Linux-Hardware Digest #777, Volume #12 Mon, 1 May 00 06:13:08 EDT
Contents:
Re: Linux woes (Compaq for one) on the horizon (Tim Hockin)
Re: Kodak 4804 CD RW (Mogens Kjaer)
Re: CD Copy (Mogens Kjaer)
Redhat 6.1 and Parallel Port Zip 100
Re: Linux woes (Compaq for one) on the horizon (Yanglong Zhu)
Re: Linux woes (Compaq for one) on the horizon (Steffen Kluge)
recording sound gives poor quality (Fritz Reichmann)
Re: Netzwerkkarte: MII PHY - this device may not operate correctly (level-one
fnc-0500tx, Winbond 840) (Arjan Drieman)
Re: IDE or SCSI CD-RW? (Anthony White)
Re: Int 13h Device Not Found, BIOS not installed ("Andy France")
Initio SCSI and Ali M1541 problems ("Chris Gifford")
Re: Linux woes (Compaq for one) on the horizon (Adrian)
----------------------------------------------------------------------------
From: Tim Hockin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Linux woes (Compaq for one) on the horizon
Date: 1 May 2000 07:06:34 GMT
In comp.os.linux.misc Yanglong Zhu <[EMAIL PROTECTED]> wrote:
: Make a Win98 bootable disk and boot from the Win98 bootable disk is not an
: issue.
: That's how I repartitioned the hard drive. The key issue is the intensional
: (I assume)
: selectivity for which systems to support and which not.
: I guess computer manufacturers behaviour is legal, but immoral, filthy, and
I have serious doubts that this is, in-fact, what is happening. I
apologize for my "blame the user first" attitude, but experience has shown
me time and time and time again that it is a 99.99% chance that the error
lies between the keyboard and chair (abbreviated as EBKAC). How many times
have people reported to me compiler bugs, CPU bugs, motherboard bugs, libc
bugs, kernel bugs, BIOS bugs. Time and time again it is something they
misunderstood or screwed up.
For a BIOS to figure out what OS was on a boot disk woul dbe pretty tough.
They;d have to look for specific key delimiters - which you can prove by
changing. I'm not sure how you are booting your floppy, but look at
/usr/src/linux/arch/i386/boot/setup.S and consider changing some of the
'signatures' and seeing what difference it makes.
--
Tim Hockin
[EMAIL PROTECTED]
[EMAIL PROTECTED]
This program has been brought to you by the language C and the number F.
------------------------------
From: Mogens Kjaer <[EMAIL PROTECTED]>
Subject: Re: Kodak 4804 CD RW
Date: Mon, 01 May 2000 09:24:06 +0200
Roberto Migliorati wrote:
...
> Unfortunately I can't post the kernel ring buffer, given that I trashed
> the machine (again!!)
...
Without this, I'm out of ideas...
Mogens
--
Mogens Kjaer, Carlsberg Laboratory, Dept. of Chemistry
Gamle Carlsberg Vej 10, DK-2500 Valby, Denmark
Phone: +45 33 27 53 25, Fax: +45 33 27 47 08
Email: [EMAIL PROTECTED] Homepage: http://www.crc.dk
------------------------------
From: Mogens Kjaer <[EMAIL PROTECTED]>
Subject: Re: CD Copy
Date: Mon, 01 May 2000 09:26:06 +0200
Matt Klein wrote:
>
> I am trying to make an exact duplicate copy of a CD (regardless of
> whether it is data or audio) under linux. I think this is possible with
> CDRecord. Does anyone know the proper command line options to use to do
> this?
I doubt this is possible.
Reading audio CD's is non-trivial, use cdparanoia for this.
For data CD's, you could:
dd if=/dev/cdrom of=tmp.iso bs=2048
and then burn the tmp.iso file using cdrecord.
Mogens
--
Mogens Kjaer, Carlsberg Laboratory, Dept. of Chemistry
Gamle Carlsberg Vej 10, DK-2500 Valby, Denmark
Phone: +45 33 27 53 25, Fax: +45 33 27 47 08
Email: [EMAIL PROTECTED] Homepage: http://www.crc.dk
------------------------------
From: <[EMAIL PROTECTED]>
Subject: Redhat 6.1 and Parallel Port Zip 100
Date: Mon, 01 May 2000 07:30:03 GMT
Hello. I'm very new with Linux, i just installed Redhat 6.1. I managed to
get all the hardwares working except for my External Parallel Zip100. I
read some articles from the net on how to setup the drive, but the
language is alien to me. Heheheh. I'm still a beginner here. Anybody can
help me out?
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Yanglong Zhu <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: Linux woes (Compaq for one) on the horizon
Date: Mon, 01 May 2000 02:47:51 -0700
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<tt>Tim Hockin wrote:</tt>
<blockquote TYPE=CITE><tt>In comp.os.linux.misc Yanglong Zhu <[EMAIL PROTECTED]>
wrote:</tt><tt></tt>
<p><tt>: Make a Win98 bootable disk and boot from the Win98 bootable disk
is not an</tt>
<br><tt>: issue.</tt>
<br><tt>: That's how I repartitioned the hard drive. The key issue is the
intensional</tt>
<br><tt>: (I assume)</tt>
<br><tt>: selectivity for which systems to support and which not.</tt><tt></tt>
<p><tt>: I guess computer manufacturers behaviour is legal, but immoral,
filthy, and</tt><tt></tt>
<p><tt>I have serious doubts that this is, in-fact, what is happening.
I</tt>
<br><tt>apologize for my "blame the user first" attitude, but experience
has shown</tt>
<br><tt>me time and time and time again that it is a 99.99% chance that
the error</tt>
<br><tt>lies between the keyboard and chair (abbreviated as EBKAC).
How many times</tt>
<br><tt>have people reported to me compiler bugs, CPU bugs, motherboard
bugs, libc</tt>
<br><tt>bugs, kernel bugs, BIOS bugs. Time and time again it is something
they</tt>
<br><tt>misunderstood or screwed up.</tt><tt></tt>
<p><tt>For a BIOS to figure out what OS was on a boot disk woul dbe pretty
tough.</tt>
<br><tt>They;d have to look for specific key delimiters - which you can
prove by</tt>
<br><tt>changing. I'm not sure how you are booting your floppy, but
look at</tt>
<br><tt>/usr/src/linux/arch/i386/boot/setup.S and consider changing some
of the</tt>
<br><tt>'signatures' and seeing what difference it makes.</tt><tt></tt>
<p><tt>--</tt>
<br><tt>Tim Hockin</tt>
<br><tt>[EMAIL PROTECTED]</tt>
<br><tt>[EMAIL PROTECTED]</tt>
<br><tt>This program has been brought to you by the language C and the
number F.</tt></blockquote>
<tt></tt>
<p><br><tt>Well, you might be the one I have been looking for, although
your EBKAC principle does seem to be hard to swallow.</tt><tt></tt>
<p><tt>I'm trying to give different parameters to circumvent the tricks.
But time has not allowed me to finish that. I hate to troubleshoot alone.
And also I want to warn people early. That's why I posted the message.
I'm not a computer whiz. But I'm not a complete newbie either. I have installed
Linux on my older computer many many times for many different reasons.
The most important and decisive test is to do what you suggested. But to
get a conclusive result and useful interpretation, I need some computer
whiz to cooperate with me to do the test. Are you the one I'm looking for?</tt>
<br><tt></tt> </html>
------------------------------
From: [EMAIL PROTECTED] (Steffen Kluge)
Crossposted-To: comp.os.linux.misc
Subject: Re: Linux woes (Compaq for one) on the horizon
Date: 1 May 2000 08:00:38 GMT
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, Yanglong Zhu <[EMAIL PROTECTED]> wrote:
>What do you want to know?
>
>I can provide this at this point.
>
>Make a Win98 bootable disk and boot from the Win98 bootable disk is not an
>issue.
>That's how I repartitioned the hard drive. The key issue is the intensional
>(I assume)
How do you come to this conclusion? What messages do you get?
Does it say "Sorry, Linux is not supported on this machine", or
something similar? Where did you install LILO? Do you get the
LILO: prompt? Did you get error messages running /sbin/lilo?
Could you create Linux native partitions on the disk? What
distribution are you using (if any), will the installer boot
(from floppy or CDROM)? Does the installer see your hard
disk(s)? Can it read the partition table on the disk?
So many questions, so little information...
Cheers
Steffen.
--
Steffen Kluge <[EMAIL PROTECTED]>
Fujitsu Australia Ltd
Keywords: photography, Mozart, UNIX, Islay Malt, dark skies
--
------------------------------
From: Fritz Reichmann <[EMAIL PROTECTED]>
Subject: recording sound gives poor quality
Date: Mon, 01 May 2000 10:20:17 +0200
I have an old noname sound card which I was able to get running with the
mad16 and mozart drivers. the output from the kernel is attached below.
If I play sounds from prerecorded mp3, realaudio or .au-Files, I get an
excellent sound quality.
The strange thing: If I try to record sound with a command (taken from
the Sound-HOWTO):
dd if=/dev/audio of=bla.au bs=4k count=4
then the sound is extremely noisy and sound as if it was played much too
slow. My voice sounds very deep and slow when I replay it with
cat bla.au > /dev/audio
I read that I need a mixer. I tried several ones, aumix, xmix and kmix.
If I start these, I hear a very clear signal on the loudspeakers. Still,
if I try to record and play the sound I get the poor quality.
I am trying to set this up to be able to telephone with speekfreely.
This programm also delivers the poor sound quality when reading in from
the microphone.
Can anyone point me into the right direction, please ?
Fritz
The mashine is a Celeron 400MHz with 64M of RAM.
Kernel boot message:
Sound initialization started
<MAD16 WSS (82C930)> at 0x534 irq 7 dma 0,1
<Mad16/Mozart> at 0x300 irq 5 dma 0
Sound initialization complete
>From /proc:
> cat /proc/sound
OSS/Free:3.8s2++-971130
Load type: Driver compiled into kernel
Kernel: Linux fritz 2.2.14 #8 Sun Apr 30 16:30:20 MEST 2000 i586
Config options: 0
Installed drivers:
Type 19: MAD16/Mozart (MSS)
Type 20: MAD16/Mozart (MPU)
Type 26: MPU-401 (UART)
Type 2: Sound Blaster
Type 29: Sound Blaster PnP
Type 7: SB MPU-401
Card config:
MAD16/Mozart (MSS) at 0x530 irq 7 drq 0,1
MAD16/Mozart (MPU) at 0x300 irq 5 drq 0
Audio devices:
0: MAD16 WSS (82C930) (DUPLEX)
Synth devices:
Midi devices:
0: Mad16/Mozart
Timers:
0: System clock
Mixers:
0: MAD16 WSS (82C930)
------------------------------
From: [EMAIL PROTECTED] (Arjan Drieman)
Crossposted-To:
comp.os.linux.networking,de.comp.os.unix.linux.hardware,de.comp.os.unix.linux.newusers,de.comp.os.unix.networking,hannover.uni.comp.linux,hanse.linux,maus.computer.linux,muc.lists.linux,z-netz.alt.linux
Subject: Re: Netzwerkkarte: MII PHY - this device may not operate correctly (level-one
fnc-0500tx, Winbond 840)
Date: 1 May 2000 08:34:05 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 30 Apr 2000 22:57:36 +0200, Gerhard Engler
<[EMAIL PROTECTED]> wrote:
>Hallo lieber Leser,
If you crosspost your huge message to this many newsgroups, please make
sure your language is appropriate for all of them.
Arjan
------------------------------
From: Anthony White <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: IDE or SCSI CD-RW?
