Received-SPF: pass (google.com: domain of linux-wireless+bounces-4030-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Message-ID: <00db4b62-4cd4-474f-bbe9-33f64d5e3853@quicinc.com>
Date: Mon, 26 Feb 2024 12:40:23 -0800
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 5/6] wifi: ath11k: invert scan flag
 WMI_SCAN_FILTER_PROBE_REQ for QCA6390/WCN6855/QCA2066
Content-Language: en-US
To: Kang Yang <quic_kangyang@quicinc.com>, <ath11k@lists.infradead.org>
CC: <linux-wireless@vger.kernel.org>
References: <20240226060203.2040444-1-quic_kangyang@quicinc.com>
 <20240226060203.2040444-6-quic_kangyang@quicinc.com>
From: Jeff Johnson <quic_jjohnson@quicinc.com>
In-Reply-To: <20240226060203.2040444-6-quic_kangyang@quicinc.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit

On 2/25/2024 10:02 PM, Kang Yang wrote:
> Current ROC scan will filter probe request. But probe request is
> necessary for P2P mode. A P2P device cannot be discovered by others if
> it doesn't respond to others' probe request.
> 
> So invert scan flag WMI_SCAN_FILTER_PROBE_REQ for
> QCA6390/WCN6855/QCA2066.
> 
> Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.37
> Tested-on: QCA2066 hw2.1 PCI WLAN.HSP.1.1-03926.13-QCAHSPSWPL_V2_SILICONZ_CE-2.52297.2
> 
> Signed-off-by: Kang Yang <quic_kangyang@quicinc.com>
> ---
>  drivers/net/wireless/ath/ath11k/wmi.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/net/wireless/ath/ath11k/wmi.c b/drivers/net/wireless/ath/ath11k/wmi.c
> index bbccddd7d729..1dd0cbdda199 100644
> --- a/drivers/net/wireless/ath/ath11k/wmi.c
> +++ b/drivers/net/wireless/ath/ath11k/wmi.c
> @@ -2317,6 +2317,9 @@ int ath11k_wmi_send_scan_start_cmd(struct ath11k *ar,
>  	ether_addr_copy(cmd->mac_addr.addr, params->mac_addr.addr);
>  	ether_addr_copy(cmd->mac_mask.addr, params->mac_mask.addr);
>  
> +	if (ar->ab->hw_params.single_pdev_only)
> +		cmd->scan_ctrl_flags ^=  WMI_SCAN_FILTER_PROBE_REQ;
> +

Why is this being done in WMI? Ideally WMI should just be doing host to
firmware translation, so seems this should be further up the stack, i.e.
in ath11k_mac_op_hw_scan() / ath11k_mac_op_remain_on_channel()

It also seems strange to invert the flag (which assumes a known starting
value) instead of just explicitly setting it to the required value.

>  	ptr += sizeof(*cmd);
>  
>  	len = params->num_chan * sizeof(u32);