First Support on Ginger and OMAP TI

[linux-ginger.git] / drivers / staging / et131x / et1310_tx.c

/*
 * Agere Systems Inc.
 * 10/100/1000 Base-T Ethernet Driver for the ET1301 and ET131x series MACs
 *
 * Copyright © 2005 Agere Systems Inc.
 * All rights reserved.
 *   http://www.agere.com
 *
 *------------------------------------------------------------------------------
 *
 * et1310_tx.c - Routines used to perform data transmission.
 *
 *------------------------------------------------------------------------------
 *
 * SOFTWARE LICENSE
 *
 * This software is provided subject to the following terms and conditions,
 * which you should read carefully before using the software.  Using this
 * software indicates your acceptance of these terms and conditions.  If you do
 * not agree with these terms and conditions, do not use the software.
 *
 * Copyright © 2005 Agere Systems Inc.
 * All rights reserved.
 *
 * Redistribution and use in source or binary forms, with or without
 * modifications, are permitted provided that the following conditions are met:
 *
 * . Redistributions of source code must retain the above copyright notice, this
 *    list of conditions and the following Disclaimer as comments in the code as
 *    well as in the documentation and/or other materials provided with the
 *    distribution.
 *
 * . Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following Disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 *
 * . Neither the name of Agere Systems Inc. nor the names of the contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * Disclaimer
 *
 * THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
 * INCLUDING, BUT NOT LIMITED TO, INFRINGEMENT AND THE IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  ANY
 * USE, MODIFICATION OR DISTRIBUTION OF THIS SOFTWARE IS SOLELY AT THE USERS OWN
 * RISK. IN NO EVENT SHALL AGERE SYSTEMS INC. OR CONTRIBUTORS BE LIABLE FOR ANY
 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
 * ON ANY THEORY OF LIABILITY, INCLUDING, BUT NOT LIMITED TO, CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
 * DAMAGE.
 *
 */

#include "et131x_version.h"
#include "et131x_defs.h"

#include <linux/pci.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>

#include <linux/sched.h>
#include <linux/ptrace.h>
#include <linux/slab.h>
#include <linux/ctype.h>
#include <linux/string.h>
#include <linux/timer.h>
#include <linux/interrupt.h>
#include <linux/in.h>
#include <linux/delay.h>
#include <linux/io.h>
#include <linux/bitops.h>
#include <asm/system.h>

#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <linux/if_arp.h>
#include <linux/ioport.h>

#include "et1310_phy.h"
#include "et1310_pm.h"
#include "et1310_jagcore.h"

#include "et131x_adapter.h"
#include "et131x_initpci.h"
#include "et131x_isr.h"

#include "et1310_tx.h"


static void et131x_update_tcb_list(struct et131x_adapter *etdev);
static void et131x_check_send_wait_list(struct et131x_adapter *etdev);
static inline void et131x_free_send_packet(struct et131x_adapter *etdev,
                                           PMP_TCB pMpTcb);
static int et131x_send_packet(struct sk_buff *skb,
                              struct et131x_adapter *etdev);
static int nic_send_packet(struct et131x_adapter *etdev, PMP_TCB pMpTcb);

/**
 * et131x_tx_dma_memory_alloc
 * @adapter: pointer to our private adapter structure
 *
 * Returns 0 on success and errno on failure (as defined in errno.h).
 *
 * Allocates memory that will be visible both to the device and to the CPU.
 * The OS will pass us packets, pointers to which we will insert in the Tx
 * Descriptor queue. The device will read this queue to find the packets in
 * memory. The device will update the "status" in memory each time it xmits a
 * packet.
 */
int et131x_tx_dma_memory_alloc(struct et131x_adapter *adapter)
{
        int desc_size = 0;
        TX_RING_t *tx_ring = &adapter->TxRing;

        /* Allocate memory for the TCB's (Transmit Control Block) */
        adapter->TxRing.MpTcbMem = (MP_TCB *)kcalloc(NUM_TCB, sizeof(MP_TCB),
                                                      GFP_ATOMIC | GFP_DMA);
        if (!adapter->TxRing.MpTcbMem) {
                dev_err(&adapter->pdev->dev, "Cannot alloc memory for TCBs\n");
                return -ENOMEM;
        }

        /* Allocate enough memory for the Tx descriptor ring, and allocate
         * some extra so that the ring can be aligned on a 4k boundary.
         */
        desc_size = (sizeof(TX_DESC_ENTRY_t) * NUM_DESC_PER_RING_TX) + 4096 - 1;
        tx_ring->pTxDescRingVa =
            (PTX_DESC_ENTRY_t) pci_alloc_consistent(adapter->pdev, desc_size,
                                                    &tx_ring->pTxDescRingPa);
        if (!adapter->TxRing.pTxDescRingVa) {
                dev_err(&adapter->pdev->dev, "Cannot alloc memory for Tx Ring\n");
                return -ENOMEM;
        }

        /* Save physical address
         *
         * NOTE: pci_alloc_consistent(), used above to alloc DMA regions,
         * ALWAYS returns SAC (32-bit) addresses. If DAC (64-bit) addresses
         * are ever returned, make sure the high part is retrieved here before
         * storing the adjusted address.
         */
        tx_ring->pTxDescRingAdjustedPa = tx_ring->pTxDescRingPa;

        /* Align Tx Descriptor Ring on a 4k (0x1000) byte boundary */
        et131x_align_allocated_memory(adapter,
                                      &tx_ring->pTxDescRingAdjustedPa,
                                      &tx_ring->TxDescOffset, 0x0FFF);

        tx_ring->pTxDescRingVa += tx_ring->TxDescOffset;

        /* Allocate memory for the Tx status block */
        tx_ring->pTxStatusVa = pci_alloc_consistent(adapter->pdev,
                                                    sizeof(TX_STATUS_BLOCK_t),
                                                    &tx_ring->pTxStatusPa);
        if (!adapter->TxRing.pTxStatusPa) {
                dev_err(&adapter->pdev->dev,
                                  "Cannot alloc memory for Tx status block\n");
                return -ENOMEM;
        }

        /* Allocate memory for a dummy buffer */
        tx_ring->pTxDummyBlkVa = pci_alloc_consistent(adapter->pdev,
                                                      NIC_MIN_PACKET_SIZE,
                                                      &tx_ring->pTxDummyBlkPa);
        if (!adapter->TxRing.pTxDummyBlkPa) {
                dev_err(&adapter->pdev->dev,
                        "Cannot alloc memory for Tx dummy buffer\n");
                return -ENOMEM;
        }

        return 0;
}

/**
 * et131x_tx_dma_memory_free - Free all memory allocated within this module
 * @adapter: pointer to our private adapter structure
 *
 * Returns 0 on success and errno on failure (as defined in errno.h).
 */
void et131x_tx_dma_memory_free(struct et131x_adapter *adapter)
{
        int desc_size = 0;

        if (adapter->TxRing.pTxDescRingVa) {
                /* Free memory relating to Tx rings here */
                adapter->TxRing.pTxDescRingVa -= adapter->TxRing.TxDescOffset;

                desc_size =
                    (sizeof(TX_DESC_ENTRY_t) * NUM_DESC_PER_RING_TX) + 4096 - 1;

                pci_free_consistent(adapter->pdev,
                                    desc_size,
                                    adapter->TxRing.pTxDescRingVa,
                                    adapter->TxRing.pTxDescRingPa);

                adapter->TxRing.pTxDescRingVa = NULL;
        }

        /* Free memory for the Tx status block */
        if (adapter->TxRing.pTxStatusVa) {
                pci_free_consistent(adapter->pdev,
                                    sizeof(TX_STATUS_BLOCK_t),
                                    adapter->TxRing.pTxStatusVa,
                                    adapter->TxRing.pTxStatusPa);

                adapter->TxRing.pTxStatusVa = NULL;
        }

        /* Free memory for the dummy buffer */
        if (adapter->TxRing.pTxDummyBlkVa) {
                pci_free_consistent(adapter->pdev,
                                    NIC_MIN_PACKET_SIZE,
                                    adapter->TxRing.pTxDummyBlkVa,
                                    adapter->TxRing.pTxDummyBlkPa);

                adapter->TxRing.pTxDummyBlkVa = NULL;
        }

        /* Free the memory for MP_TCB structures */
        kfree(adapter->TxRing.MpTcbMem);
}

