[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1282048602.5765.752.camel@bigi>
Date: Tue, 17 Aug 2010 08:36:42 -0400
From: jamal <hadi@...erus.ca>
To: Grégoire Baron <baronchon@...m.org>
Cc: netdev@...r.kernel.org
Subject: Re: [PATCH] net/sched: add ACT_CSUM action to update packets
checksums
On Tue, 2010-08-17 at 01:02 +0200, Grégoire Baron wrote:
> You're rigth. I will correct that, maybe using a macro which is
> specialised to get the protocol structure ...
> However, this macro could call a 'goto' instruction. Is it really a good
> idea? Tell me. Are you sure a function is appropriate?
Do you even need the goto fail? It seems you could short-circuit and
return 0 (at least thats what the goto seems to do)
But i really dont want to complicate this;
In the minimal i think you can make code like this
====
struct icmphdr *icmph;
+ int hl;
+ int ntkoff;
+
+ ntkoff = skb_network_offset(skb);
+
+ hl = ihl + sizeof(*icmph);
+
+ if (!pskb_may_pull(skb, ipl + ntkoff) || (ipl < hl) ||
+ (skb_cloned(skb) &&
+ !skb_clone_writable(skb, hl + ntkoff) &&
+ pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
+ goto fail;
===
reusable - return 0 if fail else return 1 etc etc
BTW, another comment: icmp and igmp have unused parameter iph
passed to them - probably cut and paste.
cheers,
jamal
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists