[LEDE-DEV] ath10k ct ack timeout/coverage class

[Ben Greear]

The settings should stick in the firmware, but they have to be set each time the
firmware starts (the ath10k-ct driver should do that now, I think).

I haven't tested this feature though, so let me know if you find any issues.

Thanks,
Ben

On 04/24/2017 05:17 PM, Nick Dennis wrote:
> Thanks. i did see something in mail-archive (don't know what source)
>
> https://www.mail-archive.com/search?l=ath10k@lists.infradead.org&q=subject:%22request%5C%3A+ACK+timing+setting+required%22&o=newest&f=1
>
> Does the debugfs setting get overwritten by the ini settings during resets and have to be re-done ?
>
> In the past, I just set the ack timeout register without adjusting any others and it worked, so I'll try that first.
>
> Nick
>
>
> On 24/04/2017 4:58 PM, Ben Greear wrote:
>> There are some un-documented hacks using the ct_special debugfs file.
>>
>> Very unlikely they are hooked into LEDE.
>>
>> See this code in wmi.h:
>>
>> /* CT firmware only, and only builds after June 26, 2015 */
>> struct wmi_pdev_set_special_cmd {
>> #define SET_SPECIAL_ID_ACK_CTS 0 /* set ack-cts-timeout register */
>> #define SET_SPECIAL_ID_SLOT    1 /* set slot-duration register */
>> #define SET_SPECIAL_ID_SIFS    2 /* set sifs-duration register */
>>
>> and the set_special logic in debugfs.
>>
>> Thanks,
>> Ben
>>
>> On 04/24/2017 03:36 PM, Nick Dennis wrote:
>>> I have AP <-> Client links on the ath10k-ct driver+ firmware on QCA 9882 radios that work fine at a couple of hundred Meters. I recently tried a link at 1.2 Kilometers, but only got <1Mbps throughput. Does anyone know if coverage class/ack timeout is implemented in the ath10k-ct driver+firmware? Thanks
>>>
>>>
>>> _______________________________________________
>>> Lede-dev mailing list
>>> Lede-dev at lists.infradead.org
>>> http://lists.infradead.org/mailman/listinfo/lede-dev
>>>
>>
>

-- 
Ben Greear <greearb at candelatech.com>
Candela Technologies Inc  http://www.candelatech.com