Linux-Hardware Digest #860, Volume #9 Sun, 28 Mar 99 16:13:47 EST
Contents:
Digital AlphaStation 200 4/233, 16MB RAM (Christine Salib)
Re: Commercial "rm -rf" recovery services for EXT2? (Jon Haugsand)
Commercial "rm -rf" recovery services for EXT2? ([EMAIL PROTECTED])
Re: SuSE 6.0 and SB 128 PCI (Jerry Normandin)
help with gcc and aout ("Michael Ruel Loehr")
Re: Linux, SCSI, RAID0 performance (Eoin)
Re: Graphics Card STB Velocity 4400 Purchase ("Mark DuPrey")
Re: Hercules Beast AGP driver under Redhat 5.2 (Jerry Normandin)
Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0 (Johan Kullstam)
Re: Linux, SCSI, RAID0 performance (Tomasz Korycki)
problem with awe 64 ("Leray Jean")
Re: Modem Woes (M. Buchenrieder)
Re: Help me spend $2,000 on a new Linux-based computer ("Cameron Spitzer")
Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0 (Shankar Unni)
Logitech First Mouse (tm) 3 Button Mouse ("Michael J. Glanovsky")
Re: Netgear Network starter kit support (Gary Momarison)
Adaptec 1510 setup in RH 5.2 (Jes Ramsing)
Intel's 28595TX chip ("Lior Langer")
Re: How to compile Kernel 2.2.2 with redhat 5.2??? (Eugene VonNiederhausern)
----------------------------------------------------------------------------
From: Christine Salib <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Digital AlphaStation 200 4/233, 16MB RAM
Date: Sun, 28 Mar 1999 13:57:12 -0500
Digital AlphaStation 200 4/233, 16MB RAM
(Linux, UNIX, NT, VMS)
http://cgi.ebay.com/aw-cgi/eBayISAPI.dll?ViewItem&item=81983765
OVERVIEW
233MHz CPU Alpha RISC (64-bit)
512KB cache
16MB RAM
1.44MB floppy disk drive
16-bit stereo quality in/out
10BT/10B2 Network interface card
Several expansion ports
PC Card Reader-Type III PMCIA slot
Desktop enclosure
Power cord and English documentation included
Compatible with Linux, Windows NT, UNIX, and VMS (not
included)
ONLY THREE DAYS LEFT FOR BIDS. BIDS START AT $350.
------------------------------
From: Jon Haugsand <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Commercial "rm -rf" recovery services for EXT2?
Date: 28 Mar 1999 21:37:39 +0200
* [EMAIL PROTECTED]
> One of our administrators just pulled a complete delete on an important ext2
> partition. The partition contains a small (8-10) set of biggish (>~330MB)
> gzipped tarballs.
I once did someting similar. I created a file system on a backup disk
containing all my home directories of a linux system when I was
upgrading my RH system. All important files were gzipped tarballs.
What I did was to write a special purpose rescue program. The program
searched for blocks that looked like (1) index blocks in the file
system, and (2) start of a gzipped tar file.
I'll gladly let you have my rescue program, but I am not sure it will
help you just to run it. In order to get my files back, I made
commands in the program, changed the behaviour of some commands, and
recompiled my program over and over. Finally, I had rescued the most
important files. There is naturally no documentation in the program,
and no documentation on the use of the program. To use it you need to
be a C programmer and have some knowledge of the EXT2 file system.
Perhaps you are better out than me. I do not know if the inode of a
deleted file is zeroed, or just marked as unused. In the former case,
you'll have all the files by just read the inodes and search the disk
for the files. (When creating a file system, all the inodes were
zeroed.)
> Does anyone on this list know of someone out there who might be able to do
> this for us, or could you provide a pointer to either a net resource or more
> appropriate forum for me to post this request?
I would think comp.os.linux.misc is a better group. Followup-to there.
--
Jon Haugsand
Norwegian Computing Center, <http://www.nr.no/engelsk/>
<mailto:[EMAIL PROTECTED]> Pho: +47 22852608 / +47 22852500,
Fax: +47 22697660, Pb 114 Blindern, N-0314 OSLO, Norway
------------------------------
From: [EMAIL PROTECTED]
Subject: Commercial "rm -rf" recovery services for EXT2?
