Linux-Hardware Digest #984, Volume #13 Mon, 4 Dec 00 18:13:05 EST
Contents:
Re: linksys network card (Frank)
Re: Where can I get a driver for my <insert brand and model> printer? (James Richard
Tyrer)
Re: testing nvidia 0.9-5 drivers ([EMAIL PROTECTED])
80G IDE drive support under 2.2.16 kernel? (Michael Shiels)
Re: HSP K56Flex Modem problem (Gareth Randall)
Re: PCI configuration questions (Frank)
Re: Serial Keyboard on Linux (Gareth Randall)
Re: Multiple serial ports on one board (Kostis Mentzelos)
Linux users should avoid Fujitsu ([EMAIL PROTECTED])
Dialing without testing the dialtone ("Filipe Bastos de Freitas")
Re: X-window problems (Skipp)
Re: parallel ("robert smith")
RHL 6.2 on a Compaq Prosignia VS?? ("John Anderson")
Any good hardware vendor Linux support? ([EMAIL PROTECTED])
----------------------------------------------------------------------------
From: Frank <[EMAIL PROTECTED]>
Subject: Re: linksys network card
Date: Mon, 04 Dec 2000 20:21:28 GMT
[EMAIL PROTECTED] wrote:
> I have the current drivers for the card on a dos floppy ( the only way
> I
> could get to the net is on my winblows machine) in a package called
> netdrivers-2.0.src.rpm How do I get these into the kernal so that my
> network card will work. Opps I am running RedHat 7.0 I am very new to
> linux and would appreciate any help.
>
> Thank You
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
Not sure if I can help, I'm running Caldera2.4 & it came with the Tulip
drivers already
installed. If you go to the kernel settings, select Tulip drivers, the
kenel should load
them at startup if you tell it to. I run 2 linksys cards, 1 for internel
net, other for internet.
Frank
------------------------------
From: James Richard Tyrer <[EMAIL PROTECTED]>
Subject: Re: Where can I get a driver for my <insert brand and model> printer?
Date: Mon, 04 Dec 2000 20:22:55 GMT
Or to be more specific, where can I get a driver for <insert distro> for
a <insert printer>.
There is a very good web site for printer information:
http://www.linuxprinting.org/
And, they have various news groups:
news://www.linuxprinting.org/linuxprinting.general
news://www.linuxprinting.org/linuxprinting.hp.general
news://www.linuxprinting.org/linuxprinting.epson.general
and others where you can ask specific questions after you look up your
printer in the large database on the web site.
In general (I hope without seeming to flame anyone -- but I do have a
bad cold and am a little cranky):
I realize that everyone that used Micro$oft Windoz has been conditioned
to think that they need a driver for their printer and, at that, a
specific driver for the specific version of MS-Windows that they are
using. But, this is not reality, it is just part of their monopolistic
plan for world domination.
Different distributions of Linux are not different versions of Linux.
Most everyone is now using Linux v. 2.2.x except for those trying out
the new pre-release of 2.4.??. But, even between these major versions
there aren't any significant difference in the printing setup.
There are several different "print filters" but these aren't "drivers".
Despite differences in which "filter" you are using (CUPS is somewhat a
special case), there are basically two choices and these come with the
distribution:
PostScript printers; Here you do NOT need anything that even resembles a
driver. You just tell the setup routine or widgets for your print
filter that you have a PostScript printer. Perhaps you will need to
tell the printer to print PostScript.
Non-PostScript printers; Here you use GhostScript as the "printer
driver". You again need to use the set up routine or widgets to tell
the print filter which printer you have. Unfortunately, at this point
there can be problems. There are a lot of new printers and GhostScript
requires a software device to support your printer. This is not really
a "printer driver" but it does take the place of one.
After you find out which GhostScript device, or driver, you need from
the Linux Printing site. Execute: "gs -h" to see if you have the needed
device. If you don't, the first step is to see if you have GhostScript
6.01. If not, you should down load it from:
ftp://ftp.cs.wisc.edu/ghost/aladdin/gs601/linux/ghostscript-6.01-1.i386.rpm
[The Source Forge site has source if you can't use an RPM]
http://download.sourceforge.net/ghostscript/
and install it. Again try "gs -h" to see if you have the needed
device. If you now have the device and you can't setup the "print
filter" you will also need to get a newer release of the one you have,
set it up manually (edit the "/etc/printcap" file), or get a new one.
Also see the Linux Printing site for info.
There is a little left out here. The case where you still can't print.
To better understand Linux Printing, I suggest that you read the HOWTO:
http://www.linuxdoc.org/HOWTO/Printing-HOWTO/index.html
Then, if you still have trouble printing, please ask your specific
questions.
And, might I also suggest that you look around the web for information?!
JRT
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: testing nvidia 0.9-5 drivers
Crossposted-To: comp.os.linux.x,comp.os.linux.misc,comp.os.linux.setup
Date: Mon, 04 Dec 2000 20:29:50 GMT
In comp.os.linux.x Mikael Gramont <[EMAIL PROTECTED]> wrote:
> ATI XPERT2000 AGP
Consider yourself lucky... Since there's no PCI gart support for
the linux kernel, PCI version work even worse.
Adam
------------------------------
From: Michael Shiels <[EMAIL PROTECTED]>
Subject: 80G IDE drive support under 2.2.16 kernel?
Date: Mon, 04 Dec 2000 20:31:43 GMT
HELP! My nice new 80G drive appears as as ~13G driver, kinda like ~65G
and 13G almost add back up to 80G, do we have another truncation problem
in the drivers/kernel?? HELP!!!!!
--
Michael Shiels
MaS Network Software and Consulting
2093 Summerwood Court
Mississauga, ON +1 (905) 8239455 voice
Canada L5K 2S8 +1 (905) 8236361 fax
------------------------------
From: Gareth Randall <[EMAIL PROTECTED]>
Subject: Re: HSP K56Flex Modem problem
Date: Mon, 04 Dec 2000 20:32:25 +0000
This is quite strange. I didn't think that ISA had sufficient bandwidth to reliably
operate a Winmodem. I thought that was why ISA was generally a safe bet with regard to
modems, whereas PCI was risky. Am I wrong here?
On your IRQ question, the IRQ10 won't show up until something actually opens the modem
to talk to it, so don't worry about this. Mine basically doesn't show up until I go
online.
Another issue is how can the following happen if there's no proper UART to talk to??
/dev/ttyS2, UART: 16550A, Port: 0x03e8, IRQ: 10
The serial driver works on IO ports to make this identification. I'd be very surprised
if a software only modem would go to the trouble of pretending to be a UART of exactly
the type we'd expect to see.
Is this really a win-modem?
[EMAIL PROTECTED] wrote:
>
> Hope this doesn't get posted twice...
>
> Well...I am pretty sure I already know the answer to this (that answer
> being "It's a Winmodem, go buy another modem"), but some of the pnpdump
> seem to have some info on the type of modem or rather partial
> recognization.
>
> The pnpdump registers the modem as "Id MWI5600, Serial Number 16777216,
> checksum 0x74."...don't know if this just happens to be a fluke or what.
>
> The pnpdump/isapnp.cponf that I get is as follows:
>
> # cat /etc/isapnp.conf
> (DEBUG)
> (READPORT 0x0273)
> (ISOLATE PRESERVE)
> (IDENTIFY *)
> (VERBOSITY 2)
> (CONFLICT (IO FATAL)(IRQ FATAL)(DMA FATAL)(MEM FATAL)) # or WARNING
>
> # Card 2: (serial identifier 74 01 00 00 00 00 56 e9 36)
> # Vendor Id MWI5600, Serial Number 16777216, checksum 0x74.
> # Version 1.0, Vendor version 0.0
> # ANSI string -->CommWave 56 kbps Data/Fax/Voice Modem<--
> #
> # Logical device id MWI5600
> # Device supports I/O range check register
> # Device supports vendor reserved register @ 0x3a
> # Device supports vendor reserved register @ 0x3b
> # Device supports vendor reserved register @ 0x3c
> # Device supports vendor reserved register @ 0x3e
> # Device supports vendor reserved register @ 0x3f
> #
> (CONFIGURE MWI5600/16777216 (LD 0
>
> # Start dependent functions: priority preferred
> # Logical device decodes 16 bit IO address lines
> # Minimum IO base address 0x03e8
> # Maximum IO base address 0x03e8
> # IO base alignment 8 bytes
> # Number of IO addresses required: 8
> (IO 0 (SIZE 8) (BASE 0x03e8) )
> # IRQ 10.
> # High true, edge sensitive interrupt (by default)
> (INT 0 (IRQ 10 (MODE +E)))
>
> # End dependent functions
> (NAME "MWI5600/16777216[0]{CommWave 56 kbps Data/Fax/Voice Modem}")
> # (ACT Y)
> ))
> # End tag... Checksum 0x00 (OK)
>
> # Returns all cards to the "Wait for Key" state
> (WAITFORKEY)
>
> But of course cat /proc/interrupts gives the following which shows no
> irq 10 which is the modem interrupt.
>
> CPU0
> 0: 27361 XT-PIC timer
> 1: 730 XT-PIC keyboard
> 2: 0 XT-PIC cascade
> 4: 26 XT-PIC serial
> 5: 3 XT-PIC MS Sound System
> 8: 1 XT-PIC rtc
> 13: 1 XT-PIC fpu
> 14: 51511 XT-PIC ide0
> 15: 2 XT-PIC ide1
> NMI: 0
>
> cat /proc/ioports gives
>
> # cat /proc/ioport
> 0000-001f : dma1
> 0020-003f : pic1
> 0040-005f : timer
> 0060-006f : keyboard
> 0070-007f : rtc
> 0080-008f : dma page reg
> 00a0-00bf : pic2
> 00c0-00df : dma2
> 00f0-00ff : fpu
> 01f0-01f7 : ide0
> 02f8-02ff : serial(set)
> 0330-0331 : mpu401
> 0370-0371 : OPL3-SAx
> 0376-0376 : ide1
> 0388-038b : Yamaha OPL3
> 03c0-03df : vga+
> 03e8-03ef : serial(set) <---- this is the io port in question
> 03f6-03f6 : ide0
> 03f8-03ff : serial(set)
> 0530-0533 : WSS config
> 0534-0537 : MS Sound System
> e000-e007 : ide0
> e008-e00f : ide1
>
> ..and setserial /dev/ttyS2 gives
>
> /dev/ttyS2, UART: 16550A, Port: 0x03e8, IRQ: 10
>
> And of course when trying to use kppp, I get that the modem is busy.
>
> I pretty much know this is hopeless, but the fact that it seem to
> partially regonize the modem made me think, maybe there is some hope.
--
======= Gareth Randall =======
------------------------------
From: Frank <[EMAIL PROTECTED]>
Subject: Re: PCI configuration questions
Date: Mon, 04 Dec 2000 20:36:33 GMT
Joshua Shaul wrote:
> Hello All,
>
> I'm hoping one of you folks out there can help me. I'm attempting to
> make changes to the PCI configuration of a device, specifically I want to
> change the base address 1 of a card that comes up with an incorrect
> configuration (because of a HW bug). I have found the means to change this
> base address using a utility called setpci. My question then is, what are
> the implications of changing the base address? Do I need to inform the OS
> that things have changed? I notice that after I change the address, the
> files /proc/pci and /proc/bus/pci/devices still show the old base address.
> Are these files where linux looks to find base address info? Is it the
> responsibility of the person who changes the base address to update these
> files as well?
>
> As you can tell I'm utterly confused here, any insight would be greatly
> appreciated.
>
> Josh Shaul
Do the setPCI, if it doesn't work completely you will need to manually modify
the /etc/modules.conf file. Part of my modules.conf looks like this....
# echo io=0x6800-0x687f irq=12 card=4 mem=e8004000-e800407f
alias eth0 tulip io=0x6800 irq=12
the echo line is just info I gathered about the card in Windows, you only
need
the alias line. Setpci should determine all the cards info.
The actual card # is determined which slot on the MB its plugged into, and
yes
bios plays an important part of where the card mem address is.
Frank
------------------------------
From: Gareth Randall <[EMAIL PROTECTED]>
Subject: Re: Serial Keyboard on Linux
Date: Mon, 04 Dec 2000 20:50:18 +0000
Seth Heckard wrote:
>
> On Sun, 3 Dec 2000 22:42:55 -0000, John D. Peedle
> <[EMAIL PROTECTED]> wrote:
> >I'm pretty sure that this is going to be a total no go. The protocol is not
> >the
> >same, the mouse connects through the keyboard and (I think) the plug won't
> >fit
> >anyway.
