last change: 29-May-2003

Linux on Toshiba Satellite 2410

This document about Linux on Toshiba Satellite 2410 notebooks is also available in French - kindly translated by Pierre-Etienne Melet.

I am using Debian GNU/Linux. I am delighted to report that it works like a charm on my Toshiba Satellite 2410-404. (Yes, UT2003 runs really smoothly.)
Though my experiences consider installation of linux on a Toshiba Satellite 2410-404 all models in Toshiba Satellite 2410 series and 2430 series should be supported. Also models in Toshiba Satellite 1410 series.

Yes, it's running Linux

Already working NVIDIA, TV-OUT, DVD, USB, FIREWIRE, IRDA (also LIRC), APM, SOUND, PCMCIA, MODEM, LAN, WLAN (internal MiniPCI), ACPI. ADDITIONAL KEYS.
Special Workaround KEYBOARD
Not working yet SD-CARD

BIOS

A word on the BIOS of your Toshiba 2410. To get into the BIOS dialog, hold down the escape key on power-up until you are asked to press F1. To set the bootdevice temporarily, hold down F12 on power-up. You will be presented a menu letting you select one of the available bootdevices for the current boot-procedure. (This won't work if you have a masterpassword set and are using the userpassword. In this case, please use the masterpassword.)
This box is capable to boot from PCMCIA. Actually to boot from a PCMCIA-ATA-card with a FAT12, FAT16 or FAT32 partition on it. I created a small linux-distro with X fitting onto a 4-MByte flashdisk for backup and rescue purposes. If you like to try it, feel free to mail me.
To alter BIOS settings (from Linux console or X) like display brightness, number of minutes until display auto-off, bios passwords, video out, fan speed, etc, you can use toshset (for commandline purpose) or The Toshiba Linux Utils from Jonathan Buzzard. Both usually come packaged with your distribution.
Setting display brightness, fan state and video out can also be handled by the special ACPI module toshiba_acpi (see ACPI paragraph).

XFree86

You can either use the Nvidia modules (highly recommended) or you can use the standard "nv" module packaged with X.
To use 3D-acceleration on your Toshiba 2410 using linux you need to get the nvidia-kernel module and the nvidia-glx module for X. Both usually come along packaged with your distribution (nvidia-glx and nvidia-kernel) but you can also get them from here on Nvidia's website.
As of version 1.0-4349 NVidia deliver their drivers as a self-extracing, automatic install file. You just need to download run it. If you are using the drivers of larger or equal version number 1.0-4349 skip the next paragraph. For older versions (or source tarball) read the next paragraph.

Install from source package

Detailed instructions on how to install can be found here or inside the packages. It will take a semi-experienced linux-user about five to ten minutes. If you have never compiled anything before, add an hour for reading some documentation ;)
Using Debian, and the kernel-headers package, you can easily rebuild the Nvidia modules as described on the #debian.de FAQ (in german, but shell commands don't bother about that ;).

Configuring XF86Config

It is a bit tricky to get the GeForce4 420 to work. Use the following hints for your XF86Config-4.
Add an entry into the "Module" section in the XF86Config: 
    Load    "glx"
Remove the following entries, if existing: 
    Load    "GLcore"
    Load    "dri"
In section "Device" replace module "nv" with "nvidia". We should also set the BusID. 
    Driver  "nvidia"
    BusID   "AGP:01:00:0"
Somehow DDC is not working properly on this laptop. So we have to put the following lines into the section "Device": 
    Option  "NoDDC" "1"
    Option  "IgnoreEDID" "1"
Using a TFT together with the commercial Nvidia-drivers will let us use the following lines. 
    Option  "GenerateRTList"   "0"
    Option  "OverridePolarity" "1"
Put the following two lines into section "Screen": 
    Option  "NoDDC" "1"
    Option  "NvAgp" "1"
If you like some eyecandy you can add a translucent shadow to the X-pointer. Add the folowing lines at the end of section "Device": 
    # Some Pointer Eyecandy
    Option		 "CursorShadow" "1"
    Option		 "CursorShadowAlpha" "63"
    Option		 "CursorShadowYOffset" "2"
    Option		 "CursorShadowXOffset" "4"