Date: Sun, 28 Mar 1999 19:10:11 GMT
I wasn't sure if this was the appropriate group. Along with my other
troubles (described below) our news server is down, so I'm fumbling around at
Deja News, and have already tried to send this once. My apologies if this is
a duplicate.
One of our administrators just pulled a complete delete on an important ext2
partition. The partition contains a small (8-10) set of biggish (>~330MB)
gzipped tarballs.
We are desperate to recover these files. We played around with mc on an
expendable disk, and were able to rather easily recover small files that had
been deleted. However, when we tried the same techniques with the disk we
need to recover, we came up short. We removed this disk from service
immediately the error occurred, to try to minimize the chances of anything
else being written anywhere upon it.
Does anyone on this list know of someone out there who might be able to do
this for us, or could you provide a pointer to either a net resource or more
appropriate forum for me to post this request?
Thanks in advance.
B.
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Jerry Normandin <[EMAIL PROTECTED]>
Subject: Re: SuSE 6.0 and SB 128 PCI
Date: Sun, 28 Mar 1999 14:41:38 -0500
The PMAC Linux people have been wrint a lot of code latley for USB.
Check out the 2.1.XXX kernels and the 2.2.X kernel trees!
Also check out the FAQ regarding upgrading from 2.0.X! My
PMAC 9600 is up to 2.1.136 and there were some libraries to bump up.
This PPRO is going to be bumped up soon since I added an 8GB UDMA drive.
I was impressed with the speed! I'm a SCSI guy.. but 32MB /Sec isn't
bad.
Now I need to bump up to support over 2GB partitions. I'm using right
now
by seting up four 2GB partions and striping them. This would be optimal
if
UDMA supported silmultaneous reads from multiple targets and if I had
multiple
8GB drives. My SCSI drives are old SCSI-2. My controller is a 2940UW
though...
just the Ultra Wide SCSI-3 drives are big bucks. When I saw a Maxtor
8.4GB drive
selling for $130.00 I couldn't resist! It's the only EIDE drive on my
box so I figured
I would only increas performance.
USB and firewire support is available! Also I am going with Zoom Air
cards
on my Linux network! They are supported!
Phil wrote:
>
> Hi,
>
> I know that I have to read the howto's and docs, but beeing a newbie I
> was not aible to get my SB PCI 128 to work under linux .
>
> I'm running kernel 2.0.36 (SuSE 6.0) and KDE desktop.
> Is there anybody who has the time and knowelegde to explain it rather
> simple pse ...
>
> Does anybody know if USB - HP720C and HP4100C (USB) will be supported
> in the near future ? At the time beeing, it's still hard to let W98
> for what it is ......
>
> Thanks in advance !
> Phil
------------------------------
From: "Michael Ruel Loehr" <[EMAIL PROTECTED]>
Subject: help with gcc and aout
Date: Sun, 28 Mar 1999 14:47:30 -0600
, I just installed redhat 5.2. I want to develop c++ programs on it. I
tried to make a hello world program but I run into a problem. It compiles
fine
g++ helloworld.cc
If i look in my directory i see the files
helloworld.cc a.out
but if i type a.out I get a bash error command not found or something like
that
If i go into the debugger, the program runs fine?
One of my friends mentioned that I have to type ./ or something, but I
couldn't get that to work either?
(if you know of a solution e mail me [EMAIL PROTECTED] )
Someone help, please
Ruel Loehr
------------------------------
From: (Eoin)
Subject: Re: Linux, SCSI, RAID0 performance
Date: Sun, 28 Mar 1999 19:47:56 GMT
Yes, 68-pins means 16-bits. IBM does claim that the UltraStar
9ES is available in 50-pin (narrow), 68-pin (wide), and 80-pin SCA2
(also wide, designed to plug directly into a backplane), but I've only
actually seen 68-pin and SCA.
