Re: [PATCH] fix potential NULL pointer dereference in yam

[Jesper Juhl]

On Sunday 14 May 2006 16:09, Alexey Dobriyan wrote:
> On Sun, May 14, 2006 at 03:12:50PM +0200, Jesper Juhl wrote:
> > Testing a pointer for NULL after it has already been dereferenced is not
> > very safe.
> > Patch below to rework yam_open() so that `dev' is not dereferenced until
> > after it has been tested for NULL.
> 
> > Found by coverity checker as #787
> 
> > --- linux-2.6.17-rc4-git2-orig/drivers/net/hamradio/yam.c
> > +++ linux-2.6.17-rc4-git2/drivers/net/hamradio/yam.c
> > @@ -845,15 +845,25 @@ static struct net_device_stats *yam_get_
> >  
> >  static int yam_open(struct net_device *dev)
> >  {
> > -	struct yam_port *yp = netdev_priv(dev);
> > +	struct yam_port *yp;
> >  	enum uart u;
> >  	int i;
> > -	int ret=0;
> > +	int ret = 0;
> 
> Please, don't make unrelated changes.
> 
Sorry.


> > -	printk(KERN_INFO "Trying %s at iobase 0x%lx irq %u\n", dev->name, dev->base_addr, dev->irq);
> > +	if (!dev) {
> > +		printk(KERN_ERR "yam_open() called without device\n");
> > +		return -ENXIO;
> > +	}
> 
> How can it be NULL here? The whole array of valid net_devices was
> allocated at module init time.
> 

It cannot. You are right, I'm wrong.
I guess removing the check makes sense then ?


Signed-off-by: Jesper Juhl <jesper.juhl@xxxxxxxxx>
---

--- linux-2.6.17-rc4-mm1-orig/drivers/net/hamradio/yam.c	2006-05-13 21:28:27.000000000 +0200
+++ linux-2.6.17-rc4-mm1/drivers/net/hamradio/yam.c	2006-05-15 22:16:32.000000000 +0200
@@ -852,7 +852,7 @@ static int yam_open(struct net_device *d
 
 	printk(KERN_INFO "Trying %s at iobase 0x%lx irq %u\n", dev->name, dev->base_addr, dev->irq);
 
-	if (!dev || !yp->bitrate)
+	if (!yp->bitrate)
 		return -ENXIO;
 	if (!dev->base_addr || dev->base_addr > 0x1000 - YAM_EXTENT ||
 		dev->irq < 2 || dev->irq > 15) {


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/