsd" for example. This happened from:
D 4.1 80/11/09 16:29:06 bill 5 4 00000/00000/00094
to
D 4.1 80/11/09 17:02:39 bill 2 1 00000/00000/00016
The previous sccs timestamps are from:
D 3.2 80/06/07 02:45:12 bill 2 1 00001/00001/00044
to
D 3.29 80/11/09 16:07:34 bill 29 28 00015/00086/00749
This seems to imply that the concept of 4.0 never existed in the source
tree. Then again, as far as I see, the SCCS didn't support or use .0 as
a revision.
This still doesn't explain why all the source files other than libpc are
same from 4.0bsd and 4.1bsd in the archives.
(I had noticed this same problem a year ago at least.)
If someone has their own 4.0BSD archive please check if different.