Linux-Hardware Digest #397, Volume #12 Fri, 3 Mar 00 18:13:07 EST
Contents:
Re: 4 Celeron motherboard? (David C.)
Re: Machine hangs 'lost interrupt' ("Daniel Avila")
Re: 3-button serial mouse (David C.)
do you have info on the HP 710C Deskjet Printer? ("Jacob Andresen")
Re: heating linux (Stefan Seyfried)
Re: unknown disk drive activity (Stefan Seyfried)
Re: CD-ROM DOS Formatted Mount on Linux (Stefan Seyfried)
Re: Advice on PartitionMagic on all-Linux system? (David C.)
Re: Modem/ppp config question (Clifford Kite)
Re: heating linux (David C.)
Re: ASUS P2B-D or TYAN S1832DL Tiger 100? (Alex Lam)
Re: CD-R/RW & Creative DVD together. (Alex Lam)
Re: Linux Distribution for Siemens RM600 (Alex Lam)
Binary compatibility ("Barry Kiernan")
Re: Where To Get Raid Info ([EMAIL PROTECTED])
Re: Where To Get Raid Info ([EMAIL PROTECTED])
Re: heating linux ("Robert W. Cunningham")
Re: Binary compatibility (David C.)
Re: VIA Apollo KX133 (Superscreamer)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (David C.)
Subject: Re: 4 Celeron motherboard?
Date: 03 Mar 2000 16:07:26 -0500
Lee Sau Dan writes:
>>
>> Personally, I would prefer to have 16x150Mhz Pentiums instead of one
>> 700Mhz Athlon, because a multiprocessor system should be a lot more
>> stable if done right. One processor hangs and corrupts its kernel,
>> 15 are still alive, and one of them restarts the dead one ...
>
> This is too ideal. The actual senario could be as bad as: One
> processor hangs and corrupts its kernel. This makes it become crazy,
> so crazy that it starts intervening the other processors, corrupting
> the in-memory code being executed by other processors. The other
> processors become crazy, too, and drives the remaining processor
> crazy.
But these kinds of problems don't happen from abberant user code. Only
broken kernel code can cause this kind of problem - and there's no way
to protect a system against broken kernel code if the hardware is SMP.
(Yes, you could go microkernel, but even then, broken microkernel code
can cause the same problems.)
I know it's possible for large scale SMP to work reliably. Sun systems
do this all the time. For instance, their Enterprise 10000 server
supports up to 64 UltraSPARC processors (at 250, 336, or 400MHz). I've
never heard of one of these having stability problems like you describe.
Of course, this says nothing about how Linux would work on a PC with as
many processors. The SPARC architecture is quite different from the x86
architecture, and Linux's architecture is quite different from
Solaris's.
My point here is only that such parallelism is definitely possible, and
exists in currently-shipping commercial products.
-- David
------------------------------
From: "Daniel Avila" <[EMAIL PROTECTED]>
Subject: Re: Machine hangs 'lost interrupt'
Date: Fri, 3 Mar 2000 13:17:12 -0800
I am having the same problem on 2 systems running redhat 6.0 with an ABIT
BE6 MB on both the IDE and IDE/66 controller on kernel 2.2.12 and 2.2.5
Dan
"Eric Parker" <[EMAIL PROTECTED]> wrote in message
news:6L5v4.93$[EMAIL PROTECTED]...
> I have recently installed RedHat 6.1. I find that very shortly after
logging
> in the machine hangs when trying to access disk and a meesage 'lost
> interrupt' is printed on the screen.
>
> Is this indicative of a hardware error ?
>
> Is it some sort of configuration problem ?
>
> Any advice would be appreciated ?
>
> TIA
>
> eric
>
>
>
>
====== 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: [EMAIL PROTECTED] (David C.)
