No subject

No subject

[Laurent]

auth 4ccee312 subscribe 9fans laurent@caladan.fdn.fr

No subject

[bogus]

and architectures.

No subject

[bogus]

system unless I know I can downgrade it if necessary.  I have to be able to
do this quickly.  And no, I do not expect for the upstream maintainers to
do this for me and I have 206 ebuilds in my private overlay (I just counted
them).  

> As nemo points out, "relax".
> 
> EBo just did a very good thing for all of us: 9vx is part of a distro.
> I think he's got some credibility at this point :-)

Thank you Ron for a call for civility.  

Reflecting over all this I think that I do not yet know how to communicate
in the 9fans' language, and that much of the misunderstandings stem from
simple miscommunication.  I'll learn with time, and I hope I will not be to
annoying in the process.  But the couple of times I have seen this kind of
vehemence in the past with other code bases it stemmed from software
systems which had no package management and difficult build/configure
systems.  The end result was that new users would either get discouraged
and drift off, or they would spend literally hundreds of hours to just get
to the point where they can dependably configure, build, and run the code. 
An interesting consequence of this is that any proposed non-trivial change
is met by the old-timers with a resounding "DON'T TOUCH IT!!!"  And there
is good reasons for this -- namely that it took some of these people a year
or more (quite literally) of training and experience to understand how to
maintain the systems.  A significant change will cause them to have to go
back and learn the new system, and they remember what happened the last two
or three times that happened.  I have to wonder if the same this applies to
the Plan 9 community.  If so, the best way to move forward will be to fork
the code.

  EBo --

No subject

[bogus]

ethos. n. the characteristic spirit or attitudes of a community,
people, or system, or of a literary work, etc.

I think Corey=27s use of the word certainly fit that definition, which I
would hope is what most people heard... whether or not one agrees.

K

>>> =22Patrick Kelly=22 <kameo76890=40gmail.com> 25/04/2010 11:22:24 am =
>>>

Ethos (=CE=B8=CE=BF=CF=82) is Greek for the accustomed place; you may have =
meant
ethikos (=E1=BC=A0=CE=B8=CE=B9=CE=BA=CF=8C=CF=82).

No subject

[bogus]

If anybody thinks pcap should be compiled by default, please let me know.

> Now, where do I find Charles' fix for the segmentation error?

I'm trying to find that one too.

-- 
- yiyus || JGL . 4l77.com

No subject

[bogus]

"ack is written purely in Perl, and takes advantage
of the power of Perl's regular expressions."

Forgive my ignorance and irrelevance to this topic,
but what are the advantages of Perl's regular
expressions, over the implementation we have
currently in Plan 9?


Thanks,
ak

On Mon, May 3, 2010 at 5:13 AM, Mathieu Lonjaret
<mathieu.lonjaret at gmail.com> wrote:
> Hello,
>
> just because reviving old threads is fun...
> I've just found out about this:
>
> http://betterthangrep.com/
>
> it does not seem to work out of the box (expecting some unix paths), but
> since there's a perl port and that thing is supposed to be more or
> less self contained (for the standalone version), maybe it's not too
> much work for someone interested enough.
>
> Cheers,
> Mathieu
>
>
>

No subject

[bogus]

I did this @the server
% aux/nfsserver -a tcp!thinktank -c /lib/ndb/nfs
% aux/pcnfsd
% aux/portmapper

@client
# mount -o soft,intr thinktank:thinktank /mnt/nfs
NFS Portmap: RPC: Program not registered

in /sys/log/nfs
thinktank Feb 13 07:07:05 get port
thinktank Feb 13 07:07:32 get port


It seems like I maybe having authentication issue, but not
partifularly sure where to check.
man nfsserver seems to indicate that the client should have a static
ip address is this correct?

the client are using dhcp to configure there ip. I am trying to add
static aliases to clients to see if will work, but wanted to check
everyone else's thought.


-- 
http://www.fernski.com

No subject

[bogus]

pretty quick.

> Coraid's primary business, and certainly the vast majority of their  
> product line, is about SAN storage: providing arrays of storage bits  
> over the network. None of that has anything to do with ZFS. A common  
> thing to do with those bits is create a file system for export (via  
> NFS or whatnot) to other hosts; this is what Coraid's Z-series line  
> does. ZFS is a popular option for how to implement that, and I gather  
> what the Z-series uses internally (or what's NetApp on about?), but  
> certainly isn't the only option. The installation I'm planning for,  
> for example, hasn't been considering ZFS.

So NetApp was basically posturing and ZFS is not fundamental to their
business.  Maybe the publisher would be willing to publish a followup
stating that ZFS is not, and was never, a fundamental part of the product
and that it is only one of something like 50 different file systems.  I get
the feeling that NetApp is afraid of their competitors bring out superior
products.

Anyway, thanks for the info.

  EBo --

No subject

[bogus]

Erik's FS does not take NVR from floppy. That's why Erik
suggested that you create a small 9fat partition (using
the Plan 9 Install/Boot CD) on your primary master HDD
and put the NVR on that. I suppose you should also be
able to put the rest of your files in that partition and just
boot directly from that.


Best,
ak

On Tue, Jan 12, 2010 at 11:31 AM, Wolfgang Kunz <woku at hush.com> wrote:
> Hi,
>
> because I don't have a free hard drive I try to but the file server
> from floppy with the nvram file on this floppy.
>
> This is my plan9.ini
>
> distname=plan9
> bootfile=fd0!dos!9askafs
> nvr=fd!0!plan9.nvr
>
> The boot works until:
>
> spurious interrupt 7, lastintr0
> config diag: unknown type --<d>
> spurious interrupt 7, lastintr0
> panic: bad bootdev:fd
>
> I changed the mkfile for this kernel to include fd support.
> Perhaps nvr on a floppy doesn't works?
>
> Regards,
> Wolfgang
>
>
>

No subject

[bogus]

No subject

[bogus]

cheaper to use than four 48kHz ones.  Smaller chip die area or less
components to attach.

Either way I believe it has to do more with manufacturing costs.

No subject

[bogus]

Thanks,
Roman.

No subject

[bogus]

No subject

[bogus]

please, this is just an uninformed opinion) but perhaps in the future<br>
I&#39;ll be able to see its defects.<br>
<br>
--<br>
<font color=3D"#888888">Hugo<br>
<br></font></blockquote><div><br></div><div>What XML buys you is a bunch of=
 tools that already work for about half of what you&#39;d want a language t=
ool to do. =A0By this I mean there&#39;s parsers already written, and lots =
of libraries to examine the tags at different levels in a hierarchy or stre=
am of XML data. =A0</div>
<div><br></div><div>You still have to define the meaning of each tag, attri=
bute and tagged text. =A0For validation of documents in XML there&#39;s oth=
er XML technologies like Schemas you can use, or you could describe the val=
id set of tags in a DTD. =A0</div>
<div><br></div><div>To add to the madness you can write XML files that tran=
slate XML files to other files (possibly other XML files) in an XML defined=
 language called XSLT. =A0XSLT is a bit like writing in a functional progra=
mming language with the worst syntax possible :-).</div>
<div><br></div><div>The reason I say &quot;worst syntax possible&quot; is t=
hat the amount of typing you&#39;ll do to express something simple in XML i=
s pretty excessive. =A0</div><div><br></div><div>Eventually you&#39;ll find=
 that the entire world became a nail for the XML hammer and that things lik=
e SOAP, XML-RPC, are just not very good due to the fact that sending XML do=
cuments on a wire for simple RPC calls is grossly inefficient, and there&#3=
9;s a lot better technology out there for these sorts of things.</div>
<div><br></div><div>That said, XML is still here, and you kind of have to l=
earn to play ball with it. =A0I just had a discussion with a coworker about=
 a configuration language for a management project here at work and had to =
argue in XML&#39;s defense (customers will more easily understand and be ab=
le to accept an XML language than whatever new cute DSL we come up with). =
=A0</div>
<div><br></div><div>I feel like I need a shower now.</div><div><br></div><d=
iv>Dave</div><div><br></div><div><br></div><div><br></div><div><br></div></=
div><br>

--0016e64f47c6893b71048a437b42--

No subject

[bogus]

do though...

thanks,
mathieu

--upas-biqxuinpowqzzlmjdqnyyupdgz
Content-Type: message/rfc822
Content-Disposition: inline

Return-Path: <mathieu.lonjaret+caf_=mpl=smgl.fr.eu.org at gmail.com>
X-Original-To: mpl at smgl.fr.eu.org
Delivered-To: mpl at smgl.fr.eu.org
Received: by serenity.smgl.fr.eu.org (Postfix, from userid 65534)
	id 230AB32CFD; Thu, 16 Sep 2010 16:01:44 +0200 (CEST)
X-Spam-Checker-Version: SpamAssassin 3.2.5 (2008-06-10) on
	serenity.smgl.fr.eu.org
X-Spam-Level: 
X-Spam-Status: No, score=-2.6 required=1.0 tests=AWL,BAYES_00 autolearn=ham
	version=3.2.5
Received: from mail-ew0-f54.google.com (mail-ew0-f54.google.com [209.85.215.54])
	by serenity.smgl.fr.eu.org (Postfix) with ESMTP id A274D32CDE
	for <mpl at smgl.fr.eu.org>; Thu, 16 Sep 2010 16:01:42 +0200 (CEST)
Received: by ewy22 with SMTP id 22so818495ewy.41
        for <mpl at smgl.fr.eu.org>; Thu, 16 Sep 2010 07:01:42 -0700 (PDT)
Received: by 10.213.4.66 with SMTP id 2mr5402646ebq.17.1284645702206;
        Thu, 16 Sep 2010 07:01:42 -0700 (PDT)
X-Forwarded-To: mpl at smgl.fr.eu.org
X-Forwarded-For: mathieu.lonjaret at gmail.com mpl at smgl.fr.eu.org
Delivered-To: mathieu.lonjaret at gmail.com
Received: by 10.14.47.208 with SMTP id t56cs54718eeb;
        Thu, 16 Sep 2010 07:01:41 -0700 (PDT)
Received: by 10.229.96.84 with SMTP id g20mr2330767qcn.122.1284645700479;
        Thu, 16 Sep 2010 07:01:40 -0700 (PDT)
Received: from gouda.swtch.com (gouda.swtch.com [67.207.142.3])
        by mx.google.com with ESMTP id d42si4767426qcs.144.2010.09.16.07.01.38;
        Thu, 16 Sep 2010 07:01:39 -0700 (PDT)
Received-SPF: pass (google.com: domain of 9fans-bounces+mathieu.lonjaret=gmail.com at 9fans.net designates 67.207.142.3 as permitted sender) client-ip=67.207.142.3;
Authentication-Results: mx.google.com; spf=pass (google.com: domain of 9fans-bounces+mathieu.lonjaret=gmail.com at 9fans.net designates 67.207.142.3 as permitted sender) smtp.mail=9fans-bounces+mathieu.lonjaret=gmail.com at 9fans.net; dkim=pass (test mode) header.i=@gmail.com
Received: from localhost ([127.0.0.1] helo=gouda.swtch.com)
	by gouda.swtch.com with esmtp (Exim 4.69)
	(envelope-from <9fans-bounces at 9fans.net>)
	id 1OwF11-0004fj-G2; Thu, 16 Sep 2010 14:00:43 +0000
Received: from mail-wy0-f177.google.com ([74.125.82.177])
	by gouda.swtch.com with esmtp (Exim 4.69)
	(envelope-from <rminnich at gmail.com>) id 1OwF0z-0004fS-EI
	for 9fans at 9fans.net; Thu, 16 Sep 2010 14:00:41 +0000
Received: by wyb38 with SMTP id 38so1567902wyb.36
	for <9fans at 9fans.net>; Thu, 16 Sep 2010 07:00:35 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma;
	h=domainkey-signature:mime-version:received:received:in-reply-to
	:references:date:message-id:subject:from:to:content-type;
	bh=zjjLdYbJEUGTlf/milDEkAFQlKGdCclI9+hVc0FyBoU=;
	b=c2r8NPjSggAXoTbLP+GwF1z/f+xzSxoS4/fdz4hAq4uS45jWZtAbugNRKU3eVAJ1y5
	BQptwMxcVbYgIs5XT00Upil8Q8CO3yTxKhxRWWAYex8nzGZc3cWGLKJ2FJwJsvCKMVIb
	UxFFXu6es9d7CxnR+p/hcZhnow41LPzb+pQ1w=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma;
	h=mime-version:in-reply-to:references:date:message-id:subject:from:to
	:content-type;
	b=O1vqotm0OusCneHFu7fyYDl6k8aVfdlXF5UXTAxwiees9KVuBHW3m3lvFfTj6Wtd1G
	Ao/9gTgKtgQ4MTgTVaTXjZYMmxz9SJtRe6vrBVUQsudAxtHvO7p3inNf0NHibtjjHiXP
	nZ/K6ftr4jLkzYVy5Tp0h0TSASIqn1t0iaIJk=
MIME-Version: 1.0
Received: by 10.216.17.194 with SMTP id j44mr2777310wej.68.1284645635430; Thu,
	16 Sep 2010 07:00:35 -0700 (PDT)
Received: by 10.216.174.199 with HTTP; Thu, 16 Sep 2010 07:00:35 -0700 (PDT)
In-Reply-To: <1fb06423f1ac109c8ba323c10a5f3b70 at iram.fr>
References: <1fb06423f1ac109c8ba323c10a5f3b70 at iram.fr>
Date: Thu, 16 Sep 2010 07:00:35 -0700
Message-ID: <AANLkTi=m6g3KxftkNXqUd++MHHyKbzKaJ44+A1c9a5Bh at mail.gmail.com>
From: ron minnich <rminnich at gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans at 9fans.net>
Content-Type: text/plain; charset=ISO-8859-1
Subject: Re: [9fans] [OT] CSP in fortran ?
X-BeenThere: 9fans at 9fans.net
X-Mailman-Version: 2.1.10
Precedence: list
Reply-To: Fans of the OS Plan 9 from Bell Labs <9fans at 9fans.net>
List-Id: Fans of the OS Plan 9 from Bell Labs <9fans.9fans.net>
List-Unsubscribe: <http://mail.9fans.net/options/9fans>,
	<mailto:9fans-request at 9fans.net?subject=unsubscribe>
List-Archive: <http://mail.9fans.net/private/9fans>
List-Post: <mailto:9fans at 9fans.net>
List-Help: <mailto:9fans-request at 9fans.net?subject=help>
List-Subscribe: <http://mail.9fans.net/listinfo/9fans>,
	<mailto:9fans-request at 9fans.net?subject=subscribe>
Sender: 9fans-bounces at 9fans.net
Errors-To: 9fans-bounces+mathieu.lonjaret=gmail.com at 9fans.net

I don't like it very much but ... have you looked at openmp? (NOT
openmpi, openmp)

ron

--upas-biqxuinpowqzzlmjdqnyyupdgz--

No subject

[bogus]

Lesson 1: To write predictably fast Haskell -- the kind that competes
with C day in and out
-- use tail recursion, and ensure all types are inferred as simple
machine types, like Int, Word,
Float or Double that simple machine representations. The performance
is there if you want it.

Lesson 2: Laziness has an overhead -- while it allows you to write new
kinds of programs
(where lists may be used as control structures), the memory traffic
that results can be a
penalty if it appears in tight inner loops. Don't rely laziness to
give you performance in your inner loops.

Lesson 3: For heavy optimisation, the C backend to GHC is still the
way to go. Later this
year a new bleeding edge native code generator will be added to GHC,
but until then,
the C backend is still an awesome weapon.

On Mon, Sep 7, 2009 at 2:24 PM, Paul
Donnelly<paul-donnelly at sbcglobal.net> wrote:
> eris.discordia at gmail.com (Eris Discordia) writes:
>
>> I whined about LISP on yet another thread. Above says precisely why I
>> did. LISP is twofold hurtful for me as a naive, below average
>> hobbyist. For one thing the language constructs do not reflect the
>> small computer primitives I was taught somewhere around the beginning
>> of my education. For another, most (simple) problems I have had to
>> deal with are far better expressible in terms of those very
>> primitives. In other words, for a person of my (low) caliber, LISP is
>> neither suited to the family of problems I encounter nor suited to the
>> machines I solve them on. Its claim to fame as the language for
>> "wizards" remains. Although, mind you, the AI paradigm LISP used to
>> represent is long deprecated (Rodney Brooks gives a good overview of
>> this deprecation, although not specifically targeting LISP, in
>> "Cambrian Intelligence: The Early History of the New AI").
>
> Consider that your introduction to Lisp may have been very poor. You're
> right that the mapping from Lisp primitives to machine primitives isn't
> as direct as that in, but Lisp doesn't represent any AI paradigm at all,
> nor a particular programming paradigm, and its name hasn't been written
> in caps for perhaps 30 years. I'm not trying to nitpick; I'm only saying
> that there are a lot of weird ideas about Lisp floating around which a
> person can hardly be blamed for picking up on, and these are the reasons
> it sounds to me like you have.
>
>> One serious question today would be: what's LISP _really_ good for?
>> That it represents a specific programming paradigm is not enough
>> justification.
>
> I think most Lispers would say it's _really_ good for anything but the
> most demanding number crunching, or perhaps A-list games
> programming. Probably you'd run into trouble in some parallel
> programming situations, for reasons more related to implementation
> support and libraries than reasons intrinsic to the language. And the
> justification would be that Lisp is an embarrassingly multiparadigm
> language, as general-purpose as they come.
>
>



-- 
Vinu Rajashekhar,
5th Year Dual Degree Student,
Deptt of Computer Science & Engg,
IIT Kharagpur,
India.

No subject

[bogus]

Which of these goals does the current contrib system meet?
Much as I like the contrib system, it still depends on replica, so,
from my point of view, none of the goals are met.
I once saw it take four hours to install openssl. That's just not workable.

Which of these goals does the gui-based contrib system meet?
This is the system that downloads .iso files and then runs replica
against them. It meets 3, to some extent, but is still too slow for
me; it sort of meets 6; but, unfortunately, it fails on 8.

Which of these goals does my extension meet?
2 -- can download/install all of hg, including all dependencies, in 3
minutes, 2 of which are hget
3 -- .1 seconds for 'deps hg'; .1 second for list packages; < .1
second for list installed
4 -- it knows the dependencies and will install everything with one command
5 -- get and install are seperate commands
6 -- ls /installed
8 -- yes -- /installed/<name> is only created when the package is
completely installed (but there are bugs still)
9 -- yes
10. there is little in the way of a db file, just a /installed
directory (which you get by a bind -a)
11. It's far faster than existing systems because it uses hget

1. is obviously not yet met. I think it would be worth doing a tiny 9,
just as we have tinycore, for terminals.
7. is still not met. Package removal is still a mess. I had hoped to
just mount the .iso's and run the tools out of them but have not
figured out all the issues yet. A simple rbind failed to do the trick.

Here are some examples.

# available packages
term% time list
4th
8169
82563
9load-e820
9win
X11
abaco

(etc.)
0.00u 0.00s 0.11r 	 list

# what packages does hg need?
term% time deps hg
z
bz2
openssl
python
0.00u 0.00s 0.10r 	 deps hg

#install tiff
term% install tiff
package is tiff
Package z already installed, no need to do it
9660srv 1151: serving /srv/tiff
FINIS

#install tiff again
term% install tiff
package is tiff
Package z already installed, no need to do it
Package tiff already installed, no need to do it
FINIS
term%

Sources to these tools, including the build script, are at
http://9grid.net/magic/webls?dir=/rminnich/src/package-tools

You can try them out -- it's all there. Packages are in
http://9grid.net/magic/webls?dir=/rminnich/src/package

I don't pretend the scripts are very good, they just represent a
starting point. Experience (mine) is that the system work well. For
example, just doing:

get openssh
install openssh

takes very little time and has worked reliably for me on 9vx.

And, since I installed hg earlier, openssh install skipped the openssh
install step. Left to the reader (or me in a bit): don't download iso
when the package is installed! -- but it's so fast I have not
bothered.

I'm able to install packages now without worrying about whether I will
be ready to disconnect my laptop and go home before the install is
done!

Next step, if this system is found to be useful, is to adapt fgb's gui program.

ron

No subject

[bogus]

On Thu, Oct 1, 2009 at 11:56 PM, Eric Van Hensbergen <ericvh at gmail.com>wrote:

> Actually, you can specify a depth argument and only get the most recent
> revision (and/or some number of revisions back) -- it does not, however,
> allow you to only grab a subdirectory (that I'm aware of) -- which is why we
> package 9p-sac in a separate repo.
>
>     -eric
>
>
> On Oct 1, 2009, at 1:19 PM, ron minnich wrote:
>
>  On Thu, Oct 1, 2009 at 6:23 AM, Sam Watkins <sam at nipl.net> wrote:
>>
>>  I tried to check out v9fs, but the compressed git repo without checkout
>>> is over
>>> 300Mb.
>>>
>>
>> That's git for you. When you go to git it, you git ALL of it. Kind of
>> like deciding to download sources and getting the entire venti arena.
>>
>> ron
>>
>>
>
>


-- 
Vinu Rajashekhar,
5th Year Dual Degree Student,
Deptt of Computer Science & Engg,
IIT Kharagpur,
India.

--000e0cd1520830b8340474e3e49a
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

No subject

[bogus]

until you forget about it once, and it also results in a huge number of
copies of the system/source laying around.  This is less an issue in this
day and age of cheap disks, but 

> Automated solutions are of course  
> possible, but I don't think there is one which solves conflicts  
> between packages to everyone's satisfaction.

So the question is what functionality are people looking for?


  EBo --

No subject

[bogus]

"4.1 Visualizing trace device output

Once we had the data, we needed a way to analyse the information. After
working with the data for a while, we realized that the output as
shown in Figure
1 would be very useful. No graphiing [sic] tool available to us in
Plan 9 or Linux
was able to create that output. In the end, we determined that gnuplot was the
most appropriate tool, but even then the data required significant processing to
get it into the proper form.

We wrote a suite of scripts usng rc, the plan 9 shell; acid, the Plan
9 debugger;
awk, and sed to generate data appropriate for plotting with gnuplot.
The createplot
script has the ability to filter out functions which ran for less than
a specified number of clock cycles, which is useful for reducing the amount of
noise in a plot. To generate a plot from the data collected earlier, discarding
functions which completed in less than 4000 cycles, we just ran:

    plots/createplot /amd64/9k8pf 4000 ./trace > plotme

and fed the input into gnuplot."

Jason Catena

No subject

[bogus]

please, this is just an uninformed opinion) but perhaps in the future
I'll be able to see its defects.

-- 
Hugo

No subject

[bogus]

authentication sounds more reasonable to me, p9p together with
something like fuse (even together with the new userspace hackery) or
perhaps a single-user v9fs combined with inferno for doing the
auth/crypt work seems a lot more reasonable to me than additional
clever hackery from the plan9 side. Not sure if somebody has something
like this working already...

No subject

[bogus]

the one byte for
each pixel.   If I use 32 bit depth, the one pixel have 4 bytes, and then=
 four windows
are shown.   Each window is the reasult of each byte from the four bytes.=
  =20
Is my guess is wrong?

For nvidia driver, I don't have any idea what's is going.   Is the 9pccd.=
gz=20
of the booting CDROM is differnt from the 9pccd.gz compiled from the sour=
ce tree?

Kenji

No subject

[bogus]

-- 
Greg Comeau / 4.3.10.1 with C++0xisms now in beta!
Comeau C/C++ ONLINE ==>     http://www.comeaucomputing.com/tryitout
World Class Compilers:  Breathtaking C++, Amazing C99, Fabulous C90.
Comeau C/C++ with Dinkumware's Libraries... Have you tried it?

No subject

[bogus]

e number of revisions back) -- it does not, however, allow you to only grab=
 a subdirectory (that I'm aware of) -- which is why we package 9p-sac i=
n a separate repo.<br>
<font color=3D"#888888">
<br>
 =A0 =A0 -eric</font><div><div></div><div class=3D"h5"><br>
<br>
On Oct 1, 2009, at 1:19 PM, ron minnich wrote:<br>
<br>
<blockquote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, =
204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Thu, Oct 1, 2009 at 6:23 AM, Sam Watkins &lt;<a href=3D"mailto:sam at nipl.=
net" target=3D"_blank">sam at nipl.net</a>&gt; wrote:<br>
<br>
<blockquote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, =
204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I tried to check out v9fs, but the compressed git repo without checkout is =
over<br>
300Mb.<br>
</blockquote>
<br>
That&#39;s git for you. When you go to git it, you git ALL of it. Kind of<b=
r>
like deciding to download sources and getting the entire venti arena.<br>
<br>
ron<br>
<br>
</blockquote>
<br>
<br>
</div></div></blockquote></div><br><br clear=3D"all"><br>-- <br>Vinu Rajash=
ekhar,<br>5th Year Dual Degree Student,<br>Deptt of Computer Science &amp; =
Engg,<br>IIT Kharagpur,<br>India.<br>

--000e0cd1520830b8340474e3e49a--

No subject

[bogus]

"To return to Knuth=92s paper: everything there---even
input conversion and sorting---is programmed
monolithically and from scratch. In particular the
isolation of words, the handling of punctuation, and
the treatment of case distinctions are built in. Even
if data-filtering programs for these exact purposes
were not at hand, these operations would well be
implemented separately: for separation of concerns,
for easier development, for piecewise debugging, and
for potential reuse. The small gain in efficiency from
integrating them is not likely to warrant the resulting
loss of flexibility. And the worst possible eventuality
eventuality---being forced to combine programs---is
not severe.

The simple pipeline given above will suffice to get
answers right now, not next week or next month. It
could well be enough to finish the job. But even for
a production project, say for the Library of Congress,
it would make a handsome down payment, useful
for testing the value of the answers and for smoking
out follow-on questions."

Jason Catena

No subject

[bogus]

"ack is written purely in Perl, and takes advantage
of the power of Perl's regular expressions."

Forgive my ignorance and irrelevance to this topic,
but what are the advantages of Perl's regular
expressions, over the implementation we have
currently in Plan 9?


Thanks,
ak

On Mon, May 3, 2010 at 5:13 AM, Mathieu Lonjaret
<mathieu.lonjaret at gmail.com> wrote:
> Hello,
>
> just because reviving old threads is fun...
> I've just found out about this:
>
> http://betterthangrep.com/
>
> it does not seem to work out of the box (expecting some unix paths), but
> since there's a perl port and that thing is supposed to be more or
> less self contained (for the standalone version), maybe it's not too
> much work for someone interested enough.
>
> Cheers,
> Mathieu
>
>
>

--upas-bgrsdpcrjdssrpccswuqevfppf--

No subject

[bogus]

ron

No subject

[bogus]

OS, but I am not able to do so because it lacks the data analysis
tools available in some other systems, like linux.
Because my work involves dealing with data coming from experiments in
astro-particle physics, I am more or less tied to data analysis
software like the R programming language, Python's Numpy, Cern's ROOT
and even gnuplot. While using them, I realized that most of the time I
deal with text files that go here and there as input or output of
small specific programs that perform a given task (I don't know if
this is the result of my Unix/Plan 9 background or just a
coincidence). Say I have a command 'clean' that removes undesired
points from a body of data, and another command 'four' that performs
the FFT; so they are used together as
clean data.txt | four > results.txt
so it occurred to me that one can create single commands to interact
among them to perform some analysis on data, just like in the original
Unix style. Awk can be used as glue among them, with some other small
glue utilities. Plotting data is another thing that I would like to
integrate into this, since plots are quite frequent while analysing
data, but I am not sure how.
Also, something similar to GSL (http://www.gnu.org/software/gsl/)
would be invaluable or maybe even indispensable.
Maybe some day I'll start to write some commands for plan 9 to begin
working on it, but I want to convince myself that this is worth the
time spent.
What do you think of this? my main concern is that perhaps the "do one
thing well" design falls short for data analysis. I've never seen
people work like this on data analysis before (but I do not think I am
the first to do it) because in general, they tend to use large data
analysis frameworks. I'd really appreciate some feedback on this from
people working on data analysis and also from the plan 9 community
(otherwise I wouldn't be writing here :-)
Saludos

-- 
Hugo

No subject

[bogus]

block from a function but you can sort of achieve the same
effect by declaring a global `block variable' and assigning a
block to it -- now you can use this block var elsewhere.

int (^ugly)();

int outer() {
    __block int outer_var = 1;
    ugly = ^{ outer_var = 42; }
}

Presumably __block says outer_var is allocated on heap so now
it can live beyond the life of a particular invocation of
outer().  outer_var will have to be freed when ugly is
assigned to a different block. So it seems GC is a
requirement now.  The original C features were "just right"
to keep the compiler simple and still provide a lot of
expressive power. IMHO GC doesn't fit that model.

Because of the heap allocation, most likely you won't get a
proper closure.  For instance, in Scheme

(define (counter)
  (let ((x 0))
       (lambda () (set! x (+ x 1)) x)))

(define c1 (counter))
(c1) => 1
(c1) => 2
(define c2 (counter))
(c2) => 1
(c1) => 3
etc.

Thus on every invocation of the counter function you get a
fresh counter x and c1 and c2 increment their own copy of x
independently. With blocks you'd render the above as
something like:

typedef int(^ctr_t)();

int counter(ctr_t*c) {
	__block int x = 0;
	*c = ^{ return ++x; }
}

ctr_t c1, c2;

void foo()
{
	counter(&c1);	// presumably taking address of a block var is allowed
	printf("%d\n", c1());
	printf("%d\n", c1());
	counter(&c2);
	printf("%d\n", c2());
	printf("%d\n", c1());
}

I bet you'd get 1 2 3 4, and not 1 2 1 3. If they do get this
right, they'd have to allocate a fresh x on every invocation
of counter and this will add to the memory garbage. Now their
GC problem is even worse!  And I haven't even used
concurrency so far!

They very carefully delineate what you can and can not do
with lexically scoped variables etc. but the model is far
more complex.

> > How much effort would it be to support a feature similar to blocks in  
> > 8c (and family)? What are your thoughts on the idea in general?
> 
> Personally I think you'd be better off exploring a connection that a 
> language called Lua has to C. In the immortal words of Casablanca it
> just could be "the begging of a beautiful friendship".

Amen! Or kill one's prejudice against parens (talk to your
parens!) and use Scheme which is about as simple as it can
get.

No subject

[bogus]

have thousands.  we're currently operating without auth (in part due
to configuration issues), so I don't know how well it will scale.  The
other aspect here is that in current configurations, every "run" has a
different machine configuration based on what you request from the job
scheduler and what you actually get.  We pretty much get different IP
addresses every time, with different front ends, different file
servers, etc. etc.
Again though - the idea is to use file systems more pervasively within
the applications as well -- so there may be multiple file servers per
node providing different services depending on workload needs at the
particular point of computation.  Read our MTAGS paper from last
year's supercomputing conference to get a bigger picture view on how
we view services coming, going, migrating, and adapting to changing
application usage and failure.

