Re: [PATCH] w1: Misuse of get_user()/put_user() reported by sparse

[Greg Kroah-Hartman]

On Fri, Nov 26, 2021 at 05:10:46PM +0100, Christophe Leroy wrote:
> 
> 
> Le 26/11/2021 à 17:00, Greg Kroah-Hartman a écrit :
> > On Fri, Nov 19, 2021 at 10:15:09AM +0100, Christophe Leroy wrote:
> > > sparse warnings: (new ones prefixed by >>)
> > > > > drivers/w1/slaves/w1_ds28e04.c:342:13: sparse: sparse: incorrect type in initializer (different address spaces) @@     expected char [noderef] __user *_pu_addr @@     got char *buf @@
> > >     drivers/w1/slaves/w1_ds28e04.c:342:13: sparse:     expected char [noderef] __user *_pu_addr
> > >     drivers/w1/slaves/w1_ds28e04.c:342:13: sparse:     got char *buf
> > > > > drivers/w1/slaves/w1_ds28e04.c:356:13: sparse: sparse: incorrect type in initializer (different address spaces) @@     expected char const [noderef] __user *_gu_addr @@     got char const *buf @@
> > >     drivers/w1/slaves/w1_ds28e04.c:356:13: sparse:     expected char const [noderef] __user *_gu_addr
> > >     drivers/w1/slaves/w1_ds28e04.c:356:13: sparse:     got char const *buf
> > > 
> > > The buffer buf is a failsafe buffer in kernel space, it's not user
> > > memory hence doesn't deserve the use of get_user() or put_user().
> > > 
> > > Access 'buf' content directly.
> > > 
> > > Reported-by: kernel test robot <lkp@xxxxxxxxx>
> > > Link: https://lore.kernel.org/lkml/202111190526.K5vb7NWC-lkp@xxxxxxxxx/T/
> > > Signed-off-by: Christophe Leroy <christophe.leroy@xxxxxxxxxx>
> > > ---
> > >   drivers/w1/slaves/w1_ds28e04.c | 10 ++--------
> > >   1 file changed, 2 insertions(+), 8 deletions(-)
> > > 
> > > diff --git a/drivers/w1/slaves/w1_ds28e04.c b/drivers/w1/slaves/w1_ds28e04.c
> > > index e4f336111edc..d75bb16fb7a1 100644
> > > --- a/drivers/w1/slaves/w1_ds28e04.c
> > > +++ b/drivers/w1/slaves/w1_ds28e04.c
> > > @@ -339,10 +339,7 @@ static BIN_ATTR_RW(pio, 1);
> > >   static ssize_t crccheck_show(struct device *dev, struct device_attribute *attr,
> > >   			     char *buf)
> > >   {
> > > -	if (put_user(w1_enable_crccheck + 0x30, buf))
> > > -		return -EFAULT;
> > > -
> > > -	return sizeof(w1_enable_crccheck);
> > > +	return sprintf(buf, "%d", w1_enable_crccheck);
> > 
> > This should be sysfs_emit(), right?
> 
> Ok
> 
> > 
> > >   }
> > >   static ssize_t crccheck_store(struct device *dev, struct device_attribute *attr,
> > > @@ -353,11 +350,8 @@ static ssize_t crccheck_store(struct device *dev, struct device_attribute *attr,
> > >   	if (count != 1 || !buf)
> > >   		return -EINVAL;
> > > -	if (get_user(val, buf))
> > > -		return -EFAULT;
> > > -
> > >   	/* convert to decimal */
> > > -	val = val - 0x30;
> > > +	val = *buf - 0x30;
> > 
> > Why not use a proper function that can parse a string and turn it into a
> > number?
> 
> I wanted to keep the change minimal. But I can also replace it with some
> scanf.
> 
> But don't we have any generic function to read and store a bool after all ?

Yes we do, please use kstrtobool().

thanks,

greg k-h