813 lines
30 KiB
Plaintext
813 lines
30 KiB
Plaintext
From: Digestifier <Linux-Development-Request@senator-bedfellow.mit.edu>
|
||
To: Linux-Development@senator-bedfellow.mit.edu
|
||
Reply-To: Linux-Development@senator-bedfellow.mit.edu
|
||
Date: Fri, 30 Sep 94 17:13:16 EDT
|
||
Subject: Linux-Development Digest #245
|
||
|
||
Linux-Development Digest #245, Volume #2 Fri, 30 Sep 94 17:13:16 EDT
|
||
|
||
Contents:
|
||
Token Ring driver for Linux (TJB4@niobbs1.em.cdc.gov)
|
||
Re: Process checkpointing (G Dinesh Dutt)
|
||
Re: i486 Word length, anyone? (Sven Goldt)
|
||
Floppy problems since 1.1.45 (wunderli@inf.ethz.ch)
|
||
Re: Could TCP/IP be implemented over SCSI? (David - Morris)
|
||
Re: How to use a host as a router - READ THIS (Alan Cox)
|
||
Re: SMail security hole? (Thomas Parmelan)
|
||
Re: fiber optic ethernet cards (Donald Becker)
|
||
Re: Could TCP/IP be implemented over SCSI? (David - Morris)
|
||
AIC-7770 mailing list created (Carl G. Riches)
|
||
Linux Console Device (pp000932@interramp.com)
|
||
Re: 1+ Gig SCSI Drives (Daniel Rogers)
|
||
Re: C-News cleanup release finds bug in GNU 'join' (Henry Spencer)
|
||
Re: Could TCP/IP be implemented over SCSI? (Dan Newcombe)
|
||
Re: Does Quicken work under DOSEMU? (Jon Freivald)
|
||
Re: 900 MHz CB band??? (Bruce Janson)
|
||
Is there a memory allocation debugging tool for Linux? (Henrik Juul Hansen)
|
||
Re: Could TCP/IP be implemented over SCSI? (Drew Eckhardt)
|
||
Re: Could TCP/IP be implemented over SCSI? (Drew Eckhardt)
|
||
How to get more kernel memory for kmalloc? (Thomas E Zerucha)
|
||
Re: Special Sale On QNX! (Cyber Link)
|
||
[DOSEMU] Running dosemu disables serial port!! (Joel M. Hoffman)
|
||
What GUI to write for? (Marten Liebster)
|
||
|
||
----------------------------------------------------------------------------
|
||
|
||
From: TJB4@niobbs1.em.cdc.gov
|
||
Subject: Token Ring driver for Linux
|
||
Date: Thu, 29 Sep 1994 20:20:19 GMT
|
||
|
||
does anyone know of a Token Ring driver for Linux??
|
||
|
||
|
||
tb
|
||
|
||
------------------------------
|
||
|
||
From: G Dinesh Dutt <brat@htilbom.ernet.in>
|
||
Subject: Re: Process checkpointing
|
||
Date: 30 Sep 1994 09:30:23 -0400
|
||
Reply-To: brat@htilbom.ernet.in
|
||
|
||
Hi,
|
||
|
||
Here is an entry from the LSM at sunsite :
|
||
|
||
Title: Checkpoint Restart support for Linux
|
||
Version: 0.01
|
||
Description: Allows the state of a running process to be captured and adds
|
||
kernel support to treat this saved state as a new executable
|
||
format and restart the process from it. Only simple processes
|
||
work in this version.
|
||
Author: Steve Lord
|
||
AuthorEmail: lord@cray.com
|
||
Maintainer: Steve Lord
|
||
MaintEmail: lord@cray.com
|
||
Required: Linux 0.99pl14
|
||
CopyPolicy: Gnu Copyleft
|
||
Keywords: checkpoint, kernel mod
|
||
Comments: Released in source form only, patch file for 0.99pl14 supplied.
|
||
Apply patch, rebuild kernel and build chkpnt binary.
|
||
DateEntered: 13DEC93
|
||
WhoEntered: Steve Lord
|
||
EmailEntered: lord@cray.com
|
||
Location1: sunsite.unc.edu: /pub/Linux/kernel/misc-patches/chkpnt_01.tpz (20179)
|
||
<A HREF="ftp://sunsite.unc.edu/pub/Linux/kernel/misc-patches/chkpnt_01.tpz">Location1</A>
|
||
===================================
|
||
|
||
I don't know if its really worth the effort to have the kind of support for
|
||
checkpointing that you are talking about i.e. to have the raytrace startup
|
||
again, TCP connections re-establish themselves etc. At least, kernel support
|
||
for them seems to be getting beyond the job of a kernel. Some of them are best
|
||
left to the application. But, some milder form of checkpointing may be good.
