Subject: Linux-Misc Digest #202
From: Digestifier <Linux-Misc-Request@senator-bedfellow.MIT.EDU>
To: Linux-Misc@senator-bedfellow.MIT.EDU
Reply-To: Linux-Misc@senator-bedfellow.MIT.EDU
Date:     Wed, 13 Oct 93 11:13:25 EDT

Linux-Misc Digest #202, Volume #1                Wed, 13 Oct 93 11:13:25 EDT

Contents:
  Questions Flamewar (Ian Jackson)
  Re: Unix manual, SLIP (Alan Cox)
  Re: [ANSWER] Re: Why only compressed kernels? (Alan Cox)
  Re: [ANSWER] Re: Why only compressed kernels? (Kai Petzke)
  Linux count: HELP with countries of domains (Harald T. Alvestrand)
  Re: PC internal modem recommendations?  (Lawrence Kirby)
  Re: Linux, Xfree with Diamond Viper (Long) (Chris Mauritz)
  Re: Kernel Panic: swaper problems (HELP) (Daniel Supernaw-Issen)
  Re: Personal info about Linus (Eckehard Stolz)
  Re: Questions Flamewar
  kernel hacker's guide - where? (Bill Mitchell)
  IMPORTANT: Island Office Package for Linux (Thomas Uhl)

----------------------------------------------------------------------------

Crossposted-To: comp.os.linux.development,news.groups
From: iwj10@cus.cam.ac.uk (Ian Jackson)
Subject: Questions Flamewar
Date: Wed, 13 Oct 1993 00:22:50 GMT

(Note: This flamewar definitely does not belong in col.development.
Trying to move it out of the col hierarchy probably won't work, so
I've set followups to col.misc.  However I'd much rather there weren't
any followups - pretend I said `Followup-To: poster'.)

I shan't try to respond item-by-item, since in the few days since I
last read col.* this issue has exploded.  I shall also not state my
agreement with those people like Lars Wirzenius, Matt Welsh and
Brandon Allbery (this list is inclusive, not exclusive!) whom I agree
with, except where it is particularly important.

Firstly, as many people have pointed out, col.development is for
*discussion* of the *development* *of* Linux.  It is not for questions
about the development of Linux (except those asked as part of the
development effort, or which are equivalent to informed suggestions).
It is not for discussion of development for Linux.  It is not for
discussion of the development of programs in general.

I posted my two postings because I had become fed up at the
nonexistant signal-to-noise ratio in col.development, which I had
hoped would be spared the idiots-influx.  Posting in the heat of my
anger was clearly a mistake, however I believe that everything I said
was justified.  My hope in posting rather than mailing would be that
it might deter people considering doing the same as Stefan Lukka.

My wording `no questions' would, if the flamewar had been avoided,
have conveyed exactly the right information to each and every reader.
The only person who should post a question to col.development is
someone who has been reading col.d for some time, and knows that their
posting isn't what I was talking about when I said `no questions.'
Nobody else should be posting questions to col.d.

However, I am not trying to make col.development an exclusive club.
One of the things that has gotten Linux where it is today is the open
development effort, which has enabled a large number of people to
contribute in a very short space of time.  I certainly wouldn't say
that you have to be a guru to post to col.d - but the rule that you
should read a group for a month or two before posting applies doubly
when you are addressing a technical audience on a subject you may not
be too familiar with.  Reading the FAQs and HOWTOs should be taken for
granted.

Tim Peoples's `RFD', which he now acknowledges he knew was bogus,
worked very well at inciting this flamewar.  I hope he's pleased with
himself.  I did agree with Matt's description of him as an idiot, as
this was more charitable than ascribing his posting to malice.
However since the malice has now been admitted I say that he is not an
idiot, but rather a net-sociopath, perhaps somewhat out of his depth.

I stand by my description of Stefan Lukka (the person who posted the
FAQ about -m486) as an `idiot with no netiquette' (this was in the
Keywords line of my followup).

Note that discussion of my opinions of Tim Peoples and Stefan Lukka
should be taken to email, as I don't read any groups which would be
appropriate for it.

When this furore has died down somewhat I shall post to col.misc so
that we may have a hopefully more fruitful and less personal
discussion about what to do with col.development's (and indeed
col.*'s) serious noise problems.

