ath10k: first qca6174 hw2.1 firmware released

[Michal Kazior]

On 24 August 2015 at 18:26, Lapo Calamandrei <calamandrei at gmail.com> wrote:
> 2015-08-17 7:52 GMT+02:00 Michal Kazior <michal.kazior at tieto.com>:
>> On 13 August 2015 at 16:00, Lapo Calamandrei <calamandrei at gmail.com> wrote:
>>> 2015-08-13 12:31 GMT+02:00 Kalle Valo <kvalo at qca.qualcomm.com>:
>>>> Hi,
>>>>
>>>> I just pushed the first qca6174  hw2.1 firmware to ath10k-firmware.git:
>>>>
>>>> https://github.com/kvalo/ath10k-firmware/commit/7f7e7dda33676ced293de477b03711199ffe5256
>>>>
>>>> Please test and provide feedback.
>>>>
>>>> --
>>>> Kalle Valo
>>>>
>>>
>>> Loads w/o skip_otp and works fine here (qca6174), except it crashes on
>>> 5GHz network connection attempt as the one created from the window
>>> drivers.
>>
>> Paste the crash register dump and `iw list` output, please.
>>
>>
>> Michał
>
> Sorry for the late reply.
>
> Here's the iw list output:
[...]
>     Band 1:
>         Capabilities: 0x19e3
>             RX LDPC
>             HT20/HT40
>             Static SM Power Save
>             RX HT20 SGI
>             RX HT40 SGI
>             TX STBC
>             RX STBC 1-stream
>             Max AMSDU length: 7935 bytes
>             DSSS/CCK HT40
>         Maximum RX AMPDU length 65535 bytes (exponent: 0x003)
>         Minimum RX AMPDU time spacing: 8 usec (0x06)
>         HT TX/RX MCS rate indexes supported: 0-15
>         VHT Capabilities (0x339031b2):
>             Max MPDU length: 11454
>             Supported Channel Width: neither 160 nor 80+80
>             RX LDPC
>             short GI (80 MHz)
>             TX STBC
>             SU Beamformee
>             MU Beamformee

Is this with the change I suggested you earlier? I guess not. I don't
think this firmware should advertise beamforming support.

Can you re-try with patch from
http://lists.infradead.org/pipermail/ath10k/2015-August/005774.html,
please? If it still doesn't work please post `iw list` and the
register dump.

You can also try commenting out calls to ath10k_mac_vif_recalc_txbf() as plan B.


[...]
> Here's the register dump:
>
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> firmware register dump:
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [00]: 0x05010000 0x000015B3 0x00939797 0x00955B31
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [04]: 0x00939797 0x00060330 0x00000000 0x00000000
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [08]: 0x00413980 0x0000FFFF 0x00000000 0x00000000
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [12]: 0x00000009 0xFFFFFFFF 0x0096C09C 0x0096C0A7
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [16]: 0x0096BDBC 0x009287BD 0x00000000 0x009287BD
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [20]: 0x40939797 0x0041A700 0x00407124 0x00000000
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [24]: 0x8093D6C4 0x0041A760 0x004059D0 0xC0939797
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [28]: 0x8094777F 0x0041A780 0x0046D5D8 0x00000001
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [32]: 0x800AA427 0x0041A7B0 0x0046D5D8 0x00000001
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [36]: 0x800AA58E 0x0041A7D0 0x00424824 0x00000001
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [40]: 0x80994D90 0x0041A7F0 0x00424824 0x0041A838
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [44]: 0x80996DC6 0x0041A820 0x0046F888 0x00412984
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [48]: 0x800B43F5 0x0041A860 0x004222F8 0x00005008
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [52]: 0x809A6ACC 0x0041A8F0 0x0042930C 0x0042CA24
> ago 24 18:10:07 localhost.localdomain kernel: ath10k_pci 0000:03:00.0:
> [56]: 0x809A6120 0x0041A930 0x0041A958 0x00427110

This definitely looks like a crash on TxBF configuration.


Michał