/* * Copyright (c) 1986 Regents of the University of California. * All rights reserved. The Berkeley software License Agreement * specifies the terms and conditions for redistribution. * * @(#)if_ix.c 7.1 (Berkeley) 6/5/86 */ #include "np.h" #if NNP > 0 /* * Interlan NP100 Ethernet Communications Controller interface */ #include "../machine/pte.h" #include "param.h" #include "systm.h" #include "mbuf.h" #include "buf.h" #include "protosw.h" #include "socket.h" #include "vmmac.h" #include "ioctl.h" #include "errno.h" #include "../net/if.h" #include "../net/netisr.h" #include "../net/route.h" #ifdef INET #include "../netinet/in.h" #include "../netinet/in_systm.h" #include "../netinet/in_var.h" #include "../netinet/ip.h" #include "../netinet/if_ether.h" #endif #ifdef NS #include "../netns/ns.h" #include "../netns/ns_if.h" #endif #include "../vax/cpu.h" #include "../vax/mtpr.h" #include "../vaxif/if_uba.h" #include "../vaxuba/ubareg.h" #include "../vaxuba/ubavar.h" #include "../vaxuba/npreg.h" #include "../vaxif/if_ix.h" int ixattach(), ixrint(), ixcint(); #define ILUNIT(x) minor(x) int ixinit(), ixoutput(), ixioctl(), ixreset(), ixwatch(); int (*IxAttach)() = ixattach; int (*IxReset)() = ixreset; /* * Ethernet software status per interface. * * Each interface is referenced by a network interface structure, * ix_if, which the routing code uses to locate the interface. * This structure contains the output queue for the interface, its address, ... * We also have, for each interface, a UBA interface structure, which * contains information about the UNIBUS resources held by the interface: * map registers, buffered data paths, etc. Information is cached in this * structure for use by the if_uba.c routines in running the interface * efficiently. */ struct ix_softc { struct arpcom ix_ac; /* Ethernet common part */ #define ix_if ix_ac.ac_if /* network-visible interface */ #define ix_addr ix_ac.ac_enaddr /* hardware Ethernet address */ int ix_flags; #define IXF_OACTIVE 0x1 /* output is active */ #define IXF_RCVPENDING 0x2 /* start rcv in ilcint */ #define IXF_GOTUBA 0x4 /* unibus resources mapped */ #define IXF_RUNNING 0x8 /* board is running */ #define IXF_SETADDR 0x10 /* physical address is changed */ #define IXF_STATPENDING 0x20 /* stat cmd pending */ #define IXF_GOTCQE 0x40 /* np resources available */ struct ifuba ix_ifuba; /* unibus resources */ u_short ix_aid; /* Access Id returned by open DDL */ u_short ix_badcqe; struct npmaster *ix_mp; /* Board physio request header */ struct npreq *ix_rrp; /* Cached npreq for recv */ struct npreq *ix_wrp; /* Cached npreq for xmit */ short ix_scaninterval; /* interval of stat collection */ #define IXWATCHINTERVAL 60 /* once every 60 seconds */ union ix_stats ix_stats; /* holds on-board statistics */ int ix_ubaddr; /* mapping registers of ix_stats */ } ix_softc[NNP]; extern struct uba_device *npdinfo[]; /* * Interface exists: make available by filling in network interface * record. System will initialize the interface when it is ready * to accept packets. We can't even get the ethernet address * or other interesting data until the board has been downloaded. * running ifconfig will attempt to start unit. */ ixattach(ui) struct uba_device *ui; { register struct ix_softc *ix = &ix_softc[ui->ui_unit]; register struct ifnet *ifp = &ix->ix_if; extern struct npmaster npmasters[]; ifp->if_unit = ui->ui_unit; ifp->if_name = "ix"; ifp->if_mtu = ETHERMTU; ifp->if_flags = IFF_BROADCAST; ifp->if_init = ixinit; ifp->if_output = ixoutput; ifp->if_ioctl = ixioctl; ifp->if_reset = ixreset; ix->ix_mp = npmasters + ui->ui_unit; ix->ix_ifuba.ifu_flags = UBA_CANTWAIT; if_attach(ifp); } struct npreq * ix_GetReq(mp, addr, len) struct npmaster *mp; caddr_t addr; { int unit = mp->unit; register struct npreq *rp; register struct CQE *ep; struct ix_softc *ix = ix_softc + unit; extern struct npreq *NpGetReq(); while ((rp = NpGetReq(mp->reqtab)) == NULL) { mp->reqtab->flags |= WANTREQ; sleep((caddr_t)(mp->reqtab), PZERO - 1); } rp->flags = KERNREQ; /* Clear flags */ ep = rp->element; /* Associated CQE */ ep->cqe_famid = (unsign32)ix; /* Process ID */ ep->cqe_wind = 0; /* Amount of buffer mapped */ ep->cqe_nbuf = 1; /* Must be 1, no buffer chain */ ep->cqe_char = 1; /* Driver owns this CQE */ ep->cqe_prot = NPDLA; /* Data Link Access protocol */ ep->cqe_bcnt = len; /* Byte count */ rp->bufaddr = (caddr_t) (UBADDRMASK & (int) addr);/* mapped buffer */ ep->cqe_dma[0] = (unsign16)LOWORD(rp->bufaddr); ep->cqe_dma[1] = (unsign16)HIWORD(rp->bufaddr); return (rp); } ix_DoReq(mp, rp, cmd, addr, len, rpb, routine) struct npmaster *mp; register struct npreq *rp; u_short cmd; caddr_t addr; int len; register u_short *rpb; int (*routine)(); { register struct CQE *ep = rp->element; register u_short *p = &ep->rpb1; u_short cnt = *rpb++; extern long NpDebug; int pri; ep->cqe_ust0 = ep->cqe_ust1 = NPCLEAR; /* Clear status */ ep->cqe_bcnt = len; /* Byte count */ rp->flags = KERNREQ; /* Clear flags */ rp->bufaddr = (caddr_t) (UBADDRMASK & (int) addr);/* mapped buffer */ rp->intr = routine; rp->user = (caddr_t) (ep->cqe_func = cmd);/* In case pissed on in CQE */ ep->cqe_dma[0] = (unsign16)LOWORD(rp->bufaddr); ep->cqe_dma[1] = (unsign16)HIWORD(rp->bufaddr); ep->cqe_lenrpb = cnt + cnt; for (; cnt > 0; cnt--) *p++ = *rpb++; if (NpDebug & DEBCQE) printf("Function is %x ep %x reqid %x\n", ep->cqe_func, ep, ep->cqe_reqid); if (NpDebug & DEBCQE) printf("irp len = %x rp = %x\n", ep->cqe_lenrpb, rp); if (routine == 0) { NpAddReq(mp->reqtab, rp); /* Queue onto active list */ while (!(rp->flags & REQDONE)) { pri = spl4(); NpAddCQE(ep, &mp->shmemp->devcq, mp); sleep((caddr_t)rp, PZERO - 1); splx(pri); } if (rp->flags & IOABORT || ep->cqe_sts != NPDONE || ep->cqe_ust0 != NPDONE || ep->cqe_ust1 != NPOK) { struct ix_softc *ix = (struct ix_softc *)ep->cqe_famid; printf("ix%d: Req failed, cmd %x, stat %x, ", ix->ix_if.if_unit, rp->user, ep->cqe_sts); printf("ust error %x,%x\n", ep->cqe_ust0, ep->cqe_ust1); } NpRemReq(rp); /* Clear request */ } else { pri = spl4(); NpAddCQE(ep, &mp->shmemp->devcq, mp); splx(pri); } } /* * Ethernet output routine. * Encapsulate a packet of type family for the local net. * Use trailer local net encapsulation if enough data in first * packet leaves a multiple of 512 bytes of data in remainder. */ ixoutput(ifp, m0, dst) struct ifnet *ifp; struct mbuf *m0; struct sockaddr *dst; { int type, s, error; u_char edst[6]; struct in_addr idst; register struct ix_softc *ix = &ix_softc[ifp->if_unit]; register struct mbuf *m = m0; register struct ether_header *il; register int off; int usetrailers; if ((ifp->if_flags & (IFF_UP|IFF_RUNNING)) != (IFF_UP|IFF_RUNNING)) { error = ENETDOWN; goto bad; } switch (dst->sa_family) { #ifdef INET case AF_INET: idst = ((struct sockaddr_in *)dst)->sin_addr; if (!arpresolve(&ix->ix_ac, m, &idst, edst, &usetrailers)) return (0); /* if not yet resolved */ off = ntohs((u_short)mtod(m, struct ip *)->ip_len) - m->m_len; if (usetrailers && off > 0 && (off & 0x1ff) == 0 && m->m_off >= MMINOFF + 2 * sizeof (u_short)) { type = ETHERTYPE_TRAIL + (off>>9); m->m_off -= 2 * sizeof (u_short); m->m_len += 2 * sizeof (u_short); *mtod(m, u_short *) = htons((u_short)ETHERTYPE_IP); *(mtod(m, u_short *) + 1) = htons((u_short)m->m_len); goto gottrailertype; } type = ETHERTYPE_IP; off = 0; goto gottype; #endif #ifdef NS case AF_NS: type = ETHERTYPE_NS; bcopy((caddr_t)&(((struct sockaddr_ns *)dst)->sns_addr.x_host), (caddr_t)edst, sizeof (edst)); off = 0; goto gottype; #endif case AF_UNSPEC: il = (struct ether_header *)dst->sa_data; bcopy((caddr_t)il->ether_dhost, (caddr_t)edst, sizeof (edst)); type = il->ether_type; goto gottype; default: printf("ix%d: can't handle af%d\n", ifp->if_unit, dst->sa_family); error = EAFNOSUPPORT; goto bad; } gottrailertype: /* * Packet to be sent as trailer: move first packet * (control information) to end of chain. */ while (m->m_next) m = m->m_next; m->m_next = m0; m = m0->m_next; m0->m_next = 0; m0 = m; gottype: /* * Add local net header. If no space in first mbuf, * allocate another. */ if (m->m_off > MMAXOFF || MMINOFF + sizeof (struct ether_header) > m->m_off) { m = m_get(M_DONTWAIT, MT_HEADER); if (m == 0) { error = ENOBUFS; goto bad; } m->m_next = m0; m->m_off = MMINOFF; m->m_len = sizeof (struct ether_header); } else { m->m_off -= sizeof (struct ether_header); m->m_len += sizeof (struct ether_header); } il = mtod(m, struct ether_header *); il->ether_type = htons((u_short)type); bcopy((caddr_t)edst, (caddr_t)il->ether_dhost, sizeof (edst)); bcopy((caddr_t)ix->ix_addr, (caddr_t)il->ether_shost, sizeof(il->ether_shost)); /* * Queue message on interface, and start output if interface * not yet active. */ s = splimp(); if (IF_QFULL(&ifp->if_snd)) { IF_DROP(&ifp->if_snd); splx(s); m_freem(m); return (ENOBUFS); } IF_ENQUEUE(&ifp->if_snd, m); if ((ix->ix_flags & IXF_OACTIVE) == 0) ixstart(ifp->if_unit); splx(s); return (0); bad: m_freem(m0); return (error); } /* * Reset of interface after UNIBUS reset. * If interface is on specified uba, reset its state. */ ixreset(unit, uban, softp) int unit, uban; caddr_t softp; { register struct uba_device *ui; int mask = IXF_SETADDR; /* Only remember new physaddr */ if (unit >= NNP || (ui = npdinfo[unit]) == 0 || ui->ui_alive == 0 || ui->ui_ubanum != uban) return; printf(" ix%d reset", unit); if (softp) mask |= IXF_GOTUBA; /* UBA mapping regs still valid; */ ix_softc[unit].ix_if.if_flags &= ~IFF_RUNNING; ix_softc[unit].ix_flags &= mask; } /* * Initialization of interface; clear recorded pending * operations, and reinitialize UNIBUS usage. */ ixinit(unit) int unit; { register struct ix_softc *ix = &ix_softc[unit]; struct uba_device *ui = npdinfo[unit]; register struct ifnet *ifp = &ix->ix_if; register struct CQE *ep; struct npreq *rp; struct npmaster *mp = ix->ix_mp; register u_short *dpmp = & mp->shmemp->statblock.sb_dpm; u_short rpb[7]; int s; /* not yet, if address still unknown */ if (ifp->if_addrlist == (struct ifaddr *)0) return; if (ix->ix_flags & IXF_RUNNING) return; if ((mp->flags & AVAILABLE) == 0 || (*dpmp & PROTOMASK(NPDLA)) == 0) { ifp->if_flags &= ~IFF_UP; return; } if ((ix->ix_flags & IXF_GOTUBA) == 0) { ix->ix_ifuba.ifu_flags = UBA_CANTWAIT; if (if_ubainit(&ix->ix_ifuba, ui->ui_ubanum, sizeof (struct ether_header), (int)btoc(ETHERMTU)) == 0) { printf("ix%d: can't initialize\n", unit); ix->ix_if.if_flags &= ~IFF_UP; return; } ix->ix_ubaddr = uballoc(ui->ui_ubanum, (caddr_t)&ix->ix_stats, sizeof (union ix_stats), 0); ix->ix_flags |= IXF_GOTUBA; } if ((ix->ix_flags & IXF_GOTCQE) == 0) { ix->ix_rrp = ix_GetReq(mp, ix->ix_ifuba.ifu_r.ifrw_info, ETHERMTU); ix->ix_wrp = ix_GetReq(mp, 0, 0); ix->ix_flags |= IXF_GOTCQE; } rp = ix->ix_wrp; ep = rp->element; /* Changing the ethernet address resets the dla module, so must do it before opening the channel */ if (ix->ix_flags & IXF_SETADDR) { register char *cp = (char *) &ix->ix_stats; int spincount; int x; *cp++ = 1; bcopy(ix->ix_addr, (caddr_t)cp, 6); rpb[0] = 1; /* RPB length */ ix_DoReq(mp, rp, IXC_LDPA, ix->ix_ubaddr, 7, rpb, 0); #ifndef TheyFinallyFixedTheBoard /* Board requires some time to reinitialize its protocols */ x = spl1(); spincount = 2000000; while (((*dpmp & PROTOMASK(NPDLA))==0) && spincount > 0) spincount--; if (spincount==0) { printf("ix%d: failed to reinitialize DLA module\n", unit); splx(x); } splx(x); #endif } rpb[0] = 6; /* RPB length */ rpb[2] = 0x10; /* Share with any smart users */ rpb[3] = 0; /* Take (a copy of) all frames */ rpb[5] = 8; /* On board rcv queue length */ rpb[6] = 0; /* XMT packets as is */ ix_DoReq(mp, rp, IXC_OPEN, 0, 0, rpb, 0); ix->ix_aid = ep->rpb1; /* Here we request our ethernet address, if we didn't reset it*/ if ((ix->ix_flags & IXF_SETADDR)==0) { rpb[0] = 2; rpb[1] = ix->ix_aid; rpb[2] = 0; /* get all stats */ ix_DoReq(mp, rp, IXC_GSTAT, /* Get Stats */ (caddr_t) ix->ix_ubaddr, sizeof(ix->ix_stats) - 8, rpb, 0); bcopy((caddr_t) &ix->ix_stats, (caddr_t) ix->ix_addr, 6); } ix->ix_if.if_flags |= IFF_RUNNING; ix->ix_flags |= IXF_RUNNING; ifp->if_watchdog = ixwatch; ifp->if_timer = ix->ix_scaninterval = IXWATCHINTERVAL; ixrint(mp, 0); } /* * Start output on interface. * Get another datagram to send off of the interface queue, * and map it to the interface before starting the output. */ ixstart(dev) dev_t dev; { int len = 0; int unit = minor(dev); register struct ix_softc *ix = &ix_softc[unit]; register struct mbuf *n; struct mbuf *m; int s, error = 0; struct npmaster *mp = ix->ix_mp; struct npreq *rp = ix->ix_wrp; struct CQE *ep; u_short rpb[8]; IF_DEQUEUE(&ix->ix_if.if_snd, m); if (m == 0) { if (ix->ix_flags & IXF_STATPENDING) { ix->ix_flags |= IXF_OACTIVE; rpb[0] = 2; rpb[1] = ix->ix_aid; rpb[2] = 0; /* get all stats */ ix_DoReq(mp, rp, IXC_GSTAT, /* general Stats */ (caddr_t) ix->ix_ubaddr, sizeof(ix->ix_stats) - 8, rpb, ixcint); } return; } /* * Ensure minimum packet length. * This makes the safe assumtion that there are no virtual holes * after the data. * For security, it might be wise to zero out the added bytes, * but we're mainly interested in speed at the moment. */ len = if_wubaput(&ix->ix_ifuba, m); if (len - sizeof(struct ether_header) < ETHERMIN) len = ETHERMIN + sizeof(struct ether_header); ix->ix_flags |= IXF_OACTIVE; /* Now setup to call np driver */ rpb[0] = 8; rpb[1] = ix->ix_aid; ix_DoReq(mp, rp, IXC_XMIT, /* send frame */ ix->ix_ifuba.ifu_w.ifrw_info, len, rpb, ixcint); } /* * Command done interrupt. (almost) */ ixcint(mp, rp) struct npmaster *mp; struct npreq *rp; { struct CQE *ep; register struct ix_softc *ix; int s = splimp(); ep = rp->element; ix = (struct ix_softc *)ep->cqe_famid; if ((ix->ix_flags & IXF_OACTIVE) == 0) { printf("ix%d: stray xmit interrupt, npreq=%x\n", ix->ix_if.if_unit, rp); } ix->ix_flags &= ~IXF_OACTIVE; switch (ep->cqe_func) { case IXC_XMIT: if (ep->cqe_sts == 1) ix->ix_if.if_opackets++; else ix->ix_if.if_oerrors++; break; case IXC_GSTAT: if (ep->cqe_sts == 1) ix->ix_if.if_collisions = ix->ix_stats.ixg.macg_xrty; break; } if (ix->ix_ifuba.ifu_xtofree) { m_freem(ix->ix_ifuba.ifu_xtofree); ix->ix_ifuba.ifu_xtofree = 0; } done: ixstart(ix->ix_if.if_unit); splx(s); } /* * Ethernet interface receiver interrupt. * If input error just drop packet. * Otherwise purge input buffered data path and examine * packet to determine type. If can't determine length * from type, then have to drop packet. Othewise decapsulate * packet based on type and pass to type specific higher-level * input routine. */ ixrint(mp, rp) struct npmaster *mp; struct npreq *rp; { struct CQE *ep; register struct ix_softc *ix = ix_softc + mp->unit; register struct ether_header *il; struct mbuf *m; int len, off, resid, s; register struct ifqueue *inq; if ((ix->ix_flags & IXF_RUNNING) == 0) return; if (rp == 0) goto setup; ix->ix_flags &= ~IXF_RCVPENDING; ep = rp->element; ix->ix_if.if_ipackets++; if (ix->ix_ifuba.ifu_flags & UBA_NEEDBDP) UBAPURGE(ix->ix_ifuba.ifu_uba, ix->ix_ifuba.ifu_r.ifrw_bdp); il = (struct ether_header *)(ix->ix_ifuba.ifu_r.ifrw_addr); len = ep->cqe_bcnt - sizeof (struct ether_header); if (ep->cqe_sts != NPDONE || ep->cqe_ust0 != NPDONE || ep->cqe_ust1 != NPOK) { printf("ixrint: cqe error %x, %x, %x\n", ep->cqe_sts, ep->cqe_ust0, ep->cqe_ust1); if (++ix->ix_badcqe > 100) { ix->ix_badcqe = 0; printf("ixrint: shutting down unix dla\n"); ix->ix_if.if_flags &= ~IFF_UP; return; } goto setup; } if ( len < 46 || len > ETHERMTU) { ix->ix_if.if_ierrors++; #ifdef notdef if (ix->ix_if.if_ierrors % 100 == 0) printf("ix%d: += 100 input errors\n", unit); #endif goto setup; } /* * Deal with trailer protocol: if type is trailer type * get true type from first 16-bit word past data. * Remember that type was trailer by setting off. */ il->ether_type = ntohs((u_short)il->ether_type); #define ildataaddr(il, off, type) ((type)(((caddr_t)((il)+1)+(off)))) if (il->ether_type >= ETHERTYPE_TRAIL && il->ether_type < ETHERTYPE_TRAIL+ETHERTYPE_NTRAILER) { off = (il->ether_type - ETHERTYPE_TRAIL) * 512; if (off >= ETHERMTU) goto setup; /* sanity */ il->ether_type = ntohs(*ildataaddr(il, off, u_short *)); resid = ntohs(*(ildataaddr(il, off+2, u_short *))); if (off + resid > len) goto setup; /* sanity */ len = off + resid; } else off = 0; if (len == 0) goto setup; /* * Pull packet off interface. Off is nonzero if packet * has trailing header; ilget will then force this header * information to be at the front, but we still have to drop * the type and length which are at the front of any trailer data. */ m = if_rubaget(&ix->ix_ifuba, len, off, &ix->ix_if); if (m == 0) goto setup; if (off) { struct ifnet *ifp; ifp = *(mtod(m, struct ifnet **)); m->m_off += 2 * sizeof (u_short); m->m_len -= 2 * sizeof (u_short); *(mtod(m, struct ifnet **)) = ifp; } switch (il->ether_type) { #ifdef INET case ETHERTYPE_IP: schednetisr(NETISR_IP); inq = &ipintrq; break; case ETHERTYPE_ARP: arpinput(&ix->ix_ac, m); goto setup; #endif #ifdef NS case ETHERTYPE_NS: schednetisr(NETISR_NS); inq = &nsintrq; break; #endif default: m_freem(m); goto setup; } s = splimp(); if (IF_QFULL(inq)) { IF_DROP(inq); m_freem(m); } else IF_ENQUEUE(inq, m); splx(s); setup: /* * Reset for next packet if possible. * If waiting for transmit command completion, set flag * and wait until command completes. */ if (rp == 0) { rp = ix->ix_rrp; rp->intr = ixrint; ep = rp->element; } len = ETHERMTU + sizeof(struct ether_header); /* Now setup to call np driver */ /* Initializations of request structure */ ep->cqe_func = IXC_RECV; /* get frame */ ep->cqe_ust0 = ep->cqe_ust1 = NPCLEAR; /* Clear status */ ep->cqe_bcnt = len; /* Byte count */ ep->cqe_lenrpb = 10; /* RPB length */ ep->rpb1 = ix->ix_aid; /* which channel */ ep->rpb2 = 65535; /* Timeout */ ix->ix_flags |= IXF_RCVPENDING; s = spl4(); NpAddCQE(ep, &mp->shmemp->devcq, mp); /* Add CQE to device's queue */ splx(s); } /* * Watchdog routine, request statistics from board. */ ixwatch(unit) int unit; { register struct ix_softc *ix = &ix_softc[unit]; register struct ifnet *ifp = &ix->ix_if; int s; if (ix->ix_flags & IXF_STATPENDING) { ifp->if_timer = ix->ix_scaninterval; return; } s = splimp(); ix->ix_flags |= IXF_STATPENDING; if ((ix->ix_flags & IXF_OACTIVE) == 0) ixstart(ifp->if_unit); splx(s); ifp->if_timer = ix->ix_scaninterval; } /* * Process an ioctl request. */ ixioctl(ifp, cmd, data) register struct ifnet *ifp; int cmd; caddr_t data; { register struct ifaddr *ifa = (struct ifaddr *)data; register struct ix_softc *ix = &ix_softc[ifp->if_unit]; int s = splimp(), error = 0; switch (cmd) { case SIOCSIFADDR: ifp->if_flags |= IFF_UP; ixinit(ifp->if_unit); if ((ifp->if_flags & IFF_UP) == 0) return (EBUSY); switch (ifa->ifa_addr.sa_family) { #ifdef INET case AF_INET: ((struct arpcom *)ifp)->ac_ipaddr = IA_SIN(ifa)->sin_addr; arpwhohas((struct arpcom *)ifp, &IA_SIN(ifa)->sin_addr); break; #endif #ifdef NS case AF_NS: { register struct ns_addr *ina = &(IA_SNS(ifa)->sns_addr); if (ns_nullhost(*ina)) { ina->x_host = * (union ns_host *) (ix_softc[ifp->if_unit].ix_addr); } else { return ix_setaddr(ina->x_host.c_host, ifp->if_unit); } break; } #endif } break; case SIOCSIFFLAGS: if ((ifp->if_flags & IFF_UP) == 0 && ix->ix_flags & IXF_RUNNING) { ix->ix_flags &= ~IXF_RUNNING; NpReset(ix->ix_mp, 0); } else if (ifp->if_flags & IFF_UP && (ix->ix_flags & IXF_RUNNING) == 0) ixinit(ifp->if_unit); break; default: error = EINVAL; } splx(s); return (error); } /* * set ethernet address for unit */ ix_setaddr(physaddr, unit) u_char *physaddr; int unit; { register struct ix_softc *ix = &ix_softc[unit]; if (! (ix->ix_flags & IXF_RUNNING)) return (EBUSY); /* The following is a big cop out due to the fact that Changing the ethernet address resets the dla module, so must re-open the channel, anyway. */ bcopy((caddr_t)physaddr, (caddr_t)ix->ix_addr, sizeof ix->ix_addr); ix->ix_flags &= ~IXF_RUNNING; ix->ix_flags |= IXF_SETADDR; ixinit(unit); NpKill(ix->ix_mp, ix->ix_rrp); } #endif