Re: Support for 16 bit PCMCIA 0x02d0, 0x042d?

2009-01-15 Thread Michael Buesch 
On Thursday 15 January 2009 18:59:46 Stefan Lippers-Hollmann wrote:
> Hi
> 
> On Donnerstag, 15. Januar 2009, Michael Buesch wrote:
> > On Thursday 15 January 2009 00:22:33 Stefan Lippers-Hollmann wrote:
> > > Is there a chance to add support for this kind of device?
> > 
> > I think you're on your own. Nobody else does have such a device
> > and remote-debugging is basically impossible for this.
> 
> Thanks, I just hoped to have missed anything obvious in regards to the ID 
> registration and ssb.

Well, no. There's no known bug.
But the PCMCIA core API is just crap. I had to fight with all those undocumented
flags for a long time to get my card working. There might be some flag needed 
(or not)
by your card that we don't set/clear, yet.
SSB failing to switch the core indicates a very basic problem with the PCMCIA 
setup, because
the coreswitch is the most basic and very first operation we do on the device.
It's also done through PCMCIA config space registers, so it indicates your 
config space
is not accessible. Which is a core failure.
So I _guess_ it could probably be a bug somewhere in 
drivers/net/wireless/b43/pcmcia.c.
However, of course, it can be everywhere. This file is just where I had the 
major problems
to get the thing running.

-- 
Greetings, Michael.
___
Bcm43xx-dev mailing list
Bcm43xx-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Re: Support for 16 bit PCMCIA 0x02d0, 0x042d?

2009-01-15 Thread Stefan Lippers-Hollmann 
Hi

On Donnerstag, 15. Januar 2009, Michael Buesch wrote:
> On Thursday 15 January 2009 00:22:33 Stefan Lippers-Hollmann wrote:
> > Is there a chance to add support for this kind of device?
> 
> I think you're on your own. Nobody else does have such a device
> and remote-debugging is basically impossible for this.

Thanks, I just hoped to have missed anything obvious in regards to the ID 
registration and ssb. I'll see what (or if) I can do (anything) about it.

Regards
Stefan Lippers-Hollmann
___
Bcm43xx-dev mailing list
Bcm43xx-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Re: Support for 16 bit PCMCIA 0x02d0, 0x042d?

2009-01-15 Thread Michael Buesch 
On Thursday 15 January 2009 00:22:33 Stefan Lippers-Hollmann wrote:
> Is there a chance to add support for this kind of device?

I think you're on your own. Nobody else does have such a device
and remote-debugging is basically impossible for this.

-- 
Greetings, Michael.
___
Bcm43xx-dev mailing list
Bcm43xx-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Support for 16 bit PCMCIA 0x02d0, 0x042d?

2009-01-14 Thread Stefan Lippers-Hollmann 
Hi

I recently got hold of an older 16 bit PCMCIA card, which is branded as 
"ADS Tech wireless PCMCIA Card 802.11g (581-150)" and doesn't seem to be 
claimed by b43/ b43legacy. 


$ /sbin/lspcmcia -vv
Socket 0 Bridge:[yenta_cardbus] (bus ID: :02:03.0)
Configuration:  state: on   ready: unknown
Voltage: 3.3V Vcc: 3.3V Vpp: 3.3V
--none--
--none--
Socket 0 Device 0:  [-- no driver --]   (bus ID: 0.0)
Configuration:  state: on
Product Name:   Broadcom 802.11g PCMCIA 4.3
Identification: manf_id: 0x02d0 card_id: 0x042d
function: 6 (network)
prod_id(1): "Broadcom" (0x620c9840)
prod_id(2): "802.11g PCMCIA" (0x30f90a13)
prod_id(3): "4.3" (0x97b52576)
prod_id(4): --- (---)


The PCMCIA ID suggests that it might be a BCM4306 variant, trying to add
this ID to b43 (2.6.28)


--- a/drivers/net/wireless/b43/pcmcia.c
+++ b/drivers/net/wireless/b43/pcmcia.c
@@ -34,6 +34,7 @@
 
 
 static /*const */ struct pcmcia_device_id b43_pcmcia_tbl[] = {
+   PCMCIA_DEVICE_MANF_CARD(0x2D0, 0x42d),
PCMCIA_DEVICE_MANF_CARD(0x2D0, 0x448),
PCMCIA_DEVICE_NULL,
 };


doesn't seem to be sufficient/ succesful (did I eventually miss any 
required SSB addition?) and results in the following dmesg output:


[...]
yenta_cardbus :02:03.0: pcmcia: parent PCI bridge I/O window: 0xa000 - 
0xbfff
pcmcia_socket pcmcia_socket0: cs: IO port probe 0xa000-0xbfff: clean.
yenta_cardbus :02:03.0: pcmcia: parent PCI bridge Memory window: 0xd000 
- 0xdfff
yenta_cardbus :02:03.0: pcmcia: parent PCI bridge Memory window: 0x9000 
- 0x9fff
pcmcia_socket pcmcia_socket0: pccard: PCMCIA card inserted into slot 0
pcmcia_socket pcmcia_pcmcia 0.0: pcmcia: registering new device pcmcia0.0
pcmcia_socket pcmcia_socket0: cs: IO port probe 0x100-0x3af: excluding 
0x230-0x237
pcmcia_socket pcmcia_socket0: cs: IO port probe 0x3e0-0x4ff: clean.
pcmcia_socket pcmcia_socket0: cs: IO port probe 0x820-0x8ff: clean.
pcmcia_socket pcmcia_socket0: cs: IO port probe 0xc00-0xcf7: clean.
pcmcia_socket pcmcia_socket0: cs: IO port probe 0xa00-0xaff: clean.
socket0: cs: memory probe 0xd000-0xdfff: excluding 
0xd000-0xd07f 0xd400-0xd7ff
[...]
ssb: Failed to switch to core 0
b43-pcmcia: Initialization failed (0, -110)
b43-pcmcia: probe of 0.0 failed with error -110
Broadcom 43xx driver loaded [ Features: PMLR, Firmware-ID: FW13 ]
ssb: Failed to switch to core 0
b43-pcmcia: Initialization failed (0, -110)
b43-pcmcia: probe of 0.0 failed with error -110
[...]


Is there a chance to add support for this kind of device?

Regards
Stefan Lippers-Hollmann

-- 
I'm sorry if you receive this mail twice, I already sent this mail at the 
4th of january, but it didn't appear in the mailing list archive yet (most 
likely because I wasn't subscribed at that time).
___
Bcm43xx-dev mailing list
Bcm43xx-dev@lists.berlios.de
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev