Linux-Hardware Digest #680, Volume #10 Tue, 6 Jul 99 04:13:39 EDT
Contents:
Re: RH6 w/TNT2, X res is about 320x200, can't change, can't navigate eaily ("mnip")
Dlink DFE-530TX.. (bono)
Re: Celeron, what's the catch? ("FM")
Re: Rackmount cases (wizard)
Re: Celeron, what's the catch? (Chris Robato Yao)
Re: Celeron, what's the catch? (kls)
Re: Yet another CDR / cdrecord question ("Whiplash!!!")
Re: No-name PnP ethernet card (Joe)
dhcp client ([EMAIL PROTECTED])
Re: Celeron, what's the catch? (Chris Robato Yao)
Re: How powerful a system do you need to run Linux as a server? ([EMAIL PROTECTED])
Re: Celeron, what's the catch? (L.Angel)
Re: Please help with Zip 100 IDE access error ("Bobby D. Bryant")
Re: Celeron, what's the catch? (Chris Robato Yao)
Re: Celeron, what's the catch? (George Macdonald)
NIC IRQ Problem ([EMAIL PROTECTED])
Re: Celeron, what's the catch? (Chris Robato Yao)
----------------------------------------------------------------------------
From: "mnip" <[EMAIL PROTECTED]>
Subject: Re: RH6 w/TNT2, X res is about 320x200, can't change, can't navigate eaily
Date: Mon, 05 Jul 1999 17:43:09 GMT
Thanks. Could you attach a copy of said file? I presume since its a
resonable-size text attachment, there won't be too much of an outcry.
> >I have tried every possible resolution and color depth configuration of
> >Xconfigurator, but the Gnome desktop with the enlightenment window
manager
> >always comes up in a ridiculously low resolution, I'm guessing 320x200.
>
> I think basically Xconfigurator does not work in this case.
> An XF86Config.TNT2 was posted to one of the Linux groups recently.
> I use it with a TNT card (Asus) and it seems to work well enough.
> Of course you must modify it for your monitor and mouse.
------------------------------
From: bono <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Dlink DFE-530TX..
Date: Tue, 06 Jul 1999 05:42:55 GMT
I don't think my Redhat 6.0 likes this still... I heard someone said it
is supported by Redhat 6 but I tried 2 of my DLink DFE-530TX and none
of them works so far... Any suggestion? I also have a DC528CT and it
also doesn't work.
Bono
------------------------------
From: "FM" <[EMAIL PROTECTED]>
Subject: Re: Celeron, what's the catch?
Date: Tue, 6 Jul 1999 01:59:36 -0400
Gene Heskett wrote:
>3; Based on the first two items, I'd sure be looking at
other sources
>for both the computer, and for the education.
Well I might have limited options for computers but I have
even more limited options for education, especially at this
point. Since I have no residence in the US, going to a
private college (with financial aid) is actually the only
affordable option. I already declined all other acceptances
during April when I chose Dartmouth.
>While *they* think its nice to have a captive audience for
their
>"Proprietary Program" thats so much better than the
competition (in
>their view) and thats much easier if all the hardware is
alike, but
>*you* will learn far more about the computer and its uses,
if *you* have
>to be the one to make the platform of your choice work
within the
>framework of *their* system.
>The school should be able to discuss the differences in an
intelligent
>manner, giving out freely the details of how this and that
works so that
>you can interface, and in doing that you'll learn a lot
that you would
>not learn in a structured classroom. If they refuse to
divulge those
>"proprietary" details, then run, do not walk, first to a
lawyer to get
>your money back, and while pressing that issue, to another
school.
>I've found that the huge majority of schools are in it for
the money.
>Some are worth it, many are not. But there is one
universal thing they
>understand, and thats taking your money elsewhere if they
don't give you
>what you *need*.
Well a few things. First, what I'm paying to Dartmouth
doesn't even cover the cost of living there, so at least as
far as I'm concerned, Dartmouth isn't in it for the money.
Second, I don't think the Dartmouth network is proprietary;
they certainly offer an option to bring one's own computer
and they will even help configure the machine if necessary.
But not everyone's as knowledgeable about computers as
people here and it's a lot easier for the support people if
most would purchase preconfigured systems from the college:
thus their recommendation. In fact the most prominent of
their core softwares, BlitzMail system clients/server are
open-source with full documentation of the protocol they
use. I doubt their intention is to limit the choice; it
seems to me like a sincere effort to promote the ease of
use, especially considering that the school requires
everyone to own a personal computer. After all they used to
be an Mac-only campus not so long ago and MacOS remains the
dominant platform. I'll be using Linux, an unsupported OS,
and I wouldn't be able to do this if the network was really
proprietary. On the other hand I'm not as confident on the
hardware side and buying from college remains an attractive
choice. Third, I believe Dartmouth offers financial aid for
computers purchased from the college (I have not yet
received the package so I'm not aware of details yet), so it
might even be an economical option. Fourth, I've already
developed a strong passion towards Dartmouth (amazing, isn't
it?).
Dan.
------------------------------
From: wizard <[EMAIL PROTECTED]>
Subject: Re: Rackmount cases
Date: Tue, 06 Jul 1999 01:58:47 -0400
Reply-To: [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
> Thanks to all that replied. I think that my solution is to get 19"
> rackmounted shelves and put a single computer on each shelf. It will
> not be as neat as I originally hoped for but better than what I have
> now.
>
> Thanks
> Scott
>
Hi Scott;
This idea is good as long as your CPUs are less than aproximately 18 "
in width. I only mentioned that because many Desktops where much
wider.
Thanks
Dave
[EMAIL PROTECTED] wrote:
> Thanks to all that replied. I think that my solution is to get 19"
> rackmounted shelves and put a single computer on each shelf. It will
> not be as neat as I originally hoped for but better than what I have
> now.
>
> Thanks
> Scott
>
> --
> Scott Boss
> Atlanta Perl Mongers Fearless Leader
> website: http://atlanta.pm.org
> community: http://www.dejanews.com/~apm
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
>
> --
> Scott Boss
> Atlanta Perl Mongers Fearless Leader
> website: http://atlanta.pm.org
> community: http://www.dejanews.com/~apm
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Chris Robato Yao)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: 6 Jul 1999 05:38:01 GMT
Reply-To: [EMAIL PROTECTED] (Chris Robato Yao)
In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls) writes:
>In article <7lrrtr$hgc$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>>
>>In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls) writes:
>>>& you have a ver poor ability of looking up benchmark results.
>>
>>
>>Where are your benchmark results by the way?
>
>not MY benchmark results. Where are YOURS? Couldn't find any eh? That
>poor look up ability of yours I suppose...
You are the one claiming a Dual Celeron setup (using 333s) can be faster
than a K6-III 450 on LInux. The Burden of Proof is on you.
But you have not shown anything to back your claims.
Have you seen http://www.cpureview.com?
Article shows a K6-III 400 is faster than a Cel @ 450 in compiling.
>
>>You have not yet answered my other question on your ratios of K6 vs. PII
>>FPUs.
>
>benchmarks, hardware & programming sites. That & I own a dog of a cpu(k6-2
I have not see any benchmark or website that suggests a .55 ratio for
the P2 FPU vs. K6. You made this up. Back it up.
On the contrary, there is plenty that suggests otherwise. Read the
K6-III review of http://www.combatsim.com. According to them, a K6-III
450 is barely equivalent to a PII 400 on non 3DNow games.
>300. old core: .55 p2 fpu). On a personal note, if the performance were so
>ubar compared to p2's, as you keep on insisting, my frame rate wouldn't drop
>to 16-9fps in warbirds when things get even slightly heavy. I must be
>dreaming eh?
I can get an Celeron overclocked to 500 to drop down to 16-9 FPS on
Jane's WWII Fighters when things get heavy. Does not mean anything.
Rgds,
Chris
>
>
(And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
------------------------------
From: [EMAIL PROTECTED] (kls)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Tue, 06 Jul 1999 05:54:14 GMT
In article <7ls3v5$p7j$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>
>In <7lqjgc$btc$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (Salem Lee
Gan
>zhorn) writes:
>>Chris Robato Yao ([EMAIL PROTECTED]) wrote:
>>: its time, while the CPU is idle. Spreadsheets and compiling won't be
>>: taking advantage of duals since most of them aren't designed for it, but
>>
>>gmake.. you will see huge gains in compiling with multiple processors.
>
>When you're doing multiple compiles at the same time, possibly.
>
>>Compiling of modules is completely independent and parallel processing is
the
>>way to g.
>
>Unless the compiling itself is multithreaded, you won't see it. As a
>matter of fact, compiling may tend to benefit more on a system with
>large caches, and the K6-III is about 2x of a Celeron.
There are threads & there are PROCESSES. & you delared I have a poor
understanding of multiprocessing:) hehe.
------------------------------
From: "Whiplash!!!" <[EMAIL PROTECTED]>
Subject: Re: Yet another CDR / cdrecord question
Date: Tue, 06 Jul 1999 05:55:01 +0000
Well, I finally got this one licked. Scott came up with the winning tip
which is pasted below...
I had to recompile the kernel to disable the ATAPI support and enable
SCSI emulation. I still don't really understand this since everything I
have read says you are supposed to be able to load modules on the fly.
However, by compiling these into the kernel everything is working and I
am burning CD's so I'm not complaining. Thanks to everyone for the
input.... now to get my DAT drive working.....
Dana
In comp.os.linux.hardware you write:
>My Redhat 6 system is configured with my boot drive on an Adaptec 2940
>SCSI card. My CDRW drive is an ATAPI device configured as the master
>drive on the first IDE channel. Can someone tell me how to load the
>ide-scsi module at boot up and have it handle the ATAPI device? At
this
>point cdrecord only sees the "real" scsi bus and not the ATAPI device.
RedHat 6.0 makes this really easy. Change directory to /usr/src/linux
and do a "make xconfig". Look around and turn off ATAPI support and
turn on SCSI emulation. When you quit it will tell you to "make
depend"
do that and then "make". Now you need to install the new kernel, since
there are many ways to set up the system, I'd recommend you check out
the HOWTOs and do what's right for yours.
You'll also need to fix your link in /dev. Once the CD drives are
being treated as SCSI they will have SCSI style names.
Scott
------------------------------
From: Joe <[EMAIL PROTECTED]>
Subject: Re: No-name PnP ethernet card
Date: Tue, 06 Jul 1999 05:09:01 GMT
On Sat, 3 Jul 1999, Will Schmid wrote:
> How does one configure a PLUG & PLAY ethernet card in slackware 3.5?
>
> At startup, BIOS detects it as: NB116P PnP
>
> Win 98 device manager says:
> IRQ: 5
> I/O: 220
>
> Thanks... and excuse me, I'm a newbie.
man isapnp
man pnpdump
------------------------------
From: [EMAIL PROTECTED]
Subject: dhcp client
Date: Tue, 06 Jul 1999 06:23:33 GMT
Hello everybody,
I want to connect my laptop running Suse 6.1 to an Windows NT network.
The IP is dynamically assigned by Dhcp. I got it working for
Windows 98, so the hardware should be OK. For linux I find a
"timed out waiting for a valid dhcp server response." messages in
/var/log/messages. Any suggestion is strongly appriciated.
Thanks
Tilman
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Chris Robato Yao)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: 6 Jul 1999 06:29:40 GMT
Reply-To: [EMAIL PROTECTED] (Chris Robato Yao)
In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls) writes:
>In article <7ls3v5$p7j$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>>
>>In <7lqjgc$btc$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (Salem Lee
>Gan
>>zhorn) writes:
>>>Chris Robato Yao ([EMAIL PROTECTED]) wrote:
>>>: its time, while the CPU is idle. Spreadsheets and compiling won't be
>>>: taking advantage of duals since most of them aren't designed for it, but
>>>
>>>gmake.. you will see huge gains in compiling with multiple processors.
>>
>>When you're doing multiple compiles at the same time, possibly.
>>
>>>Compiling of modules is completely independent and parallel processing is
>the
>>>way to g.
>>
>>Unless the compiling itself is multithreaded, you won't see it. As a
>>matter of fact, compiling may tend to benefit more on a system with
>>large caches, and the K6-III is about 2x of a Celeron.
>
>There are threads & there are PROCESSES. & you delared I have a poor
>understanding of multiprocessing:) hehe.
Sure. A single process has no benefit on a multiprocessing system
because the operating system and the hardware has no inherent capability
of dividing a single process into multiple threads (when it is not
multithreaded in the first place) so it can schedule it among different
processors.
Thus with such a process, the dual CPU is no faster than a single CPU of
the same clock speed, and in fact, it can be lower due to memory
contentions, bus contentions, the additional overhead of the OS, etc,.
In fact, the memory requirements of a multiprocessing system must be
higher than a single processor system, adding to your cost as well.
The nature of multithreading in desktop applications tend to have
threads that are more likely to be I/O or user interface bound, and so
are often waiting for input. Multithreaded operating systems also tend
to have threads that are I/O or input bound. With such, you have a
processor doing the work, and another one spinning its wheels.
Applications also tend to use locks. You cannot execute two threads
simultaneously, if one thread is going to be require the result for the
other, or if the threads compete on the same data resources. Thus the
second thread is not allowed to run until the first result is
available or access the same resource, until the first one has finished.
Again, the extra processor is spinning its wheels.