Crossposted-To: uk.comp.os.linux
Subject: Re: 3-button serial mouse
Date: 03 Mar 2000 16:22:40 -0500
[EMAIL PROTECTED] (Alex Butcher) writes:
>
> You didn't read my post properly. I said at _power-up_. Some mice are
> dual mode (usually Microsoft and MouseSystems) and default to 2-button
> Microsoft mode UNLESS a mouse button is held during power-up. Once in
> MouseSystems mode, they behave as proper, genuine, 3 button mice. A
> subset of these mice can be flipped into 3 button MouseSystems mode
> from 2 button Microsoft mode _after_ power up by toggling some of the
> modem control lines.
And then there are the (rarer) ones like my old Genius GM-6 serial mouse
which powered up in 3-button MouseSystems mode and toggled to 2-button
Microsoft mode if a button was held down on power-up.
-- David
------------------------------
From: "Jacob Andresen" <[EMAIL PROTECTED]>
Subject: do you have info on the HP 710C Deskjet Printer?
Date: Fri, 3 Mar 2000 22:26:20 +0100
I will try to do a driver for the HP 710C Deskjet printer
(must include rasterization emulation and all..)
If you have any information to share I would sure appreciate it (ie
datasheets and techical information)
Thanx in advance,
Jacob
PS. feel free to email me directly with any info you find relevant
------------------------------
From: Stefan Seyfried <[EMAIL PROTECTED]>
Subject: Re: heating linux
Date: 3 Mar 2000 11:28:29 +0100
Robert W. Cunningham <[EMAIL PROTECTED]> wrote:
> My original post, and the calculations it contains, are entirely appropriate to
> the task, and are EXTREMELY accurate, requiring very sensitive laboratory-grade
EXTREMELY is relative ;-))
> equipment to measure the difference between "practical" and "theoretical" values.
>
> Don't contradict me. Try it yourself. I have. You are talking without data,
> which is synonymous with "ignorance". Stop being ignorant: Make some
> measurements and do the math!
I did. With a power meter, my desktop PC draws around 55 Watts. It
draws an input current of 0.9 Amps at 230 Volts.
230 Volts x 0.9 Amps = 207 Watts.
It does make a difference, doesn�t it?
Stefan
--
Stefan Seyfried, [EMAIL PROTECTED], [EMAIL PROTECTED]
hi, i'm a signature-virus! copy me to your .signature to help me spread!
------------------------------
From: Stefan Seyfried <[EMAIL PROTECTED]>
Subject: Re: unknown disk drive activity
Date: 3 Mar 2000 11:23:40 +0100
Cl�ment Nodens HERMANN <[EMAIL PROTECTED]> wrote:
> Unfortunately, I don't see where the problem could be. But I have a
> question : how many swap partitions do you own ? I thought it where
> impossible to use a swap partition bigger than 128 Mo.
>
> [EMAIL PROTECTED] wrote:
>>
>> On Tue, 29 Feb 2000 20:36:31 +0100, Cl�ment "Nodens" HERMANN
><[EMAIL PROTECTED]> wrote:
>> >
>> >[EMAIL PROTECTED] wrote:
>> >>
>> >> My disk drive (actually scsi activity) light flashes every two seconds
>> >> constantly and the disk seems to make actuating noises about every ten.
>> >>
>> >> This is happeneing even when the system is idle. I tried removing any gnome
It is the bdflush daemon, nowadays it�s called kflushd, i think.
Believe me, and you can�t do very much about it.
You could read �man update� though and set some other timing
intervals, but trade in data security.
HTH, Stefan
--
Stefan Seyfried, [EMAIL PROTECTED], [EMAIL PROTECTED]
hi, i'm a signature-virus! copy me to your .signature to help me spread!
------------------------------
From: Stefan Seyfried <[EMAIL PROTECTED]>
Subject: Re: CD-ROM DOS Formatted Mount on Linux
Date: 3 Mar 2000 11:53:50 +0100
Dances With Crows <[EMAIL PROTECTED]> wrote:
> On Wed, 01 Mar 2000 21:24:40 -0800, Meir Levi
> <<[EMAIL PROTECTED]>> shouted forth into the ether:
>>I am downloading Linux files (Adobe Reader, etc.) with my Windows 95
>>system onto CD-RW burner, which is FAT32 formatted.
