e1000_main.c

来自「优龙2410linux2.6.8内核源代码」· C语言 代码 · 共 2,331 行 · 第 1/5 页

			adapter->smartspeed = 0;
		}
	} else {
		if(netif_carrier_ok(netdev)) {
			adapter->link_speed = 0;
			adapter->link_duplex = 0;
			DPRINTK(LINK, INFO, "NIC Link is Down\n");
			netif_carrier_off(netdev);
			netif_stop_queue(netdev);
			mod_timer(&adapter->phy_info_timer, jiffies + 2 * HZ);
		}

		e1000_smartspeed(adapter);
	}

	e1000_update_stats(adapter);

	adapter->hw.tx_packet_delta = adapter->stats.tpt - adapter->tpt_old;
	adapter->tpt_old = adapter->stats.tpt;
	adapter->hw.collision_delta = adapter->stats.colc - adapter->colc_old;
	adapter->colc_old = adapter->stats.colc;
	
	adapter->gorcl = adapter->stats.gorcl - adapter->gorcl_old;
	adapter->gorcl_old = adapter->stats.gorcl;
	adapter->gotcl = adapter->stats.gotcl - adapter->gotcl_old;
	adapter->gotcl_old = adapter->stats.gotcl;

	e1000_update_adaptive(&adapter->hw);

	if(!netif_carrier_ok(netdev)) {
		if(E1000_DESC_UNUSED(txdr) + 1 < txdr->count) {
			/* We've lost link, so the controller stops DMA,
			 * but we've got queued Tx work that's never going
			 * to get done, so reset controller to flush Tx.
			 * (Do the reset outside of interrupt context). */
			schedule_work(&adapter->tx_timeout_task);
		}
	}

	/* Dynamic mode for Interrupt Throttle Rate (ITR) */
	if(adapter->hw.mac_type >= e1000_82540 && adapter->itr == 1) {
		/* Symmetric Tx/Rx gets a reduced ITR=2000; Total
		 * asymmetrical Tx or Rx gets ITR=8000; everyone
		 * else is between 2000-8000. */
		uint32_t goc = (adapter->gotcl + adapter->gorcl) / 10000;
		uint32_t dif = (adapter->gotcl > adapter->gorcl ? 
			adapter->gotcl - adapter->gorcl :
			adapter->gorcl - adapter->gotcl) / 10000;
		uint32_t itr = goc > 0 ? (dif * 6000 / goc + 2000) : 8000;
		E1000_WRITE_REG(&adapter->hw, ITR, 1000000000 / (itr * 256));
	}

	/* Cause software interrupt to ensure rx ring is cleaned */
	E1000_WRITE_REG(&adapter->hw, ICS, E1000_ICS_RXDMT0);

	/* Early detection of hung controller */
	i = txdr->next_to_clean;
	if(txdr->buffer_info[i].dma &&
	   time_after(jiffies, txdr->buffer_info[i].time_stamp + HZ) &&
	   !(E1000_READ_REG(&adapter->hw, STATUS) & E1000_STATUS_TXOFF))
		netif_stop_queue(netdev);

	/* Reset the timer */
	mod_timer(&adapter->watchdog_timer, jiffies + 2 * HZ);
}

#define E1000_TX_FLAGS_CSUM		0x00000001
#define E1000_TX_FLAGS_VLAN		0x00000002
#define E1000_TX_FLAGS_TSO		0x00000004
#define E1000_TX_FLAGS_VLAN_MASK	0xffff0000
#define E1000_TX_FLAGS_VLAN_SHIFT	16

static inline boolean_t
e1000_tso(struct e1000_adapter *adapter, struct sk_buff *skb)
{
#ifdef NETIF_F_TSO
	struct e1000_context_desc *context_desc;
	unsigned int i;
	uint8_t ipcss, ipcso, tucss, tucso, hdr_len;
	uint16_t ipcse, tucse, mss;

	if(skb_shinfo(skb)->tso_size) {
		hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2));
		mss = skb_shinfo(skb)->tso_size;
		skb->nh.iph->tot_len = 0;
		skb->nh.iph->check = 0;
		skb->h.th->check = ~csum_tcpudp_magic(skb->nh.iph->saddr,
		                                      skb->nh.iph->daddr,
		                                      0,
		                                      IPPROTO_TCP,
		                                      0);
		ipcss = skb->nh.raw - skb->data;
		ipcso = (void *)&(skb->nh.iph->check) - (void *)skb->data;
		ipcse = skb->h.raw - skb->data - 1;
		tucss = skb->h.raw - skb->data;
		tucso = (void *)&(skb->h.th->check) - (void *)skb->data;
		tucse = 0;

		i = adapter->tx_ring.next_to_use;
		context_desc = E1000_CONTEXT_DESC(adapter->tx_ring, i);

		context_desc->lower_setup.ip_fields.ipcss  = ipcss;
		context_desc->lower_setup.ip_fields.ipcso  = ipcso;
		context_desc->lower_setup.ip_fields.ipcse  = cpu_to_le16(ipcse);
		context_desc->upper_setup.tcp_fields.tucss = tucss;
		context_desc->upper_setup.tcp_fields.tucso = tucso;
		context_desc->upper_setup.tcp_fields.tucse = cpu_to_le16(tucse);
		context_desc->tcp_seg_setup.fields.mss     = cpu_to_le16(mss);
		context_desc->tcp_seg_setup.fields.hdr_len = hdr_len;
		context_desc->cmd_and_length = cpu_to_le32(
			E1000_TXD_CMD_DEXT | E1000_TXD_CMD_TSE |
			E1000_TXD_CMD_IP | E1000_TXD_CMD_TCP |
			(skb->len - (hdr_len)));

		if(++i == adapter->tx_ring.count) i = 0;
		adapter->tx_ring.next_to_use = i;

		return TRUE;
	}
#endif

	return FALSE;
}

static inline boolean_t
e1000_tx_csum(struct e1000_adapter *adapter, struct sk_buff *skb)
{
	struct e1000_context_desc *context_desc;
	unsigned int i;
	uint8_t css, cso;

	if(skb->ip_summed == CHECKSUM_HW) {
		css = skb->h.raw - skb->data;
		cso = (skb->h.raw + skb->csum) - skb->data;

		i = adapter->tx_ring.next_to_use;
		context_desc = E1000_CONTEXT_DESC(adapter->tx_ring, i);

		context_desc->upper_setup.tcp_fields.tucss = css;
		context_desc->upper_setup.tcp_fields.tucso = cso;
		context_desc->upper_setup.tcp_fields.tucse = 0;
		context_desc->tcp_seg_setup.data = 0;
		context_desc->cmd_and_length = cpu_to_le32(E1000_TXD_CMD_DEXT);

		if(++i == adapter->tx_ring.count) i = 0;
		adapter->tx_ring.next_to_use = i;

		return TRUE;
	}

	return FALSE;
}

#define E1000_MAX_TXD_PWR	12
#define E1000_MAX_DATA_PER_TXD	(1<<E1000_MAX_TXD_PWR)

static inline int
e1000_tx_map(struct e1000_adapter *adapter, struct sk_buff *skb,
	unsigned int first, unsigned int max_per_txd,
	unsigned int nr_frags, unsigned int mss)
{
	struct e1000_desc_ring *tx_ring = &adapter->tx_ring;
	struct e1000_buffer *buffer_info;
	unsigned int len = skb->len;
	unsigned int offset = 0, size, count = 0, i;
	unsigned int f;
	len -= skb->data_len;


	i = tx_ring->next_to_use;

	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(mss && !nr_frags && size == len && size > 8)
			size -= 4;
#endif
		/* Workaround for potential 82544 hang in PCI-X.  Avoid
		 * terminating buffers within evenly-aligned dwords. */
		if(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(++i == tx_ring->count) i = 0;
	}

	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(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(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(++i == tx_ring->count) i = 0;
		}
	}
	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(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(tx_flags & E1000_TX_FLAGS_CSUM) {
		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D;
		txd_upper |= E1000_TXD_POPTS_TXSM << 8;
	}

	if(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(++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 long flags;
	unsigned int len = skb->len;
	int count = 0;
	unsigned int mss = 0;
	unsigned int nr_frags = 0;
	unsigned int f;
	nr_frags = skb_shinfo(skb)->nr_frags;
	len -= skb->data_len;
	if(skb->len <= 0) {
		dev_kfree_skb_any(skb);
		return 0;
	}

#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++;

	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;
	
	spin_lock_irqsave(&adapter->tx_lock, flags);
	/* need: count +  2 desc gap to keep tail from touching 
	 * head, otherwise try next time */
	if(E1000_DESC_UNUSED(&adapter->tx_ring) < count + 2 ) {
		netif_stop_queue(netdev);
		spin_unlock_irqrestore(&adapter->tx_lock, flags);
		return 1;
	}
	spin_unlock_irqrestore(&adapter->tx_lock, flags);

	if(adapter->hw.mac_type == e1000_82547) {
		if(e1000_82547_fifo_workaround(adapter, skb)) {
			netif_stop_queue(netdev);
			mod_timer(&adapter->tx_fifo_stall_timer, jiffies);
			return 1;
		}
	}

	if(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);
	}

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

	return 0;
}

/**
 * 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;

	netif_device_detach(netdev);
	e1000_down(adapter);
	e1000_up(adapter);
	netif_device_attach(netdev);
}

/**
 * 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;
}