Linux-Hardware Digest #529, Volume #13 Tue, 5 Sep 00 01:13:09 EDT
Contents:
Re: Duron, LILO, RedHat 6.2 probs? ("Richard Cornelius")
Re: Kernel Option for PIII Processor? ([EMAIL PROTECTED])
Re: APC brand UPS with apcupsd --> PANIC! Cannot talk to UPS (John-Paul Stewart)
Re: cd writer? (Lyndon)
!HLP getting a Promise ata controller to work with Kernel 2.2.14-15! ("SYTE|X")
Re: i810 Anyone succeded ? Please Help !!! ("Filip Atanassov")
Re: SMP Performance? (Karl Heyes)
Re: SMP Performance? (Karl Heyes)
Re: SMP Performance? (Karl Heyes)
Re: SMP Performance? (Karl Heyes)
Re: SMP Performance? (Karl Heyes)
Re: SMP Performance? (Karl Heyes)
Whats a good Motherboard for Linux?
Re: SupraMAX Modem Problems with Linux Mandrake (Flotsam)
betting farm on Dell Poweredge 1300 with OEM RAID/Linux? (NoWhereMan)
Upgraded motherboard and memory stopped working
Re: !HLP getting a Promise ata controller to work with Kernel 2.2.14-15! (James
Richard Tyrer)
----------------------------------------------------------------------------
Reply-To: "Richard Cornelius" <[EMAIL PROTECTED]>
From: "Richard Cornelius" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Duron, LILO, RedHat 6.2 probs?
Date: Mon, 4 Sep 2000 20:03:31 -0600
I have been putting together a couple AMD Athlon "Thunderbird" (slot A)
systems with TekRam SCSI controllers. I got the same "CPUID Serial number"
kernel panic when I tried to boot the SCSI drive with the 2.2.14 kernel
installed with RedHat 6.2. The CPUID serial number is an Intel Pentium III
"feature" that the kernel tries to disable in the AMD Duron cpu, obviously
unsuccessfully.
I got around both the CPUID-induced kernel panic and my inability to boot
from the Tekram SCSI controller by installing an old slow 8 GB EIDE hard
drive and building a new kernel as follows: Load the RedHat (6.2?)
distribution onto the EIDE drive. When it tells you to remove the floppy so
you can reboot from the hard drive, just leave the boot floppy in and reset
the system. Boot the system in "rescue" mode, so it uses the Linux kernel
from the install floppy, but uses the root file system from the hard drive.
Rebuild the Linux kernel in /usr/src/linux with the drivers that you select
for your system with "make config" (or "make menuconfig"...). The resulting
kernel can then be mapped for booting by LILO. The LILO message you
mentioned is in the LILO documentation -- the second stage boot loader
failed. Some experimenting with "/etc/lilo.conf" and your partition layout
can fix this problem. Since the system dual-boots with Windows, there is a
good chance that your Linux boot partition is not within the 1024 cylinder
limit. Again, the LILO documentation (even the mini-HOWTO!) is going to be
a lot more helpful that I can be here.
Good Luck!
"Stephen Bridges" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I put together a computer with a socket A motherboard and a Duron 700MHz
> CPU
> I installed Red Hat. And that did not work. If I turn my computer on
> now the lilo prompt is for some reason getting stuck on LI. The boot disk
> works, but only has a linux record in it.
>
> This sounds like the lilo probs with large disks from other posts on here
> and i can probably deal; with it eventually. But I also have another
> problem.
>
> The new linux partition doesn't boot. After loading the kernel in, it
boots
> and finds the Duron processor . . . .
>
> CPU: AMD AMD Duron(tm) Processor stepping 00
> Enabling extended fast FPU save and restore...done.
> Disabling CPUID Serial number...general protection fault: 0000
>
> Then lots of numbers and Stack numbers are printed in hex which i can copy
> if they help. This all finishes with a kernel panic and the whole thing
> hangs
> Widget
>
------------------------------
Subject: Re: Kernel Option for PIII Processor?
From: [EMAIL PROTECTED]
Date: 04 Sep 2000 22:09:25 -0400
"Sipke de Wal" <[EMAIL PROTECTED]> writes:
> If memory serves me right PII & PIII do support the virtual x86 box
> the generic Pentium Pro does not.
