Page 1 of 3

Posted: Mon 16 May 2005, 03:49
by nduanetesh
OK, so I downloaded and burned puppy 1.0.1, because the disappearing ndiswrapper seemed like a 1.0.2 bug to me...so, I booted 1.0.1 on a clean pup001 file, and did the file extracting, etc...now, when puppy boots, just before X starts, ndiswrapper loads correctly, then I get this:

Error for wireless request 'Set ESSID' (8B1A): SET failed on device eth1; No such device.

So, is this an indication that the pcmcia subsystem isn't seeing the card? or something else?

ND

Posted: Mon 16 May 2005, 10:01
by BarryK
I'm running 1.0.2RC right now. I typed:

# modinfo ndiswrapper

...which gives useful info. Then I typed:

# modprobe ndiswrapper

...and it loaded ok.

If you just type
# insmod ndiswrapper
it won't work, can't find the module, you would have to specify the entire path and filename. But there's more to it -- insmod in the 2.6 kernel seems to be partly crippled. It's a different modprobe package that I had to compile for k2.6.

Posted: Mon 16 May 2005, 14:12
by nduanetesh
When I was running 1.0.2RC, I first booted up with a clean pup001 file, and after I'd extracted the necessary files and such, I typed

modprobe ndiswrapper

and received no error messages...in fact, I received no message at all. It just gave me a new prompt line, so I took this as a good sign. It was only when I rebooted that I got the error message, which was apparently produced by one of the config files I installed, because it occurred as part of the boot process, rather than by any input I did. So it seems like some bug is causing ndiswrapper to disappear? I'll try all of this again at home, as I'll soon be separated from this computer, and report my findings...

Any ideas on the error message I'm getting with 1.0.1? At least ndiswrapper is loading OK there...

ND

Posted: Mon 16 May 2005, 20:56
by Pizzasgood
Sorry. We're past the extent of my knowledge. Good thing Bladehunter's back :)

Good luck!

Posted: Mon 16 May 2005, 22:02
by nduanetesh
Barry, you're right. "Insmod" gives and error in 1.0.2RC where "modprobe" does not. I'm currently in XP and don't have a utility to handle the .tar files, but I'm nearly positive that one of the configs calls ndiswrapper with insmod. I can change this line to modprobe easily enough. However, I'm suspecting that the pcmcia system isn't seeing (or identifying or something) the card, because after I manually modprobe ndiswrapper, if I then do "iwconfig", I get an output which shows me two devices (one is eth0, but sorry i can't remember the other) and says "no wireless extensions" or something like that for both of them. Sorry I'm not being more precise with the messages. is there any way i can get a list of pcmcia hardware that the system has detected?

ND

PS. i downloaded and burned Feather yesterday (good thing I brought a CDRW!) and all I had to do was run through the network setup wizard and it worked like a charm. Might they have some moves we can steal?
PPS That success makes me all the more determined to get this to work!

Posted: Tue 17 May 2005, 08:58
by Guest
Ndiswrapper does not interface with the pcmcia subsystem, basically it just needs the pcmcia drivers loaded so it can talk to what's on the otherside of the bridge. Most 32bit 'pcmcia' cards are what is know as PC Cards.

I think the command to list the attached pcmcia devices is
#cardctl ident

acutally try
#cardctl --help and it should give you a list of options.

After you load ndiswrapper have a look at the /var/log/syslog or /var/log/messages

eg
# cat /var/log/messages | grep ndis

or

# cat /var/log/syslog | grep ndis

and post that info

If you do a simple modporbe ndiswrapper your wireless interface will be
wlan0.

If you run rc.wirless the wireless interface will be whatever
DEVICE=" "
is ie eth1 wlan0 eth0 and so forth

One thing DO NOT LOAD NDISWRAPPER AND THEN RUN rc.wirless do one or the other

Posted: Wed 18 May 2005, 17:36
by nduanetesh
Thanks for the help and info, Bladehunter. I won't be able to get my hands on that 'puter until Saturday night or Sunday morning, though, so I'll post the info found in the log files then.

Thanks agian,

ND

PS. My order of operation had been to:

1. unarchive the files I'd been given to thier proper locations
2. edit the ESSID in wireless.conf (leaving everything else blank means "auto", right?)
3. add the given lines to rc.local
4. (and this is where I might have been going wrong, according to what you've just told me) modprobe ndiswrapper
4.a. receive no interesting output, actually, no output at all
5. reboot. still no internet connection.
6. reboot into XP, come back to forum to post my troubles.

any steps I need to add? anything I'm doing terribly wrong?

