initial import of NetBSD tree

1 files changed, 1380 insertions, 0 deletions

diff --git a/sys/arch/mac68k/dev/if_ae.c b/sys/arch/mac68k/dev/if_ae.c
new file mode 100644
index 00000000000..e3b755d2015
--- /dev/null
+++ b/sys/arch/mac68k/dev/if_ae.c
@@ -0,0 +1,1380 @@
+/*	$NetBSD: if_ae.c,v 1.36 1995/09/24 14:13:58 briggs Exp $	*/
+
+/*
+ * Device driver for National Semiconductor DS8390/WD83C690 based ethernet
+ * adapters.
+ *
+ * Copyright (c) 1994, 1995 Charles M. Hannum.  All rights reserved.
+ *
+ * Copyright (C) 1993, David Greenman.  This software may be used, modified,
+ * copied, distributed, and sold, in both source and binary form provided that
+ * the above copyright and these terms are retained.  Under no circumstances is
+ * the author responsible for the proper functioning of this software, nor does
+ * the author assume any responsibility for damages incurred with its use.
+ *
+ * Adapted for MacBSD by Brad Parker <brad@fcr.com>.
+ *
+ * Currently supports:
+ *	Apples NB Ethernet card
+ *	Interlan A310 Nubus Ethernet card
+ *	Cayman Systems GatorCard
+ *	Asante MacCon II/E
+ */
+
+#include "bpfilter.h"
+
+#include <sys/param.h>
+#include <sys/types.h>
+#include <sys/systm.h>
+#include <sys/errno.h>
+#include <sys/ioctl.h>
+#include <sys/mbuf.h>
+#include <sys/socket.h>
+#include <sys/syslog.h>
+#include <sys/device.h>
+
+#include <net/if.h>
+#include <net/if_dl.h>
+#include <net/if_types.h>
+#include <net/netisr.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
+
+#if NBPFILTER > 0
+#include <net/bpf.h>
+#include <net/bpfdesc.h>
+#endif
+
+#include "../mac68k/via.h"
+#include "nubus.h"
+#include <dev/ic/dp8390reg.h>
+#include "if_aereg.h"
+
+#define INTERFACE_NAME_LEN	32
+
+/*
+ * ae_softc: per line info and status
+ */
+struct ae_softc {
+	struct device	sc_dev;
+	nubus_slot	sc_slot;
+/*	struct	intrhand sc_ih;	*/
+
+	struct arpcom sc_arpcom;/* ethernet common */
+
+	char	type_str[INTERFACE_NAME_LEN];	/* type string */
+	u_short	type;		/* interface type code */
+	u_char	vendor;		/* interface vendor */
+	u_char	regs_rev;	/* registers are reversed */
+
+#define	REG_MAP(sc, reg)	((sc)->regs_rev ? (0x0f-(reg))<<2 : (reg)<<2)
+#define NIC_GET(sc, reg)	((sc)->nic_addr[REG_MAP(sc, reg)])
+#define NIC_PUT(sc, reg, val)	((sc)->nic_addr[REG_MAP(sc, reg)] = (val))
+	volatile caddr_t nic_addr;	/* NIC (DS8390) I/O bus address */
+	caddr_t rom_addr;	/* on board prom address */
+
+	u_char  cr_proto;	/* values always set in CR */
+
+	caddr_t mem_start;	/* shared memory start address */
+	caddr_t mem_end;	/* shared memory end address */
+	u_long  mem_size;	/* total shared memory size */
+	caddr_t mem_ring;	/* start of RX ring-buffer (in smem) */
+
+	u_char  txb_cnt;	/* Number of transmit buffers */
+	u_char  txb_inuse;	/* number of transmit buffers active */
+
+	u_char  txb_new;	/* pointer to where new buffer will be added */
+	u_char  txb_next_tx;	/* pointer to next buffer ready to xmit */
+	u_short txb_len[8];	/* buffered xmit buffer lengths */
+	u_char  tx_page_start;	/* first page of TX buffer area */
+	u_char  rec_page_start;	/* first page of RX ring-buffer */
+	u_char  rec_page_stop;	/* last page of RX ring-buffer */
+	u_char  next_packet;	/* pointer to next unread RX packet */
+};
+
+int aeprobe __P((struct device *, void *, void *));
+void aeattach __P((struct device *, struct device *, void *));
+void aeintr __P((void *));
+int aeioctl __P((struct ifnet *, u_long, caddr_t));
+void aestart __P((struct ifnet *));
+void aewatchdog __P(( /* short */ ));
+void aereset __P((struct ae_softc *));
+void aeinit __P((struct ae_softc *));
+void aestop __P((struct ae_softc *));
+
+void aeread __P((struct ae_softc *, caddr_t, int));
+struct mbuf *aeget __P((struct ae_softc *, caddr_t, int));
+
+#define inline			/* XXX for debugging porpoises */
+
+u_short ae_put __P((struct ae_softc *, struct mbuf *, caddr_t));
+void ae_getmcaf __P((struct arpcom *, u_char *));
+
+static inline void ae_rint __P((struct ae_softc *));
+static inline void ae_xmit __P((struct ae_softc *));
+static inline caddr_t ae_ring_copy __P((
+		/* struct ae_softc *, caddr_t, caddr_t, u_short */ ));
+
+struct cfdriver aecd = {
+	NULL, "ae", aeprobe, aeattach, DV_IFNET, sizeof(struct ae_softc)
+};
+
+#define	ETHER_MIN_LEN	64
+#define ETHER_MAX_LEN	1518
+#define	ETHER_ADDR_LEN	6
+
+static char zero = 0;
+static u_char ones = 0xff;
+
+/*
+ * XXX These two should be moved to locore, and maybe changed to use shorts
+ * instead of bytes.  The reason for these is that bcopy and bzero use longs,
+ * which the ethernet cards can't handle.
+ */
+
+void
+bszero(u_short * addr, int len)
+{
+	while (len--)
+		*addr++ = 0;
+}
+
+/*
+ * Memory copy, copies word at time.
+ */
+static inline void
+word_copy(a, b, len)
+	caddr_t a, b;
+	int     len;
+{
+	u_short *x = (u_short *) a, *y = (u_short *) b;
+
+	len >>= 1;
+	while (len--)
+		*y++ = *x++;
+}
+
+/*
+ * Memory copy, copies bytes at time.
+ */
+static inline void
+byte_copy(a, b, len)
+	caddr_t a, b;
+	int     len;
+{
+	while (len--)
+		*b++ = *a++;
+}
+
+static int
+ae_id_card(slot, sc)
+	nubus_slot	*slot;
+	struct ae_softc *sc;
+{
+	nubus_dir	dir;
+	nubus_dirent	dirent;
+	nubus_type	slottype;
+
+	nubus_get_main_dir(slot, &dir);
+
+	if (nubus_find_rsrc(slot, &dir, 0x80, &dirent) <= 0)
+		return 0;
+
+	nubus_get_dir_from_rsrc(slot, &dirent, &dir);
+
+	if (nubus_find_rsrc(slot, &dir, NUBUS_RSRC_TYPE, &dirent) <= 0)
+		return 0;
+
+	if (nubus_get_ind_data(slot, &dirent,
+		(caddr_t) &slottype, sizeof(nubus_type)) <= 0)
+		return 0;
+
+	if (slottype.category != NUBUS_CATEGORY_NETWORK)
+		return 0;
+
+	if (slottype.type != NUBUS_TYPE_ETHERNET)
+		return 0;
+
+	switch (slottype.drsw) {
+	case NUBUS_DRSW_3COM:
+	case NUBUS_DRSW_APPLE:
+	case NUBUS_DRSW_TECHWORKS:
+		sc->vendor = AE_VENDOR_APPLE;
+		break;
+	case NUBUS_DRSW_ASANTE:
+		sc->vendor = AE_VENDOR_ASANTE;
+		break;
+	case NUBUS_DRSW_FARALLON:
+		sc->vendor = AE_VENDOR_FARALLON;
+		break;
+	case NUBUS_DRSW_GATOR:
+		switch (slottype.drhw) {
+		default:
+		case NUBUS_DRHW_INTERLAN:
+			sc->vendor = AE_VENDOR_INTERLAN;
+			break;
+		case NUBUS_DRHW_KINETICS:
+			sc->vendor = AE_VENDOR_DAYNA;
+			break;
+		}
+		break;
+	default:
+		sc->vendor = AE_VENDOR_UNKNOWN;
+		return 0;
+	}
+
+	strncpy(sc->type_str, nubus_get_card_name(slot), INTERFACE_NAME_LEN);
+
+	sc->type_str[INTERFACE_NAME_LEN-1] = '\0';
+
+	return 1;
+}
+
+int
+ae_size_card_memory(sc)
+	struct ae_softc *sc;
+{
+	u_short *p;
+	u_short i1, i2, i3, i4;
+	int     size;
+
+	p = (u_short *) sc->mem_start;
+
+	/*
+	 * very simple size memory, assuming it's installed in 8k
+	 * banks; also assume it will generally mirror in upper banks
+	 * if not installed.
+	 */
+	i1 = (8192 * 0) / 2;
+	i2 = (8192 * 1) / 2;
+	i3 = (8192 * 2) / 2;
+	i4 = (8192 * 3) / 2;
+
+	p[i1] = 0x1111;
+	p[i2] = 0x2222;
+	p[i3] = 0x3333;
+	p[i4] = 0x4444;
+
+	if (p[i1] == 0x1111 && p[i2] == 0x2222 &&
+	    p[i3] == 0x3333 && p[i4] == 0x4444)
+		return 8192 * 4;
+
+	if ((p[i1] == 0x1111 && p[i2] == 0x2222) ||
+	    (p[i1] == 0x3333 && p[i2] == 0x4444))
+		return 8192 * 2;
+
+	if (p[i1] == 0x1111 || p[i1] == 0x4444)
+		return 8192;
+
+	return 0;
+}
+
+int
+aeprobe(parent, match, aux)
+	struct device *parent;
+	void   *match, *aux;
+{
+	struct ae_softc *sc = match;
+	nubus_slot *nu = (nubus_slot *) aux;
+	caddr_t	addr;
+	int     i, memsize;
+	int     flags = 0;
+
+	if (ae_id_card(nu, sc) <= 0)
+		return 0;
+
+	sc->regs_rev = 0;
+
+	addr = (caddr_t) NUBUS_SLOT_TO_BASE(nu->slot);
+
+	switch (sc->vendor) {
+	case AE_VENDOR_INTERLAN:
+		sc->nic_addr = addr + GC_NIC_OFFSET;
+		sc->rom_addr = addr + GC_ROM_OFFSET;
+		sc->mem_start = addr + GC_DATA_OFFSET;
+		if ((memsize = ae_size_card_memory(sc)) == 0)
+			return 0;
+
+		/* reset the NIC chip */
+		*((caddr_t) addr + GC_RESET_OFFSET) = (char) zero;
+
+		/* Get station address from on-board ROM */
+		for (i = 0; i < ETHER_ADDR_LEN; ++i)
+			sc->sc_arpcom.ac_enaddr[i] = *(sc->rom_addr + i * 4);
+		break;
+
+		/* apple-compatible cards */
+	case AE_VENDOR_ASANTE:
+	case AE_VENDOR_APPLE:
+		sc->regs_rev = 1;
+		sc->nic_addr = addr + AE_NIC_OFFSET;
+		sc->rom_addr = addr + AE_ROM_OFFSET;
+		sc->mem_start = addr + AE_DATA_OFFSET;
+		if ((memsize = ae_size_card_memory(sc)) == 0)
+			return (0);
+
+		/* Get station address from on-board ROM */
+		for (i = 0; i < ETHER_ADDR_LEN; ++i)
+			sc->sc_arpcom.ac_enaddr[i] = *(sc->rom_addr + i * 2);
+		break;
+
+	case AE_VENDOR_DAYNA:
+		printf("We think we are a Dayna card, but ");
+		sc->nic_addr = addr + DP_NIC_OFFSET;
+		sc->rom_addr = addr + DP_ROM_OFFSET;
+		sc->mem_start = addr + DP_DATA_OFFSET;
+		memsize = 8192;
+
+		/* Get station address from on-board ROM */
+		for (i = 0; i < ETHER_ADDR_LEN; ++i)
+			sc->sc_arpcom.ac_enaddr[i] = *(sc->rom_addr + i * 2);
+		printf("it is dangerous to continue.\n");
+		return (0);	/* Since we don't work yet... */
+		break;
+