      -eric

No subject

[bogus]

Why Desktop Multiprocessing Has Speed Limits
Computerworld (10/05/09) Vol. 43, No. 30, P. 24; Wood, Lamont

Despite the mainstreaming of multicore processors for desktops, not
every desktop application can be rewritten for multicore frameworks,
which means some bottlenecks will persist.  "If you have a task that
cannot be parallelized and you are currently on a plateau of
performance in a single-processor environment, you will not see that
task getting significantly faster in the future," says analyst Tom
Halfhill.  Adobe Systems' Russell Williams points out that performance
does not scale linearly even with parallelization on account of memory
bandwidth issues and delays dictated by interprocessor communications.
Analyst Jim Turley says that, overall, consumer operating systems
"don't do anything smart" with multicore architecture.  "We have to
reinvent computing, and get away from the fundamental premises we
inherited from von Neumann," says Microsoft technical fellow Burton
Smith.  "He assumed one instruction would be executed at a time, and
we are no longer even maintaining the appearance of one instruction at
a time." Analyst Rob Enderle notes that most applications will operate
on only a single core, which means that the benefits of a multicore
architecture only come when multiple applications are run.  "What we'd
all like is a magic compiler that takes yesterday's source code and
spreads it across multiple cores, and that is just not happening,"
says Turley.  Despite the performance issues, vendors prefer multicore
processors because they can facilitate a higher level of power
efficiency.  "Using multiple cores will let us get more performance
while staying within the power envelope," says Acer's Glenn Jystad.

http://www.computerworld.com/s/article/342870/The_Desktop_Traffic_Jam?intsrc=print_latest

No subject

[bogus]

What is the canonical way for added ("opt", "pkg" ?) stuff. Letting
the user adapt his profile to bind the added stuff he wants appearing in
his namespace?

More generally, what is the policy for add-ons? Providing rc(1)
fragments to bind the added stuff?

Cheers,
-- 
Thierry Laronde (Alceste) <tlaronde +AT+ polynum +dot+ com>
                 http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

No subject

[bogus]

information.  It can implement the traditional user-group-world
permissions.  It can implement access control lists.  It can do
a user name translation and say that Bob will always get Alice's
priviliges.  It can do anything it wants, because it's handling
the open request and will either succeed it for fail it and the
client reacts accordingly.

Another thing to note is that every file server can have a different
set of users and groups.  Your fossil file system has one set
of users and groups you've defined.  When you do a 9fs sources,
you attach to another file server with a completely different
set.  In fact, there's no requirement that the intersection of
the sets be non-empty.

Finally, if we try to make the in-kernel file servers borrow
another file server's user/group list, there are some annoying
complications.  If I have several file servers, which user list
do I use?  The first thought would be to have it know about
/adm/users, but each process might have a different, or no,
/adm/users in its name space.  Plus, there's a chicken and
egg problem.  The server which gives you /dev/sd00/nvram
has to approve of the attach when fossil wants to open its
/dev/sd00/fossil, but until fossil has opened it, there's no
way of knowing what's in /adm/users on that particular fossil.

So for in-kernel file servers, it's best to look at them as hostowner
and world and forget about groups.  For lib9p based servers,
you can link in a different implementation of hasperm() and
get whatever permissions checking you want, but the default
behavior is to assume that the named group has exactly one
member: the group leader.

BLS

No subject

[bogus]

"What compiler technology is used to build the compilers?

Gccgo has a C++ front-end with a recursive descent parser coupled to
the standard GCC back end. Gc is written in C using yacc/bison for the
parser. Although it's a new program, it fits in the Plan 9 C compiler
suite (http://plan9.bell-labs.com/sys/doc/compiler.html) and uses a
variant of the Plan 9 loader to generate ELF binaries."

So I believe it could be ported...
As for a comparison to limbo, I think someone who knows more about
limbo would be better qualified to delve into the language, but for
one thing it isn't tied to Inferno, and compiles natively, as opposed
to dis.
(it also has some limited type inference)



--
Rodrigo Miranda

"There is nothing like looking, if you want to find something=85 You
certainly usually find something, if you look, but it is not always
quite the something you were after." =96 J.R.R Tolkien

No subject

[bogus]

problems with that l2 cache.

Here you have the thread:
http://lists.arm.linux.org.uk/lurker/message/20091206.141601.bde66128.en.html

No subject

[bogus]

impression that it'd be better to not produce tokens like INC & DEC in
the (which must then be decomposed into effectively unary operators),
but to only have '+' and '-' tokens and work with them somehow...

Ok. I wrote this because I was suprised that something I had expected
just doesn't work in the way.
I don't think I am capable of actually changing the lexer & parser of
hoc, but if I try and succeed, I'll tell you.

Thanks
Ruda

No subject

[bogus]

1) All "contrib" packages write in /$objtype/bin, /rc/bin and /sys/man 
(or /man, see 3).

2) It's up the user to "bind -c" in /$objtype/bin, /rc/bin and 
/sys/man so that written files end where he wants. Depending on who
installs, so depending on the namespace, all flavors can be achieved
(system wide or individual).

3) Shouldn't /man and /rc/bin be, as /bin, "empty unwritable directories,
place holders", with an initial bind(1) (without "-c"; probably
$objtype agnostic either, since almost all programs are supposed to
exist in an $objtype flavor for man, and should not be a concern for rc)
of /sys/man and (new) /sys/rc/bin there?

Note: in this scheme /rc/bin is more a mean (an undirection) so that rc
programs end in the correct place than something usefull at use time,
since it ends probably bind'ed in /bin.
-- 
Thierry Laronde (Alceste) <tlaronde +AT+ polynum +dot+ com>
                 http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C

No subject

[@ 2000-05-26 15:25 UTC (permalink / raw)]

subscribe jj@comberg.cz

No subject

[Russ]

well that's amusing.
ignore the return address on that.

No subject

[presotto]

I started implementing IPv6 and got deflected doing other stuff.  The
IP stack in the upcoming release is still IPv4 but the interface to
user programs uses IPv6 addresses (with the embedded IPv4 format for
v4 addresses).  It's kind of odd without a v6 stack, but I was hoping
to get back to the v6 implementation and wanted to make sure I didn't
have to go and change all the user level code again.

No subject

[Christian]

subscribe 9fans chris@schwer.bewaff.net

No subject

[Christian]

subscribe chris@schwer.bewaff.net

No subject

[Christian]

subscribe chrisqschwer.bewaff.net

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
Subject: Re: [9fans] administrivia
References: <20000413211101.6535.qmail@g.bio.cse.psu.edu>
In-Reply-To: <20000413211101.6535.qmail@g.bio.cse.psu.edu>
X-loop: root@localhost 

>We've stopped accepting messages from 194.247.215.232 for the
>time being.  That should stop the apparent loop.  I have no
>idea which subscriber (if any) is connected to that address.

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132144.TAA04873@localhost.localdomain>
In-Reply-To: <200004132144.TAA04873@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>References: <200004132131.TAA04637@localhost.localdomain>
>In-Reply-To: <200004132131.TAA04637@localhost.localdomain>
>X-loop: root@localhost 

>>Ups!
>>No such user...

>>mailto: root@localhost
>>----------------------------------
>>--
>>Subject: Re: [9fans] The Disc Company could save you money on your cd's replication
>>References: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
>>In-Reply-To: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
>>X-loop: root@localhost 

>>>This is a multi-part message in MIME format.

>>>------=_NextPart_000_008A_01BF95AE.5869CC80
>>>Content-Type: text/plain;
>>>	charset="iso-8859-1"
>>>Content-Transfer-Encoding: quoted-printable

>>>Hello we are contacting you to let you know about our compact disc =
>>>replication company. We have great pricing and friendly service if we =
>>>can be of service check out our website at WWW.thedisccompany.com or =
>>>give us a call at 1-877-479-1009. If you want a quote you can email you =
>>>needs to sales@thedisccompany.com and one of our courteous staff will =
>>>assist you.


>>>Services:

>>>Compact disc replication

>>>Tapes

>>>Posters

>>>Printed Literature

>>>Bulk CDR

>>>Fulfillment services

>>>Mastering (in-house studio)

>>>Thanks and have a great day.


>>>------=_NextPart_000_008A_01BF95AE.5869CC80
>>>Content-Type: text/html;
>>>	charset="iso-8859-1"
>>>Content-Transfer-Encoding: quoted-printable

>>><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
>>><HTML><HEAD>
>>><META content=3D"text/html; charset=3Diso-8859-1" =
>>>http-equiv=3DContent-Type>
>>><META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR>
>>><STYLE></STYLE>
>>></HEAD>
>>><BODY bgColor=3D#ffffff>
>>><DIV><FONT size=3D2>
>>><P>Hello we are contacting you to let you know about our compact disc=20
>>>replication company. We have great pricing and friendly service if we =
>>>can be of=20
>>>service check out our website at </FONT><A=20
>>>href=3D"http://WWW.thedisccompany.com/"><FONT=20
>>>size=3D2>WWW.thedisccompany.com</FONT></A><FONT size=3D2> or give us a =
>>>call at=20
>>>1-877-479-1009. If you want a quote you can email you needs to </FONT><A =

>>>href=3D"mailto:sales@thedisccompany.com"><FONT=20
>>>size=3D2>sales@thedisccompany.com</FONT></A><FONT size=3D2> and one of =
>>>our courteous=20
>>>staff will assist you.</P>
>>><P></P><U>
>>><P>Services:</P></U>
>>><P>Compact disc replication</P>
>>><P>Tapes</P>
>>><P>Posters</P>
>>><P>Printed Literature</P>
>>><P>Bulk CDR</P>
>>><P>Fulfillment services</P>
>>><P>Mastering (in-house studio)</P>
>>><P>Thanks and have a great day.</P></FONT></DIV></BODY></HTML>

>>>------=_NextPart_000_008A_01BF95AE.5869CC80--

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132144.TAA04891@localhost.localdomain>
In-Reply-To: <200004132144.TAA04891@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>References: <200004132131.TAA04621@localhost.localdomain>
>In-Reply-To: <200004132131.TAA04621@localhost.localdomain>
>X-loop: root@localhost 

>>Ups!
>>No such user...

>>mailto: root@localhost
>>----------------------------------
>>--
>>Subject: Re: [9fans] Video card problem..
>>References: <200004131550.LAA10046@cse.psu.edu>
>>In-Reply-To: <200004131550.LAA10046@cse.psu.edu>
>>X-loop: root@localhost 

>>>If the card is a fairly recent ATI Rage-based one then the 'mach64' code
>>>in the Plan 9 CD distribution won't handle it. We probably have code to do it
>>>but it would require so much extra stuff it would be best to wait for it
>>>to be bundled up somehow. I believe others in the Plan 9 community have Rage
>>>cards working, perhaps someone else can help.

>>>If you look in the BIOS for another "MACH64xx" string near 0xC00F0 the two
>>>letters 'xx' would let me know if it is supported or not.

>>>--jim

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132144.TAA04882@localhost.localdomain>
In-Reply-To: <200004132144.TAA04882@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>References: <200004132131.TAA04646@localhost.localdomain>
>In-Reply-To: <200004132131.TAA04646@localhost.localdomain>
>X-loop: root@localhost 

>>Ups!
>>No such user...

>>mailto: root@localhost
>>----------------------------------
>>--
>>Subject: Re: [9fans] Video card problem..
>>References: <A%jJ4.1949$%L6.118363@monger.newsread.com>
>>In-Reply-To: <A%jJ4.1949$%L6.118363@monger.newsread.com>
>>X-loop: root@localhost 

>>>I have made progress. I was intsalling from 1st edition CD, with 2nd
>>>edition it does install correctly from CD. Now I want to get a better
>>>GUI. The card in the PC is:

>>>ATI 3D Charger IIC/4Mb RAM

>>>card documentation says that it can do dotclock to 80MHz
>>>ramdac is "3D Rage II" at 170Mhz

>>>aux/vga finds a string at

>>>0xC0085="ATI MACH64 BIOS P/N 113-40606-112"

>>>link=vga
>>>ctrl=mach64

>>>if I use the above three lines in /lib/vgadb the screen goes blank and
>>>stays blank.

>>>Any pointers?

>>>- ishwar

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132131.TAA04671@localhost.localdomain>
In-Reply-To: <200004132131.TAA04671@localhost.localdomain>
X-loop: root@localhost 

><<< No Message Collected >>>

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132131.TAA04614@localhost.localdomain>
In-Reply-To: <200004132131.TAA04614@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>Subject: Re: [9fans] Video card problem..
>References: <200004131425.KAA22193@smtp2.fas.harvard.edu>
>In-Reply-To: <200004131425.KAA22193@smtp2.fas.harvard.edu>
>X-loop: root@localhost 

>>  aux/vga finds a string at
>>  0xC0085="ATI MACH64 BIOS P/N 113-40606-112"
>>  link=vga
>>  ctrl=mach64

>>Did you add those lines?  I don't see them
>>in the vgadb on the CD.  Adding support for
>>a new card usually consists of more than just adding
>>the entry to the database -- unfortunately all 
>>Mach 64s are not the same.  Sadly the card is
>>probably just not supported.

>>Russ

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132131.TAA04621@localhost.localdomain>
In-Reply-To: <200004132131.TAA04621@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>Subject: Re: [9fans] Video card problem..
>References: <200004131550.LAA10046@cse.psu.edu>
>In-Reply-To: <200004131550.LAA10046@cse.psu.edu>
>X-loop: root@localhost 

>>If the card is a fairly recent ATI Rage-based one then the 'mach64' code
>>in the Plan 9 CD distribution won't handle it. We probably have code to do it
>>but it would require so much extra stuff it would be best to wait for it
>>to be bundled up somehow. I believe others in the Plan 9 community have Rage
>>cards working, perhaps someone else can help.

>>If you look in the BIOS for another "MACH64xx" string near 0xC00F0 the two
>>letters 'xx' would let me know if it is supported or not.

>>--jim

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132131.TAA04646@localhost.localdomain>
In-Reply-To: <200004132131.TAA04646@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>Subject: Re: [9fans] Video card problem..
>References: <A%jJ4.1949$%L6.118363@monger.newsread.com>
>In-Reply-To: <A%jJ4.1949$%L6.118363@monger.newsread.com>
>X-loop: root@localhost 

>>I have made progress. I was intsalling from 1st edition CD, with 2nd
>>edition it does install correctly from CD. Now I want to get a better
>>GUI. The card in the PC is:

>>ATI 3D Charger IIC/4Mb RAM

>>card documentation says that it can do dotclock to 80MHz
>>ramdac is "3D Rage II" at 170Mhz

>>aux/vga finds a string at

>>0xC0085="ATI MACH64 BIOS P/N 113-40606-112"

>>link=vga
>>ctrl=mach64

>>if I use the above three lines in /lib/vgadb the screen goes blank and
>>stays blank.

>>Any pointers?

>>- ishwar

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
References: <200004132131.TAA04637@localhost.localdomain>
In-Reply-To: <200004132131.TAA04637@localhost.localdomain>
X-loop: root@localhost 

>Ups!
>No such user...

>mailto: root@localhost
>----------------------------------
>--
>Subject: Re: [9fans] The Disc Company could save you money on your cd's replication
>References: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
>In-Reply-To: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
>X-loop: root@localhost 

>>This is a multi-part message in MIME format.

>>------=_NextPart_000_008A_01BF95AE.5869CC80
>>Content-Type: text/plain;
>>	charset="iso-8859-1"
>>Content-Transfer-Encoding: quoted-printable

>>Hello we are contacting you to let you know about our compact disc =
>>replication company. We have great pricing and friendly service if we =
>>can be of service check out our website at WWW.thedisccompany.com or =
>>give us a call at 1-877-479-1009. If you want a quote you can email you =
>>needs to sales@thedisccompany.com and one of our courteous staff will =
>>assist you.


>>Services:

>>Compact disc replication

>>Tapes

>>Posters

>>Printed Literature

>>Bulk CDR

>>Fulfillment services

>>Mastering (in-house studio)

>>Thanks and have a great day.


>>------=_NextPart_000_008A_01BF95AE.5869CC80
>>Content-Type: text/html;
>>	charset="iso-8859-1"
>>Content-Transfer-Encoding: quoted-printable

>><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
>><HTML><HEAD>
>><META content=3D"text/html; charset=3Diso-8859-1" =
>>http-equiv=3DContent-Type>
>><META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR>
>><STYLE></STYLE>
>></HEAD>
>><BODY bgColor=3D#ffffff>
>><DIV><FONT size=3D2>
>><P>Hello we are contacting you to let you know about our compact disc=20
>>replication company. We have great pricing and friendly service if we =
>>can be of=20
>>service check out our website at </FONT><A=20
>>href=3D"http://WWW.thedisccompany.com/"><FONT=20
>>size=3D2>WWW.thedisccompany.com</FONT></A><FONT size=3D2> or give us a =
>>call at=20
>>1-877-479-1009. If you want a quote you can email you needs to </FONT><A =

>>href=3D"mailto:sales@thedisccompany.com"><FONT=20
>>size=3D2>sales@thedisccompany.com</FONT></A><FONT size=3D2> and one of =
>>our courteous=20
>>staff will assist you.</P>
>><P></P><U>
>><P>Services:</P></U>
>><P>Compact disc replication</P>
>><P>Tapes</P>
>><P>Posters</P>
>><P>Printed Literature</P>
>><P>Bulk CDR</P>
>><P>Fulfillment services</P>
>><P>Mastering (in-house studio)</P>
>><P>Thanks and have a great day.</P></FONT></DIV></BODY></HTML>

>>------=_NextPart_000_008A_01BF95AE.5869CC80--

No subject

[root]

<<< No Message Collected >>>

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
Subject: Re: [9fans] Video card problem..
References: <A%jJ4.1949$%L6.118363@monger.newsread.com>
In-Reply-To: <A%jJ4.1949$%L6.118363@monger.newsread.com>
X-loop: root@localhost 

>I have made progress. I was intsalling from 1st edition CD, with 2nd
>edition it does install correctly from CD. Now I want to get a better
>GUI. The card in the PC is:

>ATI 3D Charger IIC/4Mb RAM

>card documentation says that it can do dotclock to 80MHz
>ramdac is "3D Rage II" at 170Mhz

>aux/vga finds a string at

>0xC0085="ATI MACH64 BIOS P/N 113-40606-112"

>link=vga
>ctrl=mach64

>if I use the above three lines in /lib/vgadb the screen goes blank and
>stays blank.

>Any pointers?

>- ishwar

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
Subject: Re: [9fans] The Disc Company could save you money on your cd's replication
References: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
In-Reply-To: <008d01bf95e0$ae46c630$2bcfe4ce@freent.com>
X-loop: root@localhost 

>This is a multi-part message in MIME format.

>------=_NextPart_000_008A_01BF95AE.5869CC80
>Content-Type: text/plain;
>	charset="iso-8859-1"
>Content-Transfer-Encoding: quoted-printable

>Hello we are contacting you to let you know about our compact disc =
>replication company. We have great pricing and friendly service if we =
>can be of service check out our website at WWW.thedisccompany.com or =
>give us a call at 1-877-479-1009. If you want a quote you can email you =
>needs to sales@thedisccompany.com and one of our courteous staff will =
>assist you.


>Services:

>Compact disc replication

>Tapes

>Posters

>Printed Literature

>Bulk CDR

>Fulfillment services

>Mastering (in-house studio)

>Thanks and have a great day.


>------=_NextPart_000_008A_01BF95AE.5869CC80
>Content-Type: text/html;
>	charset="iso-8859-1"
>Content-Transfer-Encoding: quoted-printable

><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
><HTML><HEAD>
><META content=3D"text/html; charset=3Diso-8859-1" =
>http-equiv=3DContent-Type>
><META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR>
><STYLE></STYLE>
></HEAD>
><BODY bgColor=3D#ffffff>
><DIV><FONT size=3D2>
><P>Hello we are contacting you to let you know about our compact disc=20
>replication company. We have great pricing and friendly service if we =
>can be of=20
>service check out our website at </FONT><A=20
>href=3D"http://WWW.thedisccompany.com/"><FONT=20
>size=3D2>WWW.thedisccompany.com</FONT></A><FONT size=3D2> or give us a =
>call at=20
>1-877-479-1009. If you want a quote you can email you needs to </FONT><A =

>href=3D"mailto:sales@thedisccompany.com"><FONT=20
>size=3D2>sales@thedisccompany.com</FONT></A><FONT size=3D2> and one of =
>our courteous=20
>staff will assist you.</P>
><P></P><U>
><P>Services:</P></U>
><P>Compact disc replication</P>
><P>Tapes</P>
><P>Posters</P>
><P>Printed Literature</P>
><P>Bulk CDR</P>
><P>Fulfillment services</P>
><P>Mastering (in-house studio)</P>
><P>Thanks and have a great day.</P></FONT></DIV></BODY></HTML>

>------=_NextPart_000_008A_01BF95AE.5869CC80--

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
Subject: Re: [9fans] Video card problem..
References: <200004131550.LAA10046@cse.psu.edu>
In-Reply-To: <200004131550.LAA10046@cse.psu.edu>
X-loop: root@localhost 

>If the card is a fairly recent ATI Rage-based one then the 'mach64' code
>in the Plan 9 CD distribution won't handle it. We probably have code to do it
>but it would require so much extra stuff it would be best to wait for it
>to be bundled up somehow. I believe others in the Plan 9 community have Rage
>cards working, perhaps someone else can help.

>If you look in the BIOS for another "MACH64xx" string near 0xC00F0 the two
>letters 'xx' would let me know if it is supported or not.

>--jim

No subject

[root]

Ups!
No such user...

mailto: root@localhost
----------------------------------
--
Subject: Re: [9fans] Video card problem..
References: <200004131425.KAA22193@smtp2.fas.harvard.edu>
In-Reply-To: <200004131425.KAA22193@smtp2.fas.harvard.edu>
X-loop: root@localhost 

>  aux/vga finds a string at
>  0xC0085="ATI MACH64 BIOS P/N 113-40606-112"
>  link=vga
>  ctrl=mach64

>Did you add those lines?  I don't see them
>in the vgadb on the CD.  Adding support for
>a new card usually consists of more than just adding
>the entry to the database -- unfortunately all 
>Mach 64s are not the same.  Sadly the card is
>probably just not supported.

>Russ

No subject

[rob]

From: fgergo@eik.bme.hu
Date: Mon Dec 13 11:23:22 EST 1999
To: 9fans@cse.psu.edu
Subject: [9fans] spin/trace source for plan9 scheduler


There is a paper called:

Process Sleep and Wakeup on a Shared-memory Multiprocessor
Rob Pike, Dave Presotto, Ken Thompson, and Gerard Holzmann

Is the spin/trace source still available? It would be a great help, I
would need it for a summary on plan9 at the university.

 thank you very much: Gergo

No subject

[Andrew]

Thanks,
  It was my mistake: I meant to send that message to 9fans-request@...

--andrew


   From: Scott Schwartz <schwartz@bio.cse.psu.edu>

   Andrew Pochinsky <avp@honti.mit.edu> writes:
   | get 9fans.94

   It works better to send that to majordomo.

   It's probably easier to read old message via my web page,
   http://www.cse.psu.edu/~schwartz/9fans/archives/
   (Not indexed, sorry.)

No subject

[Scott]

Andrew Pochinsky <avp@honti.mit.edu> writes:
| get 9fans.94

It works better to send that to majordomo.

It's probably easier to read old message via my web page,
http://www.cse.psu.edu/~schwartz/9fans/archives/
(Not indexed, sorry.)

No subject

[Andrew]

get 9fans.94
get 9fans.940704
get 9fans.95
get 9fans.9501
get 9fans.9502
get 9fans.9503
get 9fans.9504
get 9fans.9505
get 9fans.9506
get 9fans.9507
get 9fans.9508
get 9fans.9509
get 9fans.9510
get 9fans.9511
get 9fans.9512
get 9fans.9601
end

No subject

[presotto]

------ forwarded message follows ------

>From obelix.ee.duth.gr!apostolo Sun Jan 31 04:54:28 EST 1999
Received: from plan9.bell-labs.com ([135.104.9.2]) by plan9; Sun Jan 31 04:54:28 EST 1999
Received: from obelix.ee.duth.gr ([193.92.243.100]) by plan9; Sun Jan 31 04:54:27 EST 1999
Received: from obelix.ee.duth.gr (apostolo@localhost [127.0.0.1])
	by obelix.ee.duth.gr (8.8.7/8.8.7) with ESMTP id NAA32498
	for <presotto@plan9.bell-labs.com>; Sun, 31 Jan 1999 13:01:02 +0200
Sender: apostolo@obelix.ee.duth.gr
Message-ID: <36B437ED.3251F394@obelix.ee.duth.gr>
Date: Sun, 31 Jan 1999 13:01:01 +0200
From: Apostolos Syropoulos <apostolo@obelix.ee.duth.gr>
X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.1.119 i586)
MIME-Version: 1.0
To: presotto@plan9.bell-labs.com
Subject: the fdvol utility
Content-Type: text/plain; charset=x-UNICODE-2-0-UTF-7
Content-Transfer-Encoding: 7bit


   Dear Sirs,

   Instead of using RAWRITE  to copy disk1 - the 1.44MB diskette image -
directly onto a floppy, one may opt to use
FDVOL.EXE a utility available from
ftp://ftp.minix.org/pub/minix/dosutil/fdvol.exe
The utility is being used to create a similar diskette for the
installation of MINIX.

Yours Sincerely,
A.S.

--
****************************************************************
*Apostolos Syropoulos                                          *
*snail mail: 366, 28th October Str., GR-671 00  Xanthi, HELLAS *
*email     : apostolo@obelix.ee.duth.gr                        *
*phone num.: +-30-(0)541-28704                                  *
*home page : http://obelix.ee.duth.gr/+AH4-apostolo                *
****************************************************************

No subject

[arisawa]

Sorry I have mistaken:
>Does someone fix this problem?
Have  someone fixed this problem?
Kenji Arisawa
E-mail: arisawa@aichi-u.ac.jp

No subject

[arisawa]

Hello 9fans
The use of commands:
mv dir1 dir2
sometime makes problematic result,i.e,

term% ls mail
mail/009
mail/m001
mail/m002
mail/m003
mail/m004
mail/m005
mail/m006
mail/m007
mail/m008
mail/m009
term% rm mail/*
rm: mail/009: file does not exist
rm: mail/m001: file does not exist
rm: mail/m002: file does not exist
rm: mail/m003: file does not exist
rm: mail/m004: file does not exist
rm: mail/m005: file does not exist
rm: mail/m006: file does not exist
rm: mail/m007: file does not exist
rm: mail/m008: file does not exist
rm: mail/m009: file does not exist
term% 

Does someone fix this problem?

Kenji Arisawa
E-mail: arisawa@aichi-u.ac.jp

No subject

[arisawa]

>get 9fans 9fans.9807
>get 9fans 9fans.9808
>get 9fans 9fans.9809
Sorry I have mistaken.
Kenji Arisawa

No subject

[arisawa]

get 9fans 9fans.9807
get 9fans 9fans.9808
get 9fans 9fans.9809

No subject

[okamoto]

I don't know how to call this in English, we call it 
茶茶を入れる which means don't take this not so seriously.

>as long as FSF/Linux/FreeBSD stuff is free, and MS stuff >is effectively free, i don't
>think being better or more interesting or whatever helps >one jot. the youth of today,

I'm also a linux user, and unfortunately not so young,
however, I expect Plan 9 something different from those
free unicies.  I don't care how much the Plan 9 is
different from those of today, rather I prefer it,
if it's neccessary from technical view point.

Kenji

No subject

[Elliott]

i think you misattribute the reluctance amongst the youth of your acquaintance to
use Plan 9. i admit there were lots of people who gave up when typing 'vi' didn't
do what they expected, and when they found out that there wasn't anything like pine
and no way to 'talk'. but those were the sort of people 99% of whose computer use
consists of 'pine' and 'talk'. they're also smack-bang in the centre of the MS
target market, so we may as well forget about them.

there was more interest amongst the more technically inclined (though some of them
are damn hard to wean off vi) but as soon as they found that (a) they'd have to pay
to get the source and (b) they couldn't give away their work that was based on Plan 9,
they tended to lose interest. i know that i couldn't afford Plan 9 while i was a
student: it was only after i'd got a job that i forked out for it. i doubt very much
that if i hadn't been at your university, i'd have heard of Plan 9, let alone read
the papers and then bought it.

as long as FSF/Linux/FreeBSD stuff is free, and MS stuff is effectively free, i don't
think being better or more interesting or whatever helps one jot. the youth of today,
for whom i presume to speak, guessing that i'm the youngest 9fan, expects their
software to be free and to come with complete source, should they choose to hack
around with it. they also expect full (and fast) access to changes and the like: not
to be sat around wondering if they'll ever see Brazil.

and then there's the catch-22: the enticement of having users. no-one likes to write
software to push it to the back of their disc to rot (though i do wonder sometimes),
and the one place where you can be sure of users is the free world. if you can't
give your code away for free to run on a free platform, you're going to have fewer
users than if you could. and those users are less likely to be members of your
peer-group, because they, as we know, are all messing about writing emacs macros
and versions of tetris to run on VT100's.

but given that neither Plan 9 nor Brazil will be free, we may as well give up
complaining and be happy that we've seen the light. mind you, the question here
seems to be "will we see the second coming", which is rather appropriate at the
moment.

maybe when the religious nutcases run round killing each other, Plan 9/Brazil will
have its day, like the mammals that had to wait until the dinosaurs were out of
the way before they could come out of their holes and grow to respectable sizes?

-e

No subject

[Tom]

> When I see the codes in rdhtml.c, I found that
> the latin2utf() assumes the runes are latines, not
> more large, say three byte utfs such as Japanese.

I don't think this code makes any assumptions
about the Runes.  It just translates incoming
latin1 text into utf.

> Are there any special purpose to use this coding?

I wrote mothra, and worked on it until I left Bell
Labs two years ago.

I just assumed that since the html standard specifies
that the character set is latin1, I ought to convert
latin1 to utf on input.  Obviously if you prefer to
read JIS, you can change latin2utf appropriately.


-- 
Tom Duff.  It may be inelegant and sluglike, but bloatware sells.

No subject

[Russ]

> For examples,
>   for(l=latin;*l;l++) nutf+=runelen(*l&255);  etc.
>
> Are there any special purpose to use this coding?

Not that I have any real definitive answer, but I can hazard a guess...