Removing the ugly black stripe in X

Now we're almost there. Firing up X should work now. But there will be an ugly black vertical stripe on the right-hand side of the screen. To get rid of it put the following line into your modules.conf 
    options NVdriver NVreg_SoftEDIDs=0 NVreg_Mobile=2
Using recent drivers (16-Feb-2003 and up) this issue can also be solved by adding the following line to the Device section in your XF86Config-4. 
    Option "FlatPanelProperties" "Scaling =  native"
Remember: using Debian, please do not write stuff directly to /etc/modules.conf. The content is regenerated on upgrades. Put the line into /etc/modutils/NVdriver and call update-modules once.

Ready to use XF86Config-4

You can dowload my
XF86Config-4 - for nvidia driver (accelerated)
XF86Config-4_nv - for nv driver (no acceleration)

Debian users - listen up

The following note witten by Eduard Bloch reminded me of warning Debian users:
[..] Please note that when you install via make-install, the Nvidia crap irreversibly overwrittes your system files, and they are not preserved on package upgrades. Debian metapackages are a nifty way to work around this problem.[..]
Actually, /usr/lib/libGL.so (which is also in xlibmesa-dev) and all files like /lib/modules/<kernel-version>/modules.* (which are also in kernel-image-<kernel-version>) will be overwritten. The latter ones can be easily preserverd by depmod -a.

TV-OUT

TV-Out works with X and the commercial NVidia drivers. I for myself like using Twinview when displaying something on the TV, i.e. the TV screen shows the same content as the TFT does.
Add the following lines to your Section "Screen": 
    # TV Twinview
    Option		"TwinView" "1"
    Option		"SecondMonitorHorizSync" "30-50"
    Option		"SecondMonitorVertRefresh" "60"
    Option		"TwinViewOrientation" "Clone"
    Option		"TVOutFormat" "COMPOSITE"
    Option		"ConnectedMonitor" "DFP,TV"
    Option		"TVStandard" "PAL-B"
    Option		"MetaModes" "1024x768 @1024x768,1024x768 @1024x768; 800x600 @800x600,800x600 @800x600; 640x480 @640x480,640x480 @640x480"
Please check, what TVStandard is appropriate for your tv and whether you like composite or s-video output. You will find a perfect explanation of all named options in the README coming with the nvidia-glx-package. Dual-Head (one screen extends the other) is also possible and of course use of the CRT-Out.
Please note:
If the TV-OUT is not connected to some kind of tv or monitor while starting up X twinview will be disabled automatically.
If you do not want to run in twinview mode by default, you can add a additional Section "Screen" to your XF86Config-4. One should be configured as-is and one should be configured containing the lines above. Please give the latter one a different Identifier than the first one, eg.: "TV". You will find a preconfigured file named XF86Config-4_nvidia_twinview_tv in the files section. Just read it and try it afterwards with one of these commands: 
    X -xf86config XF86Config-4_nvidia_twinview_new -layout Twinview :1
    X -xf86config XF86Config-4_nvidia_twinview_new -screen TV :1
Note: A good place to exchange experiences about the commercial NVidia Drivers is nV News Froums - Linux And NVIDIA Graphics.

Sound

The chipset is supportet by the intel8x0 modules from alsa or the original kernel-modules.

Alsa

Look into the files section. You will find my configuration for modules.conf for ALSA 0.9. Also grab the according asound.conf to put into /etc/asound.conf

OSS

The modules you need are soundcore sound ac97 ac97_codec v_midi uart401 cmpci i810_audio. As I use these drivers for experimental issues only, I do have a small quick-and-dirty script to probe and remove them called audioscript2410. If you like to use these modules persintently, you should create the matching entries in your modules.conf

Keyboard

Bad news about the keyboard. The keyboard controller is buggy somehow. It spuriously sends keyrelease-events twice, which leads to double-letters. This problem is well-known in the Toshiba-world and has even been reportet to happen with M$-Windoze98 on older laptops. I agree, this sucks!
This issue has been discussed on one of the linux-kernel mailing lists. See here. I would like to quote Alan Cox who said on these kind of hardware bugs:
"Well there is one wya to help stop that, which is not to fix them."
The good news: there are five solutions to this.
  1. Patch the kernel.
  2. Patch the XFree-source
  3. Disable the Xkb extension in your XF86Config - a useful one.
  4. Use Xkb extension with AccessX to suppress keyboard bouncing - another useful one.
  5. Use Xkb extension with xkbset to suppress keyboard bouncing - yet another useful one.
Patching would mean we have to patch source on every release of a new kernel or XFree version. It's up to you if you want to burden this on yourself. Contact me for a kernelpatch. A nicer solution is disabling the Xkb extension. Notice: this will lead to a default US-keyboard! This is easy to change and has to be done once only. By the way: nice opportunity to implement your own needs into your keymap, e.g. mapping the windoze-buttons to something useful. ;) If you do not want to do this you can also use AccessX or xkbset to suppress the keyboard-bouncing.

Disabling Xkb Extension

Creating your own keymap

First of all, we need a keymap of your language. To get an actual one (assuming you have configured the keyboard-section in your XF86Config according to your language) type xmodmap -pke >> $HOME/Xmodmap in a xterm. If your language has additional characters (like mine), you may want to define modifier keys like [AltGr]. An excellent article on how to create customized keymaps: Accented Letters On ASCII Keyboards Under X11. This document is also available in german. If you want a german Xmodmap, you can download mine here and an italian here.
Thanks to Pierre Etienne who sent me a french Xmodmap!

Reconfigure XF86Config

Configure the keyboad section. Remove any options concerning "Xkb" and add the option "XkbDisable". It should look like this: 
    Section "InputDevice"
        Identifier      "Generic Keyboard"
        Driver          "keyboard"
        Option          "CoreKeyboard"
        Option          "XkbDisable"
    EndSection

Autoload keymap

Then we like our keymap to be set whenever we start a Xsession. According to the manual of Xsession, on startup of an Xsession, every script in /etc/X11/Xsession.d will be executed.
Get the script 40custom_load-xmodmap and copy it into /etc/X11/Xsession.d/. On startup, X will look for your customized keymap-file in $HOME/Xmodmap and set the keymap accordingly. If you like to set systemwide settings, copy your Xmodmap to /etc/X11/Xmodmap.

Using Xkb Extension with AccessX

AccessX by Dan Lindner is a set of features within the XKEYBOARD extension of the X Window System designed to make X more accessible to users with disabilities.
To use it to suppress keyboard bouncing, get it from here and execute the following commands: 
    ax +bouncekeys 
    ax -timeout 
    ax bouncedelay 5
If you want these commands to be executed on every startup add the same lines on the end of /etc/X11/xdm/Xsetup. Or you can place them into your .xinitrc file. Or put them into a script like described above.

Using Xkb Extension with xkbset

Get xkbset and execute the following lines: 
    xkbset exp bo &>/dev/null
    xkbset bo 1 &>/dev/null
To let these commands take effect on every startup, see above how to proceed.

Modem

Okay, it is a softmodem. And there's a driver for it. But one step after another.
For soft modems in AC'97 controller slots, the bus PCI-id describes the controller and the subsystem PCI-id describes the modem card inserted into it. Let's take a look at the PCI bus. 
    00:1f.6 Modem: Intel Corp. 82801CA/CAM AC'97 Modem (rev 02) (prog-if 00 [Generic])
    00:1f.6 Class 0703: 8086:2486 (rev 02)    
    Subsystem 1179:0001
So our vendor-id is 8086 (Intel). PCI-id 8086:2486 represents the Intel MC97 controller, to which various AC-Link modems could be connected. The modem itself has id 1179:0001. Subsystem vendor-id is 1179 (Toshiba). Now, what is a modem of an id 0001? I tried to find out which modem this really is, but I had not much success. Well, this information is crap.
Using windows and a terminal, ATI commands return these lines: 
    ATI1 - OK    
    ATI2 - OK
    ATI3 - TOSHIBA V.90 Data+Fax Software Modem Version 3.1.100
    ATI4 - Built on 09/26/2001 20:12:19
    ATI5 - 3.1.100, AMR Intel MB, AC97 ID:SIL REV:0x27, 06
    ATI6 - OK
    ATI7 - AMR Intel MB
Read here about Toshiba's policy concerning softmodems.

