Hi,
ath5sudo k don't work with my Notebook Wifi Card
sudo lspci -v
14:00.0 Ethernet controller: Atheros Communications, Inc. AR5006EG 802.=
11 b/g
Wireless PCI Express Adapter (rev 01)
=A0 =A0 =A0 =A0 Subsystem: Atheros Communications, Inc. Unknown device =
3067
=A0 =A0 =A0 =A0 Flags: fast devsel, IRQ 19
=A0 =A0 =A0 =A0 Memory at fe200000 (64-bit, non-prefetchable) [size=3D6=
4K]
=A0 =A0 =A0 =A0 Capabilities: [40] Power Management version 2
=A0 =A0 =A0 =A0 Capabilities: [50] Message Signalled Interrupts: Mask- =
64bit-=20
Queue=3D0/0
Enable-
=A0 =A0 =A0 =A0 Capabilities: [60] Express Legacy Endpoint IRQ 0
=A0 =A0 =A0 =A0 Capabilities: [90] MSI-X: Enable- Mask- TabSize=3D1
- in Windows XP is detected as AR5006X
dmesg
[ 1276.969782] ACPI: PCI Interrupt 0000:14:00.0[A] -> GSI 19 (level, lo=
w) ->=20
IRQ
19
[ 1276.969811] PCI: Setting latency timer of device 0000:14:00.0 to 64
[ 1277.982089] ath5k_hw_nic_wakeup: failed to resume the MAC Chip
[ 1277.984330] ACPI: PCI interrupt for device 0000:14:00.0 disabled
[ 1277.984399] ath5k_pci: probe of 0000:14:00.0 failed with error -5
I use Kubuntu 7.10 Gutsy
I Downloadet ath5k from
http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2=
=2E6.tar.bz
2 on 22.11.2007
Marian Virec
----------
AKCNA ZLAVA na Zoner Photo Studio 9 - viacej na - http://www.zps9.sk/
Hmm you might be right about this, ath_info would really help. Problem
is we din't have such reports about nic_wakeup before, not even on
2425 (AR5007EG). In all cases (except 5416/5418) at least attach
worked ! In this case attach doesn't work at all ;-(
This might be the new 2417 chip (AR5007G), if they had faulty pci id
for 5007EG they might have done the same for 5007G (but 2417 is not
pci-express !).
Majo can you plz do what Mike said and give us a dump of ath_info tool ?
Thanx in advance...
2007/11/23, Michael Renzmann <[email protected]>:
> Hi.
>
> >> 14:00.0 Ethernet controller: Atheros Communications, Inc. AR5006EG
> >> 802.11 b/g Wireless PCI Express Adapter (rev 01)
> > This is a 2424 pci-express card, we don't support this chip yet ;-(
>
> Don't be fooled by the PCI ID - we already know that these are not
> reliable to identify the chipset used on a Atheros-based card. See also
> http://madwifi.org/ticket/1192.
>
> I'd suggest Majo to use ath_info from MadWifi trunk, hopefully it allows
> to recognize the chipset more reliably in his case, too. The ticket
> mentioned above also has some instructions on how to use this tool.
>
> > (the testbed stuff would really help)
>
> Working on it, but it's still a long road. Getting non-profit status is
> just the first step for that, and that's work in progress.
>
> Bye, Mike
>
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
> Don't be fooled by the PCI ID - we already know that these are
> not reliable to identify the chipset used on a Atheros-based
> card. See also http://madwifi.org/ticket/1192.
While I was working with Windows XP Embedded (after a long
absence from Windows), I noticed that they don't just work with
PCI IDs, but also with PCI-Subsystem IDs.
$ lspci -vn
03:00.0 0200: 168c:0013 (rev 01)
Subsystem: 1186:3a12
...
If vendors would make the subsystem ID wrong, then even Windows
XP wouldn't be able to detect the card correctly. That's quite a
big incentive to get that number right :-)
Hi.
>> 14:00.0 Ethernet controller: Atheros Communications, Inc. AR5006EG
>> 802.11 b/g Wireless PCI Express Adapter (rev 01)
> This is a 2424 pci-express card, we don't support this chip yet ;-(
Don't be fooled by the PCI ID - we already know that these are not
reliable to identify the chipset used on a Atheros-based card. See also
http://madwifi.org/ticket/1192.
I'd suggest Majo to use ath_info from MadWifi trunk, hopefully it allows
to recognize the chipset more reliably in his case, too. The ticket
mentioned above also has some instructions on how to use this tool.
> (the testbed stuff would really help)
Working on it, but it's still a long road. Getting non-profit status is
just the first step for that, and that's work in progress.
Bye, Mike
2007/11/23, Majo <[email protected]>:
> Hi,
>
> ath5sudo k don't work with my Notebook Wifi Card
>
> sudo lspci -v
>
> 14:00.0 Ethernet controller: Atheros Communications, Inc. AR5006EG 802.11 b/g
> Wireless PCI Express Adapter (rev 01)
This is a 2424 pci-express card, we don't support this chip yet ;-(
(damn i thought that was only the pci express version of 5414 that we
just fixed, seems there's more to do)
> Subsystem: Atheros Communications, Inc. Unknown device 3067
> Flags: fast devsel, IRQ 19
> Memory at fe200000 (64-bit, non-prefetchable) [size=64K]
> Capabilities: [40] Power Management version 2
> Capabilities: [50] Message Signalled Interrupts: Mask- 64bit-
> Queue=0/0
> Enable-
> Capabilities: [60] Express Legacy Endpoint IRQ 0
> Capabilities: [90] MSI-X: Enable- Mask- TabSize=1
>
> - in Windows XP is detected as AR5006X
>
> dmesg
>
> [ 1276.969782] ACPI: PCI Interrupt 0000:14:00.0[A] -> GSI 19 (level, low) ->
> IRQ
> 19
> [ 1276.969811] PCI: Setting latency timer of device 0000:14:00.0 to 64
> [ 1277.982089] ath5k_hw_nic_wakeup: failed to resume the MAC Chip
Hmm it seems that 5424/2424 and later chips have a different reset
sequence we are not aware of. We 'll need traces to make it work, if
you have some time plz take a look here ->
http://madwifi.org/wiki/DevDocs/MadwifiTrace (the testbed stuff would
really help)
Thanx for the feedback ;-)
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
On Friday 23 November 2007 11:54:56 Holger Schurig wrote:
> > Don't be fooled by the PCI ID - we already know that these
> > are not reliable to identify the chipset used on a
> > Atheros-based card. See also http://madwifi.org/ticket/1192.
Hmm, I also notices that in ticket 1192, neither the output
of "lspci" nor "lshw" doesn't even output the subsystem ID. The
MODULE_ALIAS() syntax doesn't seem to have a way to specify
subsystem IDs.
I have two Atheros based cards here, one is a D-Link DWL-G650,
one a Proxim 8470-WD. Both have the PCI ID 168C:0013. The
D-Link has the subsystem ID 1186:3a12, the Proxim has 14b7:0a40.
The driver for the Proxim card contains an INF file, which has
this section:
[Vendor]
; DisplayName Section DeviceID
; ----------- ------- --------
%LEX.DeviceDesc.CB22ag_G% = LEX_CB22ag_01, PCI\VEN_168C&DEV_0013&SUBSYS_0A1014B7
%LEX.DeviceDesc.CB22ag_S% = LEX_CB22ag_02, PCI\VEN_168C&DEV_0013&SUBSYS_AA1014B7
%LEX.DeviceDesc.CB22ab_G% = LEX_CB22ab_01, PCI\VEN_168C&DEV_0012&SUBSYS_0A001B47
%LEX.DeviceDesc.CB22ab_S% = LEX_CB22ab_02, PCI\VEN_168C&DEV_0012&SUBSYS_AA001B47
%LEX.DeviceDesc.CB22ab_A% = LEX_CB22ab_03, PCI\VEN_168C&DEV_0012&SUBSYS_1A001B47
%LEX.DeviceDesc.CB22g_G% = LEX_CB22g_01, PCI\VEN_168C&DEV_0013&SUBSYS_0A4014B7
%LEX.DeviceDesc.CB22g_S% = LEX_CB22g_02, PCI\VEN_168C&DEV_0013&SUBSYS_AA4014B7
%LEX.DeviceDesc.MB22ag% = LEX_MB22ag_01, PCI\VEN_168C&DEV_0013&SUBSYS_0A6014B7
%LEX.DeviceDesc.ORiNOCO% = LEX_ORiNOCO, PCI\VEN_168C&DEV_0013&SUBSYS_04101467
And, as it turns out, the driver for the Proxim card doesn't
recognize the D-Link card, because it looks not only at the
PCI-ID, but also at the subsystem ID.
It would be interesting to get a list of the subsystem IDs of
misdetected atheros-based WLAN cards.
HI,
here are some Traces.
I dowloadet hte new snv from:
svn co http://svn.madwifi.org/madwifi/branches/madwifi-trace
make
sudo make install
Then i run the trase script from
svn co http://svn.madwifi.org/tools/trace
The Result is in Attachement.
I hope this can help you.
Majo
Dňa Pi 23. November 2007 00:38 Nick Kossifidis napísal:
> 2007/11/23, Majo <[email protected]>:
> > Hi,
> >
> > ath5sudo k don't work with my Notebook Wifi Card
> >
> > sudo lspci -v
> >
> > 14:00.0 Ethernet controller: Atheros Communications, Inc. AR5006EG 802.11
> > b/g Wireless PCI Express Adapter (rev 01)
>
> This is a 2424 pci-express card, we don't support this chip yet ;-(
>
> (damn i thought that was only the pci express version of 5414 that we
> just fixed, seems there's more to do)
>
> > Subsystem: Atheros Communications, Inc. Unknown device 3067
> > Flags: fast devsel, IRQ 19
> > Memory at fe200000 (64-bit, non-prefetchable) [size=64K]
> > Capabilities: [40] Power Management version 2
> > Capabilities: [50] Message Signalled Interrupts: Mask- 64bit-
> > Queue=0/0
> > Enable-
> > Capabilities: [60] Express Legacy Endpoint IRQ 0
> > Capabilities: [90] MSI-X: Enable- Mask- TabSize=1
> >
> > - in Windows XP is detected as AR5006X
> >
> > dmesg
> >
> > [ 1276.969782] ACPI: PCI Interrupt 0000:14:00.0[A] -> GSI 19 (level, low)
> > -> IRQ
> > 19
> > [ 1276.969811] PCI: Setting latency timer of device 0000:14:00.0 to 64
> > [ 1277.982089] ath5k_hw_nic_wakeup: failed to resume the MAC Chip
>
> Hmm it seems that 5424/2424 and later chips have a different reset
> sequence we are not aware of. We 'll need traces to make it work, if
> you have some time plz take a look here ->
> http://madwifi.org/wiki/DevDocs/MadwifiTrace (the testbed stuff would
> really help)
>
>
> Thanx for the feedback ;-)
----------
AKCIOVA PONUKA - Zoner Photo Studio + digitalna klucenka. Viacej na http://www.zps9.sk/