2003-03-01 01:49:51

by Thomas Molina

[permalink] [raw]
Subject: 2.5.63 wireless loading problem

Sorry if you've seen this before. I decided to resend it with an actual
subject and cc maintainers. Thanks for your patience.

After the module situation settled down I decided to once again try doing
2.5 on my laptop. The problem before was that modules weren't loading,
and eth0 was not being activated. Under 2.5.63-bk latest, the modules are
being loaded, but the last messages are still the same, and I'm not
getting any success in having eth0 activated.

My hardware is a Compaq Presario 12XL325, with a PIII processor. The
system is RedHat 8.0-based. I enclosed my configuration file for the
kernel. I get no errors during compile and load, but I get the following
errors in /var/log/messages:


Feb 28 16:07:26 lap pcmcia: Starting PCMCIA services:
Feb 28 16:07:26 lap kernel: Linux Kernel Card Services 3.1.22
Feb 28 16:07:26 lap kernel: options: [pci] [cardbus] [pm]
Feb 28 16:07:26 lap /sbin/hotplug: no runnable
/etc/hotplug/pcmcia_socket.agent is installed
Feb 28 16:07:26 lap kernel: PCI: Assigned IRQ 9 for device 00:0a.0
Feb 28 16:07:26 lap kernel: Yenta IRQ list 0018, PCI irq9
Feb 28 16:07:26 lap kernel: Socket status: 30000010
Feb 28 16:07:27 lap pcmcia: cardmgr.
Feb 28 16:07:27 lap cardmgr[471]: starting, version is 3.1.31
Feb 28 16:07:27 lap rc: Starting pcmcia: succeeded
Feb 28 16:07:28 lap cardmgr[471]: watching 1 sockets
Feb 28 16:07:28 lap cardmgr[471]: Card Services release does not match
Feb 28 16:07:28 lap cardmgr[471]: socket 0: Intersil PRISM2 11 Mbps
Wireless Adapter
Feb 28 16:07:28 lap kernel: cs: memory probe 0x0c0000-0x0fffff: excluding
0xc0000-0xcbfff 0xe0000-0xfffff
Feb 28 16:07:28 lap netfs: Mounting other filesystems: succeeded
Feb 28 16:07:28 lap cardmgr[471]: executing: 'modprobe hermes'
Feb 28 16:07:28 lap cardmgr[471]: executing: 'modprobe orinoco'
Feb 28 16:07:28 lap cardmgr[471]: executing: 'modprobe orinoco_cs'
Feb 28 16:07:28 lap kernel: orinoco_cs: RequestIRQ: Resource in use

I get the following from /proc/interrupts:

CPU0
0: 2570163 XT-PIC timer
1: 1476 XT-PIC i8042
2: 0 XT-PIC cascade
6: 100 XT-PIC floppy
8: 1 XT-PIC rtc
9: 0 XT-PIC Texas Instruments PCI1410 PC card Card
11: 0 XT-PIC uhci-hcd
12: 309 XT-PIC i8042
14: 8089 XT-PIC ide0
15: 2 XT-PIC ide1
NMI: 0
ERR: 0

and I get the following from lsmod:

Module Size Used by
nls_iso8859_1 3968 1
nls_cp437 5632 1
vfat 16000 1
fat 51712 1 vfat
floppy 66228 2 [unsafe]
orinoco_cs 10536 0
orinoco 60832 1 orinoco_cs
hermes 10144 2 orinoco_cs,orinoco,[permanent]
ds 13280 2 orinoco_cs
yenta_socket 21216 1
pcmcia_core 64864 3 orinoco_cs,ds,yenta_socket
af_packet 28712 0 [unsafe]
iptable_filter 2592 0
ip_tables 18080 2 iptable_filter,[unsafe]
hid 27104 0
uhci_hcd 42152 0
usbcore 123668 4 hid,uhci_hcd
rtc 21112 0
unix 27500 11 [unsafe]


Attachments:
config (21.93 kB)

2003-03-01 02:01:43

by David Hinds

[permalink] [raw]
Subject: Re: 2.5.63 wireless loading problem

On Fri, Feb 28, 2003 at 07:59:47PM -0600, Thomas Molina wrote:

> # CONFIG_ISA is not set

The PCMCIA drivers decide whether or not ISA interrupts are available
based on CONFIG_ISA so you should turn this on.

Perhaps this is a misuse of this configuration option. I'm not sure
what's the right thing to do.

-- Dave

2003-03-01 02:53:26

by Thomas Molina

[permalink] [raw]
Subject: Re: 2.5.63 wireless loading problem

On Fri, 28 Feb 2003, David Hinds wrote:

> On Fri, Feb 28, 2003 at 07:59:47PM -0600, Thomas Molina wrote:
>
> > # CONFIG_ISA is not set
>
> The PCMCIA drivers decide whether or not ISA interrupts are available
> based on CONFIG_ISA so you should turn this on.
>
> Perhaps this is a misuse of this configuration option. I'm not sure
> what's the right thing to do.

Your advice worked. I now get eth0 up and dhcp works.

IMHO it certainly violates the principle of least surprise. I must have
missed the documentation on this configuration advice.

2003-03-01 20:24:06

by John Weber

[permalink] [raw]
Subject: Re: 2.5.63 wireless loading problem

David Hinds wrote:
> On Fri, Feb 28, 2003 at 07:59:47PM -0600, Thomas Molina wrote:
>
>
>># CONFIG_ISA is not set
>
>
> The PCMCIA drivers decide whether or not ISA interrupts are available
> based on CONFIG_ISA so you should turn this on.
>
> Perhaps this is a misuse of this configuration option. I'm not sure
> what's the right thing to do.

How about calling it "PCI<->PCMCIA bridge support" and making it a
sub-menu of CONFIG_PCI?