[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

Hi,


does anyone know whether 168c:ff1c can be supported by the current ath9k
driver? It isn't listed with the PCI IDs in the source. I bought it off
eBay as "Apple" AR5008. It is a PCI Express card with 3 Antenna
connectors and lots of Apple stampings on it. lspci says:


02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)
        Physical Slot: 35
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 32 bytes
        Interrupt: pin A routed to IRQ 3
        Region 0: Memory@fddf0000 (64-bit, non-prefetchable) [size=64K]
        Capabilities: [40] Power Management version 2
                Flags: PMEClk- DSI- D1+ D2- AuxCurrent=375mA PME(D0+,D1+,D2-,D3hot+,D3cold-)
                Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit-
                Address: 00000000  Data: 0000
        Capabilities: [60] Express (v1) Legacy Endpoint, MSI 00
                DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <128ns, L1 <2us
                        ExtTag- AttnBtn- AttnInd- PwrInd- RBE- FLReset-
                DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
                        RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop-
                        MaxPayload 128 bytes, MaxReadReq 512 bytes
                DevSta: CorrErr- UncorrErr+ FatalErr- UnsuppReq+ AuxPwr- TransPend-
                LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Latency L0 <512ns, L1 <64us
                        ClockPM- Surprise- LLActRep- BwNot-
                LnkCtl: ASPM Disabled; RCB 128 bytes Disabled- Retrain- CommClk+
                        ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
                LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
        Capabilities: [90] MSI-X: Enable- Count=1 Masked-
                Vector table: BAR=0 offset=00000000
                PBA: BAR=0 offset=00000000
        Capabilities: [100 v1] Advanced Error Reporting
                UESta:  DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq+ ACSViol-
                UEMsk:  DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
                UESvrt: DLP+ SDES- TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
                CESta:  RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
                CEMsk:  RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
                AERCap: First Error Pointer: 14, GenCap+ CGenEn- ChkCap+ ChkEn-
        Capabilities: [140 v1] Virtual Channel
                Caps:   LPEVC=0 RefClk=100ns PATEntryBits=1
                Arb:    Fixed- WRR32- WRR64- WRR128-
                Ctrl:   ArbSelect=Fixed
                Status: InProgress-
                VC0:    Caps:   PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
                        Arb:    Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
                        Ctrl:   Enable+ ID=0 ArbSelect=Fixed TC/VC=01
                        Status: NegoPending- InProgress-

when I add the PCI ID to the driver, i get:
[  287.826320] ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[  287.826336] ath9k 0000:02:00.0: setting latency timer to 64
[  287.830918] ath9k 0000:02:00.0: Failed to initialize device
[  287.830944] ath9k 0000:02:00.0: PCI INT A disabled

(that is with debug=0xffffff, no idea how to get more output)


-David

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Sun, Sep 26, 2010 at 03:39:35PM +0200, David Lamparter wrote:
> 02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)
[...]
> 
> when I add the PCI ID to the driver, i get:
> [  287.826320] ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
> [  287.826336] ath9k 0000:02:00.0: setting latency timer to 64
> [  287.830918] ath9k 0000:02:00.0: Failed to initialize device
> [  287.830944] ath9k 0000:02:00.0: PCI INT A disabled
> 
> (that is with debug=0xffffff, no idea how to get more output)

I managed to get more debug output. ath9k_hw_chip_test fails:
"address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab"

when i disable hw_chip_test, the output is:
[ 2667.283596] ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[ 2667.283611] ath9k 0000:02:00.0: setting latency timer to 64
[ 2667.286328] ath: Using ANI v1
[ 2667.286331] ath: UNDEFINED -> AWAKE
[ 2667.286392] ath: serialize_regmode is 0
[ 2667.401925] ath: Read Magic = 0xFFFF
[ 2667.401928] ath: Invalid EEPROM Magic. Endianness mismatch.
[ 2667.401935] ath: Unable to initialize hardware; initialization status: -22
[ 2667.401947] ath9k 0000:02:00.0: Failed to initialize device
[ 2667.401980] ath9k 0000:02:00.0: PCI INT A disabled
[ 2667.402005] ath9k: probe of 0000:02:00.0 failed with error -22


-David

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Alexander Simon]

Am Sonntag, 26. September 2010, 15:39:35 schrieb David Lamparter:
> Hi,
> 
> 
> does anyone know whether 168c:ff1c can be supported by the current ath9k
> driver? It isn't listed with the PCI IDs in the source. I bought it off
> eBay as "Apple" AR5008. It is a PCI Express card with 3 Antenna
> connectors and lots of Apple stampings on it. lspci says:
> 
Hm.
I think ath9k should support this card...
About two weeks ago, cwr at netcom.co.uk asked about AR5008 and CRDA, so it must 
have been running at least for him.