|
||
|
||
Regards,
|
||
|
||
Dinesh
|
||
--
|
||
###############################################################################
|
||
Love your enemies: they'll go crazy trying to figure out what you're up to.
|
||
|
||
G. Dinesh Dutt, email : brat@htilbom.ernet.in
|
||
Hinditron Tektronix Instruments Ltd., voice : 8349393/8212262
|
||
SDF-2, Unit 63-A, SEEPZ, Andheri (east), Bombay - 400096.
|
||
###############################################################################
|
||
|
||
|
||
------------------------------
|
||
|
||
From: goldt@math.tu-berlin.de (Sven Goldt)
|
||
Subject: Re: i486 Word length, anyone?
|
||
Date: 30 Sep 1994 13:42:10 GMT
|
||
|
||
James P. Callison (callison@mailhost.ecn.uoknor.edu) wrote:
|
||
: assuming that NBPW is Number of Bytes Per Word. Of course, I could be
|
||
: totally wrong--I can't find any reference to NBPW in any of the references
|
||
: I have available...
|
||
|
||
: I tried setting it to 32 (since the 486 is a 32-bit processor <shrug>),
|
||
|
||
Are you sure ? 32 bytes in a word ? ;) I would take 2 bytes for a
|
||
word, but maybe they mean 4 bytes for 32bit.
|
||
|
||
--
|
||
*****************************************************************************
|
||
* # THE MOST IMPORTANT FINANCIAL QUESTION IS: Where is the money ? # *
|
||
*****************************************************************************
|
||
|
||
------------------------------
|
||
|
||
From: wunderli@inf.ethz.ch
|
||
Subject: Floppy problems since 1.1.45
|
||
Date: 30 Sep 1994 14:21:47 GMT
|
||
|
||
Hi,
|
||
|
||
I have been reading a while this newsgroup watching out for people
|
||
having floppy formating problems too. Doesn't seem so but maybe you
|
||
nevertheless can help:
|
||
|
||
Since kernel 1.1.45 (I now use 1.1.51 if I don't need to format
|
||
floppies, otherwise I use 1.1.38 :-) ) I cannot format floppy
|
||
disks. 3.5" ones, internal drive. No problem under DOS, but under
|
||
Linux using fdformat or fdformat2 (taken from sunsite) formating
|
||
always goes okay but when verifying the drive does not go beyond track
|
||
0 and fdformat exits after a while with the message: `Non-removable
|
||
media not supported'. Well, I assure you, I CAN remove the floppy
|
||
:-). Any ideas what it could be?
|
||
|
||
Martin
|
||
|
||
------------------------------
|
||
|
||
From: dwm@shell.portal.com (David - Morris)
|
||
Subject: Re: Could TCP/IP be implemented over SCSI?
|
||
Date: 29 Sep 1994 05:00:32 GMT
|
||
|
||
While ya'll are at it, shared scsi hard drives and/or CDROMs might be
|
||
intersting as well ... even if restrictions to one host r/w others r/o.
|
||
Source trees, netnets, etc. would be obvious candidates.
|
||
|
||
Also, when the discussion finishes offline, do post your results/plans.
|
||
|
||
Dave Morris
|
||
|
||
------------------------------
|
||
|
||
Crossposted-To: comp.os.linux.misc,comp.os.linux.admin,alt.uu.comp.os.linux.questions
|
||
From: iialan@iifeak.swan.ac.uk (Alan Cox)
|
||
Subject: Re: How to use a host as a router - READ THIS
|
||
Date: Wed, 28 Sep 1994 10:18:01 GMT
|
||
|
||
In article <CwLFH1.B3p@tasking.nl> fvm@tasking.nl (Frank van Maarseveen) writes:
|
||
>This was done in order to route something across the same physical network
|
||
>without the need for an extra adapter. The second "virtual" adapter was
|
||
>referred to as le0:1, the ":1" part appended to the original adapter name.
|
||
>Though a bit unusual, there's nothing wrong with this I think.
|
||
>I suppose there are no plans yet for implementing this feature in linux.
|
||
|
||
If Solaris takes the 4.4 BSD route it has a linked list of addresses for
|
||
each address family. This worries me for performance reasons. Linux can
|
||
happily do what you want with the dummy inteface (and you can add more
|
||
dummy interfaces 8)).
|
||
|
||
ifconfig dummy my.other.addr.ess up
|
||
arp -s my.other.addr.ess MY:ET:HE:RN:ET:NM
|
||
|
||
Alan
|
||
|
||
|
||
--
|
||
..-----------,,----------------------------,,----------------------------,,
|
||
// Alan Cox // iialan@www.linux.org.uk // GW4PTS@GB7SWN.#45.GBR.EU //
|
||
``----------'`----------------------------'`----------------------------''
|
||
|
||
------------------------------
|
||
|
||
Crossposted-To: comp.os.linux.help
|
||
From: tom@darshiva.efrei.fr (Thomas Parmelan)
|
||
Subject: Re: SMail security hole?