A lot of content on the Internet is in Latin1 instead of UTF,
especially at the time mothra was written.

I'm thinking that at the time of the CD distribution (April 1995)
most web servers didn't include character set info in the HTTP
headers, so Latin1 probably seemed like a good default.

You could probably extend the part of mothra that cracks the
headers so that it recognizes the encoding line in the HTTP headers
(Content-Transfer-Encoding: or something similar, I think) and
acts accordingly, probably by pushing it thru tcs.

Russ

No subject

[okamoto]

I asked a few days ago why mothra does not show Japanese text.  After that many negative answers were returned.  Yes, I know I'd better to choose another Web browser.  However, we have no choise than mothra, if we confined ourselves to Plan 9.
Furthermore, if she(Hah?) could display Japanmese, I can use it as an tool to introduce basic usage of Plan 9 system to my users.

Then, I'm still hanging on that problem.

When I see the codes in rdhtml.c, I found that
the latin2utf() assumes the runes are latines, not
more large, say three byte utfs such as Japanese.

For examples,
  for(l=latin;*l;l++) nutf+=runelen(*l&255);  etc.

Are there any special purpose to use this coding?

Kenji

No subject

[Kenji]

> This is a test mail to see what will happen when I send UTF encoded
> mail to this list.  Sorry.

Wow!  no problems!
At my greatest surprise!

Kenji  --from unix mailer

No subject

[okamoto]

This is a test mail to see what will happen when I send UTF encoded
mail to this list.  Sorry.

日本語のメ-ルをktransで書きました.
一体何がおこるのか興味があります.
勿論Plan 9のメ-ラで書いてます.

Kenji

No subject

[Russ]

a slightly more elegant way is to just always do challenge
response, and fire up netkey when you want to use a password.

having su accept a password encourages people to type their
passwords over the network.  if netkey is the only thing
that accepts passwords (aside from the boot process), then
you only have to worry about training users not to run netkey
remotely.

i've been using the following for quite a while.  i think
it's a cross between something i wrote and something tom killian wrote.

it accepts a -n option to say don't reinitialize the namespace,
and a -c option to specify a command to run instead of a shell.

#include <u.h>
#include <libc.h>
#include <auth.h>

int	debug;
Chalstate chal;
char	response[NETCHLEN];
int	nflag;

void
main(int argc, char **argv)
{
	int fd, r;
	char *cmd = nil;

	ARGBEGIN{
	case 'n':
		++nflag;
		break;
	case 'c':
		cmd = ARGF();
		break;
	case 'D':
		++debug;
		break;
	}ARGEND
	if(argc != 1){
		fprint(2, "usage: %s [-n] [-c cmd] user\n", argv0);
		exits("usage");
	}
	if(strcmp(argv[0], "none") == 0){
		fd = open("/dev/user", ORDWR);
		if(fd < 0){
			fprint(2, "%s: can't open /dev/user: %r\n", argv0);
			exits("Sorry");
		}
		r = write(fd, "none", 4);
		close(fd);
		if(r < 0){
			fprint(2, "%s: can't write /dev/user: %r\n", argv0);
			exits("Sorry");
		}
	}else{
		r = getchal(&chal, argv[0]);
		if(r < 0){
			fprint(2, "%s: %r\n", argv0);
			exits("Sorry");
		}
		print("challenge: %s\nresponse: ", chal.chal);
		read(0, response, NETCHLEN-1);
		r = chalreply(&chal, response);
		if(r < 0){
			fprint(2, "%s: %r\n", argv0);
			exits("Sorry");
		}
	}
	if(!nflag)
		if(newns(argv[0], 0)){
			fprint(2, "%s (newns): %r\n", argv0);
			exits("Sorry");
		}
	if(cmd)
		execl("/bin/rc", "rc", "-c", cmd, 0);
	else
		execl("/bin/rc", "rc", "-i", 0);
	fprint(2, "%s: exec /bin/rc failed: %r\n", argv0);
	exits("exec");
}

No subject

[arisawa]

/*
	su: substitute user
	usage: su [user]
	I felt inconvenient to kill processes of none, so I wrote.
	Su is the plan9 version of UNIX su.

	Ref: Much of the codes comes from elsewhere.

	1998/08/27
	Kenji Arisawa (Kenar)
	E-mail: arisawa@ar.aichi-u.ac.jp
*/

#include <u.h>
#include <libc.h>
#include <auth.h>

void error(char *s)
{	fprint(2,"%s\n", s);
	exits(s);
}

int
writefile(char *name, char *buf, int len)
{
	int f, n;

	f = open(name, OWRITE);
	if(f < 0)
		return -1;
	n = write(f, buf, len);
	close(f);
	return (n != len) ? -1 : 0;
}

void
prompt(char *p, char *b, int n)
{
	char *e;
	int i;

	print("%s: ", p);
	for(e = b+n; b < e;){
		i = read(0, b, e-b);
		if(i <= 0)
			exits("hungup");
		b += i;
		if(*(b-1) == '\n'){
			*(b-1) = 0;
			return;
		}
	}
}

int
readln(char *prompt, char *line, int len)
{
	char *p;
	int fd, ctl, n, nr;

	fd = open("/dev/cons", ORDWR);
	if(fd < 0)
		error("couldn't open cons");
	ctl = open("/dev/consctl", OWRITE);
	if(ctl < 0)
		error("couldn't set raw mode");
	write(ctl, "rawon", 5);
	fprint(fd, "%s", prompt);
	nr = 0;
	p = line;
	for(;;){
		n = read(fd, p, 1);
		if(n < 0){
			close(fd);
			close(ctl);
			return -1;
		}
		if(n == 0 || *p == '\n' || *p == '\r'){
			*p = '\0';
			write(fd, "\n", 1);
			close(fd);
			close(ctl);
			return nr;
		}
		if(*p == '\b'){
			if(nr > 0){
				nr--;
				p--;
			}
		}else if(*p == 21){		/* cntrl-u */
			fprint(fd, "\n%s", prompt);
			nr = 0;
			p = line;
		}else{
			nr++;
			p++;
		}
		if(nr == len){
			fprint(fd, "line too long; try again\n%s", prompt);
			nr = 0;
			p = line;
		}
	}
	return -1;
}


/*
 *  authenticate by challenge
 */
int
challuser(char *user)
{
	char nchall[NETCHLEN+32];
	char response[NAMELEN];
	Chalstate ch;
	int m=5; /* max trial */

	while(m-- > 0){
		if(getchal(&ch, user) < 0) return -1;
		sprint(nchall, "challenge: %s\r\nresponse", ch.chal);
		prompt(nchall, response, sizeof response);
		if(chalreply(&ch, response) == 0) break;
	}
	if(m < 0) return -1;
	return 0;
}

/*
*	authenticate by password
*	there may be more elegant method
*/
int passuser(char *user)
{
	char buf[32], passwd[32], key[DESKEYLEN];
	Chalstate ch;
	int n;
	int m=5; /* max trial */
	while(m-- > 0){
		n = readln("Password: ", passwd, sizeof(passwd));
		passwd[n] = 0;
		if(n == 0) return -1;
		passtokey(key, passwd);
		if(getchal(&ch, user) < 0) return -1;
		strcpy(buf,ch.chal);
		n = strtol(buf, 0, 10);
		sprint(buf, "%d", n);
		netcrypt(key, buf);
		if(chalreply(&ch, buf) == 0) break;
	}
	if(m < 0) return -1;
	return 0;
}

int authuser(char *user){
	char *s;
	/* check the host where su is running */
	s = getenv("service");
	if(s && strcmp(s, "terminal") == 0){
		if(passuser(user) < 0) return -1;
	}
	else{
		/* we apply challenge and response */
		if(challuser(user) < 0) return -1;
	}
	return 0;
}


void
main(int argc, char **argv)
{	int f;
	char *user=0;
	argv++;
	if(*argv) user=*argv;
	
	if(user){
		if(authuser(user) < 0) { 
			print("authentication failure\r\n");
			exits("authentication");
		}
		newns(user,0);
	}
	else{
		/* set user to none */
		if(writefile("/dev/user", "none", 4) < 0) exits("/dev/user");
		newns("none",0);
	}

	putenv("prompt", "su% ");

	execl("/bin/rc", "rc", "-i", 0);
}

No subject

[Elliott]

you can get a Unix version of mk from the Inferno download, provided
your Unix is one that you can get Inferno for. alternatively you could
look at unix.c in the mk source.

No subject

[Lucio]

OK, so what's the secret?  Somewhere in my configuration of a CPU and 
authentication server, I have managed to screw up.  Now how do I 
persuade the auth server not to look for itself on the network so it 
can grant itself access to the fileserver?

It eventually gives up and uses its own password, but I dislike waiting 
for it to time out first.  This is not the way "things are meant to be" 
but I can't figure out what it is that I need to change.

Please can some kind soul point me to the right place (the NET db?  I 
don't think I've broken anything in there - the KEY file, how do I 
remove the AUTH portion of it, without destroying it completely?  what 
is the real address for a dummy auth 0.0.0.1, 0.0.1.0?  I have 
conflicting values in various places :-(

Thanks a lot, everyone.

++L

No subject

[Steve]

Subject: [reminder] pointer to Plan 9 FAQ

The Plan 9 faq is posted to comp.os.plan9 at the beginning of each month.
It is also at news.answers archive sites, look for comp-os/plan9-faq

The latest hypertext version of the faq is available at url
http://www.fywss.com/plan9/plan9faq.html

No subject

[arisawa]

Hello 9fans!

The following is a tool to convert dump image file to pic format.
I hope this is more useful than dumppic.

Kenji Arisawa
E-mail: arisawa@aichi-u.ac.jp

--------------------------- dump2pic.c ----------------------
/*
*   dump2pic
*
*   Dump2pic converts dumpfile to picfile
*
*   usage: dump2pic [-m] dumpfile
*   option m produces monochrome output
*   picfile is written to stdout
*
*   Plan9 has similar tool named dumppic,
*   However dumppic does not read headers in dumpfile,
*   and what is worse, dumppic does not put cmap to the output.
*
*   Dump2pic read headers of dumpfile and cmap of
*   current screen, and put these informations to the output.
*
*   1998/03/20
*   Kenji Arisawa
*   E-mail: arisawa@aichi-u.ac.jp
*/

#include <u.h>
#include <libc.h>
#include <libg.h>
#include <fb.h>

/*
*   dump2pic convert dumpfile to picfile
*   reading dumpfile headers
*   putting screen-cmap
*
*   opt:
*       1 for 256 color output
*       0 for monochrome output
*/
void    dump2pic(char *input, char *output, int opt){
    PICFILE *pf;
    Bitmap *b;
    Rectangle r;
    RGB map[256]; /* see: rgbpix(2) */
    uchar cmap[256*3];
    int fd;
    int v;
    
    binit(0,0,0);
    fd = open(input,OREAD);
    if(!fd){
        fprint(2, "%s: file not open", input);
        exits(0);
    }
    b = rdbitmapfile(fd);
    if(!b){
        perror("rdbitmapfile error");
        exits(0);
    }
    close(fd);

    /* we use screen cmap */
    rdcolmap(&screen, map);
    /*
    *   Translate it to cmap
    *   Note that map[v].X is ulong (32 bits)
    *   where X is red, green and blue.
    *   see: /sys/src/fb/getmap.c
    */
    if(opt){/* we use screen-cmap */
        for(v = 0; v < 256; v++){
            cmap[3*v + 0] = (map[255 - v].red >> 24);
            cmap[3*v + 1] = (map[255 - v].green >> 24);
            cmap[3*v + 2] = (map[255 - v].blue >> 24);
        }
    }else {/* we create monochrome cmap */
        uchar lmap[256];
        ulong x, xr, xg, xb;
        /* be careful in overflow */
        for(v=0;v < 256;v++){
            xr = map[v].red >> 16;
            xg = map[v].green >> 16;
            xb = map[v].blue >> 16;
            x = (xr*299+xg*587+xb*114)/1000;
            /* 
            *   I found these weighting parameters
            *   in /sys/src/fb/pic2gif
            */
            lmap[v] = x >> 8;
        }
        for(v = 0; v < 256; v++){
            cmap[3*v + 0] = lmap[255 - v];
            cmap[3*v + 1] = lmap[255 - v];
            cmap[3*v + 2] = lmap[255 - v];
        }
    }

    r = b->r;
    pf=picopen_w(output, "runcode", screen.r.min.x,
        screen.r.min.y, Dx(r), Dy(r), "m", 0, (char*)cmap);
    if(pf==0){
        fprint(2,"can't create %s\n", output);
        exits("create");
    }
    if(wrpicfile(pf, b)<0){
        fprint(2, "%s: can't save\n", output);
        exits("output error");
    }
    picclose(pf);
    bfree(b);
    bclose();
}

void main(int argc, char *argv[]) {
    int opt;
    opt = 1;
    argc = getflags(argc, argv, "m:0");
    if(flag['m']) opt = 0;
    if(argc != 2) usage("dumpfile");
    dump2pic(argv[1], "OUT", opt);
}

No subject

[arisawa]

Hello 9fans!

A patch to 8.5 window is posted to:
ftp://plan9.aichi-u.ac.jp/9win.bod

This patch enables the grab.

Grab enables to copy the image in rectangle of screen.
The output is created in /tmp with the name grab.3c.
You can see the result by: fb/9v /tmp/grab.3c

Click right mouse button, then you will see new menu item "Grab".
Select Grab then button operations are as follows:
Left button determines the rectangle to cut.
Middle button and rifht button work the same;
they show menu: OK, Redo, Abort

1998/03/20
Kenji Arisawa
E-mail: arisawa@aichi-u.ac.jp (Kenji Arisawa)

No subject

[forsyth]

>>Any comments?

it's hopeless trying to offer advice without knowing
what you're actually trying to achieve at the application level
(except that i can already say i really don't approve of the proposed
change to create).

what's the aim?

No subject

[Mark]

gdb@dbsystems.COM (G. David Butler) writes:
> Mark Salyzyn <mark@bohica.net> writes:
> >The DPT controller is the only one that supports RAID, with the added bonus
> >of up to 64MB of Cache.
> RAID is ok, I guess, but Cache that isn't battery backed up or
> doesn't follow the OS's sense of "write this data now, really"
> is a BAD thing.

There is a philisophical problem with battery backup Cache, that is for sure.
DPT (internally) has a Battery Backed up PM3334 card, but none of our OEMs
want it ...

Sync (command 35 in the SCSI command set) flushes cache on both the card and
the drives. This is the command that is issued by our other drivers to ensure
clean shutdowns, critical data paths and just before a SCSI bus reset.

As for the Battery Backup issue, one will need to ensure that there is
continuous power to the drives as well since their sense of Cache also
gets lost during power outages and SCSI Bus resets. The consensus is that
since the drives have their own problems with Cache, that a UPS is a far
better investment than YABRS (Yet Another Battery Replacement Schedule) for
the server market we generally target our controllers for.

The Battery backup card, if ever released, will be in the higher end PCI
or I2O product lines.

Disclaimer: This is my opinion and does not reflect DPT's opinion.

>>All the above DPT PCI cards do circles around the 154X series of cards. In
>>fact, our lowest end ISA based controllers (I know, John's driver doesn't
>>support them, but it can since there is no difference to the interface
>>ports and command sets if you force the I/O addresses) are 40% faster than the
>>2940s and none of our cards are as slow as the 154X's as well!
> I'm interested.  Is the PM2041W a ISA/DMA busmaster?

Yes, suffering from the same ills as all ISA cards, the 16MB limit. John did
not set up any bounce buffer technology in his driver, once that is done if
the system has more than 16MB of memory, forcing the address, to lets say 01F0,
will work.

Your mileage will vary, as my comparison was a direct result of the BSDi
BSD/OS 2.1 driver.

> How many can I put in one machine (i.e. which DMA channels and IRQ's
> does is support)?  What is the DMA transfer speed?

DPT ISA cards can be placed at 01F0, 0170, 0230 and 0330. However, they only
have three possible interrupts (12, 14 and 15) limiting you to three ISA
cards in a system (some motherboards allow for ISA interrupt sharing through
a wired-or gating of the interrupt lines).

> I will try to get a technical reference manual on the PM2041W,
> and if successfull, I will give the controller a try.

Shhhh, you may wish to indicate that you intend to evaluate the card and
be modifying a driver to run with this card and that your evaluation will be
posted on the worldwide plan 9 discussion list ...

Sincerely -- Mark Salyzyn

No subject

[Rob]

A draft of the chapter discussing low-level graphics using the Draw module
is now available in HTML under
	http://inferno.bell-labs.com/inferno/book

Rob Pike
Howard Trickey

No subject

[FODEMESI]

get 9fans 9fans.9708

No subject

[Kenji]

subscribe

No subject

[Kenji]

index

No subject

[schwartz]

approve robby.the.robot unsubscribe 9fans taweili@usc.edu

No subject

[bobf]

i did not say this, someone else apparently did.

>From: bobf@plan9.bell-labs.com
>>> Following on from everyone agreeing that alef is the right way to write
>>> applications in Plan9

No subject

[bobf]

> Following on from everyone agreeing that alef is the right way to write
> applications in Plan9 and issues of porting your applications to other
> enviroment, how difficult would it be to port the IRIX version of the alef
> runtime and compiler to other posix'ish platforms for example. Does the IRIX
> version use lots of specific SGI features or given a resonable ammount of time
> would it be possible to port it ?

first of all, the Irix version of alef isn't a port; it is cross-built
from plan 9.  that is, i use the plan 9 compilers and loaders to produce
an Irix executable.

that said, "porting" the compiler and loader is the easy part.  since the
compiler and loader generate plan 9-style symbol tables, the debugger on the
target system probably won't understand the resulting executable.
you'll also have to bring along acid and some other utilities, like ar.
to get acid, you need to emulate the portion of libmach for
the target architecture and figure out how to map the plan 9 process
control model to that of the target system.  i had access to the Irix
source when i didit, so although i couldn't modify the operating system,
i was able to find enough rocks to hide process state under.  even then, the Irix
version of acid lacks some of the functionality of the plan 9 version.

of course, if you don't need a debugger for your alef programs, it should
be relatively easy to build the alef compiler and run-time library; mainly
you need to understand the target system's system call conventions and
multi-threading package.  the latter will have to be ported to plan 9.

if you want to truly port the alef compiler, then i can't be of much help
because we haven't done it yet.

No subject

[tab]

I have an Adaptec 1542CP controller that doesn't work with
the fileserver (9pcfs).  It seems to work OK with the
cpu server.  I see that their are some modifications to the
driver for the cpu server that haven't been made to 9pcfs.

Has anyone already fixed this problem?
(I thought I'd asked before I dive in)

Thanks,
Tom Bohannon
Cisco Systems, Inc.

No subject

[tab]

I'm trying to build a new set of distribution flops
from the update plan9 stuff.  The script calls
'vdsqueeze' which I can't find anywhere.  Any clues
where I can find this animal?

thanks,
Tom Bohannon
Cisco Systems, Inc.

No subject

[beto]

In <199707292134.RAA07755@cse.psu.edu>
 seanq@plan9.bell-labs.com wrote:

>
> For those of you that use Sam on Windows 95/NT:
> There is a new version which can be installed by running
> the self extracting executable available thru netlib.

what about the plan9 terminal emulator for Windows 95/NT? is it
available in the web site? It would be a very good hit if
people could connect to plan9 cpus using
win/nt boxes.

I saw it working last year so I know it's out there.

No subject

[seanq]

For those of you that use Sam on Windows 95/NT:
There is a new version which can be installed by running
the self extracting executable available thru netlib.

Go to
	http://netlib.bell-labs.com/netlib/research
and select the file sam.exe, or more directly,
	http://netlib.bell-labs.com/netlib/research/sam.exe

Major changes from the previous release include:
*) The commands that run external programs now work, i.e. '<', '>', '|', '!'.
   Included with the distribution are several command line programs
   that are useful with sam, such as: pwd, echo, fmt, and sort.
*) There is a 'B' command that enables sam to receive file open requests
   from other programs.

Let me know any problems you have.
seanq@research.bell-labs.com

No subject

[Scott]

rob@plan9.bell-labs.com writes:
| I put screen blanking portably into Brazil by just setting
| the color map to all zeros after a quiet period.  Mouse action
| restores it.

That doesn't achieve the same effect, because it doesn't invoke
the VESA power saving mode.  Modern monitors watch the HSYNC and
VSYNC signals; if one stops they blank the screen in a moderate
power saving mode, and if both stop the monitor goes to very low
power mode.

-- Scott

No subject

[rob]

I put screen blanking portably into Brazil by just setting
the color map to all zeros after a quiet period.  Mouse action
restores it.

-rob

No subject

[Scott]

Here are some suggested changes to devvga to attempt to support screen
blanking, in the style of XFree86 3.3.  (Doing this completely right
requires chipset specific code, and more extensive changes.)  The
generic vga mechanisms could probably be used via aux/vga, but
/dev/vgactl is such a nice interface that I couldn't resist using it.
(Is that the wrong approach?)

-- Scott

diff devvga.c- devvga.c
292a293,329
> vga_blank(int mode)
> {
> 	uchar clocking_mode, mode_control;
>
> 	if (mode == 0) {
> 		clocking_mode = 0;
> 		mode_control = Not_Hardware_Reset;
> 	}
> 	else if (mode == 1) {
> 		clocking_mode = Screen_Off;
> 		mode_control = Not_Hardware_Reset;
> 		/* Inspired by xfree86 3.3 vgaHW.c */
> 	}
> 	else if (mode == 2) {
> 		clocking_mode = 0;
> 		mode_control = 0;
> 		/* Wild guess.  Probably not useful. */
> 	}
> 	else if (mode == 3) {
> 		clocking_mode = Screen_Off;
> 		mode_control = 0;
> 	}
> 	else {
> 		SET(clocking_mode);
> 		SET(mode_control);
> 		error(Ebadarg);
> 	}
>
> 	vgaxo(Seqx, Sync_Reset, Reset_Start);
> 	vgaxo(Seqx, Clocking_Mode,
> 		clocking_mode | (vgaxi(Seqx, Clocking_Mode) & ~Screen_Off));
> 	vgaxo(Crtx, Mode_Ctrl,
> 		mode_control | (vgaxi(Crtx, Mode_Ctrl) & ~Not_Hardware_Reset));
> 	vgaxo(Seqx, Sync_Reset, Reset_End);
> }
>
> static void
305c342,360
< 	if(strcmp(field[0], "hwgc") == 0){
- ---
> 	if(strcmp(field[0], "screen") == 0) {
> 		int mode = 0;
>
> 		if (strcmp(field[1], "on") == 0)
> 			mode = 0;
> 		else if (strcmp(field[1], "standby") == 0)
> 			mode = 1;
> 		else if (strcmp(field[1], "suspend") == 0)
> 			mode = 2;
> 		else if (strcmp(field[1], "off") == 0)
> 			mode = 3;
> 		else
> 			error(Ebadarg);
>
> 		/* if (vgac->blank) (vgac->blank)(mode); else */
> 		vga_blank(mode);
> 		return;
> 	}
> 	else if(strcmp(field[0], "hwgc") == 0){
1187a1243
>

diff vga.h- vga.h
31a32,43
> enum {
> 	/* Seqx */
> 	Sync_Reset = 0x0,
> 		Reset_Start = 0x1,
> 		Reset_End = 0x3,
> 	Clocking_Mode = 0x1,
> 		Screen_Off = 0x20,
> 	/* Crtx */
> 	Mode_Ctrl = 0x17,
> 		Not_Hardware_Reset = 0x80,
> };
>

No subject

[ken]

> As I recall, 8l doesn't use malloc and free; it just allocates
> space on the heap and never frees it.  It tends to walk some
> fairly large lists multiple times, which adds to the thrashing
> behaviour when there isn't enough RAM.  I remember being told
> that the justification for this was efficiency; when you have
> the RAM, it is a lot faster to avoid the extra overheads of
> using a real allocator, or so the argument goes.

> I'm starting to wonder if it mightn't be a bad idea to try and
> change the way 8l does its allocation, to make compiling
> these programs possible on machines which don't have huge
> amounts of RAM.  Has anyone else looked at the linker internals
> and have an opinion on this?  It's been a while since I looked
> in there myself, so I don't have a really good feel for how
> much memory is being wasted, or how much locality of reference
> there is, or indeed how easy it would be to figure out
> what needs to be freed...

8c allocates memory and doesnt free it when unused. 8l doesn't.
almost all of the allocated memory in 8l is used in a linked
list of instructions. i doubt if you will save any space by
calling malloc and free.

if there is any space to be had, it is from rearranging the
structures Adr and Prog (which contains 2 Adr's) to be
more specific to individual instruction types. (ie most
instructions have only one full address with the other
address a register.)

No subject

[Anthony]

It tried to get the developer manual for the 2940, but those chumps from
adapted are really fucked up. I am gonna call up marketing and shout a
little a bit maybe it'll helps.

Is there any support for the following buslogic card?

BusLogic Model BT-958 PCI Wide Ultra SCSI Host Adapter
Firmware Version: 5.06I, I/O Address: 0xD800, IRQ Channel: 10/Level

--
Anthony C. Zboralski ACZ3 <frantic@sct.fr>
Immunis, 24, rue Vieille du Temple, 75004 Paris
Phone: +33 1 44 545 535, Fax: +33 1 42 775 649
KeyID 1024/ED8D8A39 
Key fingerprint = C5 27 9A 0C 56 30 10 F9  9D 54 EE DB 2C 14 2A 78

No subject

[forsyth]

>>If it really is just an evolution of plan9, then maybe the best option
>>would be offer upgrades to Inferno for Plan9 licensees, at a reduced

the source for Plan 9 is published `like a book'.
the source for Inferno most certainly is not.

No subject

[jm]

As you probably know:
	announce("tcp!mymachine!1234", adir)
won't work ; one should use instead :
	announce("tcp!*!1234", adir)
I think that the announce man page should mention that,
even if the question is with cs.

No subject

[bobf]

> I, too, am curious about the disposition of Brazil.  I actually got
> some funding approved for Plan 9 here at MIT (I'm part of a student
> ...
> Is there an `official' answer?  Is there an unofficial answer from
> someone who has put more effort than I am into trying to get an
> official answer?
 
here's an unofficial answer.
 
we've been working on other things for the past year or so,
so there has been little new work on brazil.  phil and rob
and dave say that they want to get back to working on
brazil, but we don't know when that will be.

our plan 9 system is only used to access the old worm for archival
purposes.  our new file server contains only the brazil source
tree, so brazil is our current development system.  brazil is not
in a state that is releasable, so i doubt that there will be
a brazil release in the foreseeable future.

of course, all of this could change in a wink.

No subject

[presotto]

Russ Cox made it easier to add new keyboard sequences to
plan 9.  The pertinent files are:

ftp://plan9.bell-labs.com/plan9/update/cmd/aux/mklatinkbd.c
frp://plan9.bell-labs.com/plan9/update/9/port/858968110.rc

Here's his mail explaining it:

  Subject: /lib/keyboard and latin1.c

  A while ago, Rob and I exchanged the attached email.
  It's a shame that editing /lib/keyboard doesn't work, so
  I fixed that.  Take a look at ls /usr/rsc/kbd/mklatinkbd.c.
  The way I've installed it at home is to change latin1.c to read

	struct cvlist
	{
		char	*ld;		/* must be seen before using this conversion */
		char	*si;		/* options for last input characters */
		Rune	*so;		/* the corresponding Rune for each si entry */
	} latintab[] = {
	#include "latin1.h"
		0,	0,		0
	};

  and then have mkfile generate latin1.h with 
	aux/mklatinkbd /lib/keyboard >latin1.h

  As an aside, I found two ``bugs'' in the keyboard file
  as shipped on the CD.  The sequences <= and => are used
  for both ≤, ≥ and ⇐,⇒, according to /lib/keyboard.
  It looks like they've been fixed, at least on emelie
  (haven't checked elsewhere).  The program will catch
  duplicate sequences.

    From rsc Fri Oct 25 08:36:01 1996
    Subject: re: /lib/keyboard

    how does one add more compose sequences
    in plan 9?  i tried editing /lib/keyboard
    and rebooting, but that did not work.

      From rob Fri Oct 25 09:02:01 1996
      Subject: re: /lib/keyboard

      unfortunately, it's not nearly that easy.  you need to edit
      /n/bootes/sys/src/9/port/latin1.c.  you just edited the
      documentation, a rarely successful method of change.

No subject

[presotto]

contains a fix to arpd.c.  We had a stupid mistake that
kept the resent packets from getting out of the machine
after the arp was received.

No subject

[bobf]

> Is the rest of the road data still available by ftp?  I was looking for
> the data for PA, if possible.

it should be out there now. 

No subject

[tab]

I'm looking for a POP server for plan9.  Any clues?

Tom Bohannon
Cisco Systems
tab@cisco.com

No subject

[Scott]

I noticed today that latex gets an fp exception on the following input
(derived from a much more interesting document, of course):

\documentstyle{article}
\begin{document}
Troff!
\filbreak
\eject
\end{document}

On an x86, the fault says:
  14097: math coprocessor	vlistout+0x49f	MOVBZX	0x5(AX),AX
which acid reports as:
  At pc:0x0001a9e6:vlistout+0x49f /sys/src/cmd/tex/web2c/tex/tex4.c:937
On a sparc the fault is reported as happening in the preceeding call to [z]round().

So, before I plunge into debugging this, has anyone already fixed it?

No subject

[presotto]

In case anyone doesn't know, the name plan9.att.com
is now obsolete.  plan9.bell-labs.com is the working
name.

No subject

[Steve]

> Has anyone ported the C compiler to the PowerPC processor?

yes. see the faq at
http://www.ecf.toronto.edu/plan9/plan9faq.html

No subject

[tab]

Has anyone ported the C compiler to the PowerPC processor?
Just thought I'd check before we ported it.  Any information
will be most helpful.

thanks,
Tom Bohannon
Cisco Systems, Inc.
tab@cisco.com

No subject

[tab]

I found a bug in lpdaemon.c ...
When if-defed for Plan9, the TMPDIR is defined as follows: 

	#define TMPDIR "/sys/lib/lp/tmp"

In the function 'tempfile()' an array is defined as:
	char tmpf[20];	
and later used in:
	sprintf(tmpf, "%s/lp%d.%d", TMPDIR, getpid(), tindx++);

The resulting string overruns the 'tmpf' buffer - it really 
needs to be at least 32 bytes.

later,
Tom Bohannon 
Cisco Systems, Inc.

No subject

[tab]

I'm trying to hang a HP LaserJet 4 printer off the LPT1 port
of my CPU/Authentication server (133MHz Pentium) but can't
seem to make it work.  When trying to 'lp' from a terminal,
I get a timeout.  I found the following comments in /sys/lib/lp/bin/lpsend.rc
to be disturbing...