Seems this problem has been before:
http://www.mail-archive.com/ath9k-devel at lists.ath9k.org/msg02332.html

Maybe is also a broken EEPROM?
You should try to restore it (i dont know if you actually can do that) or 
patch ath9k to ignore the EEPROM (again, i believe this should work, if you 
know the chipset revision, regdomain etc).

Or try madwifi, i think it also supports the AR5008...

Alex

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Mon, Sep 27, 2010 at 04:43:58PM +0200, Alexander Simon wrote:
> Am Sonntag, 26. September 2010, 15:39:35 schrieb David Lamparter:
> > Hi,
> > 
> > 
> > does anyone know whether 168c:ff1c can be supported by the current ath9k
> > driver? It isn't listed with the PCI IDs in the source. I bought it off
> > eBay as "Apple" AR5008. It is a PCI Express card with 3 Antenna
> > connectors and lots of Apple stampings on it. lspci says:
> > 
> About two weeks ago, cwr at netcom.co.uk asked about AR5008 and CRDA, so it must 
> have been running at least for him.

"AR5008" seems to be a label that is sticked onto different cards...
:0023 and :0024 seem to be regular product IDs.

> Seems this problem has been before:
> http://www.mail-archive.com/ath9k-devel at lists.ath9k.org/msg02332.html
> 
> Maybe is also a broken EEPROM?

No. For now, I'm assuming the eeprom is fine; the Mac OS X driver lists
:ff1c and :ff1d as product IDs for Atheros cards, so this is NOT an
indication for a broken card.

> Or try madwifi, i think it also supports the AR5008...

Hmm. This is a 11n card. I don't think madwifi supports 11n cards?


-David

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Pavel Roskin]

On 09/27/2010 11:16 AM, David Lamparter wrote:

>> Or try madwifi, i think it also supports the AR5008...
>
> Hmm. This is a 11n card. I don't think madwifi supports 11n cards?

Yes, MadWifi trunk supports AR5008 based cards, but there is no support 
for 802.11n.

-- 
Regards,
Pavel Roskin

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Alexander Simon]

Hi david,

> No. For now, I'm assuming the eeprom is fine; the Mac OS X driver lists
> 
> :ff1c and :ff1d as product IDs for Atheros cards, so this is NOT an
> 
> indication for a broken card.
Ok, i agree.
> 
> > Or try madwifi, i think it also supports the AR5008...
> 
> Hmm. This is a 11n card. I don't think madwifi supports 11n cards?
> 
> 
> -David

Hmm.. Just an idea:
Maybe to be compliant with PowerPC, apple converted the eeprom data to big 
endian. Our x86/x64 Linuxes have little endian.
So the "endian mismatch" message could be correct and the ath9k eeprom 
function just doesnt know to handle the big endian data.

If it runs with Madwifi, then its eeprom function should be able to cope with 
the other endianness, and one should be able to adapt this to ath9k...
So: another reason to try it with the madwifi or have a look at its sources :)

Alex

[ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

Note: this e-mail is on a public mailing list.

On Sun, Sep 26, 2010 at 6:39 AM, David Lamparter <equinox@diac24.net> wrote:
> Hi,
>
>
> does anyone know whether 168c:ff1c can be supported by the current ath9k
> driver?

Do you *always* get this PCI device ID upon bootup? I have heard of
some cases some older ath5k devices would bootup with some odd PCI
device IDs and the only way to support the devices was to add the
busted PCI device ID to the device ID table. I have not heard of this
for ath9k cards but your PCI ID seems iffy.

David Quan, have you heard of this PCI Device ID before for our 11n
hardware? How about issues with the device ID being read out from the
EEPROM?