> ^^^^^^^^^^^^^^^^^
> Just how do you format a burner, exactly? And if the CDs are using the
> FAT32 filesystem, there's a Big Problem... CDs are supposed to use the
> iso9660 or UDF filesystems!
Well, i burn ext2fs on my backup-cdroms, because it�s easier to keep a
1:1 copy of my harddrive with all the permissions etc.
Read in the CD-Writing-HOWTO about this.
Stefan
--
Stefan Seyfried, [EMAIL PROTECTED], [EMAIL PROTECTED]
hi, i'm a signature-virus! copy me to your .signature to help me spread!
------------------------------
From: [EMAIL PROTECTED] (David C.)
Crossposted-To: comp.os.linux.misc
Subject: Re: Advice on PartitionMagic on all-Linux system?
Date: 03 Mar 2000 16:51:37 -0500
Michael Meissner <[EMAIL PROTECTED]> writes:
>
> You don't need PM to create the initial system, but it is extremely
> helpful if you want to resize disk partitions without doing a
> dump/load. My PM has more than paid for itself for that one feature.
Can the current version (5.0) resize an ext2fs partition?
The last time I looked, PM could only move ext2fs partitions, but
couldn't resize them. If this feature has been added, I'm quite
interested.
-- David
------------------------------
From: Clifford Kite <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: Modem/ppp config question
Date: Fri, 3 Mar 2000 15:12:02 -0600
Todd Siegel <[EMAIL PROTECTED]> wrote:
> pppd[712]: Connect: ppp0 <--> /dev/ttyS3
> pppd[712]: LCP: timeout sending Config-Requests
> pppd[712]: Connection terminated.
> pppd[712]: Connect time 0.6 minutes.
> pppd[712]: Receive serial link is not 8-bit clean:
> pppd[712]: Problem: all had bit 7 set to 0
Very likely the chat script failed to do what it needed to do to get
the ISP to start PPP on it's end. The reasons for this happening are
many and varied but one common thing is a chat script that ends with
CONNECT '' instead of CONNECT '\d\c' .
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* Make a PPP log-file: Add " daemon.*;local2.* /var/log/ppp-log "
(may need Tab separators) to /etc/syslog.conf, create the file by
" touch /var/log/ppp-log ", and do " kill -HUP `pidof syslogd` ". */
------------------------------
From: [EMAIL PROTECTED] (David C.)
Subject: Re: heating linux
Date: 03 Mar 2000 17:07:05 -0500
Stefan Seyfried <[EMAIL PROTECTED]> writes:
>
> I did. With a power meter, my desktop PC draws around 55 Watts. It
> draws an input current of 0.9 Amps at 230 Volts.
>
> 230 Volts x 0.9 Amps = 207 Watts.
>
> It does make a difference, doesn�t it?
Not quite.
230 VAC x 0.9A = 207 VA
The result is only in watts if your voltage is DC.
Since we're talking AC current, we need to convert from VA to W. We
can't know the exact factor for doing this without knowing the details
of the device. If you use the standard sqrt(2) factor (assuming sine
waves everywhere and no inductance or capacitance), you end up with:
207 VA / sqrt(2) = 146 W
But this won't really be accurate either, because all power supplies
have coils and capacitors in them.
What you do know is that the power consumption will be no more than
207W. I don't think the power in VA is ever greater than the power in
W.
-- David
------------------------------
From: Alex Lam <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: ASUS P2B-D or TYAN S1832DL Tiger 100?
Date: Fri, 03 Mar 2000 15:28:25 -0800
Francisco de Borja Rodriguez wrote:
>
> Hi.....
> I'm looking for information on the current status
> (supported/unsupported) for two mainboards on linux:
>
> ASUS P2B-D
> and
> TYAN S1832DL Tiger 100
>
> Are there any known problems with these boards?
