linux-kernel - Re: [PATCH linux-next] media: i2c: st-mipid02: replace of_node_put() with _

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] [day] [month] [year] [list]

Date: Mon, 29 Apr 2024 21:32:35 +0530
From: R Sundar <prosunofficial@...il.com>
To: Benjamin Mugnier <benjamin.mugnier@...s.st.com>,
 sylvain.petinot@...s.st.com, mchehab@...nel.org,
 Laurent Pinchart <laurent.pinchart@...asonboard.com>,
 Sakari Ailus <sakari.ailus@...ux.intel.com>
Cc: linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
 skhan@...uxfoundation.org, javier.carrasco.cruz@...il.com,
 Julia Lawall <julia.lawall@...ia.fr>
Subject: Re: [PATCH linux-next] media: i2c: st-mipid02: replace of_node_put()
 with __free

On 29/04/24 14:22, Benjamin Mugnier wrote:
> Hi,
> 
> Thank you for your patch.
> 
> On 4/27/24 11:56, R Sundar wrote:
>> Use the new cleanup magic to replace of_node_put() with
>> __free(device_node) marking to auto release and to simplify the error
>> paths.
>>
>> Suggested-by: Julia Lawall <julia.lawall@...ia.fr>
>> Signed-off-by: R Sundar <prosunofficial@...il.com>
> 
> I was not aware of this kind of auto release mechanism. Thanks for
> bringing that to my eyes.
> 
> Now I looked in /drivers/media and couldn't find such structure. All
> drivers seem to follow the goto error_of_node_put style.
> As I'm unsure if we want to introduce such magic, could either Laurent
> or Sakari comment on this ?
> 
>> ---
>>   drivers/media/i2c/st-mipid02.c | 37 +++++++++-------------------------
>>   1 file changed, 9 insertions(+), 28 deletions(-)
>>
>> diff --git a/drivers/media/i2c/st-mipid02.c b/drivers/media/i2c/st-mipid02.c
>> index f250640729ca..d42a306530f3 100644
>> --- a/drivers/media/i2c/st-mipid02.c
>> +++ b/drivers/media/i2c/st-mipid02.c
>> @@ -715,31 +715,28 @@ static int mipid02_parse_rx_ep(struct mipid02_dev *bridge)
>>   	struct v4l2_fwnode_endpoint ep = { .bus_type = V4L2_MBUS_CSI2_DPHY };
>>   	struct i2c_client *client = bridge->i2c_client;
>>   	struct v4l2_async_connection *asd;
>> -	struct device_node *ep_node;
>>   	int ret;
>>   
>>   	/* parse rx (endpoint 0) */
>> -	ep_node = of_graph_get_endpoint_by_regs(bridge->i2c_client->dev.of_node,
>> -						0, 0);
>> +	struct device_node *ep_node __free(device_node) =
>> +		of_graph_get_endpoint_by_regs(bridge->i2c_client->dev.of_node, 0, 0);
>>   	if (!ep_node) {
>>   		dev_err(&client->dev, "unable to find port0 ep");
>> -		ret = -EINVAL;
>> -		goto error;
>> +		return -EINVAL;
>>   	}
>>   
>>   	ret = v4l2_fwnode_endpoint_parse(of_fwnode_handle(ep_node), &ep);
>>   	if (ret) {
>>   		dev_err(&client->dev, "Could not parse v4l2 endpoint %d\n",
>>   			ret);
>> -		goto error_of_node_put;
>> +		return ret;
>>   	}
>>   
>>   	/* do some sanity checks */
>>   	if (ep.bus.mipi_csi2.num_data_lanes > 2) {
>>   		dev_err(&client->dev, "max supported data lanes is 2 / got %d",
>>   			ep.bus.mipi_csi2.num_data_lanes);
>> -		ret = -EINVAL;
>> -		goto error_of_node_put;
>> +		return -EINVAL;
>>   	}
>>   
>>   	/* register it for later use */
>> @@ -750,7 +747,6 @@ static int mipid02_parse_rx_ep(struct mipid02_dev *bridge)
>>   	asd = v4l2_async_nf_add_fwnode_remote(&bridge->notifier,
>>   					      of_fwnode_handle(ep_node),
>>   					      struct v4l2_async_connection);
>> -	of_node_put(ep_node);
>>   
>>   	if (IS_ERR(asd)) {
>>   		dev_err(&client->dev, "fail to register asd to notifier %ld",
>> @@ -764,46 +760,31 @@ static int mipid02_parse_rx_ep(struct mipid02_dev *bridge)
>>   		v4l2_async_nf_cleanup(&bridge->notifier);
>>   
>>   	return ret;
>> -
>> -error_of_node_put:
>> -	of_node_put(ep_node);
>> -error:
>> -
>> -	return ret;
>>   }
>>   
>>   static int mipid02_parse_tx_ep(struct mipid02_dev *bridge)
>>   {
>>   	struct v4l2_fwnode_endpoint ep = { .bus_type = V4L2_MBUS_PARALLEL };
>>   	struct i2c_client *client = bridge->i2c_client;
>> -	struct device_node *ep_node;
>>   	int ret;
>>   
>>   	/* parse tx (endpoint 2) */
>> -	ep_node = of_graph_get_endpoint_by_regs(bridge->i2c_client->dev.of_node,
>> -						2, 0);
>> +	struct device_node *ep_node =
>> +		of_graph_get_endpoint_by_regs(bridge->i2c_client->dev.of_node, 2, 0);
> 
> Shouldn't there be a '__free' here too ?
> 

my Bad.
Yes, It should come here also. Thanks!!
Will update the patch.

Thanks,
Sundar