># tcp file transfer is broken on plan 9. philw thinks its a flow control problem
># try not to use tcpgate for a while. 9201021250
># tcp file transfer still broken. 9201171404

I've grepped the CHANGES.TXT file on plan9.bell-labs.com/plan9/update
and don't see any obvious fixes to this problem.

Has the problem been fixed by anyone?

Thanks,
Tom Bohannon
Cisco Systems, Inc.
tab@cisco.com

No subject

[forsyth]

i've got a pop3 implementation i can (will) give him.
it's smaller than the bsd one, but still too much for such
a simple protocol.

No subject

[presotto]

We're running our web sites (plan9.bell-labs.com and
inferno.lucent.com) on Brazil, the internal version of
Plan 9.  Sean Dorward wrote the code here.  We haven't
done anything that shouldn't run on the released version.
We should be able to give you the current versions of
the http daemon, dns, and mailer if you have a plan 9 license.

We don't have disk quota's, ken just yells at anyone that
hogs the jukebox.  You'll have to change the file system
to support them.  Alternately, you could have
a daemon that sweeps the system enforcing after the fact
or hire ken.

Both of our servers are PC's.  The more active is a 60 MHZ
pentium.  It averages about 10000 hits a day without a noticable
load.  I have no idea what it takes to support 2000 users.

What would you need for mail?  We don't support POP3 though it'ld
be trivial to give the mailer a POP3 interface, i.e., it's not that
different than the interactive one.

------ forwarded message follows ------

>From cse.psu.edu!owner-9fans Sun Dec  1 09:51:32 EST 1996
Received: from cse.psu.edu by plan9; Sun Dec  1 09:51:32 EST 1996
Received: from localhost (majordom@localhost) by cse.psu.edu (8.7.5/8.7.3) with SMTP id JAA14030; Sun, 1 Dec 1996 09:31:51 -0500 (EST)
Received: by claven.cse.psu.edu (bulk_mailer v1.5); Sun, 1 Dec 1996 09:30:51 -0500
Received: (from majordom@localhost) by cse.psu.edu (8.7.5/8.7.3) id JAA13997 for 9fans-outgoing; Sun, 1 Dec 1996 09:30:44 -0500 (EST)
X-Authentication-Warning: claven.cse.psu.edu: majordom set sender to owner-9fans using -f
Received: from tao.ai.univ-paris8.fr (tao.ai.univ-paris8.fr [192.33.156.85]) by cse.psu.edu (8.7.5/8.7.3) with ESMTP id JAA13993 for <9fans@cse.psu.edu>; Sun, 1 Dec 1996 09:30:40 -0500 (EST)
Received: from colombie.ai.univ-paris8.fr (colombie.ai.univ-paris8.fr [192.33.156.83]) by tao.ai.univ-paris8.fr (8.7.5/8.7.3) with ESMTP id PAA09999 for <9fans@cse.psu.edu>; Sun, 1 Dec 1996 15:22:21 +0100 (MET)
From: Jean Mehat <ai.univ-paris8.fr!jm>
Received: (from jm@localhost) by colombie.ai.univ-paris8.fr (8.7.5/8.7.3) id PAA11560 for 9fans@cse.psu.edu; Sun, 1 Dec 1996 15:24:53 +0100 (MET)
Date: Sun, 1 Dec 1996 15:24:53 +0100 (MET)
Message-Id: <199612011424.PAA11560@colombie.ai.univ-paris8.fr>
To: cse.psu.edu!9fans
Sender: cse.psu.edu!owner-9fans
Reply-To: cse.psu.edu!9fans
Precedence: bulk


What do you, plan9 experts, think of running a large Web site on
a plan9 network of machines ? The site is www.mygale.org, it contains
approximately 2000 accounts. Currently, it runs on a Sparc with NetBsd.

How much work do you think is it to make things run smoothly?
What impact on performances?
How enforce the 5Mbytes per user quota?
What about the plan9 mail delivery system?

No subject

[Jean]

What do you, plan9 experts, think of running a large Web site on
a plan9 network of machines ? The site is www.mygale.org, it contains
approximately 2000 accounts. Currently, it runs on a Sparc with NetBsd.

How much work do you think is it to make things run smoothly?
What impact on performances?
How enforce the 5Mbytes per user quota?
What about the plan9 mail delivery system?

No subject

[presotto]

I put the sleep/wakeup changed and the pexit fix
from ncube in 

http://plan9.bell-labs.com/plan9/update/9/port/845586056.rc
and
http://plan9.bell-labs.com/plan9/update/9/port/845586092.rc

these changes to pertain to everyone.

No subject

[presotto]

Ncube has had problems with the fact that proc.c requires all
Rendez structures to not be reallocatable, i.e., they can't
exist on stacks or in objects that are freed.  If anyone cares,
I have a rewrite of sleep/wakeup that gets around that.
The shipped kernel should be clean this way so that it shouldn't
be a problem to anyone that hasn't been hacking hard.

They also found a neat bug in pexit().  Move the unlock on line
691 to after the wakeup.  Otherwise you can blow up indirecting
through a null reference.

			p->nwait++;
			unlock(&p->exl);

			wakeup(&p->waitr);

goes to

			p->nwait++;

			wakeup(&p->waitr);
			unlock(&p->exl);

No subject

[forsyth]

No subject

[Boyd]

    From: Thomas West <west9@worldnet.att.net>

    index
    end

this is a mailing list, not a listserver (or whatever you seem to think it is).
posting control messages to humans doesn't make for interesting reading.
thank you for using... 9fans.

No subject

[Thomas]

index
end

No subject

[Thomas]

info
index
end

No subject

[Thomas]

info
index
end

No subject

[Thomas]

info
end

No subject

[Thomas]

index
end

No subject

[Thomas]

index
end

No subject

[Markus]

No subject

[James]

subscribe jblackw@umr.edu
					James
--
Only the Dreamer can Change the Dream
                                      --John Logan

No subject

[philw]

checkout http://inferno.lucent.com

phil

No subject

[scott]

One more ape bug...  A common unix idiom is bind followed by listen.
In ape, bind sets the socket's port, and but listen also tries to do
that.  To avoid an error in that case, I propose listen.c:139 be
modified as follows:

		if(lip->sin_port >= 0) {
			if (write(cfd, "bind 0", 6) < 0) {
				errno = EGREG;
				close(cfd);
				return -1;
			}
			sprintf(msg, "announce %d", ntohs(lip->sin_port));

No subject

[Scott]

philw@plan9.bell-labs.com writes:
| Put this in the library ...

I think /sys/src/alef/lib/port/rwlock.l is there already (sans
comments), although rwlock.$O isn't listed in the mkfile.

No subject

[philw]

Put this in the library ...
philw.

#include <alef.h>

/*
 * Multiple readers/Single Write lock package
*/

void
RWlock.Rlock(RWlock *l)
{
	l->x.lock();		/* wait here for writers and exclusion */
	l->lock();
	l->readers++;
	l->k.canlock();		/* block writers if we are the first reader */
	l->unlock();
	l->x.unlock();
}

void
RWlock.Runlock(RWlock *l)
{
	l->lock();
	if(--l->readers == 0)	/* last reader out allows writers */
		l->k.unlock();
	l->unlock();
}

void
RWlock.Wlock(RWlock *l)
{
	l->x.lock();		/* wait here for writers and exclusion */
	l->k.lock();		/* wait here for last reader */
}

void
RWlock.Wunlock(RWlock *l)
{
	l->k.unlock();
	l->x.unlock();
}

No subject

[scott]

Another ape bug: putenv doesn't really do that.
(/n/lucent/sys/src/ape/lib/bsd/putenv.c)
It looks like the code in putenv.c was intended to
create a new file in /env, but skipped the /env
part.  But, given the code for execve, that's not
correct either.  Dropping in the BSD putenv will
probably fix the bug with a minimum of typing.

No subject

[scott]

I've noticed that APE has some byte order problems in the bsd compatability
code. The bsd interface uses network order exclusively, while the ape code
sometimes assumed host order. There have already been some updates on
this topic: enclosed are a few additional changes.

One problem that isn't fixed is that gethostbyaddr calls gethostbyname
with the ascii dotted-quad form of the network address.  That's a good
idea, except that the connection server only asks dns to return ip
addresses, even though it returns the whole record if it came from ndb.
gethostbyname could be modified to get the answer, but I think cs is
a more appropriate place for it.

diff inet_addr.c /n/cd/sys/src/ape/lib/bsd/inet_addr.c
50d49
< 	x = htonl(x);
diff inet_ntoa.c /n/cd/sys/src/ape/lib/bsd/inet_ntoa.c
19c19
< 	x = ntohl(in.s_addr);
---
> 	x = in.s_addr;
diff gethostbyaddr.c /n/cd/sys/src/ape/lib/bsd/gethostbyaddr.c
23c23,24
< 	memcpy(&x.s_addr, addr, sizeof(x.s_addr));
---
> 	x.s_addr = (addr[0]<<24)|(addr[1]<<16)|(addr[2]<<8)|addr[3];
>
diff gethostbyname.c /n/cd/sys/src/ape/lib/bsd/gethostbyname.c
98d97
< 			x = ntohl(x);

No subject

[Rob]

A simple bug in chan.c: in the Acreate case of namec(), OCEXEC is checked
but not ORCLOSE; add

		if(omode & ORCLOSE)
			c->flag |= CRCLOSE;

No subject

[Raj]

unsubscribe
---------------------------------------------------------------
Octopus traps
summer's moonspun dreams
soon fade away.
Basho 1644

No subject

[Scott]

10 print "9fans-request@cse.psu.edu"
20 goto 10

No subject

[Min]

unsubscribe

No subject

[Michael]

unsubscribe

No subject

[Scott]

Reminder:  unsubscribe messages must go to 9fans-request.

No subject

[Raj]

unsubscribe
---------------------------------------------------------------
Octopus traps
summer's moonspun dreams
soon fade away.
Basho 1644

No subject

[Finn]

subscribe 9fans

No subject

[philw]

check out http://inferno.bell-labs.com/inferno

No subject

[forsyth]

last week i posted a `fix' to 8c to handle side-effects in an lvalue receiving
a structure literal.  at the time, i wasn't able to try it in the other compilers,
but discovered yesterday that because something is missing from side(), the fix
isn't optimal on the 386 or 68020.  worse, because it invokes code containing
an existing but rarely seen error in cgen.c,
the `fix' will also break common cases of structure literals as arguments in
any compiler where REGARG is non-zero and equal to REGRET
(that includes most of the RISC compilers).  i've got a correct repair
that i'll post shortly.

No subject

[scott]

tarfs reports checksum errors on some tarfiles (generated by gnutar, I think),
because it computes the checksum over all TBLOCK chars, but only reads the
header before doing so.  The following patch seems to fix the problem.

term% diff /n/cd/sys/src/cmd/tapefs/tarfs.c .
38c38
< 	long blkno, isabs, chksum, linkflg;
---
> 	long blkno, isabs, chksum, linkflg, t;
47c47
< 		if (read(tapefile, (char *)&dblock.dbuf, sizeof(dblock.dbuf))<sizeof(dblock.dbuf))
---
> 		if (read(tapefile, (char *)&dblock, sizeof(dblock))<sizeof(dblock))
60,61c60,61
< 		if (chksum != checksum()){
< 			fprint(1, "bad checksum on %.28s\n", dblock.dbuf.name);
---
> 		if (chksum != (t=checksum())){
> 			fprint(1, "bad checksum %o != %o on %.28s\n", chksum, t, dblock.dbuf.name);
132c132
< 		i += *cp;
---
> 		i += *cp & 0xff;

No subject

[scott]

/sys/include/ndb.h lists csgetval() as taking five char*
parameters.  There should only be four, right?

No subject

[Nigel]

> It won't respond to pings.  Just try connecting to it
> with 9p.
> 

It responds to arp ...
 
Nigel Roles

No subject

[Dave]

It won't respond to pings.  Just try connecting to it
with 9p.

No subject

[Thomas]

How can I tell if a file server is up?

Would it respond to a ping request?
Should it respond is a better question.

-Tom

---- Where theory and reality meet. 
---- Thomas Riemer, triemer@wesleyan.edu

No subject

[Thomas]

I won one today!!!  The fileserver booted.... 
How about that!  

1. You have to disable the PLUG&PLAY feature on the card, and 
   you have to disable the BIOS... and it had to be on a slower
   machine....
2. And the ethernet card swaped... 

Finally at long last, there is hope - and a bit of light at the end
of the tunnel...  

I apologize for the exuberance... Its only taken me close on a week
to get this thing to boot. 

-Tom

---- Where theory and reality meet. 
---- Thomas Riemer, triemer@wesleyan.edu

No subject

[postmaster]

>From root Thu Mar 14 14:00 EST 1996 remote from bharat
>From 9fans Thu Mar 14 13:29:49 0500 1996 remote from cse.psu.edu
Received: from cse.psu.edu by bharat.ColumbiaSC.NCR.COM; Thu, 14 Mar 96 14:00 EST
Received: by ncrcae.ColumbiaSC.NCR.COM; 14 Mar 96 13:59:58 EST
Received: from cse.psu.edu by casey.ColumbiaSC.NCR.COM; Thu, 14 Mar 96 13:58 EST
Received: from ncrgw1.UUCP (uucp@localhost) by ncrhub4.attgis.com (8.7.3/8.7.3) with UUCP id NAA04358 for casey.columbiasc.ncr.com!karth; Thu, 14 Mar 1996 13:58:09 -0500 (EST)
Received: by ncrgw1.ATTGIS.COM; 14 Mar 96 13:42:46 EST
Received: by colossus.cse.psu.edu id <78688>; Thu, 14 Mar 1996 13:30:19 -0500
Received: from krystal.com ([205.230.227.129]) by colossus.cse.psu.edu with SMTP id <78695>; Thu, 14 Mar 1996 13:30:02 -0500
Received: by krystal.com (8.7.3/8.6.9) id MAA27301; Thu, 14 Mar 1996 12:29:49 -0600 (CST)
From: cse.psu.edu!9fans 
Date: Thu, 14 Mar 1996 13:29:49 -0500
Message-Id: <199603141829.MAA27301@krystal.com>
To: cse.psu.edu!9fans  
>From: bsdi.com!prb (Paul Borman)
Subject: re: cant start 9pcfs
Content-Length: 1416
Content-Type: text
Auto-Forwarded-From: casey!karth
Auto-Forward-Count: 1
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

> >>This is most certainly due to a memory sizing problem (heaven knows
> >>we have made the rounds with this one with BSD/OS...)  For now I
> 
> are the bytes in the CMOS that give the extended memory size not standard,
> or do some machines not set them correctly?

Well, we have found that many (all?) Dell machines think that the CMOS
can never report more than 16MB, so it doesn't, even if you have more
than 16MB of memory.  Then there are machines which we can't accurately
probe for how much memory there is, but the CMOS is right!  There are
also weird caching affect (i.e. you can read/write memory beyond the
end of memory as long as it stays in cache) as well as new and unique
ways to remap addresses that are beyond the end of memory.  To give you
and idea, here are the various parameters one can set in their boot.default
file for BSD/OS:

	-basemen mem	Assume this much base memory (memory below 1M)
	-cmosmem	Limit memory search to the mount given by the CMOS
	-extended mem	Limit the amount of memory to check
	-memsize mem	Just assume this it the amount of memory
	-noflushcache	Don't flush the cache while sizing memory

Each one of these is needed for at least one machine that we have had
problems with.  I should note that we can probe the proper amount of
memory on *most* machines, but...

Finding out how much memory you have is certainly a black art.

				-Paul Borman
				 prb@bsdi.com

No subject

[Steve]

Thomas Riemer <triemer@babbitt.bernstein.com> wrote:
> 4. Who maintains the FAQ?

I do. Isn't my name at the top of it? If not, where did you get it from?

No subject

[Ken]

> Compiling gs, I ran cross a bug, distilled down to the
> following:
> 
> term% cat x.c
> void foo()
> {	
> 	char	a;
> 	int	b;
> 	char	d[1];
>         long e = a | (1 << b); 
> 
> 	while (1) d[b] = e;
> }
> 
> term% 8c x.c && 8l x.8
> foo: doasm: notfound (6)	SALL	BX,AX
> 
> foo: doasm: notfound (6)	SALL	BX,AX
> 
> foo: doasm: notfound (6)	SALL	BX,AX

the fix is to add the following lines.
in /sys/src/cmd/8c/peep.c

165a166,177
> 		case AROLB:
> 		case AROLL:
> 		case AROLW:
> 		case ARORB:
> 		case ARORL:
> 		case ARORW:
> 		case ASALB:
> 		case ASALL:
> 		case ASALW:
> 		case ASARB:
> 		case ASARL:
> 		case ASARW:

No subject

[Scott]

Compiling gs, I ran cross a bug, distilled down to the
following:

term% cat x.c
void foo()
{	
	char	a;
	int	b;
	char	d[1];
        long e = a | (1 << b); 

	while (1) d[b] = e;
}

term% 8c x.c && 8l x.8
foo: doasm: notfound (6)	SALL	BX,AX

foo: doasm: notfound (6)	SALL	BX,AX

foo: doasm: notfound (6)	SALL	BX,AX

No subject

[Dave]

 1. Where is the equivalent of the "rc" files... i.e. where do I put
 stuff so that things like "listen" get run automatically when the machine
 boots.

/rc/bin/termrc for terminals
/rc/bin/cpurc for cpu servers

 2. I set up dns by running ndb/dns - this seemed to give me some 
 form of nameservice.  If I understand the manuals correctly, this starts
 up a domain name server on the machine.   Is there an equivalent to 
 the /etc/resolv.conf idea of just specifying a domain server without
 running a local name server?  BTW, the way the nameserver is running
 currently it only seems to resolve fully qualified names...I presume
 this is because it doesn't know what domain it is in... 
 cat /dev/hostdomain produces nothing... how do I tell it what domain it
 is in?  presumeably its in /lib/ndb/local. 

We only use fully qualified names.  Sorry.

We never did a separate resolver.

No subject

[Thomas]

I've finally gotten through the majority of installation hurdles
and gotten a machine up and running... 

So here is the list of neophyte questions that I have:
1. Where is the equivalent of the "rc" files... i.e. where do I put
stuff so that things like "listen" get run automatically when the machine
boots.

2. I set up dns by running ndb/dns - this seemed to give me some 
form of nameservice.  If I understand the manuals correctly, this starts
up a domain name server on the machine.   Is there an equivalent to 
the /etc/resolv.conf idea of just specifying a domain server without
running a local name server?  BTW, the way the nameserver is running
currently it only seems to resolve fully qualified names...I presume
this is because it doesn't know what domain it is in... 
cat /dev/hostdomain produces nothing... how do I tell it what domain it
is in?  presumeably its in /lib/ndb/local. 
 

3. How do I "disable anonymous access"... The FAQ points to the -N
option in aux/telnetd... but where does one specify this -N?  (I just
answered this myself -despite the FAQ being vague... in /bin/service)

4. Who maintains the FAQ?

-Tom



---- Where theory and reality meet. 
---- Thomas Riemer, triemer@wesleyan.edu

No subject

[Luther]

Thanks. What had me fooled about the C/1996 B2 (Hyakutake) elements is
that the eccentricity has been listed as 1.0000. According to the
latest (this morning) it really is 1.0000 (to be exact, 1.000019). I
have never seen an element like this before and figured it *had* to be
a misprint.
   Expect (later this evening?) a new replacement comet.c file at
ftp.stratcom.com in directory /pub/stratcom/comet
Thanks again,
Luther

On Mon, 26 Feb 1996 15:20:47 -0500, Rob Pike wrote:

>data from Harvard Smithsonian Center for Astrophysics
>
>    C/1995 O1 (Hale-Bopp)
>Epoch 1997 Mar. 13.0 TT = JDT 2450520.5
>T 1997 Apr. 1.09192 TT                                  Nakano
>q   0.9139592            (2000.0)            P               Q
>z  +0.0053885      Peri.  130.59470     -0.13307599     -0.17036395
> +/-0.000043       Node   282.47161     +0.28240339     +0.93776083
>e   0.9950751      Incl.   89.42447     +0.95002058     -0.30262311
> From 728 observations 1993 Apr. 27-1995 Dec. 8.  Ref. MPC 26374.
>
>    C/1995 O1 (Hale-Bopp)
>Epoch 1996 Apr. 27.0 TT = JDT 2450200.5
>T 1997 Mar. 31.81638 TT                                 Nakano
>q   0.9169064            (2000.0)            P               Q
>z  +0.0035730      Peri.  130.40448     -0.13192830     -0.17127816
> +/-0.000043       Node   282.47065     +0.27940881     +0.93859521
>e   0.9967239      Incl.   89.38049     +0.95106553     -0.29950429
>
>=============
>    C/1996 B2 (Hyakutake)
>T 1996 May 1.41087 TT                                   Marsden
>q   0.2301957            (2000.0)            P               Q
>                   Peri.  130.17807     +0.57751436     +0.80823672
>                   Node   188.06079     +0.23067112     -0.02639685
>e   1.0            Incl.  124.88607     +0.78311429     -0.58826576
> From 155 observations 1996 Jan. 1-Feb. 10.  Ref. MPEC 1996-C06.
>
>

No subject

[Rob]

data from Harvard Smithsonian Center for Astrophysics

    C/1995 O1 (Hale-Bopp)
Epoch 1997 Mar. 13.0 TT = JDT 2450520.5
T 1997 Apr. 1.09192 TT                                  Nakano
q   0.9139592            (2000.0)            P               Q
z  +0.0053885      Peri.  130.59470     -0.13307599     -0.17036395
 +/-0.000043       Node   282.47161     +0.28240339     +0.93776083
e   0.9950751      Incl.   89.42447     +0.95002058     -0.30262311
 From 728 observations 1993 Apr. 27-1995 Dec. 8.  Ref. MPC 26374.

    C/1995 O1 (Hale-Bopp)
Epoch 1996 Apr. 27.0 TT = JDT 2450200.5
T 1997 Mar. 31.81638 TT                                 Nakano
q   0.9169064            (2000.0)            P               Q
z  +0.0035730      Peri.  130.40448     -0.13192830     -0.17127816
 +/-0.000043       Node   282.47065     +0.27940881     +0.93859521
e   0.9967239      Incl.   89.38049     +0.95106553     -0.29950429

=============
    C/1996 B2 (Hyakutake)
T 1996 May 1.41087 TT                                   Marsden
q   0.2301957            (2000.0)            P               Q
                   Peri.  130.17807     +0.57751436     +0.80823672
                   Node   188.06079     +0.23067112     -0.02639685
e   1.0            Incl.  124.88607     +0.78311429     -0.58826576
 From 155 observations 1996 Jan. 1-Feb. 10.  Ref. MPEC 1996-C06.

No subject

[Berry]

>>>Ken.Goldsholl.Atlas.Computer.Equipment@west.net said:,
>>>805 said:
 > Deat AT&T:
 > 
 > We are considering using Plan 9 as the OS for the communications
 > engines that we sell.  Is this system rommable?  Also, please 
 > provide information on licensing, should we decide to market
 > a product with Plan 9.
 
That's cool.  But you seem to have sent the message to 9fans, which is a 
mailing list for users of plan 9.  Further, your email return address looks 
rather bogus, hence I'm sending this back to the 9fans list in hopes that you 
will get it.

According to the plan9 web page at http://plan9.att.com/plan9/distrib.html, 
"Read the shrinkwrap license to determine whether its terms are acceptable to 
you. The major restriction is that you can't sell products or services based 
on the supplied software. However, AT&T Software Solutions Group will 
negotiate commercial licenses (telephone 800 462 8146 or 1 415 943 4076)."

So my advice is to try calling AT&T Software Solutions Group directly.

  --berry

Berry Kercheval :: kerch@parc.xerox.com :: Xerox Palo Alto Research Center

No subject

[Ken.Goldsholl.Atlas.Computer.Equipment]

Deat AT&T:

We are considering using Plan 9 as the OS for the communications
engines that we sell.  Is this system rommable?  Also, please 
provide information on licensing, should we decide to market
a product with Plan 9.

Thanks,


Ken Goldsholl

No subject

[Ken]

forsyth a has the correct to the kc bug:

diff old /sys/src/cmd/kc/cgen.c
880c880
< 			nod4.xoffset = t->offset;
---
> 			nod4.vconst = t->offset;

thanks.
ken

No subject

[Berry]

>>>steve@plan9.ecf.toronto.edu said:
 > are many 9fans going to Usenix next week?
 > 

Probably, but not me.  I broke my knee a week ago and don't relish traipsing 
all over the Marriot and standing in NP-complete dinner-choosing groups on 
crutches...

  --berry

Berry Kercheval :: kerch@parc.xerox.com :: Xerox Palo Alto Research Center

No subject

[steve]

are many 9fans going to Usenix next week?

No subject

[Nigel]

> Don't forget Plan 9 makes the perfect
> Stocking Stuffer for this Yuletide
> season.
> 

I was thinking of asking for an optical jukebox 

No subject

[philw]

Don't forget Plan 9 makes the perfect
Stocking Stuffer for this Yuletide
season.

No subject

[steve]

I get a checksum error unpacking the boddle fs/pc/813525539.rc
Has anyone else seen this:

cpu% rc fs/pc/813525539.rc -X
813525539/8253.c: 
813525539/9pcfs.c: 
813525539/fns.h: 
813525539/l.s: 
813525539/pc.c: 813525539/pc.c checksum error creating updated file
cpu% ls -l 813525539/pc.c
--rw-rw-r-- M 4693 steve sys 6028 Nov 15 22:56 813525539/pc.c
cpu% sum 813525539/pc.c
dbeaed42   6028 813525539/pc.c
cpu% diff 813525539/pc.c /sys/src/fs/pc
336d335
< /* in assembly language
348d346
< */
cpu% ls -l fs/pc/813525539.rc
--rw-rw-r-- M 4693 steve sys 7871 Oct 25 15:59 fs/pc/813525539.rc
cpu% sum fs/pc/813525539.rc
b8547512   7871 fs/pc/813525539.rc

No subject

[jim]

The infamous 'no newline at end of file' problem. Boddle doesn't complain
about input files with no newline on the end, but the extraction by ed appends
one and so the checksums differ. I've put the correct checksum into the file
fs/pc/813525539.rc. There's probablt no need for you to pull it over again,
the pc.c you have created is likely correct.

------ original message follows ------

>From cse.psu.edu!9fans-outgoing-owner Wed Nov 15 23:10:29 EST 1995
Received: by colossus.cse.psu.edu id <78838>; Wed, 15 Nov 1995 23:04:50 -0500
Received: from plan9.ecf.toronto.edu ([128.100.8.9]) by colossus.cse.psu.edu with SMTP id <78835>; Wed, 15 Nov 1995 23:04:32 -0500
From:	thrash.ecf.toronto.edu!steve
To:	cse.psu.edu!9fans
Date:	Wed, 15 Nov 1995 23:02:08 -0500
Message-Id: <95Nov15.230432est.78835@colossus.cse.psu.edu>
Sender: cse.psu.edu!owner-9fans
Precedence: bulk
Reply-To: cse.psu.edu!9fans

I get a checksum error unpacking the boddle fs/pc/813525539.rc
Has anyone else seen this:

cpu% rc fs/pc/813525539.rc -X
813525539/8253.c: 
813525539/9pcfs.c: 
813525539/fns.h: 
813525539/l.s: 
813525539/pc.c: 813525539/pc.c checksum error creating updated file
cpu% ls -l 813525539/pc.c
--rw-rw-r-- M 4693 steve sys 6028 Nov 15 22:56 813525539/pc.c
cpu% sum 813525539/pc.c
dbeaed42   6028 813525539/pc.c
cpu% diff 813525539/pc.c /sys/src/fs/pc
336d335
< /* in assembly language
348d346
< */
cpu% ls -l fs/pc/813525539.rc
--rw-rw-r-- M 4693 steve sys 7871 Oct 25 15:59 fs/pc/813525539.rc
cpu% sum fs/pc/813525539.rc
b8547512   7871 fs/pc/813525539.rc

No subject

[Thomas]

Here are the two basic questions I have before I go spend money on plan9.
(Well, ok, so I know I'm going to spend money, I just don't know if I
can do it now, or will have to wait a couple of months.)

Here goes:

1. Does plan9 run on DecStation 2100s?  DecStation 2100s are little endian
  machines on a mips chip.   I've read the docs and it says there is
  only support for big endian machines.  Is this the truth and gospel or
  out of date docs?

2. Is it possible to run plan9 on a single standalone PC?  (I'd be willing
   to take the performance hit to get it up and running - 
   (I would be thinking of getting it up and running so that I could 
    get it to run on some of my other machines.)  Obviously the strength
    of plan9 is its network capability - the hump I'm faced with is
    the hump of getting it up and running without totally wrecking
    all the other little beasties in my apartment. 
    (if it were possible, I'd go buy a brand spanking new PC to run
       the thing on.)

-Tom

No subject

[presotto]

I made a boddle for some ftpfs changes:

//plan9.att.com/plan9/update/cmd/ftpfs/816029425.rc

	- recognize the current NT ftp server directory format
	- forsyth's change to avoid files named '.' and '..', .
	- passive mode accesses

No subject

[ken]

> From: haertel@ichips.intel.com
> The "8c" compiler sometimes loses track of side effects in expressions
> whose values aren't used.  For example, the following program should
> print "1" but in fact prints "0".
> ...

thanks to haertel for finding this. my fix follows.
ken

21a22,24
> 	l = n->left;
> 	r = n->right;
> 	o = n->op;
23,24c26,31
< 		if(nn != Z)
< 			gmove(n, nn);
---
> 		if(nn == Z) {
> 			if(o == OINDEX)
> 				nullwarn(l, r);
> 			return;
> 		}
> 		gmove(n, nn);
28,30d34
< 	l = n->left;
< 	r = n->right;
< 	o = n->op;

No subject

[Scott]

Ashish Agrawal <axa0437@faraday-gw.njit.edu> writes:
| subscribe

A good place to read 9fans is the newsgroup comp.os.plan9.

Note to 9fans who maintain FAQs or Web pages:  please update them to
mention that the list is bidirectionally gatewayed to the newsgroup.

No subject

[Ashish]

subscribe

No subject

[dhog]

>I unsuccessfully tried the unsubscription procedure posted to this
>list a few months ago. I am sorry to mail this address, but wrote to
>postmaster@, also without success.

Please note the following:

|Commands should be sent in the body of an email message to
|"majordomo@cse.psu.edu"or to "<list>-request@cse.psu.edu".
|[...]
|
|Commands in the "Subject:" line NOT processed.