Just because you have multiple processes and multiple threads still does
not SMP an ideal solution. You have to see if these threads and
processes are highly CPU bound.
Rgds,
Chris
(And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: alt.linux
Subject: Re: How powerful a system do you need to run Linux as a server?
Date: 6 Jul 1999 01:35:06 -0500
I have a freind that has been running a server on an old 386 for
years. It functions as a webserver , mail server, newssever etc.
An astounding amount of mail passes through that old box due to
various mailing lists hosted on it. He also has over 300 people
with shell accounts on it.
I'm relativly new to Linux myself. I found the SuSe distro to be
superior to Redhat , and the support for SuSe is light years ahead of
Redhat's. Also if you buy the boxed cds of SuSe you get 5 cds of
programs to use.
If you set up a server make sure you secure it well. My box
gets several scans a day by hackers trying to find a vulnerable
system.
On Tue, 06 Jul 1999 01:26:15 GMT, [EMAIL PROTECTED] wrote:
>
>
> Hello, I'm new to Linux. I would like to know how powerful of a
>machine you would need to run Linux as a server? Let's say you had a
>DSL connection, would you be able to use Linux as a webserver?
>What would be the best release for me to start with? Red Hat?
>
>thanks for your time,
>
>D.
>
>
>Sent via Deja.com http://www.deja.com/
>Share what you know. Learn what you don't.
------------------------------
From: a?n?g?e?l?@lovergirl-DOT.com (L.Angel)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Tue, 06 Jul 1999 05:27:13 GMT
Reply-To: ?a?n?g?e?[EMAIL PROTECTED]
>You have not yet answered my other question on your ratios of K6 vs. PII
>FPUs.
I'm all for AMD but based on a simple Pascal DOS based program, no
pipelining obviously, running a series of 20million pre-generated
single precision operations, P2-233 takes 235msec (85.1M/s), K6-233
takes 440msec(45.45M/s), K6-2-350@400 takes 258msec(77.5M/s).
45.45:85.1 is about 0.53 : 1
(77.5/400) to (85.1/233) is about 0.53 : 1 too
Though it isn't a certified benchmark, I think it's good enuff to
say that the fpu power of the K6 to the P2 is like kls claimed. Though
I don't know where he got his numbers from.
The little lost angel & her featherhead's 2 cents of dreaminess. :)
Email : Figure out what to remove, I'm getting tired of spam
------------------------------
From: "Bobby D. Bryant" <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc
Subject: Re: Please help with Zip 100 IDE access error
Date: Tue, 06 Jul 1999 01:45:48 -0500
[EMAIL PROTECTED] wrote:
> I am running Red Hat Linux 6.0, kernel version 2.2.5-15.
>
> I have an IDE system with a 4.3Gb hard drive as the master on IDE0, an
> IOMEGA Zip 100Mb as the slave on IDE0, and a CD-Rom as the master on
> IDE1 (this is where I stopped moving cables, jumpers, and drives...).
>
> When I attempt to use the Zip drive to move data to or from the system,
> I am getting the following messges on the console and in the
> /var/log/messages file (note: this is only two 'sets' of errors, I get
> many more at different sector numbers,)
>
> Jul 5 12:28:44 alrod kernel: ide-floppy: hdb: I/O error, pc = 2a, key =
> 4, asc = 47, ascq = 0
> Jul 5 12:28:44 alrod kernel: end_request: I/O error, dev 03:44 (hdb),
> sector 7509
> Jul 5 12:28:49 alrod kernel: ide-floppy: hdb: I/O error, pc = 2a, key =
> 4, asc = 47, ascq = 0
> Jul 5 12:28:49 alrod kernel: end_request: I/O error, dev 03:44 (hdb),
> sector 7641
>
> Could anyone shed some light on this. I have been using Zip drives (IDE
> and SCSI) on older versions of Red Hat without difficulty. I need these
> drives in order to transfer data to and from client sites, and the
> machine I just put together is going to be in Mexico tomorrow (I and in
> Los Angeles, Calif.). Any help is greatly appreciated.
Unfortunately, that's a known problem with the kernel in RH6. (The 2.2.5-22
upgrade on their 'errata' site doesn't fix it either.)
There are supposed to be two possible workarounds, though I haven't gotten
around to trying either yet.
1) Update to one of the recent 'ac' patches to 2.2.10. When I first
discovered the problem I searched dejanews, and found a message from Alan
Cox reporting that he had fixed it in his kernel tree.
