Linux-Hardware Digest #398, Volume #14 Sun, 25 Feb 01 07:13:08 EST
Contents:
Re: Two sound cards? (Frank Miller)
HP 5100C Scanner problems ("Adam Warner")
Re: IRQ conflict???? ("Adam Warner")
Re: Gateway ALR 7200 sever and Redhat (John Westerdale)
Re: Shared IRQ problem? (Nader)
Re: IRQ conflict???? (Nader)
Re: Could Linux be used in this factory environment ? (Carl Fink)
Boot Redirection ([EMAIL PROTECTED])
"busy inodes on changed media" system log message (Richard Robinson)
"could not create pipe for forking process: too many files open" (Richard Robinson)
Re: Boot Redirection (Juergen Pfann)
Re: Best RAID controller for Linux (K. Nung)
video card frame buffer ([EMAIL PROTECTED])
Access to ISA bus addresses ("Harold E Stout")
Re: Scanner recommendations? (John Thompson)
Re: Access to ISA bus addresses (Glitch)
----------------------------------------------------------------------------
From: Frank Miller <[EMAIL PROTECTED]>
Subject: Re: Two sound cards?
Date: Sun, 25 Feb 2001 04:11:38 GMT
"Barry " wrote:
>
> I would like to use two sound cards together.
>
> I have an onboard Via Apollo Super AC97/Audio and a Sound blaster
> live. Hardrake gets things really wrong and plays samples out one or other
> of then unpredictably and regardless of the one you have chosen to
> configure! So I could try configuring the sound blaster and it will play
> the sample on the Via. The audio mixer has three sound cards and the one
> it uses is a Trident something or other which I don't have! The other two
> that *are* in the system are a SB Live and the via which comes up as
> unknown. All the mixers have worked at some stage but never at the same
> time.
>
> I have used sndconfig and that only finds the Via which it says is not
> supported. It seems to me that it's supported fine. It's in the hardrake
> list and I can get it working.
>
> After all this messing around with things I can no longer log into any
> window manager as root! The system just hangs and I am forced to hit the
> power! Is this related?
>
> Thanks in advance for any help.
>
> Barry
>
> System:
> Mandrake 7.2
> Kernel 2.2.17
>
> Duron 700
> 128Mb Ram PC133 CAS 2
> MSI K7T pro II (a bit too new?)
> Maxtor 15G ata100
You will need to go into the BIOS and disable the on-board sound card.
Then see what happens.
------------------------------
From: "Adam Warner" <[EMAIL PROTECTED]>
Subject: HP 5100C Scanner problems
Date: Sun, 25 Feb 2001 16:24:19 +1300
Hi all,
I have a 2.4.2 kernel with ppscsi and espt 0.92 modules compiled in. I am
using sane 1.0.4.
Using scanimage I can only copy in black and white (1 bit per pixel). If I
try "gray" or "color" I receive these errors:
"scanimage -v --mode gray > test" results in:
scanimage: scanning image of size 2550x3507 pixels at 8 bits/pixel
scanimage: acquiring gray frame scanimage: min/max graylevel value = 255/0
scanimage: sane_read: Error during device I/O
The HP 5100C starts preliminary noises and scanning for a moment before
halting and giving up with this error. I don't think the scanner is faulty
because it passes all tests and works in Windows 95 (I tested that out
to make sure).
Again if I try "scanimage -v --mode color > test":
scanimage: scanning image of size 2550x3507 pixels at 24 bits/pixel
scanimage: acquiring RGB frame scanimage: min/max graylevel value = 255/0
scanimage: sane_read: Error during device I/O
I have already discovered that ppscsi has SMP issues but this still
arises with a uni-processor kernel.
Any advice appreciated.
Regards,
Adam
------------------------------
From: "Adam Warner" <[EMAIL PROTECTED]>
Subject: Re: IRQ conflict????
Date: Sun, 25 Feb 2001 16:40:27 +1300
Hi John,
> I've got an IRQ conflict with my ISDN card and soundcard both USES IRQ9
> and both are PCI... How can I solve this problem.. this is info from
> windows 2000
>
> ISDN location 2 bus0, device 9, function 0 IRQ 9 es1371 location
> 3 bus0, device 10, function 0 IRQ 9
>
> Can I set this up in te BIOS ???? In such case am I looking for the
> location number ???
It may be possible to affect IRQ assignments using the BIOS (check the
plug and play section). It may also be possible to tell Linux which IRQ
to use for at least one of the devices
Do a cat /proc/interrupts to see what interrupts are being used. You
should also be able to compile in interrupt sharing in the kernel (I
understand it needs to be supported by your montherboard).
But the easiest solution could be to just put one of the cards in a
different PCI slot. It's quite likely that will solve your problems.
Regards,
Adam
------------------------------
From: John Westerdale <[EMAIL PROTECTED]>
Subject: Re: Gateway ALR 7200 sever and Redhat
Date: Sun, 25 Feb 2001 04:44:08 GMT
Hello Folks,
I built a machine with an Enlight case from Comp-USA and
one of the 7210 machines. Flashed Intel Bios, and popped in
a couple 600 MHz Coppermines...Got the right cables and Ultra2
drives (nice quiet IBM that will do U2 @ 80 ) and a VooDoo3
2000 PCI and its a good performer, with plenty of slots.
Have had good luck with American Design, Alan knows what
is going on. While there dont forget to pick up a STB
BT878 Video Capture card and a PCI SB-128 card for extra
fun!
Still like Asus boards Bios a little more (better control
over the IRQ assignments), but the 7210 board ($199) is a
good solid SMP board (w/o AGP FWIW). IRQ 20 is nice too :^)
Would be interested to hear of higher Freq CPU's... heard they
will do 800 coppermines... maybe 1 GHz?
J Westerdale
(think a power supply comes with it too FWIW)
# If you can't Xyleogm, Phleogm #
------------------------------
From: Nader <[EMAIL PROTECTED]>
Subject: Re: Shared IRQ problem?
Date: Sat, 24 Feb 2001 20:54:30 -0800
Upgrade your kernel to serial driver 5.05 (see
http://serial.sourceforge.net). This took care of my USR modem - HPT366
PCI IRQ sharing problem without moving any hardware!!!
cfeller wrote:
> My problem is this:
> I am running Red Hat 7 currently (I just upgraded from 6.2 last
> week). The platform is and Intel 233 MMX CPU running on an Intel VX98
> motherboard with 128MB of RAM. The modem is a Zoom 2919 56k flex modem,
> although I have tried several other modems only to have the same
> problem. (The Zoom modem does work fine under Linux in different
> boxes). The problem is this: The modem will only work if the mouse is
> moving. Thus, if I want to surf, send or receive files, I have to keep
> the mouse slowly moving back and forth. (There is no serial mouse port
> on the board, so I have to use a serial mouse.) The mouse is a serial
> mouse under ttyS0 (COM 1). The modem is under ttyS2(COM3). So I moved
> the serial mouse to ttyS1 (COM2), at which point the modem would not
> even respond at all (yes I reassigned the mouse to the new port). I
> moved the mouse back to ttyS0, and I still have the original problem. I
> was thinking I would move the modem to a different slot, but no matter
> which slot I move the modem to, it still shows up as being on ttyS2
> (this holds true under DOS too, yet there is no IRQ problem under DOS).
> When
> Red Hat is loading, I have noticed a message: "Serial Driver 4.27
> withMany_Ports_Sharing_Mulitiple_IRQ enabled". Right now this is where
> I am at...Currently, anything of any size that I want to download, I
> boot into Windows, download it, then I boot back into Linux, and access
> the Windows partition. So if you have any insight into any fix I can
> try to solve my problem, it would be appreciated.
------------------------------
From: Nader <[EMAIL PROTECTED]>
Subject: Re: IRQ conflict????
Date: Sat, 24 Feb 2001 21:03:45 -0800
John:
Try Gary's suggestion first before moving hardware around, etc. Go to
http://sourceforge.net/projects/serial/ and download the 5.05 serial driver.
Add it to your existing kernel source code and recompile. It works wonders
for most people that have PCI IRQ sharing problems.
And Gary, you're welcome. Sorry my e-mail address kept you from thanking me,
but I'm glad to hear the updated driver worked for you and that my posted
e-mail address the spammers :-).
Nader
Gary I Kahn wrote:
> John Christian Engelsen wrote:
>
> > I've got an IRQ conflict with my ISDN card and soundcard both USES IRQ9
> > and both are PCI...
> > How can I solve this problem..
> >
> > ISDN location 2 bus0, device 9, function 0 IRQ 9
> > es1371 location 3 bus0, device 10, function 0 IRQ 9
> >
>
> I had a similar problem on my machine--my modem, sound card, and the USB
> port all share IRQ 9. I didn't find a way to force any of them off of IRQ
> 9, but I did find out that it wasn't a problem after I updated the serial
> driver that comes with the linux kernel. If an ISDN card is a kind of
> serial card, you might want to look into that. If you look in the comments
> in the driver files, you'll find that the driver included in 2.4.x
> addresses PCI IRQ problems. The driver included in kernel 2.2.17 is not as
> up-to-date.
>
> The serial driver can be downloaded all by itself. I found it on
> freshmeat.net. I think that the changes needed went in with v5.02 or 5.03,
> while the driver in kernel 2.2.17 was v4.93, or something like that.
>
> I got the information from someone in this newsgroup. "Nader", I think. I
> tried to email back to him to thank him, but the email address was a fake.
>
> Gary
------------------------------
From: [EMAIL PROTECTED] (Carl Fink)
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.misc
Subject: Re: Could Linux be used in this factory environment ?
Date: 25 Feb 2001 04:47:03 GMT
Reply-To: [EMAIL PROTECTED]
Not to be too annoying, but a one-minute Google search finds Linux
versions of Solid State's MPAC and Mapics for Linux. There doesn't
appear to be an Open Source/Free Software solution, but several
respected commercial products sure seem to exist in Linux versions.
--
Carl Fink [EMAIL PROTECTED]
Manager, Dueling Modems Computer Forum
<http://dm.net>
------------------------------
From: [EMAIL PROTECTED]
Subject: Boot Redirection
Date: Sun, 25 Feb 2001 02:35:49 -0800
I've written a custom boot sector to a floppy disk which when booted, will
load the MBR from hdc1. I have MS-DOS on hda1 and Linux on hdc1. The purpose
of the disk is to allow me to choose between MS-DOS and Linux when booting.
I've tried modifying lilo.conf, but it didn't work.
However, when booting, the system hangs with LI on the screen. I suspect that
the MBR is loaded, but the BIOS fails to load the remainder of the system. I
don't know whether this is because the boot sector was originally loaded from
fd0 or because the system is somewhere other than hda1.
Is there a way for a boot sector on fd0 to make the BIOS or MBR of hdc1 load
the system from hdc1?
------------------------------
From: Richard Robinson <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc,linux.redhat
Subject: "busy inodes on changed media" system log message
Date: Sat, 24 Feb 2001 23:36:31 -0800
Hello,
syslog reports messages of "busy inodes on change media" as of a couple
days ago -- that is, as of when I loaded a CD in the CD-rom drive, then
removed it. I wasn't using the command line command, mount, but rather
the GMOME Midnight Commander File Manager. I am figuring that it mounts
the CD-rom automatically when I click on /mnt/cdrom, and unmounts it
when I click off the /mnt/cdrom folder.
Is "busy inodes on changed media" related to CD-ROM device?
Also, you'd think I could read my RedHat Documentation CD, but cannot.
Instead, the directory names are "???????" (whereas on my Windows box,
the CD contents look normal).
I'm using RedHat 7.0 on a Dell Linux.
Any help you can provide will be greatly appreciated.
-- Richard
------------------------------
From: Richard Robinson <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat,linux.redhat.misc
Subject: "could not create pipe for forking process: too many files open"
Date: Sat, 24 Feb 2001 23:45:09 -0800
Hello,
I'm getting quite a few of these messages in the system log:
"could not create pipe for forking process: too many files open"
What's strange about it (to me, anyways) is that I've really got not
much going on except a Web server serving static pages (and very few
hits at that). This is the Apache 1.3.12 that was default installed with
the install of Red Hat Linux version 7. I'm using a Dell PowerEdge 1300
Any clues?
-- Richard
------------------------------
From: Juergen Pfann <[EMAIL PROTECTED]>
Subject: Re: Boot Redirection
Date: Sun, 25 Feb 2001 09:33:00 +0100
[EMAIL PROTECTED] wrote:
>
> I've written a custom boot sector to a floppy disk which when booted, will
> load the MBR from hdc1. I have MS-DOS on hda1 and Linux on hdc1. The purpose
> of the disk is to allow me to choose between MS-DOS and Linux when booting.
> I've tried modifying lilo.conf, but it didn't work.
>
> (...)
> Is there a way for a boot sector on fd0 to make the BIOS or MBR of hdc1 load
> the system from hdc1?
It's much much simpler than that. For a boot floppy without LILO, just
to
boot a linux system from hard disk, you only need to transfer that
system's
kernel to the floppy disk :
'dd if=/boot/vmlinuz of=/dev/fd0'
Voila...
This method assumes that the kernel file we're talking of is configured
to use the root fs of the system it is compiled with (/dev/hdc1 in our
case)
- which is normally the case with a self-compiled kernel, but not
necessarily if you still use your distro's original installed kernel.
To assure this, you can issue the command "rdev /boot/vmlinuz /dev/hdc1"
before the above-mentioned transfer, or "rdev /dev/fd0 /dev/hdc1"
afterwards, or both if you like.
BTW : I hope you read at least the relevant man pages (here : "dd"
and "rdev") - especially "dd" might be a little bit dangerous in the
hand of someone not 100% knowing what he's doing...
This method has at least one disadvantage : There's no prompt to enter
options/arguments to the kernel - just loading the kernel into memory,
finding the root fs and mounting it, and so on.
But it's nice in situations where you know your linux fs is o.k, but
your MBR is not...
Juergen
------------------------------
From: [EMAIL PROTECTED] (K. Nung)
Crossposted-To: linux.dev.raid
Subject: Re: Best RAID controller for Linux
Date: 25 Feb 2001 08:50:05 GMT
In <sXdi6.73438$[EMAIL PROTECTED]> [EMAIL PROTECTED]
(Prostorage) writes:
>sort of true
>but Adaptec has come out with good Linux drivers for their new
>PCI-Ultra-160M RAID controllers, as has Mylex
>[...]
Really? I spent 1 hour now at Adaptecs web site and didn't find a
driver for the 3200S for Linux kernel 2.4.2... Are you sure that
they have one besides their old RedHat 6.X and SuSe 6.X/7.0?
--nung
------------------------------
From: [EMAIL PROTECTED]
Subject: video card frame buffer
Date: Sun, 25 Feb 2001 03:50:53 -0800
I am trying to setup X11 under Debian Potato. I want to use 800x600, 16bpp
(64K colors). The documentation states that option linear must be used along
with membase being specified as the linear address of the frame buffer.
My video adaptor is an integrated Cirrus Logic CL-GD542X VGA BIOS Version 1.41.
It has 1MB of video RAM. My machine is a 486 with an ISA bus, but the mb
manufacturer claims the card to be PCI local bus. My system has 36MB RAM. The
BIOS does not have an option for making a hole in the system memory (e.g.
excluding RAM between 15MB & 16MB). I contacted the manufacturer, but they said
they do not have this information (linear address of the frame buffer).
I don't know what the linear address of the frame buffer is. How do I find out?
Can it not be probed? Would it not be the same as the video RAM at segment
A000? I tried a membase of A0000000 and several suggested addresses and they
all froze my machine.
What is the difference between a linear address and an address specified as
segment:offset and how to you convert between them? I don't know anything about
protected mode addressing above 1MB. I only know about real mode addressing
under 1MB (i.e. 20 bit addresses).
Why is X11 this way? I've used many graphics programs under MS Windows and
MS-DOS (even protected mode programs under MS-DOS) and have never found any
program wanting to know the linear address of the frame buffer (which I had
never heard of until I met X11). X11 (and most other Linux software) is very
troublesome!
Help! Please!
Thank you.
<begin ROT13>Oenq Eubqrf oeubqrf<end ROT13>@catt.com
------------------------------
From: "Harold E Stout" <[EMAIL PROTECTED]>
Subject: Access to ISA bus addresses
Date: Sun, 25 Feb 2001 10:29:20 GMT
Hi:
Is there a user function which allows root to access the ISA bus address in
the "15-16MB memory hole". I am working on code that needs access to shared
memory at 0xF60000 on the ISA bus. The BIOS has been used to open access to
ISA memory in this region. I have heard about the function mmap(), but have
not gotten it to work. I want to read/write to shared memory using an
address in the range of 0xF60000 0xF61000. I am using the 2.2.14 kernel on a
system with 64Mb of memory.
It seems that 2.2 kernels do not allow access to this area, because the
normal memory map does not set up the hole. Does anyone have a solution to
this problem?
Will mmap() work?
How? Please help. Thanks. Hal
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: Scanner recommendations?
Date: Sat, 24 Feb 2001 18:44:42 -0600
Jason Spaceman wrote:
> I am thinking about buying a flatbed scanner for the computer. I don't
> know much about scanning in Linux, what are some Linux-friendly scanners
> out there? What sort of open source software exists for scanning?
"Sane" is the scanner interface software in linux. It has both
command-line and X user interfaces. You can compile it to work
as a plugin with the Gimp, which gives you as much, if not more
functionality than you would enjoy with the Windows software
bundle that comes with most consumer scanners.
Check the sane homepage http://www.mostang.com for specific
hardware compatibility. Many SCSI scanners will work, and some
USB scanners. Epson seems to be one of the few manufacturers to
have made the technical details of their scanner interface
available to open-source developers. I've had good luck with my
Epson 636U (USB) scanner, but I did have to update to kernel
2.2.18 to make it work smoothly.
--
-John ([EMAIL PROTECTED])
------------------------------
Date: Sun, 25 Feb 2001 07:01:10 -0500
From: Glitch <[EMAIL PROTECTED]>
Subject: Re: Access to ISA bus addresses
Harold E Stout wrote:
>
> Hi:
> Is there a user function which allows root to access the ISA bus address in
> the "15-16MB memory hole". I am working on code that needs access to shared
> memory at 0xF60000 on the ISA bus. The BIOS has been used to open access to
> ISA memory in this region. I have heard about the function mmap(), but have
> not gotten it to work. I want to read/write to shared memory using an
> address in the range of 0xF60000 0xF61000. I am using the 2.2.14 kernel on a
> system with 64Mb of memory.
>
> It seems that 2.2 kernels do not allow access to this area, because the
> normal memory map does not set up the hole. Does anyone have a solution to
> this problem?
>
> Will mmap() work?
> How? Please help. Thanks. Hal
try comp.os.linux.development.system
they could help you more.....this isn't a programming newsgroup , but
people asking about hardware compatiblity/problems
------------------------------
** 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
******************************
