Re: [PATCH v3 2/5] coresight: tnoc: Add AG tnoc standalone compatible to the platform driver

[Jie Gan <jie.gan@oss.qualcomm.com>]

On 6/30/2026 6:57 PM, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] System Trace ID is leaked on platform device removal for standalone Aggregator TNOC.
> --
> 
> commit 060ed6649d8ad0bd7b6b53c7f626563782e801df
> Author: Jie Gan <jie.gan@oss.qualcomm.com>
> 
> coresight: tnoc: Add AG tnoc standalone compatible to the platform driver
> 
> This commit adds the "qcom,coresight-agtnoc" compatible to allow the Aggregator
> TNOC to probe as a standalone platform device. It updates the platform driver
> and callback names to generic "tnoc" variants. It also modifies the default
> data initialization so that only the Interconnect TNOC skips trace ID allocation.
> 
>> diff --git a/drivers/hwtracing/coresight/coresight-tnoc.c b/drivers/hwtracing/coresight/coresight-tnoc.c
>> index 9e8de4323d285..8237467faba72 100644
>> --- a/drivers/hwtracing/coresight/coresight-tnoc.c
>> +++ b/drivers/hwtracing/coresight/coresight-tnoc.c
> [ ... ]
>> @@ -295,7 +295,7 @@ static int itnoc_probe(struct platform_device *pdev)
>>   	return ret;
>>   }
>>   
>> -static void itnoc_remove(struct platform_device *pdev)
>> +static void tnoc_platform_remove(struct platform_device *pdev)
>>   {
>>   	struct trace_noc_drvdata *drvdata = platform_get_drvdata(pdev);
>>   
>>   	coresight_unregister(drvdata->csdev);
>>   	pm_runtime_disable(&pdev->dev);
>>   }
> 
> [Severity: High]
> Does this code leak the allocated system trace ID on device removal?
> 

Thanks for reporting. We need free the trace ID in remove path.

Thanks,
Jie

> Because trace_noc_init_default_data() now allocates a system trace ID via
> coresight_trace_id_get_system_id() for the new "qcom,coresight-agtnoc"
> devices, it appears tnoc_platform_remove() needs a corresponding call to
> coresight_trace_id_put_system_id(drvdata->atid) when the driver is unbound.
> 
> Without this cleanup, repeated unbind and bind cycles could permanently
> exhaust the finite system trace ID pool.
>