Re: [PATCH v2 2/3] soc: qcom: qcom_stats: Add QMP support for syncing ddr stats

["Maulik Shah (mkshah)" <maulik.shah@oss.qualcomm.com>]

On 5/21/2025 6:22 PM, Dmitry Baryshkov wrote:
> On Wed, May 21, 2025 at 02:02:11PM +0530, Maulik Shah wrote:
>> Recent SoCs (SM8450 onwards) require QMP command to be sent before reading
>> ddr stats. The duration field of ddr stats will get populated only if QMP
>> command is sent.
>>
>> Add support to send ddr stats freqsync QMP command.
>>
>> Signed-off-by: Maulik Shah <maulik.shah@oss.qualcomm.com>
>> ---
>>  drivers/soc/qcom/qcom_stats.c | 29 ++++++++++++++++++++++++++++-
>>  1 file changed, 28 insertions(+), 1 deletion(-)
>>
>> @@ -310,6 +329,14 @@ static int qcom_stats_probe(struct platform_device *pdev)
>>  	qcom_create_subsystem_stat_files(root, config);
>>  	qcom_create_soc_sleep_stat_files(root, reg, d, config);
>>  	qcom_create_ddr_stat_files(root, reg, config);
>> +	/*
>> +	 * QMP is used for DDR stats syncing to MSG RAM for recent SoCs (SM8450 onwards).
>> +	 * The prior SoCs do not need QMP handle as the required stats are already present
>> +	 * in MSG RAM, provided the DDR_STATS_MAGIC_KEY matches.
>> +	 */
>> +	qcom_stats_qmp = qmp_get(&pdev->dev);
>> +	if (IS_ERR(qcom_stats_qmp))
>> +		qcom_stats_qmp = NULL;
> 
> Don't we need to handle QMP being not probed _yet_? In other words,
> there are several distinct cases:
> 
> - No QMP defined, proceed without it
> - QMP defined, not probed yet, return -EPROBE_DEFER here
> - QMP defined, qmp_get() returns an error, return that error code
> - QMP defined, qmp_get() returned valid pointer, proceed with using it.
> 

Yes we need to handle these cases.
I will update in v3.

Thanks,
Maulik

>>  
>>  	platform_set_drvdata(pdev, root);
>>  
>>
>> -- 
>> 2.34.1
>>
>