[PATCH v2] rpmsg: qcom_smd: Improve error handling for qcom_smd_parse_edge

[PATCH v2] rpmsg: qcom_smd: Improve error handling for qcom_smd_parse_edge

[Luca Weiss]

When the mailbox driver has not probed yet, the error message "failed to
parse smd edge" is just going to confuse users, so improve the error
prints a bit.

Cover the last remaining exits from qcom_smd_parse_edge with proper
error prints, especially the one for the mbox_chan deserved
dev_err_probe to handle EPROBE_DEFER nicely. And add one for ipc_regmap
also to be complete.

With this done, we can remove the outer print completely.

Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
---
Changes in v2:
- Rebase on qcom for-next, drop dts patches which have been applied
- Improve error printing situation (Bjorn)
- Link to v1: https://lore.kernel.org/r/20240424-apcs-mboxes-v1-0-6556c47cb501@z3ntu.xyz
---
 drivers/rpmsg/qcom_smd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/rpmsg/qcom_smd.c b/drivers/rpmsg/qcom_smd.c
index 43f601c84b4f..06e6ba653ea1 100644
--- a/drivers/rpmsg/qcom_smd.c
+++ b/drivers/rpmsg/qcom_smd.c
@@ -1369,7 +1369,8 @@ static int qcom_smd_parse_edge(struct device *dev,
 	edge->mbox_chan = mbox_request_channel(&edge->mbox_client, 0);
 	if (IS_ERR(edge->mbox_chan)) {
 		if (PTR_ERR(edge->mbox_chan) != -ENODEV) {
-			ret = PTR_ERR(edge->mbox_chan);
+			ret = dev_err_probe(dev, PTR_ERR(edge->mbox_chan),
+					    "failed to acquire IPC mailbox\n");
 			goto put_node;
 		}
 
@@ -1386,6 +1387,7 @@ static int qcom_smd_parse_edge(struct device *dev,
 		of_node_put(syscon_np);
 		if (IS_ERR(edge->ipc_regmap)) {
 			ret = PTR_ERR(edge->ipc_regmap);
+			dev_err(dev, "failed to get regmap from syscon: %d\n", ret);
 			goto put_node;
 		}
 
@@ -1501,10 +1503,8 @@ struct qcom_smd_edge *qcom_smd_register_edge(struct device *parent,
 	}
 
 	ret = qcom_smd_parse_edge(&edge->dev, node, edge);
-	if (ret) {
-		dev_err(&edge->dev, "failed to parse smd edge\n");
+	if (ret)
 		goto unregister_dev;
-	}
 
 	ret = qcom_smd_create_chrdev(edge);
 	if (ret) {

---
base-commit: 2c79712cc83b172ce26c3086ced1c1fae087d8fb
change-id: 20240423-apcs-mboxes-12ee6c01a5b3

Best regards,
-- 
Luca Weiss <luca@z3ntu.xyz>

Re: [PATCH v2] rpmsg: qcom_smd: Improve error handling for qcom_smd_parse_edge

[Dmitry Baryshkov]

On Thu, Jun 06, 2024 at 09:01:36PM +0200, Luca Weiss wrote:
> When the mailbox driver has not probed yet, the error message "failed to
> parse smd edge" is just going to confuse users, so improve the error
> prints a bit.
> 
> Cover the last remaining exits from qcom_smd_parse_edge with proper
> error prints, especially the one for the mbox_chan deserved
> dev_err_probe to handle EPROBE_DEFER nicely. And add one for ipc_regmap
> also to be complete.
> 
> With this done, we can remove the outer print completely.
> 
> Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
> ---
> Changes in v2:
> - Rebase on qcom for-next, drop dts patches which have been applied
> - Improve error printing situation (Bjorn)
> - Link to v1: https://lore.kernel.org/r/20240424-apcs-mboxes-v1-0-6556c47cb501@z3ntu.xyz
> ---
>  drivers/rpmsg/qcom_smd.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>


-- 
With best wishes
Dmitry

Re: [PATCH v2] rpmsg: qcom_smd: Improve error handling for qcom_smd_parse_edge

[Luca Weiss]

On Donnerstag, 6. Juni 2024 21:01:36 MESZ Luca Weiss wrote:
> When the mailbox driver has not probed yet, the error message "failed to
> parse smd edge" is just going to confuse users, so improve the error
> prints a bit.
> 
> Cover the last remaining exits from qcom_smd_parse_edge with proper
> error prints, especially the one for the mbox_chan deserved
> dev_err_probe to handle EPROBE_DEFER nicely. And add one for ipc_regmap
> also to be complete.
> 
> With this done, we can remove the outer print completely.

Ping, looks like this is still pending.

