interconnect: qcom: icc-rpmh: Fix an error handling path in qcom_icc_rpmh_probe()

Message ID ec929c37c655ede7bb42e426354093c8a1377a0b.1668947686.git.christophe.jaillet@wanadoo.fr
State New
Headers
Series interconnect: qcom: icc-rpmh: Fix an error handling path in qcom_icc_rpmh_probe() |

Commit Message

Christophe JAILLET Nov. 20, 2022, 12:35 p.m. UTC
  If of_platform_populate() fails, some resources need to be freed as already
done in the other error handling paths.

Fixes: 57eb14779dfd ("interconnect: qcom: icc-rpmh: Support child NoC device probe")

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
---
 drivers/interconnect/qcom/icc-rpmh.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
  

Comments

Luca Weiss Nov. 21, 2022, 7:42 a.m. UTC | #1
Hi Christophe,

On Sun Nov 20, 2022 at 1:35 PM CET, Christophe JAILLET wrote:
> If of_platform_populate() fails, some resources need to be freed as already
> done in the other error handling paths.
>
> Fixes: 57eb14779dfd ("interconnect: qcom: icc-rpmh: Support child NoC device probe")

I believe the same needs to be applied to icc-rpm.c.

Also there shouldn't be an empty line here between Fixes: and Signed-off-by:

Regards
Luca

>
> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> ---
>  drivers/interconnect/qcom/icc-rpmh.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/interconnect/qcom/icc-rpmh.c b/drivers/interconnect/qcom/icc-rpmh.c
> index fd17291c61eb..5168bbf3d92f 100644
> --- a/drivers/interconnect/qcom/icc-rpmh.c
> +++ b/drivers/interconnect/qcom/icc-rpmh.c
> @@ -235,8 +235,11 @@ int qcom_icc_rpmh_probe(struct platform_device *pdev)
>  	platform_set_drvdata(pdev, qp);
>  
>  	/* Populate child NoC devices if any */
> -	if (of_get_child_count(dev->of_node) > 0)
> -		return of_platform_populate(dev->of_node, NULL, NULL, dev);
> +	if (of_get_child_count(dev->of_node) > 0) {
> +		ret = of_platform_populate(dev->of_node, NULL, NULL, dev);
> +		if (ret)
> +			goto err;
> +	}
>  
>  	return 0;
>  err:
> -- 
> 2.34.1
  
Christophe JAILLET Nov. 21, 2022, 9:08 p.m. UTC | #2
Le 21/11/2022 à 08:42, Luca Weiss a écrit :
> Hi Christophe,
> 
> On Sun Nov 20, 2022 at 1:35 PM CET, Christophe JAILLET wrote:
>> If of_platform_populate() fails, some resources need to be freed as already
>> done in the other error handling paths.
>>
>> Fixes: 57eb14779dfd ("interconnect: qcom: icc-rpmh: Support child NoC device probe")
> 
> I believe the same needs to be applied to icc-rpm.c.

I'll give it a look and send a v2.

CJ

> 
> Also there shouldn't be an empty line here between Fixes: and Signed-off-by:
> 
> Regards
> Luca
> 
>>
>> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
>> ---
>>   drivers/interconnect/qcom/icc-rpmh.c | 7 +++++--
>>   1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/interconnect/qcom/icc-rpmh.c b/drivers/interconnect/qcom/icc-rpmh.c
>> index fd17291c61eb..5168bbf3d92f 100644
>> --- a/drivers/interconnect/qcom/icc-rpmh.c
>> +++ b/drivers/interconnect/qcom/icc-rpmh.c
>> @@ -235,8 +235,11 @@ int qcom_icc_rpmh_probe(struct platform_device *pdev)
>>   	platform_set_drvdata(pdev, qp);
>>   
>>   	/* Populate child NoC devices if any */
>> -	if (of_get_child_count(dev->of_node) > 0)
>> -		return of_platform_populate(dev->of_node, NULL, NULL, dev);
>> +	if (of_get_child_count(dev->of_node) > 0) {
>> +		ret = of_platform_populate(dev->of_node, NULL, NULL, dev);
>> +		if (ret)
>> +			goto err;
>> +	}
>>   
>>   	return 0;
>>   err:
>> -- 
>> 2.34.1
> 
>
  
Christophe JAILLET Nov. 22, 2022, 9:45 p.m. UTC | #3
Le 21/11/2022 à 08:42, Luca Weiss a écrit :
> Hi Christophe,
> 
> On Sun Nov 20, 2022 at 1:35 PM CET, Christophe JAILLET wrote:
>> If of_platform_populate() fails, some resources need to be freed as already
>> done in the other error handling paths.
>>
>> Fixes: 57eb14779dfd ("interconnect: qcom: icc-rpmh: Support child NoC device probe")
> 
> I believe the same needs to be applied to icc-rpm.c.

