Received-SPF: pass (google.com: domain of linux-kernel+bounces-183282-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Mon, 20 May 2024 00:44:40 +0300
From: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
To: Sui Jingfeng <sui.jingfeng@linux.dev>
Cc: Maxime Ripard <mripard@kernel.org>, 
	Neil Armstrong <neil.armstrong@linaro.org>, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 0/2] drm/bridge: Add 'struct device *' field to the
 drm_bridge structure
Message-ID: <6khhnjtdcjvebnffrqlavowld4gvgcpnxplcinkja5xv3yefct@vjnmj72dfqtg>
References: <20240514154045.309925-1-sui.jingfeng@linux.dev>
 <20240514-scarlet-corgi-of-efficiency-faf2bb@penduick>
 <c44480ab-8d6b-4334-8eba-83db9b30ff1a@linux.dev>
 <20240515-fair-satisfied-myna-480dea@penduick>
 <d394ee32-4fa4-41a8-a5ca-c1c7f77f44d2@linux.dev>
 <20240515-copper-chimpanzee-of-fortitude-ff3dab@penduick>
 <2c15c859-6b2b-4979-8317-698bf6cc430c@linux.dev>
 <20240516-intrepid-uptight-tench-0df95e@penduick>
 <e955b706-04dd-479f-8327-32771d94f70f@linux.dev>
 <07b931d9-b11f-4829-a6c4-d5025fc071b5@linux.dev>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <07b931d9-b11f-4829-a6c4-d5025fc071b5@linux.dev>

On Thu, May 16, 2024 at 08:04:59PM +0800, Sui Jingfeng wrote:
> 
> 
> On 5/16/24 18:40, Sui Jingfeng wrote:
> > use 'to_i2c_client(bridge->dev)' to retrieve the pointer
> 
> to_i2c_client(bridge->kdev).
> 
> Besides, this also means that we don't need to add the fwnode
> pointer into struct drm_bridge as member. Relief the conflicts
> with other reviewers if the work of switching to fwnode is still
> needed. As for majorities cases (1 to 1), of_node and fwnode can
> be retrieved with 'struct device *' easily. The aux-bridge.c and
> aux-hdp-bridge.c can also be converted too easily.
> 
> of_node, fwnode, swnode and device properties are all belong to
> the backing device structure itself. It can be more natural to use
> device_proterty_read_xxx() APIs after init time, Which in turn
> avoid the need to acquire and duplicate all properties another
> time in the driver private structure.

This doesn't sound 100% correct. This is going to drop the possibile
case when bridge driver uses child DT or FW node under the main device
node. For example of such usecase see drivers/usb/typec/tcpm/qcom/qcom_pmic_typec.c

> 
> We could do the programming around the 'struct device *.', remove
> a batch of boilerplate.

-- 
With best wishes
Dmitry