Posted: Mon 23 May 2005, 03:34
by nduanetesh
Hi guys,

VERY busy this weekend (helping with a move), but here are the results...

cardctl ident gives the following:

Socket 0:
no product info available

This seems like a very bad sign to me. Seems that the card isn't being detected? or simply cannot be identified by puppy? One light is lighting up, btw...of two.

anyway, for drill I ran

# cat /var/log/messages | grep ndis

and I get

[date and time] syslog.info klogd: ndiswrapper version 1.1 loaded (preempt=yes, smp=no)

and that's it. Thoughts?

Thanks for sticking with me...

ND

Posted: Wed 25 May 2005, 09:58
by Guest
hmmmmmm

Try
#cat /proc/pci > pci.txt

and either attach the text file (pci.txt) or cut and paste the contents.....We're looking for any references to the card

Should I assume that the machine in question has 2 pcmcia sockets ?

If so is there any info about socket 1 ?

Posted: Fri 27 May 2005, 02:31
by nduanetesh
Hi Bladehunter,

No, only one pcmcia slot in this little beastie...

I was thinking about trying to talk my girlfriend through booting puppy, running the command, finding the output file, mounting a USB key, copying the file, unmounting the key, and rebooting (all over the phone), but then decided I'd just do it myself Saturday. So, I'll post it Sat night or early Sunday.

Thanks,

ND

BTW, I'm not sure if I ever mentioned this...but it's a Dell Inspiron 5100

Posted: Sun 29 May 2005, 15:53
by nduanetesh
Ok, here is the pci.txt file. The wifi adapter is the last item listed. It's definately there, and seems to be detected with no problems (at least here)...so, any idea where I'm going wrong?

ND

EDIT: The forum won't allow me to upload a file with a txt extension, apparently ("The Extension txt is not allowed"), so here's the important part cut and pasted. If you need any more info from the file, let me know.

Bus 3, device 0, function 0:
Ethernet controller: Atheros Communications, Inc. AR5212 802.11abg NIC (rev 1).
IRQ 11.
Master Capable. No bursts. Min Gnt=10.Max Lat=28.
Non-prefetchable 32 bit memory at 0x20800000 [0x2080ffff].

Posted: Wed 01 Jun 2005, 04:57
by Guest
The most important piece of info is the PCI ID......could you post that up ?

Posted: Wed 01 Jun 2005, 05:25
by nduanetesh
Ashamed to have to ask, but how do I get it?

I know this is like, linux101.

:oops:

ND

BTW, in case you still need this:
Contents of pci.txt:

PCI devices found:
Bus 0, device 0, function 0:
Host bridge: Intel Corp. 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 3).
Prefetchable 32 bit memory at 0xe0000000 [0xe7ffffff].
Bus 0, device 1, function 0:
PCI bridge: Intel Corp. 82845G/GL[Brookdale-G]/GE/PE Host-to-AGP Bridge (rev 3).
Master Capable. Latency=32. Min Gnt=12.
Bus 0, device 29, function 0:
USB Controller: Intel Corp. 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #1 (rev 2).
IRQ 11.
I/O at 0xbf80 [0xbf9f].
Bus 0, device 29, function 1:
USB Controller: Intel Corp. 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #2 (rev 2).
IRQ 11.
I/O at 0xbf40 [0xbf5f].
Bus 0, device 29, function 2:
USB Controller: Intel Corp. 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) USB UHCI Controller #3 (rev 2).
IRQ 11.
I/O at 0xbf20 [0xbf3f].
Bus 0, device 29, function 7:
USB Controller: Intel Corp. 82801DB/DBM (ICH4/ICH4-M) USB 2.0 EHCI Controller (rev 2).
IRQ 11.
Non-prefetchable 32 bit memory at 0xf4fffc00 [0xf4ffffff].
Bus 0, device 30, function 0:
PCI bridge: Intel Corp. 82801 PCI Bridge (rev 130).
Master Capable. No bursts. Min Gnt=4.
Bus 0, device 31, function 0:
ISA bridge: Intel Corp. 82801DB/DBL (ICH4/ICH4-L) LPC Bridge (rev 2).
Bus 0, device 31, function 1:
IDE interface: Intel Corp. 82801DB/DBL (ICH4/ICH4-L) UltraATA-100 IDE Controller (rev 2).
IRQ 11.
I/O at 0xbfa0 [0xbfaf].
Non-prefetchable 32 bit memory at 0x20000000 [0x200003ff].
Bus 0, device 31, function 5:
Multimedia audio controller: Intel Corp. 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Audio Controller (rev 2).
IRQ 7.
I/O at 0xb800 [0xb8ff].
I/O at 0xbc40 [0xbc7f].
Non-prefetchable 32 bit memory at 0xf4fff800 [0xf4fff9ff].
Non-prefetchable 32 bit memory at 0xf4fff400 [0xf4fff4ff].
Bus 0, device 31, function 6:
Modem: Intel Corp. 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) AC'97 Modem Controller (rev 2).
IRQ 11.
I/O at 0xb400 [0xb4ff].
I/O at 0xb080 [0xb0ff].
Bus 1, device 0, function 0:
VGA compatible controller: ATI Technologies Inc Radeon Mobility M7 LW [Radeon Mobility 7500] (rev 0).
IRQ 11.
Master Capable. Latency=32. Min Gnt=8.
Prefetchable 32 bit memory at 0xe8000000 [0xefffffff].
I/O at 0xc000 [0xc0ff].
Non-prefetchable 32 bit memory at 0xfcff0000 [0xfcffffff].
Bus 2, device 1, function 0:
Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 1).
IRQ 7.
Master Capable. Latency=32.
Non-prefetchable 32 bit memory at 0xfaffe000 [0xfaffffff].
Bus 2, device 4, function 0:
CardBus bridge: Texas Instruments PCI4510 PC card Cardbus Controller (rev 2).
IRQ 11.
Master Capable. Latency=168. Max Lat=5.
Non-prefetchable 32 bit memory at 0x20001000 [0x20001fff].
Bus 2, device 4, function 1:
FireWire (IEEE 1394): Texas Instruments PCI4510 IEEE-1394 Controller (rev 0).
IRQ 11.
Master Capable. Latency=32. Min Gnt=2.Max Lat=4.
Non-prefetchable 32 bit memory at 0xfaffd800 [0xfaffdfff].
Non-prefetchable 32 bit memory at 0xfaff8000 [0xfaffbfff].
Bus 3, device 0, function 0:
Ethernet controller: Atheros Communications, Inc. AR5212 802.11abg NIC (rev 1).
IRQ 11.
Master Capable. No bursts. Min Gnt=10.Max Lat=28.
Non-prefetchable 32 bit memory at 0x20800000 [0x2080ffff].

Posted: Thu 02 Jun 2005, 09:48
by Guest
#lspci -n will get pci-id



But...The athros chipset is support by madwifi, which is native driver......

Posted: Fri 03 Jun 2005, 02:56
by nduanetesh
bladehunter wrote:
But...The athros chipset is support by madwifi, which is native driver......
...which means?

I should Just Work(tm)?

...I'm afraid that I'm missing something painfully obvious here, and I'll just feel like an idiot when we find it...

:?

ND

Posted: Sun 05 Jun 2005, 14:39
by nduanetesh
results of lspci -n

(final item on the output):

3:0.0 168c:0013 (rev 01)

what next?

ND

Posted: Tue 07 Jun 2005, 09:26
by Guest
Ok


Now I just got hold of a dlink dwl-510 which uses the same chipset (atheros)...At the moment I'm running a 2.6.8 kernel on Slackware with the madwifi driver in master mode (ie access point)....All seems to be working fine.Once Barry decides on a kernel version for Puppy 1.0.3 I'll compile the drivers and forward them across. The only problem is that ad-hoc mode is "broken" according to the madwifi site..
So hopefully this will take the dependency away from ndiswrapper for these chips.

BTW in master mode (AP mode) it can also be used to bridge a wireless network to a wired network...

Posted: Sun 12 Jun 2005, 22:45
by Guest
OK just a bit of an update...

Have compiled and successfully loaded and run madwifi driver on 2.4.22 kernel

Downloaded puppy-1.0.3 and will compile and test drivers and post results....Hang in there we'll get you going :D

Posted: Mon 13 Jun 2005, 05:32
by nduanetesh
Thanks for sticking with me, Bladehunter!

ND

Posted: Mon 13 Jun 2005, 09:31
by Guest
The only problem is that a need crypto option is required for this driver has been left out by Barry......If I compiled a new kernel would that be a problem for you to use ?

I don't know if you have the facilities to remaster with a new kernel