That is, you should send a message to "9fans-request@cse.psu.edu"
with "unsubscribe" in the _body_, rather than the Subject.

No subject

[schwartz]

I noticed that srv sometimes gives unhelpful error messages.
Inspection reveals that there are uninitialized variables due to
goto-lossage.  Suggested improvement:

term% diff srv.c srv.c.orig
74d73
< 	strcpy(dest, *argv);
90a90,91
> 	strcpy(dest, *argv);
> 
124c125
< 		fprint(2, "srv %s: mount(%s) failed: %s\n", dest, mtpt, err);
---
> 		fprint(2, "srv %s: mount failed: %s\n", dest, err);

No subject

[Joshua]

Please unsubscribe me from 9fans!

I unsuccessfully tried the unsubscription procedure posted to this
list a few months ago. I am sorry to mail this address, but wrote to
postmaster@, also without success.

No subject

[presotto]

I fixed the boddle, upas/send/main.rc

No subject

[forsyth]

no, an authentication server isn't needed (or used) by u9fs in the distribution.
on plan 9,
	echo $user
or
	cat /dev/user

i suspect that the user name that appears isn't registered on the unix server.
especially so if the name is `none'.


------ original message follows ------

>From cse.psu.edu!9fans-outgoing-owner Sat Sep 30 11:49:25 BST 1995
Received: by colossus.cse.psu.edu id <78589>; Sat, 30 Sep 1995 02:38:33 -0400
Received: from skat.usc.edu ([128.125.253.131]) by colossus.cse.psu.edu with SMTP id <78575>; Sat, 30 Sep 1995 02:38:17 -0400
Received: (taweil@localhost)
	by skat.usc.edu (8.6.10/8.6.12+usc)
	id XAA29138; Fri, 29 Sep 1995 23:38:11 -0700
Date:	Sat, 30 Sep 1995 02:38:11 -0400
From:	Ta-Wei Li <taweil@skat.usc.edu>
Message-Id: <199509300638.XAA29138@skat.usc.edu>
To:	9fans@cse.psu.edu
Subject: Can not mount u9fs
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

I just got my PC setup to be a Plan 9 terminal and compiled the u9fs on my NeXT
box. When I try to access the file server, after srv command, the mount fails
with error message:

mount: mount /srv/tcp!next!9fs /n/next: unknown user or group

I am wondering if this mean that I need to setup an auth server.

Thanks.

Ta-Wei Li

No subject

[jmk]

	> Gateway 2000's (4DX-{33,66}V) - we've since added 100 MHz ships to these
as befits a boat-anchor...

No subject

[presotto]

Here's a partial list of systems Plan 9 runs on.  This
is just system/mother boards, not cards in those machines.


non-portables
-------------
Dell 433TE
Dell Omniplex 560/L
AT&T 6386/SX WGS (expected since that's what I did the orignal port for)
Gateway 2000's (4DX-{33,66}V) - we've since added 100 MHz ships to these


portables
---------
Globalyst 250/NEC Versa M (same machine, 2 retailers)
TI Travelmate/Gateway Nomad (same machine, 2 retailers)
ARM

motherboard		processor	chipset
-----------		---------	-------
AMI Enterprise III	486 DX2/66	SIS
AMI Enterprise IV	486 DX2/66	SIS
AMI Excalibur		Pentium 60	Mercury
ASUS PCI/E-P54NP4	Pentium 90	Neptune
Gigabyte GA586-ID	Pentium 90	Neptune
Intel Advanced/MN	Pentium 75	Triton
*Gigabyte GA586-AT	Pentium 133	Triton

* is new, reliability unknown as yet

No subject

[ken]

|| Speaking of play-value, has anyone gotten themselves far enough
|| out of VGA hell to try the sky catalog or the geographic/road
|| databases?
|
|the The road database is really cool, but I am a bit disappointed that
|Austin, TX is not in the database.  We used to live in Naperville, Il,
|and it was even more frustrating that the database stops within a
|couple of miles of Indian Hill.  It was, however, interesting to see
|the location of y'all's espresso machine....
|
|Where can the common man obtain additional database entries?

the tiger database was not (all) on the cdrom because of
space problems. the full database includes 48 states, alaska,
hawaii, puerto rico, and various islands. about 800 meg.

anyway i just set up the plan9 www/ftp namespace to bind
in the database in the directory tiger. let me know if
there are any problems.

ken

No subject

[Scott]

I retrieved Dennis' first message from the recycling bin.
In the interest of completeness, here it is.

-- Scott


------- Forwarded Message

Date:    Thu, 21 Sep 1995 17:37:07 -0400
From:    dmr@plan9.att.com
To:      owner-9fans@cse.psu.edu
Subject: re: breakup

A partial answer to "how will this affect AT&T Bell Labs?"
is as follows--the new, still unnamed technology company
will get the bulk of core Bell Labs, which includes research,
as well as the Bell Labs name.  The exact fraction is
not decided yet but numbers in the range 60-80% have
been mentioned.  The remainder will stay with AT&T
in a new AT&T Laboratories.  In thinking about these
figures, keep in mind that about half of research is
in the physical science area and it is assumed that
essentially all of this will be in Bell Labs; that
biases the division of the rest.

In general the division will be along obvious lines of work--
people involved with services to AT&T, with products to
Bell Labs.  In some cases, including most of our own
group, it is not an obvious call, and we don't know
which it will be.  Arno Penzias did remark today that the
split will not be like the one at divestiture,
in which one or two people from each group were selected
to go to Bellcore to yield similar populations; here it will
be more in groups as a whole. (But Arno is about to
leave his position as VP of research, so this is only
his opinion.)

Incidentally the intended time scale has the
new internal organizations being planned during the
fall and in place in the spring.  The formal
split is supposed to be accomplished at the beginning
of 1997.

I don't know what will happen with Plan 9 licensing.
During the official briefing it was mentioned that
division of IP was one of the thorniest of all the
issues, but this has to do especially with patent
licencing (if AT&T has an agreement with someone now,
who inherits it?)  Given that Plan 9 is hardly of
strategic interest to either company, I would trust
that we could argue that the outside licensing should
go where we go, and I wouldn't be surprised if
the other company kept rights to use.

There's no reason at this point to expect the split
to change the nature of our work.  But a lot is going
to happen in the next year.

		Dennis

------- End of Forwarded Message

No subject

[presotto]

I just changed http://plan9.att.com/plan9/doc/install.html to correct
the installation procedure for CPU/Auth servers.  Please read.

No subject

[presotto]

!	I have received the CDROM today, and I have seen that the
!serial interface driver for PCs doesn't support the 16550 UART. (Buffered) 
!which is mandatory for high-speed modems.
!
!	I'm considering adding the 16550 support if noone else is already 
!doing it; is anyone doing it already? It's a really nice chip, with a 
!16 character input buffer, and almost identical to the 8250/16450.
!(You can even substitute a 16450 for a 16550)
!
!	Borja.

The shipped driver supports fifo's.  The comment at the top
says 16450 but we typically use 16552's.  The code will turn
on the fifo's if you turn on modem flow control
(e.g. echo m > /dev/eia0ctl).  You can make it the default
if you want, but that screws up serial mice.

No subject

[Scott]

Murali Vemulapalli <mvemulap@valhalla.cs.wright.edu> writes:
| unsubscribe

Just a reminder: the one and only way to unsubscribe is to send a
message to 9fans-request, the same address you used to subscribe in the
first place.

No subject

[Murali]

unsubscribe

No subject

[rob]

Arrange using the usual Sun mechanisms to boot file 9sscd,
not 9sscpu.  This is a kernel configured as a terminal using a
CD-ROM as its root.  Boot that file, and when it asks you for
the root file system, say
	#R6/cd6
if your CD-ROM is SCSI unit 6.  Follow your nose after that.
Of course you will be unable to use the root as a writable file
system, but you can start kfs, use ipconfig to start networking,
and so on.  Read the manuals.

-rob
======
>From some guy

Hi,

I'd like to ask you a question about Plan9 installation procedure.

I have an un-networked Sparc IPX machine, running SunOS4.1.3, which I am 
trying to configure with the Plan9 OS and I am having some difficulty 
understanding the documented Sparc install procedure. I have a CDROM attached 
to the machines SCSI port and am able to mount the cdrom drive, but i do
not understand how to boot the Plan9 code. Do I copy the 9sscpu file to
root (/) and link to vmunix, then perform: STOP-a >boot? I've tried most of
the URLs and have not found the specific information.

I would appreciate if you can help me...re-direct me.

thank  you for your time.

No subject

[jmk]

I trawled through the offerings at CompUSA a few weeks ago and found the Trio64
on the STB Powergraph 64 and some Number9 card I can't remember the number of
(perhaps Motion 330?) as well as the Diamond Stealth 64 DRAM. I think the Hercules
Terminator 64 DRAM also uses it. In any case, the chip used was clearly marked
on all the boxes I looked at.

Remember from the updated aux/vga CHANGES file:

	09-Aug-95
	=========
	1) data.c mkfile trio64.c (new) vga.h /lib/vgadb
	   Support for Trio32/Trio64.
	   There's a bug at high resolutions causing a thin white stripe to appear between
	   the right edge of the display and the border.

I'll be trying to fix this when I can get time on the systems we have with the Trio64,
they're all heavily used in Ken's house right now ('high reolution' means > 1024).

------ original message follows ------

>From cse.psu.edu!9fans-outgoing-owner Tue Aug 22 17:17:42 EDT 1995
Received: by colossus.cse.psu.edu id <45608>; Tue, 22 Aug 1995 17:02:19 -0400
Received: from galapagos.cse.psu.edu ([130.203.2.12]) by colossus.cse.psu.edu with SMTP id <45595>; Tue, 22 Aug 1995 17:02:03 -0400
Received: from localhost by galapagos.cse.psu.edu with SMTP id <12691>; Tue, 22 Aug 1995 17:02:25 -0400
To:	9fans@cse.psu.edu
In-reply-to: Your message of "Tue, 22 Aug 1995 13:30:19 EDT."
             <95Aug22.133020edt.1408@cannon.ecf.toronto.edu> 
Date:	Tue, 22 Aug 1995 17:02:23 -0400
From:	Scott Schwartz <schwartz@galapagos.cse.psu.edu>
Message-Id: <95Aug22.170225edt.12691@galapagos.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

Steve says:
| according to "APPENDIX - What We Use - Last Updated 1-Aug-95"
| at url http://www.ecf.toronto.edu/plan9/clone.html :
| ... 
| There are a number of cards out there using the S3 Trio64 chip, they're cheap
| and perform well; the single-chip solution cuts out a lot of the guesswork in
| getting the card to work. 

Ok, but which models use the S3 Trio64 chip?  The local computer store
was pretty unhelpful; I was hoping to order something that is known to
work.  vgadb lists the "Diamond Stealth 64 DRAM": have people had good
results with that, or is the "Hercules PCI Bus Dynamite" the better
choice? 

No subject

[Scott]

Steve says:
| according to "APPENDIX - What We Use - Last Updated 1-Aug-95"
| at url http://www.ecf.toronto.edu/plan9/clone.html :
| ... 
| There are a number of cards out there using the S3 Trio64 chip, they're cheap
| and perform well; the single-chip solution cuts out a lot of the guesswork in
| getting the card to work. 

Ok, but which models use the S3 Trio64 chip?  The local computer store
was pretty unhelpful; I was hoping to order something that is known to
work.  vgadb lists the "Diamond Stealth 64 DRAM": have people had good
results with that, or is the "Hercules PCI Bus Dynamite" the better
choice? 

No subject

[Steve]

Scott Schwartz <schwartz@galapagos.cse.psu.edu>:
> Speaking of vga, what card do people recommend?  

according to "APPENDIX - What We Use - Last Updated 1-Aug-95"
at url http://www.ecf.toronto.edu/plan9/clone.html :

Subjectively, the Tseng Labs ET4000/W32p gives the best performance under
Plan 9, its only drawbacks being the lack of a hardware cursor in 2x8bit mode
and a maximum pixel-clock of 135MHz when used with a suitable RAMDAC.
The Hercules Stingray 64/Video with the ARK2000pv chip looks good too and
doesn't have restrictions on the use of the hardware cursor. Unfortunately we
have't been able to get the hardware cursor to work satisfactorily under Plan 9.
Don't know why, haven't had time to investigate. 

There are a number of cards out there using the S3 Trio64 chip, they're cheap
and perform well; the single-chip solution cuts out a lot of the guesswork in
getting the card to work. 

No subject

[Scott]

Speaking of vga, what card do people recommend?  

No subject

[presotto]

As advised by 9fans, we've replaced the online distribution with
one containing some of the more important fixes, i.e., those that
allow more sites to install plan 9. disk1 and disk2.vd are new and
contain:
- the floppy DMA overrun fixes
- changes to devata.c to put up with disks returning bogus ident
information
- the current disk/prep
- the current aux/vga and /lib/vgadb.

It's tested to the extent that I tried what I had the hardware to try.
Please tell me if anyone has problems.

The old versions are still out there under the names disk1.orig
and disk{2,3,4}.vd.orig.

No subject

[Boyd]

if people are going to post source, please post diffs
as they highlight the fix rather than hiding it.

No subject

[presotto]

Unlike dossrv, ftpfs doesn't post a file in /srv.  Dossrv does that
since it can be mounted many times.  If you removed dossrv's /srv/dos
file, it too would go away since that would clunk the last reference.

------ original message follows ------

>From cse.psu.edu!9fans-outgoing-owner Mon Aug  7 23:35:50 EDT 1995
Received: by colossus.cse.psu.edu id <45492>; Mon, 7 Aug 1995 23:13:50 -0400
Received: from galapagos.cse.psu.edu ([130.203.2.12]) by colossus.cse.psu.edu with SMTP id <45517>; Mon, 7 Aug 1995 22:52:39 -0400
Received: from localhost by galapagos.cse.psu.edu with SMTP id <12712>; Mon, 7 Aug 1995 22:10:29 -0400
To:	9fans@cse.psu.edu
In-reply-to: Your message of "Mon, 07 Aug 1995 19:15:25 EDT."
             <95Aug7.192122edt.34017@psuvax1.cse.psu.edu> 
Date:	Mon, 7 Aug 1995 22:10:15 -0400
From:	Scott Schwartz <schwartz@galapagos.cse.psu.edu>
Message-Id: <95Aug7.221029edt.12712@galapagos.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

philw@plan9.att.com writes:
| >Why does this happen...  I run dossrv, and then exit, and the shell
| >doesn't go away until dossrv is killed from another window.  
| the dossrv is holding a reference to the files supplied by the
| window system (eg. /dev/cons). Windows are cleared up by
| reference counting.

That sounds like a bug.  But why doesn't ftpfs suffer the same
symptoms?  On the other hand, even
	sleep 10 >[0]/tmp/zz >[1]/tmp/zz >[2]/tmp/zz >[3]/tmp/zz &
makes the window stick around; what files is that holding open?
It should be none, right?

No subject

[dhog]

>The namespace itself is referenced counted, so when all processes associated
>with the name space have exited, any open files bound in the namespace are closed.

At which point the server says ``Honey, I Clunked the Fids''.

No subject

[seanq]

>That sounds like a bug.  But why doesn't ftpfs suffer the same
>symptoms?  On the other hand, even
>	sleep 10 >[0]/tmp/zz >[1]/tmp/zz >[2]/tmp/zz >[3]/tmp/zz &
>makes the window stick around; what files is that holding open?
>It should be none, right?

When you open a window, a new name space is created with the window's
files bound in the appropriate places, i.e. /dev and /mnt/8½.  These
binds are in effect open file descriptors and the window will not
go away until they are closed.  This is why the command line
	sleep 10 >[0]/tmp/zz >[1]/tmp/zz >[2]/tmp/zz >[3]/tmp/zz &
stops the window from closing.

The namespace itself is referenced counted, so when all processes associated
with the name space have exited, any open files bound in the namespace are closed.

The reason the window closes even though ftpfs is running is rather subtle.
When ftpfs is started, it rforks the server process with the flag RFNAMEG
set.  This puts the server process in a copy of the original namespace.
Note that since it is a copy of the original namespace, it contains
references the the window's files.
The server is accessed by a pipe that is mounted in the original namespace.
When /bin/rc exits, the original namespace closes.  At this point, the window
is held up by ftpfs's namespace.  However, the original name-space contains
the only reference to one end of the pipe to ftpfs, thus this end of
the pipe is closed.  Ftpfs notices this and dies.  This closes
the namespace associated with ftpfs, closing the files associated with the window,
which finally allows the window to close.......

The reason that the above does not occur for dossrv is that the server
process is not rforked with RFNAMEG set.  This is probably a bug.  The net
result is that rc and dossrv share the same namespace, and thus when rc exits,
the namespace is not closed and hence dossrv does not die.....In effect
dossrv holds itself up, which stops the window from closing.

No subject

[Scott]

philw@plan9.att.com writes:
| >Why does this happen...  I run dossrv, and then exit, and the shell
| >doesn't go away until dossrv is killed from another window.  
| the dossrv is holding a reference to the files supplied by the
| window system (eg. /dev/cons). Windows are cleared up by
| reference counting.

That sounds like a bug.  But why doesn't ftpfs suffer the same
symptoms?  On the other hand, even
	sleep 10 >[0]/tmp/zz >[1]/tmp/zz >[2]/tmp/zz >[3]/tmp/zz &
makes the window stick around; what files is that holding open?
It should be none, right?

No subject

[Scott]

philw@plan9.att.com writes:

| >/dev/swap is ideosyncratic... it prints punctuation and labels, where
| >none of the other console files do.  Also, it doesn't tell you what
| >the pagesize is.
| my fault, I put it in for debugging. It stuck.

Naturally: the information /dev/swap gives is really useful to have.
Are you amenable to adding the pagesize to the output and regularizing 
the format?

No subject

[philw]

>/dev/swap is ideosyncratic... it prints punctuation and labels, where
>none of the other console files do.  Also, it doesn't tell you what
>the pagesize is.
my fault, I put it in for debugging. It stuck.

No subject

[philw]

>Why does this happen...  I run dossrv, and then exit, and the shell
>doesn't go away until dossrv is killed from another window.  
the dossrv is holding a reference to the files supplied by the
window system (eg. /dev/cons). Windows are cleared up by
reference counting.

No subject

[Scott]

/dev/swap is ideosyncratic... it prints punctuation and labels, where
none of the other console files do.  Also, it doesn't tell you what
the pagesize is.

No subject

[Scott]

Why does this happen...  I run dossrv, and then exit, and the shell
doesn't go away until dossrv is killed from another window.  

No subject

[rob]

I just put our current, much stronger version of mothra on the FTP site.
/plan9/pcdist/mothra.386 is the 386 binary.  Use it instead of the version
that unpacks with the 4-diskette system.

No subject

[boyd]

what's the etymology of bootes?

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00398; Tue, 1 Aug 1995 12:24:46 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id BAA00524; Tue, 1 Aug 1995 01:45:26 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id BAA01624 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 01:47:07 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id BAA01614 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 01:47:04 -0400
Received: by psuvax1.cse.psu.edu id <33971>; Tue, 1 Aug 1995 01:28:59 -0400
Received: from goonsquad.spies.com ([192.216.22.66]) by psuvax1.cse.psu.edu with SMTP id <33958>; Tue, 1 Aug 1995 01:28:42 -0400
Received: by goonsquad.spies.com (Smail3.1.29.1 #2)
	id m0sd9qs-000245C; Mon, 31 Jul 95 22:26 PDT
Message-Id: <m0sd9qs-000245C@goonsquad.spies.com>
X-PH: V4.2@Brown.EDU
From: ahm@goonsquad.spies.com (Andreas Meyer)
Subject: Re: questions
To: 9fans@cse.psu.edu
Date: 	Mon, 31 Jul 1995 01:26:49 -0400
In-Reply-To: <95Jul27.083136edt.33979@psuvax1.cse.psu.edu> from "rob@plan9.att.com" at Jul 27, 95 08:31:11 am
Organization: The Internet Wiretap
Content-Type: text
Content-Length: 457       
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

rob@plan9.att.com writes:

> Those of you who received the first distribution were asked to mail
> plan.9@research.att.com with questions.  We are turning off that
> mail alias; use this 9fans list as a first place for queries from now on.

Not to be rude, but I see nothing BUT questions here.
Will there eventually be answers?

Thanks.
Andy
-- 
 Andreas Meyer, N2FYE       Sustain.  Endure.  Evolve.
 ahm@goonsquad.spies.com    http://www.spies.com/~ahm/

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00386; Tue, 1 Aug 1995 12:24:44 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id CAA00549; Tue, 1 Aug 1995 02:59:15 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id DAA03496 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 03:00:56 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id DAA03491 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 03:00:53 -0400
Received: by psuvax1.cse.psu.edu id <34051>; Tue, 1 Aug 1995 02:38:41 -0400
Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by psuvax1.cse.psu.edu with SMTP id <34047>; Tue, 1 Aug 1995 02:37:30 -0400
Received: from plan9.cs.su.oz.au by staff.cs.su.OZ.AU (mail from boyd for
	9fans@cse.psu.edu)
	with MHSnet (insertion MHSnet site: staff.cs.su.oz.au); Tue, 01 Aug 1995 16:24:12 +1000
Received: from lore.plan9.cs.su.oz.au. by staff.cs.su.OZ.AU.; Tue, 01 Aug 1995 16:24:09 +1000
X-PH: V4.2@Brown.EDU
From: Boyd Roberts <boyd@plan9.cs.su.oz.au>
Date: 	Tue, 1 Aug 1995 02:12:00 -0400
To: 9fans@cse.psu.edu
Subject: Re: questions
In-Reply-To: <m0sd9qs-000245C@goonsquad.spies.com>
Message-ID: <199508011612.27381.out.babun@plan9.cs.su.oz.au>
From: ahm@goonsquad.spies.com (Andreas Meyer)
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

    

ot to be rude, but I see nothing BUT questions here.
    Will there eventually be answers?
    
why do you ask?

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00372; Tue, 1 Aug 1995 12:24:42 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id CAA00542; Tue, 1 Aug 1995 02:42:25 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id CAA03137 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 02:44:06 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id CAA03132 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 02:44:04 -0400
Received: by psuvax1.cse.psu.edu id <34046>; Tue, 1 Aug 1995 02:23:59 -0400
Received: from galapagos.cse.psu.edu ([130.203.2.12]) by psuvax1.cse.psu.edu with SMTP id <33958>; Tue, 1 Aug 1995 02:23:43 -0400
Received: from localhost by galapagos.cse.psu.edu with SMTP id <12684>; Tue, 1 Aug 1995 02:23:16 -0400
X-PH: V4.2@Brown.EDU
To: 9fans@cse.psu.edu
Subject: Re: questions 
In-reply-to: Your message of "Mon, 31 Jul 1995 01:26:49 EDT."
             <m0sd9qs-000245C@goonsquad.spies.com> 
Date: 	Tue, 1 Aug 1995 02:23:04 -0400
From: Scott Schwartz <schwartz@galapagos.cse.psu.edu>
Message-Id: <95Aug1.022316edt.12684@galapagos.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

Andreas writes:
| Not to be rude, but I see nothing BUT questions here.
| Will there eventually be answers?

In response to the questions I posted I've gotten answers via private
email.  Thanks to Jim and Dave for their help!

I was going to post a summary once things were stablized a bit, but
here's a quick rundown:

My ss2 didn't like the new scsi driver.  Replacing src/fs/ss/scsi.c
with the code from the first edition fixed the disk problems I was
seeing.

Also, the ss fs gets the clock wrong.  In toy.c, I fixed it by
incrementing rtc.year by 68 after reading the clock.  After doing that,
I noticed that the terminal version of the kernel fixes this problem
slightly differently, but I haven't tried making that change to the
fileserver yet.

On a (sparc, at least) auth server, boot dies if you tell it that its
ip address is 0.0.0.0; use the real ip number instead.

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00314; Tue, 1 Aug 1995 12:24:29 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id IAA00694; Tue, 1 Aug 1995 08:50:57 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id IAA13932 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 08:52:38 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id IAA13927 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 08:52:36 -0400
Received: by psuvax1.cse.psu.edu id <34047>; Tue, 1 Aug 1995 08:30:55 -0400
Received: from crab.plan9.cs.su.oz.au ([129.78.96.3]) by psuvax1.cse.psu.edu with SMTP id <33958>; Tue, 1 Aug 1995 08:30:35 -0400
Date: 	Tue, 1 Aug 1995 08:25:37 -0400
X-PH: V4.2@Brown.EDU
From: dhog@plan9.cs.su.oz.au
To: 9fans@cse.psu.edu
Subject: The Old and the New
Message-Id: <95Aug1.083035edt.33958@psuvax1.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

Here at Basser we have a mixture of machines running the old
and new versions of Plan 9.  We would like to get rid of the old
version.  However, one of these machines is a fileserver, with
attached jukebox, containing valuable files.  Is it safe to just install
a new file server kernel on it?  I am guessing that the disk filesystem
format hasn't changed; is this a safe assumption to make?

Note for those not in the know: the 9P protocol has changed in the
second edition, due to changes in the way that Plan 9 does its
authentication.  Writing a "protocol convertor" looks infeasible,
unless you only want to log in as user "none".  Not all files on the old
system are publically readable, and neither should they be.  Currently
we have one CPU server still on the old system, allowing us to use ftpfs
to access the old files.  But this is quite clumsy, hence my desire
to just run the new system with the old files.

So...  Can it be done?

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00290; Tue, 1 Aug 1995 12:24:25 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id JAA00743; Tue, 1 Aug 1995 09:56:35 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id JAA20604 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 09:58:14 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id JAA20585 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 09:58:05 -0400
Received: by psuvax1.cse.psu.edu id <34052>; Tue, 1 Aug 1995 09:32:46 -0400
Received: from crab.plan9.cs.su.oz.au ([129.78.96.3]) by psuvax1.cse.psu.edu with SMTP id <34049>; Tue, 1 Aug 1995 09:32:28 -0400
Date: 	Tue, 1 Aug 1995 09:27:31 -0400
X-PH: V4.2@Brown.EDU
From: David Hogan <dhog@plan9.cs.su.oz.au>
To: 9fans@cse.psu.edu
Subject: Re: plan 9 for sun 3/60
Message-Id: <95Aug1.093228edt.34049@psuvax1.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

forsyth@plan9.cs.york.ac.uk writes:
>several people ported older versions of Plan 9 to bits of the Sun-3 line.
>i did one locally for the sun-3/50 and sun-3/60; since i'm still using
>it, i'll probably do the modest amount of work required to get the
>new version running.  i'm not promising anything though, since i might
>just buy some PCs instead.

I'm one of the several.  I actually have the new system running on 3/50s
and 3/60s.  At some stage I will tidy this up and make it available to
anyone who cares (and has a valid license).

I also have 80% of a port of Plan 9 to the DEC Alpha architecture.  Currently
only axp 400s and 500s.  I plan to make this available when I reach 85% :-)
I'm hoping that someone out there who has a spare alpha will help to finish it off.

Since both of these ports are derived from Plan 9, I am only allowed to give
them to other licencees.  This will mean checking with AT&T that you
actually have a licence.

Please note that I am not quite ready to distribute the alpha port -- preparing
the distribution will take time, which I am short of at the moment.  I am telling
you all this to try to avoid duplication of effort.  I'll announce to the list when
it's ready -- hopefully within a month.

No subject

[9fans-outgoing-owner]

***** UNDELIVERABLE MAIL sent to absinthe, being returned by monte!absinthe *****
mail: Error # 8 'Invalid recipient' encountered on system monte

Received: from nick.chem.brown.edu by monte.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.brown.edu> id MAA00277; Tue, 1 Aug 1995 12:24:22 -0400
Received: from Brown.EDU by nick.chem.brown.edu via ESMTP (950221.405.SGI.8.6.10/911001.SGI)
	for <absinthe@monte.chem.Brown.EDU> id MAA00922; Tue, 1 Aug 1995 12:04:04 -0400
Received: (from daemon@localhost) by Brown.EDU (8.6.10/8.6.10) id MAA04435 for absinthe@monte.chem.Brown.EDU; Tue, 1 Aug 1995 12:05:44 -0400
Received: from psuvax1.cse.psu.edu (psuvax1.cse.psu.edu [130.203.1.6]) by Brown.EDU (8.6.10/8.6.10) with SMTP id MAA04428 for <Daniel_Faken@brown.edu>; Tue, 1 Aug 1995 12:05:42 -0400
Received: by psuvax1.cse.psu.edu id <33960>; Tue, 1 Aug 1995 11:16:09 -0400
Received: from plan9.att.com ([192.20.225.252]) by psuvax1.cse.psu.edu with SMTP id <34059>; Tue, 1 Aug 1995 11:11:00 -0400
X-PH: V4.2@Brown.EDU
From: ken@plan9.att.com
To: 9fans@cse.psu.edu
Date: 	Tue, 1 Aug 1995 11:01:40 -0400
Message-Id: <95Aug1.111100edt.34059@psuvax1.cse.psu.edu>
Sender: owner-9fans@cse.psu.edu
Precedence: bulk
Reply-To: 9fans@cse.psu.edu

> Here at Basser we have a mixture of machines running the old
> and new versions of Plan 9.  We would like to get rid of the old
> version.  However, one of these machines is a fileserver, with
> attached jukebox, containing valuable files.  Is it safe to just install
> a new file server kernel on it?  I am guessing that the disk filesystem
> format hasn't changed; is this a safe assumption to make?

> Note for those not in the know: the 9P protocol has changed in the
> second edition, due to changes in the way that Plan 9 does its
> authentication.  Writing a "protocol convertor" looks infeasible,
> unless you only want to log in as user "none".  Not all files on the old
> system are publically readable, and neither should they be.  Currently
> we have one CPU server still on the old system, allowing us to use ftpfs
> to access the old files.  But this is quite clumsy, hence my desire
> to just run the new system with the old files.

> So...  Can it be done?

the disk format is the same. there should be no problem
running the new FS code on the old disk.

ken

No subject

[ken]

> Here at Basser we have a mixture of machines running the old
> and new versions of Plan 9.  We would like to get rid of the old
> version.  However, one of these machines is a fileserver, with
> attached jukebox, containing valuable files.  Is it safe to just install
> a new file server kernel on it?  I am guessing that the disk filesystem
> format hasn't changed; is this a safe assumption to make?

> Note for those not in the know: the 9P protocol has changed in the
> second edition, due to changes in the way that Plan 9 does its
> authentication.  Writing a "protocol convertor" looks infeasible,
> unless you only want to log in as user "none".  Not all files on the old
> system are publically readable, and neither should they be.  Currently
> we have one CPU server still on the old system, allowing us to use ftpfs
> to access the old files.  But this is quite clumsy, hence my desire
> to just run the new system with the old files.

> So...  Can it be done?

the disk format is the same. there should be no problem
running the new FS code on the old disk.

ken

No subject

[rob]

I sense a conspiracy.

>From uunet.uu.net!MAILER-DAEMON	Sat Jul 29 12:49
From: MAILER-DAEMON@uunet.uu.net (Mail Delivery Subsystem)
Subject: Returned mail: Cannot send message for 1 day
To: <rob@plan9.att.com>

The original message was received at Fri, 28 Jul 1995 11:35:29 -0400
from [192.20.225.252]

   ----- The following addresses had delivery problems -----
<plan9-vote@sub-rosa.com>  (unrecoverable error)

   ----- Transcript of session follows -----
<plan9-vote@sub-rosa.com>... Deferred: Connection timed out with infinity.hip.berkeley.edu.
Message could not be delivered for 1 day
Message will be deleted from queue

   ----- Original message follows -----
Return-Path: <rob@plan9.att.com>
Received: from plan9.att.com by relay4.UU.NET with SMTP 
	id QQzaiw09963; Fri, 28 Jul 1995 11:35:29 -0400
From: rob@plan9.att.com
Message-Id: <QQzaiw09963.199507281535@relay4.UU.NET>
To: plan9-vote@sub-rosa.com
Date: Fri, 28 Jul 1995 08:05:17 EDT

      I vote YES on comp.os.plan9

No subject

[Scott]

I just installed the second edition on a sparcstation 2 (fs) and some
slcs (cpu and term), using essentially the procedure for installing the
first edition (no pc required).  One problem I see is that the file
server complains about bad udp checksums, "scsiintr: not disconnect:
status 87 step/intr=418 dma=94000011", status errors from D6.0.1.0 and
D6.0.3.0, and check reporting various problems.  Has anyone else seen
this, or has my old hardware finally gone bad?

No subject

[Scott]

Rather than have everyone dig thru the majordumbo archives, I'll just
post the messages since just before the 19th.  That way external
archive sites will get them too.

Also, could everyone please check that their addresses are correct 
and not duplicated or otherwise confused?  

-------

>From 9fans-outgoing-owner Tue Jul 18 21:02:40 1995
Received: by psuvax1.cse.psu.edu id <34078>; Tue, 18 Jul 1995 20:37:52 -0400
Received: from gw2.att.com ([192.20.239.134]) by psuvax1.cse.psu.edu with SMTP id <34076>; Tue, 18 Jul 1995 20:37:17 -0400
Received: from cbgbcs.cb.att.com by ig1.att.att.com id AA08953; Tue, 18 Jul 95 20:24:50 EDT
Received: by cbgbcs.cb.att.com (5.0/EMS-1.1 Sol2)
	id AA27715; Tue, 18 Jul 1995 20:25:06 +0500
From:	Lawrence.V.Cipriani@att.com
Received: by cbgbcs.cb.att.com (5.0/EMS-1.1 Sol2)
	id AA27687; Tue, 18 Jul 1995 20:25:03 +0500
Message-Id: <9507190025.AA27687@cbgbcs.cb.att.com>
Original-From: lvc@cbgbcs.cb.att.com
To:	9fans
Date:	Tue, 18 Jul 1995 21:23:52 -0400
Subject: AT&T Plan 9 announcement
Content-Type: text
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

         AT&T ANNOUNCES *** AT&T today announced that Plan 9, a new 
         computer operating system from AT&T Bell Laboratories, is now 
         available for research and educational use.  The Plan 9 operating 
         system, named for the science-fiction cult movie "Plan 9 From 
         Outer Space," was designed by the inventors of the UNIX system, a 
         widely used operating system created at Bell Labs 25 years ago.   
         "Plan 9 is not in competition with UNIX or Windows," said Paul 
         Fillinich, marketing manager for AT&T Software Solutions.  The 
         Plan 9 system is based on the concept of distributed computing in 
         a networked, client-server environment.  The product, including 
         source code, is available for $350.  The full kit will be shipped 
         with a CD-ROM, four diskettes and two manuals.  A partial kit, 
         containing only the manuals, may be ordered separately for $125.  
         The contact number in the U.S. for information and orders is 800-
         462-8146; elsewhere, +1-415-943-4076.
--
Larry Cipriani, l.v.cipriani@att.com
Ever feel like you're being watched ? -- You will.




>From 9fans-outgoing-owner Wed Jul 19 03:43:15 1995
Received: by psuvax1.cse.psu.edu id <34076>; Wed, 19 Jul 1995 03:10:22 -0400
Received: from octarine.cc.adfa.oz.au ([131.236.1.1]) by psuvax1.cse.psu.edu with SMTP id <34081>; Wed, 19 Jul 1995 03:08:48 -0400
Received: (from cjsv@localhost) by octarine.cc.adfa.oz.au (8.6.12/8.6.12) id RAA07265; Wed, 19 Jul 1995 17:07:47 +1000
Message-Id: <199507190707.RAA07265@octarine.cc.adfa.oz.au>
X-Mailer: exmh version 1.5.1 12/2/94
Organization: Computer Science, University College, University of New 
 South Wales, Canberra
X-URI: http://www.cs.adfa.oz.au/people/cjsv.html
From:	Christopher.Vance@adfa.oz.au
To:	9fans
Subject: Re: free plan 9 trial 
In-reply-to: Your message of Tue, 18 Jul 1995 16:18:35 -0400.
             <95Jul18.163015edt.34076@psuvax1.cse.psu.edu> 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date:	Wed, 19 Jul 1995 03:07:46 -0400
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

|  The free version is available.  Ftp to
|  plan9.att.com, log in as anonymous, and get the
|  files from /plan9/pcdist.  The distrib.html
|  page now points to them.

I tried booting disk1 on two 486s around here.

One said
	/386/init: file does not exist
	panic: boot process dies: unknown
	(plus what I take to be a stack trace)
A directory listing under MS-DOS indicates \386\init does exist.

The other one said
	exception/interrupt 6
	FLAGS=10092  TRAP=6  ECODE=0  PC=80107d84
	(plus a register dump)

I successfully booted the floppy image from the previous version on both 
machines, so I'm sad to see the new version fail so soon.  Should I try 
verifying the disk image with a checksum?  Or should I try a different 
brand PC?

(I was still waiting for a local SparcStation 2 to become available so I 
can put the previous version up there.  Perhaps I should continue to wait.)

-- Christopher




>From 9fans-outgoing-owner Wed Jul 19 06:55:20 1995
Received: by psuvax1.cse.psu.edu id <33973>; Wed, 19 Jul 1995 06:11:55 -0400
Received: from goggins.bath.ac.uk ([138.38.32.13]) by psuvax1.cse.psu.edu with SMTP id <34028>; Wed, 19 Jul 1995 06:08:41 -0400
Received: from bath.ac.uk (actually host ss1.bath.ac.uk) by goggins.bath.ac.uk 
          with SMTP (PP); Wed, 19 Jul 1995 11:03:28 +0100
To:	9fans
Subject: plan9 disks in ftp.bath.ac.uk:/pub/plan9
Date:	Wed, 19 Jul 1995 06:04:45 -0400
From:	Icarus Sparry <ccsis@bath.ac.uk>
Message-ID: <9507191104.aa09064@ss1.bath.ac.uk>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans


I have put a copy of the plan9.att.com:/plan9/pcdist directory in
ftp.bath.ac.uk:/pub/plan9, for anyone who has a faster route to us than
the states.

It is also available in the alex cache of alex.niss.ac.uk for UK academic
sites.



>From 9fans-outgoing-owner Wed Jul 19 13:12:09 1995
Received: by psuvax1.cse.psu.edu id <34085>; Wed, 19 Jul 1995 13:11:03 -0400
Received: from crab.plan9.cs.su.oz.au ([129.78.96.3]) by psuvax1.cse.psu.edu with SMTP id <34080>; Wed, 19 Jul 1995 13:09:53 -0400
From:	dhog@plan9.cs.su.oz.au
Date:	Wed, 19 Jul 1995 12:46:26 -0400
To:	9fans
Subject: Booting a Sparcstation from disk
Message-Id: <95Jul19.130953edt.34080@psuvax1.cse.psu.edu>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

[Apologies if you've already seen this]

I've got my SLC booting from disk.  The command ss/home will set up
a new boot disk (takes about 24 mins).  Caveats:

- boot disk must be scsi id 3 for the ROM monitor to boot from it
- disk probably needs to have been formatted by sunos/solaris, which stores
   a lot of pointless information + a checksum in sector 0, which I have been
   leaving alone.  If this becomes a problem, I can probably provide a "blank"
   sector 0 (there's a magic number, so it can't just be all zeros)
- only V1 of the ROM monitor is supported.  The ELC's are V2.  I will
   probably fix this at a later date.  Apparently a V0 exists also...

P.S.  You really don't want to know how it works! ;-)
P.P.S. Danger!  The other "home" scripts are unaware that the interface to
    disk/prep has changed...



>From 9fans-outgoing-owner Wed Jul 19 17:33:02 1995
Received: by psuvax1.cse.psu.edu id <34154>; Wed, 19 Jul 1995 17:32:51 -0400
Received: from relay3.UU.NET ([192.48.96.8]) by psuvax1.cse.psu.edu with SMTP id <34091>; Wed, 19 Jul 1995 17:32:28 -0400
Received: from offramp.dsccc.com by relay3.UU.NET with SMTP 
	id QQyzco07430; Wed, 19 Jul 1995 17:32:14 -0400
Received: by offramp.dsccc.com (5.67b/SMI-V1.8)
	id AA23307; Wed, 19 Jul 1995 16:33:54 -0500
Received: from onramp(192.245.102.129) by offramp via smap (V1.3mjr)
	id sma023265; Wed Jul 19 16:33:14 1995
Received: from spd.dsccc.com (spdmail.spd.dsccc.com [101.25.2.34]) by camelot.dsccc.com (8.6.11/8.6.10) with SMTP id QAA29826 for <9fans@cse.psu.edu>; Wed, 19 Jul 1995 16:32:52 -0500
Received: from sun001.spd.dsccc.com by spd.dsccc.com (5.0/SMI-SVR4)
	id AA03351; Wed, 19 Jul 1995 16:34:51 -0500
From:	mmarshal@spdmail.spd.dsccc.com (Mike Marshall)
Message-Id: <9507192134.AA03351@spd.dsccc.com>
Subject: Re: AT&T Plan 9 announcement
To:	9fans
Date:	Wed, 19 Jul 1995 17:34:50 -0400
In-Reply-To: <9507190025.AA27687@cbgbcs.cb.att.com> from "Lawrence.V.Cipriani@att.com" at Jul 18, 95 09:23:52 pm
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1023      
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

> 
>          AT&T ANNOUNCES *** AT&T today announced that Plan 9, a new 
>          computer operating system from AT&T Bell Laboratories, is now 
>          available for research and educational use.  The Plan 9 operating 
>          system, named for the science-fiction cult movie "Plan 9 From 
>          Outer Space," was designed by the inventors of the UNIX system, a 
>          widely used operating system created at Bell Labs 25 years ago.   
>          "Plan 9 is not in competition with UNIX or Windows," said Paul 
>          Fillinich, marketing manager for AT&T Software Solutions.  The 
>          Plan 9 system is based on the concept of distributed computing in 
>          a networked, client-server environment.  The product, including 
>          source code, is available for $350.

Does this mean that upon purchasing the "Full Kit" I will hold a
"Source License for this product"...

Does anyone have the Legal Speak as to what I can do with this
"Source License"???

Mike
mmarshal@spd.dsccc.com



>From 9fans-outgoing-owner Wed Jul 19 18:29:37 1995
Received: by psuvax1.cse.psu.edu id <34152>; Wed, 19 Jul 1995 18:29:27 -0400
Received: from relay1.UU.NET ([192.48.96.5]) by psuvax1.cse.psu.edu with SMTP id <34155>; Wed, 19 Jul 1995 18:28:45 -0400
Received: from offramp.dsccc.com by relay1.UU.NET with SMTP 
	id QQyzcr23675; Wed, 19 Jul 1995 18:27:23 -0400
Received: by offramp.dsccc.com (5.67b/SMI-V1.8)
	id AA25636; Wed, 19 Jul 1995 17:29:03 -0500
Received: from onramp(192.245.102.129) by offramp via smap (V1.3mjr)
	id sma025624; Wed Jul 19 17:28:21 1995
Received: from spd.dsccc.com (spdmail.spd.dsccc.com [101.25.2.34]) by camelot.dsccc.com (8.6.11/8.6.10) with SMTP id RAA02810 for <9fans@cse.psu.edu>; Wed, 19 Jul 1995 17:27:59 -0500
Received: from sun001.spd.dsccc.com by spd.dsccc.com (5.0/SMI-SVR4)
	id AA04716; Wed, 19 Jul 1995 17:29:57 -0500
From:	mmarshal@spdmail.spd.dsccc.com (Mike Marshall)
Message-Id: <9507192229.AA04716@spd.dsccc.com>
Subject: Re: floppies for ftp?
To:	9fans
Date:	Wed, 19 Jul 1995 18:29:57 -0400
In-Reply-To: <9507181914.AA30577@nssdc.gsfc.nasa.gov> from "James W. Williams" at Jul 18, 95 03:14:30 pm
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 142       
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

> 
> Any word on when the floppy disk images for the binary-only version will
> be put up for ftp?

Will there be a trial for Sparc???

Mike




>From 9fans-outgoing-owner Thu Jul 20 17:58:09 1995
Received: by psuvax1.cse.psu.edu id <33977>; Thu, 20 Jul 1995 17:57:55 -0400
Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by psuvax1.cse.psu.edu with SMTP id <34023>; Thu, 20 Jul 1995 17:57:35 -0400
Received: from sour.sw.oz.au by staff.cs.su.OZ.AU (mail from jeremy for
	9fans@cse.psu.edu)
	with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Fri, 21 Jul 1995 07:57:06 +1000
Received: from suite.sw.oz.au by swallow.sw.oz.au with SMTP
	id AA29753; Fri, 21 Jul 95 07:55:40 EST (4.1/Unixware)
	(from jeremy@sour.sw.oz.au for 9fans@cse.psu.edu)
Received: from sour.sw.oz.au by suite.sw.oz.au with SMTP
	id AA25287; Fri, 21 Jul 1995 01:06:16 +1000 (5.65c/1.34)
	(from jeremy@sour.sw.oz.au for 9fans@cse.psu.edu)
Received: by sour.sw.oz.au
	id AA02178; Fri, 21 Jul 1995 01:04:50 +1000 (5.65c/1.34)
	(from jeremy@sour.sw.oz.au for 9fans@cse.psu.edu)
From:	jeremy@sour.sw.oz.au (Jeremy Fitzhardinge)
Message-Id: <199507201504.AA02178@sour.sw.oz.au>
Subject: aux/mouse: Unknown mouse type
To:	9fans (Graverobbers from Outer Space!)
Date:	Thu, 20 Jul 1995 11:04:48 -0400
Organization: Softway Pty Ltd
X-Face: 
	 '6U=%Tv\k1<Ek%ql%PN^v`Db4bakr[v~y]\u7"GbO#I=]N{l1=#P,glz$9q>l-:?\$C[D@G
	 7(vl~w8&y}!f\bh#w<Y*S~bEBTI:s&.QR>L#n,TGKh>T.c7eT5-y)Hl'i;A1z$9?*lD.k}yqshddFb
	 l[EC}c=;uc%x'}uh3E91p&oE<q$w1r&U0yw.Sb3V&uw
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 729       
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Hi all,

I've installed the PC demo, and it mostly seems to work.  However,
when it tries to start the mouse driver (?) at boot time, or if I
start it manually, it exits with the "unknown mouse type" message
after couple of seconds delay.

If I "cat /dev/eia0" it produces output when I move the mouse.  I've
tried a couple of mice, both of which work as normal 3 button/5
byte mice on other systems.  I also tried the other serial port, and
using "aux/mouse 0" explicitly.  aux/mouse ps2, tried just for the
sake of experimentation, exited immediately without a message.

A search of the man pages on plan9.att.com didn't produce anything
useful, and of course, I don't (yet) have the source for the
aux/mouse.

Thanks,
	Jeremy



>From 9fans-outgoing-owner Fri Jul 21 03:11:02 1995
Received: by psuvax1.cse.psu.edu id <34034>; Fri, 21 Jul 1995 03:09:47 -0400
Received: from staff.cs.su.OZ.AU ([129.78.8.1]) by psuvax1.cse.psu.edu with SMTP id <34033>; Fri, 21 Jul 1995 03:08:40 -0400
Received: from sour.sw.oz.au by staff.cs.su.OZ.AU (mail from jeremy for
	9fans@cse.psu.edu)
	with MHSnet (insertion MHSnet site: swallow.sw.oz.au); Fri, 21 Jul 1995 17:07:53 +1000
Received: from sour.sw.oz.au by swallow.sw.oz.au with SMTP
	id AA06258; Fri, 21 Jul 95 17:07:27 EST (4.1/Unixware)
	(from jeremy@sour.sw.oz.au for 9fans@cse.psu.edu)
Received: by sour.sw.oz.au
	id AA00204; Fri, 21 Jul 1995 17:07:21 +1000 (5.65c/1.34)
	(from jeremy@sour.sw.oz.au for 9fans@cse.psu.edu)
From:	jeremy@sour.sw.oz.au (Jeremy Fitzhardinge)
Message-Id: <199507210707.AA00204@sour.sw.oz.au>
Subject: Mouse problems on the PC again
To:	9fans (Graverobbers from Outer Space!)
Date:	Fri, 21 Jul 1995 03:07:20 -0400
Organization: Softway Pty Ltd
X-Face: 
	 '6U=%Tv\k1<Ek%ql%PN^v`Db4bakr[v~y]\u7"GbO#I=]N{l1=#P,glz$9q>l-:?\$C[D@G
	 7(vl~w8&y}!f\bh#w<Y*S~bEBTI:s&.QR>L#n,TGKh>T.c7eT5-y)Hl'i;A1z$9?*lD.k}yqshddFb
	 l[EC}c=;uc%x'}uh3E91p&oE<q$w1r&U0yw.Sb3V&uw
X-Mailer: ELM [version 2.4 PL21]
Content-Type: text
Content-Length: 520       
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Hi again,

I should have paid more attention.  The message "unknown mouse
type" really means that: when I switched the mouse into to Microsoft
mode it worked properly.  However, every single 3 button mouse with
a Logitech/MS switch ignores the middle button in MS mode.  I know
the MS protocol supports a 3rd button, but it seems that all the
mouse hardware assumes that the software will break with it enabled.

Has someone implemented a mouse driver which supports the more
common 3 button mouse protocol?

Thanks,
	J



>From 9fans-outgoing-owner Fri Jul 21 11:40:45 1995
Received: by psuvax1.cse.psu.edu id <34033>; Fri, 21 Jul 1995 11:40:28 -0400
Received: from alf.zfn.uni-bremen.de ([134.102.20.22]) by psuvax1.cse.psu.edu with SMTP id <34034>; Fri, 21 Jul 1995 11:38:33 -0400
Received: from blue.lrw.uni-bremen.de by alf.zfn.uni-bremen.de (AIX 3.2/UCB 5.64/4.940318)
          id AA43879; Fri, 21 Jul 1995 17:37:30 +0200
Received: from black.lrw.uni-bremen.de by lrw.uni-bremen.de (5.0/SMI-SVR4)
	id AA11507; Fri, 21 Jul 1995 17:42:05 +0200
Received: by black.lrw.uni-bremen.de (5.0/SMI-SVR4)
	id AA03852; Fri, 21 Jul 1995 17:40:09 +0200
Date:	Fri, 21 Jul 1995 11:40:09 -0400
From:	a82@blue.lrw.uni-bremen.de (Henner Gratz)
Message-Id: <9507211540.AA03852@black.lrw.uni-bremen.de>
To:	plan9-fans@cs.psu.edu
Content-Length: 1010
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans


Hello everybody!

Now, after Plan9 is available for everyone, I plan to order my CD-ROM from AT&T.
I want to install it on a 486/Pentium PC. Because I don't own such a beast at
the moment, I will have to buy one. And here the problems start. On what
main-boards, graphic-cards etc. Plan9 will run. The WWW-pages on plan9.att.com
aren't very detailed. I have also read the lists spread in the 9fans mailing
list, but they are from 1994 for the old distributions. 

So here's my question: Is there a list covering the new distribution?
	
Thank you very much for any answers!

Regards,

	Henner

--
===============================================================================
 
Henner Gratz                                 Email:       a82@lrw.uni-bremen.de
Leeuwarder Str. 16A                          Fax (at home): +49-(0)421-58 52 10
D-28259 Bremen                               Tel (at home): +49-(0)421-58 51 84
GERMANY
 
===============================================================================




>From 9fans-outgoing-owner Mon Jul 24 08:51:06 1995
Received: by psuvax1.cse.psu.edu id <34047>; Mon, 24 Jul 1995 08:50:41 -0400
Received: from alf.zfn.uni-bremen.de ([134.102.20.22]) by psuvax1.cse.psu.edu with SMTP id <34041>; Mon, 24 Jul 1995 08:49:53 -0400
Received: from blue.lrw.uni-bremen.de by alf.zfn.uni-bremen.de (AIX 3.2/UCB 5.64/4.940318)
          id AA45113; Mon, 24 Jul 1995 14:49:11 +0200
Received: from black.lrw.uni-bremen.de by lrw.uni-bremen.de (5.0/SMI-SVR4)
	id AA16594; Mon, 24 Jul 1995 14:53:41 +0200
Received: by black.lrw.uni-bremen.de (5.0/SMI-SVR4)
	id AA07532; Mon, 24 Jul 1995 14:51:44 +0200
Date:	Mon, 24 Jul 1995 08:51:44 -0400
From:	a82@blue.lrw.uni-bremen.de (Henner Gratz)
Message-Id: <9507241251.AA07532@black.lrw.uni-bremen.de>
To:	9fans
Content-Length: 800
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans


Hello again!

Here I am again, looking for the right configuration...

Has anybody tried to run Plan9 on a Gigabyte GA-586AT(/P256) board?

The Bi-processor boards with the Neptun chip set used by the AT&T folks
are nice, but the Trition boards are faster AND cheaper!

Any experiences?

Tschuess,

	Henner

PS: I hope this will be the last question about the `right' configuration.
--
===============================================================================
 
Henner Gratz                                 Email:       a82@lrw.uni-bremen.de
Leeuwarder Str. 16A                          Fax (at home): +49-(0)421-58 52 10
D-28259 Bremen                               Tel (at home): +49-(0)421-58 51 84
GERMANY
 
===============================================================================




>From 9fans-outgoing-owner Tue Jul 25 12:45:48 1995
Received: by psuvax1.cse.psu.edu id <34077>; Tue, 25 Jul 1995 12:44:29 -0400
Received: from cegelecproj.co.uk ([159.245.72.6]) by psuvax1.cse.psu.edu with SMTP id <34047>; Tue, 25 Jul 1995 12:43:55 -0400
Received: from vampire.cegelecproj.co.uk (cerberus.cegelecproj.co.uk) by cegelecproj.co.uk (4.1/SMI-4.1)
	id AA28738; Tue, 25 Jul 95 17:43:37 BST
Received: from spirit.cegelecproj.co.uk (spirit.limbo.cegelecproj.co.uk) by vampire.cegelecproj.co.uk (5.0/SMI-SVR4)
	id AA05923; Tue, 25 Jul 1995 17:43:32 +0100
Received: by spirit.cegelecproj.co.uk (5.0/SMI-SVR4)
	id AA01104; Tue, 25 Jul 1995 17:43:28 +0100
Date:	Tue, 25 Jul 1995 12:43:28 -0400
From:	Steve_Kilbane@cegelecproj.co.uk (Steve_Kilbane)
Message-Id: <9507251643.AA01104@spirit.cegelecproj.co.uk>
X-Planation: X-Faces images can be viewed with the XFaces program
To:	9fans
Subject: 003056-014 VGA card supported?
Content-Length: 619
X-Face: Iqsa(US9p?)Y^W+6Ff[Z]<t?\A!eaL'DG{20*#{C1;'Ct&}L}B^/1(aYI@hP)4!<}7D=2gm
	8!$T`8QNfK<te\20%A\`wm*wa2"^Up*Qs"X}KeV*3XeB2te&sKp*t`N;^BDh[6=K{ZBE=O>rM"uFE)
	lFDjag1e]\/#2
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

I've been playing with the trial distribution floppies, trying to
get the system to boot on an old Dell 316SX, here at work. I can get
it as far as attempting to configure the VGA system, but of course the
default /lib/vgadb doesn't have an entry for my machine. I've tried
hacking vgadb and just adding in an entry for it as a straightforward
640x480, but it just hangs. Sigh. I've got the machine to the state
where it can boot up and give me a shell, but there's no access to
8½, etc.

So the question is, is there any way of getting it to support this card,
or do I just give up on this particular machine?

steve



>From 9fans-outgoing-owner Wed Jul 26 16:00:44 1995
Received: by psuvax1.cse.psu.edu id <34099>; Wed, 26 Jul 1995 15:56:20 -0400
Received: from noao.edu ([140.252.1.54]) by psuvax1.cse.psu.edu with SMTP id <34164>; Wed, 26 Jul 1995 15:16:27 -0400
Received: from gemini.tuc.noao.edu by noao.edu (4.1/SAG-Noao.G102)
	id AA21757; Wed, 26 Jul 95 09:36:53 MST; for 9fans@cse.psu.edu
Received: from daikon.tuc.noao.edu.noao by gemini.tuc.noao.edu (4.1/SAG.sat.14)
	id AA17793; Wed, 26 Jul 95 09:36:52 MST; for 9fans@cse.psu.edu
Received: by daikon.tuc.noao.edu.noao (4.1/SAG.sun.8)
	id AA19352; Wed, 26 Jul 95 09:36:50 MST; for 9fans@cse.psu.edu
Date:	Wed, 26 Jul 1995 12:36:50 -0400
From:	rwolff@noao.edu (Richard Wolff)
Message-Id: <9507261636.AA19352@daikon.tuc.noao.edu.noao>
To:	9fans
Subject: Installing on second disk
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

I have a Gateway PC (486/66) with two 400 MB (IDE) disk drives.  I'd
like to install plan9 on the second drive (hd1) but retain there a
modest DOS partion (D:).  If I install the 4 disk "trial" system using
the orignal "prep", all goes well and I have a system I can boot from
hd1, except that, should I run DOS/Windows again, the D: partition is
gone (the partition table has been destroyed, presumably).  It doesn't
matter if I partition the disk with the D: partition at the start or
end of the disk.  On the other hand, if I try the same things using the
new prep, prep complains with
	error 23: Not enough room for boot area

Aside from the notion of moving most of what's on C: to D: and installing
plan9 on C:, can anyone suggest what I'm doing wrong or what I might try?

Richard



>From 9fans-outgoing-owner Wed Jul 26 16:20:17 1995
Received: by psuvax1.cse.psu.edu id <34100>; Wed, 26 Jul 1995 16:18:22 -0400
Received: from goonsquad.spies.com ([192.216.22.66]) by psuvax1.cse.psu.edu with SMTP id <34109>; Wed, 26 Jul 1995 16:02:15 -0400
Received: by goonsquad.spies.com (Smail3.1.29.1 #2)
	id m0sbC0q-00025bC; Wed, 26 Jul 95 12:21 PDT
Message-Id: <m0sbC0q-00025bC@goonsquad.spies.com>
From:	ahm@goonsquad.spies.com (Andreas Meyer)
Subject: Got it loaded. Now what?
To:	9fans (9fans)
Date:	Wed, 26 Jul 1995 15:20:59 -0400
Organization: The Internet Wiretap
Content-Type: text
Content-Length: 517       
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans


Okay, so I've grabbed the PC distribution from plan9.att.com
and seem to have it running reasonably well on my AT&T 6386/SX WGS.
(Although, it refuses to load on my 386DX40 clone. It says it doesn't see
my IDE drive?!).  

So, short of downloading the man pages _en masse_, how do I get 
more information?  I've printed out most of the documents
available on the Web page under Volume 2.

Thanks!
Andy
-- 
 Andreas Meyer, N2FYE       Sustain.  Endure.  Evolve.
 ahm@goonsquad.spies.com    http://www.spies.com/~ahm/



>From 9fans-outgoing-owner Wed Jul 26 18:48:20 1995
Received: by psuvax1.cse.psu.edu id <34104>; Wed, 26 Jul 1995 18:46:23 -0400
Received: from galapagos.cse.psu.edu ([130.203.2.12]) by psuvax1.cse.psu.edu with SMTP id <34107>; Wed, 26 Jul 1995 18:33:03 -0400
Received: by galapagos.cse.psu.edu id <12762>; Wed, 26 Jul 1995 17:40:04 -0400
From:	Scott Schwartz <schwartz@galapagos.cse.psu.edu>
To:	9fans
Message-Id: <95Jul26.174004edt.12762@galapagos.cse.psu.edu>
Date:	Wed, 26 Jul 1995 17:39:58 -0400
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Just out of curiosity, what sort of legal permission, if any, was required
to get Bela Lugosi et al's picture on the first edition cdrom?




>From 9fans-outgoing-owner Wed Jul 26 21:01:53 1995
Received: by psuvax1.cse.psu.edu id <34127>; Wed, 26 Jul 1995 21:00:28 -0400
Received: from blah.math.tu-graz.ac.at ([129.27.150.3]) by psuvax1.cse.psu.edu with SMTP id <34138>; Wed, 26 Jul 1995 20:42:39 -0400
Received: from blah.math.tu-graz.ac.at (bri@localhost [127.0.0.1]) by blah.math.tu-graz.ac.at (8.6.12/8.6.12) with ESMTP id AAA14661 for <9fans@cse.psu.edu>; Thu, 27 Jul 1995 00:16:00 +0200
Message-Id: <199507262216.AAA14661@blah.math.tu-graz.ac.at>
To:	9fans
Subject: first tries
Date:	Wed, 26 Jul 1995 18:15:59 -0400
From:	Brian Ward <bri@blah.math.tu-graz.ac.at>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Ok, so I tried to install the floppy distribution on one of my PCs today,
and here's what my experiences were (roughly):

9 doesn't seem to want to talk to my Maxtor 7546A drive at all. This is a
plain one-year-old 540MB IDE drive that works fine with linux, so there
shouldn't be any problem with using non-BIOS-oriented I/O. More precisely,
when I booted the first time, it said that it couldn't find my primary DOS
partition, which sure isn't terribly informative, but at least "you will
have to change your hardware configuration" was somewhere close to the
point. After fiddling around with the disk and booting it to a shell, I found
that there's no hint of a /dev/hd0disk, etc, even though #H had been bound.
To make sure that I wasn't going totally nuts, I tried the same floppy on a
variety of other IDE disks (which I can't use at moment because there's
stuff on them). All of the other disks were recognized (/dev entries
showing up, etc), so there's something particular with this one. I don't
know if it's something with Maxtor as a vendor, because one of the other
disks is also a Maxtor. So tomorrow I'll do some shuffling, to see if it
installs ok on another disk.

(I tried the disk on a few different controllers and computers.)

So then to my "opinionated" comments, which may or may not mean anything to
you:

There were seemingly no diagnostics at all telling me if the bind #H failed.
In fact, in cpurc, they're all > /dev/null >[2=1]ed out. I got rid of those
and still didn't get anything. The same goes for the mounts after that,
except that they actually say something when they fail -- they tell you the
command failed (but don't say that they failed because the device didn't
exist). A little more feedback from these commands would have been nice.

The install program looks like the MS-DOS install program (which I grudgingly
had to run for the first time ever today. more on this later.). Admittedly,
I haven't gotten very far into it because of the above, but it's definitely
not my definition of pretty.

And about MS-DOS: I do hope this requirement can be dropped soon. I found
setting up MS-DOS to be quite painful. I can understand the need for
bootstrapping at first (linux and the "others" also had to do the same
thing in the early days), but let's please not get too used to it. :-) In
fact, the linux boot-loader should even be able to boot plan 9 if you
dropped it enough hints on the floor.

It's of course fun to play around a little with the mini-system on the boot
disk. Maybe that's just me.

Last and certainly not least: I (finally) got to see Plan 9 From Outer
Space in a Viennese theater last Sunday. How very excellent! An experience
no one should miss out on.




>From 9fans-outgoing-owner Wed Jul 26 21:40:56 1995
Received: by colossus.cse.psu.edu id <45492>; Wed, 26 Jul 1995 21:39:45 -0400
Received: from localhost by colossus.cse.psu.edu with SMTP id <45491>; Wed, 26 Jul 1995 21:33:07 -0400
To:	9fans
Subject: Test message
Cc:	schwartz@daneel.cse.psu.edu
X-uri: <URL:http://www.cse.psu.edu/~ehrlich/>
X-Work-Address: Department of Computer Science and Engineering
		121A Pond Laboratory
		The Pennsylvania State University
		University Park, PA 16802
X-Work-Phone: +1 814 863 1142 (Voice) +1 814 865 3176 (FAX)
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:	Wed, 26 Jul 1995 21:10:38 -0400
From:	Daniel R Ehrlich <ehrlich@daneel.cse.psu.edu>
Message-Id: <95Jul26.213307edt.45491@colossus.cse.psu.edu>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Some people have reported problems with the 9fans mailing list.  I am trying to
determine what's wrong.  Please disregard this mail.

-- Dan Ehrlich	<postmaster@cse.psu.edu>




>From 9fans-outgoing-owner Thu Jul 27 07:39:09 1995
Received: by psuvax1.cse.psu.edu id <33979>; Thu, 27 Jul 1995 07:38:09 -0400
Received: from blah.math.tu-graz.ac.at ([129.27.150.3]) by psuvax1.cse.psu.edu with SMTP id <33964>; Thu, 27 Jul 1995 07:37:47 -0400
Received: from blah.math.tu-graz.ac.at (bri@localhost [127.0.0.1]) by blah.math.tu-graz.ac.at (8.6.12/8.6.12) with ESMTP id NAA14876 for <9fans@cse.psu.edu>; Thu, 27 Jul 1995 13:37:49 +0200
Message-Id: <199507271137.NAA14876@blah.math.tu-graz.ac.at>
To:	9fans
Subject: second try
Date:	Thu, 27 Jul 1995 07:37:49 -0400
From:	Brian Ward <bri@blah.math.tu-graz.ac.at>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Ok, so I tried on a different drive this morning and eventually got it to
work. The following information may or may not be useful:

I installed it on my notebook, which has a 340MB disk, 16MB RAM, and an
i486dx4/100. Linux was installed on this machine, but there was a 32MB
swap partition, so I had the bright idea to install the DOS and plan 9
partitions over that. I used one of my other linux boxen to prepare the
floppy (had to replace the "prep" command on it), and proceeded. You don't
need to have much of DOS at all on the disk; all you have to do is:

 - boot with 1st ms-dos disk and exit the setup program as soon as it starts
 - "fdisk" and make a tiny DOS partition (I used 4MB but it could be
    smaller, I'm sure)
 - computer will reboot when you exit fdisk (stupid hardware, stupid software)
 - do step 1 again (wince harder this time)
 - "format c: /s"
 - if you have linux installed on the disk, you need to do an fdisk /mbr
     (please see linux notes below!)
 - replace DOS trash-disk with plan 9 boot disk and type "b".

Here's where I ran into trouble. The place on the disk I was planning to
install plan 9 was in between my / and /usr linux partitions. The plan 9
boot disk booted and seemingly took the first partition on the disk (my /) 
to be the DOS partition and when I told it to install the plan 9 DOS files,
it waltzed right over that. I suspected this when I rebooted and there was
no plan9 directory on the DOS partition. I knew it when I tried to boot
linux from floppy. So we now know that plan 9's install doesn't seem to
know the right way to find the "primary DOS partition."

The moral of this story is to back up everything on the disk beforehand even
if it is on another partition, because plan 9 will zing it ruthlessly --
even though I installed using the "fixed" prep command, it still went ahead
and zapped my /usr partition. I have a feeling that the fix was for
multiple partitions directly after the DOS trash (which seemingly has to be
the first on the disk) and not anything that may be after where you intend
to put plan 9. I'll try to confirm this later.

Ok, so now after / was gone, I just went ahead and put the DOS partition at
the beginning of the disk (as I described above) and went ahead with the
install. ("C'mon Bullwinkle.. that trick NEVER works.." "This time for
sure, Rocky!") Ok, this time it did work. However:

I couldn't get it to recognize the built-in PS/2 trackball or a serial Mouse
Systems mouse. Finally, I grudgingly pulled the Logitech MouseMan out of
my other Linux box, and that did work. I suppose I have to buy another
MouseMan, anyway (they cost $65 here, you know.). It didn't do 640x480x8
vga right, but it does do monochrome ok, as it should. I'm not surprised by
this because my notebook has a Cirrus 6440 chipset and that's still kind of a
rare beast (although XFree86 3.1.1 supports it). All of this fiddling
around meant a lot of rebooting and changing the plan9.ini file (through
the plan 9 install program, probably a silly thing to do). When booting the
install program, the kernel correctly says that I have a 2-slotted intel
xxxxx PCMCIA chipset in the notebook, but when I put in my 3Com 3C589 and
configure it in, it screws up and identifies it as a 1-slot Cirrus Logic
controller. Oh well, I'll see about it later.

It's really cool. In particular, acme feels great and is extremely
well-done. I can't find my break key (tried plokta) - don't know if this is
due to the smaller notebook keyboard or what. Anyway, it's an excellent
job, but I'd definitely recommend getting another hard disk before
installing.




>From 9fans-outgoing-owner Thu Jul 27 08:32:33 1995
Received: by psuvax1.cse.psu.edu id <33985>; Thu, 27 Jul 1995 08:31:53 -0400
Received: from plan9.att.com ([192.20.225.252]) by psuvax1.cse.psu.edu with SMTP id <33979>; Thu, 27 Jul 1995 08:31:36 -0400
From:	rob@plan9.att.com
To:	9fans
Date:	Thu, 27 Jul 1995 08:31:11 -0400
Message-Id: <95Jul27.083136edt.33979@psuvax1.cse.psu.edu>
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans

Those of you who received the first distribution were asked to mail
plan.9@research.att.com with questions.  We are turning off that
mail alias; use this 9fans list as a first place for queries from now on.
Thanks.




>From 9fans-outgoing-owner Thu Jul 27 15:59:14 1995
Received: by colossus.cse.psu.edu id <45588>; Thu, 27 Jul 1995 15:56:44 -0400
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Organization: Computer Science and Engineering, Penn State University 
From:	ehrlich
To:	9fans-outgoing@cse
Subject: Test email
Reply-to: ehrlich
Versions: dmail 1.9d/makemail 2.3
Message-Id: <95Jul27.155644edt.45588@colossus.cse.psu.edu>
Date:	Thu, 27 Jul 1995 15:56:42 -0400


I have heard that folks are not receiving messages from 9fans@cse.psu.edu.
If you are reading this, please reply and let me know.

Thanks,
Dan Ehrlich	<postmaster@cse.psu.edu>
--
Dan Ehrlich - Systems Analyst - PSU Computer Science and Engineering
"Universities should be safe havens where ruthless examination of realities
 will not be distorted by the aim to please or inhibited by the risk of
 displeasure." - Kingman Brewster
2.6 <EE26E805> fingerprint =  5C 01 7F 57 B0 AB 68 72  04 23 B9 BD 27 AD 85 60 
echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc


>From 9fans-outgoing-owner Thu Jul 27 18:23:30 1995
Received: by colossus.cse.psu.edu id <45595>; Thu, 27 Jul 1995 17:57:52 -0400
Received: by colossus.cse.psu.edu id <45594>; Thu, 27 Jul 1995 17:57:36 -0400
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Organization: Computer Science and Engineering, Penn State University 
From:	ehrlich
To:	9fans
Subject: Mailing list meltdown
Versions: dmail 1.9d/makemail 2.3
Message-Id: <95Jul27.175736edt.45594@colossus.cse.psu.edu>
Date:	Thu, 27 Jul 1995 17:57:34 -0400
Sender: owner-9fans
Precedence: bulk
Reply-To: 9fans


It seems that sometime between 15:00 on July 18th and 09:45 on July 19th
the file that holds the 9fans mailing list got truncated to zero bytes.

I have restored the list as of the 18th and added anyone who subscribed
since then.  Sorry for the inconvienece.  All messages are stored in the
archive.  So just send majorodomo@cse.psu.edu a message containing

	index 9fans

for a current list.

-- Dan Ehrlich	<postmaster@cse.psu.edu>
--
Dan Ehrlich - Systems Analyst - PSU Computer Science and Engineering
"Universities should be safe havens where ruthless examination of realities
 will not be distorted by the aim to please or inhibited by the risk of
 displeasure." - Kingman Brewster
2.6 <EE26E805> fingerprint =  5C 01 7F 57 B0 AB 68 72  04 23 B9 BD 27 AD 85 60 
echo '[q]sa[ln0=aln256%Pln256/snlbx]sb3135071790101768542287578439snlbxq'|dc

No subject

[rob]

Those of you who received the first distribution were asked to mail
plan.9@research.att.com with questions.  We are turning off that
mail alias; use this 9fans list as a first place for queries from now on.
Thanks.

No subject

[Scott]

Just out of curiosity, what sort of legal permission, if any, was required
to get Bela Lugosi et al's picture on the first edition cdrom?

No subject

[Henner]

Hello again!

Here I am again, looking for the right configuration...

Has anybody tried to run Plan9 on a Gigabyte GA-586AT(/P256) board?

The Bi-processor boards with the Neptun chip set used by the AT&T folks
are nice, but the Trition boards are faster AND cheaper!

Any experiences?

Tschuess,

	Henner

PS: I hope this will be the last question about the `right' configuration.
--
===============================================================================
 
Henner Gratz                                 Email:       a82@lrw.uni-bremen.de
Leeuwarder Str. 16A                          Fax (at home): +49-(0)421-58 52 10
D-28259 Bremen                               Tel (at home): +49-(0)421-58 51 84
GERMANY
 
===============================================================================

No subject

[Henner]

Hello everybody!

Now, after Plan9 is available for everyone, I plan to order my CD-ROM from AT&T.
I want to install it on a 486/Pentium PC. Because I don't own such a beast at
the moment, I will have to buy one. And here the problems start. On what
main-boards, graphic-cards etc. Plan9 will run. The WWW-pages on plan9.att.com
aren't very detailed. I have also read the lists spread in the 9fans mailing
list, but they are from 1994 for the old distributions. 

So here's my question: Is there a list covering the new distribution?
	
Thank you very much for any answers!

Regards,

	Henner

--
===============================================================================
 
Henner Gratz                                 Email:       a82@lrw.uni-bremen.de
Leeuwarder Str. 16A                          Fax (at home): +49-(0)421-58 52 10
D-28259 Bremen                               Tel (at home): +49-(0)421-58 51 84
GERMANY
 
===============================================================================

No subject

[Frank]

>>Anyone have any ideas on how to get the discount price?
>place the order through a local bookshop, giving them the AT&T contact number
and ISBN?


Hmm, I guess I'm todays bone head award candidate.

We are going to need the publishers name, and the exact title to
order. I'll track it down, if I can find a bookstore that will sell if for
a reasonable markup I'll post a contact and phone #.
 

Frank Gleason
frankg@halcyon.com

No subject

[rob]

Someone jumped the gun putting the information into the Web page
before the printer/distributor had its end in order.

By tomorrow things should have settled down.  The price will be $350
and availability is whatever they say, but I hope it's less than three weeks
since we have copies in hand.

No subject

[philw]

>I called to order the package and the price for an individual is
>$465, they will sell for $350 if you are a bookstore or other blessed
>group. Anyone have any ideas on how to get the discount price?

This is a mistake. It will be fixed today.

No subject

[Frank]

Subject: release info, more info

Oh, I forgot to add that the books are not back from the
printers yet. They say about 3 weeks.

Frank Gleason
frankg@halcyon.com

No subject

[Frank]

Subject: release info

I got this from distrib.html on plan9.att.com:

>The distribution will be available on July 18th.
>The contact number for ordering in the U.S., Canada,
>and the Caribbean is 800-841-9938, and +1 407-345-3800
>elsewhere.
>The cost for the full kit is $350 plus postage,
>ISBN number 0-03-017143-1.
>The manuals can be ordered by themselves for
>$125 plus postage, ISBN number
>0-03-017142-3.


I called to order the package and the price for an individual is
$465, they will sell for $350 if you are a bookstore or other blessed
group. Anyone have any ideas on how to get the discount price?

Frank Gleason
frankg@halcyon.com

No subject

[rob]

here is a more thorough but still rudimentary fix to the eqn problem.
david hogan's fix solves the problem but doesn't address the issue.
as a side note, the textc() routine illustrates why plan 9 doesn't by
default use ansi's specification for multibyte and wide characters.

eqn% diff text.c /n/dump/1995/0401/sys/src/cmd/eqn
40,57d39
< int textc(void)	/* read next UTF rune from psp */
< {
< 	wchar_t r;
< 	int w;
< 
< 	w = mbtowc(&r, psp, 3);
< 	if(w == 0){
< 		psp++;
< 		return 0;
< 	}
< 	if(w < 0){
< 		psp += 1;
< 		return 0x80;	/* Plan 9-ism */
< 	}
< 	psp += w;
< 	return r;
< }
< 
92c74
< 		for (psp = p1; (c = textc()) != '\0'; ) {
---
> 		for (psp = p1; (c = *psp++) != '\0'; ) {
115,124d96
< int isalpharune(int c)
< {
< 	return ('a'<=c && c<='z') || ('A'<=c && c<='Z');
< }
< 
< int isdigitrune(int c)
< {
< 	return ('0'<=c && c<='9');
< }
< 
129c101
< 	if (isalpharune(c) && ft == ITAL && c != 'f' && c != 'j') {	/* italic letter */
---
> 	if (isalpha(c) && ft == ITAL && c != 'f' && c != 'j') {	/* italic letter */
134c106
< 	if (isalpharune(c) && ft != ITAL) {		/* other letter */
---
> 	if (isalpha(c) && ft != ITAL) {		/* other letter */
139c111
< 	if (isdigitrune(c)) {
---
> 	if (isdigit(c)) {
290c262
< void cadd(int c)		/* add character c to end of cs */
---
> void cadd(int c)		/* add char c to end of cs */
293d264
< 	int w;
315,317c286
< 	w = wctomb(csp, c);
< 	if(w > 0)	/* ignore bad characters */
< 		csp += w;
---
> 	*csp++ = c;

No subject

[Scott]

I was just playing with ftpfs, grepping a huge file from gatekeeper.dec.com,
watching the ftp cache grow and grow when my sparcstation went catatonic.
Is this a bug? :-)  What sort of bug report would be helpful in a case
like this? 

No subject

[Scott]

Just wondering... what do folks without WORM jukeboxes do for
backups?  I've just been copying important files over to 
Unix, but that clearly doesn't scale, and doesn't work well
if you have just one machine (at home, say.)  Tar doesn't
do multivolume dumps, so I guess you just need a big DAT
and small disks.

-- Scott

No subject

[Scott]

SUBSCRIBE srhansen@gloria.cord.edu

If this message was sent to the wrong address - first of all sorry!  
Secondly, some guidance would be appreciated!

-Scott

    ///// ///// ///// ///// ///// | Email:  hansen@cobber.cord.edu
   /     /     /   /   /     /    |	    srhansen@gloria.cord.edu
  ///// /     /   /   /     /     | 	    scott@triston.cord.edu
     / /     /   /   /     /      | Addrs:  P.O. Box 895, Moorhead, MN 56562
///// ///// /////   /     /       | Phone:  (218) 299-4595

No subject

[Scott]

Listen gang, please don't turn 9fans into linux-advocacy.  We all 
know the story with GPL, so don't bother rehashing it here.

No subject

[Bob]

In message <95Apr6.132314edt.34060@psuvax1.cse.psu.edu>, rob@plan9.att.com writ
es:
>> Go on. What happened to it? 
>
>they generate raw quadrature, but modern machines use serial input.
>the company that makes them sells a serial version, but only two buttons
>work.
>
Why not plug the Depraz mouse into the parallel port and write a suitable
driver?

Bob.

No subject

[rob]

there's been a lot of attention recently to the files in ftp://plan9.att.com/plan9/doc,man.
they are from the old distribution; they are in some cases 5 years old.  we're preparing
documents for the new distribution. when they are in place we will let you know.
in the meantime, you probably don't want to grab what's there.

-rob

No subject

[rob]

Details about the release are still unclear.  There are still some questions
about pricing, distribution, licensing, etc.  I will say something here as soon
as there is something firm to say.  It won't be long now.

No subject

[Amos]

It takes a lot of disk space to keep all the sources, but except that,
Plan9 can run on anything (I have seen it run on a laptop too).

--
	Amos Shapir		Net: amos@cs.huji.ac.il
Paper: The Hebrew Univ. of Jerusalem, Dept. of Comp. Science.
       Givat-Ram, Jerusalem 91904, Israel
Tel: +972 2 585706,586950        GEO: 35 11 46 E / 31 46 21 N

No subject

[Scott]

If someone were to get a PC for his birthday and wanted to run Plan 9
on it, what would you all recommend getting?  I've read Paul Vixie's
web page full of recommendations with respect to BSD/OS, and the Linux
FAQ; how much of that applies?

No subject

[Scott]

So are we agreed then?  comp.os.plan9, moderated, gatewayed
to 9fans.  We had a charter posted, now all we need is a moderator,
and someone to post the charter and request for discussion.

No subject

[philw]

I suspect we are unlikely to read
an umoderated group. They drift off
subject too much.

No subject

[rob]

and you wonder why we want a moderated newsgroup

>From cse.psu.edu!9fans-outgoing-owner	Fri Mar 31 09:14
From:	"Dirk Vleugels" <vleugels@do.isst.fhg.de>
In-Reply-To: Geoff Langdale <Geoff_Langdale@GS10.SP.CS.CMU.EDU>
        "comp.os.plan9" (Mar 30,  1:32pm)
To:	9fans@cse.psu.edu
Subject: Re: comp.os.plan9
Sender: owner-9fans@cse.psu.edu
Reply-To: 9fans@cse.psu.edu

On Mar 31,  3:29am, Heiko Wengler wrote:
> Subject: [CLUE-PICKUP:617] ET vs World this sunday at 1pm EST
> Hi.
>
> ET wants to face the world this sunday at 1pm EST/8pm CET
> on
>
> -h fisher.psy.vu.nl -p 4577 ( 5000 obs)
>
> CU on sunday!
>
> 	Heiko
> 	Captain ET
>
> PS: Please be ontime.
>-- End of excerpt from Heiko Wengler

Hi.

Wie war bitte die Raumnummer von future? Und habt ihr da telefon?

Dirk

No subject

[Berry]

>>>"John C. Orthoefer" said:
 > 
 > > [news group]
 > 
 > I'm not sure why you feel it needs to be moderated.

I nominate Elizabeth Bimmler as comp.os.plan9 moderator.

  --berry

Berry Kercheval :: Xerox Palo Alto Research Center

No subject

[Frederick]

>First, we'll be looking for a few people to test our installation setup on
>their PCs
>to help us ferret out obvious problems before we publish the software.
>We can't handle too many requests (we're up to our ears as it is) but if you're
>interested, send me mail and in a few days I'll send (maybe not all of) you a
>a distribution setup to try.

We are very interested. We are two students working in the Computer Science
field and our diploma thesis concerns about Plan 9, so it would be very
interesting for both of us. As a matter of fact we use the PC's right now.

Thanks,

Technical University of Berne
Frederick L. Born and
Juraj Kis
Brunnadernstrasse 3
CH-3006 Berne, Switzerland



| Frederick L. Born                           |                          |
| Brunnadernstr. 3                            | In this world, time is   |
| CH-3000 Bern 6   PHONE: ++41 31 / 352 58 25 |   a local phenomenon.    |
| Switzerland      EMAIL: born@dial.eunet.ch  |                          |

No subject

[Mark]

>First, we'll be looking for a few people to test our installation setup on
>their PCs
>to help us ferret out obvious problems before we publish the software.
>We can't handle too many requests (we're up to our ears as it is) but if you're
>interested, send me mail and in a few days I'll send (maybe not all of) you a
>a distribution setup to try.

Love too.  We have a bunch of PCs around here that we have been been using
for portability testing.  From noname ISA bus 386/33 to PCI bus Pentium.
I would be happy to give you stuff a try on the different platforms which
we have around here.

>Second, we really don't want to be in the support business, and we're hoping
>we can use the net to help us out here.  It would be nice if there was a
>moderated
>newsgroup set up where we could get an FAQ established right from the
>beginning.
>We're looking for someone outside to volunteer to set that up and moderate it.
>Any takers?


I would think that you wouldn't need the newsgroup to be moderated at the
beginning, I expect the traffic will be managable.

--Mark Verber
  Xerox PARC

No subject

[Scott]

| I'm not sure why you feel it needs to be moderated.  At the moment I
| don't believe that there will be a huge number of people who wil begin
| running plan 9.  

Yeah, but all the rest will want to talk about it. :-)

| Although we could use the members of this list as yes votes
| (are we over 100 members yet?)

190, according to majordomo.  Quite a few of those are exploders
to other lists.

No subject

[John]

> [news group]

I'm not sure why you feel it needs to be moderated.  At the moment I
don't believe that there will be a huge number of people who wil begin
running plan 9.  

Most people who are willing to work with a new OS are skillful enough
figure out most problems.  At least I would hope so.  I would think
the list as it is would work fine.

We could gateway this list into an alt group with out much problem.  I
could do that here at BBN.  A comp group would take a month or so to
create.  Although we could use the members of this list as yes votes
(are we over 100 members yet?)

Johno


-
John Orthoefer   | Take this out and a Unix Demon will dog your steps from 
<jco@bbn.com>    | now until the time_t's wrap around.
617-873-6188     |  -- Curse from the tunefs(8) man page source

No subject

[Christopher.Vance]

|  First, we'll be looking for a few people to test our installation setup on their PCs
|  to help us ferret out obvious problems before we publish the software.
|  We can't handle too many requests (we're up to our ears as it is) but if you're
|  interested, send me mail and in a few days I'll send (maybe not all of) you a
|  a distribution setup to try.

I'm interested.  In fact you suggested I get this one.  (I've tried with 
the previous one, but haven't got it networking off the PC floppy yet.)  
Time for some packet snooping or other debugging.
 
|  Second, we really don't want to be in the support business, and we're hoping
|  we can use the net to help us out here.  It would be nice if there was a moderated
|  newsgroup set up where we could get an FAQ established right from the beginning.
|  We're looking for someone outside to volunteer to set that up and moderate it.

Setting up a newsgroup (whether moderated or not) in the main hierarchies 
(say comp.os.plan9) takes a fair bit of time and effort - there needs to 
be discussion, voting, and that sort of thing.

One could try to set up a group under alt.*, but it may have propagation 
problems, as many sites refuse to carry them or delay creation until 
somebody complains...

If it's just a FAQ, it can be posted in a number of existing groups.  If 
there's frequent new material, or lots of questions, it probably deserves 
a real newsgroup, bnut may not need moderation.  I would suggest 
comp.os.plan9, but haven't started a comp.* proposal before.

An alternative might be to make a moderated mailing list.  If run under a 
list manager like majordomo, it might not be too hard to set up or 
maintain.  I could probably arrange that here.

-- Christopher

No subject

[Cort]

Actually I have few other interestes, I'll volunteer.

} interested in setting up the faq...how much work is it likely to be?  I
} have some spare time, but I have other interests too.

No subject

[Steve]

> we'll be looking for a few people to test our installation setup

I'd like to be involved with that. I just got a 486 with an Adaptec
scsi controller and 1Gig disk that I was planning to load Plan 9 onto.

I'll also volunteer to maintain the FAQ (I already do another faq),
but I'll leave it to someone else to handle the discussion and voting
to get the newsgroup established.

No subject

[Cort]

I'm interested in setting up the faq...how much work is it likely to be?  I
have some spare time, but I have other interests too.

No subject

[Andrew]

From: rob@plan9.att.com
Subject: 
Date: Wed, 29 Mar 1995 20:18:24 -0500

> interested, send me mail and in a few days I'll send (maybe not all of) you a
> a distribution setup to try.

I'm interested.

Andy Cowell
1533 Laurel Ave, Apt 16
Knoxville, TN 37916

--
Andrew E. B. Cowell <cowell@cs.utk.edu> | "And the mountainside opened, a
Sys Admin, Computer Science Department  |  moment to pray for all the souls
The University of Tennessee, Knoxville  |  he'd come to save...now he couldn't
WWW: http://www.cs.utk.edu/~cowell/     |  save himself" [Legendary Pink Dots]

No subject

[rob]

The new distribution is imminent, and that brings up a couple of questions.

First, we'll be looking for a few people to test our installation setup on their PCs
to help us ferret out obvious problems before we publish the software.
We can't handle too many requests (we're up to our ears as it is) but if you're
interested, send me mail and in a few days I'll send (maybe not all of) you a
a distribution setup to try.

Second, we really don't want to be in the support business, and we're hoping
we can use the net to help us out here.  It would be nice if there was a moderated
newsgroup set up where we could get an FAQ established right from the beginning.
We're looking for someone outside to volunteer to set that up and moderate it.
Any takers?

Thanks to all....

-the plan 9 guys

No subject

[Scott]

Ok, /proc/n/wait sounds good.

No subject

[philw]

>What's the recommended way to go about coding modules that have to fork
>and wait and that don't stomp on each other?

There is now a wait file /proc that allows this. Which
does some really cool things (rob said this). You can
wait for anything based on file permission regardless
of who started what.

No subject

[Scott]

Phil writes:
| >Question for the Labs guys... why no waitpid()?  Simulating that
| none of us want it.

I guessed that part; I was wondering *why*.  (Calling it rwait() would
be more parimonious, I suppose.)

What's the recommended way to go about coding modules that have to fork
and wait and that don't stomp on each other?

No subject

[philw]

>Question for the Labs guys... why no waitpid()?  Simulating that
none of us want it.

No subject

[Scott]

Question for the Labs guys... why no waitpid()?  Simulating that
requires keeping a list of reaped children, and making sure that
everyone who calls wait checks that list first, and everyone who calls
fork clears it out; no such convention seems to be current, though.

No subject

[mccartyb]

	id m0rpMe4-001YzxC; Thu, 16 Mar 95 12:59 PST
Date: Thu, 16 Mar 1995 12:59:48 -48000
From: Bill McCarty <mccartyb@linux>
Subject: 
To: 9fans@cse.psu.edu
Message-ID: <Pine.3.89.9503161208.D5115-0100000@linux>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

get v00.n001
get v01.n001
get v01.n002
get v01.n003
get v01.n004
get v01.n005
end

----------------------------------------------------------
Bill McCarty, Associate Professor
Department of Computer Science
Azusa Pacific University               Ph.    818 815-5311
901 East Alosta Avenue                 Secty. 818 815-5310
Azusa, CA 91702 USA                    Fax    818 815-5323

No subject

[rob]

Here's a thrilling chance to gain fame by contributing
useful stuff to the new release.  If you have a file for
/adm/timezone that is not one of GMT or American
EST, CST, or PST, send it to us and we might include it.

-rob

No subject

[rob]

It's nice to be in a position where people apologize because they
assume there's humor in your work, based on past experience,
but they're not sure where it is.

No subject

[rob]

	failed.c:7 syntax in macro expansion: x

I don't understand why you keep beating on this.
The documentation states clearly that the C compilers
do not honor ANSI's standard for preprocessing.
It also says that if you want a real preprocessor,
use cpp, which is provided.

One reason is protest: the ANSI preprocessor
definition is needlessly baroque.  The real reason,
though, is simplification,  not only in the compiler,
but in the programs that result.  I could argue back
that making a variable have the same name as a
macro is bad style.  If you want to test whether the
letter of  the law is met by the preprocessor in the
C compiler, you're going to be disappointed.  If you'd
like a fast compiler that discourages tricky
preprocessing, you'll be happy.

As for us, we're happy.

No subject

[Fariborz]

>  From:	yodaiken@sphinx.nmt.edu (Victor Yodaiken)
>  Date:	Thu, 8 Sep 1994 15:44:16 -0400
>  reply_to: yodaiken@chelm.nmt.edu
>  X-Mailer: Mail User's Shell (7.2.5 10/14/92)
>  Sender: owner-plan9-fans@cse.psu.edu
>  Precedence: bulk
>  Reply-To: plan9-fans@cse.psu.edu
>
>  What is the status of any new plan9 distribution. I am considering
>  incorporating Plan9 into several of our courses and would be
>  trilled to learn something more about possible futue developments?

How about a public release of Plan9 at the same time that this new
film about Ed Wood comes out?

Just a thought

No subject

[Victor]

What is the status of any new plan9 distribution. I am considering
incorporating Plan9 into several of our courses and would be
trilled to learn something more about possible futue developments?

No subject

[bobf]

> #include <alef.h>
> 
> void
> main(int argc, byte **argv)
> {
> 	 rescue  {
> 		print("1");
> 		return;
> 	 }
> 	 rescue  {
> 		print("2");
> 		raise ;
> 	 } 
> 	 raise ;	
> }

this is a compiler bug that has been fixed.

> enum {TEXT, AUDIO, VIDEO, MULTI};
> 
> void (*f[]) = { /* I'm not sure about f or Mesg.f */
> 	[TEXT]  Mime.ldtext(),
> 	[AUDIO] Mime.ldaudio(),
> 	[VIDEO] Mime.ldvideo(),
> 	[MULTI] Mime.ldmulti(),
> };
> 
> and later do (*f[type])(); /* this avoid switch or if else */

use the form .<typename>.function to refer to the functions.

	[TEXT]  .Mime.ldtext,
	[AUDIO] .Mime.ldaudio,
	[VIDEO] .Mime.ldvideo,
	[MULTI] .Mime.ldmulti,

works with our compiler.

> adt Mime{
> 	.....
> 	void write(*Mime,int);
> };
> 
> adt Mesg {
> 	.....
> 	Mime;
> 	void write(*Mesg,int);
> };
> 
> If I want to access Mime.write through a Mesg object, Is it
> save to use casting? 
>
> void
> Mesg.write(Mesg *m,int)
> 
> 	((Mime *)m)->write(0); /* in order to call MIME.write */
> }

unnamed members can be referenced by their type name.  use

	m->Mime.write(0);



we are working on a new plan 9 distribution.  it contains an Alef
compiler with many bug fixes, some new features, and improved
documentation.

No subject

[rob]

the file /dev/window is a bitmap of the current window.
the file /dev/screen is a bitmap of the whole screen.
these files are documented in bit(3) and 8½(4).
their format is documented in bitmap(6).

-rob

No subject

[rob]

salus got the story just wrong. he appears to have conflated
the sale of USL to Novell with rumblings about a general
release of plan 9.  the book perpetuates the error, but he
has promised to correct the mistake in the next printing.

No subject

[forsyth]

>>Plan 9 on intel. Can you set up an auth server, file server using intel? 
>>Last I heard, there was support only for the mips/sparc machines as 
>>fileservers.

it isn't on the CDROM, but given /sys/src/9/pc and a little background
knowledge, producing /sys/src/fs/pc wasn't too hard.
producing 9pccpu (CPU server configuration) is even easier.  any cpu server
can be used as an authentication server, so that's that!

(thanks to the Labs we are back in the `use the source' era!)

a 486dx2/66 seems quite a bit faster than a Sparc Classic; i use a dx2/66
as a CPU server, and even with a 24Mb classic as a terminal i tended
to do (cross)compilations on the 486.  i use a very lowly 16Mb 486sx33
as a file server (it doesn't use floating point).  it's adequate for
a small number of users (certainly not the bottleneck at present).

>>The problem with PCs is that if users have them on their desks, most of them 
>>want/demand Windows.

i was expecting that many people here would replace old Suns by PCs to run Windows,
because they don't program much (or they like Visual Basic),
so the intention wasn't to preclude that,
BUT for those who preferred a Unix environment i thought it might be
nice to consider more interesting hardware than Sun provides.

of course, the PC configuration will be chosen to allow people to
run Linux, Amoeba and of course Plan 9 if that takes their fancy.
as cs.wisc.edu!bob observed, the faster Intel and non-Intel clones are
competitive in performance and usually much cheaper than equivalent
RISC machines:

>>I highly recommend the new Pentium machines. For under $US4k you
>>can get a machine that is very powerful (~40mips, 32Meg mem, 540M disk).

it's also nice on the PC to be able to plug in a wide range of cheap special-purpose
cards (although register-level programming information isn't always available).
furthermore, most fast PCs come with a reasonable amount of fast
second-level cache, which can make a difference.

even so, i don't think the 386/486 architecture has got enough truly
general purpose registers for some things i'd like to do
(not really enough for basic bitblt, come to that).
the 64k banking of most non-PCI graphics cards is also a disincentive, and
makes graphics implementation harder and slower.  on a more personal level,
i get dizzy every time i get near the register level programming of SVGA.

90 and 100Mhz Pentium machines, especially with PCI,
should make very good, cheap file and CPU servers.
there are even multi-processor variants.
in those applications weak graphics performance matters not a bit.

unless you are considering graphics or multimedia work (both of
which interest me a bit) and possibly floating-point
intensive work (traditionally a weak area for Intel machines),
the underlying awfulness of the Intel architecture doesn't appear at all
with either Plan 9 (or Linux, come to that), and with those
exceptions i'd certainly consider a fast PC as a Plan 9 terminal.
(in fact, i use a 486 at home.)
it is indeed hard to match the price.

No subject

[Scott]

Phil writes:
| However, while the file
| system could be modified to do this it does not seem unreasonable
| to expect the application programmer to print adequate information
| especially with the werrstr library function and %r format.

I agree.  I want to do even better and print (or at least read) messages
that say which component caused the error.  As you say, that requires
the fileservers to follow some convention for communicating the information
back.  I was imagining hacking the print routines to clip off the 
proposed trailing :number and return it in some other % verb, but even 
without that it would be an improvement just to see the raw information.

(I mistakenly used perror in the example before.  I really was thinking 
about the %r format, but my fingers are more used to typing the other.)

No subject

[philw]

>From: euler.Berkeley.EDU!serge
>
>>Brazil allows file systems to be mounted with a recover option
>>which does exactly what you ask.
>
>Pardon me, can you tell me what Brazil is?  Thank you very much.

Brazil is the name of the research version on Plan 9 we are currently
working on.

>From:	Scott Schwartz <groucho.cse.psu.edu!schwartz>
>
>One annoying thing in Unix is when you say something like
>open("/no/such/directory/or/file",0), it tells you that it something
>doesn't exist, but doesn't say what, which sometimes makes for
>uninformative error messages.  Now that we have errstr, instead of just
>errno, it would be delightful if it would encode something about
>what went wrong, so that perror could say something like
>	open: failed because directory 'such' does not exist

This has nothing to do with the plan 9 kernel. The errstr is produced by
whatever file system you are connected to. However, while the file
system could be modified to do this it does not seem unreasonable
to expect the application programmer to print adequate information
especially with the werrstr library function and %r format.

No subject

[philw]

The other day, just for fun, I rebooted my file server while I was
still logged on to the terminal and cpu.  Very shortly they noticed
that the connection had timed out, and required rebooting themselves.
In a minimal configuration like this, that's not a big deal, but what
does one do in a larger configuration, with several cpu and file
servers?  NFS has lots of shortcomings, but its one good point is that
recovery is automatic.

Brazil allows file systems to be mounted with a recover option
which does exactly what you ask. The recovery is entirely transparent
to the user.

No subject

[rob]

the sun is perhaps a poor choice for this test because its
mmu and caches are pessimal for anything involving
lots of processes.  the cost of flushing caches and mmu
will dominate any operating system performance.
in other words, the hardware is so expensive that
software hardly matters.

i tried the test on our challenge multiprocessors.

10,000 fork/exit/wait.

plan 9: real time 1.32 seconds.
unix: real time 41.1 seconds.

the unix system had more load, but was still
pretty quiet.  system time was 27.9 seconds.

on our seven-year-old SGI power machines,
the test takes 6.26 seconds.  the time is about
4 seconds on a single-processor MIPS magnum
of the same vintage, so the multiprocessor
part isn't very important.

now scott's test was 1,000 forks on a sun; my test
was 10,000.  on our at&t gnots, with 25MHz 68020's
and awful mmu, it takes 75 seconds to do 10,000.
our modern sparcstation-2 machine takes 34 seconds.
(a newer machine than the magnum that takes 4 seconds.)
you see, suns suck.  they really do.  the processors are
borderline OK but the rest of the system hurts.  yesterday,
quite independently, i was doing some tests of large-scale
memory bandwidth and the suns were laughable.

as for compiler names, the scheme we have is simple,
consistent, and parsimonious.  if crypticness is the
complaint, remember you're talking about a C compiler.

No subject

[Vijay]

This is the deal here:

---cut
Subject: Re: prices for Entre PCs
	[too expensive for our taste]

The only options available are 1) to wait till summer. We will be doing a
new rfp and awarding the contract to a mainstream player yet to be decided.
2) bid something they don't have. I just got in a pentium a few weeks ago
because they weren't selling them at the time, or 3) Look at buying a
machine that is officially supported by something like nextstep or
solaris and state on the requisition you are planning to run that OS
---cut

I have approval to get 4 machines but the local dealer with wich we have 
a contract with is gouging us, so to get out of the coils, we have to get 
machines that are NeXTSTEP/Solaris certified, and most of these machines 
come with weird mainstream cards.  I need your help in getting a list of 
video cards that work.  I am not that good (yet) with intel hardware and 
would prefer the minimum of problems getting it up to run first.

So if you have any YES - This card runs, please mail me it.  I would 
really love if the ATI UltraPro or something like that worked.  Diamond 
is not too well liked here so thats out.  I was hoping for something like 
the list of supported cards direct from the horses mouth so to speak.

The only NeXTSTEP machine that has ET4000/W32 etc chipset is the NEC 
Image series, which are not as much bang for the buck as Gateway 2000 or 
the other clones.  I know Compaq Qvision probably won't work, so we can't 
get the Compaq Deskpro's. (Local bus is good)

Any idea what hardware the plan9 folks are running (x86).

Vijay Gill                         |The (paying) customer is always right.
wrath@cs.umbc.edu                  |                  - Piercarlo Grandi
vijay@gl.umbc.edu                  | Eagles may soar, but weasels don't get
These are my opinions only.        | sucked into jet engines.

No subject

[Sean]

There is a problem with the plan9 memory management system for the PC
due to one of the brain dead features of the 386 architecture.
In particular, the R/W bit in a user page table entry is ignored when
the processor is in supervisor mode.  This means that copy on write will
not work when the kernel is copying into user space.
For example the following code segment exhibits the problem.

#include <u.h>
#include <libc.h>

char    buf[1024];

main()
{
        int i;

        if (fork() == 0) {
                /* make sure memory is allocated RONLY */
                i = buf[0];

                read(0, buf, 1024);
                print("%s\n", buf);
                exits(0);
        }

        /* wait a second so the child completes */

        sleep(1000);

        buf[1023] = 0;

        print("parent -> %s\n", buf);
}

On plan 9 for the pc, the data read into buf will incorrectly appear in both
the child and parent process.

The 486 (and higher?) fixed the problem but to maintain compatibility
it defaults to the 386 behavior.  On a 486 you need to set bit 16 in CR0
to get the correct behavior for the R/W bit.  Of course, if you have a
386, then copy on write can not be fixed, but you can still run plan9
with a copy on reference scheme... unfortunately, this means your 386
machine runs slower and requires more memory...

The complete fix of the problem, thanks to Dave Presotto, follows:

Here's the changes.  It works on a couple of 486's and 386's so its
probably right.

In l.s (around line 131) change

        ORL     $0X80000000,AX
        ANDL    $~(0x8|0x2),AX  /* TS=0, MP=0 */

to

        ORL     $0X80010000,AX
        ANDL    $~(0x40000000|0x20000000|0x8|0x2),AX    /* CD=0, NW=0, TS=0, MP=0 */

This also turns on internal caching in case it was off.

In l.s add the routine

/*
 *  return cpu type (what is a pentium?)
 */
TEXT    x86(SB),$0

        PUSHFL
        MOVL    0(SP),AX
        ORL     $0x40000,AX
        PUSHL   AX
        POPFL
        PUSHFL
        POPL    AX
        ANDL    $0x40000,AX
        JZ      is386
        MOVL    $486,AX
        JMP     done
is386:
        MOVL    $386,AX
done:
        POPFL
        RET

This returns the number 386 or 486 depending.  I guess it'll return
586 also when I get a chance at a Pentium.

In fns.h add a definition for it:

        int     x86(void);

In main.c in confinit() change

        conf.copymode = 0;      /* copy on write */

to

        switch(x86()){
        case 386:
                conf.copymode = 1;      /* copy on reference */
                break;
        default:
        case 486:
                conf.copymode = 0;      /* copy on write */
                break;
        }

and you're done.

No subject

[rob]

be a weenie:

fn cd{
	builtin cd $* && prompt = (`{pwd}^'% '  '	')
}

also man pbd

No subject

[rob]

there's a background project here to do some scaling font stuff.
we may use them one day, if the results are good enough, on our
higher-resolution (>130 dpi) monitors.  you need the higher res.
the problem is that antialiased fonts are often poor for day-to-day
use because their fuzziness bothers the eye.  programs like nEwS (or
your favorite pOSTsCRIPT interpreter, for different reasons)
try not to use scaled fonts when they can, but rather refer to
precompiled rasterized, hand-tuned, unfuzzy fonts.  the results
are faster, sharper, more pleasing, and easier on the eye.

-rob

No subject

[rob]

c'mon, guys, take this editor discussion somewhere else please....

-rob

No subject

[geoff]

Scott Deerwester:

>I'd like to suggest that, the next time you put out an RFD,
>you proselytize a little more aggressively.  I'm *very* interested
>in comp.os.plan9, and have been quite surprised that there isn't
>one already.  I'd vote yes (early and often, being from Chicago),
>but I've never seen the RFD.  Perhaps you could put out the RFD
>to a wider audience, certainly including this mailing list, and
>maybe including comp.os.linux.misc, comp.sys.notebook, etc.
>I'd think that many of the people who are interested in Linux
>(c'mon now, running a more or less full-blown UNIX on something
>that you can carry around in your backpack is hardly completely
>devoid of interest...) would be interested in Plan 9 as well, ne?

There were over 40 newsgroups that `could have been relevant' to
a RFD for Plan 9. I'd rather not launch a huge campaign, as OS
evangelism is a fairly irritating behavior (even when you are
running the |<00L3St 0S in the whole wide world :-).

I should have sent the RFD to this list, but I was planning
to do that only if there was a reasonable support from the rest of
the net - after all, if all the support we got was from users of the
mailing list, why not just stay with the mailing list? Still as an
experiment: if you're reading this list, and you'd go to the trouble
of voting on comp.os.plan9, mail me and let me what you'd vote (and
if you'd vote `no', then what your reasons are).

Support from the people at the Labs is another issue. I would think
a newsgroup would reduce the amount of support mail to the Labs, if
there was a decent FAQ there and people could post answers about getting
Plan 9 up and running.

There's also a the fact that of traffic on this list being hardly
overwhelming (these last couple weeks are an exception).

>I've saved an 80Mb partition to load Plan 9 onto on my notebook...
>but haven't had the time to figure out what to do to get it up
>and running... 

Perhaps this should be our first FAQ? PC configuration type stuff?

(I await answers to gary's pc file server question  with interest).

Geoff.

No subject

[Vijay]

The purpose of this list is to maintain a database of hardware that is
known to work with plan 9.  This should make life easier for those
people contemplating the purchase of new hardware.  Most of this list
is taken directly from what Forsyth sent me.  Please do not deluge him
with mail, reply to me and I'll try to set up a mechanism to distribute
this stuff.

 The list is starting out for the most selfish of reasons, since I am
in the market for some hardware to run plan 9, I thought it would be a
good idea to find out what to get.  This list will concentrate mostly
on the 386/486 clone hardware.  While the 386/486 machines are looked
upon with derision, they provide the cheapest hardware to run plan 9.
However, most of the hardware is only tested with, and comes with
drivers for, DOS and Windows, and thus may not work with plan 9.  This
list details the stuff that works with plan 9.

BeginBlurb:  
Plan 9 on a Mac.  No more futzing around. 
I could deal with plan 9 on a Powerbook 180.
End Blurb.

Contributors:

Most of this hardware list comes from forsyth@minster.york.ac.uk who
has done yeoman work to contributing to this list.  Actually, come to
think of it, his is about the only contribution ;)

Here is the list of hardware that will work with plan 9.  Other
hardware may work with plan 9, but this list will only list stuff that
is known to work or stuff that I need confirmation on.

-----------------------------------------------------
Bus:
	ISA
	VESA
	EISA (may work with ISA Cards, need confirmation)
-----------------------------------------------------
CPU:
	Intel
	AMD - To be tested by Forsyth
	Cyrix - Ditto
	IBM SLC - Ditto
------------------------------------------------------
Video Cards:

This is the trickiest part.  SVGA cards all have different ways of
configuring and operating high-resolution modes (higher than 640x480
VGA).

	ET4000 SVGA cards work.
	S3 911 cards work
	Paradise cards (reportedly) work.
	Diamond Stealth Pro cards work, with some changes
		to devvga.c and lib/vga.  This is from
		Gary Chapel.

------------------------------------------------------
Ethernet cards:

	WD
	SMC
	NE2000 clones (Reset port address it uses
		       might not work on some clones)
	3Com 3C509 Etherlink III (need to get some fixes
		       to use the driver under very heavy
		       loads)
	3Com 3C503 -- Avoid this one


------------------------------------------------------------
Hard Disk Controllers:

	Ultrastor 14F SCSI (driver is available by ftp)
	Adaptec 1542B/Buslogic 542B (driver is available,
		but hasn't been used very much)
	Forsyth has added the driver for the Ultrastor 14F.
	He has used it on both cpu and terminal machines,
	and also on the file server, running Fujitsu 2624FA
	disks, Seagate 3283N and Sun CDROM drives.

	Standard IDE controllers should work.
--------------------------------------------------------------
Memory:

	8-16 megs should be adequate.  More is better.

--------------------------------------------------------------

Modems:

	High speed modems will work but the kernel does not take
	advantage of the FIFO's in the 16550 UARTS.  Forsyth can
	supply the diffs and the appropriate changes that will work
	for `hayes', which work fine upto 38400.
	
--------------------------------------------------------------------------
Diskettes:

	Stick with 3.5" 1.44 meg or 5.25".  Plan 9 does not
	work too well with 2.88 meg 3.5".

---------------------------------------------------------------------------

Disks:

The pc system on the distribution (Jan 1993 version) supports
IDE disks.  The IDE driver has a bug in it that causes it to fail
on very old Seagate 42 Meg drives (fixed by Forsyth).  Modern IDE
drives work just fine.

----------------------------------------------------------------------------
This is the list as it stands.  If you have plan 9 running on
different hardware, please send me details.

War stories welcomed, to go into the war stories list. I presently
have Forsyth's war story. More input welcomed.  I will be looking at
putting this stuff up for anon ftp when I have sufficient volume.  If
there is sufficient interest, I'll post up the war stories that I
have.

--
Vijay Gill                         |The (paying) customer is always right.
wrath@cs.umbc.edu                  |                  - Piercarlo Grandi
vijay@gl.umbc.edu                  | Eagles may soar, but weasels don't get
These are my opinions only.        | sucked into jet engines.

No subject

[rob]

isspace() is an ASCII-ism.  ANSI C has nothing helpful to say on the topic of
collation and expanded character sets.  if you have input and are prepared to
defend your position, comp.std.character-set.flames.R.us is your oyster.

-rob

No subject

[Scott]

| hey, unicode has a gajillion space characters to choose from,
| and only one of them is disallowed in file names.

But it is the one that 8.5 connects to my spacebar. :-)

| get with the program, mon.

Ouch.  I know about the funny spaces, but I'm uncertain what the
program is.  Why doesn't isspace() recognise them, for example?  When
awk goes to split input into fields, I can see why one might not want
it to split on the non-breaking-space or the graphic-for-space, but
what about the others?

No subject

[rob]

hey, unicode has a gajillion space characters to choose from,
and only one of them is disallowed in file names.
here are some of the others, from grep space /lib/unicode.
get with the program, mon.

00a0	non-breaking space; = iso no-break space; x (space - 0020)
0400		x (non-breaking space - 00a0)
0600		x (non-breaking space - 00a0)
2002	en space
2003	em space
2004	three-per-em space
2005	four-per-em space
2006	six-per-em space
2007	figure space
2008	punctuation space
2009	thin space
200a	hair space
200b	zero width space
2422	blank; graphic for space; x (latin small letter b bar - 0180)
3000	ideographic space; x (space - 0020)
303f	ideographic half fill space

No subject

[rob]

postscript versions of the two papers we presented at the USENIX
conference last week are now available by anonymous FTP from
research.att.com, directory dist/plan9man.
an old version of acid is part of the plan 9 distribution.
acme is unobtainable at the moment.

-rob

No subject

[philw]

We will bring at least two laptops.
They both have acme and acid stuff.

No subject

[Scott]

| if you want a bof, schedule it and see what happens.
| i'm a little curious myself.

Will you guys bring your laptops and have show-and-tell?

No subject

[rob]

if you want a bof, schedule it and see what happens.
i'm a little curious myself.

-rob

No subject

[Scott]

It doesn't look like there's much interest in a bof;
maybe next time.

No subject

[rob]

all right all right all right!
i should know better than to say anything about the contents
of RFC's without checking first, but i was trying to avoid bothering
my local mail expert.  so much for good intentions.
i will consider changing the local mail programs (there are two here)
to reply to Reply-To if it exists, but i won't enjoy doing it.

-rob

No subject

[Scott]

| The example that the RFC cites is a special mailbox where errors are
| to be sent could be used as the envelope address (e.g. the mailing
| list administrator's address).

That's exactly what we do.

| I think that your sysadmins are right: the UA should reply to the From: 
| address stored within the message.  I think that RFC821 specifies other 
| headers that can be used, too, such as Reply-To:.

RFC 822 says that.  See sections 4.3.1 and 4.4.4.

No subject

[Neil]

In message <93Oct21.200627edt.2516@groucho.cse.psu.edu>, Scott Schwartz writes:
> 
> | not quite... upas uses the SMTP information, not the message contents,
> | as i think it should.
> 
> Well, ok, but the local postmasters tell me that the internet
> conventions are otherwise.

I don't have RFC821 in front of me, but (unless I'm confused) it
indicates that the SMTP envelope address may be different than the
sender's address (and the sender's address is where one would
presumably want a reply to go).  The example that the RFC cites is a
special mailbox where errors are to be sent could be used as the
envelope address (e.g. the mailing list administrator's address).

I think that your sysadmins are right: the UA should reply to the From: 
address stored within the message.  I think that RFC821 specifies other 
headers that can be used, too, such as Reply-To:.

Regards,
Neil

No subject

[Scott]

| not quite... upas uses the SMTP information, not the message contents,
| as i think it should.

Well, ok, but the local postmasters tell me that the internet
conventions are otherwise.

No subject

[rob]

not quite... upas uses the SMTP information, not the message contents,
as i think it should.

-rob

No subject

[philw]

>We are running plan9 on both pc and sparc machines.  We also have
>a cpu server (sparc) up and running.  When the cpu server is running,
>a "who" on the file server, will show that both "adm" and "bootes" are
>running.  Fine.  Frequently, this same inquirery will show 
>"none" and "adm"!  When this happens, the distributed system is *very*
>sluggish.  We have to reboot the cpu server to get rid of "none".
Next time it happens mail me a stack strace
of the process by using db.

No subject

[Scott]

------- Forwarded Message

Return-Path: psuvax1.cse.psu.edu!medizin.uni-ulm.de!Thomas.Nau
Received: from psuvax1.cse.psu.edu ([130.203.2.4]) by groucho.cse.psu.edu with SMTP id <2516>; Wed, 20 Oct 1993 02:59:40 -0400
Received: from struppi.medizin.uni-ulm.de ([134.60.79.2]) by psuvax1.cse.psu.edu with SMTP id <291785>; Wed, 20 Oct 1993 02:59:03 -0400
Received: by struppi.medizin.uni-ulm.de id AA04735
  (5.65+/IDA-1.4.4); Wed, 20 Oct 93 07:58:42 +0100
From:	Thomas Nau <Thomas.Nau@medizin.uni-ulm.de>
Message-Id: <9310200658.AA04735@struppi.medizin.uni-ulm.de>
Subject: access to cdrom from PC
To:	plan9-fans-request@cse.psu.edu
Date:	Wed, 20 Oct 1993 02:58:42 -0400
Mailer: Elm [revision: 70.85]

Hi,
I would like to install Plan 9 from a cdrom mounted on our UNIX host.
The problem is that every filename.
Is there a way to copy files from the CD to the 386 (using mkfs ...)
without having a Plan 9 fileserver ?

regards
Thomas
**********************************************************************
                            Thomas Nau 
       (Thomas.Nau@medizin.uni-ulm.de, nau@medizin.uni-ulm.de)
       Departement of Anaesthesia, University of Ulm, Germany

                    think about 42 and DONT PANIC
**********************************************************************


------- End of Forwarded Message

No subject

[presotto]

% diff /sys/src/cmd/upas/send/rewrite.c /n/juke/plan.9/sys/src/cmd/upas/send
94,96d93
< 		re = rule_parse(s_restart(line));
< 		if(re == 0)
< 			continue;
102a100
> 		re = rule_parse(s_restart(line));

No subject

[philw]

>Then from /sys/src/alef I ran "objtype=architecture mk install";
>all the libraries (/$objtype/lib/alef/*.a) seem to compile fine,
>(assuming none generated for the 68020 & only libA.a, libbio.a for
>the 386), but $objtype=mips generates the error:
>val -w div64.l
>div64.l":12: syntax error near symbol 'lint'
>mk: val -w div64.l  : exit status=rc 8989:val 8991 errors 
>mk: for (i in ...  : exit status:=rc 8191:mk 8999:error
>mk: @{cd v; mk ...  : exit status=rc 8049:val 8991:errors
>mk: for (i in ...  : exit status=rc 8191:mk 8988:error
>mk: @{cd v; mk ...  : exit status=rc 8042:mk 8190:error
>mk: mk $objtype.install  : exit status=rc 8039:mk 8041:error
div64.l is for a 64 bit mips compiler I have been working on.
Just delete the file from the library. When or If I do a 64
bit distribution I will include everything necessary.

>once the installation is complete, will the example code which
>came on the 2nd rev cdrom, (/sys/src/alef/test, /sys/src/alef/test/Y),
No. The /sys/src/alef/test/Y I included with the new compilers
I shipped you compile and run. Some of the tests are supposed
to generate compiler errors.

philw.

No subject

[forsyth]

>> Has anyone ported the plan9 file server to a PC?
>> thanks
>> sean quinlan

i did that a few months ago.  it only supports ISA at the moment
(i haven't got an EISA yet to try), and thus only 16 Mbytes
since i don't compensate for the limitations of the pc's DMA addressing.
(it wouldn't be too hard: the SCSI adapter could allocate some
space below 16meg and copy the data; whether that's sensible
is another matter.)

i run the file server on a 16Mb 486sx33 with an Ultrastor 14F
and two Fujitsu SCSI-2 drives.  (it boots from a floppy.)
the ethernets supported are those supported by the cpu/terminal kernel
and also NE2100 clones, which use the AMD Lance
(another device with addressing problems!).
the 486sx33 serves a 486dx2-66 CPU server and a few terminals.
i can't compare its performance to an SS2; i improvised
a bootstrap for Plan 9 that went straight to the pc via 9sscd.

the only host adapter supported is the Ultrastor 14F.  the driver should
work with the 34F with at most minor changes, if any, but my 34F
and not-quite-VESA motherboard don't agree, so i haven't been able to test it.
a bus-mastering VESA device can avoid the 16Mb DMA restrictions, however,
making an ISA/VESA pc more useful as a file server.
i have an Adaptec 1542B/Bustek 542B driver, but it doesn't drive discs
(it runs the 1542B in target mode, not initiator mode);
still, most of what is required is there.

No subject

[rob]

HP took far to long to release the documentation for the snake,
so that port, although planned, was never done.

-rob

No subject

[rob]

please don't write.  we're working on getting it released.

-rob

No subject

[plan.9]

>> From cse.psu.edu!9fans-owner Mon Jul 26 08:11:55 EDT 1993
>> Received: by inet.att.com; Mon Jul 26 08:11 EDT 1993
>> Received: from relay.pipex.net ([158.43.128.1]) by psuvax1.cse.psu.edu with SMTP id <292433>; Mon, 26 Jul 1993 08:04:41 -0400
>> Received: from pipex.net by relay.pipex.net with SMTP (PP) 
>>           id <28301-0@relay.pipex.net>; Mon, 26 Jul 1993 13:03:56 +0100
>> To:	9fans@cse.psu.edu
>> From:	Tim.Goodwin@pipex.net
>> Subject: Obtaining Plan 9 as an individual
>> Date:	Mon, 26 Jul 1993 07:55:00 -0400
>> Message-ID: <"relay.pipe.335:26.06.93.12.04.18"@pipex.net>
>> 
>> Does anyone know what my chances of obtaining Plan 9 are as an
>> individual?  Would it be worth writing to Bell Labs?
>> 
>> Failing that, maybe I'll have to find a university who'd take me on a
>> as a (very) part time assistant OS hacker...
>> 
>> Tim.

it might become possible for individuals to acquire the system.
steps are being taken in that direction.  the decision depends
on various factors that are not influencable by outside campaigns
and there is no promise of a successful conclusion, however.

No subject

[Bob]

DOS doesn't like a file or directory name of "aux" or the files
/dev/c: and /dev/a:

The way to do it is to boot from the floppy, use the floppy as
your file system and then copy the files and directories
from the floppy to /n/c: using plan9.

(ie mkdir /n/c:/386/bin/aux, etc)

Bob.

No subject

[Jerry]

Hi 9'ers,
 my question is of the pc installation. We've
always used the original 1.44 floppy. In
"Configuring for a PC", it is discussed a way
to copy the boot files to a disk to use as the
boot disk. DOS doesn't like the AUX directory
in /386/bin. Even if the files are zipped, then
expanded into DOS, it refuses to create or 
maintain this directory. How have others got 
around this? Similarly, one may copy the image 
(pcdisk) to a floppy, but of what utility is this?
  Thanks in advance, --Jerry

No subject

[jwjohn]

Hi 9'ers,
does anyone have any examples of ALEF code they might
share? (Or know of an FTP site?) It needn't be all that 
complex, just something to demonstrate the concurrency 
primitives. I intend to write a simple demonstration 
program; like the "classic" dining philosophers.

Also, is SPARC and MIPS the only supported platforms?
Specifically, has anyone ported the compiler to the PC?

Any input is appreciated.
             Thanks in advance. --Jerry

No subject

[rob]

if you want emacs, you can run it almost anywhere
else; why bother to run it here?

if you trust us enough to try our operating system,
why not trust us enough to try our other software,
like the window system and the editor?  why not see
if we have a good idea in our structuring of the
libraries and header files?  why marvel at the
compatibilities of the system when some of its
best ideas might be incompatibilities?  why install
a replica of your familiar playground with all these
new toys around?

-rob pike