Their are other issues in the error handling path of this file.

1)	if (desc->has_bus_pd) {
		ret = dev_pm_domain_attach(dev, true);
is unbalanced, both in the error handling path and in the remove function.


2)	clk_bulk_prepare_enable()
is not balanced in all error handling paths.


3) the same issue about error handling if of_platform_populate() fails.


These issues have been introduced in different commits.


Would you prefer several patches, each related to a given Fixes: tag, or 
one bigger "fix all error handling paths"?


Anyway, fixing the points above would require moving some code around. 
Usually I try to avoid it, because it is not always easy to see 
associated side effects.

CJ

> 
> Also there shouldn't be an empty line here between Fixes: and Signed-off-by:
> 
> Regards
> Luca
> 
>>
>> Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
>> ---
>>   drivers/interconnect/qcom/icc-rpmh.c | 7 +++++--
>>   1 file changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/interconnect/qcom/icc-rpmh.c b/drivers/interconnect/qcom/icc-rpmh.c
>> index fd17291c61eb..5168bbf3d92f 100644
>> --- a/drivers/interconnect/qcom/icc-rpmh.c
>> +++ b/drivers/interconnect/qcom/icc-rpmh.c
>> @@ -235,8 +235,11 @@ int qcom_icc_rpmh_probe(struct platform_device *pdev)
>>   	platform_set_drvdata(pdev, qp);
>>   
>>   	/* Populate child NoC devices if any */
>> -	if (of_get_child_count(dev->of_node) > 0)
>> -		return of_platform_populate(dev->of_node, NULL, NULL, dev);
>> +	if (of_get_child_count(dev->of_node) > 0) {
>> +		ret = of_platform_populate(dev->of_node, NULL, NULL, dev);
>> +		if (ret)
>> +			goto err;
>> +	}
>>   
>>   	return 0;
>>   err:
>> -- 
>> 2.34.1
> 
>
  
Luca Weiss Nov. 23, 2022, 7:32 a.m. UTC | #4
Hi Christophe,

On Tue Nov 22, 2022 at 10:45 PM CET, Christophe JAILLET wrote:
> > I believe the same needs to be applied to icc-rpm.c.
>
> Their are other issues in the error handling path of this file.
>
> 1)	if (desc->has_bus_pd) {
> 		ret = dev_pm_domain_attach(dev, true);
> is unbalanced, both in the error handling path and in the remove function.
>
>
> 2)	clk_bulk_prepare_enable()
> is not balanced in all error handling paths.
>
>
> 3) the same issue about error handling if of_platform_populate() fails.
>
>
> These issues have been introduced in different commits.
>
>
> Would you prefer several patches, each related to a given Fixes: tag, or 
> one bigger "fix all error handling paths"?

I don't really have an opinion on this, I guess if the issues were
introduced in different commits, make separate fixes?
If it's the wrong way somebody will tell you anyways ;)

Regards
Luca

>
>
> Anyway, fixing the points above would require moving some code around. 
> Usually I try to avoid it, because it is not always easy to see 
> associated side effects.
>
> CJ
  

Patch

diff --git a/drivers/interconnect/qcom/icc-rpmh.c b/drivers/interconnect/qcom/icc-rpmh.c
index fd17291c61eb..5168bbf3d92f 100644
--- a/drivers/interconnect/qcom/icc-rpmh.c
+++ b/drivers/interconnect/qcom/icc-rpmh.c
@@ -235,8 +235,11 @@  int qcom_icc_rpmh_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, qp);
 
 	/* Populate child NoC devices if any */
-	if (of_get_child_count(dev->of_node) > 0)
-		return of_platform_populate(dev->of_node, NULL, NULL, dev);
+	if (of_get_child_count(dev->of_node) > 0) {
+		ret = of_platform_populate(dev->of_node, NULL, NULL, dev);
+		if (ret)
+			goto err;
+	}
 
 	return 0;
 err: