Bug in CCK hw_rate definitions?

Bug in CCK hw_rate definitions?

[Ben Greear]

ath10k hw.h has this:

enum ath10k_hw_rate_cck {
	ATH10K_HW_RATE_CCK_LP_11M = 0,
	ATH10K_HW_RATE_CCK_LP_5_5M,
	ATH10K_HW_RATE_CCK_LP_2M,
	ATH10K_HW_RATE_CCK_LP_1M,

But, at least in 10.1 firmware, it appears that CCK hw_rates are 0x40, 0x41, etc
(For those of you with firmware, see ar600P_phy.c, 'RC' column of the rate table.)

Am I missing something or is the driver code just wrong?

Thanks,
Ben

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


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Bug in CCK hw_rate definitions?

[Michal Kazior]

On 15 September 2015 at 23:35, Ben Greear <greearb@candelatech.com> wrote:
> ath10k hw.h has this:
>
> enum ath10k_hw_rate_cck {
>         ATH10K_HW_RATE_CCK_LP_11M = 0,
>         ATH10K_HW_RATE_CCK_LP_5_5M,
>         ATH10K_HW_RATE_CCK_LP_2M,
>         ATH10K_HW_RATE_CCK_LP_1M,
>
> But, at least in 10.1 firmware, it appears that CCK hw_rates are 0x40, 0x41,
> etc
> (For those of you with firmware, see ar600P_phy.c, 'RC' column of the rate
> table.)
>
> Am I missing something or is the driver code just wrong?

These macros are used only to decode L_SIG_RATE now. It can take values 0..15.

If you need/want to change ath10k_hw_rate_cck you'll need to make sure
to update ath10k_mac_hw_rate_to_idx() or its callsite accordingly.


Michał

_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: Bug in CCK hw_rate definitions?

[Ben Greear]

On 09/21/2015 12:32 AM, Michal Kazior wrote:
> On 15 September 2015 at 23:35, Ben Greear <greearb@candelatech.com> wrote:
>> ath10k hw.h has this:
>>
>> enum ath10k_hw_rate_cck {
>>          ATH10K_HW_RATE_CCK_LP_11M = 0,
>>          ATH10K_HW_RATE_CCK_LP_5_5M,
>>          ATH10K_HW_RATE_CCK_LP_2M,
>>          ATH10K_HW_RATE_CCK_LP_1M,
>>
>> But, at least in 10.1 firmware, it appears that CCK hw_rates are 0x40, 0x41,
>> etc
>> (For those of you with firmware, see ar600P_phy.c, 'RC' column of the rate
>> table.)
>>
>> Am I missing something or is the driver code just wrong?
>
> These macros are used only to decode L_SIG_RATE now. It can take values 0..15.
>
> If you need/want to change ath10k_hw_rate_cck you'll need to make sure
> to update ath10k_mac_hw_rate_to_idx() or its callsite accordingly.

Just to close the loop on this:

I was confused about exactly how this was being used.

In the end, I have this working to my satisfaction without having
to change the enums.

Thanks,
Ben

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


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k