Linux-Hardware Digest #742, Volume #14 Tue, 8 May 01 06:13:08 EDT
Contents:
I want to write a simple device driver module: where do I start? (Bruce Allen)
Re: Sound Blaster 16 PCI (Dances With Crows)
Re: kppp problems... (Dances With Crows)
Re: VIA 686B bug fix (Dances With Crows)
Re: VIA 686B bug fix (MindPatrol)
USB printer under 2.2.19 (Rich Carreiro)
Re: BP6 + Mandrake + reboot problem (Juergen Pfann)
Re: Backing up Windows on Linux (Michael Meissner)
Re: Backing up Windows on Linux (hac)
Re: HDD problem on an ASUS a7v133 (Michael Meissner)
Re: VIA 686B bug fix ("Anthony")
Re: VIA 686B bug fix (Nils Holland)
HW recommendations for >4 GB RAM (Cyrille Artho)
Re: Backing up Windows on Linux (John Thompson)
DLT tape drives (Florian Lorenzn)
Re: VIA 686B bug fix (M. Buchenrieder)
----------------------------------------------------------------------------
From: Bruce Allen <[EMAIL PROTECTED]>
Subject: I want to write a simple device driver module: where do I start?
Date: Mon, 07 May 2001 21:19:03 -0500
I was hoping to get a pointer on writing a simple device driver module.
I want to monitor
and control the status of two fans on an intel D815EEA2 motherboard.
These are logged
and controlled by an SMSC chip
http://www.smsc.com/main/catalog/lpc47m13x.html
The complete chip manual is:
http://www.smsc.com/main/datasheets/lpc47m13x.pdf
These manuals describe the exact locations and function of the fan
control registers.
By reading these, one can find the fan rotation speed. By writing them,
one can control
the fan speed.
I have written thousands of lines of C, but do not know how to write
even a simple kernel
module. I need to know:
(a) how to read data from these registers
(b) how to write data to these registers
(c) how to create a /proc file system entry to report the register
contents and perhaps to control
them.
If someone would be willing to help me get started, or to point me to
the right thing to read, it would
help. Please don't just tell me to read the lm_sensors code. I have
looked at it but don't know if
this is applicable (eg, is the LMC chip on the I2C bus) or where to
start.
Cheers,
Bruce Allen
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Sound Blaster 16 PCI
Reply-To: [EMAIL PROTECTED]
Date: 08 May 2001 02:33:56 GMT
On 07 May 2001 17:41:30 GMT, Marc Rosen staggered into the Black Sun and said:
>recently I bought a SB 16 PCI sound card.
>anyone have any success in getting this to work?
>I am renning redhat 6.1 compiled w/sound enabled
>
>postings indicate that I should be using the es1370.o module, but it
>doesn't seem to work
>
>/proc/pci shows:
> Bus 0, device 9, function 0:
> Multimedia audio controller: Ensoniq Unknown device (rev 2).
> Vendor id=1274. Device id=5880.
cd /usr/src/linux/drivers/sound && grep -r 5880 *
[[es1371.c returns 11 matches, es1370.c reutrns none]]
You want to use the es1371 module, and you definitely want to use the
newest kernel you can find. I've heard conflicting reports about the
5880 cards, but they should work with the es1371 module and a kernel
newer than about 2.2.17 . Use http://groups.google.com/ and search this
NG for keywords "5880" and "sound" for more information, HTH....
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: kppp problems...
Reply-To: [EMAIL PROTECTED]
Date: 08 May 2001 02:35:42 GMT
On Mon, 07 May 2001 13:54:23 GMT, Jonadab the Unsightly One staggered
into the Black Sun and said:
>"Karim R. Sobhi" <[EMAIL PROTECTED]> wrote:
>
>> Apr 24 21:52:38 Edgar pppd[927]: Connect: ppp0 <--> /dev/ttyS1
>> Apr 24 21:52:46 Edgar pppd[927]: Peer is not authorized to use remote
>> address 217.52.7.110
>
>That line right there is the important one.
>
>Unfortunately, I'm not certain whether "peer" in this case is you or
>your ISP. Is your connection governed by DHCP? Are you trying to use
>217.52.7.110 as your IP address? Or is your ISP using that address
>maybe and your system isn't accepting that for some reason? One of
>those two things is probably your problem.
The OP should check his /etc/ppp/options file and make sure that the
line which says "noauth" is not commented out. If you don't have
"noauth" somewhere in that file, then pppd will require the peer to
authenticate itself. This is not what you want if you are just dialing
in to your ISP.
(I could swear I posted a reply to this guy's message several days ago.
Bloody news swerver.)