Date: Mon, 01 May 2000 18:41:07 -0700
"David C." wrote:
> Anthony White <[EMAIL PROTECTED]> writes:
> > Michael O'Reilly wrote:
> >>
> >> This is not a Linux specific question but I do have a dual boot
>
> > SCSI is always better than IDE or EIDE when more than 2 devices are
> > used. This is because SCSI can handle accessing several devices
> > (Disks) at the same time without dragging down the performance of
> > other devices.
>
> An oversimplification.
I think I said in my reply that things are not as simple as I put
them:-)
> You are still limited to the bus speed (10, 20, 40, 80 or 160M bytes/s,
> depending on the kind of controler and drives used). If you have a pair
> of hard drives on a 10M SCSI bus, each one will impact the performance
> of the other - the isn't any way for this not to happen.
It is just like the Net at large, it is no good if you have a 1.5Mb/s
link to the back bone and trying to get 10 users to achieve 400Mb/s
at the same time, they will have to share what is there...
If you use a modern SCSI drive which can easily transfer 2-3Mb/s sustained
then 7 of these drives on a 10 or 20Mb SCSI BUS will "fill it up".
Using 40, 80 or even the latest 160Mb/s SCSI interfaces will allow
more disks to run without saturating the BUS. An other issue to
consider is the limitation of the 33Mb/s PCI BUS.
>
> The big deal is when the bus is fast enough to accomodate all the drives
> at once. With SCSI, they can all take their share of the bandwidth.
> With IDE, two drives can not process commands at once, so extra
> bandwidth only helps if the drive is fast enough to use all that
> bandwidth by itself.
>
> (Does UDMA-66 avoid this problem? I don't think there are any hard
> drives out which are faster than a UDMA-33 bus.)
I have seen a couple of UDMA 66 drives but they are few and far apart.
> > The reason for this is the more advanced protocols used in SCSI which
> This is the difference between port-IO (PIO) and direct memory access
> (DMA). But this has nothing to do with the interface technology. It
> has to do with the interface between the controller circuit and the
> motherboard.
In simple terms the more advanced protocols of SCSI allow for more
efficient usage of the SCSI BUS including transferring data between
devices on the same BUS.
True multitasking operating systems like Linux benefit from using
SCSI over IDE in many cases, especially when many drives are used.
> The big deal about SCSI is that a device, once it gets a command from
> the host adapter, can disconnect from the bus, and then reconnect when
> it is ready to transmit data. While disconnected, another device can
> use the bus. SCSI disconnect works regardless of the I/O interface used
>
That is also correct....
> > IDE interfaces suffer from a need to use more CPU power to drive them.
>
> Not necessarily true. A DMA-based IDE interface (like UDMA-33 and
> UDMA-66) doesn't need to use any more CPU power than a DMA-based SCSI
> host adapter.
I have not checked these latest interfaces so I do not know the
impact on CPU.
> > In short, if you have 1 or 2 IDE (EIDE) devices (1 on the primary
> > controller and 1 on the secondary controller) you will get as good or
>
> Right, but for the wrong reasons.
To explain SCSI versus IDE in detail would take a large chunk out
of the available disk space on the news servers....
> > And yes it is more costly but not as bad as it used to be.
>
> But still more costly than it should be. The main reason IDE is cheap
> is that it's bundled with every motherboard. It costs less to make a
> chip if you're making millions of them instead of only thousands.
That is true, it also has to do with competition. Look at Adaptec
who are the leading supplier of SCSI controllers and chip sets for those
controllers. In the last year or two a lot of other companies have
started making SCSI controllers based on other than Adaptec chip sets.
This has forced prices down on Adaptec products. You can get controllers
that are as good as those made by Adaptec at 50% or less cost.
An other thing that has helped the competition compete with Adaptec was the
reluctance of Adaptec to release specs for their chip sets so proper
drivers could be made for Linux.
Since so many servers run Linux these days there has been a surge in cheaper
SCSI controller solutions - I guess that Adaptec decided that the Linux
business was to good to let go....
> > An other thing to consider is the more consistent interface to SCSI.
> Note that this isn't unique to SCSI. USB and firewire also have
> consistent interfaces and generic device support. The ATAPI spec for
> IDE also accomplishes much of this (although not all IDE devices are
> ATAPI compliant.)
I have used SCSI under almost every conceivable OS and found it
easier to get going in more configurations than using IDE. In
some cases IDE was not even supported.
Anthony
------------------------------
From: "Andy France" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi
Subject: Re: Int 13h Device Not Found, BIOS not installed
Date: Mon, 1 May 2000 20:48:09 +1200
<[EMAIL PROTECTED]> wrote in message news:8eivks$4a$[EMAIL PROTECTED]...
> I just installed a Adaptec SCSI adapter into my computer which is
> an HP Vectra VL8, Pentium II 450, 128MB SDRAM. Anyway, I keep getting
> this message when I boot up the computer, but it seems the scsi adapter
> is able to detect a tape drive and cdrom writer that I have on the
> internal scsi chain, this is the 50pins connector.
>
> Here is what I see
>
> AHA-2940 BIOS v1.2
> (c) 1995 Adaptec
>
> Int 13h Device Not Found
> BIOS not installed
>
>
> Please can someone tell me how to get rid of this message and make it
> normal. It is plain annoying. Thanx
This message is only telling you that there is no bootable device on SCSI ID
0 or 1, so the card is not loading the overlays required to boot off SCSI
hard drives... it may be annoying but certainly isn't an error! Int 13h is
usually provided by the IDE controller so the card would intercept it if
required.
If you press CTRL-A when prompted, you should be able to find an option to
permanently disable booting on this card - I don't have an Adaptec handy so
can't tell you exactly what to look for.
Regards,
Andy.
------------------------------
From: "Chris Gifford" <[EMAIL PROTECTED]>
Subject: Initio SCSI and Ali M1541 problems
Date: Mon, 1 May 2000 11:06:00 +0100
Hi,
I have two problems with my SuSE 6.3 installation. My Gigabyte GA-5AX mother
board has the Ali Alladin chipset M1541 and apparently there is no DMA
support unless I patch the kernel. I have seen the files on the 'net
somewhere before but can't locate them now. Anybody know where I can find
them?
Also, I have a Domex 3194U/UW SCSI card with the Initio IN-91xx chipset
which IS supported by SuSE, but when I try to "modprobe initio" my system
hangs. Anyone have any tips please?
Cheers,
[EMAIL PROTECTED]
------------------------------
From: Adrian <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Linux woes (Compaq for one) on the horizon
Date: Mon, 01 May 2000 03:09:20 -0700
Yanglong Zhu wrote:
>
> Make a Win98 bootable disk and boot from the Win98 bootable disk is not an issue.
> That's how I repartitioned the hard drive. The key issue is the intensional (I
>assume)
> selectivity for which systems to support and which not. I guess computer
>manufacturers
> behaviour is legal, but immoral, filthy, and greedy. They want make computing as
> complicated as possible so that more people will ask for support, therefore buy their
> support. They lure people into their trap by selling their computers a little bit
> cheaper. What a reputation. We need to expose these companies. Relentlessly.
When you say it will not boot "free OS" systems, do you mean it won't
boot a RedHat installation floppy/CDrom or just that it won't boot LILO?
There are several of these new BIOSes floating around (to date I've only
heard of them it laptops) that are brain damaged and won't boot LILO.
This doesn't appear to be a case of malficense on the part of computer
manafacturer, more like ignorance on the part of BIOS programers in
basic assumptions about the future use of the BIOS. Fortunately, other
"LILO" like boot loaders like Chos do work.
Adrian
--
- I just tried this on my old Packard Bell 486/66 w/4MB (Hey ...
- shut-up! I was young, ignorant, and didn't know anything about
- hardware or quality manufacturers.).
------------------------------
** 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
******************************