> It isn't listed with the PCI IDs in the source. I bought it off
> eBay as "Apple" AR5008. It is a PCI Express card with 3 Antenna
> connectors and lots of Apple stampings on it. lspci says:
>
>
> 02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)
> ? ? ? ?Physical Slot: 35
> ? ? ? ?Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
> ? ? ? ?Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
> ? ? ? ?Latency: 0, Cache Line Size: 32 bytes
> ? ? ? ?Interrupt: pin A routed to IRQ 3
> ? ? ? ?Region 0: Memory at fddf0000 (64-bit, non-prefetchable) [size=64K]
> ? ? ? ?Capabilities: [40] Power Management version 2
> ? ? ? ? ? ? ? ?Flags: PMEClk- DSI- D1+ D2- AuxCurrent=375mA PME(D0+,D1+,D2-,D3hot+,D3cold-)
> ? ? ? ? ? ? ? ?Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
> ? ? ? ?Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit-
> ? ? ? ? ? ? ? ?Address: 00000000 ?Data: 0000
> ? ? ? ?Capabilities: [60] Express (v1) Legacy Endpoint, MSI 00
> ? ? ? ? ? ? ? ?DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <128ns, L1 <2us
> ? ? ? ? ? ? ? ? ? ? ? ?ExtTag- AttnBtn- AttnInd- PwrInd- RBE- FLReset-
> ? ? ? ? ? ? ? ?DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
> ? ? ? ? ? ? ? ? ? ? ? ?RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop-
> ? ? ? ? ? ? ? ? ? ? ? ?MaxPayload 128 bytes, MaxReadReq 512 bytes
> ? ? ? ? ? ? ? ?DevSta: CorrErr- UncorrErr+ FatalErr- UnsuppReq+ AuxPwr- TransPend-
> ? ? ? ? ? ? ? ?LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Latency L0 <512ns, L1 <64us
> ? ? ? ? ? ? ? ? ? ? ? ?ClockPM- Surprise- LLActRep- BwNot-
> ? ? ? ? ? ? ? ?LnkCtl: ASPM Disabled; RCB 128 bytes Disabled- Retrain- CommClk+
> ? ? ? ? ? ? ? ? ? ? ? ?ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
> ? ? ? ? ? ? ? ?LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
> ? ? ? ?Capabilities: [90] MSI-X: Enable- Count=1 Masked-
> ? ? ? ? ? ? ? ?Vector table: BAR=0 offset=00000000
> ? ? ? ? ? ? ? ?PBA: BAR=0 offset=00000000
> ? ? ? ?Capabilities: [100 v1] Advanced Error Reporting
> ? ? ? ? ? ? ? ?UESta: ?DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq+ ACSViol-
> ? ? ? ? ? ? ? ?UEMsk: ?DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
> ? ? ? ? ? ? ? ?UESvrt: DLP+ SDES- TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
> ? ? ? ? ? ? ? ?CESta: ?RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
> ? ? ? ? ? ? ? ?CEMsk: ?RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
> ? ? ? ? ? ? ? ?AERCap: First Error Pointer: 14, GenCap+ CGenEn- ChkCap+ ChkEn-
> ? ? ? ?Capabilities: [140 v1] Virtual Channel
> ? ? ? ? ? ? ? ?Caps: ? LPEVC=0 RefClk=100ns PATEntryBits=1
> ? ? ? ? ? ? ? ?Arb: ? ?Fixed- WRR32- WRR64- WRR128-
> ? ? ? ? ? ? ? ?Ctrl: ? ArbSelect=Fixed
> ? ? ? ? ? ? ? ?Status: InProgress-
> ? ? ? ? ? ? ? ?VC0: ? ?Caps: ? PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
> ? ? ? ? ? ? ? ? ? ? ? ?Arb: ? ?Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
> ? ? ? ? ? ? ? ? ? ? ? ?Ctrl: ? Enable+ ID=0 ArbSelect=Fixed TC/VC=01
> ? ? ? ? ? ? ? ? ? ? ? ?Status: NegoPending- InProgress-
>
> when I add the PCI ID to the driver, i get:
> [ ?287.826320] ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
> [ ?287.826336] ath9k 0000:02:00.0: setting latency timer to 64
> [ ?287.830918] ath9k 0000:02:00.0: Failed to initialize device
> [ ?287.830944] ath9k 0000:02:00.0: PCI INT A disabled
>
> (that is with debug=0xffffff, no idea how to get more output)

In order to get debug prints you must have ATH_DEBUG enabled on your
kernel, do you have that?

  Luis

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Mathieu Olivari]

Hi Luis,

You're talking about ath5k, but actually AR5008 refers to our first 11n chip family.
Therefore, you would probably have more chances to make it work using ath9k than ath5k.
I checked in our code, and ff1c refers to "owl emulation  device id". Therefore, it will probably work out of the box with the 11n driver.

Hope this helps :)
Mathieu


-----Original Message-----
From: atheros.eng.projects.ath9k-devel-bounces@atheros.com [mailto:atheros.eng.projects.ath9k-devel-bounces at atheros.com] On Behalf Of Luis R. Rodriguez
Sent: Monday, September 27, 2010 7:19 PM
To: David Lamparter
Cc: atheros.eng.projects.ath9k-devel at mailman.users.atheros.com; David Quan; ath9k-devel at venema.h4ckr.net
Subject: Re: [internal-ath9k-devel] [ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

Note: this e-mail is on a public mailing list.

On Sun, Sep 26, 2010 at 6:39 AM, David Lamparter <equinox@diac24.net> wrote:
> Hi,
>
>
> does anyone know whether 168c:ff1c can be supported by the current ath9k
> driver?

Do you *always* get this PCI device ID upon bootup? I have heard of
some cases some older ath5k devices would bootup with some odd PCI
device IDs and the only way to support the devices was to add the
busted PCI device ID to the device ID table. I have not heard of this
for ath9k cards but your PCI ID seems iffy.

David Quan, have you heard of this PCI Device ID before for our 11n
hardware? How about issues with the device ID being read out from the
EEPROM?

> It isn't listed with the PCI IDs in the source. I bought it off
> eBay as "Apple" AR5008. It is a PCI Express card with 3 Antenna
> connectors and lots of Apple stampings on it. lspci says:
>
>
> 02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)
> ? ? ? ?Physical Slot: 35
> ? ? ? ?Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
> ? ? ? ?Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
> ? ? ? ?Latency: 0, Cache Line Size: 32 bytes
> ? ? ? ?Interrupt: pin A routed to IRQ 3
> ? ? ? ?Region 0: Memory at fddf0000 (64-bit, non-prefetchable) [size=64K]
> ? ? ? ?Capabilities: [40] Power Management version 2
> ? ? ? ? ? ? ? ?Flags: PMEClk- DSI- D1+ D2- AuxCurrent=375mA PME(D0+,D1+,D2-,D3hot+,D3cold-)
> ? ? ? ? ? ? ? ?Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
> ? ? ? ?Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit-
> ? ? ? ? ? ? ? ?Address: 00000000 ?Data: 0000
> ? ? ? ?Capabilities: [60] Express (v1) Legacy Endpoint, MSI 00
> ? ? ? ? ? ? ? ?DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s <128ns, L1 <2us
> ? ? ? ? ? ? ? ? ? ? ? ?ExtTag- AttnBtn- AttnInd- PwrInd- RBE- FLReset-
> ? ? ? ? ? ? ? ?DevCtl: Report errors: Correctable- Non-Fatal- Fatal- Unsupported-
> ? ? ? ? ? ? ? ? ? ? ? ?RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop-
> ? ? ? ? ? ? ? ? ? ? ? ?MaxPayload 128 bytes, MaxReadReq 512 bytes
> ? ? ? ? ? ? ? ?DevSta: CorrErr- UncorrErr+ FatalErr- UnsuppReq+ AuxPwr- TransPend-
> ? ? ? ? ? ? ? ?LnkCap: Port #0, Speed 2.5GT/s, Width x1, ASPM L0s L1, Latency L0 <512ns, L1 <64us
> ? ? ? ? ? ? ? ? ? ? ? ?ClockPM- Surprise- LLActRep- BwNot-
> ? ? ? ? ? ? ? ?LnkCtl: ASPM Disabled; RCB 128 bytes Disabled- Retrain- CommClk+
> ? ? ? ? ? ? ? ? ? ? ? ?ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
> ? ? ? ? ? ? ? ?LnkSta: Speed 2.5GT/s, Width x1, TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
> ? ? ? ?Capabilities: [90] MSI-X: Enable- Count=1 Masked-
> ? ? ? ? ? ? ? ?Vector table: BAR=0 offset=00000000
> ? ? ? ? ? ? ? ?PBA: BAR=0 offset=00000000
> ? ? ? ?Capabilities: [100 v1] Advanced Error Reporting
> ? ? ? ? ? ? ? ?UESta: ?DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq+ ACSViol-
> ? ? ? ? ? ? ? ?UEMsk: ?DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- MalfTLP- ECRC- UnsupReq- ACSViol-
> ? ? ? ? ? ? ? ?UESvrt: DLP+ SDES- TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ MalfTLP+ ECRC- UnsupReq- ACSViol-
> ? ? ? ? ? ? ? ?CESta: ?RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
> ? ? ? ? ? ? ? ?CEMsk: ?RxErr- BadTLP- BadDLLP- Rollover- Timeout- NonFatalErr-
> ? ? ? ? ? ? ? ?AERCap: First Error Pointer: 14, GenCap+ CGenEn- ChkCap+ ChkEn-
> ? ? ? ?Capabilities: [140 v1] Virtual Channel
> ? ? ? ? ? ? ? ?Caps: ? LPEVC=0 RefClk=100ns PATEntryBits=1
> ? ? ? ? ? ? ? ?Arb: ? ?Fixed- WRR32- WRR64- WRR128-
> ? ? ? ? ? ? ? ?Ctrl: ? ArbSelect=Fixed
> ? ? ? ? ? ? ? ?Status: InProgress-
> ? ? ? ? ? ? ? ?VC0: ? ?Caps: ? PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
> ? ? ? ? ? ? ? ? ? ? ? ?Arb: ? ?Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
> ? ? ? ? ? ? ? ? ? ? ? ?Ctrl: ? Enable+ ID=0 ArbSelect=Fixed TC/VC=01
> ? ? ? ? ? ? ? ? ? ? ? ?Status: NegoPending- InProgress-
>
> when I add the PCI ID to the driver, i get:
> [ ?287.826320] ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
> [ ?287.826336] ath9k 0000:02:00.0: setting latency timer to 64
> [ ?287.830918] ath9k 0000:02:00.0: Failed to initialize device
> [ ?287.830944] ath9k 0000:02:00.0: PCI INT A disabled
>
> (that is with debug=0xffffff, no idea how to get more output)

