Re: [PATCH] soc: amlogic: canvas: Fix meson_canvas_get when probe failed

From: Kevin Hilman <khilman@baylibre.com>
To: Maxime Jourdan <mjourdan@baylibre.com>,
	Neil Armstrong <narmstrong@baylibre.com>
Cc: linux-amlogic@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH] soc: amlogic: canvas: Fix meson_canvas_get when probe failed
Date: Thu, 07 Feb 2019 16:30:31 -0800	[thread overview]
Message-ID: <7h4l9fglzs.fsf@baylibre.com> (raw)
In-Reply-To: <CAMO6nawKrH3d2X=TikdJy8yjiPyuJJRnpaBjyfPYuRx3d=7vVA@mail.gmail.com>

Maxime Jourdan <mjourdan@baylibre.com> writes:

> On Mon, Feb 4, 2019 at 10:49 AM Neil Armstrong <narmstrong@baylibre.com> wrote:
>>
>> When probe fails, a platforn_device is still associated to the node,
>> but dev_get_drvdata() returns NULL.
>>
>> Handle this case by returning a consistent error.
>>
>> Fixes: d4983983d987 ("soc: amlogic: add meson-canvas driver")
>> Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
>> ---
>>  drivers/soc/amlogic/meson-canvas.c | 13 ++++++++++++-
>>  1 file changed, 12 insertions(+), 1 deletion(-)
>>
>> This fix depends on "soc: amlogic: add missing of_node_put()" at [1]
>>
>> [1] https://patchwork.kernel.org/patch/10794545/
>>
>> diff --git a/drivers/soc/amlogic/meson-canvas.c b/drivers/soc/amlogic/meson-canvas.c
>> index 87a6c6e23591..151aa84bc593 100644
>> --- a/drivers/soc/amlogic/meson-canvas.c
>> +++ b/drivers/soc/amlogic/meson-canvas.c
>> @@ -51,6 +51,7 @@ struct meson_canvas *meson_canvas_get(struct device *dev)
>>  {
>>         struct device_node *canvas_node;
>>         struct platform_device *canvas_pdev;
>> +       struct meson_canvas *canvas;
>>
>>         canvas_node = of_parse_phandle(dev->of_node, "amlogic,canvas", 0);
>>         if (!canvas_node)
>> @@ -63,7 +64,17 @@ struct meson_canvas *meson_canvas_get(struct device *dev)
>>         }
>>
>>         of_node_put(canvas_node);
>> -       return dev_get_drvdata(&canvas_pdev->dev);
>> +
>> +       /*
>> +        * If priv is NULL, it's probably because the canvas hasn't
>> +        * properly initialized. Bait out with -EINVAL because, in the
>
> Bait -> Bail ?

Fixed this up locally.

[...]

> Thanks for the patch, I missed the fact that a driver could have a
> pdev yet with a failed probe.
>
> Reviewed-by: Maxime Jourdan <mjourdan@baylibre.com>

Queud for v5.1 with Maxime's tag. (branch: v5.1/drivers)

Kevin

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic