RE: [PATCH v5 4/8] char: rpmb: provide a user space interface

[Winkler, Tomas]

\
> Subject: Re: [PATCH v5 4/8] char: rpmb: provide a user space interface
> 
> On Sun 2016-09-04 11:35:33, Winkler, Tomas wrote:
> >
> > > On Thu, Sep 01, 2016 at 08:05:26PM +0000, Winkler, Tomas wrote:
> > > >
> > > > >
> > > > > On Sun, Aug 07, 2016 at 09:44:03AM +0000, Winkler, Tomas wrote:
> > > > > > >
> > > > > > > On Mon 2016-07-18 23:27:49, Tomas Winkler wrote:
> > > > > > > > The user space API is achieved via two synchronous IOCTL.
> > > > > > >
> > > > > > > IOCTLs?
> > > > > >
> > > > > > Will fix
> > > > > >
> > > > > > > > Simplified one, RPMB_IOC_REQ_CMD, were read result cycles
> > > > > > > > is
> > > > > > > performed
> > > > > > > > by the framework on behalf the user and second,
> > > > > > > > RPMB_IOC_SEQ_CMD
> > > > > > > where
> > > > > > > > the whole RPMB sequence including RESULT_READ is supplied
> > > > > > > > by the
> > > > > caller.
> > > > > > > > The latter is intended for  easier adjusting  of the
> > > > > > > > applications that use MMC_IOC_MULTI_CMD ioctl.
> > > > > > >
> > > > > > > Why "  "?
> > > > > > Not sure I there is enough clue in your question.
> > > > > > >
> > > > > > > >
> > > > > > > > Signed-off-by: Tomas Winkler <tomas.winkler@xxxxxxxxx>
> > > > > > >
> > > > > > > > +
> > > > > > > > +static long rpmb_ioctl(struct file *fp, unsigned int cmd,
> > > > > > > > +unsigned long arg) {
> > > > > > > > +	return __rpmb_ioctl(fp, cmd, (void __user *)arg); }
> > > > > > > > +
> > > > > > > > +#ifdef CONFIG_COMPAT
> > > > > > > > +static long rpmb_compat_ioctl(struct file *fp, unsigned int cmd,
> > > > > > > > +			      unsigned long arg) {
> > > > > > > > +	return	__rpmb_ioctl(fp, cmd, compat_ptr(arg));
> > > > > > > > +}
> > > > > > > > +#endif /* CONFIG_COMPAT */
> > > > > > >
> > > > > > > Description of the ioctl is missing,
> > > > > > Will add.
> > > > > >
> > > > > > and it should certainly be designed in a way
> > > > > > > that it does not need compat support.
> > > > > >
> > > > > > The compat_ioctl handler just casts the compat_ptr, I believe
> > > > > > this should be done unless the ioctl is globaly registered in
> > > > > > fs/compat_ioctl.c, but I might be wrong.
> > > > >
> > > > > You shouldn't need a compat ioctl for anything new that is
> > > > > added, unless your api is really messed up.  Please test to be
> > > > > sure, and not use a compat ioctl at all, it isn't that hard to do.
> > > >
> > > > compat_ioctl is called anyhow when CONFIG_COMPAT is set, there is
> > > > no way around it, or I'm missing something?  Actually there is no
> > > > more than that for the COMPAT support in this code.
> > >
> > > If you don't provide a compat_ioctl() all should be fine, right?
> >
> > No,  this doesn't work the driver has to provide compat_ioctl
> >
> > You would expect something like
> > if (!f.file->f_op->compat_ioctl) {
> >                         error =
> > f_op->f.file->f_op->unlocked_ioctl((f.file, cmd, compat_ptr(arg)) }
> > But there is no such code under  fs/compat_ioctl.c
> >
> > The translation has to implemented by the device driver or registered
> > directly in fs/compat_ioct.c in do_ioctl_trans or ioctl_pointer[]
> >
> > If compat_ioct is not provided the application is receiving
> > : ioctl failure -1: Inappropriate ioctl for device
> 
> Care to submit a patch? We should not really have to include compat_ioctl
> support if it is already compatible...
> 
> Or maybe better provide empty function drivers can fill in when
> compatible...?
> 
I'm not sure it is so simple to generallize, because ioctl selection logic in compat_ioctl has already a default actions and second because 'arg' my not be a pointer. 
Maybe a beter option would be to provide a function in spirit generic_ fs functions let say compat_generic_ioct() that would just cal unlocked_ioctl(fp, cmd, compat_ptr(arg))

Thanks
Tomas