Yes, I'd be absolutely interested.  This is for more of a proof of concept
system and not production, so maintenance is not a concern.  
Unfortunately here in Hawaii, my choices of locations to get PCMCIA
ethernet cards is limited and this was the only one on the shelves where I
tried.

Thanks,

Chris

> On Wed, 2009-03-04 at 14:10 -1000, Chris Richmond wrote:
>> Here is the last few messages when inserting the card from the logs:
>>
>> Mar  4 14:04:05 RH-OUTBOUND kernel: pccard: CardBus card inserted into
>> slot
>> 0
>> Mar  4 14:04:05 RH-OUTBOUND kernel: 8139cp 0000:04:00.0: This (id
>> 10ec:8139
>> rev 10) is not an 8139C+ compatible chip
>> Mar  4 14:04:05 RH-OUTBOUND kernel: 8139cp 0000:04:00.0: Try the
>> "8139too"
>> driver instead.
>> Mar  4 14:04:05 RH-OUTBOUND kernel: PCI: Enabling device 0000:04:00.0
>> (0000
>> -> 0003)
>> Mar  4 14:04:05 RH-OUTBOUND kernel: ACPI: PCI Interrupt 0000:04:00.0[A]
>> ->
>> GSI 19 (level, low) -> IRQ 185
>> Mar  4 14:04:05 RH-OUTBOUND kernel: 8139too 0000:04:00.0: cannot remap
>> MMIO,
>> aborting
>> Mar  4 14:04:05 RH-OUTBOUND kernel: ACPI: PCI interrupt for device
>> 0000:04:00.0 disabled
>> Mar  4 14:04:05 RH-OUTBOUND kernel: 8139too: probe of 0000:04:00.0
>> failed
>> with error -5
>
> OK, some good info here.  You can see the kernel tries the 8139cp
> driver, which is older and doesn't work, then tries the 8139too driver,
> for newer cards, which tries to work but fails because it can't setup
> MMIO.  My guess the vendor implementation on this chipset on your
> particular card only supports PIO mode.
>
> The 8139too driver included with RHEL5 uses MMIO by default but does
> include vendor strings for particular cards that only support PIO to
> force PIO mode on those device.  Unfortunately it doesn't look like your
> card is listed as a PIO only card so it simply tries MMIO and then
> aborts.
>
> There is actually a compile time flag that can be set to make the driver
> use PIO by default, however, it can't be changed at runtime in the
> version of the driver included with RHEL5 (more recent versions of the
> driver include a "use_io" module option that allow you to force PIO mode
> at driver load time by just adding an option to modprobe.conf).
>
> It probably wouldn't be too difficult to build a slightly newer version
> of the 8139too driver for the current kernel but I'm not sure how
> involved your wanting to get into this since that would require
> additional maintenance for future kernel updates.  If you want to try it
> let me know and I'll post some instructions.
>
> Later,
> Tom
>
>
> _______________________________________________
> rhelv5-list mailing list
> [email protected]
> https://www.redhat.com/mailman/listinfo/rhelv5-list
>


_______________________________________________
rhelv5-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv5-list

Reply via email to