Received-SPF: pass (google.com: domain of linux-wireless+bounces-8084-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.199.223 as permitted sender) client-ip=147.75.199.223;
Message-ID: <024e2594-309a-4549-9c69-d527717c5dd4@quicinc.com>
Date: Mon, 27 May 2024 15:08:40 +0530
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH v2 5/5] wifi: ath12k: Dump additional Tx PDEV HTT stats
To: Kalle Valo <kvalo@kernel.org>
CC: <ath12k@lists.infradead.org>, <linux-wireless@vger.kernel.org>
References: <20240510050806.514126-1-quic_rgnanase@quicinc.com>
 <20240510050806.514126-6-quic_rgnanase@quicinc.com>
 <87le43d26u.fsf@kernel.org>
Content-Language: en-US
From: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
In-Reply-To: <87le43d26u.fsf@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit


On 5/21/2024 1:30 PM, Kalle Valo wrote:
> Ramya Gnanasekar <quic_rgnanase@quicinc.com> writes:
> 
>> Support to dump additional Tx PDEV stats through HTT stats debugfs.
>> Following stats dump are supported:
>>         1. PDEV control path stat to dump Tx management frame count
>>         2. Tx PDEV SIFS histogram stats
>>         3. Tx MU MIMO PPDU stats for 802.11ac, 802.11ax and 802.11be
>>
>> Sample Output:
>> --------------
>> echo 1 > /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats_type
>> cat /sys/kernel/debug/ath12k/pci-0000\:06\:00.0/mac0/htt_stats
>> HTT_TX_PDEV_STATS_CMN_TLV:
>> mac_id = 0
>> comp_delivered = 0
>> self_triggers = 13
>> ......
>> ......
>> HTT_TX_PDEV_STATS_CTRL_PATH_TX_STATS:
>> fw_tx_mgmt_subtype =  0:1, 1:0, 2:0, 3:0, 4:38, 5:0, 6:0, 7:0, 8:0, 9:0, 10:0, 11:1, 12:0, 13:7, 14:0, 15:0,
>>
>> HTT_TX_PDEV_STATS_SIFS_HIST_TLV:
>> sifs_hist_status =  0:237, 1:185, 2:1, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>>
>> HTT_TX_PDEV_MU_PPDU_DISTRIBUTION_STATS:
>> ac_mu_mimo_num_seq_posted_nr4 = 0
>> ac_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ac_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ac_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> ac_mu_mimo_num_seq_posted_nr8 = 0
>> ac_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ac_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ac_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> ax_mu_mimo_num_seq_posted_nr4 = 0
>> ax_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ax_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ax_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> ax_mu_mimo_num_seq_posted_nr8 = 0
>> ax_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ax_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> ax_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> be_mu_mimo_num_seq_posted_nr4 = 0
>> be_mu_mimo_num_ppdu_posted_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> be_mu_mimo_num_ppdu_completed_per_burst_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> be_mu_mimo_num_seq_term_status_nr4 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> be_mu_mimo_num_seq_posted_nr8 = 0
>> be_mu_mimo_num_ppdu_posted_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> be_mu_mimo_num_ppdu_completed_per_burst_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0, 9:0,
>> be_mu_mimo_num_seq_term_status_nr8 =  0:0, 1:0, 2:0, 3:0, 4:0, 5:0, 6:0, 7:0, 8:0,
>>
>> Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.0.1-00029-QCAHKSWPL_SILICONZ-1
>> Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
>>
>> Signed-off-by: Ramya Gnanasekar <quic_rgnanase@quicinc.com>
> 
> [...]
> 
>> +static inline void
>> +ath12k_htt_print_tx_pdev_mu_ppdu_dist_stats_tlv(const void *tag_buf,
>> +						struct debug_htt_stats_req *stats_req)
>> +{
>> +	const struct ath12k_htt_tx_pdev_mu_ppdu_dist_stats_tlv *htt_stats_buf = tag_buf;
>> +	char *mode;
>> +	u8 j, hw_mode, i, str_buf_len;
>> +	u8 *buf = stats_req->buf;
>> +	u32 len = stats_req->buf_len;
>> +	u32 buf_len = ATH12K_HTT_STATS_BUF_SIZE;
>> +	u32 stats_value;
>> +	u8 max_ppdu = ATH12K_HTT_STATS_MAX_NUM_MU_PPDU_PER_BURST;
>> +	u8 max_sched = ATH12K_HTT_STATS_MAX_NUM_SCHED_STATUS;
>> +	char str_buf[ATH12K_HTT_MAX_STRING_LEN];
>> +
>> +	hw_mode = le32_to_cpu(htt_stats_buf->hw_mode);
>> +
>> +	switch (hw_mode) {
>> +	case ATH12K_HTT_STATS_HWMODE_AC:
>> +		len += scnprintf(buf + len, buf_len - len,
>> +				 "HTT_TX_PDEV_MU_PPDU_DISTRIBUTION_STATS:\n");
>> +		mode = "ac";
>> +		break;
>> +	case ATH12K_HTT_STATS_HWMODE_AX:
>> +		mode = "ax";
>> +		break;
>> +	case ATH12K_HTT_STATS_HWMODE_BE:
>> +		mode = "be";
>> +		break;
>> +	default:
>> +		return;
>> +	}
> 
> Why are we not adding "HTT_TX_PDEV_MU_PPDU_DISTRIBUTION_STATS:\n" with
> ax and be modes?
> 
Sorry for the delayed response. I was on OoO for a week.
We will receive this TLV for each hw modes. Since
"HTT_TX_PDEV_MU_PPDU_DISTRIBUTION_STATS:\n" is header and it would be
suffice to print it once, hence added it inside hw mode ac which will be
the first hw mode integrated inside the TLV.