e1000_main.c

COPE the first practical network coding scheme which is developped on click

		/* Workaround for potential 82544 hang in PCI-X.  Avoid
		 * terminating buffers within evenly-aligned dwords. */
		if(unlikely(adapter->pcix_82544 &&
		   !((unsigned long)(skb->data + offset + size - 1) & 4) &&
		   size > 4))
			size -= 4;

		buffer_info->length = size;
		buffer_info->dma =
			pci_map_single(adapter->pdev,
				skb->data + offset,
				size,
				PCI_DMA_TODEVICE);
		buffer_info->time_stamp = jiffies;

		len -= size;
		offset += size;
		count++;
		if(unlikely(++i == tx_ring->count)) i = 0;
	}

#ifdef MAX_SKB_FRAGS
	for(f = 0; f < nr_frags; f++) {
		struct skb_frag_struct *frag;

		frag = &skb_shinfo(skb)->frags[f];
		len = frag->size;
		offset = frag->page_offset;

		while(len) {
			buffer_info = &tx_ring->buffer_info[i];
			size = min(len, max_per_txd);
#ifdef NETIF_F_TSO
			/* Workaround for premature desc write-backs
			 * in TSO mode.  Append 4-byte sentinel desc */
			if(unlikely(mss && f == (nr_frags-1) && size == len && size > 8))
				size -= 4;
#endif
			/* Workaround for potential 82544 hang in PCI-X.
			 * Avoid terminating buffers within evenly-aligned
			 * dwords. */
			if(unlikely(adapter->pcix_82544 &&
			   !((unsigned long)(frag->page+offset+size-1) & 4) &&
			   size > 4))
				size -= 4;

			buffer_info->length = size;
			buffer_info->dma =
				pci_map_page(adapter->pdev,
					frag->page,
					offset,
					size,
					PCI_DMA_TODEVICE);
			buffer_info->time_stamp = jiffies;

			len -= size;
			offset += size;
			count++;
			if(unlikely(++i == tx_ring->count)) i = 0;
		}
	}
#endif

	i = (i == 0) ? tx_ring->count - 1 : i - 1;
	tx_ring->buffer_info[i].skb = skb;
	tx_ring->buffer_info[first].next_to_watch = i;

	return count;
}

static inline void
e1000_tx_queue(struct e1000_adapter *adapter, int count, int tx_flags)
{
	struct e1000_desc_ring *tx_ring = &adapter->tx_ring;
	struct e1000_tx_desc *tx_desc = NULL;
	struct e1000_buffer *buffer_info;
	uint32_t txd_upper = 0, txd_lower = E1000_TXD_CMD_IFCS;
	unsigned int i;

	if(likely(tx_flags & E1000_TX_FLAGS_TSO)) {
		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D |
		             E1000_TXD_CMD_TSE;
		txd_upper |= (E1000_TXD_POPTS_IXSM | E1000_TXD_POPTS_TXSM) << 8;
	}

	if(likely(tx_flags & E1000_TX_FLAGS_CSUM)) {
		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D;
		txd_upper |= E1000_TXD_POPTS_TXSM << 8;
	}

	if(unlikely(tx_flags & E1000_TX_FLAGS_VLAN)) {
		txd_lower |= E1000_TXD_CMD_VLE;
		txd_upper |= (tx_flags & E1000_TX_FLAGS_VLAN_MASK);
	}

	i = tx_ring->next_to_use;

	while(count--) {
		buffer_info = &tx_ring->buffer_info[i];
		tx_desc = E1000_TX_DESC(*tx_ring, i);
		tx_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
		tx_desc->lower.data =
			cpu_to_le32(txd_lower | buffer_info->length);
		tx_desc->upper.data = cpu_to_le32(txd_upper);
		if(unlikely(++i == tx_ring->count)) i = 0;
	}

	tx_desc->lower.data |= cpu_to_le32(adapter->txd_cmd);

	/* Force memory writes to complete before letting h/w
	 * know there are new descriptors to fetch.  (Only
	 * applicable for weak-ordered memory model archs,
	 * such as IA-64). */
	wmb();

	tx_ring->next_to_use = i;
	E1000_WRITE_REG(&adapter->hw, TDT, i);
}

/**
 * 82547 workaround to avoid controller hang in half-duplex environment.
 * The workaround is to avoid queuing a large packet that would span
 * the internal Tx FIFO ring boundary by notifying the stack to resend
 * the packet at a later time.  This gives the Tx FIFO an opportunity to
 * flush all packets.  When that occurs, we reset the Tx FIFO pointers
 * to the beginning of the Tx FIFO.
 **/

#define E1000_FIFO_HDR			0x10
#define E1000_82547_PAD_LEN		0x3E0

static inline int
e1000_82547_fifo_workaround(struct e1000_adapter *adapter, struct sk_buff *skb)
{
	uint32_t fifo_space = adapter->tx_fifo_size - adapter->tx_fifo_head;
	uint32_t skb_fifo_len = skb->len + E1000_FIFO_HDR;

	E1000_ROUNDUP(skb_fifo_len, E1000_FIFO_HDR);

	if(adapter->link_duplex != HALF_DUPLEX)
		goto no_fifo_stall_required;

	if(atomic_read(&adapter->tx_fifo_stall))
		return 1;

	if(skb_fifo_len >= (E1000_82547_PAD_LEN + fifo_space)) {
		atomic_set(&adapter->tx_fifo_stall, 1);
		return 1;
	}

no_fifo_stall_required:
	adapter->tx_fifo_head += skb_fifo_len;
	if(adapter->tx_fifo_head >= adapter->tx_fifo_size)
		adapter->tx_fifo_head -= adapter->tx_fifo_size;
	return 0;
}

#define TXD_USE_COUNT(S, X) (((S) >> (X)) + 1 )
static int
e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
{
	struct e1000_adapter *adapter = netdev->priv;
	unsigned int first, max_per_txd = E1000_MAX_DATA_PER_TXD;
	unsigned int max_txd_pwr = E1000_MAX_TXD_PWR;
	unsigned int tx_flags = 0;
	unsigned int len = skb->len;
	unsigned long flags;
	unsigned int nr_frags = 0;
	unsigned int mss = 0;
	int count = 0;
#ifdef MAX_SKB_FRAGS
	unsigned int f;
	len -= skb->data_len;
#endif

	if(unlikely(skb->len <= 0)) {
		dev_kfree_skb_any(skb);
		return NETDEV_TX_OK;
	}

#ifdef NETIF_F_TSO
	mss = skb_shinfo(skb)->tso_size;
	/* The controller does a simple calculation to
	 * make sure there is enough room in the FIFO before
	 * initiating the DMA for each buffer.  The calc is:
	 * 4 = ceil(buffer len/mss).  To make sure we don't
	 * overrun the FIFO, adjust the max buffer len if mss
	 * drops. */
	if(mss) {
		max_per_txd = min(mss << 2, max_per_txd);
		max_txd_pwr = fls(max_per_txd) - 1;
	}

	if((mss) || (skb->ip_summed == CHECKSUM_HW))
		count++;
	count++;	/* for sentinel desc */
#else
	if(skb->ip_summed == CHECKSUM_HW)
		count++;
#endif
	count += TXD_USE_COUNT(len, max_txd_pwr);

	if(adapter->pcix_82544)
		count++;

#ifdef MAX_SKB_FRAGS
	nr_frags = skb_shinfo(skb)->nr_frags;
	for(f = 0; f < nr_frags; f++)
		count += TXD_USE_COUNT(skb_shinfo(skb)->frags[f].size,
				       max_txd_pwr);
	if(adapter->pcix_82544)
		count += nr_frags;
#endif

#ifdef NETIF_F_LLTX
	local_irq_save(flags);
	if (!spin_trylock(&adapter->tx_lock)) {
		/* Collision - tell upper layer to requeue */
		local_irq_restore(flags);
		return NETDEV_TX_LOCKED;
	}
#else
	spin_lock_irqsave(&adapter->tx_lock, flags);
#endif

	/* need: count + 2 desc gap to keep tail from touching
	 * head, otherwise try next time */
	if(unlikely(E1000_DESC_UNUSED(&adapter->tx_ring) < count + 2)) {
		netif_stop_queue(netdev);
		spin_unlock_irqrestore(&adapter->tx_lock, flags);
		return NETDEV_TX_BUSY;
	}

	if(unlikely(adapter->hw.mac_type == e1000_82547)) {
		if(unlikely(e1000_82547_fifo_workaround(adapter, skb))) {
			netif_stop_queue(netdev);
			mod_timer(&adapter->tx_fifo_stall_timer, jiffies);
			spin_unlock_irqrestore(&adapter->tx_lock, flags);
			return NETDEV_TX_BUSY;
		}
	}

