> -----Original Message-----
> From: Greg KH [mailto:gre...@linuxfoundation.org]
> Sent: Thursday 6 June 2019 14:29
> To: Dragan Cvetic <drag...@xilinx.com>
> Cc: a...@arndb.de; Michal Simek <mich...@xilinx.com>;
> linux-arm-ker...@lists.infradead.org; robh...@kernel.org;
> mark.rutl...@arm.com; devicet...@vger.kernel.org;
> linux-kernel@vger.kernel.org; Derek Kiernan <dkier...@xilinx.com>
> Subject: Re: [PATCH V4 04/12] misc: xilinx_sdfec: Add open, close and ioctl
>
> On Sat, May 25, 2019 at 12:37:17PM +0100, Dragan Cvetic wrote:
> > +static int xsdfec_dev_open(struct inode *iptr, struct file *fptr)
> > +{
> > + return 0;
> > +}
> > +
> > +static int xsdfec_dev_release(struct inode *iptr, struct file *fptr)
> > +{
> > + return 0;
> > +}
>
> empty open/close functions are never needed, just drop them.
Accepted.
Will remove them.
>
> > +
> > +static long xsdfec_dev_ioctl(struct file *fptr, unsigned int cmd,
> > + unsigned long data)
> > +{
> > + struct xsdfec_dev *xsdfec;
> > + void __user *arg = NULL;
> > + int rval = -EINVAL;
> > +
> > + xsdfec = container_of(fptr->private_data, struct xsdfec_dev, miscdev);
> > + if (!xsdfec)
> > + return rval;
>
> It is impossible for container_of() to return NULL, unless something
> very magical and rare just happened. It's just doing pointer math, you
> can never check the return value of it.
Accepted.
Will remove if statement.
>
> > +
> > + if (_IOC_TYPE(cmd) != XSDFEC_MAGIC)
> > + return -ENOTTY;
>
> How can this happen?
Accepted.
Will be removed.
>
> > +
> > + /* check if ioctl argument is present and valid */
> > + if (_IOC_DIR(cmd) != _IOC_NONE) {
> > + arg = (void __user *)data;
> > + if (!arg)
> > + return rval;
> > + }
> > +
> > + switch (cmd) {
> > + default:
> > + /* Should not get here */
> > + dev_warn(xsdfec->dev, "Undefined SDFEC IOCTL");
>
> Nice that userspace has a way to fill up the kernel log :(
>
> Just return the correct error here, don't log it.
Accepted.
Will remove log.
>
> thanks,
>
> greg k-h