|
||
Date: Wed, 28 Sep 1994 23:46:49 GMT
|
||
|
||
John Henders <20>crivait:
|
||
> If smail is doing this, it must be misconfigured. In the
|
||
>transports file there should be a line in the appendfile driver that
|
||
>tells smail to do the file append as the user who the mail is addressed
|
||
>to. The syntax is append_as_user. The above .forward hack doesn't work
|
||
>here, as the following debug extract will show.
|
||
>
|
||
>director <dotforward> matched <jh>
|
||
> directed jh --> /etc/passwd ... send to file transport
|
||
>transport file uses driver appendfile
|
||
> appendfile: write to file /etc/passwd
|
||
>insert_addr_list() called:
|
||
> ERR133: transport file: failed to open output file: Permission denied
|
||
> /etc/passwd
|
||
>write_log: /etc/passwd ... failed: (ERR_133) transport file: failed to
|
||
>open output file: Permission denied
|
||
|
||
This is because you already have a file named /etc/passwd.
|
||
Try it with a file that doesn't already exist (eg. /this.is.a.test).
|
||
It really works :-(
|
||
|
||
(Tip: if you don't have a /.rhosts, create an empty one...)
|
||
|
||
--
|
||
Thomas Parmelan - tom@darshiva.efrei.fr
|
||
|
||
------------------------------
|
||
|
||
From: becker@cesdis.gsfc.nasa.gov (Donald Becker)
|
||
Subject: Re: fiber optic ethernet cards
|
||
Date: 29 Sep 1994 21:14:16 -0400
|
||
|
||
In article <corey.6bae@bbs.xnet.com>, Corey Sweeney <corey@bbs.xnet.com> wrote:
|
||
>has anyone seen a fiber optic ethernet driver, or does anyone have a intention
|
||
>of creating one?
|
||
>
|
||
>I havn't been able to find one.
|
||
|
||
If you mean ethernet adaptors that have 10baseF interfaces, Allied Telesis
|
||
has 10baseF versions of their AT1500 (79C960 LANCE) and AT1700 (Fujitsu
|
||
MB86965) ethercards.
|
||
|
||
There are no supported FDDI/CDDI, or ATM network cards at this time.
|
||
|
||
--
|
||
Donald Becker becker@cesdis.gsfc.nasa.gov
|
||
USRA-CESDIS, Center of Excellence in Space Data and Information Sciences.
|
||
Code 930.5, Goddard Space Flight Center, Greenbelt, MD. 20771
|
||
301-286-0882 http://cesdis.gsfc.nasa.gov/pub/people/becker/whoiam.html
|
||
|
||
------------------------------
|
||
|
||
From: dwm@shell.portal.com (David - Morris)
|
||
Subject: Re: Could TCP/IP be implemented over SCSI?
|
||
Date: 29 Sep 1994 05:07:02 GMT
|
||
|
||
btway ... I am aware of a case where a special purpose computer had a spare
|
||
scsi adapter which was used to connect to a 'normal' DOS PC. It is in
|
||
regular use. The objective was to improve on 9.6kbs data transfers.
|
||
To avoid bus length problems, the guy who did it slowed the data rate way
|
||
down (don't know how) and uses 50kbyte/sec. Just to confirm that
|
||
computer adapter to adapter is possible.
|
||
|
||
------------------------------
|
||
|
||
Crossposted-To: comp.os.linux.announce
|
||
From: cgr@poplar1.cfr.washington.edu (Carl G. Riches)
|
||
Subject: AIC-7770 mailing list created
|
||
Reply-To: cgr@poplar1.cfr.washington.edu (Carl G. Riches)
|
||
Date: Fri, 30 Sep 1994 15:01:32 GMT
|
||
|
||
Announcing the creation of a mailing list for developers and testers of
|
||
device drivers for AIC-7770-based SCSI devices. To subscribe, send mail
|
||
to:
|
||
|
||
aic7770-list-request@poplar1.cfr.washington.edu
|
||
|
||
with the body of the message containing the following:
|
||
|
||
SUBSCRIBE AIC7770-LIST <your name here, without the angle brackets>
|
||
|
||
do _not_ include anything else, not even a sig.
|
||
the mailing lists's welcome message follows . . .
|
||
|
||
carl
|
||
|
||
carl g. riches
|
||
college of forest resources internet: cgr@poplar1.cfr.washington.edu
|
||
university of washington ar-10 voice: 206-543-2764
|
||
seattle, wa 98195 fax: 206-543-3254
|
||
|
||
|
||
=======================================================================
|
||
|
||
Welcome to the AIC-7770 electronic mailing list. This list is for
|
||
people who are working on device drivers for AIC-7770-based controllers
|
||
for Linux. Use this list to communicate with other people who are
|
||
developing or testing AIC-7770 device drivers (AHA-274x/AHA-284x/AIC-7770
|
||
motherboard).
|
||
|
||
Your messages to aic7770-list@poplar1.cfr.washington.edu will be broadcast
|
||
to the entire membership. We need a high signal-to-noise ratio -- please
|
||
help conserve bandwidth.
|
||
|
||
you can get help about the AIC-7770 mailing list manager and learn how to
|
||
use it by sending e-mail to listproc@poplar1.cfr.washington.edu. put only
|
||
the word "help" (without the quotes) in the body of your message. send
|
||
reports about problems to cgr@poplar1.cfr.washington.edu, who will try to
|
||
help resolve the problem.
|
||
|
||
carl
|
||
|
||
(cgr@poplar1.cfr.washington.edu)
|
||
|
||
--
|
||
Send submissions for comp.os.linux.announce to: linux-announce@tc.cornell.edu
|
||
Be sure to include Keywords: and a short description of your software.
|
||
|
||
------------------------------
|
||
|
||
From: pp000932@interramp.com
|
||
Subject: Linux Console Device
|
||
Date: Thu, 29 Sep 94 06:31:47 PDT
|
||
|
||
|
||
I have been playing around a little with the Linux console and am curious how
|
||
or if it can be switched into an 8 bit character mode and how to get it to
|
||
support color ANSI sequences (similar to 'ansipc' emulation present in
|
||
Coherent's console driver). I can use DEC SI/SO to switch in/out DEC
|
||
graphics character sets, and it seems to perform correctly as a vt100
|
||
console, but I am looking for ANSI color support. Is there an escape sequence
|
||
which controls the console's mode of operation?
|
||
|
||
I have looked a little at the source for console, but I have not been able to
|
||
devine a mechanism to accomplish this from it yet.
|
||
|
||
|
||
------------------------------
|
||
|
||
Subject: Re: 1+ Gig SCSI Drives
|
||
From: rogersd@epaus.island.net (Daniel Rogers)
|
||
Date: 29 Sep 1994 10:28:41 -0700
|
||
|
||
In article <36dal0$iei@tuba.cit.cornell.edu>,
|
||
S. Lee <sl14@crux3.cit.cornell.edu> wrote:
|
||
>If what I've experienced is the same to yours, you can just patch the
|
||
>Buslogic driver to make it return 'translation off' instead of on.
|
||
>Actually, OFF was the default when the driver was introduced into the
|
||
>kernel, but somewhere around 1.1.46 it changed, and I remember it
|
||
>breaking LILO on my system (I have a Micropolis 1.7G but the boot
|
||
>partition is on < 1G).
|
||
|
||
Well, the first symptom was that LILO wouldn't come up. I though "Hmmm,
|
||
thats odd." and then just booted it with a floppy.
|
||
|
||
Then the crashes began, and it would run e2fsck each time, getting worse and
|
||
worse, turned out the partitions were overlapping between the 1 gig news
|
||
partition and the 20 meg swap partition, so as soon as the news partition
|
||
got full, it began writing into the swap space.
|
||
|
||
Not good.
|
||
|
||
>If you're interested I can dig out the patch for you. But you seem to
|
||
>have already found it, no? It's just a one-line change.
|
||
|
||
Actually, I just copied it off to another drive, turned on the extended
|
||
translation, re-partitioned and re-formatted the drive. It works fine now,
|
||
but I suppose I could have just changed that in the source.
|
||
|
||
It would be nice if that were a little better documented somewhere.
|
||
|
||
Dan.
|
||
--
|
||
Daniel Rogers | "Good tea, nice house." - Worf
|
||
(rogersd@epaus.island.net) | Linux - The choice of a GNU generation.
|
||
|
||
------------------------------
|
||
|
||
Crossposted-To: news.software.b
|
||
From: henry@zoo.toronto.edu (Henry Spencer)
|
||
Subject: Re: C-News cleanup release finds bug in GNU 'join'
|
||
Date: Fri, 30 Sep 1994 16:57:40 GMT
|
||
|
||
In article <36dlu4$e8g@Thunder.McRCIM.McGill.EDU> steve@CIM.McGill.CA (Steve Robbins) writes:
|
||
>...[Another annoying thing on linux: the standard includes
|
||
>insist on #define'ing atol(), which breaks the declaration in
|
||
>c-news/include/libc.h...
|
||
|
||
Grr, I thought I'd stamped out C News's bad habit of trying to declare the
|
||
system functions. I'll have to be more thorough.
|
||
|
||
My thanks to Steve for tracking down the GNU join bug. I haven't yet
|
||
determined whether this is the source of the "awk problem" some people
|
||
have reported.