/**
 * ConfigTxDmaRegs - Set up the tx dma section of the JAGCore.
 * @etdev: pointer to our private adapter structure
 */
void ConfigTxDmaRegs(struct et131x_adapter *etdev)
{
        struct _TXDMA_t __iomem *txdma = &etdev->regs->txdma;

        /* Load the hardware with the start of the transmit descriptor ring. */
        writel((uint32_t) (etdev->TxRing.pTxDescRingAdjustedPa >> 32),
               &txdma->pr_base_hi);
        writel((uint32_t) etdev->TxRing.pTxDescRingAdjustedPa,
               &txdma->pr_base_lo);

        /* Initialise the transmit DMA engine */
        writel(NUM_DESC_PER_RING_TX - 1, &txdma->pr_num_des.value);

        /* Load the completion writeback physical address
         *
         * NOTE: pci_alloc_consistent(), used above to alloc DMA regions,
         * ALWAYS returns SAC (32-bit) addresses. If DAC (64-bit) addresses
         * are ever returned, make sure the high part is retrieved here before
         * storing the adjusted address.
         */
        writel(0, &txdma->dma_wb_base_hi);
        writel(etdev->TxRing.pTxStatusPa, &txdma->dma_wb_base_lo);

        memset(etdev->TxRing.pTxStatusVa, 0, sizeof(TX_STATUS_BLOCK_t));

        writel(0, &txdma->service_request);
        etdev->TxRing.txDmaReadyToSend = 0;
}

/**
 * et131x_tx_dma_disable - Stop of Tx_DMA on the ET1310
 * @etdev: pointer to our adapter structure
 */
void et131x_tx_dma_disable(struct et131x_adapter *etdev)
{
        /* Setup the tramsmit dma configuration register */
        writel(ET_TXDMA_CSR_HALT|ET_TXDMA_SNGL_EPKT,
                                        &etdev->regs->txdma.csr);
}

/**
 * et131x_tx_dma_enable - re-start of Tx_DMA on the ET1310.
 * @etdev: pointer to our adapter structure
 *
 * Mainly used after a return to the D0 (full-power) state from a lower state.
 */
void et131x_tx_dma_enable(struct et131x_adapter *etdev)
{
        u32 csr = ET_TXDMA_SNGL_EPKT;
        if (etdev->RegistryPhyLoopbk)
                /* TxDMA is disabled for loopback operation. */
                csr |= ET_TXDMA_CSR_HALT;
        else
                /* Setup the transmit dma configuration register for normal
                 * operation
                 */
                csr |= PARM_DMA_CACHE_DEF << ET_TXDMA_CACHE_SHIFT;
        writel(csr, &etdev->regs->txdma.csr);
}

/**
 * et131x_init_send - Initialize send data structures
 * @adapter: pointer to our private adapter structure
 */
void et131x_init_send(struct et131x_adapter *adapter)
{
        PMP_TCB pMpTcb;
        uint32_t TcbCount;
        TX_RING_t *tx_ring;

        /* Setup some convenience pointers */
        tx_ring = &adapter->TxRing;
        pMpTcb = adapter->TxRing.MpTcbMem;

        tx_ring->TCBReadyQueueHead = pMpTcb;

        /* Go through and set up each TCB */
        for (TcbCount = 0; TcbCount < NUM_TCB; TcbCount++) {
                memset(pMpTcb, 0, sizeof(MP_TCB));

                /* Set the link pointer in HW TCB to the next TCB in the
                 * chain.  If this is the last TCB in the chain, also set the
                 * tail pointer.
                 */
                if (TcbCount < NUM_TCB - 1) {
                        pMpTcb->Next = pMpTcb + 1;
                } else {
                        tx_ring->TCBReadyQueueTail = pMpTcb;
                        pMpTcb->Next = (PMP_TCB) NULL;
                }

                pMpTcb++;
        }

        /* Curr send queue should now be empty */
        tx_ring->CurrSendHead = (PMP_TCB) NULL;
        tx_ring->CurrSendTail = (PMP_TCB) NULL;

        INIT_LIST_HEAD(&adapter->TxRing.SendWaitQueue);
}

/**
 * et131x_send_packets - This function is called by the OS to send packets
 * @skb: the packet(s) to send
 * @netdev:device on which to TX the above packet(s)
 *
 * Return 0 in almost all cases; non-zero value in extreme hard failure only
 */
int et131x_send_packets(struct sk_buff *skb, struct net_device *netdev)
{
        int status = 0;
        struct et131x_adapter *etdev = NULL;

        etdev = netdev_priv(netdev);

        /* Send these packets
         *
         * NOTE: The Linux Tx entry point is only given one packet at a time
         * to Tx, so the PacketCount and it's array used makes no sense here
         */

        /* Queue is not empty or TCB is not available */
        if (!list_empty(&etdev->TxRing.SendWaitQueue) ||
            MP_TCB_RESOURCES_NOT_AVAILABLE(etdev)) {
                /* NOTE: If there's an error on send, no need to queue the
                 * packet under Linux; if we just send an error up to the
                 * netif layer, it will resend the skb to us.
                 */
                status = -ENOMEM;
        } else {
                /* We need to see if the link is up; if it's not, make the
                 * netif layer think we're good and drop the packet
                 */
                /*
                 * if( MP_SHOULD_FAIL_SEND( etdev ) ||
                 *  etdev->DriverNoPhyAccess )
                 */
                if (MP_SHOULD_FAIL_SEND(etdev) || etdev->DriverNoPhyAccess
                    || !netif_carrier_ok(netdev)) {
                        dev_kfree_skb_any(skb);
                        skb = NULL;

                        etdev->net_stats.tx_dropped++;
                } else {
                        status = et131x_send_packet(skb, etdev);

                        if (status == -ENOMEM) {

                                /* NOTE: If there's an error on send, no need
                                 * to queue the packet under Linux; if we just
                                 * send an error up to the netif layer, it
                                 * will resend the skb to us.
                                 */
                        } else if (status != 0) {
                                /* On any other error, make netif think we're
                                 * OK and drop the packet
                                 */
                                dev_kfree_skb_any(skb);
                                skb = NULL;
                                etdev->net_stats.tx_dropped++;
                        }
                }
        }
        return status;
}

/**
 * et131x_send_packet - Do the work to send a packet
 * @skb: the packet(s) to send
 * @etdev: a pointer to the device's private adapter structure
 *
 * Return 0 in almost all cases; non-zero value in extreme hard failure only.
 *
 * Assumption: Send spinlock has been acquired
 */
static int et131x_send_packet(struct sk_buff *skb,
                              struct et131x_adapter *etdev)
{
        int status = 0;
        PMP_TCB pMpTcb = NULL;
        uint16_t *shbufva;
        unsigned long flags;

        /* All packets must have at least a MAC address and a protocol type */
        if (skb->len < ETH_HLEN) {
                return -EIO;
        }

        /* Get a TCB for this packet */
        spin_lock_irqsave(&etdev->TCBReadyQLock, flags);

        pMpTcb = etdev->TxRing.TCBReadyQueueHead;

        if (pMpTcb == NULL) {
                spin_unlock_irqrestore(&etdev->TCBReadyQLock, flags);
                return -ENOMEM;
        }

        etdev->TxRing.TCBReadyQueueHead = pMpTcb->Next;

        if (etdev->TxRing.TCBReadyQueueHead == NULL)
                etdev->TxRing.TCBReadyQueueTail = NULL;

        spin_unlock_irqrestore(&etdev->TCBReadyQLock, flags);

        pMpTcb->PacketLength = skb->len;
        pMpTcb->Packet = skb;

        if ((skb->data != NULL) && ((skb->len - skb->data_len) >= 6)) {
                shbufva = (uint16_t *) skb->data;

                if ((shbufva[0] == 0xffff) &&
                    (shbufva[1] == 0xffff) && (shbufva[2] == 0xffff)) {
                        pMpTcb->Flags |= fMP_DEST_BROAD;
                } else if ((shbufva[0] & 0x3) == 0x0001) {
                        pMpTcb->Flags |=  fMP_DEST_MULTI;
                }
        }

        pMpTcb->Next = NULL;

