Geoff Simmons schrieb:
> On Fri, Jan 22, 2010 at 11:57:48AM +0100, Stefan Welte wrote:
>> the cardbus/pcmcia-wlan-card (11ab:1faa) doesn't work on my laptop
>> with debian lenny, 2.6.26-2-486:
>
> What laptop do you have?
lshw:
description: Computer
width: 32 bits
*-core
description: Motherboard
physical id: 0
*-memory
description: System memory
physical id: 0
size: 80MiB
*-cpu
product: Pentium MMX
vendor: Intel Corp.
physical id: 1
bus info: c...@0
version: 5.4.3
size: 200MHz
width: 32 bits
capabilities: f00f_bug fpu fpu_exception wp vme de pse tsc msr mce
cx8 mmx
*-pci
description: Host bridge
product: 82C557 [Viper-M]
vendor: OPTi Inc.
physical id: 100
bus info: p...@0000:00:00.0
version: 14
width: 32 bits
clock: 33MHz
*-isa
description: ISA bridge
product: 82C558 [Viper-M ISA+IDE]
vendor: OPTi Inc.
physical id: 1
bus info: p...@0000:00:01.0
version: 02
width: 32 bits
clock: 33MHz
capabilities: isa bus_master
configuration: latency=0
*-display UNCLAIMED
description: VGA compatible controller
product: F65550
vendor: Chips and Technologies
physical id: 6
bus info: p...@0000:00:06.0
version: c6
width: 32 bits
clock: 33MHz
capabilities: vga_controller
configuration: latency=0
*-pcmcia:0
description: CardBus bridge
product: PCI1130
vendor: Texas Instruments
physical id: 7
bus info: p...@0000:00:07.0
version: 04
width: 32 bits
clock: 33MHz
capabilities: pcmcia bus_master
configuration: driver=yenta_cardbus latency=176 maxlatency=5
mingnt=192
module=yenta_socket
*-network
description: 10/100 PC Card
product: 1.0
vendor: Dual Speed
physical id: 0
slot: Socket 0
resources: irq:7
*-pcmcia:1
description: CardBus bridge
product: PCI1130
vendor: Texas Instruments
physical id: 7.1
bus info: p...@0000:00:07.1
version: 04
width: 32 bits
clock: 33MHz
capabilities: pcmcia bus_master
configuration: driver=yenta_cardbus latency=176 maxlatency=5
mingnt=64
module=yenta_socket
*-ide
description: IDE interface
product: 82C621 [Viper-M/N+]
vendor: OPTi Inc.
physical id: 14
bus info: p...@0000:00:14.0
version: 12
width: 32 bits
clock: 33MHz
capabilities: ide bus_master
configuration: driver=Opti621_IDE latency=0 module=opti621
*-ide
description: IDE Channel 0
physical id: 0
bus info: i...@0
logical name: ide0
clock: 33MHz
*-disk
description: ATA Disk
product: TOSHIBA MK3205MAV
vendor: Toshiba
physical id: 0
bus info: i...@0.0
logical name: /dev/hda
Looking at pcmcia:0 in detail:
description: CardBus bridge
product: PCI1130
vendor: Texas Instruments
width: 32 bits
Seems like the cardbus controller/bridge is 32-bit capable?
According to [1] it is.
physical id: 7
bus info: p...@0000:00:07.0
clock: 33MHz
capabilities: pcmcia bus_master
configuration: driver=yenta_cardbus latency=176 maxlatency=5
mingnt=192
According to [2] i need a "resource database", as yenta_socket is my
(complicated) socket driver
AND the second block of "0000:00:07.0" is zero.
Do you think, it's a possible solution to compile pcmciautils with STARTUP =
true ?
> If this is correct, are you using BIOS version 1.38? [1]
Getting this BIOS from rock or the latest BIOS from highscreen/vobis for the
laptop seems
impossible as both manufacturers removed/deleted them, when i asked them for
the files.
>> dmesg shows:
>> cs: pcmcia_socket0: cardbus cards are not supported.
> [..]
>> another card is working in the pcmcia-slots, so the slot is ok.
The slot is okay for 16bit, but maybe not completely for 32-bit, as an
interrupt pin is possibly
not connected... [3]
>> A differnce is that damnsmalllinux uses cardmgr, debian not.
>> Can I set up cardmgr on Debian Lenny somehow?
>
> cardmgr (part of pcmcia-cs) is deprecated and only usable with Linux 2.4
> and earlier kernel versions. It was removed from Debian post-Etch.
>> Have cardbus cards been supported earlier/in former versions and now
>> support stopped completely?
>
> CardBus devices continue to be supported by the Linux kernel, subject to
> driver availability (which you intend to workaround using NDISwrapper).
Seems like support stopped partially in kernels >= 2.6 [3]:
"... The pcmcia-cs socket driver will route ISA interrupts for
CardBus cards if no PCI interrupt is available. The kernel subsystem
doesn't support this option."
dmesg confirms this:
[ 0.184192] PCI: No IRQ known for interrupt pin A of device 0000:00:07.0.
Please try using
pci=biosirq.
[ 0.184502] PCI: No IRQ known for interrupt pin B of device 0000:00:07.1.
Please try using
pci=biosirq.
boot-parameter pci=biosirq is not a solution:
[ 0.210488] PCI: No IRQ known for interrupt pin A of device 0000:00:07.0.
[ 0.210663] PCI: No IRQ known for interrupt pin B of device 0000:00:07.1.
So, why cannot a 2.6 kernel also route an ISA-interrupt like
pcmcia-cs-socket/package, if there
is no PCI interrupt?
The question, who/what produces dmesg message "cs: pcmcia_socket0: cardbus
cards are not
supported" remains? Is it directly from the kernel? Who can I ask for why there
is this message
with my 32-bit card?
Stefan Welte
[1]
http://www.thefreelibrary.com/SystemSoft%27s+CardSoft%2F32+PCMCIA+Software+Supports+The+Texas...-a017815934
[2] http://www.kernel.org/pub/linux/utils/kernel/pcmcia/howto.html
[3] http://lists.infradead.org/pipermail/linux-pcmcia/2004-October/001182.html
General Information:
[4] http://www.kernel.org/pub/linux/utils/kernel/pcmcia/pcmcia.html
--
To UNSUBSCRIBE, email to debian-laptop-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
