linux-kernel - Re: [PATCH -next 1/5] drm/rockchip: Use for_each_child_of_node

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]

Message-ID: <ab3c09e7-43cc-b946-0f7c-44ea7f4111f2@huawei.com>
Date: Mon, 26 Aug 2024 17:04:09 +0800
From: Jinjie Ruan <ruanjinjie@...wei.com>
To: Jonathan Cameron <Jonathan.Cameron@...wei.com>
CC: <hjc@...k-chips.com>, <heiko@...ech.de>, <andy.yan@...k-chips.com>,
	<maarten.lankhorst@...ux.intel.com>, <mripard@...nel.org>,
	<tzimmermann@...e.de>, <airlied@...il.com>, <daniel@...ll.ch>,
	<dri-devel@...ts.freedesktop.org>, <linux-arm-kernel@...ts.infradead.org>,
	<linux-rockchip@...ts.infradead.org>, <linux-kernel@...r.kernel.org>,
	<krzk@...nel.org>, <jic23@...nel.org>
Subject: Re: [PATCH -next 1/5] drm/rockchip: Use
 for_each_child_of_node_scoped()



On 2024/8/23 19:32, Jonathan Cameron wrote:
> On Fri, 23 Aug 2024 17:20:49 +0800
> Jinjie Ruan <ruanjinjie@...wei.com> wrote:
> 
>> Avoids the need for manual cleanup of_node_put() in early exits
>> from the loop.
>>
>> Signed-off-by: Jinjie Ruan <ruanjinjie@...wei.com>
> 
> There is more to do here, and looking at the code, I'm far from
> sure it isn't releasing references it never had.
> 
>> ---
>>  drivers/gpu/drm/rockchip/rockchip_lvds.c | 8 +++-----
>>  1 file changed, 3 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/rockchip/rockchip_lvds.c b/drivers/gpu/drm/rockchip/rockchip_lvds.c
>> index 9a01aa450741..f5b3f18794dd 100644
>> --- a/drivers/gpu/drm/rockchip/rockchip_lvds.c
>> +++ b/drivers/gpu/drm/rockchip/rockchip_lvds.c
>> @@ -548,7 +548,7 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master,
>>  	struct drm_encoder *encoder;
>>  	struct drm_connector *connector;
>>  	struct device_node *remote = NULL;
>> -	struct device_node  *port, *endpoint;
> 
> Odd extra space before *port in original. Clean that up whilst here.
> 
> 
>> +	struct device_node  *port;
> 
> Use __free(device_node) for *port as well.
> 
> So where the current asignment is.
> 	struct device_node *port = of_graph_get_port_by_id(dev->of_node, 1);
> 
>>  	int ret = 0, child_count = 0;
>>  	const char *name;
>>  	u32 endpoint_id = 0;
>> @@ -560,15 +560,13 @@ static int rockchip_lvds_bind(struct device *dev, struct device *master,
>>  			      "can't found port point, please init lvds panel port!\n");
>>  		return -EINVAL;
>>  	}
>> -	for_each_child_of_node(port, endpoint) {
>> +	for_each_child_of_node_scoped(port, endpoint) {
>>  		child_count++;
>>  		of_property_read_u32(endpoint, "reg", &endpoint_id);
>>  		ret = drm_of_find_panel_or_bridge(dev->of_node, 1, endpoint_id,
>>  						  &lvds->panel, &lvds->bridge);
>> -		if (!ret) {
>> -			of_node_put(endpoint);
>> +		if (!ret)
>>  			break;
> 
> This then can simply be
> 			return dev_err_probe(dev, ret,
> 					     "failed to find pannel and bridge node\n");
>> -		}

Thank you! I'll resend and cleanup them.

> 
> Various other paths become direct returns as well.
> 
>>  	}
> 
> The later code with remote looks suspect as not obvious who got the reference that
> is being put but assuming that is correct, it's another possible place for __free based
> cleanup.
> 
> 
>>  	if (!child_count) {
>>  		DRM_DEV_ERROR(dev, "lvds port does not have any children\n");
>