Linux-Hardware Digest #589, Volume #14 Sun, 8 Apr 01 19:13:05 EDT
Contents:
Re: HP Pavilion Sound and Redhat ("Cyberbear")
Re: HP Pavilion Sound and Redhat (Dances With Crows)
Re: Safe hdparm settings? (Dances With Crows)
Re: Safe hdparm settings? (James K. Wiggs)
Re: Linux on Xbox? (Thomas Tonino)
Re: Parport problem (Dances With Crows)
SCSI issue with two cards (Don Gingrich)
Re: beeps (Norman Levin)
Re: Safe hdparm settings? (James K. Wiggs)
RH7.0 ignores /etc/resolv.conf (Walter Dnes)
Re: how to set time in cmos clock so it doesn't revert on reboot? (Walter Dnes)
Re: SCSI issue with two cards ("Peter T. Breuer")
Re: beeps ("Peter T. Breuer")
Re: this is very weird!!!!!!!!! ("BetrOffDed")
----------------------------------------------------------------------------
From: "Cyberbear" <[EMAIL PROTECTED]>
Subject: Re: HP Pavilion Sound and Redhat
Date: Sun, 08 Apr 2001 20:18:19 GMT
The Riptide sound card and modem used in many Pavilions is not supported as
far as I know. If you go to www.opensound.com, the have a driver that "sort
of" works. I was able to play CD audio, but the system sounds did not work
for me. I am biting the bullet and installing a Sound Blaster card. I hope
the model I found will work :-)
Good luck with yours!
"Charlie Bailey" <[EMAIL PROTECTED]> wrote in message
news:U%2A6.693946$[EMAIL PROTECTED]...
> Has anyone had any luck getting Redhat to recognize the HP onboard sound?
> From what it looks like, it is not on the list of supported sound
hardware.
> Thanks.
>
> Charlie
>
>
>
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: HP Pavilion Sound and Redhat
Reply-To: [EMAIL PROTECTED]
Date: 08 Apr 2001 20:32:49 GMT
On Sun, 08 Apr 2001 19:31:00 GMT, Charlie Bailey staggered into the
Black Sun and said:
>Has anyone had any luck getting Redhat to recognize the HP onboard
>sound? From what it looks like, it is not on the list of supported
>sound hardware. Thanks.
More info, please. What's the output of "cat /proc/pci" that relates to
sound cards? If there's nothing there for the sound card, then the
sound is attached to the ISA bus, and you will have to use pnpdump or
open up the case and go hunting for the sound chip in order to find out
what you need to do.
At the very least, provide the model# of your machine. There are quite
a few different kinds of HP Pavilions, and not all of them have the same
sound chips!
--
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: Safe hdparm settings?
Reply-To: [EMAIL PROTECTED]
Date: 08 Apr 2001 20:32:52 GMT
On 8 Apr 2001 19:20:37 GMT, James K. Wiggs staggered into the Black Sun
and said:
> I've got an older ASUS P2L97 motherboard with an IBM 13 GB
>IDE drive model IBM-DJNA-371350. The question I'm trying to find an
>answer for is simple: Is it *safe* to use the -u1 parameter to hdparm
>for this disk/MB combination? I've spent the last 2 and a half hours
>trying to find some documentation somewhere to tell me if that
>motherboard has one of the buggy IDE interfaces mentioned in
>Documentation/ide.txt. The ASUS website doesn't say.
Check the output of "cat /proc/pci" as it relates to IDE interfaces. It
should cough up something similar to the following:
Bus 0, device 7, function 1:
IDE interface: VIA Technologies VT 82C586 Apollo IDE (rev 16).
Medium devsel. Fast back-to-back capable. Master Capable. Latency=32.
I/O at 0xd000 [0xd001].
If it's not a CMD640 or RZ1000, then it should be safe to use -u1. The
buggy interfaces mentioned are often found in 486 or early Pentium
systems, not recent boards.
>motherboard manuals don't say. I can't find a pertinent posting on any
>newsgroup using groups.google.com or in any archived mailing list I can
>track down. Can *anyone* give me a straight answer on this one?
The man page for hdparm errs on the side of blind, screaming paranoia.
This is probably a good thing, since people *have* lost data by pushing
dodgy hardware. I think you'll be all right.