#ifndef NETIF_F_LLTX
	spin_unlock_irqrestore(&adapter->tx_lock, flags);
#endif

#ifdef NETIF_F_HW_VLAN_TX
	if(unlikely(adapter->vlgrp && vlan_tx_tag_present(skb))) {
		tx_flags |= E1000_TX_FLAGS_VLAN;
		tx_flags |= (vlan_tx_tag_get(skb) << E1000_TX_FLAGS_VLAN_SHIFT);
	}
#endif

	first = adapter->tx_ring.next_to_use;
	
	if(likely(e1000_tso(adapter, skb)))
		tx_flags |= E1000_TX_FLAGS_TSO;
	else if(likely(e1000_tx_csum(adapter, skb)))
		tx_flags |= E1000_TX_FLAGS_CSUM;

	e1000_tx_queue(adapter,
		e1000_tx_map(adapter, skb, first, max_per_txd, nr_frags, mss),
		tx_flags);

	netdev->trans_start = jiffies;

#ifdef NETIF_F_LLTX
	/* Make sure there is space in the ring for the next send. */
	if(unlikely(E1000_DESC_UNUSED(&adapter->tx_ring) < MAX_SKB_FRAGS + 2))
		netif_stop_queue(netdev);

	spin_unlock_irqrestore(&adapter->tx_lock, flags);
#endif

	return NETDEV_TX_OK;
}

/**
 * e1000_tx_timeout - Respond to a Tx Hang
 * @netdev: network interface device structure
 **/

static void
e1000_tx_timeout(struct net_device *netdev)
{
	struct e1000_adapter *adapter = netdev->priv;

	/* Do the reset outside of interrupt context */
	schedule_work(&adapter->tx_timeout_task);
}

static void
e1000_tx_timeout_task(struct net_device *netdev)
{
	struct e1000_adapter *adapter = netdev->priv;

	e1000_down(adapter);
	e1000_up(adapter);
}

/**
 * e1000_get_stats - Get System Network Statistics
 * @netdev: network interface device structure
 *
 * Returns the address of the device statistics structure.
 * The statistics are actually updated from the timer callback.
 **/

static struct net_device_stats *
e1000_get_stats(struct net_device *netdev)
{
	struct e1000_adapter *adapter = netdev->priv;

	e1000_update_stats(adapter);
	return &adapter->net_stats;
}

/**
 * e1000_change_mtu - Change the Maximum Transfer Unit
 * @netdev: network interface device structure
 * @new_mtu: new value for maximum frame size
 *
 * Returns 0 on success, negative on failure
 **/

static int
e1000_change_mtu(struct net_device *netdev, int new_mtu)
{
	struct e1000_adapter *adapter = netdev->priv;
	int old_mtu = adapter->rx_buffer_len;
	int max_frame = new_mtu + ENET_HEADER_SIZE + ETHERNET_FCS_SIZE;

	if((max_frame < MINIMUM_ETHERNET_FRAME_SIZE) ||
		(max_frame > MAX_JUMBO_FRAME_SIZE)) {
			DPRINTK(PROBE, ERR, "Invalid MTU setting\n");
			return -EINVAL;
	}

	if(max_frame <= MAXIMUM_ETHERNET_FRAME_SIZE) {
		adapter->rx_buffer_len = E1000_RXBUFFER_2048;

	} else if(adapter->hw.mac_type < e1000_82543) {
		DPRINTK(PROBE, ERR, "Jumbo Frames not supported on 82542\n");
		return -EINVAL;

	} else if(max_frame <= E1000_RXBUFFER_4096) {
		adapter->rx_buffer_len = E1000_RXBUFFER_4096;

	} else if(max_frame <= E1000_RXBUFFER_8192) {
		adapter->rx_buffer_len = E1000_RXBUFFER_8192;

	} else {
		adapter->rx_buffer_len = E1000_RXBUFFER_16384;
	}

	if(old_mtu != adapter->rx_buffer_len && netif_running(netdev)) {
		e1000_down(adapter);
		e1000_up(adapter);
	}

	netdev->mtu = new_mtu;
	adapter->hw.max_frame_size = max_frame;

	return 0;
}

/**
 * e1000_update_stats - Update the board statistics counters
 * @adapter: board private structure
 **/

void
e1000_update_stats(struct e1000_adapter *adapter)
{
	struct e1000_hw *hw = &adapter->hw;
	unsigned long flags;
	uint16_t phy_tmp;

#define PHY_IDLE_ERROR_COUNT_MASK 0x00FF

	spin_lock_irqsave(&adapter->stats_lock, flags);

