[255] in NetBSD-Development
Lola & project/netbsd changes
daemon@ATHENA.MIT.EDU (jhawk@MIT.EDU)
Thu Dec 8 04:29:43 1994
From: jhawk@MIT.EDU
Date: Thu, 8 Dec 1994 04:28:46 -0500
To: netbsd-dev@MIT.EDU
'Couple o' things...
Mycroft seems to be hanging around the office occasionally. This is nice
and useful, so treat 'im nice :-)
The lola kernel is back to the sb kernel, mycroft having found that
adding delays makes vat work (this we found by running it under ktrace...).
The project/netbsd tree now has the -current (as of a few days ago)
wd driver, which, among other things, provides logical bus addressing (lba)
on lola-granola, so if you make a kernel from there (like the current
lola kernel), that'll be there. Ditto for the "improved" sb.c.
src/gnu/usr.bin/gdb/gdb/arch/i386/i386b-nat.c is the -current version,
and the gdb has been rebuilt. This means that (supposedly) kgdb should
work.
I spent somet time tonight trying to make lola work correctly with test
accounts and other dotfiles that aren't people who use this machine
regularly's. I created /usr/athena/lib/init/RCS, and threw in Yoav's changes
to cshrc. I also modified manpath and athena_manpath to correctly
handle X things, removing /srvd/patch and some other nonapplicable stuff,
and added /usr/local and the sbins. Check the RCS if you'd like to know.
In doing this, In order to make dash work, I symlinked a few files in
/usr/athena/include/X11/bitmaps/ to files on in /afs/athena/system/
pmax_ul4/srvd; I'm sure this is the wrong way to go about fixing it,
but it's not apparent to me what I _ought_ to do...
I observed that in some situations (when /etc/master.passwd.orig exists),
if a user who is not in the password file attempts to login, they fail the
first time (w/ a pwd_mkdb error complaining about the existance of the
above file), and then login sucessfully the 2nd time. rm-ing the .orig file
seemed to fix the problem, but I'm not too sure...
I finally got tired of the WARNING: running as root with dot in path''
nonsense, and patched src/bin/sh/var.c to NOT include :/bin:/usr/bin but
rather /bin:/usr/bin. I hope this doesn't break anything.
That's about it for changes; now a question:
Why is it that /bin/athena is a symlink to /usr/athena/bin? Under the
standard Athena paradigm, /bin/athena is supposed to include everything
you need to get to /usr/athena/bin (i.e. attach, kinit, aklog, etc.).
Unfortunately, this is not the case. As a result, /usr/athena is merely
a symlink to the syspacks instead of an "attach" symlink. Can someone
explain to me the rationale for breaking the paradigm?
There's very little space in lola-granola's root. This is annoying;
if I'm feeling bold I may fix this.
--jhawk
