Re: [PATCH v2 4/4] drm/bridge: nwl-dsi: Add error handling of nwl_dsi_mode_set()

Thank you for your contribution! Sashiko AI review found 1 potential issue(=
s) to consider:
- [Medium] Handling `nwl_dsi_mode_set()` error by jumping to `runtime_put` =
causes a `pm_runtime` usage counter underflow.
--

--- Patch [4]: [PATCH v2 4/4] drm/bridge: nwl-dsi: Add error handling of nw=
l_dsi_mode_set() ---
commit 7d9af739876057a4f21f4e98151a3fe153456a02
Author: Esben Haabendal <[email protected]>

drm/bridge: nwl-dsi: Add error handling of nwl_dsi_mode_set()

This commit introduces error handling for nwl_dsi_mode_set() in
nwl_dsi_bridge_mode_set(). If the function fails, it jumps to the
runtime_put label to release the PM runtime reference.

> diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nw=
l-dsi.c
> index 226036d22e3f8..08e87d166b17a 100644
> --- a/drivers/gpu/drm/bridge/nwl-dsi.c
> +++ b/drivers/gpu/drm/bridge/nwl-dsi.c
> @@ -942,7 +942,9 @@ nwl_dsi_bridge_mode_set(struct drm_bridge *bridge,
>  	}
> =20
>  	/* Step 2 from DSI reset-out instructions */
> -	nwl_dsi_mode_set(dsi);
> +	ret =3D nwl_dsi_mode_set(dsi);
> +	if (ret < 0)
> +		goto runtime_put;

[Severity: Medium]
Since nwl_dsi_bridge_mode_set() returns void, does the DRM core know that t=
his
setup failed?

If the DRM core continues treating the bridge as active, it will unconditio=
nally
call nwl_dsi_bridge_atomic_disable() during teardown, which also drops the
reference:

drivers/gpu/drm/bridge/nwl-dsi.c:nwl_dsi_bridge_atomic_disable() {
	...
	pm_runtime_put(dsi->dev);
	...
}

Could this double-put cause a pm_runtime usage counter underflow, since we=
=20
already dropped the reference in this new error path?

> =20
>  	/* Step 3 from DSI reset-out instructions */
>  	ret =3D reset_control_deassert(dsi->rst_esc);

--=20
Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20260803-nwl-dsi-fi=
[email protected]?part=3D4