[Q] LL_RESERVED_SPACE(dev) vs HH_DATA_ALIGN(dev->hard_header_len)

[YOSHIFUJI Hideaki / 吉藤英明 <yoshfuji@xxxxxxxxxxxxxx>]

Hello.

There're many places like:
  (dev->hard_header_len + 15) & ~15
or
  dev->hard_header_len + 15.

I guess the second one trying to express maximum number of the 
first one and it seems that
  HH_DATA_ALIGN(dev->hard_header_len)
  LL_SPACE_RESERVED(dev)
(defined in include/linux/netdevice.h) will do for us respectively.

Question: 
  How do you think about the following patch?
  Or, is it okay to simply use LL_RESERVED_SPACE(dev) in both places?

Thanks in advance.

===== net/econet/af_econet.c 1.30 vs edited =====
--- 1.30/net/econet/af_econet.c Mon Jan 26 04:05:06 2004
+++ edited/net/econet/af_econet.c       Sun Feb  8 17:18:51 2004
@@ -318,12 +318,12 @@
 #ifdef CONFIG_ECONET_NATIVE
                dev_hold(dev);
                
-               skb = sock_alloc_send_skb(sk, len+dev->hard_header_len+15, 
+               skb = sock_alloc_send_skb(sk, len + LL_RESERVED_SPACE(dev),
                                          msg->msg_flags & MSG_DONTWAIT, &err);
                if (skb==NULL)
                        goto out_unlock;
                
-               skb_reserve(skb, (dev->hard_header_len+15)&~15);
+               skb_reserve(skb, HH_DATA_ALIGN(dev->hard_header_len));
                skb->nh.raw = skb->data;
                
                eb = (struct ec_cb *)&skb->cb;

-- 
Hideaki YOSHIFUJI @ USAGI Project <yoshfuji@xxxxxxxxxxxxxx>
GPG FP: 9022 65EB 1ECF 3AD1 0BDF  80D8 4807 F894 E062 0EEA