|
||
--
|
||
Justice for groups that doesn't include justice | Henry Spencer
|
||
for individuals is a mockery. | henry@zoo.toronto.edu
|
||
|
||
------------------------------
|
||
|
||
From: newcombe@aa.csc.peachnet.edu (Dan Newcombe)
|
||
Subject: Re: Could TCP/IP be implemented over SCSI?
|
||
Date: Wed, 28 Sep 1994 14:25:24 UNDEFINED
|
||
|
||
In article <ianm.780705652@miles> ianm@qualcomm.com (Ian McCloghrie) writes:
|
||
>lim@vector.gs.tandem.com (myers_lincoln) writes:
|
||
>> I read in the SCSI FAQ that two SCSI hosts can share SCSI peripherals
|
||
>>on the same bus. Is it possible for these two hosts to send commands to each
|
||
>>other?
|
||
>It is possible, yes. I've heard of it being done before. It's not
|
||
>terribly practical though, because the maximum length of a scsi bus
|
||
>is on the other of 6 meters, which doesn't give you much to work
|
||
>with... You've also gotta be careful not to have the two systems
|
||
>both try to access one drive at the same time.
|
||
|
||
But even with the 6 metre limitation, it still would be great for several PC's
|
||
in the same room.
|
||
|
||
>>having SoundCardNet. Sound Cards would record each other's audio output from
|
||
>>across the room. True short range wireless communication, though sleeping in
|
||
>Ulch. Your error rate would be atrocious, I would imagine.
|
||
>"Hey! John! Turn your music down, it's making me drop packets!"
|
||
|
||
Well, I suppose that you could interconnect the line in/line out's.
|
||
|
||
Now, you put this all on a mixer and:
|
||
"Hey! John! Could you mix me in a 50% mix of sunsite and tsx-11 on the left
|
||
channel and netnews, gopher, and WWW on the right channel."
|
||
|
||
:)
|
||
|
||
--
|
||
Dan Newcombe newcombe@aa.csc.peachnet.edu
|
||
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
|
||
"And the man in the mirror has sad eyes." -Marillion
|
||
|
||
------------------------------
|
||
|
||
From: jaf@jaflrn.liii.com (Jon Freivald)
|
||
Subject: Re: Does Quicken work under DOSEMU?
|
||
Date: Fri, 30 Sep 1994 02:00:45 GMT
|
||
|
||
Justin Edmond Zaglio (jez5@bonjour.cc.columbia.edu) wrote:
|
||
> The title pretty much says it all: doe any of the DOS flavors of
|
||
> Quicken work under DOSEMU? If not, are there any financial packages
|
||
> that *do* work under it?
|
||
|
||
I'm using v6 almost daily under dosemu 0.5pl1. No problems.
|
||
|
||
Jon
|
||
--
|
||
Jon Freivald ( jaf@jaflrn.liii.com )
|
||
PGP V2 - 22A829/40 DA 9E 8E C0 A1 59 B2 46 3B 73 81 2B 7B 83 1F
|
||
Nothing is impossible for the man who doesn't have to do it.
|
||
|
||
------------------------------
|
||
|
||
From: bruce@cs.su.oz.au (Bruce Janson)
|
||
Subject: Re: 900 MHz CB band???
|
||
Date: 30 Sep 1994 11:52:35 +1000
|
||
|
||
In article <36ef6j$891@portal.gmu.edu>,
|
||
Shawn C. Masters <smasters@bzy.gmu.edu> wrote:
|
||
>Vassili Leonov (vassili@cs.sunysb.edu) wrote:
|
||
>: the WaveLan card, designed by NCR, that gives you about 2.5 Kbit/sec
|
||
>: link as far as you can beam 1Watt at around 910 MHz. This band is
|
||
>: at least for Spread Spectrum operation, which WaveLan is using.
|
||
>
|
||
>: p.s. Well - you CAN build a backbone using this stuff. Though they are
|
||
>: rather expensive - about $700 per card...
|
||
>
|
||
> I think that is 2.5 Mbit/sec, ..
|
||
>..
|
||
|
||
The NCR WaveLAN cards give a maximum throughput of 2 Mbits/sec.
|
||
The cards come in ISA, EISA, Micro Channel, PCMCIA and parallel
|
||
port(?) flavours.
|
||
In the USA they signal (using spread spectrum encoding) at 915 MHz.
|
||
Indoors, using just the omni-directional antennas that accompany
|
||
each unit, they can be used to connect devices within a large,
|
||
open-plan office. Outdoors, using (extra cost) directional antennas,
|
||
they are able to support point-to-point links, apparently up to 8 km.
|
||
I understand that DEC, Solectec and Digital Ocean also sell rebadged
|
||
versions of the same hardware.
|
||
|
||
Cheers,
|
||
bruce.