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: VIA 686B bug fix
Reply-To: [EMAIL PROTECTED]
Date: 08 May 2001 02:33:55 GMT
On Mon, 7 May 2001 17:41:23 +0200, jay staggered into the Black Sun and
said:
>I have a MSI K7T Turbo motherboard with the VIA 686B Southbridge chip
>(KT-133A chipset). Is there a bug fix for Linux? If so how do I correct
>the problem. I'm currently using SuSE 7.0 but want to fix the problem
>before upgrading to 7.1.
? Which bug were you referring to? The 2.4.x series had problems with
IDE DMA on the VIA 686 causing data corruption in isolated instances.
They said this bug was fixed in 2.4.3, and they say it is *really* fixed
in 2.4.4. It was never present in the 2.2 series, so "Stay with 2.2 for
a couple of weeks and wait for bug reports to trickle in from the 2.4.4
users before upgrading," would be my advice.
This bug is entirely within the kernel; you can fix any problems you are
having with IDE DMA at any time without doing a full 7.0->7.1 upgrade.
>The Windows bug fix does not help Linux at all
Well, yes. Nailing horseshoes to your Porsche's tires will not help it
run down the bridle path.... :-]
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (MindPatrol)
Subject: Re: VIA 686B bug fix
Date: Tue, 08 May 2001 03:29:50 GMT
Reply-To: [EMAIL PROTECTED]
On Mon, 7 May 2001 17:41:23 +0200, jay <[EMAIL PROTECTED]>
wrote:
>I have a MSI K7T Turbo motherboard with the VIA 686B Southbridge chip
>(KT-133A chipset). Is there a bug fix for Linux? If so how do I correct the
>problem. I'm currently using SuSE 7.0 but want to fix the problem before
>upgrading to 7.1.
>
>
>The Windows bug fix does not help Linux at all, I have already tried it.
>Windows works without any problems anymore but Linux is still messed up.
>
>Regards,
>
>Jay
>
Fixed in the latest kernel ( 2.4.4)
--
John
------------------------------
From: Rich Carreiro <[EMAIL PROTECTED]>
Subject: USB printer under 2.2.19
Date: 07 May 2001 23:45:46 -0400
Can a 2.2.19 kernel drive a printer via its USB port
(HP Deskjet 952C, specifically)? And if so, what
special stuff do you have to do with /etc/printcap,
and so forth to get it to work?
How about a 2.4.x kernel?
Thanks!
--
Rich Carreiro [EMAIL PROTECTED]
------------------------------
From: Juergen Pfann <[EMAIL PROTECTED]>
Subject: Re: BP6 + Mandrake + reboot problem
Date: Tue, 08 May 2001 05:30:08 +0200
Tim Moore wrote:
>
> (...)
> Jerry Broszkowski wrote:
> >
> > I've had this problem since Mandrake 7.1, now
> > running LM8.0.
> >
> > When I get into a console & su then do a reboot,
> > services shut down properly but then my machine "hangs".
> > Reset button does not work neither does power off. I have
> > to turn off the power supply, wait 6 secs, turn the
> > power supply on then use the "On" button. Everything
> > boots fine after this. "Restart" from WinME does not
> > cause this behavior. Anyone know what causes this?
> >
> > MB: ABit BP6
> > CPU: dual 366Cel's overclocked to 550
> > RAM: 512M
> > OS: LM8.0
> > (...)
>
1st of all : sorry, didn't find Jerry's OP any more, so I reply
to Tim's f'up, but like to address the OP...
2 comments / recommendations :
a.Do you use APM ? You shouldn't do so in SMP systems. If you still
use Mandrake's orig. SMP kernel,hopefully it's compiled without APM
support, or even ACPI if you use a 2.4.x kernel; check the kernel
config file for that. Also, better disable APM & ACPI in BIOS.
b.In my local "Bootprompt HOWTO" (which is a bit outdated I admit -
dealing with 2.0), there is a boot option "reboot={cold|warm}".
Check that, if it still applies to 2.2 or 2.4 series, and feel
free to try that out.
BTW : I don't see such problems with my BP6 (2xCel. 466 not o/c,
768 MB RAM, AGP, 3xPCI, 1xISA card, altogether 10 drives w/300W
PS) - neither in 2.0 UP, nor 2.2 or 2.4 SMP kernels.
The power off at shutdown even worked with freshly-installed
RH 6.1 and SuSE 7.0, but I disabled that later ('cause of
the potential APM probs, see above).
HTH
Juergen
------------------------------
Subject: Re: Backing up Windows on Linux
From: Michael Meissner <[EMAIL PROTECTED]>
Date: 08 May 2001 00:56:00 -0400
Steve Smith <[EMAIL PROTECTED]> writes:
> Yes, I can do this, but...
>
> I'm producing a list of files and directories to back up via "ls", and
> filtering with grep to eliminate certain directories. I would need to
> substantially complicate my script (so far as I can tell) to add quotes or
> escape. I'm wondering if anyone has an easier solution.
>
> Restating it: I don't want to give tar a fixed list of files and directories.
> I want to create the list each time, with the exception of certain
> directories.
>
> Maybe what you're saying is that the easy way is to have a fixed backup list,
> rather than a changing one. This is probably a reasonable idea.
>
> Steve
I would seriously look into using find -print0, grep -Z, perl -0, tar -T
--null, and/or cpio -o -0 instead of ls. Find -0 will print names with a null
terminator instead of newline (to guard against files with whitespace in the
name). For example:
find . -type f -print0 | perl -n0e 'print if (! /foobar/);' | tar -cf foo.tar
-T --null
would tar up all simple files that don't contain the 'foobar' string. Of
course if all you want to do exclude certain directories and are using tar, you
probably just want to use the --exclude option.
--
Michael Meissner, Red Hat, Inc. (GCC group)
PMB 198, 174 Littleton Road #3, Westford, Massachusetts 01886, USA
Work: [EMAIL PROTECTED] phone: +1 978-486-9304
Non-work: [EMAIL PROTECTED] fax: +1 978-692-4482
------------------------------
From: hac <[EMAIL PROTECTED]>
Subject: Re: Backing up Windows on Linux
Date: Tue, 08 May 2001 04:59:39 GMT
Steve Smith wrote:
>
> This seemed like an obvious idea until I got to the part about file
> names with spaces, like "My Documents".
>
> I want to write a script to back up all the directories *except* certain
> ones, like Windows and Recycled. I'm using ls to generate a list of file
> names, and tar tries to interpret "My Documents" as two file names,
> neither of which exist.
>
> I can figure out some hard ways to do this; anyone know any easy ones?
>
Use find with the exclude switch, and pipe it into cpio. Works for
me.
--
Howard Christeller Irvine, CA [EMAIL PROTECTED]
------------------------------
Subject: Re: HDD problem on an ASUS a7v133
From: Michael Meissner <[EMAIL PROTECTED]>
Date: 08 May 2001 02:13:44 -0400
"Aaron Berg" <[EMAIL PROTECTED]> writes:
> I have a brand new a7v133 motherboard with a 30 gig ata100 quatum hdd on the
> ata100 controller primary master. I am trying to install linux on the hard
> drive from booting on a homemade linux kernel on a floppy. When booting the
> kernel it goes through different hardware driver loads and checks until it
> comes to checking the partitions
> Checking partitions:
> hde
> hde: status=0x50 { DriveReady SeekComplete Error }
> hde dma disabled (at this point the computer continues to boot using PIO
> mode 4 for transfers)
>
> I got this message for a while trying different things in bios and different
> hardware configurations. Nothing seemed to change. reset the bios to its
> defaults and now the kernel boots till
> Checking partitions:
> hde (at this point the computer hard crashes)
>
> From what I have read and what I am guessing is that the timings are off
> between the DMA on the HDD and the DMA on the controller. The hard drive is
> brand new. I just took it out of its packaging three days ago. I believe
> that it is a bad hard drive. I tested the computer with a western digital
> ata66 20 gig hard drive and that had no problems.
>
> could anyone confirm exactly what the problem is and if it is infact a
> bad/failing hard drive.
> Thanks,
> Aaron
>
> PS. The hard drive appears to work perfectly in pio mode.
There is a disk corruption bug that shows up with motherboards with the VIA
chipsets that is being discussed on the kernel development mailing list
([EMAIL PROTECTED]). Since I don't have such a motherboard, I
haven't been paying attention to the details much.
--
Michael Meissner, Red Hat, Inc. (GCC group)
PMB 198, 174 Littleton Road #3, Westford, Massachusetts 01886, USA
Work: [EMAIL PROTECTED] phone: +1 978-486-9304
Non-work: [EMAIL PROTECTED] fax: +1 978-692-4482
------------------------------
From: "Anthony" <[EMAIL PROTECTED]>
Subject: Re: VIA 686B bug fix
Date: Tue, 08 May 2001 15:05:04 +0800
In article <3af75b12$0$42883$[EMAIL PROTECTED]>, "Dances With
Crows" <[EMAIL PROTECTED]> wrote:
> ? Which bug were you referring to? The 2.4.x series had problems with
> IDE DMA on the VIA 686 causing data corruption in isolated instances. They
> said this bug was fixed in 2.4.3, and they say it is *really* fixed in
> 2.4.4. It was never present in the 2.2 series, so "Stay with 2.2 for a
> couple of weeks and wait for bug reports to trickle in from the 2.4.4
> users before upgrading," would be my advice.
>
> This bug is entirely within the kernel; you can fix any problems you are
> having with IDE DMA at any time without doing a full 7.0->7.1 upgrade.
NO. This is a HARDWARE bug in the 686B chipset.
Reference:
http://www.theregister.co.uk/content/3/18267.html
The so-called fix is just a work around: disable some setting in the
chipset that triggered data corruption. Otherwise, there won't be a
Windows fix.
--
Linux 2.2.19 #2 Sat Apr 14 12:45:42 CST 2001 i586 unknown
1:35pm up 4:09, 1 user, load average: 0.01, 0.01, 0.00
====== 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: Nils Holland <[EMAIL PROTECTED]>
Subject: Re: VIA 686B bug fix
Date: Tue, 8 May 2001 10:28:22 +0200
Dances With Crows wrote:
> ? Which bug were you referring to? The 2.4.x series had problems with
> IDE DMA on the VIA 686 causing data corruption in isolated instances.
YES, here for example!
> They said this bug was fixed in 2.4.3, and they say it is *really* fixed
> in 2.4.4.
Starting with 2.4.3, I didn't have any problems anymore. 2.4.4 also works.
When booting, I see a message about the VIA686 bugfix being enabled, so it
seems to be present on my system and working correctly.
Greetings,
Nils
------------------------------
Date: Mon, 07 May 2001 18:46:19 +0200
From: Cyrille Artho <[EMAIL PROTECTED]>
Subject: HW recommendations for >4 GB RAM
I am looking for a 64 bit machine that features, above all, fast RAM
access. The computer will have around 16 GB RAM, give or take one binary
magnitude. Both fast (and large) first and second level cache and a fast
overall access are important. Also, ALU performance (for simple integer
calculations) matters somewhat. The rest (FPU etc.) are not relevant.
What hardware do you recommend (at a reasonable cost)? The two obvious
solutions are 64 bit Alphas and Suns, what is best? Also, would you
recommend using Linux as the kernel for such a platform, or use Tru64
Unix/Solaris, respectively? The latter is less important as a dual boot
solution will probably be used anyway.
--
Regards,
Cyrille Artho
A classic is something that everyone wants to have read
and nobody wants to read.
-- Mark Twain, "The Disappearance of Literature"
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: Backing up Windows on Linux
Date: Mon, 07 May 2001 20:09:55 -0500
Steve Smith wrote:
> Restating it: I don't want to give tar a fixed list of files and directories.
> I want to create the list each time, with the exception of certain
> directories.
Use the "find" command to build the file list. See "man find"
and give special attention to the "-prune" directive. Feed the
list to tar through a pipe.
--
-John ([EMAIL PROTECTED])
------------------------------
From: Florian Lorenzn <[EMAIL PROTECTED]>
Subject: DLT tape drives
Date: Tue, 08 May 2001 11:49:06 +0000
Hi there,
do anyone know whether the DLT HP SureStore Tape 80i 40-80GB SCSI device
is supported by the Linux kernel? In the Hardware HOWTO is said that
nearly all SCSI tapes can be used with Linux with the appropriate
SCSI-controller driver (I have got that one) and the /dev/st(whatever)
device-file. Has anybody some experiences with those HP-drives and
Linux? I am usinf SuSE 6.3 but kernel shall soon be updated to 2.4 and I
would not care upgraing other packages, too.
Thanks in advance,
Florian
------------------------------
From: [EMAIL PROTECTED] (M. Buchenrieder)
Subject: Re: VIA 686B bug fix
Date: Tue, 8 May 2001 06:49:01 GMT
[EMAIL PROTECTED] (Dances With Crows) writes:
>On Mon, 7 May 2001 17:41:23 +0200, jay staggered into the Black Sun and
>said:
>>I have a MSI K7T Turbo motherboard with the VIA 686B Southbridge chip
>>(KT-133A chipset). Is there a bug fix for Linux? If so how do I correct
>>the problem. I'm currently using SuSE 7.0 but want to fix the problem
>>before upgrading to 7.1.
>? Which bug were you referring to?
The broken design of the VIA 686B southbridge.
[...]
>This bug is entirely within the kernel; you can fix any problems you are
>having with IDE DMA at any time without doing a full 7.0->7.1 upgrade.
Nope. While there may have been bugs in the driver as well, the
VIA southbridge itself is buggy and may cause severe data loss when
transferring huge amounts of data from channel 1 to channel 2 on
the IDE subsystem. You'll need to search for a BIOS update from
the manufacturer of your board.
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.
------------------------------
** 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 by posting to comp.os.linux.hardware.
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
******************************