-- 
Ian Jackson, at home  <ijackson@nyx.cs.du.edu> or <iwj10@cus.cam.ac.uk>
PGP2 public key available on server.  Urgent email: <iwj10@phx.cam.ac.uk>
2 Lexington Close, Cambridge, CB4 3LS, England;  phone: +44 223 64238

------------------------------

From: iiitac@swan.pyr (Alan Cox)
Subject: Re: Unix manual, SLIP
Date: Wed, 13 Oct 1993 10:46:03 GMT

In article <forb0004.12.00148057@gold.tc.umn.edu> forb0004@gold.tc.umn.edu (Eric J. Forbis) writes:
>For those who don't have the net2 files, the net2-howto suggests picking up 5 
>or 6 files from 3 sites. I'd like to avoid this, if possible- I'm not Linux- 
>savvy enough yet to deal with variations on a theme I don't know. 
You need to pick up about 3 things to get net2 working properly if you don't have
the bits. If you have SLS with the kernel from 0.99.12 or higher you have the
bits but the SLS setup is a bit broken.
>How about the net2-debugged file? Does this require files from multiple 
>sources?
Net2Debugged is a single archive that replaces the kernel tcp/ip code with
a patched up set. You still need the same network tools as before. If you
are not sure how to procede with getting all the bits drop me an email.

Alan




------------------------------

From: iiitac@swan.pyr (Alan Cox)
Subject: Re: [ANSWER] Re: Why only compressed kernels?
Date: Wed, 13 Oct 1993 10:49:24 GMT

In article <29eu4l$a94@news.u.washington.edu> tzs@stein3.u.washington.edu (Tim Smith) writes:
>Why not load the kernel entirely above 1 meg?  On a *286* this is somewhat
>hard to do, unless you want to use the undocumented LOADALL instruction,
Its farirly easy on a 286 if you load the blocks under 1Mb and then copy them
up with the BIOS routines.
>but on a 386 or 486, it is no problem.  (In fact, doesn't SCO do this?  I'm
>pretty sure ISC did this, too).  The problem with the 286 was that it was
>hard to switch back to real mode from protected mode (which you need to
>do, assuming that you are using INT 13h to read the kernel from disk),
>but on the 386 it is easy to switch back.
Indeed. In fact all you have to do to get Linux to have this wonderful facility
is download a copy and rewrite the boot loader code (which is in 8086). The
8086 assembler/linker is included with Linux (as86,ld86) and then submit it
to Linus. This is your chance..

Alan


------------------------------

From: wpp@marie.physik.tu-berlin.de (Kai Petzke)
Subject: Re: [ANSWER] Re: Why only compressed kernels?
Date: 13 Oct 1993 12:15:11 GMT

In <1993Oct12.162106.11329@gmd.de> veit@borneo.gmd.de (Holger Veit) writes:

>I believe the question "Why?" aims at another thing, namely:
>If the 640K limit is a restriction, then *why* don't you load above
>1MB, where you find sufficient contiguous memory? The stuff to
>move code to there is already in the BIOS (INT 15h), even if it is
>slow (actually, since the [34]86 can get as easily out of PM as into,
>in contrast to the 286, only BIOS access itself is slow).
>The bootstrap loader can, and does, reside in low memory (I doubt it will
>ever reach the 640K size limit :-)), and when all necessary stuff has
>been done, it switches to PM and calls the kernel at (physical) 1M.
>Stuff is necessary in mm, of course to reclaim the now unused 640K
>area and the inter-adaptor area.


This is exactly, what the uncompression routine does:

It loads things into normal BIOS acc essable memory.

It uncompresses it into the memory starting at 1 Meg.

It calls the kernel at 1 Meg.

The uncompression is fast.  It will even get faster, soon (wait for a new
release of gzip).

The uncompression is usable with everything, whether you boot from floppy
or disk.

There is still an uncompressed kernel image in the file
/usr/src/linux/tools/zSystem.
--
Kai
wpp@marie.physik.tu-berlin.de
Advertisement by Microsoft in a well-known German magazine:
        If you don't like our programmes, than make your own ones.
However, they expect you to use Microsoft products for this -:)

------------------------------

From: hta@uninett.no (Harald T. Alvestrand)
Subject: Linux count: HELP with countries of domains
Date: 13 Oct 1993 12:48:05 GMT

Hello, world!

The greatest number of comments to the Linux count has been on my
allocation of domains to countries.
(The count is now 2928 users)

So, I did some homework, and now have the following configuration
file for mapping domains to countries: 

# Mapping file for domain -> country
# Each line is a country + domains that map to this country
# More than one line per country can exist
de: sub.org germany.eu.net xlink.net
no: hydro.com
gb: uk
fr: fr.net
nl: nl.net
se: swip.net
ca: north.net
# The US is a special problem
us: edu com gov mil org
# Subdomains of NET that are known to be in the US
us: uu.net tyrell.net nwnet.net
# The world - use private-use country code
xw: net int bitnet uucp unknown
# Europe
xe: eu.net

Any comments/corrections to these mappings are appreciated!

Now, these are the domains currently allocated to "world",
"Europe", or simply "unknown" (the number is number of users):

  8 xe relay.eu.net
  2 xw access.digex.net
  1 xw ans.net
  1 xw avix.uucp
  1 xw banjo.concert.net
  1 xw blackbox.uucp
  1 xw brufpb.bitnet
  2 xw cfruni51.bitnet
  1 xw cfruni52.bitnet
  1 xw chaserv.uucp
  1 xw clark.net
  1 xw cube.net
  1 xw cyber.net
  4 xw dbnrhrz1.bitnet
  1 xw elvis.concert.net
  2 xw esoc.bitnet
  1 xw eugene.uucp
  1 xw front.uucp
  1 xw greg.uucp
  1 xw holonet.net
  1 xw hotblack.uucp
  1 xw hustle.rahul.net
  1 xw monroe.bitnet
  1 xw moundst.uucp
  1 xw phaffle.uucp
  1 xw pilot.njin.net
  1 xw pioneer.ci.net
  1 xw pipex.net
  1 xw rniil.uucp
  2 xw rnima.uucp
  1 xw rniwh.uucp
  2 xw rock.concert.net
  1 xw small.uucp
  1 xw spatzi.uucp
  1 xw speedway.net
  1 xw timbuktu.uucp
  1 xw tower.nimbus.uucp
  2 xw waggen.uucp
  8 xx baghira
  1 xx brand
  1 xx dada.sd113-w2
  1 xx ender
  1 xx glenn
  1 xx laura
  1 xx snow-white

I also have 96 lines of "com" domains - if anyone really feels like it, he
can tell me which of the following does NOT belong in the US:

.3com.com
.adobe.com
.amdahl.com
.ankle.com
.atlas.com
.att.com
.bbc.com
.bellcore.com
.bms.com
.boeing.com
.bull.com
.ccur.com
.cdc.com
.celestial.com
.citicorp.com
.comsat.com
.connected.com
.convex.com
.cray.com
.csc.com
.cts.com
.cv.com
.cypress.com
.dec.com
.delcoelect.com
.dg.com
.dgtl.com
.dhl.com
.dpw.com
.era.com
.es.com
.esys.com
.fluent.com
.fmc.com
.fns.com
.ford.com
.ge.com
.gpsemi.com
.gs.com
.gtegsc.com
.gwl.com
.harris.com
.honeywell.com
.hp.com
.ibm.com
.idsoftware.com
.ingr.com
.inri.com
.intel.com
.kodak.com
.lockheed.com
.loral.com
.meaddata.com
.mentorg.com
.mmc.com
.mot.com
.msen.com
.ncr.com
.neosoft.com
.nokia.com
.northrop.com
.novell.com
.nsc.com
.ods.com
.oracle.com
.paramax.com
.parcplace.com
.portal.com
.psi.com
.rain.com
.ricoh.com
.rockwell.com
.saic.com
.sas.com
.sbc.com
.sccsi.com
.shl.com
.siemens.com
.slb.com
.sosi.com
.sra.com
.std.com
.stortek.com
.stratus.com
.stx.com
.sun.com
.tek.com
.ti.com
.trw.com
.tti.com
.uswest.com
.valmet.com
.vitek.com
.wa.com
.xerox.com
.zds.com

Yours for a better Linux count,
-- 
                   Harald Tveit Alvestrand
                Harald.T.Alvestrand@uninett.no
      G=Harald;I=T;S=Alvestrand;O=uninett;P=uninett;C=no
                      +47 7 59 70 94
My son's name is Torbjxrn. The letter between "j" and "r" is o with a slash.

------------------------------

Crossposted-To: comp.dcom.modems
From: fred@genesis.demon.co.uk (Lawrence Kirby)
Subject: Re: PC internal modem recommendations? 
Date: Wed, 13 Oct 1993 11:53:15 +0000

In article <29ghnm$he0@marr.ecs.soton.ac.uk> mrd@ecs.soton.ac.uk writes:

>
>I'm thinking of getting a modem to work with linux on my PC.
>
>My existing serial I/O is 8250 based and on the motherboard, so buying
>a fast external modem would mean upgrading my serial ports too.
>
>With internal modems I gather some of them try and emulate a 16550A
>UART.  Do others actually have a real 16550A on the card?

I don't think it is that common yet to emulate a 16550A. Some emulate a 16450
but use extra 'invisible' buffering on board. Others use a true 16550A.