|
||
|
||
Bruce Janson Email: bruce@cs.su.oz.au
|
||
Basser Department of Computer Science Phone: +61-2-692-3423
|
||
University of Sydney, N.S.W., 2006, AUSTRALIA Fax: +61-2-692-3838
|
||
|
||
------------------------------
|
||
|
||
From: hjh@snake8.imsor.dth.dk (Henrik Juul Hansen)
|
||
Subject: Is there a memory allocation debugging tool for Linux?
|
||
Date: 30 Sep 1994 15:42:43 GMT
|
||
|
||
Are there any tools available for debugging memory allocation in
|
||
c-programs running Linux. I know of a commercial program called `purify`
|
||
but if there is a GNU or public domain program I would be happy if
|
||
somebody could point me to it.
|
||
|
||
To check the use of `malloc`, use of allocated memory etc.
|
||
|
||
|
||
Please send a copy of reply by E-mail to:
|
||
hjh@imm.dtu.dk
|
||
|
||
|
||
|
||
Sincerely,
|
||
|
||
|
||
Henrik Juul Hansen
|
||
|
||
|
||
|
||
------------------------------
|
||
|
||
From: drew@frisbee.cs.Colorado.EDU (Drew Eckhardt)
|
||
Subject: Re: Could TCP/IP be implemented over SCSI?
|
||
Date: 28 Sep 1994 18:24:35 GMT
|
||
|
||
In article <NEWTNews.24475.780779924.jbarrett@xterm1.fone.com>,
|
||
<jbarrett@onramp.net> wrote:
|
||
>
|
||
><lim@vector.gs.tandem.com> writes:
|
||
>>
|
||
>> I read in the SCSI FAQ that two SCSI hosts can share SCSI peripherals
|
||
>> on the same bus. Is it possible for these two hosts to send commands to each
|
||
>> other?
|
||
|
||
Yes.
|
||
|
||
>> I am asking because I would like to know how viable it would be to add
|
||
>> support to Linux for TCP/IP over SCSI, which might be practical for two or
|
||
>> three machines which already have SCSI support.
|
||
>>
|
||
|
||
There are some ugly implications of this, and it isn't possible with some
|
||
intelligent boards which have hidden enough of the underlying details from
|
||
the programmer and don't support a target mode, but otherwise it is
|
||
possible.
|
||
|
||
>We did something like this about 7 years back at one company I worked for....
|
||
>Passing messages over the SCSI bus.... so here goes...
|
||
>
|
||
>1. Each host on the SCSI bus must have a separate SCSI ID.. The more hosts the
|
||
>less devices... only 7 ID's available (i.e. host1=7 host2=6 cdrom1=5 hd0=0)
|
||
>
|
||
|
||
This is the first problem you're going to run into -
|
||
|
||
You need some mechanism for telling SCSI boards what their
|
||
addresses are.
|
||
|
||
Furthurmore, you need a mechanism to limit the devices detected
|
||
on power up/boot if you want to avoid conflicts.
|
||
|
||
>2. The Linux SCSI driver would have to be hacked to accept async command
|
||
>message from the bus (just like an HD does..
|
||
|
||
You're probably going to want to implement the SCSI Processor
|
||
device model, since this is the only model the firmware on some
|
||
intelligent boards can support in target mode.
|
||
|
||
The simplest method is to choose a sender-driven model,
|
||
where the sender connects to a device and sends a
|
||
SEND command to it.
|
||
|
||
Alternateively, you could have it receiver driven, where the
|
||
receivers execute a RECEIVE command on each sender, which in
|
||
turn disconnects, and reconnects when data is available, although
|
||
this is more complicated.
|
||
|
||
>but most drivers don't support the
|
||
>capability by default)...
|
||
|
||
None of the Linux SCSI drivers have any notion of support for target mode.
|
||
|
||
>whats worse.. it takes an inteligent SCSI card to do
|
||
>this efficiently (Adaptec 154x or better, 1520 wont handle it).
|
||
|
||
Not necessarily - most of the dumb boards will interrupt when a
|
||
SCSI selection or reselection occurs. Given a typical interrupt
|
||
latency of about 60us, amortized over a large MTU (say 64K), this
|
||
isn't too bad.
|
||
|
||
> Next, the SCSI
|
||
>card needs to interrupt when a packet arrives (1520 works in polled mode so it
|
||
>would take buku cpu time to monitor for messages)...
|
||
|
||
Selection/reselection generally trigger interrupts.
|
||
|
||
> (?? Or am I lying and the Linux SCSI DRIVERS will accept ASYNC CMDS ??)
|
||
>
|
||
>3. Treat the interface just like an ethernet card.. but with no broadcast
|
||
>capability, and fixed ARP tables on each machine.. Assign the SCSI bus its own
|
||
>IP Network or SubNet.. then the ARP tables tell you the SCSI ID of the target
|
||
>host... Sending a packet is just like issuing a command to a HD... and if the
|
||
>target device times out responding to the command... drop the packet on the
|
||
>floor... Just like ENet does if the target host is not available...