+	case AE_VENDOR_FARALLON:
+		sc->regs_rev = 1;
+		sc->rom_addr = addr + FE_ROM_OFFSET;
+		sc->nic_addr = addr + AE_NIC_OFFSET;
+		sc->mem_start = addr + AE_DATA_OFFSET;
+		if ((memsize = ae_size_card_memory(sc)) == 0)
+			return (0);
+
+		/* Get station address from on-board ROM */
+		for (i = 0; i < ETHER_ADDR_LEN; ++i)
+			sc->sc_arpcom.ac_enaddr[i] = *(sc->rom_addr + i);
+		break;
+
+	default:
+		return (0);
+		break;
+	}
+
+	sc->cr_proto = ED_CR_RD2;
+
+	/* Allocate one xmit buffer if < 16k, two buffers otherwise. */
+	if ((memsize < 16384) || (flags & AE_FLAGS_NO_DOUBLE_BUFFERING))
+		sc->txb_cnt = 1;
+	else
+		sc->txb_cnt = 2;
+
+	sc->tx_page_start = 0;
+	sc->rec_page_start = sc->tx_page_start + sc->txb_cnt * ED_TXBUF_SIZE;
+	sc->rec_page_stop = sc->tx_page_start + (memsize >> ED_PAGE_SHIFT);
+	sc->mem_ring = sc->mem_start + (sc->rec_page_start << ED_PAGE_SHIFT);
+	sc->mem_size = memsize;
+	sc->mem_end = sc->mem_start + memsize;
+
+	/* Now zero memory and verify that it is clear. */
+	bszero((u_short *) sc->mem_start, memsize / 2);
+
+	for (i = 0; i < memsize; ++i)
+		if (sc->mem_start[i]) {
+			printf("%s: failed to clear shared memory at %x - check configuration\n",
+			    sc->sc_dev.dv_xname,
+			    sc->mem_start + i);
+			return (0);
+		}
+
+	bcopy(nu, &sc->sc_slot, sizeof(nubus_slot));
+
+	return (1);
+}
+
+/*
+ * Install interface into kernel networking data structures
+ */
+void
+aeattach(parent, self, aux)
+	struct device *parent, *self;
+	void   *aux;
+{
+	struct ae_softc *sc = (void *) self;
+	struct nubus_hw *nu = aux;
+	struct cfdata *cf = sc->sc_dev.dv_cfdata;
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+
+	/* Set interface to stopped condition (reset). */
+	aestop(sc);
+
+	/* Initialize ifnet structure. */
+	ifp->if_unit = sc->sc_dev.dv_unit;
+	ifp->if_name = aecd.cd_name;
+	ifp->if_start = aestart;
+	ifp->if_ioctl = aeioctl;
+	ifp->if_watchdog = aewatchdog;
+	ifp->if_flags =
+	    IFF_BROADCAST | IFF_SIMPLEX | IFF_NOTRAILERS | IFF_MULTICAST;
+
+	/* Attach the interface. */
+	if_attach(ifp);
+	ether_ifattach(ifp);
+
+	/* Print additional info when attached. */
+	printf(": address %s, ", ether_sprintf(sc->sc_arpcom.ac_enaddr));
+
+	printf("type %s, %dk mem.\n", sc->type_str, sc->mem_size / 1024);
+
+#if NBPFILTER > 0
+	bpfattach(&ifp->if_bpf, ifp, DLT_EN10MB, sizeof(struct ether_header));
+#endif
+
+	/* make sure interrupts are vectored to us */
+	add_nubus_intr(sc->sc_slot.slot, aeintr, sc);
+
+	/*
+	 * XXX -- enable nubus interrupts here.  Should be done elsewhere,
+	 *        but that currently breaks with some nubus video cards'
+	 *	  interrupts.  So we only enable nubus interrupts if we
+	 *	  have an ethernet card...  i.e., we do it here.
+	 */
+	enable_nubus_intr();
+}
+
+/*
+ * Reset interface.
+ */
+void
+aereset(sc)
+	struct ae_softc *sc;
+{
+	int     s;
+
+	s = splimp();
+	aestop(sc);
+	aeinit(sc);
+	splx(s);
+}
+
+/*
+ * Take interface offline.
+ */
+void
+aestop(sc)
+	struct ae_softc *sc;
+{
+	int     n = 5000;
+
+	/* Stop everything on the interface, and select page 0 registers. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STP);
+
+	/*
+	 * Wait for interface to enter stopped state, but limit # of checks to
+	 * 'n' (about 5ms).  It shouldn't even take 5us on modern DS8390's, but
+	 * just in case it's an old one.
+	 */
+	while (((NIC_GET(sc, ED_P0_ISR) & ED_ISR_RST) == 0) && --n);
+}
+
+/*
+ * Device timeout/watchdog routine.  Entered if the device neglects to generate
+ * an interrupt after a transmit has been started on it.
+ */
+static int aeintr_ctr = 0;
+void
+aewatchdog(unit)
+	int     unit;
+{
+	struct ae_softc *sc = aecd.cd_devs[unit];
+
+#if 1
+/*
+ * This is a kludge!  The via code seems to miss slot interrupts
+ * sometimes.  This kludges around that by calling the handler
+ * by hand if the watchdog is activated. -- XXX (akb)
+ */
+	int     i;
+
+	i = aeintr_ctr;
+
+	(*via2itab[1]) (1);
+
+	if (i != aeintr_ctr) {
+		log(LOG_ERR, "ae%d: device timeout, recovered\n", unit);
+		return;
+	}
+#endif
+
+	log(LOG_ERR, "%s: device timeout\n", sc->sc_dev.dv_xname);
+	++sc->sc_arpcom.ac_if.if_oerrors;
+
+	aereset(sc);
+}
+
+/*
+ * Initialize device.
+ */
+void
+aeinit(sc)
+	struct ae_softc *sc;
+{
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+	int     i;
+	u_char  command;
+	u_char  mcaf[8];
+
+	/*
+	 * Initialize the NIC in the exact order outlined in the NS manual.
+	 * This init procedure is "mandatory"...don't change what or when
+	 * things happen.
+	 */
+
+	/* Reset transmitter flags. */
+	ifp->if_timer = 0;
+
+	sc->txb_inuse = 0;
+	sc->txb_new = 0;
+	sc->txb_next_tx = 0;
+
+	/* Set interface for page 0, remote DMA complete, stopped. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STP);
+
+	/*
+	 * Set FIFO threshold to 8, No auto-init Remote DMA, byte
+	 * order=80x86, word-wide DMA xfers,
+	 */
+	NIC_PUT(sc, ED_P0_DCR,
+	    ED_DCR_FT1 | ED_DCR_WTS | ED_DCR_LS);
+
+	/* Clear remote byte count registers. */
+	NIC_PUT(sc, ED_P0_RBCR0, 0);
+	NIC_PUT(sc, ED_P0_RBCR1, 0);
+
+	/* Tell RCR to do nothing for now. */
+	NIC_PUT(sc, ED_P0_RCR, ED_RCR_MON);
+
+	/* Place NIC in internal loopback mode. */
+	NIC_PUT(sc, ED_P0_TCR, ED_TCR_LB0);
+
+	/* Initialize receive buffer ring. */
+	NIC_PUT(sc, ED_P0_TPSR, sc->rec_page_start);
+	NIC_PUT(sc, ED_P0_PSTART, sc->rec_page_start);
+
+	NIC_PUT(sc, ED_P0_PSTOP, sc->rec_page_stop);
+	NIC_PUT(sc, ED_P0_BNRY, sc->rec_page_start);
+
+	/*
+	 * Clear all interrupts.  A '1' in each bit position clears the
+	 * corresponding flag.
+	 */
+	NIC_PUT(sc, ED_P0_ISR, 0xff);
+
+	/*
+	 * Enable the following interrupts: receive/transmit complete,
+	 * receive/transmit error, and Receiver OverWrite.
+	 *
+	 * Counter overflow and Remote DMA complete are *not* enabled.
+	 */
+	NIC_PUT(sc, ED_P0_IMR,
+	    ED_IMR_PRXE | ED_IMR_PTXE | ED_IMR_RXEE | ED_IMR_TXEE |
+	    ED_IMR_OVWE);
+
+	/* Program command register for page 1. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_1 | ED_CR_STP);
+
+	/* Copy out our station address. */
+	for (i = 0; i < ETHER_ADDR_LEN; ++i)
+		NIC_PUT(sc, ED_P1_PAR0 + i, sc->sc_arpcom.ac_enaddr[i]);
+
+	/* Set multicast filter on chip. */
+	ae_getmcaf(&sc->sc_arpcom, mcaf);
+	for (i = 0; i < 8; i++)
+		NIC_PUT(sc, ED_P1_MAR0 + i, mcaf[i]);
+
+	/*
+	 * Set current page pointer to one page after the boundary pointer, as
+	 * recommended in the National manual.
+	 */
+	sc->next_packet = sc->rec_page_start + 1;
+	NIC_PUT(sc, ED_P1_CURR, sc->next_packet);
+
+	/* Program command register for page 0. */
+	NIC_PUT(sc, ED_P1_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STP);
+
+	i = ED_RCR_AB | ED_RCR_AM;
+	if (ifp->if_flags & IFF_PROMISC) {
+		/*
+		 * Set promiscuous mode.  Multicast filter was set earlier so
+		 * that we should receive all multicast packets.
+		 */
+		i |= ED_RCR_PRO | ED_RCR_AR | ED_RCR_SEP;
+	}
+	NIC_PUT(sc, ED_P0_RCR, i);
+
+	/* Take interface out of loopback. */
+	NIC_PUT(sc, ED_P0_TCR, 0);
+
+	/* Fire up the interface. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STA);
+
+	/* Set 'running' flag, and clear output active flag. */
+	ifp->if_flags |= IFF_RUNNING;
+	ifp->if_flags &= ~IFF_OACTIVE;
+
+	/* ...and attempt to start output. */
+	aestart(ifp);
+}
+
+/*
+ * This routine actually starts the transmission on the interface.
+ */
+static inline void
+ae_xmit(sc)
+	struct ae_softc *sc;
+{
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+	u_short len;
+
+	len = sc->txb_len[sc->txb_next_tx];
+
+	/* Set NIC for page 0 register access. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STA);
+
+	/* Set TX buffer start page. */
+	NIC_PUT(sc, ED_P0_TPSR, sc->tx_page_start +
+	    sc->txb_next_tx * ED_TXBUF_SIZE);
+
+	/* Set TX length. */
+	NIC_PUT(sc, ED_P0_TBCR0, len);
+	NIC_PUT(sc, ED_P0_TBCR1, len >> 8);
+
+	/* Set page 0, remote DMA complete, transmit packet, and *start*. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_TXP | ED_CR_STA);
+
+	/* Point to next transmit buffer slot and wrap if necessary. */
+	sc->txb_next_tx++;
+	if (sc->txb_next_tx == sc->txb_cnt)
+		sc->txb_next_tx = 0;
+
+	/* Set a timer just in case we never hear from the board again. */
+	ifp->if_timer = 2;
+}
+
+/*
+ * Start output on interface.