2) Build a custom kernel and fake it into thinking your IDE/ATAPI Zip is a
SCSI Zip. This sounds screwy, but someone in this group reported success
with it, so you may want to skim the messages for the last couple of weeks
and find his report.
Red Hat is aware of the problem, because the same guy told me he was
talking to them about it. There hasn't been any word of a kernel update
coming out for it, though. (A bit of a surprise -- surely thousands of
potential RH6 users have internal IDE Zip drives?)
Dejanews is your friend: search for 'zip' 'linux' 'error' 'ATAPI', etc, and
you should turn up the instructions for the two fixes.
Good luck,
Bobby Bryant
Austin, Texas
------------------------------
From: [EMAIL PROTECTED] (Chris Robato Yao)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: 6 Jul 1999 06:55:47 GMT
Reply-To: [EMAIL PROTECTED] (Chris Robato Yao)
In <7ls7k6$krk$[EMAIL PROTECTED]>, a?n?g?e?l?@lovergirl-DOT.com (L.Angel) writes:
>>You have not yet answered my other question on your ratios of K6 vs. PII
>>FPUs.
>
> I'm all for AMD but based on a simple Pascal DOS based program, no
>pipelining obviously, running a series of 20million pre-generated
>single precision operations, P2-233 takes 235msec (85.1M/s), K6-233
>takes 440msec(45.45M/s), K6-2-350@400 takes 258msec(77.5M/s).
> 45.45:85.1 is about 0.53 : 1
>(77.5/400) to (85.1/233) is about 0.53 : 1 too
>
> Though it isn't a certified benchmark, I think it's good enuff to
>say that the fpu power of the K6 to the P2 is like kls claimed. Though
>I don't know where he got his numbers from.
Do you ever see the possibility that your small benchmark might actually
all fit within the PII's cache (and is thus much faster)?
Rgds,
Chris
>
>
>The little lost angel & her featherhead's 2 cents of dreaminess. :)
>Email : Figure out what to remove, I'm getting tired of spam
>
(And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
------------------------------
From: fammacd=!SPAM^[EMAIL PROTECTED] (George Macdonald)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Tue, 06 Jul 1999 07:16:01 GMT
No I'm not talking particularly about MIP models - believe me, there are
many LP models being run in the 4K and less equations. Quite a few use
SLP (Sequential LP) methods to handle non-linearities.
I agree that the PII is probably a better choice for scientific work.
On Mon, 5 Jul 1999 18:49:00 -0400, "Suleyman Karabuk" <[EMAIL PROTECTED]>
wrote:
>If you are talking about MIP models then you are right but I still think
>that even for those problems regular PII will outperform the Celeron.
>Usually you will need high computing power for bigger problems thogh so I
>believe that PII is a better choice for scientific computing.
>
>
>> Nobody?:-) There are many smaller models in the ~4K equation size - many
>> of them quite difficult numerically.
>
>
>
Rgds, George Macdonald
"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??
------------------------------
From: [EMAIL PROTECTED]
Subject: NIC IRQ Problem
Date: Tue, 06 Jul 1999 07:14:16 GMT
Any help appreciated:
I have been trying to setup an ISA network card on my machine. All
available IRQs are taken, and I have to enable COM ports in the BIOS in
order to get my USR PNP modem card to work with the right IRQ under
Win95.
This means that when Linux boots, it assigns IRQ3 to a COM port, which
is a problem as it then can't see the ISA Network card (which can ONLY
work on IRQ3).
So: how do I tell the kernel not to assign IRQ3 to the COM port so I can
use the Network card OK?
Thanks in advance,
Matt
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Chris Robato Yao)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: 6 Jul 1999 05:24:53 GMT
Reply-To: [EMAIL PROTECTED] (Chris Robato Yao)
In <7lqjgc$btc$[EMAIL PROTECTED]>, [EMAIL PROTECTED] (Salem Lee Ganzhorn)
writes:
>Chris Robato Yao ([EMAIL PROTECTED]) wrote:
>: its time, while the CPU is idle. Spreadsheets and compiling won't be
>: taking advantage of duals since most of them aren't designed for it, but
>
>gmake.. you will see huge gains in compiling with multiple processors.
When you're doing multiple compiles at the same time, possibly.
>Compiling of modules is completely independent and parallel processing is the
>way to g.
Unless the compiling itself is multithreaded, you won't see it. As a
matter of fact, compiling may tend to benefit more on a system with
large caches, and the K6-III is about 2x of a Celeron.
Rgds,
Chris
>
>--
>Salem Lee Ganzhorn... [EMAIL PROTECTED]
(And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.hardware) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Hardware Digest
******************************