|
||
>
|
||
>Voila... 5 MByte per second TCP/IP! (or more for SCSI-2)
|
||
>
|
||
>That should cover the essentials.. NOW:
|
||
> HOW MANY SCSI DEVICE DRIVERS NEED TO BE HACKED TO SUPPORT THIS ??
|
||
|
||
All the ones you want to use.
|
||
|
||
> I'm not a kernel guru.. but I'll take a look at the SCSI drivers over
|
||
>the next week or so and see if there is any hope... I've already looked at the
|
||
>network driver interfaces, and they appear to be no problem...
|
||
|
||
I've thought about it some, and it's _definately_ possible. In fact,
|
||
it appears to be a nice way to do distributed communications.
|
||
|
||
>So as final answer to Lincoln's question:
|
||
> Yes its is POSSIBLE to run TCP/IP over SCSI...
|
||
> Ready to get to work writing it ???
|
||
|
||
If anyone is interested, we should probably take the discussion to
|
||
E-mail, perhaps on the SCSI channel of the mailing list.
|
||
|
||
--
|
||
Since our leaders won't respect The Constitution, the highest law of our
|
||
country, you can't expect them to obey lesser laws of any country.
|
||
Boycott the United States until this changes.
|
||
|
||
------------------------------
|
||
|
||
From: drew@frisbee.cs.Colorado.EDU (Drew Eckhardt)
|
||
Subject: Re: Could TCP/IP be implemented over SCSI?
|
||
Date: 28 Sep 1994 18:29:23 GMT
|
||
|
||
In article <JASON.94Sep28095634@idiom.berkeley.ca.us>,
|
||
Jason Venner <jason@idiom.berkeley.ca.us> wrote:
|
||
>
|
||
>With differential scsi, your scsi bus can extend quite a distance.
|
||
>With SCSI 2 fast wide/SCSI3 I think you are limited to 256 id's and 256 feet
|
||
>for the scsi bus.
|
||
|
||
Nope. Differential busses are limited to 25 meters total versus
|
||
six for single ended. 16 bit WIDE busses allow for 16 devices.
|
||
|
||
>So: given a little change in technology, networking via scsi will make a lot
|
||
>of sense.
|
||
|
||
Not really -
|
||
|
||
1. Cabling with 25 pair shielded cable is _not_ cheap.
|
||
|
||
2. 100 base T is a 100Mbit/sec technology which can be wired with
|
||
4 pair cables.
|
||
|
||
However, it would be a reasonable way to connect a few boxes on a
|
||
desktop to get a poor-man's distributed computing system.
|
||
--
|
||
Since our leaders won't respect The Constitution, the highest law of our
|
||
country, you can't expect them to obey lesser laws of any country.
|
||
Boycott the United States until this changes.
|
||
|
||
------------------------------
|
||
|
||
From: zerucha@shell.portal.com (Thomas E Zerucha)
|
||
Subject: How to get more kernel memory for kmalloc?
|
||
Date: 30 Sep 1994 02:16:31 GMT
|
||
|
||
To continue with my Telnet troubles (it stops transmission), I am almost
|
||
sure that it is happening because of (ultimately) a failed kmalloc. Telnet
|
||
tries a send, and gets a EWILLBLOCK/EAGAIN. There are only two places this
|
||
error happens in tcp.c in two places (in tcp_write), once when waiting for
|
||
a previously open connection to close, and when a socket buffer malloc fails.
|
||
|
||
I can understand with the various protocols, drivers, devices, etc. how I can
|
||
be running out from a default config, but:
|
||
|
||
HOW DO I INCREASE THE KMALLOCABLE MEMORY?
|
||
---
|
||
zerucha@shell.portal.com - main email address
|
||
|
||
|
||
------------------------------
|
||
|
||
Crossposted-To: comp.os.386bsd.apps,comp.os.386bsd.bugs,comp.os.386bsd.misc,comp.os.386bsd.questions,comp.os.linux.admin,comp.os.linux.help,comp.os.minix
|
||
From: cl@dfw.net (Cyber Link)
|
||
Subject: Re: Special Sale On QNX!