In order to get debug prints you must have ATH_DEBUG enabled on your
kernel, do you have that?

  Luis

_______________________________________________
Atheros.eng.projects.ath9k-devel mailing list
Atheros.eng.projects.ath9k-devel at mailman.users.atheros.com
http://mailman/mailman/listinfo/atheros.eng.projects.ath9k-devel

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

On Mon, Sep 27, 2010 at 10:28 AM, Mathieu Olivari
<Mathieu.Olivari@atheros.com> wrote:
> Hi Luis,
>
> You're talking about ath5k, but actually AR5008 refers to our first 11n chip family.
> Therefore, you would probably have more chances to make it work using ath9k than ath5k.

Well I referred to ath5k only because I had heard of those type of
devices having some issues with the PCI device ID.

> I checked in our code, and ff1c refers to "owl emulation ?device id". Therefore, it will probably work out of the box with the 11n driver.

That is strikingly odd. The thing is the user (CC'd and again, this
e-mail is on a public mailing list) gets a failure upon
initialization.

David - what kernel did you try with? Can you try compat-wireless-2.6.36-rc3-1 ?

http://wireless.kernel.org/en/users/Download/stable/

  Luis

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Mon, Sep 27, 2010 at 10:40:22AM -0700, Luis R. Rodriguez wrote:
> On Mon, Sep 27, 2010 at 10:28 AM, Mathieu Olivari
> <Mathieu.Olivari@atheros.com> wrote:
> > You're talking about ath5k, but actually AR5008 refers to our first 11n chip family.
> > Therefore, you would probably have more chances to make it work using ath9k than ath5k.
> 
> Well I referred to ath5k only because I had heard of those type of
> devices having some issues with the PCI device ID.

The device reliably comes up as ff1c. In the meantime i noticed another
odd problem: on booting up, the kernel emits the following message
exactly once:

[   21.548449] ath9k 0000:02:00.0: Refused to change power state, currently in D1

this comes immediately before the remainder of the (failing)
initialization. However, rmmod/modprobe ath9k does not cause it to be
emitted again. Basically this message is once per boot; I will reboot
later and check if it re-appears after a reboot.