--Eoin
On Sun, 28 Mar 1999 13:22:23 +0000, Thomas Dorris <[EMAIL PROTECTED]>
wrote:
>Eoin wrote:
>
>> Wait a minute, you've got the 50-pin-connector drives?
>
>Nope. 68-pin connectors. Does a 68-pin connector imply a 16-bit word
>size even if the drives are listed only as Ultra2 drives (not U2W)? I
>was under the impression that Ultra2 meant an 8-bit word size at 40mhz
>while Ultra2Wide meant 16-bit. Could be wrong, though, as this is my
>first venture into SCSI-land since 1985 with my Apple ][gs (which barely
>counts as SCSI, but I guess technically it was).
>
>ThomasD
Addressable as Bullwinkle in the
commercial domain of mostlysunny
------------------------------
From: "Mark DuPrey" <[EMAIL PROTECTED]>
Subject: Re: Graphics Card STB Velocity 4400 Purchase
Date: Sun, 28 Mar 1999 14:08:43 -0500
To use an STB Velocity 4400 you have to make sure you have a version of X
3.3.3 or later. Use the SVGA Xserver and select the Nvidia chipset. It works
on my machine and I've read of others in the newsgroups who have had success
like this.
Good luck
Mark DuPrey
<[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello,
>
> I am going to by a new graphics card. I am looking at 16 MB 128 bit
> Riva TNT cards to work with both Linux and Win98. I have an IBM Aptiva
> 2138 with PCI slots. I have been considering the STB Velocity 4400 card
> or the Diamond Viper 550. I have read messages where folks have got the
> Viper card to work with Linux. Has any one had any experience with the
> STB Velocity 4400 card?? Also, any suggestions of cards that are
> possibly a little better, but not to pricey.
>
> Thanks for any input,
> Ken
>
------------------------------
From: Jerry Normandin <[EMAIL PROTECTED]>
Subject: Re: Hercules Beast AGP driver under Redhat 5.2
Date: Sun, 28 Mar 1999 14:32:59 -0500
Check out the specs on Xfree86 3.3.3! It supports MANY AGP cards.
Also check out kernel 2.1.136, it supports fbconsole.
Ken wrote:
>
> I really need help!
> My computer is using a new AGP display card: Hercules Beast Terminator with
> the S3 Savage3D chip. 250MHz DAC.
> Where can i get the driver for Redhat 5.2? I cannot find it in
> www.hercules.com
> Thank you.
>
> Ken
------------------------------
Crossposted-To:
comp.os.linux.misc,linux.redhat.misc,alt.linux,alt.os.linux,comp.os.linux.development.system
Subject: Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: 28 Mar 1999 14:47:07 -0500
Tomasz Korycki <[EMAIL PROTECTED]> writes:
> Johan Kullstam wrote:
> >
> > d s f o x @ c o g s c i . u c s d . e d u (David Fox) writes:
> >
> > > "Idea Man" <[EMAIL PROTECTED]> writes:
> > >
> > > > Does anyone else think this would be a good idea? Keep the i386 tree, and
> > > > add an i686 tree that is optimized for P-II/Celeron/P-III processors.
> > > >
> > > > This might be a pain in the butt for the mirrors (more hard drive space
> > > > used), but for some mirrors this would be just fine. This would also make
> > > > Linux higher performing for all the people with flashy new Pentium-III
> > > > machines...
> > >
> > > How much performance improvement would there be?
> >
> > based on my experience with egcs over the past year, not much.
> >
> > the pentium classic seems to be hypersensitive to scheduling, but the
> > i686 (i have a pentiumpro) seems have roughly the same performance
> > (using time on a few of my programs) for compiles with -march=i386,
> > i486 or pentiumpro. -march=pentium hurt speed by about 10%.
> >
> Actually, if You have FP-heavy code (graphics, raytracing, FEA, CFD,
> Stats, simulation), P6-optimised code can gain You as much as 15-20%,
> compared to i386.
here are some results from a number cruncher (specifically a turbo
code simulation). it uses floating point arithmetic - mostly multiply-
accumulates. while i used C++, i did not use anything out of the
stdc++ library.
using -S and -fverbose-asm, one of my files has this header
.file "wsiso.c"
.version "01.01"
# GNU C++ version egcs-2.91.66 19990314 (egcs-1.1.2 release) (i686-linux) compil
ed by GNU C version egcs-2.91.66 19990314 (egcs-1.1.2 release).
# options passed: -march=pentiumpro -march=XXXX -O3 -Wall
# -fstrict-aliasing -fno-rtti -fomit-frame-pointer -fno-exceptions
# -fverbose-asm
# options enabled: -fdefer-pop -fomit-frame-pointer -fcse-follow-jumps
# -fcse-skip-blocks -fexpensive-optimizations -fthread-jumps
# -fstrength-reduce -fpeephole -fforce-mem -ffunction-cse
# -finline-functions -finline -fkeep-static-consts -fcaller-saves
# -fpcc-struct-return -fgcse -frerun-cse-after-loop -frerun-loop-opt
# -fschedule-insns2 -fcommon -fverbose-asm -fgnu-linker -fregmove
# -foptimize-register-move -fargument-alias -fstrict-aliasing -m80387
# -mhard-float -mno-soft-float -mieee-fp -mfp-ret-in-387
# -mschedule-prologue -mcpu=XXXX -march=XXXX
XXXX is one of i386, i486, pentium and pentiumpro. (i hacked the gcc
spec file to give -march=pentiumpro by default, but it does get
overridden by an explicit -march as far as i can tell.)
here are the outputs from bash's time on a lightly loaded pentiumpro
system.
-march=i386
real 0m54.673s
user 0m53.500s
sys 0m0.050s
-march=i486
real 0m53.713s
user 0m52.910s
sys 0m0.060s
-march=pentium
real 0m52.957s
user 0m52.300s
sys 0m0.050s
-march=pentiumpro
real 0m54.563s
user 0m53.810s
sys 0m0.120s
i can't see *any* significant difference. looks like it's all in the
noise here. do i have problems with the spec default
-march=pentiumpro option? can anyone else post benchmarks?
--
J o h a n K u l l s t a m
[[EMAIL PROTECTED]]
Don't Fear the Penguin!
------------------------------
From: Tomasz Korycki <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Linux, SCSI, RAID0 performance
Date: Sun, 28 Mar 1999 14:34:57 -0500
Thomas Dorris wrote:
>
> Leif Thuresson wrote:
>
> > thing as "Narrow Ultra2". Ultra2 is always 16-bit even though
> > they sometimes drop the "W"
>
> Ah. That explains one of my questions, then. Thanks!
>
> Thomas Dorris
If You _really_ want to see spped-up due to striping, put each pv on
separate SCSI bus. The effects can actually be quite dramatic. Remember
that there is always an emphasis on _either_ R or W performance in each
RAID level. That's why, when choosing Your level You have to decide:
1. performance vs. data integrity, then
2. read or write performance, then
3. cost vs. benefit
Each step enables You to consider the next one fully, without knowing
Your answer to 1 _fully_, You cannot even start thinking about 2 in
intelligent fashion...
------------------------------
From: "Leray Jean" <[EMAIL PROTECTED]>
Subject: problem with awe 64
Date: Sun, 28 Mar 1999 21:54:39 +0200
Hi i've just put linux red hat on my computer but when i tried to set up my
awe 64 my computer said this
"
LD settid verify failed,this may not be a problem
Try adding (verify N) to the top of your script
Error occured executing request 'LD 7' on or around line 632
Further action aborded
OK
"
Thank u in advance
Leray Jean
[EMAIL PROTECTED]
Nantes,France
------------------------------
From: [EMAIL PROTECTED] (M. Buchenrieder)
Subject: Re: Modem Woes
Date: Sun, 28 Mar 1999 14:37:26 GMT
Clay Lawrence <[EMAIL PROTECTED]> writes:
>Help...Linux won't see either of my modems :-(
>I've got a Rockwell HCF 56K Data/Fax RTAD PCI Modem amd a U.S.
^^^^
Not supported at all. Hopeless.