+ * We make two assumptions here:
+ *  1) that the current priority is set to splimp _before_ this code
+ *     is called *and* is returned to the appropriate priority after
+ *     return
+ *  2) that the IFF_OACTIVE flag is checked before this code is called
+ *     (i.e. that the output part of the interface is idle)
+ */
+void
+aestart(ifp)
+	struct ifnet *ifp;
+{
+	struct ae_softc *sc = aecd.cd_devs[ifp->if_unit];
+	struct mbuf *m0;
+	caddr_t buffer;
+	int     len;
+
+	if ((ifp->if_flags & (IFF_RUNNING | IFF_OACTIVE)) != IFF_RUNNING)
+		return;
+
+outloop:
+	/* See if there is room to put another packet in the buffer. */
+	if (sc->txb_inuse == sc->txb_cnt) {
+		/* No room.  Indicate this to the outside world and exit. */
+		ifp->if_flags |= IFF_OACTIVE;
+		return;
+	}
+	IF_DEQUEUE(&ifp->if_snd, m0);
+	if (m0 == 0)
+		return;
+
+	/* We need to use m->m_pkthdr.len, so require the header */
+	if ((m0->m_flags & M_PKTHDR) == 0)
+		panic("aestart: no header mbuf");
+
+#if NBPFILTER > 0
+	/* Tap off here if there is a BPF listener. */
+	if (ifp->if_bpf)
+		bpf_mtap(ifp->if_bpf, m0);
+#endif
+
+	/* txb_new points to next open buffer slot. */
+	buffer = sc->mem_start + ((sc->txb_new * ED_TXBUF_SIZE) << ED_PAGE_SHIFT);
+
+	len = ae_put(sc, m0, buffer);
+#if DIAGNOSTIC
+	if (len != m0->m_pkthdr.len)
+		printf("aestart: len %d != m0->m_pkthdr.len %d.\n",
+			len, m0->m_pkthdr.len);
+#endif
+	len = m0->m_pkthdr.len;
+
+	m_freem(m0);
+	sc->txb_len[sc->txb_new] = max(len, ETHER_MIN_LEN);
+
+	/* Start the first packet transmitting. */
+	if (sc->txb_inuse == 0)
+		ae_xmit(sc);
+
+	/* Point to next buffer slot and wrap if necessary. */
+	if (++sc->txb_new == sc->txb_cnt)
+		sc->txb_new = 0;
+
+	sc->txb_inuse++;
+
+	/* Loop back to the top to possibly buffer more packets. */
+	goto outloop;
+}
+
+/*
+ * Ethernet interface receiver interrupt.
+ */
+static inline void
+ae_rint(sc)
+	struct ae_softc *sc;
+{
+	u_char  boundary, current;
+	u_short len;
+	u_char  nlen, *lenp;
+	struct ae_ring packet_hdr;
+	caddr_t packet_ptr;
+
+loop:
+	/* Set NIC to page 1 registers to get 'current' pointer. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_1 | ED_CR_STA);
+
+	/*
+	 * 'sc->next_packet' is the logical beginning of the ring-buffer - i.e.
+	 * it points to where new data has been buffered.  The 'CURR' (current)
+	 * register points to the logical end of the ring-buffer - i.e. it
+	 * points to where additional new data will be added.  We loop here
+	 * until the logical beginning equals the logical end (or in other
+	 * words, until the ring-buffer is empty).
+	 */
+	current = NIC_GET(sc, ED_P1_CURR);
+	if (sc->next_packet == current)
+		return;
+
+	/* Set NIC to page 0 registers to update boundary register. */
+	NIC_PUT(sc, ED_P1_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STA);
+
+	do {
+		/* Get pointer to this buffer's header structure. */
+		packet_ptr = sc->mem_ring +
+		    ((sc->next_packet - sc->rec_page_start) << ED_PAGE_SHIFT);
+
+		/*
+		 * The byte count includes a 4 byte header that was added by
+		 * the NIC.
+		 */
+		packet_hdr = *(struct ae_ring *) packet_ptr;
+		lenp = (u_char *) &((struct ae_ring *) packet_ptr)->count;
+		len = lenp[0] | (lenp[1] << 8);
+		packet_hdr.count = len;
+
+		/*
+		 * Try do deal with old, buggy chips that sometimes duplicate
+		 * the low byte of the length into the high byte.  We do this
+		 * by simply ignoring the high byte of the length and always
+		 * recalculating it.
+		 *
+		 * NOTE: sc->next_packet is pointing at the current packet.
+		 */
+		if (packet_hdr.next_packet >= sc->next_packet)
+			nlen = (packet_hdr.next_packet - sc->next_packet);
+		else
+			nlen = ((packet_hdr.next_packet - sc->rec_page_start) +
+			    (sc->rec_page_stop - sc->next_packet));
+		--nlen;
+		if ((len & ED_PAGE_MASK) + sizeof(packet_hdr) > ED_PAGE_SIZE)
+			--nlen;
+		len = (len & ED_PAGE_MASK) | (nlen << ED_PAGE_SHIFT);
+#ifdef DIAGNOSTIC
+		if (len != packet_hdr.count) {
+			printf("%s: length does not match next packet pointer\n",
+			    sc->sc_dev.dv_xname);
+			printf("%s: len %04x nlen %04x start %02x first %02x curr %02x next %02x stop %02x\n",
+			    sc->sc_dev.dv_xname, packet_hdr.count, len,
+			    sc->rec_page_start, sc->next_packet, current,
+			    packet_hdr.next_packet, sc->rec_page_stop);
+		}
+#endif
+
+		/*
+		 * Be fairly liberal about what we allow as a "reasonable"
+		 * length so that a [crufty] packet will make it to BPF (and
+		 * can thus be analyzed).  Note that all that is really
+		 * important is that we have a length that will fit into one
+		 * mbuf cluster or less; the upper layer protocols can then
+		 * figure out the length from their own length field(s).