No.
The other feature is, of course, SIMD streaming extensions; MMX++.
--
Eric McCoy <[EMAIL PROTECTED]>
------------------------------
From: John-Paul Stewart <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: APC brand UPS with apcupsd --> PANIC! Cannot talk to UPS
Date: Tue, 05 Sep 2000 02:17:24 GMT
[EMAIL PROTECTED] wrote:
>
> Does anyone have apcupsd (apcupsd-3.5.8-2.i386.rpm) working?
>
> Whenever I try to start I get the error "PANIC! Cannot talk to UPS".
>
> Thanks,
> Richard
First off, different models from APC use different signalling methods over the
serial ports. There's 'dumb' mode which toggles serial port control lines, and
'smart' mode which actually sends messages over the serial port data lines.
(There might be a third signalling method, too, plus USB as well.) Are your UPS
and software both configured for the same method?
Second, are you using a cable from APC? I've read elsewhere on this newsgroup
(6-8 weeks ago, I think) that the APCs require their own cable, *not* a standard
serial cable. Moreover, when I checked out the APC website last month, they
were offering different cables depending on whether you were using Linux or
Windows.
HTH,
J-P Stewart
p.s. -- my other e-mail address is [EMAIL PROTECTED] ;)
------------------------------
From: Lyndon <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.questions,comp.os.linux.setup
Subject: Re: cd writer?
Date: Mon, 04 Sep 2000 19:26:49 -0700
Go to /etc/fstab
Go to line that deals with your cdrw
change 'hdc' to 'scd0'
Hope this helps.
Have a nice day.
Bob R wrote:
>
> I just loaded Mandrake 7.0 from CD. Everything seemed okay, but now when i
> log on as user or root
> I cannot access the CD. The CD is a HP 8250i CD Writer. Mandrake sees it
> as an IDE/HDC
> I tried changing the properties of the device, but that did not help. If I
> try to view the files through KDE Explorer
> it says it cannot open them. It is mounted, although it says it is a locked
> directory.
>
> Any assistance will be appreciated. I would like to install some more
> software that I have, but I need to access
> the CD.
>
> My system is an K6-2/350mhz
> 96 megs of ram
> 2 HD
> Windows on first HD
> Linux on 2nd.
> HP-CD Burner (only CD on system)(IDE type on second IDE channel as Master)
------------------------------
From: "SYTE|X" <[EMAIL PROTECTED]>
Subject: !HLP getting a Promise ata controller to work with Kernel 2.2.14-15!
Date: Tue, 5 Sep 2000 05:08:11 +0200
Hi,
I'am really stuck here with my promise Ultra-66 PCI.
I downloaded the driver from www.promise.com, and used it happy with a
older linux kernel. Now I had to upgrade to 2.2.14-15.0 after a crash.
But now I can't get that card to work anymore, please help!
When I insmod -f ultra66.o , it runs almost.. but crashes. I have tryed to
compile the old sourcecode, but without luck.
I'am using RedHat 6.2.
The promise card has bios 1.14 (build 0728)
Desparate for help (took all night) and regards,
Leroy
Reply to [EMAIL PROTECTED]
also
------------------------------
From: "Filip Atanassov" <[EMAIL PROTECTED]>
Subject: Re: i810 Anyone succeded ? Please Help !!!
Date: Tue, 05 Sep 2000 03:18:47 GMT
I followed the instructions from
http://support.intel.com/support/graphics/intel815/ and everything is
working fine for me.
Regards,
Filip
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:24:15 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:24:15 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:24:28 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:24:28 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:21:08 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: Karl Heyes <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.hardware,hk.comp.os.linux,comp.os.linux.misc,comp.os.linux.networking
Subject: Re: SMP Performance?
Date: Tue, 05 Sep 2000 04:21:08 +0000
In article <[EMAIL PROTECTED]>,
Beggar <[EMAIL PROTECTED]> wrote:
> Hi all,
>
> I want to ask what's the performance of SMP on Linux compare to Win NT/2000?
> My application is mostly Java native thread that will spawn around 3-400
> threads.
>
> And I think that most of the CPU time in single CPU is waste in context
> switching. Will SMP benefit in this case?
>
> How much should I expect will increase when using SMP in Linux?? Is it worth
> for SMP or should I setup two mahines for load balancing??
>
What you are asking is very vague, there are many issues wrt SMP. I'm not a
guru on this, but some of the points that have been raised are.
linux 2.2 is ok at the kernel level to 4 processors. linux 2.4 ok to 8
processors. This means nothing if your application doesn't use it effectively.
WRT java, big issue, thread creation API is easy, many people think the one
thread by client concept is ok, but it doesn't scale. You didn't state whether
thats 400 running threads, either way under whatever OS, you will be hitting
cache hard, and VM tricks (ie context switching) does have a penalty.
I/O versus computation work. 8 processors doing image rendering sounds nice.
linux does try to keep tasks on the same CPU, more cache hits that way, but if
you ask 2 processors to schedule 300 jobs and each processor has 64k cache
then whats the average share of the cache. Memory speed is slow to a CPU.
karl.
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Whats a good Motherboard for Linux?
Date: Tue, 05 Sep 2000 03:29:05 GMT
I am going to build my first Linux machine. I have a Intel Celoron 333,
and a Intel Pentium 233 lying around. What would be a good motherboard to
use with Red Hat 6.2, and uses IDE.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: [EMAIL PROTECTED] (Flotsam)
Crossposted-To:
ahn.tech.linux,alt.linux.sucks,alt.os.linux,alt.os.linux.madrake,alt.os.linux.mandrake,at.linux,es.comp.os.linux.misc,han.comp.os.linux.misc
Subject: Re: SupraMAX Modem Problems with Linux Mandrake
Reply-To: [EMAIL PROTECTED]
Date: Tue, 05 Sep 2000 04:20:36 GMT
On Mon, 4 Sep 2000 14:48:51 -0700, mohan <[EMAIL PROTECTED]> wrote:
>I'm having a problem detecting my SupraMax Internal PCI Modem. To my
>knowledge it is not a Winmodem so I dont know why it is giving me a problem.
>When I try to query it, it says it can't detect it.
>If anyone knows what could be wrong could you please post a followup or
>email me. It would be extremely helpful.
Helpful? Now there's a thought. You can go to http://linmodems.org
and at the bottom is Rob Clark's "gromit" page linked...
http://www.o2.net/~gromitkc/winmodem.html or something like that.
On the latter page you'll find a link to Rick's Rants. Interesting
reading on quite a few topics. Somewhere in there is a Rant about
modems, and a list of (I think) eight models of internal PCI modem
which turn out NOT to be winmodems. Sorry about all the quibbling. 8-))
F.
------------------------------
From: NoWhereMan <[EMAIL PROTECTED]>
Subject: betting farm on Dell Poweredge 1300 with OEM RAID/Linux?
Date: Tue, 05 Sep 2000 00:18:17 -0400
I *think* the RAID controller that Dell ships with
the PowerEdge 1300 will be an OEM version
of the AMI MegaRAID.
v
Will the open source megaraid.c drivers
listed at www.linuxhq.com work with the Dell
OEM version of this card, or do I have to
use the kernel specific binary version
that *I think* Dell ships with the RH 6.2
pre-install. What version of the driver is
Dell shipping? I've seen 0.92 -> 1.05
mentioned elsewhere.
v
I usually use Caldera e-Server on Dell
Poweredge single processor 2200s.
The 1300 will have 2 processors active.
v
If good source drivers exist for the OEM
RAID on the Poweredge 1300, would I be
better off with the pre-installed RH 6.2 from
Dell, or with adding a Mylex AcelleRaid 150
to a no-raid 1300 myself?
v
None of my SCSI/Linux servers have
crashed in years (actually *not ever*) ,
and I can't afford any instability surprises
from my first attempt at RAID. I'll want
to put together RAID 5 with 3 drives
and a 4th drive as a hot spare.
v
I don't have a good route to ami.com
right now.
---end---
Nothing is ever tested until it is under "real"
load. Nothing ever works, until I see it work.
All of my stuff works *so far*, but none of
us are ever more than a few keystrokes
away from retirement. You're only as
good as your last gig. I've never had
more than a news reader and a good
search engine for "support."
====================================================================
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Upgraded motherboard and memory stopped working
Date: 4 Sep 2000 23:42:49 -0500
Hello.
I bought a PII-333MhZ system about 2 years ago. It came with an Intel 440BX
motherboard and 64 MB of memory. I recently added another 64 MB DIMM module.
Both were, at least according to their labels, 64 Mb 100 MhZ SDRAM parts.
They worked fine with that old motherboard.
Then today, I bought Motherboard (it says 993AS on it) with a PIII-750 MhZ processor
and moved all my stuff from the "old" motherboard to this one. However, when I booted
up
my RH 6.2 installation I was getting intermittent OOPSes and what
resembled kernel panics and mention of a problem trying to access virtual
memory. Playing around (rather blindly- the documentation is terrible) with
the CMOS settings I found a setting "Memory Hole" which had two choices:
None or 15-16 MB. Desprate to figure out why my kernel kept
panicing (even when I loaded straight from the rescue disk), I set
this option to "15-16MB". Then things worked fine, except that the system was
obscenely slow (it acted like my old 486/66). It turned out that that
CMOS setting limited me to 15 MB system memory, but at least it didn't panic/crash
on me. On a hunch, I took old the older of the two DIMM modules, unset the memory hole
option,
and now the system is working fine (with only 64 meg). If I have the other memory chip
in, then no matter how much I play with the timings, I still get a panic unless I have
that "15-16 MB hole" option set.
Now for my question:
Is there any known problem with running Linux on PC100 DIMMs at 750 MhZ? I'm thinking
that the 'old' memory (which had originally come with the system) may have actually
a 67 MhZ part instead of a 100 MhZ part. However, setting the system to HOSTCLK-33M
didn't fix the problem. If it had, I'd just replace the memory, except that it worked
fine
up to this point and didn't work at the lowered speed (did I actually lower the speed
with that setting)? I may try running only on the 'old' memory.
Thoughts? Oh what I'd give for some good documentation. :-)
Thanks,
Roger
====== Posted via Newsfeeds.Com, Uncensored Usenet News ======
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
======= Over 80,000 Newsgroups = 16 Different Servers! ======
------------------------------
From: James Richard Tyrer <[EMAIL PROTECTED]>
Subject: Re: !HLP getting a Promise ata controller to work with Kernel 2.2.14-15!
Date: Tue, 05 Sep 2000 05:07:39 GMT
This is a multi-part message in MIME format.
==============56DF6DBDBB89DCAFF1C5C633
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
SYTE|X wrote:
> Hi,
>
> I'am really stuck here with my promise Ultra-66 PCI.
> I downloaded the driver from www.promise.com, and used it happy with a
> older linux kernel. Now I had to upgrade to 2.2.14-15.0 after a crash.
> But now I can't get that card to work anymore, please help!
> When I insmod -f ultra66.o , it runs almost.. but crashes. I have tryed to
> compile the old sourcecode, but without luck.
>
> I'am using RedHat 6.2.
>
> The promise card has bios 1.14 (build 0728)
>
> Desparate for help (took all night) and regards,
> Leroy
>
> Reply to [EMAIL PROTECTED]
> also
You could resort to using the 2.2.16 Kernel and the ide patch.
This is what I am using with my Promise Ultra 66 PCI card.
Don't forget the "linear" global parameter in lilo.conf.
JRT
==============56DF6DBDBB89DCAFF1C5C633
Content-Type: text/x-vcard; charset=us-ascii;
name="tyrerj.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for James Richard Tyrer
Content-Disposition: attachment;
filename="tyrerj.vcf"
begin:vcard
n:Tyrer;James Richard
tel;fax:call first
tel;home:(520) 648-3720
x-mozilla-html:FALSE
org:JRT Engineering
adr:;;1468 North Rio Sonora;Green Valley;Arizona;85614-4007;USA
version:2.1
email;internet:[EMAIL PROTECTED]
x-mozilla-cpt:;2
fn:James Richard Tyrer
end:vcard
==============56DF6DBDBB89DCAFF1C5C633==
------------------------------
** 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
******************************
