Subject: Snapshot report - June 24th tar_files
To: None <current-users@NetBSD.ORG>
From: Alistair G. Crooks <email@example.com>
Date: 06/29/1995 07:55:24
Just time for a quick report.
Thanks to firstname.lastname@example.org (Paul Kranenburg) for various items this week:
1. After last week's new rpcgen import, a few programs wouldn't compile.
Paul gave us the definitive way out of this problem:
> The recent upgrade of rpcgen(1) warrants some special actions to be taken
> to avoid errors during "make worlds".
> 1) Get -current sources from June 26 or thereabouts.
> 2) `make clean' in usr.bin/rpcgen, lib/librpcsvc, usr.sbin/rpc.*,
> 3) `make & make install' in usr.bin/rpcgen.
> 4) `make & make install' in lib/librpcsvc
> 5) `make' ...
> The reason for this is that rpcgen(1) now generates prototypes in the
> generated header files in librpcsvc.
To help those of you who don't sup, I've gathered up the necessary files
that have changed for the rpcgen thang, and stuffed them on
2. Paul's also implemented a merge switch (-m) on ldconfig, so that
we don't have to maintain a separate file containing the directories
already used by ldconfig. A great many thanks, Paul.
3. I've re-done the instructions to upgrade to 1.0a (aka -current),
and put them up for ftp. The instructions now have a date on them,
so that people can see what they're doing is out of date. I'm
surprised people are actually using these instructions, and not using
a binary snapshot to upgrade, but that's life for you - perhaps I
didn't realise so many people owned Adaptec 1540s. (:-))
4. These damn snapshot reports are taking me a long time to complete,
what with my work, the baby etc, and it's fairly depressing when I get
mail saying that I should distribute the reports to current-users on
a weekly basis - in addition to all the people who send messages to the
mailing list that are old, old problems. I concluded that either:
a) the warm weather in England this week has meant that I'm getting
very run-down (actually I am, but it's mainly through lack of sleep)
b) no-one finds these reports any good, so I'll stop doing producing them.
c) the mailing lists have suddenly had an influx of trigger-happy people.
So if you want these to continue, folks, give me some feedback.
5. There is a depressingly low number of packages on ftp.netbsd.org,
and I know I'm one of the ones to blame for that. In addition, I heard
that some of the X11 binaries for the sparc port weren't good - so can
people try and take an interest in this one, please, and support your
local package demi-god, John Kohl.
6. How fares the dosemu? Or the rolling in to the source tree of the
7. I recently came up with a new way of doing packages, this time by
means of bmake. If anyone's interested in these matters, please mail me,
and we can share thoughts and comments, etc.
8. There still seems to be a problem with "options ISO" in the kernel,
whereby netiso/tuba_usrreq.c still won't compile.
9. If you can't run X these days (hangs during mouse initialisation)
various people have recommended that you revert back to version 1.56
of com.c, which doesn't have this problem (1.59 does).
10. Frank van der Linden's added some a generic, machine-independent
ELF loader (thanks, Frank), and I think Christos has some Solaris
compatibility stuff on the way. Watch this space.
Other `notable' changes this week are:
1. Wrap __flt_rounds() declaration with __BEGIN_DECLS/__END_DECLS in
- Support rip version 2.
- Add ansi prototypes.
- Be less internet centric.
- Convert queues to use <sys/queue.h> so we don't need -lcompat
Initialise after malloc (fixes PR#1149).
Also, correct comparison when looking in `prefered' paths.
Add a `merge' option.
Use err() & warn().
Allow exporting regular files.
Generic mi ELF loader; delete Linux and Svr4 compat conf entries and
add generic ELF entry to exec_conf.c
And to sign off, as usual Herb hits it right on the button:
> I thought Carson Harding (email@example.com) said it best:
> Linux/FreeBSD users: "weeeee. I can use Unix on my PC!"
> NetBSD users: "hmm. I think I'll use my PC to run Unix".
(Just so's you don't think I'm a dork, Herb - is that proof enough?)
Alistair G. Crooks (firstname.lastname@example.org) +44 125 234 6377
Amdahl European HQ, Dogmersfield Park, Hartley Wintney, Hants RG27 8TE, UK.
[These are only my opinions, and certainly not those of Amdahl Corporation]