Re: [PATCH 2/2] drm/exynos: dp: fix wrong return type

[Javier Martinez Canillas <javier@osg.samsung.com>]

[adding Ajay Kumar who added the bridge support]

Hello Inki,

On 11/26/2015 09:47 AM, Inki Dae wrote:
> This patch fixes wrong return type when dt binding of bridge device
> failed.
> 
> If a board has a bridge device then of_graph_get_remote_port_parent
> function shouldn't be NULL. So this patch will return a proper error
> type so that the deferred probe isn't triggered.
> 
> Signed-off-by: Inki Dae <inki.dae@samsung.com>
> ---
>  drivers/gpu/drm/exynos/exynos_dp_core.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c
> index 0b53045..c77fb83 100644
> --- a/drivers/gpu/drm/exynos/exynos_dp_core.c
> +++ b/drivers/gpu/drm/exynos/exynos_dp_core.c
> @@ -1436,8 +1436,10 @@ static int exynos_dp_probe(struct platform_device *pdev)
>  			of_node_put(bridge_node);
>  			if (!dp->ptn_bridge)
>  				return -EPROBE_DEFER;
> -		} else
> -			return -EPROBE_DEFER;
> +		} else {
> +			DRM_ERROR("no port node for bridge device.\n");
> +			return -ENXIO;
> +		}
>  	}
>  

As I mentioned in the other thread, I wonder if -ENXIO is the best errno
code in this case. Shouldn't -EINVAL be more appropriate since is about
an invalid DTB?

>  	pm_runtime_enable(dev);
> 

Best regards,
-- 
Javier Martinez Canillas
Open Source Group
Samsung Research America