from 386bsd0.1 to FreeBSD or NetBSD 0.9
M.C. Wong
I am currently using 386bsd 0.1 + pk 0.2.4! My question is how easy
it is to upgrade from :
i) 386bsd to FreeBSD
ii) 386bsd to NetBSD
My question is do I have to reinstall everything to go from existing system
to either one of them ? Also , Since FreeBSD incorporate all of the patchkits,
does it mean, it will cut down the overall installed size ? (since we no longer
need to keep a separate dir for the patchkit)!
I read some discussion about FreeBSD and NetBSD are quite similar, and
before I go ahead ftp megabytes of stuff, I will like people that has
done the above transition to give me some experience.
Basically, I will like to get all of the features I am having in the old
system, ie sio supports, XFree 1-3, syscons-0.2, slip support etc etc ...!
Apart from ease of transition to a newer platform, I wonder if anyone can
summarize some improvements over 386bsd, eg performance wise, less bugs etc ?
Last question is, if 386bsd 0.2 is ever going to come out, is there an
easy upgrade from 386bsd 0.1 to it? How about from FreeBSD to 386bsd 0.2 ?
I am waiting for a big change over but hope to stick with 386bsd 0.2 (when
it comes out). And is has been months that I wait for the announcement for
its birth. I wonder if anyone knows of any development of it, and when it
will be released by estimation ?
Many thanks in advance!
Jun-Lin Chen
: Hi,
: I am currently using 386bsd 0.1 + pk 0.2.4! My question is how easy
: Apart from ease of transition to a newer platform, I wonder if anyone can
: summarize some improvements over 386bsd, eg performance wise, less bugs etc ?
: Last question is, if 386bsd 0.2 is ever going to come out, is there an
: easy upgrade from 386bsd 0.1 to it? How about from FreeBSD to 386bsd 0.2 ?
: I am waiting for a big change over but hope to stick with 386bsd 0.2 (when
: it comes out). And is has been months that I wait for the announcement for
: its birth. I wonder if anyone knows of any development of it, and when it
: will be released by estimation ?
I would like to install a BSD UNIX in PC for our school Campus Information
System, if anyone has experience for NetBSD or FreeBSD, please give me
some comments about them.
Thank you in advance.
J.L. Chen
Sep. 9, 93
===============================================================================
Jun-Lin Chen National Center for
Tel:886-2-737-6759 High Performance Computing (Taiwan)
E-Mail:nch...@ccms.ntu.edu.tw Research and Development Group
=======================================================================
Robert Baron
groups that FreeBSD and NetBSD are going to merge together soon.
I'd like to hear that confirmed here.
I find it particularly sad that the community has chosen to fragment
it self so. Doing a "diff -r" of both /usr/src's I find 1.58Meg of
differences. Doing a slightly mor careful look at /usr/src/sys, I
find 600K differences. Now a lot of the differences are just RCS
headers being different. But there are substantive code differences
as well. And there are even trivial annoying differences -- like
wether Debugger() takes an argument or not.
I'd certainly like a clear mandate as to which code base to work with.
For as it turns out, I have an EISA machine with a DPT controller
(scsi emulating wd1003); they do a slightly quirky emulation --
they typically raise "done" before they set other bits in a register.
So this was working fine with 0.1+0.2.3. I cut over to FreeBSD there
and I got as far as trashing my disk, but I can not get FreeBSD to
successfully come up. On the other hand, NetBsd does not even see
the wd drives, so I lose there too. And sadly, I now have nothing.
Well, which kernel do I hack to fix this?
Andrew Moore
>Hi,
> I am currently using 386bsd 0.1 + pk 0.2.4! My question is how easy
>it is to upgrade from :
>
> i) 386bsd to FreeBSD
> ii) 386bsd to NetBSD
There is an upgrade script from any 386BSD version to FreeBSD.
> My question is do I have to reinstall everything to go from existing system
>to either one of them ? Also , Since FreeBSD incorporate all of the patchkits,
>does it mean, it will cut down the overall installed size ? (since we no longer
>need to keep a separate dir for the patchkit)!
I don't have exact figures at the moment.
The overall installed size is quite large still, of course.
> I read some discussion about FreeBSD and NetBSD are quite similar, and
>before I go ahead ftp megabytes of stuff, I will like people that has
>done the above transition to give me some experience.
Whether you install FreeBSD or NetBSD, please wait for the new install
disks and/or the upgrade script. Both are being readied for release.
> Basically, I will like to get all of the features I am having in the old
>system, ie sio supports, XFree 1-3, syscons-0.2, slip support etc etc ...!
I believe these are supported in both NetBSD and FreeBSD. sio you may
have to add yourself to NetBSD.
> Apart from ease of transition to a newer platform, I wonder if anyone can
>summarize some improvements over 386bsd, eg performance wise, less bugs etc ?
Chris has done a good job of maintaining a change log for NetBSD.
Nothing comparable for FreeBSD, unfortunately. The emphasis in both
systems has been usability.
> Last question is, if 386bsd 0.2 is ever going to come out, is there an
>easy upgrade from 386bsd 0.1 to it? How about from FreeBSD to 386bsd 0.2 ?
>
> I am waiting for a big change over but hope to stick with 386bsd 0.2 (when
>it comes out). And is has been months that I wait for the announcement for
>its birth. I wonder if anyone knows of any development of it, and when it
>will be released by estimation ?
The inside information is that it was to be released last February :)
Bill reminds us that BSD stands for "Berkeley Software's Delayed"
-AM
Jordan K. Hubbard
>groups that FreeBSD and NetBSD are going to merge together soon.
>I'd like to hear that confirmed here.
And the "official" answer is: We don't know.
What I can say is that we've discussed it (to death), and we've made
up a charter detailing how such a merger should proceed. Now we
simply need to decide if and when it should proceed - there are still
some sticking points to be ironed out.
I can say that for the short-term there will be a FreeBSD 1.0,
available for ftp, or on cdrom/tape/etc, and that everyone is
encouraged to run it as it is a massive improvement over 386BSD 0.1
(also, an upgrade script for 386bsd->freebsd has just been released
and is in the tools area on freebsd.cdrom.com).
Those running 1.0 will not be left in the lurch, if and when we ever
do merge with NetBSD, since it will be done as a transition from the
existing FreeBSD 1.0 (all of us are running it, and we don't want to
nuke ourselves either).
All in all, I wouldn't recommend waiting at all, since this isn't
necessarily going to save you anything either.
>So this was working fine with 0.1+0.2.3. I cut over to FreeBSD there
>and I got as far as trashing my disk, but I can not get FreeBSD to
>successfully come up. On the other hand, NetBsd does not even see
I'm very sorry to hear that Robert has been having these troubles,
even more that they caused data loss. All I can say in regards to
this is "FOLKS, YOU GOTTA LET US KNOW ABOUT SUCH THINGS!"
Specifically in regar to the wd driver, since we have about 3
different mutations floating around at the moment that we've been
sending people who have problems. By letting us interactively solve
your problem, you're also aiding us in the effort to find the One True
WD driver that works for everyone. This will also benefit NetBSD,
since they can share in the data we collect, if not the results.
>I'd certainly like a clear mandate as to which code base to work with.
>
>Well, which kernel do I hack to fix this?
This, like many other things, must remain a matter of taste. I cannot
give you the mandate you wish, I'm afraid to say. I can certainly say
that we're interested in any work you may do.
Jordan
--
Jordan Hubbard j...@violet.berkeley.edu, j...@al.org, j...@whisker.lotus.ie
Adam David
>So this was working fine with 0.1+0.2.3. I cut over to FreeBSD there
>and I got as far as trashing my disk, but I can not get FreeBSD to
>successfully come up. On the other hand, NetBsd does not even see
>the wd drives, so I lose there too. And sadly, I now have nothing.
I would suggest for now, install FreeBSD with a pk0.2.[34] kernel,
the old kernel should boot correctly in the new environment. Then rename
any driver sources that don't work in the FreeBSD tree, and put pk0.2.[34]
sources in their place. The scsi driver is everything in /sys/scsi and
whichever low-level scsi driver that you are using in /sys/i386/isa,
The wd driver is probably only /sys/i386/isa/wd.c
Once you have compiled the new kernel you will be using a FreeBSD system
except for whatever old driver you had to use. Then you are pretty much
sorted out.
I am running a FreeBSD system with pk0.2.4 scsi drivers, because my disks
are not recognised by the new scsi code. It's nonetheless FreeBSD.
Nate Williams
>My best advice is to just wait. I've heard on some other news
>groups that FreeBSD and NetBSD are going to merge together soon.
>I'd like to hear that confirmed here.
You heard Jordan's words already on what is going/not-going to happen.
>Well, which kernel do I hack to fix this?
You hack whichever kernel you feel most comfortable in, and if you're code
is good then both the NetBSD and FreeBSD folks will use it. I think the
diversity has helped in many ways, as it has caused both groups to see things
that the other group has been lacking in.
Opinion follows:
For one example of this.
I believe the FreeBSD folks do a better job of packaging the release into
a more palatable state for users to use, while the NetBSD spend more time
trying to make the system into a more palatable system for developer's to
use. It's just a difference in priorities.
This isn't to say that the FreeBSD folks don't care what the system looks
like, nor does it imply that the NetBSD folks don't package their system
well, it's just that the amount of time spent in those areas are different
for the different groups.
Alot of folks may say, "Hey, you should get together, since you do
different things, and together you could be a better team". That may be
true, but because of those strengths, I (a lame kernel hacker by all
standards) would never be allowed to touch the kernel again. Right now,
I like being a little fish in a small pond. But, if I were to join
NetBSD today, I (have been told) that I am to keep my grubby little
fingers out of the kernel. So, I choose to stay where I only get yelled
at for doing stupid kernel changes, (but at least I can make them. :-)
So, more than just talents are at stake. I *like* playing with *BSD. I
haven't gotten a dime since I started doing this when 0.1 was released,
and I don't expect to get anything. So, when it starts becoming less
fun, it's not worth it to me.
A merge may happen, and then again, it may not. But, I don't think
anyone is *losing* out except for those folks who think there has to be
one unified release. If you want one unified release, buy BSDI. There
release is a lot less fragmented and more stable than any of the free
versions. In the meantime, I'm gonna have some fun, and learn on the
way. If you like what I do, I'm more than welcome to share it with you,
and even ask for your help, but you don't have to accept my work, or
even like it for me to feel fulfilled. (But I *do* appreciate when
people appreciate me and my work. :-)
Off my soap-box and off to bed,
Nate
--
na...@bsd.coe.montana.edu | In the middle of it ........ again.
na...@cs.montana.edu | Running/supporting one of many freely available
work #: (406) 994-4836 | Operating Systems for [34]86 machines.
home #: (406) 586-0579 | (based on Net/2, name changes all the time :-)
--
na...@bsd.coe.montana.edu | In the middle of it ........ again.
na...@cs.montana.edu | Running/supporting one of many freely available
work #: (406) 994-4836 | Operating Systems for [34]86 machines.
home #: (406) 586-0579 | (based on Net/2, name changes all the time :-)
Andrew Moore
>>My best advice is to just wait. I've heard on some other news
>>groups that FreeBSD and NetBSD are going to merge together soon.
>>I'd like to hear that confirmed here.
>
>And the "official" answer is: We don't know.
It won't be for at least six months. And between now and then, the
release of 0.2 is sure to complicate the matter. It is best to forget
about a merge.
In the mean time, I think it is safe to say that FreeBSD is putting
more effort into i386-specific support.
-AM
Theo de Raadt
In the mean time, I think it is safe to say that FreeBSD is putting
more effort into i386-specific support.
In the meantime, I think it's completely stupid to say anything of the
kind.
I hope you've fixed the swtch() bug in locore.s, BTW.
--
This space not left unintentionally unblank. der...@fsa.ca
Nate Williams
Theo de Raadt <der...@fsa.ca> wrote:
>In article <almCD6...@netcom.com> a...@netcom.com (Andrew Moore) writes:
> In the mean time, I think it is safe to say that FreeBSD is putting
> more effort into i386-specific support.
>
>In the meantime, I think it's completely stupid to say anything of the
>kind.
>
>I hope you've fixed the swtch() bug in locore.s, BTW.
In general, you can ignore anything Theo says after you make any
comparisons to NetBSD, because NetBSD is perfect, and anything that is
compared to NetBSD is a piece of s*it. (And you will notice little
digs anytime you compare anything to NetBSD)
But, he is a pretty good hacker, so most of the time it makes up for it.
Nate
(Hi Theo, can't wait for the response to this one ;-)
Jordan K. Hubbard
>kind.
>
>I hope you've fixed the swtch() bug in locore.s, BTW.
I would ask that everyone join me in ignoring any subsequent traffic on
this topic, we don't need it and I doubt that NetBSD needs it. I
assure you that both groups have more than their share of strengths
and weaknesses, and trying to make any assessment of "better or worse"
is bound to come down to being little more than personal opinion.
Articles like the above merely cast doubt on the credibility of the
poster.
This is the last and only thing I'm going to say publically on this
topic, and I hope that people in both camps find it within themselves
to show similar restraint.
Thanks.
Charles Hannum
In the mean time, I think it is safe to say that FreeBSD is putting
more effort into i386-specific support.
That comment is totally unfounded.
I am personally working on redoing all the autoconfig stuff in NetBSD,
since the current stuff is rather braindead, and FreeBSD's isn't any
better. The new code is based on Chris Torek's new config, and deals
with address conflicts and automatic determining of IRQs.
I've looked through the FreeBSD commitlog. Many of the kernel patches
were taken from NetBSD, and the few that weren't and were correct and
useful have already been incorporated into NetBSD. I cannot say the
same in reverse; there are many important changes that have not been
incorporated into FreeBSD.
I did *not* want to have this argument.
Charles Hannum
In article <26t5dt$8...@pdq.coe.montana.edu> na...@bsd.coe.montana.edu
(Nate Williams) writes:
In article <DERAADT.93...@newt.fsa.ca>, Theo de Raadt
<der...@fsa.ca> wrote:
In article <almCD6...@netcom.com> a...@netcom.com (Andrew
Moore) writes:
In the mean time, I think it is safe to say that FreeBSD is
putting more effort into i386-specific support.
In the meantime, I think it's completely stupid to say anything
of the kind.
In general, you can ignore anything Theo says after you make any
comparisons to NetBSD, [...]
In general, you can ignore anything any of the FreeBSD or NetBSD
people say after such a comparison, because they are all (including
myself) biased assholes who just want to push their own system.
Well, most of the time, anyway.
Actually, I agree with Theo. Saying that FreeBSD is putting any more
work into i386 support is ludicrous, and is only worsened by the fact
that where you have stepped you've introduced new bugs.
Have you integrated the changes for >640K kernels (without losing 640K
of memory)? No, but it's in NetBSD-current. Have you fixed any of
the spl stupidities (broken swtch(), broken drivers)? No, but it's in
NetBSD-current. Who fixed the broken boot blocks? Me. Who fixed the
GCC 2 bug that was causing kernels to hang? Me. Who is actually
working on the stupid probe bugs? Me. Who is distributing GCC in
such a way that it violates the GPL? You.
Should I stop now? Yes. I won't even mention your crusty ISOFS, PCFS
and other things.
because NetBSD is perfect, and anything that is compared to NetBSD
is a piece of s*it.
And I suppose that FreeBSD *is* perfect? We never claimed we didn't
have problems; Hell, we have a publicly available bug list for anyone
who wants to read it. Some of the bugs are pretty embarrassing, and
I'm pretty sure they (and a host of others that we have fixed) are all
in FreeBSD, too.
But, he is a pretty good hacker, so most of the time it makes up
for it.
Nothing `makes up' for rudeness; neither from Theo nor from me nor
from you. However, sometimes other things are more important.
Sean Eric Fagan
>Who is distributing GCC in
>such a way that it violates the GPL? You.
Last time I checked, sources for gcc were included in the source
distribution. That is all that is necessary.
Charles Hannum
In article <26p8ul$1...@pdq.coe.montana.edu> na...@bsd.coe.montana.edu
(Nate Williams) writes:
But, if I were to join NetBSD today, I (have been told) that I am
to keep my grubby little fingers out of the kernel.
This is a non-sequitur. What you're referring to is the strawman
proposal for merging, not joining the NetBSD group outright. If your
memory isn't too rusty, you'll recall that I disagreed with that part,
too. (In fact, it's the primary reason why I rejected the proposal.)
Cor
and one thing just keeps popping in my mind. Does all this arguing
do any of us any good? Ive seen this kinda arguments all the way
since the 64 flamed msx and amiga flamed atari..etc etc.
Wouldnt it be way more profitable for everyone to stick yer heads
together and work as a team? It seems both the netbsd and the freebsd
people have been doing great work. So..get together and work on it
together instead of bringing down the other side.
But..im just a low end 386bsd admin (yes..still using 386bsd and
i have no complaints really :) so I probably dont know all the
intimate details there :)
cor
--
| c...@hacktic.nl | Hack-Tic System Management | +31-20-6001480-3 (VMB) |
| -------------------------------------------------------------------------|
| ######### Signature Virus Running. Contamination Complete ######### |
+--------------------------------------------------------------------------+
Charles Hannum
In article <CD7w8...@kithrup.com> s...@kithrup.com (Sean Eric Fagan)
writes:
The GPL specifically defines `source' as `the preferred form for
editing'. FreeBSD includes only pre-generated insn-* files, not the
original machines descriptions.
I have told the FreeBSD people about this a couple of times. They
could have simply picked up the GCC 2 from NetBSD 0.9 (which I fixed
myself), but they have not done so.
Peter da Silva
> In the mean time, I think it is safe to say that FreeBSD is putting
> more effort into i386-specific support.
>In the meantime, I think it's completely stupid to say anything of the
>kind.
In the meantime, let's grow a bit and quit (a) making patriotic statements and
soundbites, and (b) jumping on minor offenses with two feet and a search
warrant.
Pax.
Jordan K. Hubbard
could have simply picked up the GCC 2 from NetBSD 0.9 (which I fixed
myself), but they have not done so.
Sigh. People, people, can we STOP this already? In my last article I
said that I wouldn't join in this ridiculous "comparison war" some of
our mutual members are having, and I've no intention of changing my
mind, but I'd like to at least request that while you hurl epithets at
the "other side", and call each other names, that you at least try to
avoid giving false impressions like the one above, especially on so
charged a subject as "Violating the GPL".
We're well aware of the conditions of the GPL, long before you or
anyone else deemed it necessary to point it out. We've undertaken to
make the *original* sources available to anyone that asks, from the
same sources (and the CD will have an unadulterated gcc-2.4.5.tar.gz
on it), thus upholding the GPL paragraph that says:
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have. You must make sure that they, too, receive or can get the
source code. And you must show them these terms so they know their
rights.
And the COPYING file is part of the distribution, so they are well
able to "know their rights" as well.
Perhaps NetBSD's gcc2 release is superior, I have not looked nor do I
dispute such a "fact" (how can I dispute what I have not even looked
at?), and perhaps even some day FreeBSD will be happy to take the gcc2
release, just as you have taken some of "our" GNU ports. Where's
the problem?
What I do know is that I have neither the time nor the inclination to
look into this now, as I have more pressing matters to attend to than
looking into NetBSD's gcc2 port. Just because I've not done so,
however, does not mean that FreeBSD's own release is somehow illegal.
I said I wouldn't get into the comparison war, but this doesn't stop
me from saying a few POSITIVE words, and I'd like to state, for the
record, that I think NetBSD is a great effort and that I have NOTHING
AT ALL AGAINST IT. A very few people working on NetBSD I would
probably cross the street to avoid (or struggle not to reach for my
.45 :), but this does not mean I have any technical grievances with
the "product" itself, which seems to be the whole topic of continuing
debate.
Let's just knock it off. Both teams are doing some good work, which
can only benefit the public (the public: remember the public? The
folks we're supposed to be doing this all for?), whether all people on
both sides admit it or not.
I've got an idea: Why not let the USERS decide how well we're doing,
rather than debating this endlessly between ourselves? If Lotus
and Microsoft were actually allowed to write the other company's
"BYTE" and "PC Magazine" reviews, I'm sure this is very much what
it would look like! Fortunately, such reviews are generally left
to people with the outsider's viewpoint.
Amancio Hasty Jr
>In article <DERAADT.93...@newt.fsa.ca>, Theo de Raadt <der...@fsa.ca> wrote:
>>In article <almCD6...@netcom.com> a...@netcom.com (Andrew Moore) writes:
>> In the mean time, I think it is safe to say that FreeBSD is putting
>> more effort into i386-specific support.
>
>>In the meantime, I think it's completely stupid to say anything of the
>>kind.
>
Lets see:
386bsd-0.1
386bsd-0.1 + random patches
386bsd-0.1 + patchkit
FreeBSD + complete patched distribution + development
NetBSD + complete distribution + development + incompatible binaries
386bsd-0.2 + the great unknown
This situation is sad. It appears that the only way to consolidate
our efforts is for the 386bsd community to choose an OS.
Look, I am not saying that NetBSD or FreeBSD are inferior or that
one group is less talented than any other, rather it would
be to our benefit to provide a single alternate solution
to 386bsd-0.1 or to 386bsd-0.2 whenever it comes out.
Peace and I am going back to enjoy my system :-)
Amancio
--
This message brought to you by the letters X and S and the number 3
Amancio Hasty |
Home: (415) 495-3046 | ftp-site depository of all my work:
e-mail ha...@netcom.com | sunvis.rtpnc.epa.gov:/pub/386bsd/incoming
Andrew Moore
>people say after such a comparison, because they are all (including
>myself) biased assholes who just want to push their own system.
Stating the intent of FreeBSD developers, as I understand them, is not
intended to push FreeBSD, nor to detract from NetBSD. You call it
misinformation, but these are simply expressed interests.
By i386-specific support, I mean, for instance, integrating the Future
Domain driver, dealing with DOS partitions during install, supporting
old IDE drives and so on. This is thankless work, but all the kernel
mods in the world aren't going to help those who need this more basic
stuff.
If NetBSD is in fact addressing these and like issues, then I
gladly retract my claim and fully endorse NetBSD independent of FreeBSD.
-AM
Theo de Raadt
> der...@newt.fsa.ca (Theo de Raadt) writes:
> > >a...@netcom.com (Andrew Moore) writes:
> > >In the mean time, I think it is safe to say that FreeBSD is putting
> > >more effort into i386-specific support.
> >
> >In the meantime, I think it's completely stupid to say anything of the
> >kind.
>
> comparisons to NetBSD, because NetBSD is perfect, and anything that is
> compared to NetBSD is a piece of s*it. (And you will notice little
> digs anytime you compare anything to NetBSD)
I didn't say anything of the kind, Nate. Read it all again... (and by
the way, your posting didn't attribute correctly)
The point I tried to make was that people who make comparisons between
NetBSD and FreeBSD are just stirring the fires. Andrew Moore's original
posting is quite similar to the ``Which is better? XXXBSD or Linux?''
postings we see periodically.
This not a dig against Andrew, since I know he didn't mean any harm.
Perhaps though, Nate could in the future sit back and realize how
futile it is for inside developers to compare in this way. Like i
said:
>In the meantime, I think it's completely stupid to say anything of the
>kind.
Sigh.
<tdr.
Theo de Raadt
> 386bsd-0.1
> 386bsd-0.1 + random patches
> 386bsd-0.1 + patchkit
> FreeBSD + complete patched distribution + development
> NetBSD + complete distribution + development + incompatible binaries
> 386bsd-0.2 + the great unknown
(I'm not attacking you, Amancio, just adding details.)
NOTE: I'm not posting this to create a flame war, so if anyone is
about to reply to this: STOP. If you disagree, send me mail or skip
to the next article.
NOTE: The only mention of FreeBSD made by me in this article occurs
in this sentence.
1. NetBSD 0.9 has all the important patches from the "complete patched
distribution" (as you call it).
2. NetBSD 0.9 generates incompatible binaries by default, but it can also
link and execute old-style executables. The new executable format saves
4K per executable (nearly 1 Mbyte of diskspace saved in /usr/bin alone),
and causes *NULL to cause a SIGSEGV (clearly a win.)
> Look, I am not saying that NetBSD or FreeBSD are inferior or that
> [...]
I'm not saying anything of the kind either, no matter how much Nate
or Jordan twist my words around. Both of them could learn how to
properly attribute articles too.
Comparing things on the Net is really a rather dull idea -- it does not
work. But if you insist on comparing, please get the facts straight.
<tdr.
----------------------------------------------------------------------
details for the finicky:
1a. SIO is the only large patch that was left out. That's because it does
not stick to conventional kernel interfaces to do it's job. It intrudes
into the interrupt subsystem, plays freely with the ring buffer
implementation, saves tty characteristics across open/close invocations,
and perhaps a few other things I have forgotten. Hopefully this will be
resolved soon.
2a. A few old-style executables did things that clearly weren't right.
Let's not even talk about those.
Otherwise, all old-style executable modes should work except the Jolitz
"screwballmode". I think the only executables of this kind are on the
original 386BSD install disk... OMAGIC and NMAGIC should serve the
(limited) purpose of this whacked a.out format once they are implimented
in NetBSD... (ie. not yet)
To generate old style executables use "-Xlinker -Z" in your CFLAGS
variable. The capability to execute old-style executables is enabled by
the kernel option "COMPAT_NOMID", and all shipped kernels have this option
enabled.
By the way, NetBSD also runs BSDI executables.
Elizabeth Haley
All those out there who have or are administering a mixed platform
base of computers raise your hands?
(Don't really, just being catchy...)
Having run a small lab with 10 DOS machines, not to mention some
machines produced by a well known somewhat litigious company, I have
quite a bit of experience with the users of these machines. In my
experience, one could almost have an operating system to suit every
personality. Some want a system that is ever unchanging, they know the
bugs and know how to deal with them. Others want more features, more
speed, more everything, and are willing to change how they work to get
it.
Most are in-between.
Having two separate (but communicating groups) in *BSD land is a good
idea, because different priorities produce different brain storms,
which everyone can eventually benefit from. Just because the NetBSD
folks are trying to be cutting edge doesn't mean they don't want way
stable code. I know from experience that occasionally you can find an
extremely stable version of something that can be tweaked to use
platform dependent features that let it scream. That is, of course,
the reason for having all those #ifdefs.
On the other hand, the FreeBSD, is unlikely to turn down a screaming
piece of code, if it proves stable and robust.
This results in two flavors which appeal to a wider range of users.
I prefer FreeBSD because my current projects don't have to do with
kernel hacking or what not, and I want to be sure that the majority of
my time is spent in emacs hacking MY code.
Remember, quite a few of the advances in automotive design came from
things designed to go MUCH faster, like racing cars and rockets.
--
If you love your fun...
|[{(<=--=>)}]|David Charles Todd, tHE mAN wITH tHREE fIRST nAMES|[{(<=--=>)}]|
|||||||||||||||||||||||||hac...@headcheese.daa.uc.edu||||||||||||||||||||||||
...Die for it!
Kaleb Keithley
>Remember, quite a few of the advances in automotive design came from
>things designed to go MUCH faster, like race cars and rockets.
Umh, yeah, do you mean like the tailfins on a '59 Cadillac? Not that I
don't love 'em; I do, but they're entirely useless!
Now I'm trying real hard, but I fail to see the analogy. Cars have improved
almost continuously, quality control notwithstanding; although the basic
technology remains unchanged from what Daimler and Olds were building 100
years ago. What is it that rocket technology has added that makes cars go
faster? Cars didn't get better because of rocket technology, they got
better in spite of it.
Ditto for race cars. There's too many varieties of race cars to say anything
meaningful; but if you stick to the race cars that most closely resemble
street cars, I have yet to see a production street car that that can sustain
200 MPH (~300 KPH) and survive a full speed collision and protect the
occupants so well that they can walk away from the crash.
But I digress.
What makes an operating system fast(er)? Either a faster CPU, or fewer
instructions to execute. There's a lot of reasons why getting a faster
CPU isn't always an option. So how do you get fewer instructions to
execute? A) Rewrite in assembler. Since we're talking about a family of
operating systems that is written in a high-level language for specific
reasons, we can pretty well rule that out. B) Get a compiler with a better
optimizer. Not many options there, gcc is as gcc does. Anybody got an
alternative? C) Get better, more efficient algorithms.
And that's where competing operating systems comes in. That's what the
free market is all about. If I think I can do better then Stallman, Torvalds,
Jolitz, et al, then I can take my algorithm, plug it into a freely-
redistributable-source operating system, call it <my>BSD, and may the best
operating system win.
Okay, so the world doesn't need 20+ different flavors of BSD, but the market
will take care of that. Some will take off and be successful, and some will
wither on the vine for a variety of reasons. If nothing else, having several
products improves the chances that someone with an existing operating
system might be willing give the new algorithm a try. And if there's no one
out there that recognizes the genius of the new algorithm, then it'll be
a testament to just how strongly the person believes in the superiority
of the algorithm if they're willing to give it a go and roll their own
operating system.
What this all boils down to is, it's way too early to talk about merging the
varieties of <xxx>BSD. Or maybe you think the world would be a better place
if we could go down to the CheChryForVolksRenVolvFiatToyDatHon dealer and
buy a homogenousmobile?
--
Kaleb Keithley
Chris Torek
(Kaleb Keithley) writes:
>What this all boils down to is, it's way too early to talk about merging the
>varieties of <xxx>BSD. Or maybe you think the world would be a better place
>if we could go down to the CheChryForVolksRenVolvFiatToyDatHon dealer and
>buy a homogenousmobile?
Actually, the latter might well be true. People like to talk about how
those outside the computing field work woth modular, interchangeable
parts (`stand on each other's shoulders') while those in computing work
with against each other (`stand on each other's toes'). This simply is
not true. Just try buying a replacement window crank handle for your
car---assuming it has a crank handle. Heck, if you want confusion,
look up the NEMA standards for power connectors. Even once you pin
down the voltage and number of phases, there are still half a dozen
varieties to choose from. :-) (There are good reasons for that, as
well as bad ones, but never mind that.)
We get in trouble when the competition/cooperation ratio gets too
seriously out of balance. Too much competition leads to confusion and
incompatibility; too little leads to inflexible, insufficient, and/or
over-expensive systems. The decisions people make as to how much
competition is `right' are partly a matter of taste, and there is
certainly a workable continuum.
--
In-Real-Life: Chris Torek, Lawrence Berkeley Lab CSE/EE (+1 510 486 5427)
Berkeley, CA Domain: to...@ee.lbl.gov
James Tsillas
Having two very similar (if not three) operating systems seems to me
like a incredible waste of personal resource and will give rise to much
duplicate (triplicate?) work for what appear to be simply religious
reasons.
There are no good reasons that I can see for keeping the efforts
separate. The arguments that we need an experimental release of the OS
in parallel to the stable release is no excuse for the parties to work
independently. What is needed is a good source control/configuration
management system and some good will among all the parties involved. The
first is easy to obtain, the second is much more difficult.
Professionally, I work as the SC/CM specialist amongst a group of thirty
software developers. Our product is the network OS of a high speed
router-bridge. It is not uncommon for us to have three major release and
one beta release in the field and under development (whether
maintenance or new development) at any given time. In addition to this
we have some five or six divergent/experimental efforts towards the
development of new features which our customers demand. We are also
pursuing different hardware architectures and there is usually at least
one porting effort ongoing. Our customers demand this aggressive release
and development policy and we've put alot of resources into perfecting
the underlying methodology and process. My point is: it can be done. You
can have the experimental branch and the stable branch working side by
side and borrowing from each other in a coherent and managed way. It's
not easy, and it takes at least a few dedicated individuals to keep
things on track. But the results are much, much better than keeping the
efforts separate.
I urge the developers of all 386BSD based efforts to carefully consider
the benefits of a parallel release strategy using a good SC/CM system. I
urge them to give a high priority to merging the two OS's and to reach a
common code base. From there the source can be entered into a SC system
to provide a branch structure for the parallel efforts. The major (or
"main" branch) can be used to keep the latest stable source for the next
stable release while the several minor branches can be used to keep
unstable source for experimental releases which may be merged to the
major branch on a selective basis. I believe that RCS can provide 100%
of what you need to make this happen.
regards,
-Jim.
--
*** James Tsillas jts...@damon.ccs.northeastern.edu ***
*** Work: (508)898-2800, Home: (617)641-0513 ***
*** "He is after me. Jim is after him." ***
*** - Hop on Pop, Dr. Seuss ***
Elizabeth Haley
>Having two very similar (if not three) operating systems seems to me
>like a incredible waste of personal resource and will give rise to much
>duplicate (triplicate?) work for what appear to be simply religious
>reasons.
Ahh, but that is the point, it is not for "religious" reasons. There
is nothing at all wrong with having two sets of priorities in this
case. No-one is getting paid for this specifically, so no jobs are on
the line. This is the product of many fertile minds on the internet,
getting together to make something for society, as well as add to
their own experience. The fact that there are divergent groups is
merely a sign that those involved (most anyway) can disagree amiably.
One person says "Faster" someone else says "Stabler". They smile and
say "Can I see yer code once in a while", and then "Sure".
>There are no good reasons that I can see for keeping the efforts
>separate. The arguments that we need an experimental release of the OS
>in parallel to the stable release is no excuse for the parties to work
>independently. What is needed is a good source control/configuration
I would suggest that the groups are not exactly independent. If I
recall, there are those with close connections to both.
>management system and some good will among all the parties involved. The
>first is easy to obtain, the second is much more difficult.
This may be true. However, it is possible that portions would become
so divergent that it would amount to a waste of time to produce
deltas. On the other hand, I think someone said that the differences
were smallish right now. Back on the first hand, I would be somewhat
miffed if support for Freebsd were dropped because I can't easily
change over to NetBSD.
>one porting effort ongoing. Our customers demand this aggressive release
>and development policy and we've put alot of resources into perfecting
>the underlying methodology and process. My point is: it can be done. You
>can have the experimental branch and the stable branch working side by
>side and borrowing from each other in a coherent and managed way. It's
>not easy, and it takes at least a few dedicated individuals to keep
>things on track. But the results are much, much better than keeping the
>efforts separate.
1. A few dedicated individuals to keep things on track. This possibly
would mean someone spending time coordinating when they would rather
be hacking. This usually adds up to half assed coordination.
2. Experimental and stable working side by side. Despite the
wonderfulness of the internet, communication of the sort you are
suggesting requires MUCH interaction, of the sort one can get easily
at a work site such as yours. But the various members of the *BSD
teams are scattered across the country or world. Cgd is in CA, Nate is
in Montana, Charles Hannum is in Boston, Julian is in Australia!(?)
Sure there's the telephone. You want to pay their phone bills?
>I urge the developers of all 386BSD based efforts to carefully consider
>the benefits of a parallel release strategy using a good SC/CM system. I
>urge them to give a high priority to merging the two OS's and to reach a
>common code base. From there the source can be entered into a SC system
Actually there was talk back a long while about releasing patches as
RCS deltas.
Anyway, I am not suggesting that your ideas couldn't be implemented
somehow, but I do think that the amount of extra work is much greater
than you think.
Dyane Bruce
>Ive been reading this NetBSD, FreeBSD 'flamewar' for a while now..
>and one thing just keeps popping in my mind. Does all this arguing
ditto....
Maybe I will just go and use linux. A bunch of uncoordinated hackers,
but at least they are working together.
I have heard reports that they have a version of WABI, and are close
to having a version that runs SCO binaries. They already have auto
IRQ detection as well as a crufty version of shared libs. (I don't
care much for their implementation. very very stupid. But at least
they aren't flaming each other...) Not to mention MSDOG emulator
(A _very_ bad emulator, but at least they have it)
I could see the justification for Netbsd over 386 0.1 as 0.1 is useless
without the patch kit... But Freebsd as well? Come on guys, BSD has
already lost the war. Linux has won, and I _hate_ linux. Merge the
base kernel at least. Get the important things into BSD that
will keep it alive for researchers at least. You know how
embarassing it is for me to tell confirmed linux users how much
better BSD is? Which (free)BSD?
Bunch of little children in a sandbox...
--
Dyane Bruce d...@diana.ocunix.on.ca
29 Vanson Ave. Nepean On, K2E 6A9 So who first started the tradition of
613-225-9920 putting witty sayings in sigs anyway?
Dave Burgess
>In article <MYCROFT.93...@duality.gnu.ai.mit.edu> myc...@duality.gnu.ai.mit.edu (Charles Hannum) writes:
>>In general, you can ignore anything any of the FreeBSD or NetBSD
>>people say after such a comparison, because they are all (including
>>myself) biased assholes who just want to push their own system.
>
>Stating the intent of FreeBSD developers, as I understand them, is not
>intended to push FreeBSD, nor to detract from NetBSD. You call it
>misinformation, but these are simply expressed interests.
>
I call it a complete waste of time.
Remember why I put the 'Is 386bsd better than (your favorite operating
system)?' in section 0 of the FAQ??
Because the question, no matter how it is asked, or how it is answered
nearly always turns into a mud-slinging, nasty, name-calling contest
that does nothing but alienate somebody and generally make everyone that
supports NetBSD, FreeBSD and 386BSD look like a raving lunatic.
It doesn't make any more sense to compare FreeBSD to NetBSD than it does
to ccompare twins. Each effort has an agenda, each effort has many
talented people supporting it, and each effort is better for the
competition/collaboration.
Now, I will go back and see if I can get this 'auto-faq' thing to work
without choking on something. Thank you for your time.
--
------
TSgt Dave Burgess
NCOIC AL/Management Information Systems Office
Brooks AFB, TX
Paul
>
>386bsd-0.1
>386bsd-0.1 + random patches
>386bsd-0.1 + patchkit
Well, let's face it, with FreeBSD and NetBSD now out there I think we should
forget about 0.1 and any patches that go with it. Even Bill doesn't think 0.1
is in a fit state to be usable and the patchkit mechanism, while it it
served us well for a while, should now be quickly forgotten about.
No-one should even consider getting 386BSD 0.1 anymore and we should forget
about it until Bill releases 0.2, people should upgrade to Net or Free BSD as
soon as possible.
>FreeBSD + complete patched distribution + development
>NetBSD + complete distribution + development + incompatible binaries
>386bsd-0.2 + the great unknown
>
>This situation is sad. It appears that the only way to consolidate
>our efforts is for the 386bsd community to choose an OS.
I believe that we will merge and somewhat sooner than in 6 months time. It
may be 6 months before the first merged release appears but that's because
there are a large number of issues to be resolved.
>
>Look, I am not saying that NetBSD or FreeBSD are inferior or that
>one group is less talented than any other, rather it would
>be to our benefit to provide a single alternate solution
>to 386bsd-0.1 or to 386bsd-0.2 whenever it comes out.
Agreed, I'm confident that by the time 0.2 gets released (if the Dr Dobbs
report is correct) we will have merged.
Despite a handfull of individuals who are vehemently patriotic about their
own release, most people in the Net/Free BSD teams are very integrated. In
fact, I would say that most people are in both teams and that code
development taking place now gets migrated very quickly.
--
Paul Richards, University of Wales, College Cardiff
Internet: pa...@isl.cf.ac.uk
JANET: RICHA...@CARDIFF.AC.UK
Andre Skarzynski
: I've got an idea: Why not let the USERS decide how well we're doing,
: rather than debating this endlessly between ourselves? If Lotus
: and Microsoft were actually allowed to write the other company's
: "BYTE" and "PC Magazine" reviews, I'm sure this is very much what
: it would look like! Fortunately, such reviews are generally left
: to people with the outsider's viewpoint.
Hi there all,
I have been playing with both NetBSD-0.9 and FreeBSD. Bofore that I was just
using 386BSD+patchkits. I did try NetBSD-0.8, but had far to many problems with
the wd drives to keep it. I was waiting for 386BSD-0.1.5 or FreeBSD to arrive
when NetBSD-0.9 came on the scene. Goody I said, and installed it. I was
disappointed, the install was easy enough, and the distribution comes in a nice
modular format, so you can install what you want/need, but there where still
bugs in the install. Missing permissions and files for /var/log/aculog etc.
Also they NetBSD-0.9 now tries to use /dev/tty00, but this is no good, because
if you are using a terminal on that port, "w","who", "finger" etc will not
report anyone logged in on that port unless you set the ttys entry to use com0.
I know this not a big problem, but it is a nuisance. Also I had troubles with
wd driver. Not bad, but now and then my wd1003 would hang.
When FreeBSD-1.0-GAMMA arrived, I installed it. I found that the installation
was as easy as NetBSD, but a lot more informative of what was going on.
Unfortantly it is not as modular as NetBSD, where the kernel source is separate
from the complete source, but I did notice that all of the binary distribution
files where there. (like sendamil.cf :).
I will see how it goes with FreeBSD. So far things are working well. I am also
a bit concerned about the /dev/com00 of FreeBSD! I still have lots of testing
to do though.
--
Andre B. Skarzynski -- Information Technology, University of Stellenbosch --
a...@itu1.sun.ac.za ------ Tel: +27 2231 774293 Fax: +27 2231 774102 -------
Kaleb Keithley
>In response to what I wrote:
>>What this all boils down to is, it's way too early to talk about merging the
>>varieties of <xxx>BSD. Or maybe you think the world would be a better place
>>if we could go down to the CheChryForVolksRenVolvFiatToyDatHon dealer and
>>buy a homogenousmobile?
>Actually, the latter might well be true. People like to talk about how
>those outside the computing field work woth modular, interchangeable
>parts (`stand on each other's shoulders') while those in computing work
>with against each other (`stand on each other's toes'). This simply is
>not true. Just try buying a replacement window crank handle for your
>car---assuming it has a crank handle. ...
Ah, the old "Don't just do something, stand there!" argument, i.e. we
can't move forward because we might leave someone behind. To continue
with your crankhandle analogy, the fact that you even need to buy a
replacement connotes opportunities for competitors to produce better
products; that presumably won't break or will be better in some other
tangible way. Knowing that brand X is inferior to brand Y will probably
affect your future "purchase" decisions.
(BTW, I'd like to know in which industries you think competitors really
are cooperating? Excluding those who must because of government mandate,
e.g. the phone companies; from where I stand I see very few!)
>We get in trouble when the competition/cooperation ratio gets too
>seriously out of balance. Too much competition leads to confusion and
>incompatibility;
In the short term perhaps. In the long run everything works itself out.
("In the long run we're all dead." John Maynard Keynes) To make a trite
analogy, you've got to break some eggs to make a cake. A little short
term confusion is eminently worth it if you get a better product out at
the end.
>too little leads to inflexible, insufficient, and/or
>over-expensive systems.
As the great Professor Spooner may once have said: "Vuja de!" And just how
do you think we've reached the point we're at?
>The decisions people make as to how much
>competition is `right' are partly a matter of taste, and there is
>certainly a workable continuum.
It all comes back to tastes (and preferences). Sounds like an economist at
heart. :-)
Charles Hannum
In article <haley.747894008@husc8> ha...@husc8.harvard.edu (Elizabeth
Haley) writes:
But the various members of the *BSD teams are scattered across the
country or world.
That's why we chose the name `NetBSD'. It is a tribute to the
internet, without which our combined work would not be possible.
The world changes, line by line...
Robert Withrow
jts...@damon.ccs.northeastern.edu (James Tsillas) writes:
| Having two very similar (if not three) operating systems seems to me
| like a incredible waste of personal resource and will give rise to much
| duplicate (triplicate?) work for what appear to be simply religious
| reasons.
Sigh. I don't even think it is religion anymore. I claim this is all the
result of a few ``personality challenged'' individuals building empires.
You can detect this in the arrogant and postured missives we get from
two of the camps.
And a third camp (trying to steer a middle-course) is beset by active
non-cooperation from the other two, and thus hasn't been able to unite.
As a result they are (unintentially) helping to fragment.
If Palestine and Israel can come to an agreement, why can't the BSD
warriors?
--
Robert Withrow, Tel: +1 617 598 4480, Fax: +1 617 598 4430, Net: wi...@rwwa.COM
R.W. Withrow Associates, 21 Railroad Ave, Swampscott MA 01907-1821 USA
Amancio Hasty Jr
efforts we are evolving and coming along very well.
In article <1993Sep13....@cm.cf.ac.uk> pa...@myrddin.isl.cf.ac.uk (Paul) writes:
>
>I believe that we will merge and somewhat sooner than in 6 months time. It
>may be 6 months before the first merged release appears but that's because
>there are a large number of issues to be resolved.
>
Care to elaborate or at least list the issues?
Thank you,
Bob Costa
So why did these separate groups form in the first place?
What would cause someone to decide to go their own way and not contribute
to an "already established" group? Are the co-ordinators of these
"already established" groups pig-headed? Do they refuse to listen when
advise is given? Or are the people giving the advise pushy and demaning
and thefore go their own way?
Just wondering, because it seems to me that the only way you can have a
split is if there is disatisfaction with what is already there.
Sorry, I don't mean to stir the pot. I'm just wondering why this thread
is so long...
bc
Jaye Mathisen
Bob Costa <co...@benton.prepress.com> wrote:
>
>So why did these separate groups form in the first place?
>
>What would cause someone to decide to go their own way and not contribute
>to an "already established" group? Are the co-ordinators of these
>"already established" groups pig-headed? Do they refuse to listen when
>advise is given? Or are the people giving the advise pushy and demaning
>and thefore go their own way?
The groups didn't split off from a main group if that's what you mean.
To summarize (again).
386bsd 0.1 comes along. A group of people start working on a patchkit,
to fix some of the problems. In that same time period, for reasons
that the NetBSD group should explain, NetBSD was formed. I think part
of it was unhappiness with 386bsd (which was a disaster), a lack of
participation from the developer, and a different vision for what
a free PC based unix should become.
So NetBSD is off doing it's thing.
Meanwhile, the patchkit people are still working on 386bsd, and decided
to create an "interim" release of 386bsd, as WFJ was taking an incredible
amount of time to get 0.2 out.
Eventually, the interim group reached its limit wrt wfj, and how things
were proceeding, and decided to put their interim effort out as FreeBSD.
So FreeBSD is not a group of people that got together, and decided that
NetBSD wasn't good enough and wanted to do their own thing, they're two
separate evolving groups.
There has been talk of merging between NetBSD and FreeBSD, but I'm not
sure it will happen, there's some strong personalities that make it
difficult to see how working together would work.
--
Jaye Mathisen, COE Systems Manager (406) 994-4780
410 Roberts Hall,Dept. of Computer Science
Montana State University,Bozeman MT 59717 os...@cs.montana.edu
Jaye Mathisen
Thankfully, I didn't save all that mail, it was too painful.
Quite honestly, it's mainly personnel issues, as to who should do what.
There is a document describing the proposed merge, but I don't feel
at liberty to post it, as it does mention names and such.
There aren't many technical reasons, that a merge can't happen.
A few minor differences exist:
FreeBSD feels sio is stable enough, NetBSD doesn't.
NetBSD has the interest of ports to non-PC platforms, I'm not sure that
FreeBSD does.
NetBSD has the definite intent of moving towards 4.4, FreeBSD doesn't.
There are others. Hell, these aren't even major ones, they're just some.
Frankly, I feel there's room for both. Since it's a volunteer effort, there's
no way to "force" somebody to work with somebody else, and becoming
dictatorial about it is a no-win situation, as it makes other people
unhappy.
People that complain that the divisiveness causes a problem need to
state their case better. For the most part, applications will drop right
into both, as long as it doesn't muck around in the kernel much.
People that want to hack can pick and choose.
Nothing keeps FreeBSD from picking up the best i386 specific parts of
NetBSD and incorporating them and pushing the product.
A Wizard of Earth C
>If Palestine and Israel can come to an agreement, why can't the BSD
>warriors?
The only thing standing between Palestine and Israel is their beliefs...
for God's sake Bob, we're talking *code*! }B-).
Terry Lambert
te...@icarus.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Jordan K. Hubbard
> Maybe I will just go and use linux. A bunch of uncoordinated hackers,
That would be fine and you have our blessings, go!
Linux is a fine operating system (meant sincerely), with many
admirable features. Don't sit here whining about what a bunch of
jerks we are, defect! Vote with your feet!
> I could see the justification for Netbsd over 386 0.1 as 0.1 is useless
>without the patch kit... But Freebsd as well? Come on guys, BSD has
Actually, the conception of both projects was pretty close to
simultaneous, and born from the same stimulus (inadequate patchkit,
total lack of participation from Bill). We're doing this for free;
you want to whine at me, give me some money. Give us all some money
while you're at it, then we'll at least listen politely and nod our
heads at the right places.
Jordan K. Hubbard
>was as easy as NetBSD, but a lot more informative of what was going on.
>Unfortantly it is not as modular as NetBSD, where the kernel source is separate
Kind words - and the whole install has just been *significantly*
reworked, based on the many useful bug reports we received on the
first set. See recent announcement about availability on freefall.
Vax
Robert Withrow <wi...@rwwa.com> wrote:
>In article <JTSILLA.93...@damon.ccs.northeastern.edu>,
> jts...@damon.ccs.northeastern.edu (James Tsillas) writes:
>
>Sigh. I don't even think it is religion anymore. I claim this is all the
>result of a few ``personality challenged'' individuals building empires.
>You can detect this in the arrogant and postured missives we get from
>two of the camps.
>
Since the post before this one (this was the last of the thread for me)
was about tastes and preferences, I just thought I'd throw in my
$.02 about the subject.
As long as FreeBSD and NetBSD are close enough to 386BSD, my willingness
to switch to any of them is proportional to the relative advantage,
and indirectly proportional to the ease of switchover. If I were confident
that a switchover to Net or Free was easy, like a patchkit kind of thing,
I'd do it - but I've had troubles simpy restoring my dump off a variable
blocking tape drive, so things that often seem simple are not.
NetBSD, last time I looked (0.8) was a whole bunch of binary data cut up
into little chunks - hardly what I would call an easy upgrade from
a working, 300+MB 386BSD system! Let me tell you, I -really-
look forward to dumping the whole system to tape, formatting a bazillion
floppies and unpacking everything by hand (mcopy a:\* /tmp x 100000)
and then hoping that a restore from a Tandberg SCSI tape drive will work.
(Smell that? (sniff sniff)... That's sarcasm!)
Now, if I saw an SLS-type system, with binary disks that might actually
-work- for my SCSI tape drive, then I might think about it....
Is it too much to ask? Is my system so strange that most bootdisks won't
read the SCSI stuff properly? It's just an AHA-1542B with a TDC-3800 on the
end!
Shoot, BSD land is so fragmented I'm going to experiment with Linux when
I get my new hard drive.
--
Protect our endangered bandwidth - reply by email. NO BIG SIGS!
VaX#n8 v...@ccwf.cc.utexas.edu - Don't blame me if the finger daemon is down
Amancio Hasty Jr
I hate to see two separate groups doing almost the same thing.
If you think that there is room for both think again -- there are
many areas that *BSD is lacking such that if we had a single
effort we could perhaps pick-up, ISDN support, Multi-media
applications, hardware support for 3d-graphics, fddi support,
BIOS support, WABI, etc... Is anyone looking into CORBA?
What about nifty database applications? the list goes on...
Fine, lets mirror the tiny Unix market,perhaps for the same reasons.
It is clear that if there is to be a single *BSD distribution
that the selection rests with the user community since the principal
developers can not come to an agreement.
The future now rests with the users...
Nate Williams
Vax <v...@sneezy.cc.utexas.edu> wrote:
>
>As long as FreeBSD and NetBSD are close enough to 386BSD, my willingness
>to switch to any of them is proportional to the relative advantage,
>and indirectly proportional to the ease of switchover. If I were confident
>that a switchover to Net or Free was easy, like a patchkit kind of thing,
>I'd do it - but I've had troubles simpy restoring my dump off a variable
>blocking tape drive, so things that often seem simple are not.
If you're interested in *upgrading* to FreeBSD, there exists a set of
scripts that will do it for you, ala patchkit style of script. (But
unfortunately it's not done like a patchkit, but there are reasons for
that outlined in the README)
I have had harddly any feedback from the BETA testers, but what I have had
has been positive, so you are welcome to try them out, or at least
download them and look if you're interested.
It doesn't require a complete-reinstall, it just updates your entire
machine to FreeBSD from 386BSD.
If you are interested, send me some email, prefferably to:
freebsd-...@freefall.cdrom.com OR
na...@bsd.coe.montana.edu
And I'll send you some email stating where you can find them.
Sorry, but they are only available for those who ask right now, so that
I can know who they are going out to, in order to get feedback on the
scripts.
>Let me tell you, I -really-
>look forward to dumping the whole system to tape, formatting a bazillion
>floppies and unpacking everything by hand (mcopy a:\* /tmp x 100000)
>and then hoping that a restore from a Tandberg SCSI tape drive will work.
Well, my Tanberg 3600 tape work last time I checked it (was a while ago),
so I suspect it will still work.
>Now, if I saw an SLS-type system, with binary disks that might actually
>-work- for my SCSI tape drive, then I might think about it....
Your SCSI tape should work fine, if it works at all. If it is flakey,
chances are it's a hard-ware problem. (Bad tapes, bad firmware, etc)
>
>Is it too much to ask? Is my system so strange that most bootdisks won't
>read the SCSI stuff properly? It's just an AHA-1542B with a TDC-3800 on the
>end!
Huh? What do you mean, read the SCSI stuff properly. Can you be a little
more specific?
Nate
--
na...@bsd.coe.montana.edu | In the middle of it ........ again.
na...@cs.montana.edu | Running/supporting one of many freely available
work #: (406) 994-4836 | Operating Systems for [34]86 machines.
home #: (406) 586-0579 | (based on Net/2, name changes all the time :-)
Paul
>efforts we are evolving and coming along very well.
>
>In article <1993Sep13....@cm.cf.ac.uk> pa...@myrddin.isl.cf.ac.uk (Paul) writes:
>>
>>I believe that we will merge and somewhat sooner than in 6 months time. It
>>may be 6 months before the first merged release appears but that's because
>>there are a large number of issues to be resolved.
>>
>
>
>Care to elaborate or at least list the issues?
>
Not really :-)
However, there are quite a few differences in the actual code. Just
mergeing the two source trees will take considerable time but before
we even start on that we have to merge the two teams and that
inevitably involves some politics.
Richard Tobin
>2. NetBSD 0.9 generates incompatible binaries by default, but it can also
> link and execute old-style executables.
This is true and good, but the change in db format has made a lot of
old executables useless. A large proportion of interesting programs
access the password file, for example.
Also, programs whose data segment isn't a multiple of the page size don't
work any more (eg kcl saved states).
-- Richard
--
"For thousands of years, [homoeopathic magic] was known to the sorcerors of
ancient India, Babylon and Egypt, as well as of Greece and Rome, and at this
day it is still resorted to by cunning and malignant savages in Australia,
Africa and Scotland." - J G Frazer, The Golden Bough
Jaye Mathisen
>that a switchover to Net or Free was easy, like a patchkit kind of thing,
>I'd do it - but I've had troubles simpy restoring my dump off a variable
>blocking tape drive, so things that often seem simple are not.
Yes, tape drive stuff is being fixed and worked on, somebody else on
the hacker list is having the same problem. However, at the source/binary
upgrade level, Nate's scripts work just fine.
>NetBSD, last time I looked (0.8) was a whole bunch of binary data cut up
>into little chunks - hardly what I would call an easy upgrade from
NetBSD so far has not focused much on packaging, as opposed to functionality.
>
>Now, if I saw an SLS-type system, with binary disks that might actually
>-work- for my SCSI tape drive, then I might think about it....
Well, good look getting disks to work in your tape, I just tried it, and
it didn't work for me. I'm not sure where you feel the time savings would
be, if there was an SLS-style distribution for NetBSD, it would still be
on a kazillion floppies, if only because of the lack of shared libs.
>Shoot, BSD land is so fragmented I'm going to experiment with Linux when
>I get my new hard drive.
Well, that's your perogative, have a nice time. I run NetBSD, FreeBSD, and
Linux all at home, and Linux is nice, but it's very subject to patch-of-the-day
syndrome. (which is appropriate given the relative age).
Frankly, I don't see what the big deal is. There are many advantages to
FreeBSD and NetBSD over 386bsd0.1. Applications port back and forth easily,
what the hell is the big deal? The only serious difference is in stuff
in the kernel, and still, what's the big deal? Pick one and run with it.
Dyane Bruce
>
>> Maybe I will just go and use linux. A bunch of uncoordinated hackers,
>
>That would be fine and you have our blessings, go!
I have no desire to use a second rate operating system.
>Linux is a fine operating system (meant sincerely), with many
>admirable features. Don't sit here whining about what a bunch of
>jerks we are, defect! Vote with your feet!
Unfortunately for BSD, a lot of people have already voted with
their feet... BSD needs co-operation now, healthy competition
can come later.
>> I could see the justification for Netbsd over 386 0.1 as 0.1 is useless
>>without the patch kit... But Freebsd as well? Come on guys, BSD has
>
>Actually, the conception of both projects was pretty close to
>simultaneous, and born from the same stimulus (inadequate patchkit,
>total lack of participation from Bill). We're doing this for free;
I understand that. Bill has abrogated leadership for BSD.
Linus has not (as far as I know) abrogated leadership for
the Kernel of Linux.
>you want to whine at me, give me some money. Give us all some money
>while you're at it, then we'll at least listen politely and nod our
>heads at the right places.
I'm sorry, but I'd rather hack kernels. I don't have any money
to contribute. I wish I did. Right now I am too busy scrambling to
stay afloat in the rough job market up here to even do that....
Incidentally, there is a thread in the linux groups about
contributing a machine to Linus for development. Ideally it
would be nice to resurrect CSRG, but that isn't going to happen.
I am also glad to see cooler minds prevail, and the hot heads
calmed down in this thread. I am glad to see a merge of the
core kernel being discussed, not flame throwers being used.
Mark Sienkiewicz
>>386bsd-0.1 + random patches
>>386bsd-0.1 + patchkit
>
>Well, let's face it, with FreeBSD and NetBSD now out there I think we should
>forget about 0.1 and any patches that go with it. Even Bill doesn't think 0.1
>is in a fit state to be usable and the patchkit mechanism, while it it
>served us well for a while, should now be quickly forgotten about.
The way to make this happen is to remove it from the archive sites.
Consider this typical scenario-
Somebody hears about 386bsd, maybe at a party, maybe from a back
issue of DDJ, maybe it was a few months ago, but now's the time
to check it out.
% archie 386bsd
>>tons of archive sites reported<<
% Pnews
Subject: I can't get 386bsd 0.1 to work...
If you are an archive operator, at least put a note in the 386bsd tree
that points out that it is obsolete and directs people to Netbsd and Freebsd.
If you know an archive operator, please pass the word along.
Amancio Hasty Jr
>In article <2730dk$5...@sneezy.cc.utexas.edu>,
>Vax <v...@sneezy.cc.utexas.edu> wrote:
>syndrome. (which is appropriate given the relative age).
>
Well, the last time I took a peek at c.o.l they had over 50k postings
must be very interesting to sit on front of the screen and grab a patch :-)
I think our strenght now is FreeBSD and NetBSD's complete OS and hopefully
applications distributions.
Cheers,
Amancio Hasty Jr
Not bad,
Now the next step, someone should publish a paper to a magazine
such as Byte or Unix World and describe what NetBSD and FreeBSD
represents along with very detail information on how to install
the packages. For instance, "It is beyond the scope of this
document to tell you how to use the dd command", is not a
very good instruction for the Unix beginer.
Also, I don't know if this is a local phenomena but the
San Francisco newspaper, San Francisco Chronicles are
publishing more and more technical issues in the
business sections -- other major newspapers may have
a computing section
Do you show off the strenght of the X distribution including XS3.
If someone tells me that I can have the same of better graphics
performance than a sparc 2 and at higher resolution, you bet
I am going to pay attention. Disk thruput is another big one,
details on how to achieve the 1.44mb disk thruput or higher
being observed over at sun-lamp.cs.berkeley.edu will help
a lot specially if the cdrom installation can fine tune
the file system for you.
Along with a very long list of currently ported applications
available in the cdrom.
Several months ago, on one of the popular Unix magazines there
was a whole article about Linux and how to installed it.
If we anchor the article with an address of where to get
the cdrom distribution and of course a few ftp sites it
will be like icing on a cake.
How hard is this do?
Not hard, just elaborate a bit more on the current installation
notes and mailed it :-)
Good Luck,
Larry Felawka
>>>386bsd-0.1 + random patches
>>>386bsd-0.1 + patchkit
>>
>>Well, let's face it, with FreeBSD and NetBSD now out there I think we should
>>forget about 0.1 and any patches that go with it. Even Bill doesn't think 0.1
>>is in a fit state to be usable and the patchkit mechanism, while it it
>>served us well for a while, should now be quickly forgotten about.
>
>The way to make this happen is to remove it from the archive sites.
Oh, kind Sir .. please don't! I found the 386bsd binaries very useful
for bringing up FreeBSD when it first came out. I used goodies like mread,
kermit, cksum (the FreeBSD version seems to use a different algorithm),
emacs, tcsh and so on. Ahh, binary compatibilty is a wonderful thing!
Larry F.
Herb Peyerl
NOTE: This is a repost of message <276kq0$2...@fw.novatel.ca> (T.P.)
James Tsillas (jts...@damon.ccs.northeastern.edu) wrote:
: Having two very similar (if not three) operating systems seems to me
: like a incredible waste of personal resource and will give rise to much
: duplicate (triplicate?) work for what appear to be simply religious
: reasons.
: There are no good reasons that I can see for keeping the efforts
: separate. The arguments that we need an experimental release of the OS
I don't understand what all the whining is about... I see nothing wrong
with a little bit of parallelism... After all; where would the world be
if Macdonald's had had no competition. Would we have one Hamburger to
choose from? Would there be One car? (as was brought up earlier). How
'bout the "Yellow God" (Kodak). There's all kinds of examples. In fact;
our society (well; at least the one I'm a member of) is one big huge
example of competition at work. Without competition; it becomes difficult
to find a stimulus for improving something which is already sufficiently
functional. Agreeably; competition can be taken too far and have dire
consequences (See: Microsoft/Xerox/Apple) but I think in our case(s);
I don't think anyone has anything to fear... Last I recall; there weren't
any rabid-kill-programmers in either camp holding loaded semi-automatic
weapons ready to shoot at sight any member of the other camp (Though
cgd *does* live in California and drives a truck so one never knows....).
In any case; I think we're all safe. I think we're all having fun;
and whoever isn't can (as stated by jordan) "Vote with their feet".
--
hpe...@novatel.ca | NovAtel Commnications Ltd.
hpe...@fsa.ca | <nothing I say matters anyway>
<NetBSD: A drinking group with a serious computing problem!>
Gene Stark
I just wanted to add that, in upgrading my system last weekend from
386BSD 0.1 + pk 0.2.4 to FreeBSD, I decided to "recycle" my archival floppies
used for the 386BSD 0.1 distribution. However, there were several things
I kept:
a virgin Tiny BSD floppy
a virgin Fixit floppy
a copy of the 386BSD kernel source tree, with all the old
versions of files created by the patchkit
*all* my "etc" distribution floppies.
I decided not to recycle the "etc" floppies because there were several
things on there that I use regularly which did not find their way into
the FreeBSD tree. In particular, kermit and mtools. There were a couple
of other things on that distribution that are also useful. I don't care
about the vast majority of stuff in that distribution, and if I could
collect demonstrably working versions of the things I do use (with source),
then I would gladly recycle these floppies as well.
I kept the Tiny BSD floppy and the Fixit floppy because I almost got
burned by the 0.2.4 wd driver not being able to disklabel an unlabeled
disk. I did have to revert to the Tiny BSD floppy at one point to get
my system back into a usable state.
I kept the kernel source tree with old versions for possible comparison
with later versions, in case of questions about later changes or maybe
even copyright notices.
- Gene Stark
--
st...@cs.sunysb.edu
Amancio Hasty Jr
>
>NOTE: This is a repost of message <276kq0$2...@fw.novatel.ca> (T.P.)
>
>James Tsillas (jts...@damon.ccs.northeastern.edu) wrote:
>
>: Having two very similar (if not three) operating systems seems to me
>: like a incredible waste of personal resource and will give rise to much
>: duplicate (triplicate?) work for what appear to be simply religious
>: reasons.
>: There are no good reasons that I can see for keeping the efforts
>: separate. The arguments that we need an experimental release of the OS
>
>I don't understand what all the whining is about... I see nothing wrong
>with a little bit of parallelism... After all; where would the world be
We do have a very strong competition and it is called Linux.
If we wish to win against linux, we are going to need everything
that we got plus a bit more like in terms of marketing to spread
the *bsd word around.
Alex R.N. Wetmore
partitioning by Amancio Hasty Jr@netcom.
> If we wish to win against linux, we are going to need everything
> that we got plus a bit more like in terms of marketing to spread
> the *bsd word around.
I really don't get this whole bsd vs linux thing. They are two totally
different ways of doing this stuff. People should just choose what they
feel they are most happy with, and we really shouldn't be out there to
convert them.
Its not like anyone is making money off of this. We aren't loosing
"customers" per se by having people running linux at this point. I like
the *bsd crowd because I feel like they have a better idea of what is
going on and how to do stuff, since the majority of the implementers are
unix people putting unix on a pc, instead of pc people trying to find
something to do besides dos or os/2.
alex
Nate Williams
>In article <277do6$4...@fw.novatel.ca> hpe...@novatel.ca (Herb Peyerl) writes:
>>I don't understand what all the whining is about... I see nothing wrong
>>with a little bit of parallelism... After all; where would the world be
>
>We do have a very strong competition and it is called Linux.
>If we wish to win against linux, we are going to need everything
>that we got plus a bit more like in terms of marketing to spread
>the *bsd word around.
>
Competition? I don't think you can call two freely available *nix
operating systems *competition*. I would much rather see folks be using
Linux than be moving towards using NT, or DOS. That is where the
*competition* should be.
Long live *ALL* free uni*es.
Nate
--
na...@bsd.coe.montana.edu | In the middle of it ........ again.
na...@cs.montana.edu | Running/supporting one of many freely available
work #: (406) 994-4836 | Operating Systems for [34]86 machines.
home #: (406) 586-0579 | (based on Net/2, called FreeBSD today :-)
Dave Burgess
>
>Competition? I don't think you can call two freely available *nix
>operating systems *competition*. I would much rather see folks be using
>Linux than be moving towards using NT, or DOS. That is where the
>*competition* should be.
>
>Long live *ALL* free uni*es.
>
Here, here.
I agree with this whole heartedly. There is plenty enough RAW talent to
go around. With a 'market' of two or three free uni*es, there is plenty
of fertile ground to plant and cultivate new ideas. Loadable Kernel
Modules are neat, and a part of NetBSD. The new install procedures for
FreeBSD are on their way to becoming an 'industry' standard. The Linux
way of doing things is definitely different, but it is no less right or
wrong than the way the BNR/2 systems do their thing.
Does anyone honestly believe that the developers of software for *BSD
completely ignore Linux? Or that the NetBSD folks completely ignore the
FreeBSD stuff? I would be willing to guess that for half of that group
they would spend a lot of time ignoring themselves.
We aren't talking about competing product lines with FreeBSD and NetBSD;
we are talking about two systems, conceived for the same reason, at
about the same time, with two distinct and seperate purposes. I Like
it.
--
------
TSgt Dave Burgess
NCOIC AL/Management Information Systems Office
Brooks AFB, TX
M.C. Wong
>In article <CD190...@latcs1.lat.oz.au> wo...@latcs5.lat.oz.au (M.C. Wong) writes:
>>Hi,
>> I am currently using 386bsd 0.1 + pk 0.2.4! My question is how easy
>>it is to upgrade from :
>>
>> i) 386bsd to FreeBSD
>> ii) 386bsd to NetBSD
>
>There is an upgrade script from any 386BSD version to FreeBSD.
>
Hi,
So where is the upgrade script, does it have full installation note or the
like ?
Thanks in advance!
- wo...@latcs1.lat.oz.au (M.C Wong)
--
- wo...@latcs1.lat.oz.au
Brett Lymn
James> Nntp-Posting-Host: damon.ccs.northeastern.edu
[rave about the joys of change control software deleted]
James> unstable source for experimental releases which may be merged to the
James> major branch on a selective basis. I believe that RCS can provide 100%
James> of what you need to make this happen.
IMHO RCS will not cut it. It is too fiddly to try and maintain the
status of all those files manually. If a CM system was to be used
then I would strongly suggest that CVS would be a better way to go.
Apart from being able to tell CVS "check out all the files to do with
ps" it has other nice features like being able to automatically
generate patch files. Of course this all assumes you have the disk
space to hold > 2*`du -s /usr/src`. If you are *serious* about doing
kernel hacking you should have.
--
Brett Lymn
Nate Williams
Brett Lymn <bl...@awadi.com.au> wrote:
>>>>>> On Mon, 13 Sep 1993 02:37:41 GMT, jts...@damon.ccs.northeastern.edu (James Tsillas) said:
James> Nntp-Posting-Host: damon.ccs.northeastern.edu
>
[rave about the joys of change control software deleted]
>
James> unstable source for experimental releases which may be merged to the
James> major branch on a selective basis. I believe that RCS can provide 100%
James> of what you need to make this happen.
>
Brett>status of all those files manually. If a CM system was to be used
Brett>then I would strongly suggest that CVS would be a better way to go.
Brett>Apart from being able to tell CVS "check out all the files to do with
Brett>ps" it has other nice features like being able to automatically
Brett>generate patch files. Of course this all assumes you have the disk
Brett>space to hold > 2*`du -s /usr/src`. If you are *serious* about doing
Brett>kernel hacking you should have.
Both the FreeBSD and NetBSD teams use CVS for all their source code
control issues, and it works rather well for concurrent editing of
files. (Though there are still multiple developers editing one file
issues)
Nate
--
na...@bsd.coe.montana.edu | Freely available *nix clones benefit everyone,
na...@cs.montana.edu | so let's not compete with each other, let's
work #: (406) 994-4836 | compete with folks who try to tie us down to
home #: (406) 586-0579 | proprietary O.S.'s (Microsloth) - Me
Noriyuki Soda
> Both the FreeBSD and NetBSD teams use CVS for all their source code
> control issues, and it works rather well for concurrent editing of
> files. (Though there are still multiple developers editing one file
> issues)
Why FreeBSD and NetBSD teams doesn't use CVS branch to share one
common source repository ? I know first cvs-importing to share tree is
painfull work. But the sooner, the less pain.
IMHO, one that already has much cvs-commits must provide repository,
and one that has less cvs-commits must become branch.
--
so...@sra.co.jp SRA システム開発６部 曽田哲之 (Soda Noriyuki)
Dyane Bruce
but kept the References... ]
In article <hastyCD...@netcom.com> ha...@netcom.com (Amancio Hasty Jr) writes:
>In article <277do6$4...@fw.novatel.ca> hpe...@novatel.ca (Herb Peyerl) writes:
>>
>>NOTE: This is a repost of message <276kq0$2...@fw.novatel.ca> (T.P.)
>>
>>James Tsillas (jts...@damon.ccs.northeastern.edu) wrote:
>>
>>: Having two very similar (if not three) operating systems seems to me
>>: like a incredible waste of personal resource and will give rise to much
>>: duplicate (triplicate?) work for what appear to be simply religious
>>: reasons.
>>: There are no good reasons that I can see for keeping the efforts
>>: separate. The arguments that we need an experimental release of the OS
>>
>>I don't understand what all the whining is about... I see nothing wrong
>>with a little bit of parallelism... After all; where would the world be
>
>We do have a very strong competition and it is called Linux.
>If we wish to win against linux, we are going to need everything
>that we got plus a bit more like in terms of marketing to spread
>the *bsd word around.
>
> Amancio
Exactly what I said in the thread of NetBSD vs. FreeBSD. I have to
repeat myself. Cooperation now is important for the survival
of BSD, healthy competition can come later....
Linux just about has "WINE" (A WABI lookalike), and now runs
(Or very shortly will run) SCO binaries has a DOS emulator etc.
Running some SCO binaries makes it commercially viable...
I evaluated Linux a few months ago (As well as NetBSD 0.8) for
a client. At that time Linux made me puke :-) (The source code
for some of the drivers *sucked* big time. For example figure out
whether you are a NE2000 or not by reading the high order bytes
out of the boards ether address and using that... *yuck* and this
was only one example...) But maybe I will have to re-evaluate the latest
and greatest of Linux....
The point is, Bill has abrogated leadership. Cooperate on a stable
Kernel. There are competing _distributions_ of Linux (SLS, Slackware
come to mind...) but they both use the same Kernel. What is wrong
with having a merged Kernel but different install/distribution
policies. One for *hackers* and one for newbies?
Hell, once I get back on my feet financially I'd be quite
willing to write in an ELF or COFF loader. Thinking about Writing
a WABI emulator makes me sick, then I'd have to understand
the Windows API.... :-)
Alex R.N. Wetmore
merge urged! by Dyane Br...@diana.ocunix
The Wine project is being worked on for both Linux and NetBSD. Look at
the post recently put on 386bsd.announce.
alex
Shawn F. Mckay
to be *VERY* stable. Some folks like to tinker, but dont like
the idea that one day it may all just BLOW UP / STOP WORKING.
Although I see little mail about 386bsd, I see LOTS of people
complaining about NetBSD, or FeeBSD for one reason or another.
Its all well and good to be "developing" an operating system, but
there should be something around for folks like me who like to
have a reasonable "development" system, while still actully
USING the O/S.
I look forward to 386bsd 1.0 when it is released. And have
hopes it will be as stable as 0.1/0.2.4 has proven to be.
- Shawn
Amancio Hasty Jr
>I can only speak for myself, but I find 386BSD 0.1 (PK 0.2.4)
>to be *VERY* stable. Some folks like to tinker, but dont like
>the idea that one day it may all just BLOW UP / STOP WORKING.
>
>Although I see little mail about 386bsd, I see LOTS of people
>complaining about NetBSD, or FeeBSD for one reason or another.
>
The problem with NetBSD and FreeBSD is that the applications
deployment lacks behind the OS release also many of us are
accustomed to having certain basic applications working:
tk, perl, X, emacs, epoch, etc...
Hopefully, with freeBSD's user packages it will contribute to a more
stable and usable OS distribution.
The art of an OS distribution is evolving and in a few more months
I would expect awesome OS releases.
Cheers,
Dyane Bruce
I'll have to check my newsfeed. The only thing I currently
have in "comp.os.386bsd.announce" is the FAQ by Dave Burgess...
It's also possible that it hasn't arrived here yet. (I am
a uucp only "leaf" node)
That's great news.... Gives me something to tell my Linux
bigoted friends.. I have also heard that a port of the DOS emulator
was underway to NetBSD as well... BSD still needs to be able
to run SCO binaries though...
Charles Hannum
In article <1993Sep18.0...@diana.ocunix.on.ca>
d...@diana.ocunix.on.ca (Dyane Bruce) writes:
[...] I'd be quite willing to write in an ELF or COFF
loader.
Look at the way execve() works in NetBSD-current. It should be easier
to plug in different executable formats, and with a little work
alternate syscall tables.
Thinking about Writing a WABI emulator makes me sick, [...]
Someone claims he was running Windows Solitaire under NetBSD-current.
I assume this requires some changes to Wine; I don't know if they've
been incorporated into their code yet.
Eric Youngdale
> Thinking about Writing a WABI emulator makes me sick, [...]
>
>Someone claims he was running Windows Solitaire under NetBSD-current.
>I assume this requires some changes to Wine; I don't know if they've
>been incorporated into their code yet.
They have been incorporated into the distribution. Incidentally, if
anyone wants to join in and help out, an extra set of hands is always welcome.
There is a lot left to be done, and people with Windows or X experience are
especially welcome. People who complain, flame or argue all of the time are
asked not to sign on. To get a copy of the latest development snapshot, you
should first sign onto the mailing list that we use to communicate ideas.
To subscribe to the mailing list, send a mesage to
linux-activ...@joker.cs.hut.fi, and put the following line in the
header:
X-MN-Admin: join WABI
When you send a message to the mailing list, CC it to
linux-a...@niksula.hut.fi. You must include the line:
X-Mn-Key: WABI
in the header. To leave the channel, send a message to the admin channel with
the line:
X-Mn-Admin: leave WABI
in the header.
-ERic
--
"When Gregor Samsa woke up one morning from unsettling dreams, he
found himself changed in his bed into a lawyer."
Jon Hilton
>I can only speak for myself, but I find 386BSD 0.1 (PK 0.2.4)
>to be *VERY* stable. Some folks like to tinker, but dont like
>the idea that one day it may all just BLOW UP / STOP WORKING.
I agree, I also use 386BSD PK 0.2.4
>Although I see little mail about 386bsd, I see LOTS of people
>complaining about NetBSD, or FeeBSD for one reason or another.
Also true - I'm waiting for some of the dust to settle
>Its all well and good to be "developing" an operating system, but
>there should be something around for folks like me who like to
>have a reasonable "development" system, while still actully
>USING the O/S.
Absolutely - and I think that your wishes will be fulfilled.
>I look forward to 386bsd 1.0 when it is released. And have
>hopes it will be as stable as 0.1/0.2.4 has proven to be.
Do you mean 0.2? Remember what 0.1 was like BEFORE the patchkits?
Remember also that many of the kind folk who worked on the patchkits
are also involved in one (or both?) of the netBSD/FreeBSD efforts.
Just my $2/100
> - Shawn
-Jon
--
--------------------------------------------------------------
Jon Hilton, 536 Tooronga Road, East Hawthorn, Victoria, 3123
Phone/Fax : 61+(0)3 882 9643
EMail: j...@kite.med.monash.edu.au
Szabolcs Szigeti (PinkPanther)
>In article <Ygam_WG00...@andrew.cmu.edu> "Alex R.N. Wetmore" <aw...@andrew.cmu.edu> writes:
> That's great news.... Gives me something to tell my Linux
>bigoted friends.. I have also heard that a port of the DOS emulator
>was underway to NetBSD as well... BSD still needs to be able
>to run SCO binaries though...
>
he has SCO emulation running, but can't release the code, cause it has
some not public stuff in it. Would be great!!! (correct me if i'm wrong)
On the other hand i don't realy understand why on earth would one want to
run DOS/Windows under NetBSD. I don't think that it would be so fast/compatible
that serious work could be done under it. (Ready, steady, flame!)
---
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
|Too high baud rate always means: | Szabolcs Szigeti |
| @%^gg^@* . !# jjEjWPo[[23@ | Internet: pi...@fsz.bme.hu |
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Dyane Bruce
>In article 29...@diana.ocunix.on.ca, d...@diana.ocunix.on.ca (Dyane Bruce) writes:
>>In article <Ygam_WG00...@andrew.cmu.edu> "Alex R.N. Wetmore" <aw...@andrew.cmu.edu> writes:
>> That's great news.... Gives me something to tell my Linux
...
>I remember somebody (Terry Lambert?) mentioned a couple of months ago that
>he has SCO emulation running, but can't release the code, cause it has
You are right... It was Terry (Wizard of Earth C, I've read that
book too Terry :-) )
[I've ruthlessly trimmed to save bandwidth, maybe too much trimming.
The attributions might now be a little unclear. My apologies. ]
Sender: ne...@fcom.cc.utah.edu
Organization: Weber State University, Ogden, UT
References: <1993Jul27....@spcvxb.spc.edu> <1993Jul28....@fcom.cc.utah.edu> <1993Jul28....@spcvxb.spc.edu>
Date: Thu, 29 Jul 93 01:07:58 GMT
>In article <1993Jul28....@fcom.cc.utah.edu>, te...@cs.weber.edu (A Wizard of Earth C) writes:
...
>> LKM is "Loadable Kernel Modules".. it allows you to load/unload device
...
>> An execution class is basically a program loader. For instance, using code
>> I can't give out, I can run ISC 386 and SCO Xenix binaries on my machine
>> (if I'd spent another week, the Xenix stuff would be distributable) as long
>> as they are statically linked.
>
> This is certainly interesting. What percentage of these binaries are static-
>ally linked? (Out of curiousity - I really don't know the answer)
For SCO Xenix -- all of them. For ISC UNIX 3.1 (the one I was interested
in running binaries from) most of them. I can't distribute the ISC code
because it's derived from some unclean sources. And now that we have a
console driver that emulates the SCO console driver, a little hacking lets
me run the commercial version of WordPerfect on my machine... and SCO
Professional (a Lotus clone) and SCO FoxBase (a DBase III+ clone). I still
...
[End of quoted article]
I guess I'll just be patient... I would have been willing to write a
SCO program loader, but haven't even got a 386 at home yet.. :-(
Little matter of personal finances due to a start up consulting
firm I'm involved with... :-) :-)
>On the other hand i don't realy understand why on earth would one want to
>run DOS/Windows under NetBSD. I don't think that it would be so fast/compatible
DOS will always be a dog to run under unix... DOS has no concept
of being to do things in the background. No concept of asynchronous I/O
completion i.e. it spin loops waiting for I/O completion,
waiting for keyboard ready is brutal. Still, it can be very usable
for running the "dusty deck" old Miserable DOS program, or
the occasional word processing. You aren't going to compile
anything anyway under DOS... You'd be using the native unix.
It also sure beats rebooting under dos (ugh) for some stupid little
thing..
Re-writing the virtual BIOS can help a lot, as anytime a BIOs
trap is taken you can suspend the virtual DOS...
Windows on the other time can be faster under unix as WABI has shown.
The WINE project will result in a PD version of WABI.
It can be argued that the Windows API can't be any worse then the X11
API. :-) ( I have been told that the Windows API "grew" rather than
being designed, so I find that a little hard to believe. :-) )
It is going to take a decent DOS/Windows emulator to wean people
away from an program loader system of the 60's to a real
operating system, of at least the 70's :-)
Of course, the dreaded legal challenge is always possible.... There
is a legal case on in the states that might enable MS to
make WABI illegal.. (As they can argue that the Windows API can't
be copied.)
Volker A. Brandt
>In article <kaleb.7...@kanga.x.org> ka...@expo.lcs.mit.edu
>(Kaleb Keithley) writes:
>>varieties of <xxx>BSD. Or maybe you think the world would be a better place
>>if we could go down to the CheChryForVolksRenVolvFiatToyDatHon dealer and
>>buy a homogenousmobile?