+		 */
+		if (len <= MCLBYTES &&
+		    packet_hdr.next_packet >= sc->rec_page_start &&
+		    packet_hdr.next_packet < sc->rec_page_stop) {
+			/* Go get packet. */
+			aeread(sc, packet_ptr + sizeof(struct ae_ring),
+			    len - sizeof(struct ae_ring));
+			++sc->sc_arpcom.ac_if.if_ipackets;
+		} else {
+			/* Really BAD.  The ring pointers are corrupted. */
+			log(LOG_ERR,
+			    "%s: NIC memory corrupt - invalid packet length %d\n",
+			    sc->sc_dev.dv_xname, len);
+			++sc->sc_arpcom.ac_if.if_ierrors;
+			aereset(sc);
+			return;
+		}
+
+		/* Update next packet pointer. */
+		sc->next_packet = packet_hdr.next_packet;
+
+		/*
+		 * Update NIC boundary pointer - being careful to keep it one
+		 * buffer behind (as recommended by NS databook).
+		 */
+		boundary = sc->next_packet - 1;
+		if (boundary < sc->rec_page_start)
+			boundary = sc->rec_page_stop - 1;
+		NIC_PUT(sc, ED_P0_BNRY, boundary);
+	} while (sc->next_packet != current);
+
+	goto loop;
+}
+
+/* Ethernet interface interrupt processor. */
+void
+aeintr(arg)
+	void *arg;
+{
+	struct ae_softc *sc = arg;
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+	u_char  isr;
+
+	aeintr_ctr++;
+
+	/* Set NIC to page 0 registers. */
+	NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STA);
+
+	isr = NIC_GET(sc, ED_P0_ISR);
+	if (!isr)
+		return;
+
+	/* Loop until there are no more new interrupts. */
+	for (;;) {
+		/*
+		 * Reset all the bits that we are 'acknowledging' by writing a
+		 * '1' to each bit position that was set.
+		 * (Writing a '1' *clears* the bit.)
+		 */
+		NIC_PUT(sc, ED_P0_ISR, isr);
+
+		/*
+		 * Handle transmitter interrupts.  Handle these first because
+		 * the receiver will reset the board under some conditions.
+		 */
+		if (isr & (ED_ISR_PTX | ED_ISR_TXE)) {
+			u_char  collisions = NIC_GET(sc, ED_P0_NCR) & 0x0f;
+
+			/*
+			 * Check for transmit error.  If a TX completed with an
+			 * error, we end up throwing the packet away.  Really
+			 * the only error that is possible is excessive
+			 * collisions, and in this case it is best to allow the
+			 * automatic mechanisms of TCP to backoff the flow.  Of
+			 * course, with UDP we're screwed, but this is expected
+			 * when a network is heavily loaded.
+			 */
+			(void) NIC_GET(sc, ED_P0_TSR);
+			if (isr & ED_ISR_TXE) {
+				/*
+				 * Excessive collisions (16).
+				 */
+				if ((NIC_GET(sc, ED_P0_TSR) & ED_TSR_ABT)
+				    && (collisions == 0)) {
+					/*
+					 * When collisions total 16, the P0_NCR
+					 * will indicate 0, and the TSR_ABT is
+					 * set.
+					 */
+					collisions = 16;
+				}
+				/* Update output errors counter. */
+				++ifp->if_oerrors;
+			} else {
+				/*
+				 * Update total number of successfully
+				 * transmitted packets.
+				 */
+				++ifp->if_opackets;
+			}
+
+			/* Done with the buffer. */
+			sc->txb_inuse--;
+
+			/* Clear watchdog timer. */
+			ifp->if_timer = 0;
+			ifp->if_flags &= ~IFF_OACTIVE;
+
+			/*
+			 * Add in total number of collisions on last
+			 * transmission.
+			 */
+			ifp->if_collisions += collisions;
+
+			/*
+			 * Decrement buffer in-use count if not zero (can only
+			 * be zero if a transmitter interrupt occured while not
+			 * actually transmitting).
+			 * If data is ready to transmit, start it transmitting,
+			 * otherwise defer until after handling receiver.
+			 */
+			if (sc->txb_inuse > 0)
+				ae_xmit(sc);
+		}
+		/* Handle receiver interrupts. */
+		if (isr & (ED_ISR_PRX | ED_ISR_RXE | ED_ISR_OVW)) {
+			/*
+			 * Overwrite warning.  In order to make sure that a
+			 * lockup of the local DMA hasn't occurred, we reset
+			 * and re-init the NIC.  The NSC manual suggests only a
+			 * partial reset/re-init is necessary - but some chips
+			 * seem to want more.  The DMA lockup has been seen
+			 * only with early rev chips - Methinks this bug was
+			 * fixed in later revs.  -DG
+			 */
+			if (isr & ED_ISR_OVW) {
+				++ifp->if_ierrors;
+#ifdef DIAGNOSTIC
+				log(LOG_WARNING,
+				    "%s: warning - receiver ring buffer overrun\n",
+				    sc->sc_dev.dv_xname);
+#endif
+				/* Stop/reset/re-init NIC. */
+				aereset(sc);
+			} else {
+				/*
+				 * Receiver Error.  One or more of: CRC error,
+				 * frame alignment error FIFO overrun, or
+				 * missed packet.
+				 */
+				if (isr & ED_ISR_RXE) {
+					++ifp->if_ierrors;
+#ifdef AE_DEBUG
+					printf("%s: receive error %x\n",
+					    sc->sc_dev.dv_xname,
+					    NIC_GET(sc, ED_P0_RSR));
+#endif
+				}
+				/*
+				 * Go get the packet(s)
+				 * XXX - Doing this on an error is dubious
+				 * because there shouldn't be any data to get
+				 * (we've configured the interface to not
+				 * accept packets with errors).
+				 */
+				ae_rint(sc);
+			}
+		}
+		/*
+		 * If it looks like the transmitter can take more data, attempt
+		 * to start output on the interface.  This is done after
+		 * handling the receiver to give the receiver priority.
+		 */
+		aestart(ifp);
+
+		/*
+		 * Return NIC CR to standard state: page 0, remote DMA
+		 * complete, start (toggling the TXP bit off, even if was just
+		 * set in the transmit routine, is *okay* - it is 'edge'
+		 * triggered from low to high).
+		 */
+		NIC_PUT(sc, ED_P0_CR, sc->cr_proto | ED_CR_PAGE_0 | ED_CR_STA);
+
+		/*
+		 * If the Network Talley Counters overflow, read them to reset
+		 * them.  It appears that old 8390's won't clear the ISR flag
+		 * otherwise - resulting in an infinite loop.
+		 */
+		if (isr & ED_ISR_CNT) {
+			(void) NIC_GET(sc, ED_P0_CNTR0);
+			(void) NIC_GET(sc, ED_P0_CNTR1);
+			(void) NIC_GET(sc, ED_P0_CNTR2);
+		}
+		isr = NIC_GET(sc, ED_P0_ISR);
+		if (!isr)
+			return;
+	}
+}
+
+/*
+ * Process an ioctl request.  This code needs some work - it looks pretty ugly.
+ */
+int
+aeioctl(ifp, cmd, data)
+	register struct ifnet *ifp;
+	u_long  cmd;
+	caddr_t data;
+{
+	struct ae_softc *sc = aecd.cd_devs[ifp->if_unit];
+	register struct ifaddr *ifa = (struct ifaddr *) data;
+	struct ifreq *ifr = (struct ifreq *) data;
+	int     s, error = 0;
+
+	s = splimp();
+
+	switch (cmd) {
+
+	case SIOCSIFADDR:
+		ifp->if_flags |= IFF_UP;
+
+		switch (ifa->ifa_addr->sa_family) {
+#ifdef INET
+		case AF_INET:
+			aeinit(sc);
+			arp_ifinit(&sc->sc_arpcom, ifa);
+			break;
+#endif
+#ifdef NS
+			/* XXX - This code is probably wrong. */
+		case AF_NS:
+			{
+				register struct ns_addr *ina = &IA_SNS(ifa)->sns_addr;
+
+				if (ns_nullhost(*ina))
+					ina->x_host =
+					    *(union ns_host *) (sc->sc_arpcom.ac_enaddr);
+				else
+					bcopy(ina->x_host.c_host,
+					    sc->sc_arpcom.ac_enaddr,
+					    sizeof(sc->sc_arpcom.ac_enaddr));
+				/* Set new address. */
+				aeinit(sc);
+				break;
+			}
+#endif
+		default:
+			aeinit(sc);
+			break;
+		}
+		break;
+
+	case SIOCSIFFLAGS:
+		if ((ifp->if_flags & IFF_UP) == 0 &&
+		    (ifp->if_flags & IFF_RUNNING) != 0) {
+			/*
+			 * If interface is marked down and it is running, then
+			 * stop it.
+			 */
+			aestop(sc);
+			ifp->if_flags &= ~IFF_RUNNING;
+		} else
+			if ((ifp->if_flags & IFF_UP) != 0 &&
+			    (ifp->if_flags & IFF_RUNNING) == 0) {
+				/*
+				 * If interface is marked up and it is stopped, then
+				 * start it.
+				 */
+				aeinit(sc);
+			} else {
+				/*
+				 * Reset the interface to pick up changes in any other
+				 * flags that affect hardware registers.
+				 */
+				aestop(sc);
+				aeinit(sc);
+			}
+		break;
+
+	case SIOCADDMULTI:
+	case SIOCDELMULTI:
+		/* Update our multicast list. */
+		error = (cmd == SIOCADDMULTI) ?
+		    ether_addmulti(ifr, &sc->sc_arpcom) :
+		    ether_delmulti(ifr, &sc->sc_arpcom);
+
+		if (error == ENETRESET) {
+			/*
+			 * Multicast list has changed; set the hardware filter
+			 * accordingly.
+			 */
+			aestop(sc);	/* XXX for ds_setmcaf? */
+			aeinit(sc);
+			error = 0;
+		}
+		break;
+
+	default:
+		error = EINVAL;
+		break;
+	}
+
+	splx(s);
+	return (error);
+}
+
+/*
+ * Retreive packet from shared memory and send to the next level up via
+ * ether_input().  If there is a BPF listener, give a copy to BPF, too.
+ */
+void
+aeread(sc, buf, len)
+	struct ae_softc *sc;
+	caddr_t buf;
+	int len;
+{
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+	struct mbuf *m;
+	struct ether_header *eh;
+
+	/* Pull packet off interface. */
+	m = aeget(sc, buf, len);
+	if (m == 0) {
+		ifp->if_ierrors++;
+		return;
+	}
+
+	ifp->if_ipackets++;
+
+	/* We assume that the header fits entirely in one mbuf. */
+	eh = mtod(m, struct ether_header *);
+
+#if NBPFILTER > 0
+	/*
+	 * Check if there's a BPF listener on this interface.
+	 * If so, hand off the raw packet to bpf.
+	 */
+	if (ifp->if_bpf) {
+		bpf_mtap(ifp->if_bpf, m);
+
+		/*
+		 * Note that the interface cannot be in promiscuous mode if
+		 * there are no BPF listeners.  And if we are in promiscuous
+		 * mode, we have to check if this packet is really ours.
+		 */
+		if ((ifp->if_flags & IFF_PROMISC) &&
+		    (eh->ether_dhost[0] & 1) == 0 &&	/* !mcast and !bcast */
+		    bcmp(eh->ether_dhost, sc->sc_arpcom.ac_enaddr,
+			sizeof(eh->ether_dhost)) != 0) {
+			m_freem(m);
+			return;
+		}
+	}
+#endif
+
+	/* Fix up data start offset in mbuf to point past ether header. */
+	m_adj(m, sizeof(struct ether_header));
+	ether_input(ifp, eh, m);
+}
+
+/*
+ * Supporting routines.
+ */
+/*
+ * Given a source and destination address, copy 'amount' of a packet from the
+ * ring buffer into a linear destination buffer.  Takes into account ring-wrap.