>Robotics 56K Fax Int. Neither of course are winmodems. Both are
>installed in a new build with a Soyo D61BA mainboard and dual
>PII-400's. The USR is an ISA. They are on Com1 and Com2 respectively
>in windows at 0x2F8, IRQ3 and 0x3F8,IRQ4.
[...]
The USR is a PNP model, and you'll have to use the isapnptools
to get it working in Linux.
Michael
--
Michael Buchenrieder * [EMAIL PROTECTED] * http://www.muc.de/~mibu
Lumber Cartel Unit #456 (TINLC) & Official Netscum
Note: If you want me to send you email, don't munge your address.
------------------------------
From: "Cameron Spitzer" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.networking
Subject: Re: Help me spend $2,000 on a new Linux-based computer
Date: 28 Mar 1999 17:20:07 GMT
In article <StjL2.8002$[EMAIL PROTECTED]>,
Eugene <[EMAIL PROTECTED]> wrote:
>Here is my dream machine:
>
>Celeron 366 overclocked to 550
>128 Mb SDRAM (PC-100) (one piece)
This machine will be unreliable at best. "Overclocking" is for machines
that contain no important data and do no important work,
and where you don't mind installing from scratch all the time.
There is an "Overclocking FAQ" floating around the 'net, with my name
in it. It's a FORGERY. I said approximately the opposite of what
the "FAQ" says, and the FAQ's author reversed the sense of what I said
because he didn't agree with me about the issue.
Then I told him he couldn't use my name, and he ignored that.
I think you can assume the whole thing is a crock, made up by some
crooked PC builder who wanted to blur the issue so he could sell
processors that were slower than he claimed.
In other words, anyone who tries to sell you on "overclocking" is
probably a scammer.
There have been times when semiconductor manufacturers mark parts
as slower than they can be guaranteed to be, to fill orders for the slower
parts, or to protect the price premium on the faster ones.
It's the exception, not the rule. Rumors of a vast conspiracy to
overprice processors and memory by concealing their true reliable
speed are hogwash.
Cameron
------------------------------
From: Shankar Unni <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.misc,linux.redhat.misc,alt.linux,alt.os.linux,comp.os.linux.development.system
Subject: Re: Idea: Make a seperate "i686" tree for Redhat Linux 6.0
Date: Sun, 28 Mar 1999 12:22:51 -0800
In comp.os.linux.misc Enkidu <[EMAIL PROTECTED]> wrote:
> > Redhat do not develop anything, they "just" package it.
They certainly develop a lot more than you do, you berk..
And then Chris Mauritz chimed in:
> It would be nice if they would "package" a source-based distribution
> similar to what you get with FreeBSD's source.
Tell you what: why don't you actually go look into a RedHat (or any
such) distribution (which has sources packages for everything), unpack
the kernel tree and rebuild it (a simple task, by the way) before
spouting off mindless drivel like this?
Now, about the "separate 686 tree" idea:
1. a separate 686 tree is stupid unless you have *LOTS* of *source*
that's separate between a 686 and a 386. Most of the *necessary*
differences can be accommodated with simple ifdefs.
2. you are free to build your source tree with the compiler of your
choice, giving it whatever code-generation flags you deem appropriate.
If you find a combination of compiler and flags that gives you an
appreciable difference in kernel performance on anything newer than a
386, let us know.
3. the PII/III (and K6 and M2) processors are very forgiving about
integer code scheduling, and you'll see very little difference between
386-scheduled code and 686-scheduled code on these processors. The only
processors on which fancy optimizations made any difference at all were
the Pentium and the original PPro processors (to a slight extent).
4. Kernels, and 99% of the applications out there, don't have much FP
code (not enough to make a difference - it's not like you're running
Livermore loops or fpppp). So doing fancy FP scheduling (even if GCC
were very good at it) would not get you anything.
> Then, if I wanted 686 optimization, I could replace gcc with egcs,
> use the appropriate cflags, type "make world", go to bed, and wake
> up to a system where every single binary had been recompiled from
> scratch on the running system.
You can do a make clean and rebuild and install everything every night
if you choose. With a minor effort, you can switch the entire build
around each night to a different compiler or flags, and see if it makes
a blazing difference..
--
Shankar Unni [EMAIL PROTECTED]
------------------------------
From: "Michael J. Glanovsky" <[EMAIL PROTECTED]>
Subject: Logitech First Mouse (tm) 3 Button Mouse
Date: Sun, 28 Mar 1999 20:04:07 GMT
In mouseconfig for this mouse...
logical choices could be...
Generic 3 Button Mouse (serial)
Logitech Mouse (serial, old C7 type)
Logitech MouseMan / First Mouse (serial)
Any choice more appropriate than another?
Thanks for any suggestions,
Mike
[EMAIL PROTECTED]
------------------------------
From: Gary Momarison <[EMAIL PROTECTED]>
Subject: Re: Netgear Network starter kit support
Date: 28 Mar 1999 09:22:56 -0800
"Wayne" <[EMAIL PROTECTED]> writes:
> I am considering purchasing the Netgear Network Starter kit SB104. I want
> to use this for a duel Winblows 98 and Redhat 5.2 configuration. Does
> anyone know if the Netgear cards are supported in Linux? The kit contains 2
> dual-speed network cards (fa310tx) and a 4 port hub (en104tp).
> Thanks for the help.
The FA310TX with the DEC Tulip chip is supported. I think I heard there
are some with a different chip; I can't say whether it is supported.
I've never used a hub, but AFAIK, they're all supported, since they
don't really need support.
--
Look for Linux info at http://www.dejanews.com/home_ps.shtml and in
Gary's Encyclopedia at http://www.aa.net/~swear/pedia/index.html
------------------------------
From: Jes Ramsing <[EMAIL PROTECTED]>
Subject: Adaptec 1510 setup in RH 5.2
Date: Sun, 28 Mar 1999 22:07:09 +0200
I am sorry to admit it but i am stuck. How do i get the d... thing to do
as i please?
I have tried to get lilo.conf to start i up, but nothing happens and i
would like not to recompile the kernel since some programs are running
on the machine.
Got some good ideas anyone :-)
Jes Ramsing
------------------------------
From: "Lior Langer" <[EMAIL PROTECTED]>
Subject: Intel's 28595TX chip
Date: Sun, 28 Mar 1999 21:38:10 +0200
Hello
I am trying to use Olicom OC2220 pcmcia network card.
I was informed by the vompany that it uses Intel's 28595TX chip.
Does anybody knows what other pcmcia network cards uses
the same chip?
I could not find that information t Intel web site
Thanks
--
Lior Langer
work: [EMAIL PROTECTED]
24h: [EMAIL PROTECTED]
mac: [EMAIL PROTECTED]
cell: 972-53-644237
------------------------------
From: Eugene VonNiederhausern <[EMAIL PROTECTED]>
Crossposted-To: aus.computers.linux,comp.os.linux.networking,comp.os.linux.setup
Subject: Re: How to compile Kernel 2.2.2 with redhat 5.2???
Date: Sun, 28 Mar 1999 15:01:42 -0600
Aaron Saikovski wrote:
> Please Help!!
> I have followed the how-to's to the letter, I have used the Linus
> pocketbook.
> I cannot get the 2.2.2 kernel to work correctly under RedHat 5.2.
>
> I compiled the kernel, removed static links to linux, I have made the
> modules, made the modules_install.
> I have moved the zImage to /boot.
> When I reboot and try to boot the new kernel, it causes a total system
> reboot..
> In the /boot directory there is still initrd etc files...
>
> Is there an updated version of the kernel how-to that focuses primarily on
> redhat5.2 and the latest kernel build?
> I have followed the how-to to the letter and it doesn't workj with redhat
> 5.2
> Please help!!
>
> Thanks!
>
> Aaron
> email: [EMAIL PROTECTED]
There are some redhat updates for kernel 2..2.x (init script, and some
others). I had troubles getting kernel 2.2.0 to work until I made those
updates.
------------------------------
** 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
******************************
