[PATCH v6 11/26] drm/bridge: samsung-dsim: use devm_drm_of_get_bridge[_by_node] to find the out_bridge

[Maxime Ripard]

On Thu, Feb 06, 2025 at 07:14:26PM +0100, Luca Ceresoli wrote:
> In order to support panels described either via graph links or via a
> subnode (e.g. "panel at 0"), this driver uses low-level deprecated functions
> to find the next bridge. The resulting logic is complex and duplicates code
> already present in the DRM bridge core. Switch to the new APIs in DRM
> bridge core that allow to do the same in a much cleaner way.
> 
> Note there are two slight changes in the new logic intended to improve the
> final result:
> 
>  * the old code looks for a subnode with any name except "port" or "ports",
>    while the new code uses the node passed as a parameter
> 
>  * the old code looks for a subnode first and falls back to a graph link,
>    while the new code uses the reverse order because graph links are the
>    recommended device tree representation now
> 
> The first change makes the code more robust by avoiding the risk of using
> an unrelated node which is not describing a panel and not names "port" or
> "ports".
> 
> The second change is not expected to expose regressions because, in the
> cases where both a subnode and a graph link are used to describe a panel,
> the graph link should point to the subnode itself, such as in
> arch/arm64/boot/dts/qcom/sdm850-lenovo-yoga-c630.dts
> 
> As a further cleanup, use a temporary variable to assign dsi->out_bridge
> only on success. This avoids the risk of leaving a non-NULL value in
> dsi->out_bridge when samsung_dsim_host_attach() fails.
> 
> Suggested-by: Dmitry Baryshkov <dmitry.baryshkov at linaro.org>
> Signed-off-by: Luca Ceresoli <luca.ceresoli at bootlin.com>
> 
> ---
> 
> This patch was added in v6.
> ---
>  drivers/gpu/drm/bridge/samsung-dsim.c | 55 ++++++-----------------------------
>  1 file changed, 9 insertions(+), 46 deletions(-)
> 
> diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
> index f8b4fb8357659018ec0db65374ee5d05330639ae..bbd0a4f5a3f52b61bf48f10d6e8ca741bffa5e46 100644
> --- a/drivers/gpu/drm/bridge/samsung-dsim.c
> +++ b/drivers/gpu/drm/bridge/samsung-dsim.c
> @@ -1704,55 +1704,16 @@ static int samsung_dsim_host_attach(struct mipi_dsi_host *host,
>  	const struct samsung_dsim_plat_data *pdata = dsi->plat_data;
>  	struct device *dev = dsi->dev;
>  	struct device_node *np = dev->of_node;
> -	struct device_node *remote;
> -	struct drm_panel *panel;
> +	struct drm_bridge *out_bridge;
>  	int ret;
>  
> -	/*
> -	 * Devices can also be child nodes when we also control that device
> -	 * through the upstream device (ie, MIPI-DCS for a MIPI-DSI device).
> -	 *
> -	 * Lookup for a child node of the given parent that isn't either port
> -	 * or ports.
> -	 */
> -	for_each_available_child_of_node(np, remote) {
> -		if (of_node_name_eq(remote, "port") ||
> -		    of_node_name_eq(remote, "ports"))
> -			continue;
> +	out_bridge = devm_drm_of_get_bridge(dev, np, 1, 0);
> +	if (IS_ERR(out_bridge) && PTR_ERR(out_bridge) != -EPROBE_DEFER)
> +		out_bridge = devm_drm_of_get_bridge_by_node(dev, device->dev.of_node);
>  

For the same reason I mentioned earlier, this is inherently unsafe if
the bridge device goes away but the DRM device doesn't.

Maxime
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 273 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20250210/23284f0f/attachment.sig>