Messages in this thread | | | From | Otel Florian-Daniel <> | Date | Tue, 13 Jul 1999 02:19:31 +0200 (MET DST) | Subject | Re: OOPs: 2.2.10 + pcmcia-cs-3.0.13 (ThinkPad390E + 3CCFE575BT) |
| |
"Steven N. Hirsch" <shirsch@adelphia.net> writes: > On Sun, 11 Jul 1999, Otel Florian-Daniel wrote:
>> I just tried to splash RH6.0 on a ThinkPad390E with a 3Com Megahertz >> combo card. The stock RH pcmcia.img OOPsed (same as below) upon installation >> couldn't get the card running. I still installed the system, then upgraded >> to the latest (kernel to 2.2.10 and pcmcia-cs 3.0.13). Even after >> upgrading, when booting or when issuing a 'cardctl ident' i get >> OOPsed. Booting __without__ the card inserted is O.K.: The host >> controler is identified and the cardmgr is started: >>
> I'm having zero luck with pcmcia-cs-3.0.13, .14 or anything, with kernel > 2.2.10-ac10.
> Looks like the same thing you're reporting. cardmgr triggers kernel oops > at init time. I'm unable to update the kernel on that machine.
After reading carefully the HOWTO (doh!) , my best bet is that it is a memory probe failure: The cardmgr is started properly and the core modules (ds, i82365, pcmcia_core) are loaded properly (see the /var/log/messages excerpt in my previous postings). _Then_ inserting a card I get OOOPsed.
So I tried the following:
Marcel Schmidt <schmidt@osys.ch> writes: > Hello > In /etc/pcmcia/config.opts remove the high memory range. Just say > "include memory 0xc000-0xfffff".
But i still got the same OOPs. I wil try to install a Win* to see the card working there and __maybe__ i can get the params properly from there (i.e. include only a small mem range).
In the first place I thought it's a 3c575_cb driver problem, but (many thanks to Donald B.) seems it's the general PCMCIA/CardBus control code problem.
> It's an IBM ThinkPad 365X, RedHat 5.2 + kernel-2.2.x with a Linksys > 10/100Base card.
Again, my card is a 3Com Megahertz 3CCFE575BT, so AFAICT it's a different driver than the one for your card. But this tells me once more is the CardBus code to blame and not a driver specific bug.
> Steve
All the best,
Florian.
P.S: Sorry for re-postings and for huge attachments. I didn't realized that vger has such a lag (albeit understandable considering the amount of traffic exchanged). Apologies once more.
- 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/
| |