Getting the modem to work

As of release 2.7.8 of the SmartLink Modem drivers for linux, the modem is now supported for outgoing calls. Version 2.7.10 works in a stable manner while 2.7.14 (available since May 24th 2003) does not. As mentioned above there's a driver under construction. You can get it from http://linmodems.technion.ac.il/packages/smartlink/.
Probing the modules leads to the following messages (syslog): 
    Jan  7 18:55:56 valea kernel: Smart Link AMRMO modem.
    Jan  7 18:55:56 valea kernel: amrmo: probe 8086:2486 Intel Corp. 82801CA/CAM AC'97 Modem : ICH card...
    Jan  7 18:56:59 valea kernel: This driver is not compatible with the installed modem codec.
Though the codec does not match it is possible to set up a working ppp-connection, i.e. to dial in to your ISP.

IrDA

IrDA is supported. But with a little twist. Damn Toshiba crap ;)
Toshiba Satellite 2410 laptops are provided with a SMCS LPC47N227 SuperIO chip which is IrDA SIR/FIR capable. AFAIK, the chip is connected through an PCI-ISA bridge. The IrDA subsystem of the SuperIO chip is supported by the smc-ircc Linux kernel module. Unfortunately the BIOS neither configurates the SuperIO chip IrDA subsystem (SIR port, FIR port, dma, irq, IrDA mode, power) nor sets the PCI-ISA bridge to decode any usable port. Linux kernel is thus prevented to detect the second UART making impossible to use it in SIR mode. For the same reason, the FIR module smc-ircc is able to detect the SuperIO chip but, once found the IrDA subsystem unconfigured, fails to install.

Solution

The problem is widely known concerning newer Toshiba notebooks. To set parameters of the IrDA-chip you can use Rob Miller's smcinit. His solutions was designed for a Toshiba Satellite 5100 but works for the 2410 series as well. Just get the init-routine from his page, compile, install and run it. After that, the module smc-ircc should probe finely.

IrDA-Hardware and LIRC

LIRC is a package that allows you to decode and send infra-red signals of many (but not all) commonly used remote controls. I am using it to control MPlayer, XMMS or to skip forward my presentations.
The SMCS LPC47N227 chip is supported by the kernel-module lirc_sir which you might have to compile from source. Please note: simultaneous use of IrDA and LIRC ist not possible. You have to remove all modules concerning infrared before probing lirc_sir. Furthermore you have to initialize the SMSC chip like described above.
The lirc_sir module needs to know where to find the SIR-port (SIR = Serial InfraRed) of the infrared-hardware, i.e. I/O port and IRQ. For a Toshiba Satellite 2410 these values are: I/O port 0x03f8, IRQ 3. After Installing the lirc-package and it's kernel-modules try a 
    # modprobe lirc_sir port io=0x03f8 irq=3
Then follow the instrucions on the lirc homepage how to check your setup. If the parameters (I/O port and IRQ) should not match for you, remove the module lirc_sir, modprobe module smc-ircc and get the parameters from syslog, which should now contain some lines like: 
    kernel: found SMC SuperIO Chip (devid=0x5a rev=00 base=0x002e): LPC47N227
    kernel: SMC IrDA Controller found
    kernel:  IrCC version 2.0, firport 0x130, sirport 0x3f8 dma=3, irq=3
Here are our parameters: sirport 0x3f8 dma=3, irq=3.
To use lirc persistentliy, put the following lines into your modules.conf: 
    pre-install lirc_sir /usr/local/sbin/tosh5100-smcinit
    alias char-major-61 lirc_sir
    options lirc_sir io=0x3f8 irq=3
Using Debian? Please do it the debian-way: Put these lines into /etc/modutils/lirc and run update-modules.
If you are wondering about the first line, please read the paragraph about IrDA.

