Re: [Fwd: extra spinlocking in forcedeth 0.25]

Andrew,

could you have a look at the attached mail regarding spinlocking in your WOL patch for forcedeth?

In 0.23 -> 0.25 forcedeth diff I noticed this:

+       case ETHTOOL_GWOL:
+       {
+               struct ethtool_wolinfo wolinfo;
+               memset(&wolinfo, 0, sizeof(wolinfo));
+               wolinfo.supported = WAKE_MAGIC;

(quoted diff) +10 -0

+
+               spin_lock_irq(&np->lock);
+               if (np->wolenabled)
+                       wolinfo.wolopts = WAKE_MAGIC;
+               spin_unlock_irq(&np->lock);
+
+               if (copy_to_user(useraddr, &wolinfo, sizeof(wolinfo)))
+                       return -EFAULT;
+               return 0;
+       }

IMHO this pair of lock()/unlock() is not needed
and can be safely removed.