I've assembled some notes from old manuals and other sources
on the formats used for on-disk file systems through the
Additional notes, comments on style, and whatnot are welcome.
(It may be sensible to send anything in the last two categories
directly to me, rather than to the whole list.)
> SCO is blustering more and more as the open source community exposes
> them for the fruads that they have become.
In the for what it is worth department, I happen to know that this
stuff is more complex than it seems. For instance, I am pretty sure
that ATT should have won their lawsuit over the BSD stuff and if you
doubt that I'd suggest that you go compare the UFS code against the 32v
or v7 code. bmap() is a good place to look. Any suggestions that that
was completely rewritten are patently false, at least in my opinion.
I'm a file system guy, I've done a lot of work in UFS, I'm intimately
familiar with the code. In fact, I defended UFS against LFS when Kirk
wouldn't (LFS is a friggin' joke, any file system hacker knows that the
allocation policy is 90% of the file system).
I do not have knowledge of the code it is that SCO says infringes. And I
think that SCO is about as astute as I am in terms of public relations
(we both tend to be our own worst enemies and I thought I was without
peer in that department :-) But I suspect that there is at least some
merit to what they are claiming. I have to believe that nobody is stupid
enough to have zero data and jump out in public like they are doing.
That's just way too far over the top. Anything is possible I guess,
but doesn't it seem just a little unlikely that a corporation would
commit that public a suicide? I'll probably be proved wrong but I'm
a CEO, running a small company, much smaller than SCO, and there is
no way I'd stick my neck out that far with no data to back it up.
I'd like to think I'm smarter than they are but I tend to doubt it,
they have more experience.
Larry McVoy lm at bitmover.comhttp://www.bitmover.com/lm
On Friday 30 May 2003 11:50 am, Greg 'groggy' Lehey wrote:
> On Thursday, 29 May 2003 at 6:33:54 -0600, M. Warner Losh wrote:
> > In message: <BAFBB8B1.118%rob(a)vetsystems.com>
> > Robert Tillyard <rob(a)vetsystems.com> writes:
> >> I believe the legal action is over breach on contract with IBM and
> >> not on copyright issues.
> > All of SCO's statements to the court have been contractual. Their
> > statements to the press have been inflated to include things that
> > aren't actually alledged in the court filings.
> What's not very clear here is that there seem to be two issues. The
> IBM issue is, as you say, a contractual one which about which they
> have been remarkably vague. The suspension of Linux distribution is a
> different matter. From http://www.lemis.com/grog/sco.html:
> On Tuesday, 27 May 2003, I spoke to Kieran O'Shaughnessy, managing
> director of SCO Australia. He told me that SCO had entrusted three
> independent companies to compare the code of the UnixWare and Linux
> kernels. All three had come back pointing to significant
> occurrences of common code ("UnixWare code", as he put it) in both
> In view of the long and varied history of UNIX, I wondered whether
> the code in question might have been legally transferred from an
> older version of UNIX to Linux, so I asked him if he really meant
> UnixWare and not System V.4. He stated that it was specifically
> UnixWare 7.
> >> But if it turns out the IBM is guilty of lifting SCO code and
> >> putting it into Linux I think SCO does have the right to get a bit
> >> upset about it, after all I wouldn't be to happy if I had to
> >> compete with a product that's just about free and contains code
> >> that I wrote.
> > That's the rub. Do they, in point of fact, actually have any code
> > they own the Copyright to or the patent rights to?
> Of course they have lots of code with their own copyright. The
> release of JFS was one example. Probably the majority of AIX was
> developed by IBM, not by AT&T. It's rather similar to the issue with
> 4BSD in the early 90s: with a little bit of work you could probably
> replace the entire AT&T code in AIX and have a system which did not
> require an SCO license.
I would say that that is entirely likely. AIX was developed by IBM for
IBM-specific machines running in IBM-style environments, and I can imagine
that SysVRx just _doesn't_ _cut_ _the_ _mustard_.
So, SCO's latching on the IBM for Monterey - RS-6000 was 64-bit, or am I
getting confused? - probably gave SCO much more than it gave IBM. So
ironically, if IBM donated stuff to Monterey under the terms of the agreement
and later incorporated the same stuff into Linux, it _could_ look as if they
had taken stuff from SysVRx/Unixware - stuff that SCO had never had the
opportunity to develop if it hadn't been for Monterey and IBM's pre-existing
Just some thoughts - but if that is so, I can see why IBM's not getting too
het up about the whole muck-up.
> If you mean "is there IBM copyright code in Linux?", I think the
> answer is again yes, but it's under the GPL or possibly IPL, IBM's
> attempt at a compromise between proprietary licenses and the GPL. I
> think they've given up on the IPL now.
> For what it's worth, I'd be astounded if SCO's claims were found to be
Mau e ki, "He aha te mea nui?"
You ask, "What is the most important thing?"
Maku e ki, "He tangata, he tangata, he tangata."
I reply, "It is people, it is people, it is people."
Interesting. I suggest everyone interested in this fracas read the
whole scoop at (to repeat Kenneth Stailey's pointer)
Here's a question of interest not to the Linux community but to
the TUHS one: if, as Novell now claim, the 1995 agreement didn't
convey the UNIX copyrights to SCO, under what right did SCO issue
the Ancient UNIX Source Code agreements, whether the restrictive
version of early 1998 or the do-as-you-like Caldera letter of early
2002? Are those agreements really valid?
M. Warner Losh:
There's another article that is saying that there are 10-15
line snippets scattered all through the kernel. Give me a break.
That claim is so absurd as to be not credible on its face. I can see
one or two files, maybe stretching my disbelief to its limits, but I
can't see anything more pervasive than that.
I agree that it sounds unlikely, and I won't give it much credit
until SCO makes its evidence generally available. But it's by no
means absurd. Suppose SCO invented some whizzy data structure and
associated code conventions to afford especially efficient
interprocessor locks. That could show up in fragments scattered
throughout the kernel, and the idea itself could in fact be
valuable intellectual property and the fragments a demonstration
that the idea was stolen.
Or suppose the issue at hand was a particular way to implement a
file system switch. I was involved in adding such a thing to an
old-fashioned kernel myself; it touches many little pieces of
code all over the kernel that happen to do certain things to or
with in-core i-nodes. If I was worried that someone had stolen
such work wholesale, part of what I would look for would indeed
be scattered fragments.
As I say, there's no useful evidence on view at all, therefore
there is no useful evidence that what I am describing is what
the fuss is about.
I have a cont.a.z I would like to extract. When I run it through Solaris
unpack(1) there are no complaints but then I go to unarchive it with either my
3BSD derived ar11 port or Warren's 2.9BSD newoldar and get:
$ file cont.a
cont.a: old PDP-11 archive
$ ar11 tv cont.a
rwx---r-- 2/0 3505 Aug 20 17:07 1976 alog.mat
rw----r-- 2/0 273 Jan 3 05:14 1978 assem
rwx---r-- 2/0 6332 Aug 20 17:07 1976 atan.mat
r-s--x-w- 9/49170995977 Oct 22 01:48 1974 1
ar11: phase error on 1
Same thing with newoldar. I'm thinking Solaris unpack was incompatible with
the pack that was used to make the cont.a.z. Possibly endian issues.
I go poking around for pack(1) in V7 and PWB and 2.11BSD but can't find
anything. Any ideas?
Do you Yahoo!?
Yahoo! Calendar - Free online calendar with sync to Outlook(TM).
In article by Robertdkeys(a)aol.com:
> Warren... is there a non-broken 4.3BSD-Tahoe set somewhere?
As in a bootable 4.3BSD-Tahoe kit? As far as I know, no. The Unix Archive
has a broken copy in 4BSD/Distributions/4.3BSD-Tahoe, indicating that
both usr.tar and src.tar are broken.
It might/should be possible to merge files from the CRSG CD set from Kirk
to recreate these tar files.
Anybody out there have an unbroken Tahoe release?
I just saw this:
This either Very Smart or Very Dumb, I'm not sure which.
I will just point out the recursive conflict when he says
I can't talk about how this information will be applied, nor by
who. You'll have to trust me, or at any rate my record as
ambassador to the community ...
I find it hard to take a secret No Secrets campaign seriously.
If I am to be used as an example of something or to promote some
cause, I think it's only fair that the campaigner tell me just
what he's up to first.
That such a campaign exists in the current context also sounds to
me like an admission that substantial parts of Linux were in fact
lifted directly from a licensed UNIX. That that might be so seems
surprising; that someone would want to prove it was OK even more so.
> I find it hard to take a secret No Secrets campaign seriously.
That says it well.
> That such a campaign exists in the current context also sounds to
> me like an admission that substantial parts of Linux were in fact
> lifted directly from a licensed UNIX. That that might be so seems
> surprising; that someone would want to prove it was OK even more so.
I think it's more of a "How dare anyone even think licensed code was
lifted? We all know better than that. I'm gonna show you that you don't
even have a leg to stand on."
Whatever. Anyone with any sense knows that SCO hasn't got much ground
to stand on. Sadly, "anyone with any sense" likely doesn't cover
the judge and the jury.
Sigh. (Fade to black, as Dionne Warwick sings "Deja Vu" in the
I just saw this:
This either Very Smart or Very Dumb, I'm not sure which. I don't know
that it need be discussed to death on this list, either, but I do figure
that the list members will at least be interested in it.