Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <d6f794e3-7463-da7d-721e-00869faf4c17@quicinc.com>
Date:   Wed, 24 May 2023 14:33:03 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
 Thunderbird/102.10.1
Subject: Re: [PATCH v3] wifi: ath11k: Add rx histogram stats
Content-Language: en-US
To:     Tyler Stachecki <stachecki.tyler@gmail.com>
CC:     "open list:QUALCOMM ATHEROS ATH11K WIRELESS DRIVER" 
        <ath11k@lists.infradead.org>,
        "open list:NETWORKING DRIVERS (WIRELESS)" 
        <linux-wireless@vger.kernel.org>, <quic_akalaise@quicinc.com>,
        <quic_mpubbise@quicinc.com>
References: <CAC6wqPVvke=Gy59yvYcaps2dwVzO=wfwdSTGNxFzAOWc5OpoDw@mail.gmail.com>
From:   Maharaja Kennadyrajan <quic_mkenna@quicinc.com>
In-Reply-To: <CAC6wqPVvke=Gy59yvYcaps2dwVzO=wfwdSTGNxFzAOWc5OpoDw@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 5/21/2023 11:21 PM, Tyler Stachecki wrote:
> Was this really tested on QCN9074 as the commit text suggests...?
>
>> const struct ath11k_hw_ops ipq6018_ops = {
>> @@ -1132,6 +1147,7 @@ const struct ath11k_hw_ops wcn6750_ops = {
>>   .rx_desc_get_msdu_payload = ath11k_hw_qcn9074_rx_desc_get_msdu_payload,
>>   .reo_setup = ath11k_hw_wcn6855_reo_setup,
>>   .mpdu_info_get_peerid = ath11k_hw_ipq8074_mpdu_info_get_peerid,
>> + .mpdu_info_get_mpdu_len = ath11k_hw_qcn9074_mpdu_info_get_mpdu_len,
> ...
>
>> +static u32 ath11k_hal_rx_mpduinfo_get_mpdu_len(struct ath11k_base *ab,
>> +       struct hal_rx_mpdu_info *mpdu_info)
>> +{
>> + return ab->hw_params.hw_ops->mpdu_info_get_mpdu_len(mpdu_info);
>> +}
> I think you want to put this under qcn9074_ops. As of now, when
> QCN9074 is present, it attempts to jump to a NULL pointer as
> mpdu_info_get_mpdu_len remains uninitialized for qcn9074_ops.
>
> And, do you not need to define mpdu_info_get_mpdu_len for all the
> other hw_ops? If so, please be careful about defining it for
> WCN6855/WCN6750 as there was a recent regression due to how the RX
> MPDU info is provided by those firmwares as it differed from
> IPQ8074/QCN9074. I personally do not have the appropriate literature
> to determine whether or not this is consequential or not here as well,
> though it seems like it would be:
> https://lore.kernel.org/linux-wireless/20230404072234.18503-3-quic_youghand@quicinc.com/
>
> Tyler

Thanks for your comments. Will fix this in the upcoming patchset.