SD-Card Slot

AFAIK, the SD-card controller is a proprietary one manufactured by toshiba. There are no drivers for use with linux. It worked using windoze using original SD cards but failed with any new large MMC cards.

Firewire

 Firewire should work with the module ohci1394 which probes smoothly. But I do not own any firewire-devices to test it.

USB

 USB works out of the box. Uses the module usb-uhci.

PCMCIA

 PCMCIA works also out of the box. Uses module yenta_socket together with ds.

APM

 APM runs smoothly. Except you use X11 with the nvidia-drivers. The module NVdriver will prevent your box to suspend. You'll have to leave X and unload the module NVdriver before suspend.
There is a possibility to get the commercial nvidia-drivers to work with apm properly, but you have to modify the driver sources and in a way that makes the use of AGP impossible. This has been reported to not work very stable on other notebooks.
Before compiling the commercial nvidia kernel-drivers, alternate the file nv.c. Browse the code to this section: 
#else
        case PM_RESUME:
        case PM_SUSPEND:
        //    printk("NVRM: avoiding suspend request, don't want to shutdown!!\n");
            return 1;
#endif
        default:
        //    printk("NVRM: received unknown PM event: 0x%x\n", rqst);
            return 1;
    }
    return 0;
}
#endif
and alternate both return statements from return 1 to return 0: 
#else
        case PM_RESUME:
        case PM_SUSPEND:
        //    printk("NVRM: avoiding suspend request, don't want to shutdown!!\n");
            return 0;
#endif
        default:
        //    printk("NVRM: received unknown PM event: 0x%x\n", rqst);
            return 0;
    }
    return 0;
}
#endif
Please do not forget to deactivate use of AGP in your XF86Config-4: Option "NvAGP" "0"!
Also note it is not a good idea to have agp-support compiled into the kernel when using this workaround.

ACPI

 ACPI is supported by John Belmonte's Toshiba ACPI Extras Driver - module toshiba_acpi. The code of this module is implemented in the current development kernels. To use it with a 2.4.x kernel, you have to get a fresh kernel from kernel.org and apply the according ACPI patch from http://sourceforge.net/projects/acpi/. After that apply the patch from John Belmonte and compile you kernel. Please do not use the kernel-sources from your distro as they often already contain patches.

What to expect from ACPI

Well, as Toshiba is as well as Compaq, Intel, Microsoft and Phoenix inventor and publisher of the current ACPI specifications we may expect the specifications are correctly implemented in Toshiba's BIOSes. Scared by a lot of crappy implementaions I had to deal with (and even more I heard and read about) in past times I am glad to own some hardware manufactured by one of these companies.
Concerning the Toshiba Satellite 2410-404, the nifty things like additional keys, setting display brightness in eight steps, setting video ouput work with the above metioned patch. But any powermanagement options of OSPM (Operating System-directed configuration and Power Management) do not do much on my laptop by now because the sleep states S2, S3 and S4 are not implemented in kernels ofversion 2.4.x . OSPM is a part of ACPI. I am not telling OSPM power management does not work with Linux by now. You'll have to use a development kernel or wait for kernel 2.6.
As I use my notebook every day I am currently using apm for power management as well as the toshset utility (see bios paragraph) for setting display brightness and video output. This works in a reasonably stable manner.

Additional Keys

They are supported by ACPI. Namely by the module designed especially for Toshiba laptops: toshiba_acpi. John Belmonte wrote also a tool (written in python) to handle the additional keys. Grab it from here.

LAN - Onboard Network

The onboard NIC is an EtherExpress PRO/100 VE. The module eepro100 will get your NIC to work.

WLAN (internal MiniPCI)

 The Wireless LAN Mini PCI Card (PA3212U-2MPC) will be recognized as PCMCIA device in slot 0.

orinoco_cs

It is supported by the module orinoco_cs. You can use the kernel-pcmcia-drivers or get get the latest PCMCIA-CardServices from SourceForge. Patches exist for the latter one to be able to change your MAC and support monitor mode. Grab them from shmoo.com.

wavelan2_cs

The Wireless LAN Mini PCI Card is also supported by module wavelan2_cs.o. Drivers are available from Agere. You will find detailled installation instructions on how to install within the package from Agere.
Davyd 'proXy' Madeley sent me the following note:
[..] I have [..] found that the internal wireless card seems to work quite a bit better with the open source orinoco_cs driver, rather then the closed and mildly broken wavelan2_cs.

