suggested patches for netinet6/

From: Luigi Rizzo (rizzo_at_icir.org)
Date: 04/09/04

  • Next message: Ruslan Ermilov: "Re: Testers wanted: polling(4) support for vr(4)" 
    Date: Fri, 9 Apr 2004 04:27:20 -0700
To: ume@freebsd.org, net@freebsd.org, ru@freebsd.org, andre@freebsd.org

    While adapting to ipv6 the new arp table code I am developing
    following Andre's ideas, i hit a few places that would deserve a
    fix independently of that:

    + nd6_nud_hint() is only called as nd6_nud_hint(NULL, NULL, 0);
      in some cases from netinet/tcp_input.c
      With these args, the routine is a NOP. I propose to remove it
      (and the associated field, ln_byhint, in struct llinfo_nd6)

    + In a couple of places, the logic to compute 'olladdr' and 'llchange'
      is rather convoluted, and it could be greatly simplified (see below)

    + nd6_output() contains a tail-recursive call which certainly can be
      removed by using a 'goto' near the beginning of the function
      (or maybe in a better way!)

    Also:
    + in nd6_na_input() we are responding to an nd6 message on interface 'ifp',
      so i guess in the routine rt->rt_ifp == ifp, and we can use the latter
      when needed instead of rt->rt_ifp ?
     
    + is it ok to remove the __P() from the header files, ANSIfy
      the function declarations and make them static as appropriate ?
      Of course this ought to be done as a separate step.

    I am attaching the relevant snippets of the diff... Can anyone comment ?
    Any objection if i commit the trivial fixes to the first 3 issues above ?
    What about the style changes ?

            cheers
            luigi

    Index: nd6.c
    ===================================================================
    - olladdr = (sdl->sdl_alen) ? 1 : 0;
    - if (olladdr && lladdr) {
    - if (bcmp(lladdr, LLADDR(sdl), ifp->if_addrlen))
    - llchange = 1;
    - else
    - llchange = 0;
    - } else
    - llchange = 0;
    + olladdr = ln->ln_state >= ND6_LLINFO_REACHABLE;
    + llchange = olladdr && lladdr && bcmp(lladdr, &ln->ll_addr, ifp->if_addrlen);
     
    @@ -1844,11 +1836,8 @@ nd6_output(ifp, origifp, m0, dst, rt0)
    +again:
            /*
             * next hop determination. This routine is derived from ether_outpout.
             */
            if (rt) {
                    if ((rt->rt_flags & RTF_UP) == 0) {
                            rt0 = rt = rtalloc1((struct sockaddr *)dst, 1, 0UL);
                             if (rt != NULL) {
                                     RT_REMREF(rt);
                                     RT_UNLOCK(rt);
    - if (rt->rt_ifp != ifp) {
    - /* XXX: loop care? */
    - return nd6_output(ifp, origifp, m0,
    - dst, rt);
    - }
    + if (rt->rt_ifp != ifp)
    + goto again; /* rt has changed. */
                             } else
                                     senderr(EHOSTUNREACH);
                     }
     
    Index: nd6_nbr.c
    ===================================================================
    @@ -653,17 +650,9 @@ nd6_na_input(m, off, icmp6len)
                     /*
                      * Check if the link-layer address has changed or not.
                      */
    - if (!lladdr)
    - llchange = 0;
    - else {
    - if (sdl->sdl_alen) {
    - if (bcmp(lladdr, LLADDR(sdl), ifp->if_addrlen))
    - llchange = 1;
    - else
    - llchange = 0;
    - } else
    - llchange = 1;
    - }
    + llchange = lladdr &&
    + ( (ln->ln_state < ND6_LLINFO_REACHABLE) ||
    + bcmp(lladdr, &ln->l3_addr, ifp->if_addrlen) );
     

    @@ -744,7 +728,7 @@ nd6_na_input(m, off, icmp6len)
                              * context. However, we keep it just for safety.
                              */
                             s = splnet();
    - dr = defrouter_lookup(in6, rt->rt_ifp);
    + dr = defrouter_lookup(&ln->l3_addr, ifp);
                             if (dr)
                                     defrtrlist_del(dr);
                             else if (!ip6_forwarding && ip6_accept_rtadv) {
    @@ -755,21 +739,25 @@ nd6_na_input(m, off, icmp6len)
                                      * (e.g. redirect case). So we must
                                      * call rt6_flush explicitly.
                                      */
    - rt6_flush(&ip6->ip6_src, rt->rt_ifp);
    + rt6_flush(&ip6->ip6_src, ifp);
                             }
                             splx(s);
                     }
                     ln->ln_router = is_router;
             }

    _______________________________________________
    freebsd-net@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-net
    To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"

  • Next message: Ruslan Ermilov: "Re: Testers wanted: polling(4) support for vr(4)"