Re: [PATCH] hso.c hso_net_open bugfix

Denis Joseph Barrow wrote:

Hi Paulus & others,
Sorry I nearly completely agree with you the code can be left as is
However netdev_priv check is superflous from my understanding of the code,
admittedly if they changed the netdev_priv function the check might
become valid.

I agree with you that odev checks in hso_net_open and hso_net_tx_timeout
are pointless and should be removed.

Please add netdev@vger.kernel.org mailing list also.

Admittedly dev_open would crash if dev!=NULL so I agree with you it
must be checked further up the stack.


Paulius Zaleckas wrote:

quoted

Denis Joseph Barrow wrote:

quoted

netdev_priv will return a non NULL pointer even if the net_device
is NULL & cause a crash without this fix.

Wrong. If net_device will be NULL hso_net_open won't get called.

quoted

Signed-off-by: Denis Joseph Barrow
[off-list ref]
---
Index: linux-2.6.27-rc3.patch/drivers/net/usb/hso.c
===================================================================

--- linux-2.6.27-rc3.patch.orig/drivers/net/usb/hso.c    2008-08-20

12:12:28.000000000 +0200

+++ linux-2.6.27-rc3.patch/drivers/net/usb/hso.c    2008-08-20

12:26:35.000000000 +0200

@@ -611,14 +611,12 @@
 /* called when net interface is brought up by ifconfig */
 static int hso_net_open(struct net_device *net)
 {
-    struct hso_net *odev = netdev_priv(net);
+    struct hso_net *odev;
     unsigned long flags = 0;
 
-    if (!odev) {
-        dev_err(&net->dev, "No net device !\n");
+    if (!net)
         return -ENODEV;
-    }

--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html