>The linux serial FAQ mentions there can be problems with the emulation
>provided by the Rockwell chipset. Are the problems serious or can one
>live with them? What does linux try and do that DOS doesn't?

I believe these are down to timing as much as anything.

>I'm considering a USR sportster 14400 (internal). Does anyone know if it
>works well with linux? Any other internal recommendations? I would like
>V32bis (with V42 and V42bis on top) but FAX is optional.

The Sportster should be fine - it uses a real 16550A.

>Finally, I'm in the UK. Is it OK to use US modems here?

They work but they won't be approved so it is illegal to connect them to the
PSTN. Why not get a UK model Sportster? They're almost reasonably priced now
and you'll get a USR warranty and tech. support in the UK.

=========================================
Lawrence Kirby | fred@genesis.demon.co.uk
Wilts, England | 70734.126@compuserve.com
=========================================

------------------------------

Crossposted-To: biz.zeos.general
From: mauritz_c@spcuna.spc.edu (Chris Mauritz)
Subject: Re: Linux, Xfree with Diamond Viper (Long)
Date: Wed, 13 Oct 1993 13:51:57 GMT

Erik Nygren (nygren@athena.mit.edu) wrote:
: --

: I would like to post an unfortunate warning to those considering purchasing
: computers from Zeos or any other company which includes Diamond video cards
: with the intent of Linux or any of the other free versions of UN*X
: OS's.

In short, don't do it (buy a Diamond card for use with Xfree).

: If you want to run X windows with Diamond cards other than the Viper,
: it is possible using 3rd party hacks.  However, this may be very difficult
: to get working, and the group developing the X-server for all of the free
: x86-based UN*X's does not support Diamond cards because of Diamond's
: unwillingness to help developers.

They're happy to help developers IF you pay them for the programming
information and then sign a non-disclosure agreement.  So then you
have everything you need to write the driver, BUT you can only release
a binary.  Releasing source violates the non-disclosure agreement.
It seems Diamond can not figure out if thy're in the hardware or the
software business.

: I bought my Zeos 486/66 a few months ago with the intention of running Linux,
: and I was not pleasantly suprised when I found the problems I had with
: the Viper (which is not a cheap card).  

Well, you SHOULD have read the FAQ.  The situation with Diamond cards has
been very widely documented for well over a year.  You could have saved 
yourself a lot of problems by getting the box with a different brand of
video card.  I ended up switching vendors (bought a Gateway) since Zeos
wouldn't let me choose a non-Diamond card (and their machine was alos
more expensive).

The bottom line is that if you run DOS+Win, Diamond boards are nice.
If you run OS/2 or ANY flavor of unix+X you are out of luck.

Caveat emptor.

: is growing VERY rapidly.  It would be ashame for Zeos to lose out on this
: rapidly growing market because of policy of a vendor that supplies Zeos with
: components.

They already have lost out.

: Another warning:  The Adaptek SCSI Protocol Chip (AIC-6250) that is
: available in Zeos computers for $45 is not yet supported by Linux.
: It may be in a few months, but it is not yet.  As a result, there is no way
: to access SCSI hard drives and CDROM drives connected through it.

If you want to use any type of linux on the machine, it doesn't make
any sense to scrimp on a SCSI adapter.  For an additional $100-150,
you can get the Adaptec 1542B/C which is much faster, supported by
virtually every OS I've run across, and portable (you can pull it out
and put it in another machine later if you want...not so with the
"SCSI on a chip").

Cheers,

Chris
-- 
==========================+==================================
Christopher Mauritz       | This space reserved for some
mauritz_c@spcvxa.spc.edu  | future witty saying.
mauritz_c@spcvxe.spc.edu  |

------------------------------

From: daniel@austin.ibm.com (Daniel Supernaw-Issen)
Subject: Re: Kernel Panic: swaper problems (HELP)
Date: Wed, 13 Oct 1993 13:34:06 GMT


If this kernel panic occured after upgrading the kernel from <=pl10 to >pl10
the solution is simple.  In this case, the panic is caused by the keymaps
being incompatable between the kernels.  If you are using a US keyboard, the 
solution is very simple - delete /etc/kbmap and reboot.  This will force the 
kernel to use its default - US keymapping. If you are not using a US keyboard - 
the solution is still relatively simple - get the appropiate keymaps in the 
same directory you got the latest kernel - install,reboot and enjoy.  If your
problem has nothing to do with all this - I apologise for wasting your time.

Daniel Supernaw-Issen
please send all mail/flames/etc to danielsi@cs.utexas.edu

I speak for nobody but myself.

------------------------------

From: stolz@Informatik.TU-Muenchen.DE (Eckehard Stolz)
Subject: Re: Personal info about Linus
Date: Wed, 13 Oct 1993 14:32:53 GMT


In article <29cjldINN4v3@xs4all.hacktic.nl>, fro@hacktic.nl (fro) writes:
|> Hi Linux Lovers!  
|> 
|> I'm just a newbie, attracted to this nice, free Unix version, and I'd like
|> to explore it to increase my knowledge of Unix (currently NULL) - inside
|> and outside. 
|> 
|> I wonder if anybody has some personal data of Linus: What is his
|> profession, what did he study, is he a professor... I'd really like to
|> know a bit more about the man that created this OS. 


Well, as far as I know, Linus studies computer science in Finland ! He started
Linux while trying to experiment with the protected mode of 386-CPU's

"First" version of Linux consisted of two processes printing "AAA" and "BBB" !
And that was all !!!

But if you wnat to know more, please look in the linux/docs/linus directory on
the TSX-11 ! There are some really :) nice GIF-pictures of Linus !

(sorry, Linus :-) !

There is also a file called "linux-history" in linux/docs, where (almost)
everything is told about Linux and its beginning !

hope it helped

cu

Eckehard
stolz@fiffi.sta.sub.org
stolz@informatik.tu-muenchen.de

------------------------------

From: jwest@jwest.ecen.okstate.edu ()
Subject: Re: Questions Flamewar
Date: Wed, 13 Oct 1993 14:48:20 GMT

In article <1993Oct13.002250.22588.chiark.ijackson@nyx.cs.du.edu> iwj10@cus.cam.ac.uk (Ian Jackson) writes:
>
>I posted my two postings because I had become fed up at the
>nonexistant signal-to-noise ratio in col.development, which I had
>hoped would be spared the idiots-influx.  Posting in the heat of my
                           ^^^^^^^^^^^^^

Is everyone who makes an honest (but possibly stupid) mistake an idiot? 
Gosh, there are a lot of idiots in the world. 


>However since the malice has now been admitted I say that he is not an
>idiot, but rather a net-sociopath, perhaps somewhat out of his depth.
                     ^^^^^^^^^^^^^

Hmm...interesting observation. Seems to apply to several people on both
sides of the issue.

>I stand by my description of Stefan Lukka (the person who posted the
>FAQ about -m486) as an `idiot with no netiquette' (this was in the
>Keywords line of my followup).


Attempts at public humiliation (name-calling qualifies in this case)
don't follow good netiquette either.


People, if someone repeatly ignores *polite* reminders to use the proper 
group (posted or emailed), then a public flame is called for. If you get
to this point, put *in the flame* "this is the nth time you've done this".
If you are too angry to give a polite message to a first offender, simply
ignore it, someone else surely will. If you are so consumed with anger 
that you feel you must publically flame a first offender, then you should
seriously consider getting some therapy  (1/2 :-).

Jim

==========
Jim West
Associate Professor
Electrical and Computer Engineering
Oklahoma State University
jwest@master.ceat.okstate.edu

------------------------------

From: mitchell@mdd.comm.mot.com (Bill Mitchell)
Subject: kernel hacker's guide - where?
Reply-To: mitchell@mdd.comm.mot.com (Bill Mitchell)
Date: Wed, 13 Oct 1993 14:30:31 GMT

I couldn't find it on a browse through sunsite and tsx-11.  A pointer would
be appreciated.

-- 
mitchell@mdd.comm.mot.com (Bill Mitchell)


------------------------------

From: tul@tonga.heidelbg.ibm.com (Thomas Uhl)
Subject: IMPORTANT: Island Office Package for Linux
Date: Wed, 13 Oct 1993 11:54:38 GMT
Reply-To: tul%tonga.heidelbg.ibm.com@ibmpa.awdpa.ibm.com

As I wrote in my posting about two weeks ago, I want to convince the
Island company to port their office software (word processor, drawing
and painting tool) to Linux. 

In the last moths many people have seem to be intereted in a commercial
powerfull word processor for Linux. I asked all members of the Linux 
Internet community to send me an e-mail if there is still an interest.

I am _VERY_ disappointed: There were only about 30 answers. I don't think
it would make any sense to continue if there is not much more interest.


   IF YOU ARE INTERESTED IN A WORDPROCESSOR FOR LINUX, PLEASE SEND ME A
                      E-MAIL TO THE FOLLOWING ADDRESS:

                        uhl@sun1.rz.fh-heilbronn.de

Please ask also your friends, especially if they are not connected to the
internet.

Thomas

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.misc) via:

    Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Misc Digest
******************************
