Mikael Pettersson wrote:
>
> On Sun, 02 Jul 2000 20:42:08 +1000, Andrew Morton wrote:
>
> >Fingers crossed, this is the final 3c59x patch for Linux 2.4. But I
> >have heard very little from users of the Cardbus NICs.
>
> In my case, it's because the kernel's pcmcia/cardbus and 3c59x
> drivers don't work nearly as well as David Hinds' pcmcia-cs package,
> to the point that I rarely even bother testing the kernel drivers.
Yes, it does appear that 2.4.0's cardbus is broke, and it is not clear
why, nor is it clear who is going to fix it, nor how, nor when.
> Disregarding problems with cardbus bridges on some laptops, I've had the
> following problems with kernel-based pcmcia/cardbus and 3c59x:
>
> 1. Naming confusion and incompatibilities with the pcmcia-cs
> cardmgr. I was able to use the kernel's drivers with pcmcia-cs
> between (roughly) 2.3.48 and 2.3.99-pre1. Later, perhaps
> in 2.3.99-pre9, I tested the kernel drivers again. 3c59x
> did recognize my 3c575_cb card, but cardmgr failed to `up' it,
> apparently due to some naming confusion (eth0 vs 3c575_cb).
> A manual ifup did work, but requiring manual intervention is
> not acceptable.
I kept the pci driver's name as "3c575_cb" in attempt to prevent this
sort of thing, but it appears that fundamental information which the
PCMCIA package needs has been removed from 2.4.
I did manage to get it working, but I had to kludge a few PCMCIA scripts
to do it. Described in
http://www.uow.edu.au/~andrewm/linux/3c59x-2.3-changelog.txt
> ...
> Jul 5 16:46:22 xxx kernel: PCI: No IRQ known for interrupt pin A of device 03:00.0. Please try using pci=biosirq.
I saw that as well. But it works OK.
> Jul 5 16:46:22 xxx cardmgr[448]: executing: 'modprobe cb_enabler'
> Jul 5 16:46:22 xxx cardmgr[448]: + modprobe: Can't locate module cb_enabler
> Jul 5 16:46:22 xxx cardmgr[448]: modprobe exited with status 255
> Jul 5 16:46:22 xxx cardmgr[448]: module /lib/modules/2.4.0-test3-pre2-test/pcmcia/cb_enabler.o not available
> Jul 5 16:46:22 xxx cardmgr[448]: executing: 'modprobe 3c575_cb'
> Jul 5 16:46:22 xxx kernel: 3c59x.c:v0.99L+LK1.1.6 28 May 2000 Donald Becker and others. http://www.scyld.com/network/vortex.html $Revision: 1.97 $
> Jul 5 16:46:22 xxx kernel: eth0: 3Com PCI 3c575 Boomerang CardBus at 0x1400, <3>eth0: Cannot reserve I/O resource 0x80 @ 0x1400, aborting
Oh. What does /proc/ioports have to say about 0x1400-0x147f?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jul 07 2000 - 21:00:18 EST