> >
> >I have to ask the question 'why?'
> >
> >
>
> It is possible -- I've seen a kernel module that can understand the Type
> 5 keyboard. I doubt I bookmarked the site, but a search on google or
> dejanews should turn something useful up.
>
> As for why, if you've ever used a Sun Type 5 keyboard (I like Type 6
> keyboards better yet), you'll understand.
>
> Seth Heckard / [EMAIL PROTECTED]
>
> >Jali <[EMAIL PROTECTED]> wrote in message
> >news:90e93c$orm$02$[EMAIL PROTECTED]...
> >> Hi!
> >>
> >> I am currently experimenting with some exotic hardware. I have a Sun
> >> Keyboard Type 5, I'd like to try on my Intel machine. Does anyone know
> >> about a special Keyboard driver for this, serial-based, keyboard?
If you find one, please post a location here!
Funnily enough I want to do exactly the same thing!
I can tell you a few things that might interest you.
Right, the signal from the mouse is just a plain serial stream. You have three wires:
+5V, GND and serial out.
The signal from the keyboard is the same, except that there is a serial stream going
back to the keyboard to control the LEDs i.e. +5V, GND, serial in, serial out.
Want the pinouts? Look here:
http://docs.sun.com:80/ab2/coll.240.2/UGJAVELIN/@Ab2PageView/12717?Ab2Lang=C&Ab2Enc=iso-8859-1
The fact that both keyboard and mouse go through the same plug just gives the illusion
of complexity - they are two separate interfaces.
So what is needed is this:
Kernel reads keyboard from a serial port (normal keyboard port appears to need a
clocked input by the pinouts, which we don't have)
Kernel still outputs to screen. i.e. this isn't the "serial console" option.
Mouse is read in through another serial port. A user-mode program could maybe be used
to translate the protocol. Not sure whether its 7-bit or 8-bit.
I'd like to know what progress you make.
--
======= Gareth Randall =======
------------------------------
From: Kostis Mentzelos <[EMAIL PROTECTED]>
Subject: Re: Multiple serial ports on one board
Date: Mon, 04 Dec 2000 23:48:44 +0200
Steve
1. read Hardware HOWTO (multiports)
2. I have successfully used:
a. digiboard classic (only ISA)
b. digiboard 16e (ISA)
c. digiboard Xr 920 (PCI)
d. digiboard Xem (ISA,PCI)
e. PCCOM (ISA - best choice)
f. cyclades YeP
and more
kostis.
Steve Wolfe wrote:
>
> I am looking for a board with 4-8 serial ports on it, to turn a linux
> machine into a terminal server. However, I haven't found much info - any
> recommendations would be appreciated.
>
> steve
>
> --
> ------------------
> domain for replies is "codon"
------------------------------
From: [EMAIL PROTECTED]
Subject: Linux users should avoid Fujitsu
Date: Mon, 04 Dec 2000 22:01:17 GMT
Just got off the phone with Fujitsu. I was able to get my call
escalated to L3 before being told to bug off. The reason for my call
was to get spec parameters for the display panel so I could tweak my
XFconfig file.
Perhaps someday we will hear a loud poping sound when Fujitsu pulls
its head out of their collective ###.
In the meantime if someone knows of a resource where I could get this
info it would be much appreciated. Specifically, I am looking for info
for a laptop using the Trident Cyber 9388 to feed the 13.3 TFT display
panel. I found this one
http://www.sanpei.org/Laptop-X/Laptop-X/Fujutsu_Lifebook_E340 and with
a little tweaking I was able to get it to run in 16 bit color mode. 32
bit has a harsh red tinge.
--
I think, therefore, ken_i_m
Chief Gadgeteer, Elegant Innovations
http://shadowfax.linuxfromscratch.org/
learn Linux from the ground up
------------------------------
From: "Filipe Bastos de Freitas" <[EMAIL PROTECTED]>
Subject: Dialing without testing the dialtone
Date: Mon, 4 Dec 2000 22:05:07 -0000
Hello,
How can i configurate my modem to skip the dialtone test, when i dial a
number?
Filipe
------------------------------
From: Skipp <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.linux,alt.os.linux,comp.os.linux.redhat,linux.redhat.install
Subject: Re: X-window problems
Date: 4 Dec 2000 22:13:23 GMT
Hi there,
Your not going to like my answer, but it worked for me. I removed my high
end video card and replaced it with a generic SVGA card I had
layiing around. Red hat Linux picked up the change the next boot and made
the basic config changes for the os.
I then configured the system with the generic card and wrote the X-Server
driver file. It tested and worked just fine... Xconfigurator can/should
be used for the setup.
Then I removed the generic card and replaced my high end card back in
doing the same as above to the new high end card. It worked right after
a new Xconfigurator file was written with the original card back in place.
For some strange reason, Xconfigurator gets all hung up about your video
card and won't let you write a new file once it zeros in (finds) on it.
You can never seem to get out of the loop unless you try what I did, which
worked for me.
cheers
[EMAIL PROTECTED]
: In alt.comp.linux Stiaan <[EMAIL PROTECTED]> wrote:
: Hi there,
: I installed RH 7.0 and all went fine until I want to enter Gnome.
: It does not start the X-server. I went back to Xconfigurator and
: reconfigured everything. The funny thing is that it automatically picks up
: my display adapter (Intel i810, 1024KB shared RAM) and monitor (LG
: Studioworks 45V), but when it starts up the X-server it fails and tell me
: there is a problem. Please help!
: Best Stiaan
------------------------------
From: "robert smith" <[EMAIL PROTECTED]>
Subject: Re: parallel
Date: Mon, 4 Dec 2000 22:16:15 -0000
so what you're saying then is that parallel processing is useless if you
dont have the program(s) which can run on x number of processors
simultaneously.
however i dont think that this is the end of it. why cant you have a single
server if you like divide the code up. or even to be more level headed we
could attempt to have each cpu running background tasks, now this surely has
to be more resonable and simpler in the long run than trying to divide code
up into many segments for an array of cpus to process. i know parallel
processing is useful for databases etc, and other things like that, but its
harder for the programming one would assume to code this way.
the itanium's epic or whatever is designed around this, but to me its
senseless unless everything is re-written.
rob.
<[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED]...
> "robert smith" <[EMAIL PROTECTED]> writes:
>
> > why not?
>
> A P2-200 (if there is such a beast) is faster than a P-200. It may be
> roughly comparable in speed to a PPro-200 with the same-size cache.
>
> Also, two CPUs do not exactly double your processing capabilities.
> For one thing, there's a certain overhead required for managing both
> CPUs and dispatching tasks appropriately. If you've ever taken a Comp
> Sci class on parallel processing, you'll see that, even if much of
> this is handled by hardware, it can still be a real nightmare.
>
> > and what would i have then?
>
> You would have two Pentiums.
>
> You would be able to run two identical tasks at nearly full speed for
> one chip. A single task would run at nearly full speed for one chip.
>
> --
> Eric McCoy <[EMAIL PROTECTED]>
>
> In one gloss of the cut interstellarly I must immovable protect the
> universe.
------------------------------
From: "John Anderson" <[EMAIL PROTECTED]>
Subject: RHL 6.2 on a Compaq Prosignia VS??
Date: Mon, 04 Dec 2000 22:33:30 GMT
Will RHL 6.2 operate on a Compaq Prosignia VS server?
We're having trouble getting it to recognize 106MB of RAM.
The machine sees it on POST.
It worked on a Netware 4.2 file server.
mem=xM at boot does not help.
The machine is running S L O W ! ! ! !
Any suggestions would be appreciated.
ja
------------------------------
From: [EMAIL PROTECTED]
Subject: Any good hardware vendor Linux support?
Date: Mon, 04 Dec 2000 22:42:55 GMT
I'm about to buy a linux web server from either Dell, VA Linux or SGI.
How would people who've dealt with these companies' Linux offerings
rate their support? This is the most important factor in my purchase
decision. I know it will be anecdotal but I need something to go on.
Also, anyone know of any websites where I might find information
pertaining to this?
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
******************************