Is it possible the device needs some kind of magic to power up / leave
standby correctly?

> > I checked in our code, and ff1c refers to "owl emulation ?device id". Therefore, it will probably work out of the box with the 11n driver.

owl emulation?

> That is strikingly odd. The thing is the user (CC'd and again, this
> e-mail is on a public mailing list) gets a failure upon
> initialization.
> 
> David - what kernel did you try with? Can you try compat-wireless-2.6.36-rc3-1 ?

I'm on wireless-2.6 HEAD 78b110 (3 days old, 2.6.36-rc5-wl+, i assume
that includes all the newest drivers?


-David


P.S.: to re-cite the progress from my other mail:
I managed to get more debug output. ath9k_hw_chip_test fails:

address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab

_after_ i disable hw_chip_test in the source code, the output is:

ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
ath9k 0000:02:00.0: setting latency timer to 64
ath: Using ANI v1
ath: UNDEFINED -> AWAKE
ath: serialize_regmode is 0
ath: Read Magic = 0xFFFF
ath: Invalid EEPROM Magic. Endianness mismatch.
ath: Unable to initialize hardware; initialization status: -22
ath9k 0000:02:00.0: Failed to initialize device
ath9k 0000:02:00.0: PCI INT A disabled
ath9k: probe of 0000:02:00.0 failed with error -22

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

On Mon, Sep 27, 2010 at 12:54 PM, David Lamparter <equinox@diac24.net> wrote:
> On Mon, Sep 27, 2010 at 10:40:22AM -0700, Luis R. Rodriguez wrote:
>> On Mon, Sep 27, 2010 at 10:28 AM, Mathieu Olivari
>> <Mathieu.Olivari@atheros.com> wrote:
>> > You're talking about ath5k, but actually AR5008 refers to our first 11n chip family.
>> > Therefore, you would probably have more chances to make it work using ath9k than ath5k.
>>
>> Well I referred to ath5k only because I had heard of those type of
>> devices having some issues with the PCI device ID.
>
> The device reliably comes up as ff1c. In the meantime i noticed another
> odd problem: on booting up, the kernel emits the following message
> exactly once:
>
> [ ? 21.548449] ath9k 0000:02:00.0: Refused to change power state, currently in D1
>
> this comes immediately before the remainder of the (failing)
> initialization. However, rmmod/modprobe ath9k does not cause it to be
> emitted again. Basically this message is once per boot; I will reboot
> later and check if it re-appears after a reboot.
>
> Is it possible the device needs some kind of magic to power up / leave
> standby correctly?
>
>> > I checked in our code, and ff1c refers to "owl emulation ?device id". Therefore, it will probably work out of the box with the 11n driver.
>
> owl emulation?
>
>> That is strikingly odd. The thing is the user (CC'd and again, this
>> e-mail is on a public mailing list) gets a failure upon
>> initialization.
>>
>> David - what kernel did you try with? Can you try compat-wireless-2.6.36-rc3-1 ?
>
> I'm on wireless-2.6 HEAD 78b110 (3 days old, 2.6.36-rc5-wl+, i assume
> that includes all the newest drivers?

that's fine

> -David
>
>
> P.S.: to re-cite the progress from my other mail:
> I managed to get more debug output. ath9k_hw_chip_test fails:
>
> address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab


could be  busted card, does the card work with another OS? If so when
did you last try?

  Luis

> _after_ i disable hw_chip_test in the source code, the output is:
>
> ath9k 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
> ath9k 0000:02:00.0: setting latency timer to 64
> ath: Using ANI v1
> ath: UNDEFINED -> AWAKE
> ath: serialize_regmode is 0
> ath: Read Magic = 0xFFFF
> ath: Invalid EEPROM Magic. Endianness mismatch.
> ath: Unable to initialize hardware; initialization status: -22
> ath9k 0000:02:00.0: Failed to initialize device
> ath9k 0000:02:00.0: PCI INT A disabled
> ath9k: probe of 0000:02:00.0 failed with error -22

If the chip test fails, better not keep trying.

  Luis

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Mon, Sep 27, 2010 at 12:58:30PM -0700, Luis R. Rodriguez wrote:
> > address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab
> 
> could be  busted card, does the card work with another OS? If so when
> did you last try?

I don't have any other OS at my disposal currently; if I can't get it to
work I will ask around if I find someone with a Mini-PCIe slot and
Windows / Mac OS X.

> If the chip test fails, better not keep trying.

I've disabled it because it already was disabled for the 9271, so I
guessed my card might need the same workaround.

The chip test also checks various revision values, could it be that it
is assuming the wrong revision for this card?


-David

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

On Mon, Sep 27, 2010 at 01:08:39PM -0700, David Lamparter wrote:
> On Mon, Sep 27, 2010 at 12:58:30PM -0700, Luis R. Rodriguez wrote:
> > > address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab
> > 
> > could be  busted card, does the card work with another OS? If so when
> > did you last try?
> 
> I don't have any other OS at my disposal currently; if I can't get it to
> work I will ask around if I find someone with a Mini-PCIe slot and
> Windows / Mac OS X.

My guess so far based on your debug logs is the card is bust.

> > If the chip test fails, better not keep trying.
> 
> I've disabled it because it already was disabled for the 9271, so I
> guessed my card might need the same workaround.
> 
> The chip test also checks various revision values, could it be that it
> is assuming the wrong revision for this card?

chip tests runs prior to eeprom checksum, my guess is the eeprom
checksum will fail too and your card is just bust.

  Luis

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Mon, Sep 27, 2010 at 01:25:02PM -0700, Luis R. Rodriguez wrote:
> On Mon, Sep 27, 2010 at 01:08:39PM -0700, David Lamparter wrote:
> > On Mon, Sep 27, 2010 at 12:58:30PM -0700, Luis R. Rodriguez wrote:
> > > > address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab
> > > 
> > > could be  busted card, does the card work with another OS? If so when
> > > did you last try?
> > 
> > I don't have any other OS at my disposal currently; if I can't get it to
> > work I will ask around if I find someone with a Mini-PCIe slot and
> > Windows / Mac OS X.
> 
> My guess so far based on your debug logs is the card is bust.

Even assuming the card is bust, there are two atheros device IDs - ff1c
and ff1d - that designate AR5008 cards (according to pci.ids and google)
and that are not supported by any driver currently.

Are these cards driver compatible to the 0023/0024 devices? Do they need
different workarounds?


-David

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[David Lamparter]

On Mon, Sep 27, 2010 at 10:44:47PM +0200, David Lamparter wrote:
>                                  there are two atheros device IDs - ff1c
> and ff1d - that designate AR5008 cards (according to pci.ids and google)
> and that are not supported by any driver currently.
> 
> Are these cards driver compatible to the 0023/0024 devices? Do they need
> different workarounds?

On Sun, Sep 26, 2010 at 03:39:35PM +0200, David Lamparter wrote:
> 02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)

I've now tried madwifi and ndiswrapper. To sum up:

- ath9k: ath9k_hw_chip_test fails:
  (ath9k modified to add PCI ID 168c:ff1c)
  "address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab"

- madwifi trunk:
  (madwifi modified to add PCI ID 168c:ff1c)
  "MadWifi: unable to attach hardware: '' (HAL status 1434461696)
  (1434461696 = 0x55802600)

- ndiswrapper with 7.7.0.523 WinXP 64-bit driver:
  (driver lists 168c:ff1c in .inf file, no modification)
  ndiswrapper (link_pe_images:566): fixing KI_USER_SHARED_DATA address in the driver
  ndiswrapper: driver netathwx (,06/04/2010,7.7.0.523) loaded
  ndiswrapper 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
  ndiswrapper 0000:02:00.0: setting latency timer to 64
  ndiswrapper (NdisWriteErrorLogEntry:190): log: C000138B, count: 25, return_address: ffffffffa021851b
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x19
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xa020e750
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xc000138b
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7bcc7000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xffffffff
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x224e94c6
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x225750a8
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xff1c168c
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x4
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x738d3484
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x19
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xa020e750
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xc000138b
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7bcc7000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xffffffff
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
  ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7b84c560
  ndiswrapper (mp_init:219): couldn't initialize device: C0010006
  ndiswrapper (pnp_start_device:435): Windows driver couldn't initialize the device (C0000001)
  ndiswrapper (mp_halt:262): device ffff880076a04800 is not initialized - not halting

Any hints on what this elusive 168c:ff1c device (and :ff1d) actually
is? Mac and Windows drivers *do* list these IDs...


-David

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

On Wed, Sep 29, 2010 at 02:42:30PM -0700, David Lamparter wrote:
> On Mon, Sep 27, 2010 at 10:44:47PM +0200, David Lamparter wrote:
> >                                  there are two atheros device IDs - ff1c
> > and ff1d - that designate AR5008 cards (according to pci.ids and google)
> > and that are not supported by any driver currently.
> > 
> > Are these cards driver compatible to the 0023/0024 devices? Do they need
> > different workarounds?
> 
> On Sun, Sep 26, 2010 at 03:39:35PM +0200, David Lamparter wrote:
> > 02:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR5008 Wireless Network Adapter [168c:ff1c] (rev 01)
> 
> I've now tried madwifi and ndiswrapper. To sum up:
> 
> - ath9k: ath9k_hw_chip_test fails:
>   (ath9k modified to add PCI ID 168c:ff1c)
>   "address test failed addr: 0x00008000 - wr:0xaaaaaaaa != rd:0xaaaaaaab"
> 
> - madwifi trunk:
>   (madwifi modified to add PCI ID 168c:ff1c)
>   "MadWifi: unable to attach hardware: '' (HAL status 1434461696)
>   (1434461696 = 0x55802600)
> 
> - ndiswrapper with 7.7.0.523 WinXP 64-bit driver:
>   (driver lists 168c:ff1c in .inf file, no modification)
>   ndiswrapper (link_pe_images:566): fixing KI_USER_SHARED_DATA address in the driver
>   ndiswrapper: driver netathwx (,06/04/2010,7.7.0.523) loaded
>   ndiswrapper 0000:02:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
>   ndiswrapper 0000:02:00.0: setting latency timer to 64
>   ndiswrapper (NdisWriteErrorLogEntry:190): log: C000138B, count: 25, return_address: ffffffffa021851b
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x19
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xa020e750
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xc000138b
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7bcc7000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xffffffff
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x224e94c6
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x225750a8
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xff1c168c
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x4
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x738d3484
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x19
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x21bd5000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xa020e750
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xc000138b
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7bcc7000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0xffffffff
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x76aec000
>   ndiswrapper (NdisWriteErrorLogEntry:193): code: 0x7b84c560
>   ndiswrapper (mp_init:219): couldn't initialize device: C0010006
>   ndiswrapper (pnp_start_device:435): Windows driver couldn't initialize the device (C0000001)
>   ndiswrapper (mp_halt:262): device ffff880076a04800 is not initialized - not halting

Your device is bust then it seems.

> Any hints on what this elusive 168c:ff1c device (and :ff1d) actually
> is? Mac and Windows drivers *do* list these IDs...

No, not sure, perhaps a mistake upon programming the EEPROM somehow?

  Luis

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Peter Stuge]

Luis R. Rodriguez wrote:
> > Any hints on what this elusive 168c:ff1c device (and :ff1d) actually
> > is? Mac and Windows drivers *do* list these IDs...
> 
> No, not sure, perhaps a mistake upon programming the EEPROM somehow?

Except that Mathieu already confirmed it as a known ID. Maybe no more
can be said. Too bad.


//Peter

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Adrian Chadd]

On 30 September 2010 05:42, David Lamparter <equinox@diac24.net> wrote:

> I've now tried madwifi and ndiswrapper. To sum up:

Have you tried windows?


Adrian

[ath9k-devel] [internal-ath9k-devel] PCI device 168c:ff1c ("AR5008", possibly Apple-branded)

[Luis R. Rodriguez]

On Wed, Sep 29, 2010 at 03:16:12PM -0700, Peter Stuge wrote:
> Luis R. Rodriguez wrote:
> > > Any hints on what this elusive 168c:ff1c device (and :ff1d) actually
> > > is? Mac and Windows drivers *do* list these IDs...
> > 
> > No, not sure, perhaps a mistake upon programming the EEPROM somehow?
> 
> Except that Mathieu already confirmed it as a known ID.

Peter, Mathieu's comment was saying how that PCI Device ID is used
internally to support devices used during emulation. Emulation devices
are not production devices, they are not applicable to the real world.

> Maybe no more can be said. Too bad.

I already noted how the device chip test failed, and how this gets
run before even the EEPROM gets checksummed and checked, I suspected
the device is bust and the tests run indicate it very likely is.

  Luis