Notes

Take a look on the files section.
Work on this document is in progress. If you like to contact me concerning help or information exchange, feel free to do so!

Thorsten Haas mail@thorstenhaas.de

Appendix - Changelog

13-Nov-2002Initial document.
17-Nov-2002Added apm, modem, irda, firewire.
24-Nov-2002Added paragraph about AccesX.
5-Dec-2002Added french Xmodmap sent to me by Pierre Etienne.
Added paragraph about SD-card slot.
8-Dec-2002Added solution of IrDA problem.
Added paragraph about ACPI.
10-Dec-2002Added paragraph about Wireless LAN Mini PCI Card.
2-Jan-2003Document correction: the author of the Toshiba ACPI Extra Drivers is John Belmonte.
7-Jan-2003Added paragraph about TV-OUT. Updated paragraph about the modem.
9-Jan-2003Added paragraph about LIRC. Updated paragraph about APM and Audio.
13-Jan-2003Updated paragraph about keyboard and TV-OUT.
24-Jan-2003Added italian Xmodmap set to me by Antonello Facchetti
15-Feb-2003Updated link to SmartLink drivers.
16-Feb-2003Updated paragraph about the commercial NVidia drivers.
09-May-2003Updated paragraph about WLAN drivers, the commercial NVidia drivers, validated links, several updates.
29-May-2003Updated paragraph about modem.

Appendix - Credits

I receive mails concerning this page every day from people who have questions or just like to thank for writing this page. I highly appreciate that and like to thank them, because these are the people who give this site a sense.
Thanks to
Pierre-Etienne Meletfor the french translation and for sending me a french Xmodmap.
Sebastian Roesnerfor sending me a link to Rob Miller's smcinit and a link to John Belmote's toshiba acpi driver.
Eduard Blochfor sending me a note about the commercial nvidia drivers.
Alexander Forschnerfor mailing me the WLAN (MiniPCI) instructions.
John Belmontefor correcting me concerning the author of the Toshiba ACPI Extra Drivers.
Damian Weber for writing a page on using a Toshiba Satellite 2410-404 with SuSE
Tony for heavy testing of theses instructions.
Antonello Facchetti for sending me an italian Xmodmap.
Davyd Madeley for lively communication and for reporting a dead-link to the SmartLink drivers as well as providing a new link.
John Dragun for sending me another solution to the "ugly black stripe"-issue.
You for belonging to the elite capable of reading documentation this far!

Appendix - System Information

Information about all PCI buses in the system and all devices connected to them.
00:00.0 Host bridge: Intel Corp. 82845 845 (Brookdale) Chipset Host Bridge (rev 
04)
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, fast devsel, latency 0
    Memory at e0000000 (32-bit, prefetchable) [size=256M]
    Capabilities: 

00:01.0 PCI bridge: Intel Corp. 82845 845 (Brookdale) Chipset AGP Bridge (rev 
04) (prog-if 00 [Normal decode])
    Flags: bus master, 66Mhz, fast devsel, latency 64
    Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
    Memory behind bridge: fd000000-fdffffff
    Prefetchable memory behind bridge: dbf00000-dfffffff

00:1d.0 USB Controller: Intel Corp. 82801CA/CAM USB (Hub #1) (rev 02) (prog-if 
00 [UHCI])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 0, IRQ 11
    I/O ports at efe0 [size=32]

00:1d.1 USB Controller: Intel Corp. 82801CA/CAM USB (Hub #2) (rev 02) (prog-if 
00 [UHCI])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 0, IRQ 11
    I/O ports at ef80 [size=32]

00:1e.0 PCI bridge: Intel Corp. 82801BAM/CAM PCI Bridge (rev 42) (prog-if 00 
[Normal decode])
    Flags: bus master, fast devsel, latency 0
    Bus: primary=00, secondary=02, subordinate=05, sec-latency=64
    I/O behind bridge: 0000d000-0000dfff
    Memory behind bridge: fce00000-fcefffff

00:1f.0 ISA bridge: Intel Corp. 82801CAM ISA Bridge (LPC) (rev 02)
    Flags: bus master, medium devsel, latency 0

00:1f.1 IDE interface: Intel Corp. 82801CAM IDE U100 (rev 02) (prog-if 8a 
[Master SecP PriP])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 0, IRQ 11
    I/O ports at cff8 [size=8]
    I/O ports at cff4 [size=4]
    I/O ports at cfe8 [size=8]
    I/O ports at cfe4 [size=4]
    I/O ports at cfa0 [size=16]
    Memory at 20000000 (32-bit, non-prefetchable) [size=1K]

00:1f.5 Multimedia audio controller: Intel Corp. 82801CA/CAM AC'97 Audio (rev 
02)
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 0, IRQ 11
    I/O ports at ce00 [size=256]
    I/O ports at cdc0 [size=64]

00:1f.6 Modem: Intel Corp. 82801CA/CAM AC'97 Modem (rev 02) (prog-if 00 
[Generic])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 0, IRQ 11
    I/O ports at ca00 [size=256]
    I/O ports at c980 [size=128]

01:00.0 VGA compatible controller: nVidia Corporation NV17 [GeForce4 420 Go] 
(rev a3) (prog-if 00 [VGA])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, 66Mhz, medium devsel, latency 248, IRQ 10
    Memory at fd000000 (32-bit, non-prefetchable) [size=16M]
    Memory at dc000000 (32-bit, prefetchable) [size=64M]
    Memory at dbf80000 (32-bit, prefetchable) [size=512K]
    Expansion ROM at  [disabled] [size=128K]
    Capabilities: 

02:07.0 FireWire (IEEE 1394): Texas Instruments TSB43AB22 1394a-2000 Controller 
(prog-if 10 [OHCI])
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, medium devsel, latency 64, IRQ 11
    Memory at fceff800 (32-bit, non-prefetchable) [size=2K]
    Memory at fcef8000 (32-bit, non-prefetchable) [size=16K]
    Capabilities: 

02:08.0 Ethernet controller: Intel Corp. 82801CAM (ICH3) Chipset Ethernet 
Controller (rev 42)
    Subsystem: Toshiba America Info Systems EtherExpress PRO/100 VE
    Flags: bus master, medium devsel, latency 64, IRQ 11
    Memory at fcef7000 (32-bit, non-prefetchable) [size=4K]
    I/O ports at df40 [size=64]
    Capabilities: 

02:0b.0 CardBus bridge: Toshiba America Info Systems ToPIC95 PCI to Cardbus 
Bridge with ZV Support (rev 32)
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, slow devsel, latency 168, IRQ 11
    Memory at fce00000 (32-bit, non-prefetchable) [size=4K]
    Bus: primary=02, secondary=04, subordinate=04, sec-latency=0
    I/O window 0: 0000d000-0000d0ff
    I/O window 1: 0000d400-0000d4ff
    16-bit legacy interface ports at 0001

02:0b.1 CardBus bridge: Toshiba America Info Systems ToPIC95 PCI to Cardbus 
Bridge with ZV Support (rev 32)
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: bus master, slow devsel, latency 168, IRQ 11
    Memory at fce01000 (32-bit, non-prefetchable) [size=4K]
    Bus: primary=02, secondary=05, subordinate=05, sec-latency=0
    I/O window 0: 0000d800-0000d8ff
    I/O window 1: 0000dc00-0000dcff
    16-bit legacy interface ports at 0001

02:0d.0 System peripheral: Toshiba America Info Systems: Unknown device 0805 
(rev 03)
    Subsystem: Toshiba America Info Systems: Unknown device 0001
    Flags: medium devsel, IRQ 11
    Memory at fcef6e00 (32-bit, non-prefetchable) [size=512]
    Capabilities:

Thorsten Haas mail@thorstenhaas.de