Please note: This project is no longer active. The website is kept online for historic purposes only.
If you´re looking for a Linux driver for your Atheros WLAN device, you should continue here .

Ticket #1981 (new defect)

Opened 13 years ago

Last modified 12 years ago

Problem using Routerboard 14 (PCI to 4 mini-PCI adapter) with madwifi

Reported by: thisisrv <rv.vijay@gmail.com> Assigned to:
Priority: major Milestone:
Component: madwifi: driver Version:
Keywords: routerboard 14, soekris Cc:
Patch is attached: 0 Pending: 0

Description (Last modified by mentor)

I am using Routerboard 14 (PCI to 4 miniPCI adapter) on a 433MHz AMD Geode processor running 2.6.23 linux kernel (it is a Soekris net5501 board). I use an Atheros-based miniPCI wireless adapter (CM9:5004MP) driven by madwifi. Though my wireless card is getting detected, I could not associate with an essid or change the channel of operation. Nor am I able to assign an IP address. Please find the series of commands that I typed and the corresponding output:

$ wlanconfig ath3 destroy

$ wlanconfig ath3 create wlandev wifi0 wlanmode adhoc
ath3

$ ifconfig ath3 down

$ iwpriv ath3 mode 1

$ifconfig ath3 up

scan_next: chan 65535->36
wifi0: ath_chan_set: unable to reset channel 36 (5180 MHz) flags 0x140
'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->40
scan_next: chan 65535->44
wifi0: ath_chan_set: unable to reset channel 44 (5220 MHz) flags 0x140
'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->48
scan_next: chan 65535->149
wifi0: ath_chan_set: unable to reset channel 149 (5745 MHz) flags
0x140 'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->153
scan_next: chan 65535->157
wifi0: ath_chan_set: unable to reset channel 157 (5785 MHz) flags
0x140 'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->161
scan_next: chan 65535->165
wifi0: ath_chan_set: unable to reset channel 165 (5825 MHz) flags
0x140 'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->52
scan_next: chan 65535->56
wifi0: ath_chan_set: unable to reset channel 56 (5280 MHz) flags 0x140
'Hardware didn't respond as expected' (HAL status 3)
scan_next: chan 65535->60
scan_next: chan 65535->64
wifi0: ath_chan_set: unable to reset channel 64 (5320 MHz) flags 0x140
'Hardware didn't respond as expected' (HAL status 3)

However, it works well with the Routerboard 11 adapter (PCI to a single mini-PCI adapter). I have no problems with Routerboard 11. The problem is only with Routerboard 14. I also tried using an older version of Linux (2.4.26) and I got similar problems. The logs for 2.4.26 are as follows:

$ ifconfig ath0 192.168.100.7
ath0: unable to reset hardware; hal status 3
SIOCSIFFLAGS: Input/output error

Please let me know if anything need to be done to get Routerboard 14 working on my machine. The problem seems to be with madwifi as the adapter (Routerboard 14) seems to work fine on Windows. I am using the latest version of madwifi (not ng).

Change History

06/07/08 01:35:21 changed by mentor

  • description changed.

06/07/08 11:51:16 changed by anonymous

what does dmesg, /proc/interrupt and (lspci -t && lspci -vvn) look like?

(follow-up: ↓ 5 ) 06/07/08 11:53:00 changed by anonymous

it should be lspci -vvnn sorry!

(follow-up: ↓ 6 ) 06/07/08 12:03:59 changed by strasak@bubakov.net

thisisrv , does your RB14 work ok under windows on the very same machine? We use here over 50 of these beast and it is sometimes tricky to get them run ok - aka on regular PC boards they don't work well in every PCI slot, they are bitchy about irq assigment, sometimes they don't work at all on some chipsets or cause HW lockups etc. etc. . We use them on a couple of GEODE based boards - alix 1C, where they work great, even RB18 works there, but with problems.

(in reply to: ↑ 3 ) 06/07/08 22:17:31 changed by thisisrv <rv.vijay@gmail.com>

Replying to anonymous:

it should be lspci -vvnn sorry!

Please find the /proc/interrupt and lspci outputs below.

$ cat /proc/interrupts
  
          CPU0
  0:      23569          XT-PIC  timer
  1:          0          XT-PIC  keyboard
  2:          0          XT-PIC  cascade
  4:       1503          XT-PIC  serial
  7:          0          XT-PIC  ath1
  9:          0          XT-PIC  eth2
 10:          0          XT-PIC  ath0
 11:          6          XT-PIC  eth0
 14:      14710          XT-PIC  ide0
 15:          0          XT-PIC  usb-ohci
NMI:          0
LOC:          0
ERR:          0
MIS:          0

I have connected two miniPCI cards to the RB14 adapter, ath0 and ath1. They are assigned IRQs 10 and 7 respectively. The lspci outputs are consistent with this as seen below (I have edited the log below to contain just the relevant info),

$ lspci

00:0e.0 PCI bridge: Hint Corp: Unknown device 0021 (rev 15)
01:08.0 Ethernet controller: Unknown device 168c:0013 (rev 01)
01:09.0 Ethernet controller: Unknown device 168c:0013 (rev 01)

$ lspci -t
-[00]-+-09.0
      +-0e.0-[01]--+-08.0
                   \-09.0
      
$ lspci -vvnn

00:0e.0 Class 0604: 3388:0021 (rev 15)
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B-
        Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 64, cache line size 08
        Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
        I/O behind bridge: 0000d000-0000dfff
        Memory behind bridge: a4000000-a7ffffff
        BridgeCtl: Parity- SERR- NoISA- VGA- MAbort- >Reset- FastB2B-
        Capabilities: [80] Power Management version 2
                Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
                Bridge: PM- B3+
        Capabilities: [90] #06 [0000]
        Capabilities: [a0] Vital Product Data

01:08.0 Class 0200: 168c:0013 (rev 01)
        Subsystem: 185f:1012
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B-
        Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 168 (2500ns min, 7000ns max), cache line size 08
        Interrupt: pin A routed to IRQ 10
        Region 0: Memory at a4000000 (32-bit, non-prefetchable) [size=64K]
        Capabilities: [44] Power Management version 2
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=2 PME-

01:09.0 Class 0200: 168c:0013 (rev 01)
        Subsystem: 185f:1012
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B-
        Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 168 (2500ns min, 7000ns max), cache line size 08
        Interrupt: pin A routed to IRQ 7
        Region 0: Memory at a4010000 (32-bit, non-prefetchable) [size=64K]
        Capabilities: [44] Power Management version 2
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=375mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=2 PME-

I even tried using just one miniPCI card with the same bad luck! As you can see from the above logs, the devices are getting detected, the interrupts are assigned!

(in reply to: ↑ 4 ) 06/07/08 22:31:55 changed by thisisrv <rv.vijay@gmail.com>

Replying to strasak@bubakov.net:

thisisrv , does your RB14 work ok under windows on the very same machine? We use here over 50 of these beast and it is sometimes tricky to get them run ok - aka on regular PC boards they don't work well in every PCI slot, they are bitchy about irq assigment, sometimes they don't work at all on some chipsets or cause HW lockups etc. etc. . We use them on a couple of GEODE based boards - alix 1C, where they work great, even RB18 works there, but with problems.

No, as I mentioned before, the linux machine that I use is actually a Soekris net5501 board (soekris.com/net5501.htm). Whereas the Windows machine is actually my PC with a Pentium processor. I just tried associating the card to a wireless network and it seemed to work.

06/08/08 22:15:09 changed by anonymous

Well, I had "bad luck" with everything every chipset&cpu combination that doesn't have a IO-APIC and APIC with current HALs on my RB14 (0.9.30 & 0.10.5.6) ... maybe someone else could reproduce this by booting with noapic nolapic.

(follow-ups: ↓ 9 ↓ 11 ) 06/18/08 22:55:44 changed by thisisrv <rv.vijay@gmail.com>

I got a reply from one of the Mikrotik guys saying that I have to set the madwifi driver in "silent mode". However, the guy said that he does know the details of it. Any idea what that is and how to set it?

(in reply to: ↑ 8 ) 06/19/08 04:00:59 changed by mentor

Replying to thisisrv <rv.vijay@gmail.com>:

I got a reply from one of the Mikrotik guys saying that I have to set the madwifi driver in "silent mode". However, the guy said that he does know the details of it. Any idea what that is and how to set it?

I have absolutely no idea what he is talking about.

(in reply to: ↑ 8 ; follow-up: ↓ 12 ) 09/29/08 19:58:10 changed by anonymous

Replying to thisisrv <rv.vijay@gmail.com>:

I got a reply from one of the Mikrotik guys saying that I have to set the madwifi driver in "silent mode". However, the guy said that he does know the details of it. Any idea what that is and how to set it?

I don't think there is a "silent mode" in madwifi. However, your problem might be related to insufficient electrical power - when you fire up the card, it is eating too much and the hardware gets stuck in some invalid state.

[whoops. forgot to login before posting -- georg]

(in reply to: ↑ 11 ; follow-up: ↓ 13 ) 09/29/08 20:31:23 changed by thisisrv

Is there a way to get around this? i am still stuck up with that problem!

(in reply to: ↑ 12 ) 09/30/08 15:38:33 changed by georg

  • pending changed.

Replying to thisisrv:

Is there a way to get around this? i am still stuck up with that problem!

Use a lower number of wifi cards, wifi cards which consume less power, or a better power supply? Sorry, I can't tell for sure.

10/19/09 14:46:41 changed by il_gregg

I have the same problem... no way to solve it? Thanks