Date: Tue, 24 Apr 2007 22:35:47 GMT From: Marko Zec <zec@FreeBSD.org> To: Perforce Change Reviews <perforce@FreeBSD.org> Subject: PERFORCE change 118758 for review Message-ID: <200704242235.l3OMZlkX033326@repoman.freebsd.org>
next in thread | raw e-mail | index | archive | help
http://perforce.freebsd.org/chv.cgi?CH=118758 Change 118758 by zec@zec_tca51 on 2007/04/24 22:35:11 A hack to allow for lo0 to be destroyed when cleaning up a vnet instance. Affected files ... .. //depot/projects/vimage/src/sys/net/if_loop.c#12 edit Differences ... ==== //depot/projects/vimage/src/sys/net/if_loop.c#12 (text+ko) ==== @@ -211,12 +211,24 @@ static int vnet_loif_idetach(void) { INIT_VNET_NET(curvnet); - struct lo_softc *lop, *nlop; + struct lo_softc *sc, *nsc; + + LIST_FOREACH_SAFE(sc, &V_lo_list, sc_next, nsc) { + struct ifnet *ifp = sc->sc_ifp; - LIST_FOREACH_SAFE(lop, &V_lo_list, sc_next, nlop) { - if (lop->sc_ifp == V_loif) + if (ifp == V_loif) { + /* + * A hack to allow lo0 to be detached: + * bump if_unit number from 0 to 1. By + * setting V_loif to NULL we prevent queuing + * of routing messages that would have + * m_pkthdr.rcvif pointing to a nonexisting + * ifnet, i.e. the lo0 we just destroyed. + */ + ifp->if_dunit = 1; V_loif = NULL; - lo_clone_destroy(lop->sc_ifp); + } + if_clone_destroy(ifp->if_xname); } return 0; }
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200704242235.l3OMZlkX033326>