linux-kernel - Re: [PATCH v3 linux-next] leds: powernv: replace of_node_put to _

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Fri, 7 Jun 2024 08:25:38 -0300
From: Marilene Andrade Garcia <marilene.agarcia@...il.com>
To: Pavel Machek <pavel@....cz>, Lee Jones <lee@...nel.org>,
 Michael Ellerman <mpe@...erman.id.au>, Nicholas Piggin <npiggin@...il.com>,
 Christophe Leroy <christophe.leroy@...roup.eu>,
 "Naveen N. Rao" <naveen.n.rao@...ux.ibm.com>,
 Julia Lawall <julia.lawall@...ia.fr>
Cc: Shuah Khan <skhan@...uxfoundation.org>,
 Javier Carrasco <javier.carrasco.cruz@...il.com>,
 linux-leds@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 linux-next] leds: powernv: replace of_node_put to
 __free

On 01/06/2024 00:17, MarileneGarcia wrote:
> Use __free for device_node values, and thus drop calls to
> of_node_put.
> 
> The variable attribute __free adds a scope based cleanup to
> the device node. The goal is to reduce memory management issues
> in the kernel code.
> 
> The of_node_put calls were removed, and the
> for_each_available_child_of_node was replaced to the equivalent
> for_each_available_child_of_node_scoped which use the __free.
> 
> Suggested-by: Julia Lawall <julia.lawall@...ia.fr>
> Signed-off-by: MarileneGarcia <marilene.agarcia@...il.com>
> ---
> Hello,
> Thank you for the feedback.
> 
> The line-break strategy was fixed, and now it is according to
> the top one suggested.
> 
> The __free is a wrapper to __attribute__((__cleanup__())) so
> the variavel definition is needed. And according to Julia, it
> is preferred to combine the declaration and the allocation to
> ensure that there is no return that can occur after the declaration,
> but before the allocation (or more precisely the initialization).
> If there is no other option for the initialization of the variable,
> then it should be NULL.
> 
>   drivers/leds/leds-powernv.c | 28 +++++++++-------------------
>   1 file changed, 9 insertions(+), 19 deletions(-)
> 
> diff --git a/drivers/leds/leds-powernv.c b/drivers/leds/leds-powernv.c
> index 4f01acb75727..49ab8c9a3f29 100644
> --- a/drivers/leds/leds-powernv.c
> +++ b/drivers/leds/leds-powernv.c
> @@ -246,29 +246,25 @@ static int powernv_led_classdev(struct platform_device *pdev,
>   	const char *cur = NULL;
>   	int rc = -1;
>   	struct property *p;
> -	struct device_node *np;
>   	struct powernv_led_data *powernv_led;
>   	struct device *dev = &pdev->dev;
>   
> -	for_each_available_child_of_node(led_node, np) {
> +	for_each_available_child_of_node_scoped(led_node, np) {
>   		p = of_find_property(np, "led-types", NULL);
>   
>   		while ((cur = of_prop_next_string(p, cur)) != NULL) {
>   			powernv_led = devm_kzalloc(dev, sizeof(*powernv_led),
>   						   GFP_KERNEL);
> -			if (!powernv_led) {
> -				of_node_put(np);
> +			if (!powernv_led)
>   				return -ENOMEM;
> -			}
>   
>   			powernv_led->common = powernv_led_common;
>   			powernv_led->loc_code = (char *)np->name;
>   
>   			rc = powernv_led_create(dev, powernv_led, cur);
> -			if (rc) {
> -				of_node_put(np);
> +			if (rc)
>   				return rc;
> -			}
> +
>   		} /* while end */
>   	}
>   
> @@ -278,12 +274,11 @@ static int powernv_led_classdev(struct platform_device *pdev,
>   /* Platform driver probe */
>   static int powernv_led_probe(struct platform_device *pdev)
>   {
> -	struct device_node *led_node;
>   	struct powernv_led_common *powernv_led_common;
>   	struct device *dev = &pdev->dev;
> -	int rc;
> +	struct device_node *led_node
> +		__free(device_node) = of_find_node_by_path("/ibm,opal/leds");
>   
> -	led_node = of_find_node_by_path("/ibm,opal/leds");
>   	if (!led_node) {
>   		dev_err(dev, "%s: LED parent device node not found\n",
>   			__func__);
> @@ -292,20 +287,15 @@ static int powernv_led_probe(struct platform_device *pdev)
>   
>   	powernv_led_common = devm_kzalloc(dev, sizeof(*powernv_led_common),
>   					  GFP_KERNEL);
> -	if (!powernv_led_common) {
> -		rc = -ENOMEM;
> -		goto out;
> -	}
> +	if (!powernv_led_common)
> +		return -ENOMEM;
>   
>   	mutex_init(&powernv_led_common->lock);
>   	powernv_led_common->max_led_type = cpu_to_be64(OPAL_SLOT_LED_TYPE_MAX);
>   
>   	platform_set_drvdata(pdev, powernv_led_common);
>   
> -	rc = powernv_led_classdev(pdev, led_node, powernv_led_common);
> -out:
> -	of_node_put(led_node);
> -	return rc;
> +	return powernv_led_classdev(pdev, led_node, powernv_led_common);
>   }
>   
>   /* Platform driver remove */

Hello,
Did you have a chance to look at the patch after the requested change?

Thank you,
Marilene