+ */
+static inline caddr_t
+ae_ring_copy(sc, src, dst, amount)
+	struct ae_softc *sc;
+	caddr_t src, dst;
+	u_short amount;
+{
+	u_short tmp_amount;
+
+	/* Does copy wrap to lower addr in ring buffer? */
+	if (src + amount > sc->mem_end) {
+		tmp_amount = sc->mem_end - src;
+
+		/* Copy amount up to end of NIC memory. */
+		byte_copy(src, dst, tmp_amount);
+
+		amount -= tmp_amount;
+		src = sc->mem_ring;
+		dst += tmp_amount;
+	}
+	byte_copy(src, dst, amount);
+
+	return (src + amount);
+}
+
+/*
+ * Copy data from receive buffer to end of mbuf chain allocate additional mbufs
+ * as needed.  Return pointer to last mbuf in chain.
+ * sc = ae info (softc)
+ * src = pointer in ae ring buffer
+ * dst = pointer to last mbuf in mbuf chain to copy to
+ * amount = amount of data to copy
+ */
+struct mbuf *
+aeget(sc, src, total_len)
+	struct ae_softc *sc;
+	caddr_t src;
+	u_short total_len;
+{
+	struct ifnet *ifp = &sc->sc_arpcom.ac_if;
+	struct mbuf *top, **mp, *m;
+	int len;
+
+	MGETHDR(m, M_DONTWAIT, MT_DATA);
+	if (m == 0)
+		return 0;
+	m->m_pkthdr.rcvif = ifp;
+	m->m_pkthdr.len = total_len;
+	len = MHLEN;
+	top = 0;
+	mp = &top;
+
+	while (total_len > 0) {
+		if (top) {
+			MGET(m, M_DONTWAIT, MT_DATA);
+			if (m == 0) {
+				m_freem(top);
+				return 0;
+			}
+			len = MLEN;
+		}
+		if (total_len >= MINCLSIZE) {
+			MCLGET(m, M_DONTWAIT);
+			if (m->m_flags & M_EXT)
+				len = MCLBYTES;
+		}
+		m->m_len = len = min(total_len, len);
+		src = ae_ring_copy(sc, src, mtod(m, caddr_t), len);
+		total_len -= len;
+		*mp = m;
+		mp = &m->m_next;
+	}
+
+	return top;
+}
+/*
+ * Compute the multicast address filter from the list of multicast addresses we
+ * need to listen to.
+ */
+void
+ae_getmcaf(ac, af)
+	struct arpcom *ac;
+	u_char *af;
+{
+	struct ifnet *ifp = &ac->ac_if;
+	struct ether_multi *enm;
+	register u_char *cp, c;
+	register u_long crc;
+	register int i, len;
+	struct ether_multistep step;
+
+	/*
+	 * Set up multicast address filter by passing all multicast addresses
+	 * through a crc generator, and then using the high order 6 bits as an
+	 * index into the 64 bit logical address filter.  The high order bit
+	 * selects the word, while the rest of the bits select the bit within
+	 * the word.
+	 */
+
+	if (ifp->if_flags & IFF_PROMISC) {
+		ifp->if_flags |= IFF_ALLMULTI;
+		for (i = 0; i < 8; i++)
+			af[i] = 0xff;
+		return;
+	}
+	for (i = 0; i < 8; i++)
+		af[i] = 0;
+	ETHER_FIRST_MULTI(step, ac, enm);
+	while (enm != NULL) {
+		if (bcmp(enm->enm_addrlo, enm->enm_addrhi,
+			sizeof(enm->enm_addrlo)) != 0) {
+			/*
+			 * We must listen to a range of multicast addresses.
+			 * For now, just accept all multicasts, rather than
+			 * trying to set only those filter bits needed to match
+			 * the range.  (At this time, the only use of address
+			 * ranges is for IP multicast routing, for which the
+			 * range is big enough to require all bits set.)
+			 */
+			ifp->if_flags |= IFF_ALLMULTI;
+			for (i = 0; i < 8; i++)
+				af[i] = 0xff;
+			return;
+		}
+		cp = enm->enm_addrlo;
+		crc = 0xffffffff;
+		for (len = sizeof(enm->enm_addrlo); --len >= 0;) {
+			c = *cp++;
+			for (i = 8; --i >= 0;) {
+				if (((crc & 0x80000000) ? 1 : 0) ^ (c & 0x01)) {
+					crc <<= 1;
+					crc ^= 0x04c11db6 | 1;
+				} else
+					crc <<= 1;
+				c >>= 1;
+			}
+		}
+		/* Just want the 6 most significant bits. */
+		crc >>= 26;
+
+		/* Turn on the corresponding bit in the filter. */
+		af[crc >> 3] |= 1 << (crc & 0x7);
+
+		ETHER_NEXT_MULTI(step, enm);
+	}
+	ifp->if_flags &= ~IFF_ALLMULTI;
+}
+/*
+ * Copy packet from mbuf to the board memory
+ *
+ * Currently uses an extra buffer/extra memory copy,
+ * unless the whole packet fits in one mbuf.
+ *
+ */
+u_short
+ae_put(sc, m, buf)
+	struct ae_softc *sc;
+	struct mbuf *m;
+	caddr_t buf;
+{
+	u_char *data, savebyte[2];
+	int     len, wantbyte;
+	u_short totlen = 0;
+
+	wantbyte = 0;
+
+	for (; m ; m = m->m_next) {
+		data = mtod(m, u_char *);
+		len = m->m_len;
+		totlen += len;
+		if (len > 0) {
+			/* Finish the last word. */
+			if (wantbyte) {
+				savebyte[1] = *data;
+				word_copy(savebyte, buf, 2);
+				buf += 2;
+				data++;
+				len--;
+				wantbyte = 0;
+			}
+			/* Output contiguous words. */
+			if (len > 1) {
+				word_copy(data, buf, len);
+				buf += len & ~1;
+				data += len & ~1;
+				len &= 1;
+			}
+			/* Save last byte, if necessary. */
+			if (len == 1) {
+				savebyte[0] = *data;
+				wantbyte = 1;
+			}
+		}
+	}
+
+	if (wantbyte) {
+		savebyte[1] = 0;
+		word_copy(savebyte, buf, 2);
+	}
+	return (totlen);
+}