>
>[...] Heck, if you want confusion,
>look up the NEMA standards for power connectors. Even once you pin
>down the voltage and number of phases, there are still half a dozen
>varieties to choose from. :-)
Ever heard of the Europlug? Buy an appliance in France, plug it in in Denmark,
and it will work.
I'd like to see a merger, because then all that talent that is now split apart
could be used to improve one OS. I want shared libs, DMA support for > 16M,
true international keyboards, stable serial I/O, fast ethernet drivers, a
Rock Ridge fs driver, etc etc etc etc.
Since I have neither the knowledge nor the time to do all of these things by
myself, I have to wait until all those kind people out there do it. Having
two BSDs means that I have to wait longer.
Let's work on the Europlug OS.
-- Volker
--
----------------------------------------------------------------------------
Bitnet: UNM409@DBNRHRZ1 Volker A. Brandt
Internet: vol...@sfb256.iam.uni-bonn.de Angewandte Mathematik
Phone: +49 228 73 3427 (Bonn, Germany)
A Wizard of Earth C
>he has SCO emulation running, but can't release the code, cause it has
>some not public stuff in it. Would be great!!! (correct me if i'm wrong)
Before I get deluged with mail, a preemptive strike (sorry if this is terse,
but I have been hit up a lot on these things):
o Yes, I can run SCO Xenix 286 binaries (some ioctl()'s don't work).
o Yes, I can run SCO Xenix 386 binaries.
o Yes, I can run SCO UNIX, ISC UNIX, Microport, Cubix, and Altos
binaries for SVR3 derivitive UNIX.
o Yes, I can run *statically linked* SVR4 binaries.
o Yes, I can run *statically linked* Linux binaries.
o Yes, I have a real streams (some problems with priority banding) that
can run the Lachman TCP/IP code and the sample streams code in the
SVR4 docs.
o Yes, I have real shared libraries not derived from Sun code.
o Yes, I have a user space threads implementation which is source
compatable with Sun's LWP.
o Yes, my console is Unicode and can handle Japanese, Russian, Korean,
etc. with only data set changes.
o Yes, my filesystem is localizable so that you can rename well known
files like "/etc" or "/etc/passwd" to the non-English equivalents
and programs like "passwd" can still find them.
o Yes, my VM has been fixed to eliminate ETXTBSY and all the failure
cases therein (related to NFS, etc).
o Yes, I have a stackable attributed filesystem with data compression
built in on a per file block basis.
o Yes, I have install disks that don't care about translated drive
geometries and *just work* with DOS.
o Yes, I have a DOS FS that can mount extended partitions and doesn't
need changes to the disklabel to find DOS (it reads the partition
table).
o Yes, my serial ports work correctly with flow control.
NO, I CAN NOT GIVE OUT CODE WITHOUT MY EMPLOYERS PERMISSION.
NO, MY EMPLOYER WOULD NOT LOOK KINDLY ON PEOPLE LOBBYING FOR PERMISSION.
NO, I DON'T KNOW WHEN OR IF PERMISSION WILL BE FORTHCOMING.
NO, I DON'T WANT A BUNCH OF MAIL ASKING ME TO "JUST SNEAK A COPY".
NO, YOU DON'T FIND THIS AS FRUSTRATING AS I DO.
[ ... ]
>On the other hand i don't realy understand why on earth would one want to
>run DOS/Windows under NetBSD. I don't think that it would be so fast/compatible
>that serious work could be done under it. (Ready, steady, flame!)
WABI on OS/2 is *faster* than Windows 3.1 by about 30% on the same hardware;
WABI on a box using X for the libs will allow you to run your windows apps
on X terminals on the same screen with UNIX apps, and if DDE is implemented
correctly, cut and paste between Windows apps and X apps.
Terry Lambert
te...@icarus.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Jordan K. Hubbard
Actually, I look at us more as the ERM of the BSD world.
:-)
Jordan
--
(Jordan K. Hubbard) j...@violet.berkeley.edu, j...@al.org, j...@whisker.lotus.ie
I do not speak for Lotus, nor am I even a Lotus employee. I am an independent
contractor.
Vax
Jaye Mathisen <os...@cs.montana.edu> wrote:
>Yes, tape drive stuff is being fixed and worked on, somebody else on
>the hacker list is having the same problem.
I got NetBSD-0.9 and after installation (it's unfortunate restore(1) is not
on any floppy disk images!) and downloading for a full day or two,
NetBSD was up and running, and (would ya believe it?) it read teh tape.
Not that I had much left to restore (I downloaded all the distributions!).
But at least I could restore my mud. I think installing all of NetBSD-0.9
was easier than trying to get those (explicative deleted) 386bsd-0.1
floppies to read the tape. :-)
>>NetBSD, last time I looked (0.8) was a whole bunch of binary data cut up
>>into little chunks - hardly what I would call an easy upgrade from
>NetBSD so far has not focused much on packaging, as opposed to functionality.
And I agree with that theology/philosophy. I evidently didn't see the boot
floppies when I checked out the 0.8 distribution before.
>Well, that's your perogative, have a nice time. I run NetBSD, FreeBSD, and
>Linux all at home, and Linux is nice, but it's very subject to patch-of-the-day
>syndrome. (which is appropriate given the relative age).
>Frankly, I don't see what the big deal is. There are many advantages to
>FreeBSD and NetBSD over 386bsd0.1. Applications port back and forth easily,
>what the hell is the big deal? The only serious difference is in stuff
>in the kernel, and still, what's the big deal? Pick one and run with it.
I'm sure I've pounded this issue into the ground, but when you have over
a thousand hours in a programming project, you'd really like to get it
restored. NetBSD got it, thanks to their fine effort. Although now I'm
going to run off and attempt to make a floppy with restore(1) on it,
and this time I think I'll figure out how to test it. :-)
--
Protect our endangered bandwidth - reply by email. NO BIG SIGS!
VaX#n8 v...@ccwf.cc.utexas.edu - Don't blame me if the finger daemon is down
Marc WANDSCHNEIDER
A Wizard of Earth C <te...@cs.weber.edu> wrote:
>o Yes, I can run SCO UNIX, ISC UNIX, Microport, Cubix, and Altos
> binaries for SVR3 derivitive UNIX.
>o Yes, I can run *statically linked* SVR4 binaries.
>o Yes, I can run *statically linked* Linux binaries.
drooling deleted]
>NO, I CAN NOT GIVE OUT CODE WITHOUT MY EMPLOYERS PERMISSION.
>NO, MY EMPLOYER WOULD NOT LOOK KINDLY ON PEOPLE LOBBYING FOR PERMISSION.
>NO, I DON'T KNOW WHEN OR IF PERMISSION WILL BE FORTHCOMING.
>NO, I DON'T WANT A BUNCH OF MAIL ASKING ME TO "JUST SNEAK A COPY".
>NO, YOU DON'T FIND THIS AS FRUSTRATING AS I DO.
Damn. Now you've gone and made ME frustrated. Sounds like
you're running one HELL of a system....
Toodlepip!
Marc 'em.
--
----------------------------------------------------------------------------
Marc Wandschneider Seattle, WA
"Satan has a stranglehold on my toilet and he won't let go!"
- frm the Weekly World News
Nate Williams
Shawn F. Mckay <sh...@eddie.mit.edu> wrote:
>I can only speak for myself, but I find 386BSD 0.1 (PK 0.2.4)
>to be *VERY* stable. Some folks like to tinker, but dont like
>the idea that one day it may all just BLOW UP / STOP WORKING.
If you like the patchkit, you'll like FreeBSD even more. It
contains the entire patchkit + more bug fixes + a LARGE portion
of updated software.
>Although I see little mail about 386bsd, I see LOTS of people
>complaining about NetBSD, or FeeBSD for one reason or another.
Most of those complaints are from installation problems, and from
problems that would still exist if those folks would be running 0.2.4
systems. (Especially in the FreeBSD arena). Basically, if you run
0.2.4 well, then FreeBSD will run better.
>Its all well and good to be "developing" an operating system, but
>there should be something around for folks like me who like to
>have a reasonable "development" system, while still actully
>USING the O/S.
That is the purpose of FreeBSD, and that is one of the reasons that it's
kernel hasn't changed as much as the NetBSD kernel.
Brett Lymn
Dyane> In article <CDn8x...@sztaki.hu> pi...@fsz.bme.hu writes:
>In article 29...@diana.ocunix.on.ca, d...@diana.ocunix.on.ca (Dyane Bruce) writes:
>>In article <Ygam_WG00...@andrew.cmu.edu> "Alex R.N. Wetmore" <aw...@andrew.cmu.edu> writes:
[..snip...]
Dyane> DOS will always be a dog to run under unix...
Not true. On a commercial unix with a commercial dos-under-unix I
could run four dos tasks at once and still be comparable with a 386/25
running DOS (my machine was a 486/25)
Dyane> DOS has no concept
Dyane> of being to do things in the background. No concept of asynchronous I/O
Dyane> completion i.e. it spin loops waiting for I/O completion,
Dyane> waiting for keyboard ready is brutal.
True but busy waits can be detected and the process priority lowered.
Dyane> Still, it can be very usable
Dyane> for running the "dusty deck" old Miserable DOS program, or
Dyane> the occasional word processing.
Again, on a commercial unix, I *preferred* using Microsoft Word on a
dos-under-unix to using it on a 286 machine because the d-u-u process
was *much* faster.
Dyane> You aren't going to compile
Dyane> anything anyway under DOS... You'd be using the native unix.
Wrong... just think of this: using the unix development environment,
i.e. your favourite editor (vi/emacs/whatever), awk, sed, whatever.
Doing your software build using a unix make file that fires off a dos
process to do the compile and link. I have done this, basically to
avoid having to deal with some compiler author's idea of an integrated
development environment (Wordstar command keys for doing
cut/paste...gimme a break!) AND their brain-dead/naive idea of what a
make should do (like regenerate dependencies *every* time it makes ...
life is too short). By running d-u-u and being able to use the unix
tools I am familiar with I won in terms of being able to get on with
the job instead of struggling with a concept on an editor that hails
from the early '80's.
Dyane> Re-writing the virtual BIOS can help a lot, as anytime a BIOs
Dyane> trap is taken you can suspend the virtual DOS...
Hmmm I think you would have problems with this concept, some DOS
things still expect timely events...
Dyane> It is going to take a decent DOS/Windows emulator to wean people
Dyane> away from an program loader system of the 60's to a real
Dyane> operating system, of at least the 70's :-)
Ooops, I sugest a dose one of the folk-lore newsgroups for you... unix
precedes DOS by a good margin, in fact if you reverse your 60 and 70
you would be closer to the mark.
Dyane> Of course, the dreaded legal challenge is always possible.... There
Dyane> is a legal case on in the states that might enable MS to
Dyane> make WABI illegal.. (As they can argue that the Windows API can't
Dyane> be copied.)
Sun is not small fish so Microsoft may have their hands full trying to
win that case. It will be interesting to watch.
--
Brett Lymn
David Holcomb
& FreeBSD merge, or is this really in the works?
Dave Holcomb
Vax
>accustomed to having certain basic applications working:
>tk, perl, X, emacs, epoch, etc...
>
of Perl I could find (4.036?) compiled for NetBSD (it does alot of
accessing the db files). If I knew where I could upload it, I would.
In addition, same thing goes for bash and ghostsscript.
Caveat: I haven't gotten that crazy XFree86-1.3u1 to work yet,
so I haven't tested out ghostscript. I'd have it print to my HP LJ-2,
but it doesn't seem to want to.
Bruce 'Loodvrij' Keeler
I think I can safely speak for entire comunity when I say
Wow. What a ***BUMMER***!!!!!!!!!!!
> Terry Lambert
> te...@icarus.weber.edu
>---
>Any opinions in this posting are my own and not those of my present
>or previous employers.
Bruce
--
Bruce J. Keeler (907) 337-8193 | "...and Bruce is in charge |
Internet: lood...@cyb.fred.com | of the sheep dip!" |