You could've avoided wasting 2 hours by making sure you had a current
backup (make one! NOW!), entering "hdparm -m16 -u1 -c1 /dev/hdX", and
attempting to read and write large files. If nothing bad happens
within the first 5 minutes, you're OK. 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] (James K. Wiggs)
Subject: Re: Safe hdparm settings?
Date: 8 Apr 2001 20:43:26 GMT
On Sun, 08 Apr 2001 19:36:39 GMT, Rinaldi J. Montessi
<[EMAIL PROTECTED]> wrote:
>James K. Wiggs wrote:
>
>> Folks,
>>
>> I've got an older ASUS P2L97 motherboard with an IBM 13 GB
>> IDE drive model IBM-DJNA-371350. The question I'm trying to find
>> an answer for is simple: Is it *safe* to use the -u1 parameter
>> to hdparm for this disk/MB combination? I've spent the last 2
>> and a half hours trying to find some documentation somewhere to
>> tell me if that motherboard has one of the buggy IDE interfaces
>> mentioned in Documentation/ide.txt. The ASUS website doesn't
>> say. The ASUS motherboard manuals don't say. I can't find a
>> pertinent posting on any newsgroup using groups.google.com or
>> in any archived mailing list I can track down. Can *anyone*
>> give me a straight answer on this one?
>>
>> many thanks,
>> Jim Wiggs
>
>I have an acceptable chipset (Intel 440xx) and acceptable drives, but only
>notice a small increase in speed with the -u1 setting. What you're doing
>with -u1 is not shutting down other interrupts when there is disk activity.
>Is the extra 5 to 6 (ymmv) mb/s worth the risk? Here that's only =< 6%
The objective is not to speed up the drive, but to cut down the
number of serial input overruns. This box is used to receive a very
high-volume realtime data feed, and it is dropping a completely
unacceptable number of bytes. Anything we can do to improve the
situation has to be investigated. We're using a Comtrol RocketPort
PCI card running at 460800, and still losing a lot of data whenever
there is any disk I/O. Switching to SCSI is not feasible at this
time, so this is the only option I really have available.
>The defaults seem adequate here. What is the output of hdparm /dev/hdx ?
>And what does hdparm -Tt /dev/hdx show?
Here is the output you request:
[root@milestone /root]# hdparm -iv /dev/hda
/dev/hda:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 1650/255/63, sectors = 26520480, start = 0
Model=IBM-DJNA-371350, FwRev=J76OA30K, SerialNo=GM0GMT9M221
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=26310/16/63, TrkSize=0, SectSize=0, ECCbytes=34
BuffType=3(DualPortCache), BuffSize=1966kB, MaxMultSect=16, MultSect=off
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
CHS=1650/255/63 LBA Native, LBAsects=26520480
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:240,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 *mode2 mode3 mode4
Drive Supports : ATA/ATAPI-4 T13 1153D revision 17 : ATA-1 ATA-2 ATA-3 ATA-4
and:
[root@milestone /root]# hdparm -Tt /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 2.01 seconds =63.68 MB/sec
Timing buffered disk reads: 64 MB in 16.92 seconds = 3.78 MB/sec
[root@milestone /root]# hdparm -Tt /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 1.77 seconds =72.32 MB/sec
Timing buffered disk reads: 64 MB in 16.61 seconds = 3.85 MB/sec
[root@milestone /root]# hdparm -Tt /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 1.77 seconds =72.32 MB/sec
Timing buffered disk reads: 64 MB in 16.54 seconds = 3.87 MB/sec
Unfortunately, the -Tt readings are unlikely to be accurate;
there are several running background applications that I *can't*
shut down to run these tests. They use little CPU but a lot of
RAM. Even so, those buffered disk reads look pretty pathetic,
eh?
>Rinaldi
>--
>We all know Linux is great...it does infinite loops in 5 seconds.
>--Linus Torvalds
------------------------------
From: Thomas Tonino <[EMAIL PROTECTED]>
Subject: Re: Linux on Xbox?
Date: Sun, 08 Apr 2001 23:27:33 +0200
Hermann Samso wrote:
> So noone finds interesting to see such a powerful and cheap
> system running Linux?
Don't know how much RAM it has, but it could be interesting indeed. It
has an nVidia card - that should work. The only problem could be input.
I guess that should be the easy part.
But we'll know as soon as we see the machine. MS will be selling it
below cost most likely, so I would not be surprised if there is some
structure that makes it impossible to run software unless you've paid
for some license.
Thomas
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Parport problem
Reply-To: [EMAIL PROTECTED]
Date: 08 Apr 2001 21:25:38 GMT
On Sun, 08 Apr 2001 20:02:14 GMT, Gunnar staggered into the Black Sun
and said:
>Dances With Crows wrote:
>> Is the "lp" module loaded, and does "ls -l /dev/lp0" return the
>> following?
>> crw-rw---- 1 root lp 6, 0 Jul 29 2000 /dev/lp0
>Here's my output:
>k62:~# ls -l /dev/lp*
>crw-rw---- 1 root lp 6, 0 Nov 30 16:23 /dev/lp0
>> You must have the "lp" module loaded to get any output from the printer.
>> If it doesn't exist in /lib/modules/$VERSION/misc/ , then you need to
>> compile it.
>The only modules I see is parport.o and parport_pc.o
>> When you tried to access /dev/lp0, the kernel module loader
>> should've loaded the module automatically. Put the following line in
>> /etc/modules.conf if it isn't there already:
>> alias char-major-6 lp
>A cat to /dev/lp0 does not load any module.
>I'm not a module person so could you shortly tell me what
>alias char -major -6 lp
>would do.
>What if I compile it into the kernel (I have tried that too, but it didn't
>work).
The "lp" module is in a funny place. Character Devices->Parallel
Printer Support is the option you should set to "Y" or "M". It is not
with the parallel port options for reasons of symmetry, balance, and
logic (a printer is a character device, after all, and it makes sense
for all the character devices to be in one place) but lp is needed for
doing just about anything with the parallel port on the x86
architecture. (Yes, this bit me in the arse many months ago, and I
flailed in confusion for an hour before figuring it out.)
The alias line mentioned above would register that the character device
with major number 6 would be handled by the lp module. Then, when you
try to access the device with major number 6, the kernel module loader
would check to make sure that the lp module was loaded, and load it if
it's not there.
>> The hardware is there; just get that lp module loaded and you should be
>> in business. If you wish to use interrupt-driven mode for printing,
>> which might help performance, try doing "tunelp /dev/lp0 -T on". HTH
>I'm glad the hardware is there, this box is not that "upgradeable"...
>Gunnar the still confused...
This is odd, as I thought the stock Debian install gave you all the
modules you would ever need and then some. Oh well, see above for the
option you need to set. 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: Don Gingrich <"gingrich"@melbpc(dot)org(dot).au>
Subject: SCSI issue with two cards
Date: Mon, 09 Apr 2001 07:36:19 +1000
The problem is that the kernel seems to get
confused with more than one card in the system.
I have too many devices for a single card(at least
at the moment) and I need to be able to force
which one is the primary since there are HDDs
on both and thus the order of drives changes,
depending on which is primary.
I've got an adaptec 1542c and a 2940U. Obviously,
(to a human) the 2940 is the primary. I'm thinking
that one option may be to put the driver for the
2940 in the kernel and make the 1542 a module.
Does this sound like a good strategy? Any other
thoughts or suggestions?
I looked for a FAQ but didn't see one -- sorry
if this is a repeat.
-Don
--
Don Gingrich Unix SysAdmin,Comp Sci Dept
RMIT - Melbourne, Australia
gingrich(at)melbpc.org.au gingrich(at)acm.org
gingrich(at)acslink.net.au gingrich(at)cs.rmit.edu.au
------------------------------
Date: Sun, 08 Apr 2001 17:06:49 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: beeps
Vlad wrote:
>
> Hello all!
>
> Where on earth do i switch off these irritating system beeps, like when i
> backspace too far,
** that doesn't sound like the 'system', but vi. vi loves to tell people
that they are stupid when then try to move left of the left margin or to
the right of the end of the line. Total arrogance is what I call it.
Have to use some hardware tool to turn volume of beeps down. I don't know
if vi(m) has a sound control.
or do something similar. These speaker beeps drive me
> nuts!
>
> Thanks!
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
From: [EMAIL PROTECTED] (James K. Wiggs)
Subject: Re: Safe hdparm settings?
