Re: [PATCH 1/1] drivers/i2c: fix a potential null pointer dereference

["Théo Lebrun" <theo.lebrun@bootlin.com>]

Hello Linus, Yuanjun,

On Wed May 14, 2025 at 12:10 AM CEST, Linus Walleij wrote:
> On Tue, May 13, 2025 at 2:38 PM Yuanjun Gong <ruc_gongyuanjun@163.com> wrote:
>
>> Add check to *priv to make sure the input *dev is not NULL,
>> and to avoid a potential null pointer dereference.
>>
>> Signed-off-by: Yuanjun Gong <ruc_gongyuanjun@163.com>
>
> Why is this a problem? Reading nmk_i2c_probe():
>
>         priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
>         if (!priv)
>                 return -ENOMEM;
>
> OK so priv is not NULL.
>
>         amba_set_drvdata(adev, priv);
>
> It is unconditionally assigned to the AMBA device.
>
> Surely runtime resume isn't called before probe() has commenced
> successfully?
>
> I don't understand when this condition can occur.
>
> Also including Theo on this, he is the major industrial user of this driver now.

Thanks for the Cc; I agree with your reading of the code and don't see
the issue.

Thanks,

--
Théo Lebrun, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com