> (configuration would be something like
> 2x500PIII, Ati Rage Fury, 256Mb RAM)
> What mainboar is better, ASUS P2B-D or
> TYAN S1832DL Tiger 100?
>
> Thanks in advance......
I'm running two Asus SMP Dual P-2/3
Both are running U2WSCSI. 512MB RAM.
One running Linux, one running FreeBSD.
Love them.
Alex Lam.
------------------------------
From: Alex Lam <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.misc
Subject: Re: CD-R/RW & Creative DVD together.
Date: Fri, 03 Mar 2000 15:30:23 -0800
Richard Gaywood wrote:
>
> > How can I get Linux to identify and list both CD devices together?
> > I have a Philips CD-R/RW (secondary EIDE controller as slave), and a
> > Creative DVD Dxr2 (secondary EIDE controller as primary.)
>
> Both drives should have entries in /dev. The CDR should be /dev/hdd and the
> DVD should be /dev/hdc. If you go into your /dev directory, I imagine there
> will be symlink call /dev/cdrom pointing to one of these.
>
> Just create a new symlink for whichever device doesn't work. Assuming this
> is the DVD drive:
>
> To create a symlink, use (as root, in the /dev directory)
> ln -s hdc dvd
>
> You will then have a device called /dev/dvd which is an alias for your dvd
> drive. If you then edit /etc/fstab (again as root) and change the reference
> to /dev/cdrom to /dev/dvd, that should make your dvd work. Flip this around
> if it's the CD-R that doesn't work.
>
> My RH6.1 system has nifty CDROM automounting; I'm not sure if this will work
> on the second drive automatically. I suspect not, but my X is broken at the
> moment, so I can't check. Any takers?
>
Thx. I'll try it this weekend.
Alex Lam.
> --
>
> -=R=-
>
> "Is this what you thought married life would
> be like, Homer?"
> "Yeah, pretty much. 'Cept we drove around in
> a van solving mysteries."
------------------------------
From: Alex Lam <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Linux Distribution for Siemens RM600
Date: Fri, 03 Mar 2000 15:35:47 -0800
AXK wrote:
>
> Hello,
>
> i search an Linux Distribution for SIEMENS RM600,
> or what Distribution can i use?
>
> Regards
>
> Axel
Try SuSE. .
It's a distro from Germany. So you should have better luck with
Siemens...
Alex Lam.
------------------------------
From: "Barry Kiernan" <[EMAIL PROTECTED]>
Subject: Binary compatibility
Date: Fri, 3 Mar 2000 22:22:15 -0000
Hope somebody can teach me something here.
In order to appreciate the value of COM (component object model), I need to
understand why it's hard to get compiled code to talk to other code. It
would help if somebody could explain
1. Linux and Windows both run on Intel x86 processors. What stops code
compiled on one platform from not running on the other (ie why does the OS
make a difference)?
2. Even on the same platform (e.g win95) the object code generated from one
source (C say) doesn't work with object code from another source (pascal for
instance). Why not?
3. When you run your program, what is the extent of the interaction with the
OS? (eg the OS stops user programs from accessing memory allocated to other
processes. Does it have to vet each single machine code instruction in order
to do this?)
If it helps in pitching your explanation, I'm familiar with assembly
language (eg calling subroutines by pushing parameters/local vars onto the
stack) though I don't know what the assembly language equivalent of dynamic
memory allocation is.
PS saying the OS performs this task doesn't help; after all, the OS boils
down to some assembly program, no?
Thanks for any help.
Barry Kiernan
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Where To Get Raid Info
Date: Fri, 03 Mar 2000 22:17:39 GMT
raid 0 and 1 are only there for redundance and are not very helpful,
you need raid 5...how many drives do you have? you must have 3 for
raid5 otherwise it is just not worth it...the howto was enough for
me...but then again, i haven't setup a raid system, i was going to
mirror (raid 1) because i'm tired of a hard drive crashing and me
losing everything and wasting a lot of time getting back where i
started, but i decided instead to get a tape drive because its just
more economical (20gb tape drive was only 260 bucks, and its scsi)
hope this helps
Mike
In article <[EMAIL PROTECTED]>,
mike <[EMAIL PROTECTED]> wrote:
> Hi,
> I want to setup raid for the purposes spanning a number of
> small old hard drives to make one large drive to install Linux on.