	/* these counters are modified from e1000_adjust_tbi_stats,
	 * called from the interrupt context, so they must only
	 * be written while holding adapter->stats_lock
	 */

	adapter->stats.crcerrs += E1000_READ_REG(hw, CRCERRS);
	adapter->stats.gprc += E1000_READ_REG(hw, GPRC);
	adapter->stats.gorcl += E1000_READ_REG(hw, GORCL);
	adapter->stats.gorch += E1000_READ_REG(hw, GORCH);
	adapter->stats.bprc += E1000_READ_REG(hw, BPRC);
	adapter->stats.mprc += E1000_READ_REG(hw, MPRC);
	adapter->stats.roc += E1000_READ_REG(hw, ROC);
	adapter->stats.prc64 += E1000_READ_REG(hw, PRC64);
	adapter->stats.prc127 += E1000_READ_REG(hw, PRC127);
	adapter->stats.prc255 += E1000_READ_REG(hw, PRC255);
	adapter->stats.prc511 += E1000_READ_REG(hw, PRC511);
	adapter->stats.prc1023 += E1000_READ_REG(hw, PRC1023);
	adapter->stats.prc1522 += E1000_READ_REG(hw, PRC1522);

	adapter->stats.symerrs += E1000_READ_REG(hw, SYMERRS);
	adapter->stats.mpc += E1000_READ_REG(hw, MPC);
	adapter->stats.scc += E1000_READ_REG(hw, SCC);
	adapter->stats.ecol += E1000_READ_REG(hw, ECOL);
	adapter->stats.mcc += E1000_READ_REG(hw, MCC);
	adapter->stats.latecol += E1000_READ_REG(hw, LATECOL);
	adapter->stats.dc += E1000_READ_REG(hw, DC);
	adapter->stats.sec += E1000_READ_REG(hw, SEC);
	adapter->stats.rlec += E1000_READ_REG(hw, RLEC);
	adapter->stats.xonrxc += E1000_READ_REG(hw, XONRXC);
	adapter->stats.xontxc += E1000_READ_REG(hw, XONTXC);
	adapter->stats.xoffrxc += E1000_READ_REG(hw, XOFFRXC);
	adapter->stats.xofftxc += E1000_READ_REG(hw, XOFFTXC);
	adapter->stats.fcruc += E1000_READ_REG(hw, FCRUC);
	adapter->stats.gptc += E1000_READ_REG(hw, GPTC);
	adapter->stats.gotcl += E1000_READ_REG(hw, GOTCL);
	adapter->stats.gotch += E1000_READ_REG(hw, GOTCH);
	adapter->stats.rnbc += E1000_READ_REG(hw, RNBC);
	adapter->stats.ruc += E1000_READ_REG(hw, RUC);
	adapter->stats.rfc += E1000_READ_REG(hw, RFC);
	adapter->stats.rjc += E1000_READ_REG(hw, RJC);
	adapter->stats.torl += E1000_READ_REG(hw, TORL);
	adapter->stats.torh += E1000_READ_REG(hw, TORH);
	adapter->stats.totl += E1000_READ_REG(hw, TOTL);
	adapter->stats.toth += E1000_READ_REG(hw, TOTH);
	adapter->stats.tpr += E1000_READ_REG(hw, TPR);
	adapter->stats.ptc64 += E1000_READ_REG(hw, PTC64);
	adapter->stats.ptc127 += E1000_READ_REG(hw, PTC127);
	adapter->stats.ptc255 += E1000_READ_REG(hw, PTC255);
	adapter->stats.ptc511 += E1000_READ_REG(hw, PTC511);
	adapter->stats.ptc1023 += E1000_READ_REG(hw, PTC1023);
	adapter->stats.ptc1522 += E1000_READ_REG(hw, PTC1522);
	adapter->stats.mptc += E1000_READ_REG(hw, MPTC);
	adapter->stats.bptc += E1000_READ_REG(hw, BPTC);

	/* used for adaptive IFS */

	hw->tx_packet_delta = E1000_READ_REG(hw, TPT);
	adapter->stats.tpt += hw->tx_packet_delta;
	hw->collision_delta = E1000_READ_REG(hw, COLC);
	adapter->stats.colc += hw->collision_delta;

	if(hw->mac_type >= e1000_82543) {
		adapter->stats.algnerrc += E1000_READ_REG(hw, ALGNERRC);
		adapter->stats.rxerrc += E1000_READ_REG(hw, RXERRC);
		adapter->stats.tncrs += E1000_READ_REG(hw, TNCRS);
		adapter->stats.cexterr += E1000_READ_REG(hw, CEXTERR);
		adapter->stats.tsctc += E1000_READ_REG(hw, TSCTC);
		adapter->stats.tsctfc += E1000_READ_REG(hw, TSCTFC);
	}

	/* Fill out the OS statistics structure */

	adapter->net_stats.rx_packets = adapter->sta