|
||
Date: Fri, 30 Sep 1994 02:16:10 GMT
|
||
|
||
Well i am still trying to get Tnet to work on my machine... I have 1.5.10
|
||
and have applyed the patches. (I wrote a script to do it) now the kernel
|
||
will not compile!! when i do a make TNET it says i don't have enough
|
||
memory to run cem. I am curently workign on adding the -F to the CFLAGS
|
||
of all the Makefile's..... In other words this has been a long two week
|
||
project and i still can not see the end! Any Help from anyone would be
|
||
apreacheated(SP)
|
||
|
||
|
||
L8r
|
||
From
|
||
CL (CL@DFW.NET)
|
||
|
||
|
||
----Lots of other peoples Stuff------
|
||
|
||
Jay Ashworth (jra@zeus.IntNet.net)
|
||
wrote: : scheidel@gate.net (Michael S. Scheidell) writes:
|
||
: >scheidel@gate.net wrote:
|
||
: >NO HE DID NOT!
|
||
: >LOOK AT PATHS. THIS IS A FORGERY!
|
||
|
||
: So it is. My apologies.
|
||
|
||
: >PLEASE READ THIS. I DID NOT MAKE THAT POST!
|
||
: >This posting did not come from our office, but we are tracing it.
|
||
: >If you look carefully at headers you will find sites that don't exist,
|
||
: >and see that it does not trace back to scheidel@gate.net.
|
||
|
||
: I can never be sure... but they sure in hell don't match the headers on
|
||
: _this_ post.
|
||
|
||
: >Again, I did not do this post, and we have taken great pains to avoid
|
||
: >offending anyone, and to date this is my first flame, even if undeserved.
|
||
: >I thank those who looked at headers and saw it was a forgery, and look
|
||
: >forward to a solution to this.
|
||
|
||
: Again... sorry about that. At least _my_ flame was polite. :-)
|
||
|
||
: Cheers,
|
||
: -- jra
|
||
: --
|
||
: Jay R. Ashworth High Technology Systems Comsulting Ashworth
|
||
: Designer Linux: The Choice of a GNU Generation & Associates
|
||
: ka1fjx/4
|
||
: jra@baylink.com "Hey! Do any of you guys know how to Madison?" 813 790 7592
|
||
|
||
------------------------------
|
||
|
||
From: joel@wam.umd.edu (Joel M. Hoffman)
|
||
Subject: [DOSEMU] Running dosemu disables serial port!!
|
||
Date: Fri, 30 Sep 1994 13:28:26 GMT
|
||
|
||
I'm running DOSEMU 0.52, and it's GREAT, except that when DOSEMU is
|
||
active on the console, nothing else can use the serial port! If I
|
||
have kermit or rz/sz running on the serial port, communication simply
|
||
stops. If I don't --- even worse --- mgetty goes crazy. Incoming
|
||
FAX's and logins are not processed.
|
||
|
||
I've already taken all the serial lines out of dosemu.conf. Is this a
|
||
known problem? Has it been fixed in any of the pre_53 releases? Is
|
||
there a workaround?
|
||
|
||
Thanks.
|
||
|
||
-Joel
|
||
(joel@wam.umd.edu
|
||
|
||
|
||
|
||
--
|
||
=============================================================================
|
||
|_|~~ Germany, Europe. 1943. "The diameter of the bomb was 30 centimeters,
|
||
__|~| 16 Million DEAD. and the diameter of its destruction, about 7
|
||
meters, and in it four killed and 11 wounded.
|
||
cnc Bosnia, Europe. 1993. And around these, in a larger circle of pain
|
||
cnc HOW MANY MORE? and time, are scattered two hospitals and one
|
||
cemetery. But the young woman who was buried in
|
||
the place from where she came, at a distance of more than
|
||
than 100 kilometers, enlarges the circle considerably. And the
|
||
lonely man who is mourning her death in a distant country incorporates
|
||
into the circle the whole world. And I won't speak of the cry of the orphans
|
||
that reaches God's chair and from there makes the circle endless and godless."
|
||
=============================================================================
|
||
Tell Clinton to stop the genocide: president@whitehouse.gov
|
||
|
||
------------------------------
|
||
|
||
From: mmarten@panix.com (Marten Liebster)
|
||
Subject: What GUI to write for?
|
||
Date: 30 Sep 1994 11:58:29 -0400
|
||
|
||
I want to write a X application or two. At first they would be for
|
||
personal use, but eventually I might make them availble for the
|
||
public to use.
|
||
|
||
I am not sure which GUI/toolkit to use. It would be nice to keep it
|
||
portable to use under various UIs. Do I have to use Xlib? or can I
|
||
write them using XView?
|
||
|
||
I would appreciate any guidence I could recieve. Thanks for any and
|
||
all help.
|
||
|
||
Marten
|
||
|
||
--
|
||
========================================
|
||
Marten M. Liebster Please no flames for spelling,
|
||
mmarten@panix.com I already know I can't spell!!
|
||
|
||
------------------------------
|
||
|
||
|
||
** 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-Development-Request@NEWS-DIGESTS.MIT.EDU
|
||
|
||
You can send mail to the entire list (and comp.os.linux.development) via:
|
||
|
||
Internet: Linux-Development@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-Development Digest
|
||
******************************
|