> I have looked at the RAID-HOWTO and it seems that there is
> not enough info there to be able to set it up. I think that
> the spanning requires raid 0 or 1. Is there any other doc somewhere.
>
> Thanks
> Mike
>
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Where To Get Raid Info
Date: Fri, 03 Mar 2000 22:17:30 GMT
raid 0 and 1 are only there for redundance and are not very helpful,
you need raid 5...how many drives do you have? you must have 3 for
raid5 otherwise it is just not worth it...the howto was enough for
me...but then again, i haven't setup a raid system, i was going to
mirror (raid 1) because i'm tired of a hard drive crashing and me
losing everything and wasting a lot of time getting back where i
started, but i decided instead to get a tape drive because its just
more economical (20gb tape drive was only 260 bucks, and its scsi)
hope this helps
Mike
In article <[EMAIL PROTECTED]>,
mike <[EMAIL PROTECTED]> wrote:
> Hi,
> I want to setup raid for the purposes spanning a number of
> small old hard drives to make one large drive to install Linux on.
> I have looked at the RAID-HOWTO and it seems that there is
> not enough info there to be able to set it up. I think that
> the spanning requires raid 0 or 1. Is there any other doc somewhere.
>
> Thanks
> Mike
>
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Robert W. Cunningham" <[EMAIL PROTECTED]>
Subject: Re: heating linux
Date: Fri, 03 Mar 2000 22:34:32 GMT
Stefan Seyfried wrote:
> Robert W. Cunningham <[EMAIL PROTECTED]> wrote:
> > My original post, and the calculations it contains, are entirely appropriate to
> > the task, and are EXTREMELY accurate, requiring very sensitive laboratory-grade
>
> EXTREMELY is relative ;-))
>
> > equipment to measure the difference between "practical" and "theoretical" values.
> >
> > Don't contradict me. Try it yourself. I have. You are talking without data,
> > which is synonymous with "ignorance". Stop being ignorant: Make some
> > measurements and do the math!
>
> I did. With a power meter, my desktop PC draws around 55 Watts. It
> draws an input current of 0.9 Amps at 230 Volts.
>
> 230 Volts x 0.9 Amps = 207 Watts.
>
> It does make a difference, doesn�t it?
How did you measure each of those values? Is your equipment working correctly? Are
you using the equipment in accordance with the manufacturer's instructions?
There must be a problem with your measurements, your technique, or your equipment,
since there is no way you could have ANY realistic power-factor value that would
result in nearly a 4:1 power ratio!
If possible, make your measurements with ALL the equipment connected at once, thereby
ensuring all measurements are taking place in the same environment. Some equipment
can introduce its own errors.
Also, ensure you have selected a scale that permits the reading to be taken in the
upper portion of the range, where you will have the greatest number of significant
digits. A single-digit reading is worse than useless.
As a first guess, I'd say your current reading was way off. Second guess would be the
power meter, even though the value it gives is at the low end of reasonable for a
desktop PC. Remember that all values must be RMS, which if your meter doesn't give
that to you directly, will be .707 times the peak-to-peak voltage of the sine wave.
-BobC
------------------------------
From: [EMAIL PROTECTED] (David C.)
Subject: Re: Binary compatibility
Date: 03 Mar 2000 17:44:36 -0500
"Barry Kiernan" <[EMAIL PROTECTED]> writes:
>
> In order to appreciate the value of COM (component object model), I need to
> understand why it's hard to get compiled code to talk to other code. It
> would help if somebody could explain
This is the wrong newsgroup for your questions. This group is meant for
people asking questions about how to get their hardware to work under
Linux.
Nevertheless, here's an answer for you.
> 1. Linux and Windows both run on Intel x86 processors. What stops code
> compiled on one platform from not running on the other (ie why does
> the OS make a difference)?
Back in the good ol' days of DOS and CP/M, operating systems didn't
amount to much. They'd manage files on disks and do little else.
The result was that programs had free access to the hardware. They
could, and did, access any piece of memory and any I/O device they
wanted to.
But this practice doesn't work well under a multitasking environment.
Your program could crash the system if it accesses a device or a piece
of memory that some other program is trying to access at the same time.
So people put more advanced resource-management facilities into
operating systems. Under this scenario (which almost all operating
systems use), applications can no longer directly access the hardware.
The OS accesses the hardware on behalf of the applications. The
applications, when they need the OS to do this access, make requests to
the OS. These requests are made in the form of system calls.
This is why apps aren't immediately compatible. Every OS has a
different set of system calls. A program on Windows that wants to
request 1M of memory from the OS will make one kind of system call. A
program on UNIX or OS/2 that wants to make that same request will make a
different kind of system call.
Furthermore, the mechanism used to make system calls themselves differs
from OS to OS, and between hardware platforms that run the "same" OS.
> 2. Even on the same platform (e.g win95) the object code generated
> from one source (C say) doesn't work with object code from another
> source (pascal for instance). Why not?
While the instruction to add two 32-bit memory locations together may be
identical, the system calls used are completely different. Without
system calls, a program can't do much of anything. Some of the commonly
used calls include:
- request and release memory
- read/write files on the disk
- read input from the keyboard and mouse
- write output to the screen, sound card and printer
- read/write the network
etc.
> 3. When you run your program, what is the extent of the interaction
> with the OS? (eg the OS stops user programs from accessing memory
> allocated to other processes. Does it have to vet each single machine
> code instruction in order to do this?)
Modern CPUs have hardware for this. Memory is allocatd in pages
(typically 4K chunks on x86 processors). Each page of memory has
various protection bits set, which the OS can modify. If an app tries
to read or write a page that it doesn't have access to, the OS is
alerted by the hardware (usually via an interrupt.) The OS can then
take action - which usually means that the app is shut down, although
other actions could also be taken.
Similarly, I/O ports and some assembly instructions can also be
protected from applications.
The processor typically runs in two or more protection modes. (The x86
processor supports four, but few operating systems use all four.) The
main modes are kernel mode and user mode.
When the processor is in kernel mode, it can do anything. It can touch
any memory, access any I/O port, and execute any assembly instruction.
When the processor is in user mode, various faults (interrupts) will be
triggered if the program accesses memory, I/O ports or instructions that
are protected.
Hardware interrupts (say, from timers, or faults, or I/O devices) throw
the processor into kernel mode. This way, the kernel's interrupt
handler can take action. This typically involves a call into a device
driver. When it is done handlingt the interrupt, it resets itself back
to user mode before returning control to the application.
Apps are protected from each other with this mechanism. When the kernel
is entered (either via an interrupt or a system call), the kernel will
decide which app should get the CPU next. If it decides to switch from
the active app to another, it stores the first app's state away
somewhere, updates the tables of which memory pages are protected, and
loads the second app's state - then it switches to user mode and
"returns" to the second app - which will run until a system call or
interrupt causes a return to kernel mode.
For more information, check out any good book on operating system
design.
-- David
------------------------------
From: Superscreamer <[EMAIL PROTECTED]>
Subject: Re: VIA Apollo KX133
Date: Fri, 03 Mar 2000 22:50:45 GMT
I found this article:
http://www.linuxtests.org/articles/motherboards/epox/ep-7kxa/index.htm
In article <[EMAIL PROTECTED]>,
Stefan Brabec <[EMAIL PROTECTED]> wrote:
>
> Hi,
>
> does anybody have experience in running Linux
> on a VIA Apollo KX133 mainboard (e.g. Tyan Trinity K7) ?
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
** 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
******************************
