Re: [PATCH v3 03/14] irqchip/csky-mpintc: Fix potential resource leaks

Re: [PATCH v3 03/14] irqchip/csky-mpintc: Fix potential resource leaks

[Markus Elfring]

> There exists potential resource leaks in the error path, fix them.

I suggest to improve this change description.
How do you think about another wording variant?

   Specific system resources were not released in a few error cases.
   Thus add jump targets for the completion of the desired exception handling.


…
+++ b/drivers/irqchip/irq-csky-mpintc.c
@@ -247,8 +247,10 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
…
 		INTCG_base = ioremap(mfcr("cr<31, 14>"),
 				     INTCL_SIZE*nr_cpu_ids + INTCG_SIZE);

See also:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/coding-style.rst?id=3e08a95294a4fb3702bb3d35ed08028433c37fe6#n257


…
+		if (INTCG_base == NULL) {


Would you like to use the following code variant?

See also:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/scripts/checkpatch.pl?id=3e08a95294a4fb3702bb3d35ed08028433c37fe6#n5756

+		if (!INTCG_base) {


…
> @@ -270,12 +274,22 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
>
>  #ifdef CONFIG_SMP
>  	ipi_irq = irq_create_mapping(root_domain, IPI_IRQ);
…
> +	if (!ipi_irq) {
> +		ret = -EIO;
> +		goto err_domain_remove;
> +	}
…

Can the function call “irq_domain_remove(root_domain)” become relevant
only if the preprocessor symbol “CONFIG_SMP” was defined?

Regards,
Markus

Re: [PATCH v3 03/14] irqchip/csky-mpintc: Fix potential resource leaks

[Tiezhu Yang]

On 06/24/2020 07:43 PM, Markus Elfring wrote:
>> There exists potential resource leaks in the error path, fix them.
> I suggest to improve this change description.
> How do you think about another wording variant?
>
>     Specific system resources were not released in a few error cases.
>     Thus add jump targets for the completion of the desired exception handling.

OK, thank you, it looks good to me.
Maybe I can use this description for other patches of this series.

>
>
> …
> +++ b/drivers/irqchip/irq-csky-mpintc.c
> @@ -247,8 +247,10 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
> …
>   		INTCG_base = ioremap(mfcr("cr<31, 14>"),
>   				     INTCL_SIZE*nr_cpu_ids + INTCG_SIZE);
>
> See also:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/process/coding-style.rst?id=3e08a95294a4fb3702bb3d35ed08028433c37fe6#n257
>
>
> …
> +		if (INTCG_base == NULL) {
>
>
> Would you like to use the following code variant?
>
> See also:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/scripts/checkpatch.pl?id=3e08a95294a4fb3702bb3d35ed08028433c37fe6#n5756
>
> +		if (!INTCG_base) {
>
>
> …
>> @@ -270,12 +274,22 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
>>
>>   #ifdef CONFIG_SMP
>>   	ipi_irq = irq_create_mapping(root_domain, IPI_IRQ);
> …
>> +	if (!ipi_irq) {
>> +		ret = -EIO;
>> +		goto err_domain_remove;
>> +	}
> …
>
> Can the function call “irq_domain_remove(root_domain)” become relevant
> only if the preprocessor symbol “CONFIG_SMP” was defined?

OK, I will do it in the next version.

>
> Regards,
> Markus

Re: [v3 03/14] irqchip/csky-mpintc: Fix potential resource leaks

[Markus Elfring]

>> How do you think about another wording variant?
>>
>>     Specific system resources were not released in a few error cases.
>>     Thus add jump targets for the completion of the desired exception handling.
>
> OK, thank you, it looks good to me.

Thanks for your positive feedback.


> Maybe I can use this description for other patches of this series.

Would you like to make any of the affected commit messages more “specific”?

Regards,
Markus

[PATCH v3 03/14] irqchip/csky-mpintc: Fix potential resource leaks

[Tiezhu Yang]

There exists potential resource leaks in the error path, fix them.

Fixes: d8a5f5f79122 ("irqchip: add C-SKY SMP interrupt controller")
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
---
 drivers/irqchip/irq-csky-mpintc.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/irqchip/irq-csky-mpintc.c b/drivers/irqchip/irq-csky-mpintc.c
index a1534ed..c195e24 100644
--- a/drivers/irqchip/irq-csky-mpintc.c
+++ b/drivers/irqchip/irq-csky-mpintc.c
@@ -247,8 +247,10 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
 	if (INTCG_base == NULL) {
 		INTCG_base = ioremap(mfcr("cr<31, 14>"),
 				     INTCL_SIZE*nr_cpu_ids + INTCG_SIZE);
-		if (INTCG_base == NULL)
-			return -EIO;
+		if (INTCG_base == NULL) {
+			ret = -EIO;
+			goto err_free;
+		}
 
 		INTCL_base = INTCG_base + INTCG_SIZE;
 
@@ -257,8 +259,10 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
 
 	root_domain = irq_domain_add_linear(node, nr_irq, &csky_irqdomain_ops,
 					    NULL);
-	if (!root_domain)
-		return -ENXIO;
+	if (!root_domain) {
+		ret = -ENXIO;
+		goto err_iounmap;
+	}
 
 	/* for every cpu */
 	for_each_present_cpu(cpu) {
@@ -270,12 +274,22 @@ csky_mpintc_init(struct device_node *node, struct device_node *parent)
 
 #ifdef CONFIG_SMP
 	ipi_irq = irq_create_mapping(root_domain, IPI_IRQ);
-	if (!ipi_irq)
-		return -EIO;
+	if (!ipi_irq) {
+		ret = -EIO;
+		goto err_domain_remove;
+	}
 
 	set_send_ipi(&csky_mpintc_send_ipi, ipi_irq);
 #endif
 
 	return 0;
+
+err_domain_remove:
+	irq_domain_remove(root_domain);
+err_iounmap:
+	iounmap(INTCG_base);
+err_free:
+	kfree(__trigger);
+	return ret;
 }
 IRQCHIP_DECLARE(csky_mpintc, "csky,mpintc", csky_mpintc_init);
-- 
2.1.0