[PATCH] net: usb: asix: Fix crash on skb alloc failure

[PATCH] net: usb: asix: Fix crash on skb alloc failure

[David B. Robins]

If asix_rx_fixup_internal() fails to allocate rx->ax_skb, it will return
but not clear rx->size. rx points to driver private data. A later call
assumes that nonzero size means ax_skb was allocated and passes a null
ax_skb to skb_put. Changed allocation failure return to clear size first.

Found testing board with AX88772B devices.

Signed-off-by: David B. Robins <linux@davidrobins.net>
---
 drivers/net/usb/asix_common.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/usb/asix_common.c b/drivers/net/usb/asix_common.c
index 75d6f26..079069a 100644
--- a/drivers/net/usb/asix_common.c
+++ b/drivers/net/usb/asix_common.c
@@ -91,8 +91,10 @@ int asix_rx_fixup_internal(struct usbnet *dev, struct sk_buff *skb,
 			}
 			rx->ax_skb = netdev_alloc_skb_ip_align(dev->net,
 							       rx->size);
-			if (!rx->ax_skb)
+			if (!rx->ax_skb) {
+				rx->size = 0;
 				return 0;
+			}
 		}
 
 		if (rx->size > dev->net->mtu + ETH_HLEN + VLAN_HLEN) {
-- 
1.9.1

Re: [PATCH] net: usb: asix: Fix crash on skb alloc failure

[David Miller]

From: "David B. Robins" <linux@davidrobins.net>
Date: Wed, 30 Sep 2015 16:20:04 -0400

> If asix_rx_fixup_internal() fails to allocate rx->ax_skb, it will return
> but not clear rx->size. rx points to driver private data. A later call
> assumes that nonzero size means ax_skb was allocated and passes a null
> ax_skb to skb_put. Changed allocation failure return to clear size first.
> 
> Found testing board with AX88772B devices.
> 
> Signed-off-by: David B. Robins <linux@davidrobins.net>

Applied, thanks.

Re: [PATCH] net: usb: asix: Fix crash on skb alloc failure

[David B. Robins]

On 2015-10-05 06:31, David Miller wrote:
> From: "David B. Robins" <linux@davidrobins.net>
> Date: Wed, 30 Sep 2015 16:20:04 -0400
> 
>> If asix_rx_fixup_internal() fails to allocate rx->ax_skb, it will 
>> return
>> but not clear rx->size. rx points to driver private data. A later call
>> assumes that nonzero size means ax_skb was allocated and passes a null
>> ax_skb to skb_put. Changed allocation failure return to clear size 
>> first.
>> 
>> Found testing board with AX88772B devices.
>> 
>> Signed-off-by: David B. Robins <linux@davidrobins.net>
> 
> Applied, thanks.

While I am happy for the patch I submitted to be applied, and it is 
consistent with existing error handling and does fix the specific issue, 
I believe a later series of 5 patches by Dean Jenkins that more 
comprehensively address the issue (grouped under subject "Improve ASIX 
RX memory allocation error handling") should be preferred.