From mboxrd@z Thu Jan 1 00:00:00 1970 From: maitysanchayan@gmail.com (maitysanchayan at gmail.com) Date: Sun, 23 Aug 2015 21:01:48 +0530 Subject: [PATCH v4 3/3] touchscreen: colibri-vf50-ts: Add touchscreen support for Colibri VF50 In-Reply-To: References: <915c9122cfba83c6fbbc519c10ae2d101e27bc5b.1440162473.git.maitysanchayan@gmail.com> Message-ID: <20150823153148.GB3935@Sanchayan-Arch> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 15-08-22 18:52:28, Stefan Agner wrote: > Hi Sanchayan, > > On 2015-08-21 06:26, Sanchayan Maity wrote: > > +static int vf50_ts_probe(struct platform_device *pdev) > > +{ > > + struct input_dev *input; > > + struct iio_channel *channels; > > + struct device *dev = &pdev->dev; > > + struct vf50_touch_device *touchdev; > > + int error; > > + > > + channels = iio_channel_get_all(dev); > > + if (IS_ERR(channels)) > > + return PTR_ERR(channels); > > Hm, we expect here that four channels are returned, however a faulty > device tree could contain less. Access to the fourth channel above would > lead to a bug. > > Can you check the amount of channels returned? The returned list is > explicitly terminated with a null entry, you can rely on that. Something > similar to hwmon should do the job. > http://lxr.free-electrons.com/source/drivers/hwmon/iio_hwmon.c#L86 I agree. It did be nice to have this error check. Thanks. I will add this check so we can bail out if less channels are specified. - Sanchayan. > > > Otherwise, Acked-by: Stefan Agner > > -- > Stefan