[PATCH] USB GeneLink/NDIS tweaks

USB GeneLink/NDIS tweaks

This patch, from Oliver, discards slightly less data in cases
where memory is very tight.  The GeneLink framing can batch
several packets in one transmission, so it's always possible to
at least report the last one, since it's already sitting in an SKB
and we won't need a new one.

Random note, this seems to be the same issue I saw
mentioned somewhere else recently:  MSFT host drivers
with high latency for submitting bulk packets, so rather
than just queuing them as they arrive, it turns out to be a
win on the MSFT side to batch packets.

drivers/usb/net/usbnet.c

@@ -538,25 +538,25 @@ static int genelink_rx_fixup (struct usbnet *dev, struct sk_buff *skb)
 
 		// allocate the skb for the individual packet
 		gl_skb = alloc_skb (size, GFP_ATOMIC);
-		if (gl_skb == 0)
-			return 0;
+		if (gl_skb) {
 
-		// copy the packet data to the new skb
-		memcpy (gl_skb->data, packet->packet_data, size);
+			// copy the packet data to the new skb
+			memcpy (gl_skb->data, packet->packet_data, size);
 
-		// set skb data size
-		gl_skb->len = size;
-		gl_skb->dev = &dev->net;
+			// set skb data size
+			gl_skb->len = size;
+			gl_skb->dev = &dev->net;
 
-		// determine the packet's protocol ID
-		gl_skb->protocol = eth_type_trans (gl_skb, &dev->net);
+			// determine the packet's protocol ID
+			gl_skb->protocol = eth_type_trans (gl_skb, &dev->net);
 
-		// update the status
-		dev->stats.rx_packets++;
-		dev->stats.rx_bytes += size;
+			// update the status
+			dev->stats.rx_packets++;
+			dev->stats.rx_bytes += size;
 
-		// notify os of the received packet
-		status = netif_rx (gl_skb);
+			// notify os of the received packet
+			status = netif_rx (gl_skb);
+		}
 
 		// advance to the next packet
 		packet = (struct gl_packet *)