coresight: cti: Check if the CPU activated for the CPU CTI

Message ID 1679910560-23469-1-git-send-email-quic_taozha@quicinc.com
State New
Headers
Series coresight: cti: Check if the CPU activated for the CPU CTI |

Commit Message

Tao Zhang March 27, 2023, 9:49 a.m. UTC
  Check whether the CPU corresponding to the CPU CTI is activated.
If it is not activated, the CPU CTI node should not exist, and
an error will be returned in the initialization function.

Signed-off-by: Tao Zhang <quic_taozha@quicinc.com>
---
 drivers/hwtracing/coresight/coresight-cti-core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
  

Comments

Suzuki K Poulose March 27, 2023, 9:52 a.m. UTC | #1
On 27/03/2023 10:49, Tao Zhang wrote:
> Check whether the CPU corresponding to the CPU CTI is activated.
> If it is not activated, the CPU CTI node should not exist, and
> an error will be returned in the initialization function.
> 
> Signed-off-by: Tao Zhang <quic_taozha@quicinc.com>
> ---
>   drivers/hwtracing/coresight/coresight-cti-core.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
> index 277c890..aaa83ae 100644
> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
> @@ -899,10 +899,12 @@ static int cti_probe(struct amba_device *adev, const struct amba_id *id)
>   	drvdata->config.hw_powered = true;
>   
>   	/* set up device name - will depend if cpu bound or otherwise */
> -	if (drvdata->ctidev.cpu >= 0)
> +	if (drvdata->ctidev.cpu >= 0) {
> +		if (!cpu_active(drvdata->ctidev.cpu))
> +			return -ENXIO;
>   		cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>   					       drvdata->ctidev.cpu);

But why ? As long as we do not enable or touch any CPU specific bits in 
the probe, why do we need to fail this ? What are you trying to fix ?

Please could you share the log if you are hitting something ? This looks
like masking a problem.

Suzuki


> -	else
> +	} else
>   		cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
>   	if (!cti_desc.name)
>   		return -ENOMEM;
  
Suzuki K Poulose March 27, 2023, 2:57 p.m. UTC | #2
On 27/03/2023 15:28, Tao Zhang wrote:
> Hi Suzuki,
> 
> On 3/27/2023 5:52 PM, Suzuki K Poulose wrote:
>> On 27/03/2023 10:49, Tao Zhang wrote:
>>> Check whether the CPU corresponding to the CPU CTI is activated.
>>> If it is not activated, the CPU CTI node should not exist, and
>>> an error will be returned in the initialization function.
>>>
>>> Signed-off-by: Tao Zhang <quic_taozha@quicinc.com>
>>> ---
>>>   drivers/hwtracing/coresight/coresight-cti-core.c | 6 ++++--
>>>   1 file changed, 4 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c 
>>> b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> index 277c890..aaa83ae 100644
>>> --- a/drivers/hwtracing/coresight/coresight-cti-core.c
>>> +++ b/drivers/hwtracing/coresight/coresight-cti-core.c
>>> @@ -899,10 +899,12 @@ static int cti_probe(struct amba_device *adev, 
>>> const struct amba_id *id)
>>>       drvdata->config.hw_powered = true;
>>>         /* set up device name - will depend if cpu bound or otherwise */
>>> -    if (drvdata->ctidev.cpu >= 0)
>>> +    if (drvdata->ctidev.cpu >= 0) {
>>> +        if (!cpu_active(drvdata->ctidev.cpu))
>>> +            return -ENXIO;
>>>           cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
>>>                              drvdata->ctidev.cpu);
>>
>> But why ? As long as we do not enable or touch any CPU specific bits 
>> in the probe, why do we need to fail this ? What are you trying to fix ?
>>
>> Please could you share the log if you are hitting something ? This looks
>> like masking a problem.
>>
>> Suzuki
> 
> We found that when the CPU core is disabled, for example, CPU3 is 
> disabled, but
> 
> CPU3 CTI node corresponding to CPU3 still exists. In fact, in this case, 
> CPU3 CTI
> 
> has been unable to trigger CPU3 properly since CPU3 is in an inactive 
> state. This change
> 
> is to avoid configuring the CPU CTI of the CPU that has been disabled in 
> this case.

Who is configuring the trigger ? Shouldn't we skip "enabling" the CTI
when the associated CPU is inactive instead ? Disabling the probe with
an error doesn't solve the problem. What if the CPU becomes active later 
? What makes sure that the CTI is probed then ?

Suzuki


> 
> Tao
> 
>>
>>
>>> -    else
>>> +    } else
>>>           cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, 
>>> dev);
>>>       if (!cti_desc.name)
>>>           return -ENOMEM;
>>
  

Patch

diff --git a/drivers/hwtracing/coresight/coresight-cti-core.c b/drivers/hwtracing/coresight/coresight-cti-core.c
index 277c890..aaa83ae 100644
--- a/drivers/hwtracing/coresight/coresight-cti-core.c
+++ b/drivers/hwtracing/coresight/coresight-cti-core.c
@@ -899,10 +899,12 @@  static int cti_probe(struct amba_device *adev, const struct amba_id *id)
 	drvdata->config.hw_powered = true;
 
 	/* set up device name - will depend if cpu bound or otherwise */
-	if (drvdata->ctidev.cpu >= 0)
+	if (drvdata->ctidev.cpu >= 0) {
+		if (!cpu_active(drvdata->ctidev.cpu))
+			return -ENXIO;
 		cti_desc.name = devm_kasprintf(dev, GFP_KERNEL, "cti_cpu%d",
 					       drvdata->ctidev.cpu);
-	else
+	} else
 		cti_desc.name = coresight_alloc_device_name(&cti_sys_devs, dev);
 	if (!cti_desc.name)
 		return -ENOMEM;