Regards
Luca

> 
> Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
> ---
> Changes in v2:
> - Rebase on qcom for-next, drop dts patches which have been applied
> - Improve error printing situation (Bjorn)
> - Link to v1: https://lore.kernel.org/r/20240424-apcs-mboxes-v1-0-6556c47cb501@z3ntu.xyz
> ---
>  drivers/rpmsg/qcom_smd.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/rpmsg/qcom_smd.c b/drivers/rpmsg/qcom_smd.c
> index 43f601c84b4f..06e6ba653ea1 100644
> --- a/drivers/rpmsg/qcom_smd.c
> +++ b/drivers/rpmsg/qcom_smd.c
> @@ -1369,7 +1369,8 @@ static int qcom_smd_parse_edge(struct device *dev,
>  	edge->mbox_chan = mbox_request_channel(&edge->mbox_client, 0);
>  	if (IS_ERR(edge->mbox_chan)) {
>  		if (PTR_ERR(edge->mbox_chan) != -ENODEV) {
> -			ret = PTR_ERR(edge->mbox_chan);
> +			ret = dev_err_probe(dev, PTR_ERR(edge->mbox_chan),
> +					    "failed to acquire IPC mailbox\n");
>  			goto put_node;
>  		}
>  
> @@ -1386,6 +1387,7 @@ static int qcom_smd_parse_edge(struct device *dev,
>  		of_node_put(syscon_np);
>  		if (IS_ERR(edge->ipc_regmap)) {
>  			ret = PTR_ERR(edge->ipc_regmap);
> +			dev_err(dev, "failed to get regmap from syscon: %d\n", ret);
>  			goto put_node;
>  		}
>  
> @@ -1501,10 +1503,8 @@ struct qcom_smd_edge *qcom_smd_register_edge(struct device *parent,
>  	}
>  
>  	ret = qcom_smd_parse_edge(&edge->dev, node, edge);
> -	if (ret) {
> -		dev_err(&edge->dev, "failed to parse smd edge\n");
> +	if (ret)
>  		goto unregister_dev;
> -	}
>  
>  	ret = qcom_smd_create_chrdev(edge);
>  	if (ret) {
> 
> ---
> base-commit: 2c79712cc83b172ce26c3086ced1c1fae087d8fb
> change-id: 20240423-apcs-mboxes-12ee6c01a5b3
> 
> Best regards,
> 

Re: [PATCH v2] rpmsg: qcom_smd: Improve error handling for qcom_smd_parse_edge

[Luca Weiss]

On 10/19/24 12:06 PM, Luca Weiss wrote:
> On Donnerstag, 6. Juni 2024 21:01:36 MESZ Luca Weiss wrote:
>> When the mailbox driver has not probed yet, the error message "failed to
>> parse smd edge" is just going to confuse users, so improve the error
>> prints a bit.
>>
>> Cover the last remaining exits from qcom_smd_parse_edge with proper
>> error prints, especially the one for the mbox_chan deserved
>> dev_err_probe to handle EPROBE_DEFER nicely. And add one for ipc_regmap
>> also to be complete.
>>
>> With this done, we can remove the outer print completely.
> 
> Ping, looks like this is still pending.

Ping again, still pending.

Regards
Luca

> 
> Regards
> Luca
> 
>>
>> Signed-off-by: Luca Weiss <luca@z3ntu.xyz>
>> ---
>> Changes in v2:
>> - Rebase on qcom for-next, drop dts patches which have been applied
>> - Improve error printing situation (Bjorn)
>> - Link to v1: https://lore.kernel.org/r/20240424-apcs-mboxes-v1-0-6556c47cb501@z3ntu.xyz
>> ---
>>   drivers/rpmsg/qcom_smd.c | 8 ++++----
>>   1 file changed, 4 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/rpmsg/qcom_smd.c b/drivers/rpmsg/qcom_smd.c
>> index 43f601c84b4f..06e6ba653ea1 100644
>> --- a/drivers/rpmsg/qcom_smd.c
>> +++ b/drivers/rpmsg/qcom_smd.c
>> @@ -1369,7 +1369,8 @@ static int qcom_smd_parse_edge(struct device *dev,
>>   	edge->mbox_chan = mbox_request_channel(&edge->mbox_client, 0);
>>   	if (IS_ERR(edge->mbox_chan)) {
>>   		if (PTR_ERR(edge->mbox_chan) != -ENODEV) {
>> -			ret = PTR_ERR(edge->mbox_chan);
>> +			ret = dev_err_probe(dev, PTR_ERR(edge->mbox_chan),
>> +					    "failed to acquire IPC mailbox\n");
>>   			goto put_node;
>>   		}
>>   
>> @@ -1386,6 +1387,7 @@ static int qcom_smd_parse_edge(struct device *dev,
>>   		of_node_put(syscon_np);
>>   		if (IS_ERR(edge->ipc_regmap)) {
>>   			ret = PTR_ERR(edge->ipc_regmap);
>> +			dev_err(dev, "failed to get regmap from syscon: %d\n", ret);
>>   			goto put_node;
>>   		}
>>   
>> @@ -1501,10 +1503,8 @@ struct qcom_smd_edge *qcom_smd_register_edge(struct device *parent,
>>   	}
>>   
>>   	ret = qcom_smd_parse_edge(&edge->dev, node, edge);
>> -	if (ret) {
>> -		dev_err(&edge->dev, "failed to parse smd edge\n");
>> +	if (ret)
>>   		goto unregister_dev;
>> -	}
>>   
>>   	ret = qcom_smd_create_chrdev(edge);
>>   	if (ret) {
>>
>> ---
>> base-commit: 2c79712cc83b172ce26c3086ced1c1fae087d8fb
>> change-id: 20240423-apcs-mboxes-12ee6c01a5b3
>>
>> Best regards,
>>
> 
> 
> 
>