Date: 8 Apr 2001 22:13:59 GMT
On 08 Apr 2001 20:32:52 GMT, Dances With Crows <[EMAIL PROTECTED]> wrote:
>On 8 Apr 2001 19:20:37 GMT, James K. Wiggs staggered into the Black Sun
>and said:
>> I've got an older ASUS P2L97 motherboard with an IBM 13 GB
>>IDE drive model IBM-DJNA-371350. The question I'm trying to find an
>>answer for is simple: Is it *safe* to use the -u1 parameter to hdparm
>>for this disk/MB combination? I've spent the last 2 and a half hours
>>trying to find some documentation somewhere to tell me if that
>>motherboard has one of the buggy IDE interfaces mentioned in
>>Documentation/ide.txt. The ASUS website doesn't say.
>
>Check the output of "cat /proc/pci" as it relates to IDE interfaces. It
>should cough up something similar to the following:
>
>Bus 0, device 7, function 1:
> IDE interface: VIA Technologies VT 82C586 Apollo IDE (rev 16).
> Medium devsel. Fast back-to-back capable. Master Capable. Latency=32.
> I/O at 0xd000 [0xd001].
>
>If it's not a CMD640 or RZ1000, then it should be safe to use -u1. The
>buggy interfaces mentioned are often found in 486 or early Pentium
>systems, not recent boards.
Damn! Why didn't I think of /proc/pci? And yup, here's the
result:
Bus 0, device 4, function 1:
IDE interface: Intel 82371AB PIIX4 IDE (rev 1).
Medium devsel. Fast back-to-back capable. Master Capable. Latency=32.
I/O at 0xd800 [0xd801].
Doesn't look like one of the bad ones. Is anyone aware of
any issues with this driver? Running 2.2.14 on that box.
>>motherboard manuals don't say. I can't find a pertinent posting on any
>>newsgroup using groups.google.com or in any archived mailing list I can
>>track down. Can *anyone* give me a straight answer on this one?
>
>The man page for hdparm errs on the side of blind, screaming paranoia.
>This is probably a good thing, since people *have* lost data by pushing
>dodgy hardware. I think you'll be all right.
>
>You could've avoided wasting 2 hours by making sure you had a current
>backup (make one! NOW!), entering "hdparm -m16 -u1 -c1 /dev/hdX", and
>attempting to read and write large files. If nothing bad happens
>within the first 5 minutes, you're OK. HTH,
There's no backup device attached to that box. Backups consist
of nightly SQL database dumps and tar's of several directories, which
are then remotely copied to another machine where they are eventually
burned onto CDs. I still live in fear of having to rebuild that box.
We are planning a major hardware upgrade for it, which will include
getting an OnStream IDE drive into a box on the same network for use
as a backup device, but that's at least 3-6 weeks off. I *really*
don't like it, but those are the financial constraints we are working
within right now.
>--
>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] (Walter Dnes)
Subject: RH7.0 ignores /etc/resolv.conf
Date: Sun, 8 Apr 2001 22:28:22 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
I'm trying to salvage a frankenstein monster consisting of parts from
a few dead computers, and use it as my RedHat 7.0 testbed, learn
iptables, etc. Since RH7 can't auto detect an old ISA network card, I
had to figure out the correct modules.conf settings. I've got that
part working...
> [root@waltdnes70 root]# cat /etc/modules.conf
> alias parport_lowlevel parport_pc
> alias eth0 ne
> options eth0 irq=10 io=0x300
> alias sound-slot-0 sb
> options sb irq=5 dma=1 io=0x220 esstype=1788
> options opl3 io=0x388
eth0 comes up OK. I've ftp'd stuff from my main computer, so I know
that most networking works. I've manually set up the NIC with a
DOS-based utility, and it works when dual-booted to Win98 with no
resource confilcts at all. My resolv.conf, copied straight from a
working RH6.2 machine, lists out like so...
> [root@waltdnes70 root]# cat /etc/resolv.conf
> nameserver=204.101.251.1
> nameserver=204.101.251.2
But DNS is totally broken, and it can't connect to any names...
> [root@waltdnes70 root]# nslookup www.cnn.com
> Server: waltdnes70
> Address: 0.0.0.0
>
> *** waltdnes70 can't find www.cnn.com: No response from server
It's trying to use my local machine for DNS, which fails. Just in
case you think my networking is totally broken, here's what happens when
I manually specify a nameserver for nslookup...
> [root@waltdnes70 root]# nslookup www.cnn.com 204.101.251.1
> Server: dns1.sympatico.ca
> Address: 204.101.251.1
>
> Non-authoritative answer:
> Name: cnn.com
> Addresses: 207.25.71.6, 207.25.71.20, 207.25.71.22, 207.25.71.23
> 207.25.71.24, 207.25.71.25, 207.25.71.26, 207.25.71.27, 207.25.71.28
> 207.25.71.29, 207.25.71.30, 207.25.71.5
> Aliases: www.cnn.com
So it does work when manually forced, but ignores /etc/resolv.conf,
which makes it useless on the net. In 25 words or less HELP!!!.
--
Walter Dnes
<[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED] (Walter Dnes)
Crossposted-To: comp.os.linux.networking
Subject: Re: how to set time in cmos clock so it doesn't revert on reboot?
Date: Sun, 8 Apr 2001 22:28:24 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Fri, 06 Apr 2001 02:35:13 GMT, Q, <[EMAIL PROTECTED]> wrote:
> I am running redhat 6.2 and I do the rdate -s time.server.com and
> it updates the clock but when I reboot, it reverts. I compiled
> in Enhanced Real Time Clock Support in the kernel but that
> doesn't help. How do I do this?
Quoting from "man setclock"...
DESCRIPTION
setclock sets the hardware clock on the current system to
the current time stored in the system clock.
Here's the simple two-liner that I use on my machine...
rdate -s time.chu.nrc.ca
setclock
It's a simple 2-step process. First set the kernel's clock from an
external time-server, then immediately set your hardware clock from the
kernel clock.
--
Walter Dnes
<[EMAIL PROTECTED]>
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: SCSI issue with two cards
Date: Mon, 9 Apr 2001 00:25:39 +0200
Don Gingrich <"gingrich"@melbpc(dot)org(dot).au> wrote:
> The problem is that the kernel seems to get
> confused with more than one card in the system.
It only seems that way to you. Perhaps it is you who are confused? My
kernels have always worked fine with two scsi adapters in the system.
And two nics, and two soundcards ...
> I have too many devices for a single card(at least
Impossible .. you would need >15, no? But I understand .. I use two
controllers just to keep the cable lengths and socket numbers per
cable down.
> at the moment) and I need to be able to force
> which one is the primary since there are HDDs
> on both and thus the order of drives changes,
> depending on which is primary.
But you can force the detection order .. pci
cards are detected in ascending slot order. Slot 1 (furthest from cpu)
is detected first. You can reverse the pci scan order too if you
prefer. It's a boot param.
In general you can force the detection order between two differnet
drivers by just choosing which driver you load first!
> I've got an adaptec 1542c and a 2940U. Obviously,
> (to a human) the 2940 is the primary. I'm thinking
Whichever one you like may be considered primary by a human!
But the 1542c must be ISA? That's useless. Get a symbios for $30
and put it in a pci slot.
> that one option may be to put the driver for the
> 2940 in the kernel and make the 1542 a module.
> Does this sound like a good strategy? Any other
> thoughts or suggestions?
Just load the driver you want loaded first, first.
Peter
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: beeps
Date: Mon, 9 Apr 2001 00:32:34 +0200
Norman Levin <[EMAIL PROTECTED]> wrote:
> Vlad wrote:
>> Where on earth do i switch off these irritating system beeps, like when i
>> backspace too far,
> ** that doesn't sound like the 'system', but vi. vi loves to tell people
> that they are stupid when then try to move left of the left margin or to
> the right of the end of the line. Total arrogance is what I call it.
> Have to use some hardware tool to turn volume of beeps down. I don't know
> if vi(m) has a sound control.
: set noerrorbells
and/or in vim at least
: set visualbell
(error bell is flashing lights in las vegas)
Peter
------------------------------
From: "BetrOffDed" <[EMAIL PROTECTED]>
Subject: Re: this is very weird!!!!!!!!!
Date: Sun, 08 Apr 2001 22:51:50 GMT
In article <9aqe7m$27g$[EMAIL PROTECTED]>, "P�man Malekzadeh"
<[EMAIL PROTECTED]> wrote:
> I am looking for the driver for the Hercules 3D Prophet II MX video card
> for linux. I am running mandrake 7 , I do not have XF86 4. if anyone
> can help I greatly appreciate it.
>
You really need X4+.
Follow the instructions here:
http://www.linuxdoc.org/HOWTO/mini/Nvidia-OpenGL-Configuration/index.html
------------------------------
** 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
******************************