        /* Call the NIC specific send handler. */
        if (status == 0)
                status = nic_send_packet(etdev, pMpTcb);

        if (status != 0) {
                spin_lock_irqsave(&etdev->TCBReadyQLock, flags);

                if (etdev->TxRing.TCBReadyQueueTail) {
                        etdev->TxRing.TCBReadyQueueTail->Next = pMpTcb;
                } else {
                        /* Apparently ready Q is empty. */
                        etdev->TxRing.TCBReadyQueueHead = pMpTcb;
                }

                etdev->TxRing.TCBReadyQueueTail = pMpTcb;
                spin_unlock_irqrestore(&etdev->TCBReadyQLock, flags);
                return status;
        }
        WARN_ON(etdev->TxRing.nBusySend > NUM_TCB);
        return 0;
}

/**
 * nic_send_packet - NIC specific send handler for version B silicon.
 * @etdev: pointer to our adapter
 * @pMpTcb: pointer to MP_TCB
 *
 * Returns 0 or errno.
 */
static int nic_send_packet(struct et131x_adapter *etdev, PMP_TCB pMpTcb)
{
        uint32_t loopIndex;
        TX_DESC_ENTRY_t CurDesc[24];
        uint32_t FragmentNumber = 0;
        uint32_t thiscopy, remainder;
        struct sk_buff *pPacket = pMpTcb->Packet;
        uint32_t FragListCount = skb_shinfo(pPacket)->nr_frags + 1;
        struct skb_frag_struct *pFragList = &skb_shinfo(pPacket)->frags[0];
        unsigned long flags;

        /* Part of the optimizations of this send routine restrict us to
         * sending 24 fragments at a pass.  In practice we should never see
         * more than 5 fragments.
         *
         * NOTE: The older version of this function (below) can handle any
         * number of fragments. If needed, we can call this function,
         * although it is less efficient.
         */
        if (FragListCount > 23) {
                return -EIO;
        }

        memset(CurDesc, 0, sizeof(TX_DESC_ENTRY_t) * (FragListCount + 1));

        for (loopIndex = 0; loopIndex < FragListCount; loopIndex++) {
                /* If there is something in this element, lets get a
                 * descriptor from the ring and get the necessary data
                 */
                if (loopIndex == 0) {
                        /* If the fragments are smaller than a standard MTU,
                         * then map them to a single descriptor in the Tx
                         * Desc ring. However, if they're larger, as is
                         * possible with support for jumbo packets, then
                         * split them each across 2 descriptors.
                         *
                         * This will work until we determine why the hardware
                         * doesn't seem to like large fragments.
                         */
                        if ((pPacket->len - pPacket->data_len) <= 1514) {
                                CurDesc[FragmentNumber].DataBufferPtrHigh = 0;
                                CurDesc[FragmentNumber].word2.bits.
                                    length_in_bytes =
                                    pPacket->len - pPacket->data_len;

                                /* NOTE: Here, the dma_addr_t returned from
                                 * pci_map_single() is implicitly cast as a
                                 * uint32_t. Although dma_addr_t can be
                                 * 64-bit, the address returned by
                                 * pci_map_single() is always 32-bit
                                 * addressable (as defined by the pci/dma
                                 * subsystem)
                                 */
                                CurDesc[FragmentNumber++].DataBufferPtrLow =
                                    pci_map_single(etdev->pdev,
                                                   pPacket->data,
                                                   pPacket->len -
                                                   pPacket->data_len,
                                                   PCI_DMA_TODEVICE);
                        } else {
                                CurDesc[FragmentNumber].DataBufferPtrHigh = 0;
                                CurDesc[FragmentNumber].word2.bits.
                                    length_in_bytes =
                                    ((pPacket->len - pPacket->data_len) / 2);

                                /* NOTE: Here, the dma_addr_t returned from
                                 * pci_map_single() is implicitly cast as a
                                 * uint32_t. Although dma_addr_t can be
                                 * 64-bit, the address returned by
                                 * pci_map_single() is always 32-bit
                                 * addressable (as defined by the pci/dma
                                 * subsystem)
                                 */
                                CurDesc[FragmentNumber++].DataBufferPtrLow =
                                    pci_map_single(etdev->pdev,
                                                   pPacket->data,
                                                   ((pPacket->len -
                                                     pPacket->data_len) / 2),
                                                   PCI_DMA_TODEVICE);
                                CurDesc[FragmentNumber].DataBufferPtrHigh = 0;

                                CurDesc[FragmentNumber].word2.bits.
                                    length_in_bytes =
                                    ((pPacket->len - pPacket->data_len) / 2);

                                /* NOTE: Here, the dma_addr_t returned from
                                 * pci_map_single() is implicitly cast as a
                                 * uint32_t. Although dma_addr_t can be
                                 * 64-bit, the address returned by
                                 * pci_map_single() is always 32-bit
                                 * addressable (as defined by the pci/dma
                                 * subsystem)
                                 */
                                CurDesc[FragmentNumber++].DataBufferPtrLow =
                                    pci_map_single(etdev->pdev,
                                                   pPacket->data +
                                                   ((pPacket->len -
                                                     pPacket->data_len) / 2),
                                                   ((pPacket->len -
                                                     pPacket->data_len) / 2),
                                                   PCI_DMA_TODEVICE);
                        }
                } else {
                        CurDesc[FragmentNumber].DataBufferPtrHigh = 0;
                        CurDesc[FragmentNumber].word2.bits.length_in_bytes =
                            pFragList[loopIndex - 1].size;

                        /* NOTE: Here, the dma_addr_t returned from
                         * pci_map_page() is implicitly cast as a uint32_t.
                         * Although dma_addr_t can be 64-bit, the address
                         * returned by pci_map_page() is always 32-bit
                         * addressable (as defined by the pci/dma subsystem)
                         */
                        CurDesc[FragmentNumber++].DataBufferPtrLow =
                            pci_map_page(etdev->pdev,
                                         pFragList[loopIndex - 1].page,
                                         pFragList[loopIndex - 1].page_offset,
                                         pFragList[loopIndex - 1].size,
                                         PCI_DMA_TODEVICE);
                }
        }

        if (FragmentNumber == 0)
                return -EIO;

        if (etdev->linkspeed == TRUEPHY_SPEED_1000MBPS) {
                if (++etdev->TxRing.TxPacketsSinceLastinterrupt ==
                    PARM_TX_NUM_BUFS_DEF) {
                        CurDesc[FragmentNumber - 1].word3.value = 0x5;
                        etdev->TxRing.TxPacketsSinceLastinterrupt = 0;
                } else {
                        CurDesc[FragmentNumber - 1].word3.value = 0x1;
                }
        } else {
                CurDesc[FragmentNumber - 1].word3.value = 0x5;
        }

        CurDesc[0].word3.bits.f = 1;

        pMpTcb->WrIndexStart = etdev->TxRing.txDmaReadyToSend;
        pMpTcb->PacketStaleCount = 0;

        spin_lock_irqsave(&etdev->SendHWLock, flags);

        thiscopy = NUM_DESC_PER_RING_TX -
                                INDEX10(etdev->TxRing.txDmaReadyToSend);

        if (thiscopy >= FragmentNumber) {
                remainder = 0;
                thiscopy = FragmentNumber;
        } else {
                remainder = FragmentNumber - thiscopy;
        }

        memcpy(etdev->TxRing.pTxDescRingVa +
               INDEX10(etdev->TxRing.txDmaReadyToSend), CurDesc,
               sizeof(TX_DESC_ENTRY_t) * thiscopy);

        add_10bit(&etdev->TxRing.txDmaReadyToSend, thiscopy);

        if (INDEX10(etdev->TxRing.txDmaReadyToSend)== 0 ||
            INDEX10(etdev->TxRing.txDmaReadyToSend) == NUM_DESC_PER_RING_TX) {
                etdev->TxRing.txDmaReadyToSend &= ~ET_DMA10_MASK;
                etdev->TxRing.txDmaReadyToSend ^= ET_DMA10_WRAP;
        }

        if (remainder) {
                memcpy(etdev->TxRing.pTxDescRingVa,
                       CurDesc + thiscopy,
                       sizeof(TX_DESC_ENTRY_t) * remainder);

                add_10bit(&etdev->TxRing.txDmaReadyToSend, remainder);
        }

        if (INDEX10(etdev->TxRing.txDmaReadyToSend) == 0) {
                if (etdev->TxRing.txDmaReadyToSend)
                        pMpTcb->WrIndex = NUM_DESC_PER_RING_TX - 1;
                else
                        pMpTcb->WrIndex= ET_DMA10_WRAP | (NUM_DESC_PER_RING_TX - 1);
        } else
                pMpTcb->WrIndex = etdev->TxRing.txDmaReadyToSend - 1;

        spin_lock(&etdev->TCBSendQLock);

        if (etdev->TxRing.CurrSendTail)
                etdev->TxRing.CurrSendTail->Next = pMpTcb;
        else
                etdev->TxRing.CurrSendHead = pMpTcb;

        etdev->TxRing.CurrSendTail = pMpTcb;

        WARN_ON(pMpTcb->Next != NULL);

        etdev->TxRing.nBusySend++;

        spin_unlock(&etdev->TCBSendQLock);

        /* Write the new write pointer back to the device. */
        writel(etdev->TxRing.txDmaReadyToSend,
               &etdev->regs->txdma.service_request);

        /* For Gig only, we use Tx Interrupt coalescing.  Enable the software
         * timer to wake us up if this packet isn't followed by N more.
         */
        if (etdev->linkspeed == TRUEPHY_SPEED_1000MBPS) {
                writel(PARM_TX_TIME_INT_DEF * NANO_IN_A_MICRO,
                       &etdev->regs->global.watchdog_timer);
        }
        spin_unlock_irqrestore(&etdev->SendHWLock, flags);

        return 0;
}


/**
 * et131x_free_send_packet - Recycle a MP_TCB, complete the packet if necessary
 * @etdev: pointer to our adapter
 * @pMpTcb: pointer to MP_TCB
 *
 * Assumption - Send spinlock has been acquired
 */
inline void et131x_free_send_packet(struct et131x_adapter *etdev,
                                                        PMP_TCB pMpTcb)
{
        unsigned long flags;
        TX_DESC_ENTRY_t *desc = NULL;
        struct net_device_stats *stats = &etdev->net_stats;

        if (pMpTcb->Flags & fMP_DEST_BROAD)
                atomic_inc(&etdev->Stats.brdcstxmt);
        else if (pMpTcb->Flags & fMP_DEST_MULTI)
                atomic_inc(&etdev->Stats.multixmt);
        else
                atomic_inc(&etdev->Stats.unixmt);

        if (pMpTcb->Packet) {
                stats->tx_bytes += pMpTcb->Packet->len;

                /* Iterate through the TX descriptors on the ring
                 * corresponding to this packet and umap the fragments
                 * they point to
                 */
                do {
                        desc =
                            (TX_DESC_ENTRY_t *) (etdev->TxRing.pTxDescRingVa +
                                INDEX10(pMpTcb->WrIndexStart));

                        pci_unmap_single(etdev->pdev,
                                         desc->DataBufferPtrLow,
                                         desc->word2.value, PCI_DMA_TODEVICE);

                        add_10bit(&pMpTcb->WrIndexStart, 1);
                        if (INDEX10(pMpTcb->WrIndexStart) >=
                            NUM_DESC_PER_RING_TX) {
                                pMpTcb->WrIndexStart &= ~ET_DMA10_MASK;
                                pMpTcb->WrIndexStart ^= ET_DMA10_WRAP;
                        }
                } while (desc != (etdev->TxRing.pTxDescRingVa +
                                INDEX10(pMpTcb->WrIndex)));

                dev_kfree_skb_any(pMpTcb->Packet);
        }

        memset(pMpTcb, 0, sizeof(MP_TCB));

        /* Add the TCB to the Ready Q */
        spin_lock_irqsave(&etdev->TCBReadyQLock, flags);

        etdev->Stats.opackets++;

        if (etdev->TxRing.TCBReadyQueueTail) {
                etdev->TxRing.TCBReadyQueueTail->Next = pMpTcb;
        } else {
                /* Apparently ready Q is empty. */
                etdev->TxRing.TCBReadyQueueHead = pMpTcb;
        }

        etdev->TxRing.TCBReadyQueueTail = pMpTcb;

        spin_unlock_irqrestore(&etdev->TCBReadyQLock, flags);
        WARN_ON(etdev->TxRing.nBusySend < 0);
}

/**
 * et131x_free_busy_send_packets - Free and complete the stopped active sends
 * @etdev: pointer to our adapter
 *
 * Assumption - Send spinlock has been acquired
 */
void et131x_free_busy_send_packets(struct et131x_adapter *etdev)
{
        PMP_TCB pMpTcb;
        struct list_head *entry;
        unsigned long flags;
        uint32_t FreeCounter = 0;

        while (!list_empty(&etdev->TxRing.SendWaitQueue)) {
                spin_lock_irqsave(&etdev->SendWaitLock, flags);

                etdev->TxRing.nWaitSend--;
                spin_unlock_irqrestore(&etdev->SendWaitLock, flags);

                entry = etdev->TxRing.SendWaitQueue.next;
        }

        etdev->TxRing.nWaitSend = 0;

        /* Any packets being sent? Check the first TCB on the send list */
        spin_lock_irqsave(&etdev->TCBSendQLock, flags);

        pMpTcb = etdev->TxRing.CurrSendHead;

        while ((pMpTcb != NULL) && (FreeCounter < NUM_TCB)) {
                PMP_TCB pNext = pMpTcb->Next;

                etdev->TxRing.CurrSendHead = pNext;

                if (pNext == NULL)
                        etdev->TxRing.CurrSendTail = NULL;

                etdev->TxRing.nBusySend--;

                spin_unlock_irqrestore(&etdev->TCBSendQLock, flags);

                FreeCounter++;
                et131x_free_send_packet(etdev, pMpTcb);

                spin_lock_irqsave(&etdev->TCBSendQLock, flags);

                pMpTcb = etdev->TxRing.CurrSendHead;
        }

        WARN_ON(FreeCounter == NUM_TCB);

        spin_unlock_irqrestore(&etdev->TCBSendQLock, flags);

        etdev->TxRing.nBusySend = 0;
}

/**
 * et131x_handle_send_interrupt - Interrupt handler for sending processing
 * @etdev: pointer to our adapter
 *
 * Re-claim the send resources, complete sends and get more to send from
 * the send wait queue.
 *
 * Assumption - Send spinlock has been acquired
 */
void et131x_handle_send_interrupt(struct et131x_adapter *etdev)
{
        /* Mark as completed any packets which have been sent by the device. */
        et131x_update_tcb_list(etdev);

        /* If we queued any transmits because we didn't have any TCBs earlier,
         * dequeue and send those packets now, as long as we have free TCBs.
         */
        et131x_check_send_wait_list(etdev);
}

/**
 * et131x_update_tcb_list - Helper routine for Send Interrupt handler
 * @etdev: pointer to our adapter
 *
 * Re-claims the send resources and completes sends.  Can also be called as
 * part of the NIC send routine when the "ServiceComplete" indication has
 * wrapped.
 */
static void et131x_update_tcb_list(struct et131x_adapter *etdev)
{
        unsigned long flags;
        u32 ServiceComplete;
        PMP_TCB pMpTcb;
        u32 index;

        ServiceComplete = readl(&etdev->regs->txdma.NewServiceComplete);
        index = INDEX10(ServiceComplete);

        /* Has the ring wrapped?  Process any descriptors that do not have
         * the same "wrap" indicator as the current completion indicator
         */
        spin_lock_irqsave(&etdev->TCBSendQLock, flags);

        pMpTcb = etdev->TxRing.CurrSendHead;

        while (pMpTcb &&
               ((ServiceComplete ^ pMpTcb->WrIndex) & ET_DMA10_WRAP) &&
               index < INDEX10(pMpTcb->WrIndex)) {
                etdev->TxRing.nBusySend--;
                etdev->TxRing.CurrSendHead = pMpTcb->Next;
                if (pMpTcb->Next == NULL)
                        etdev->TxRing.CurrSendTail = NULL;

                spin_unlock_irqrestore(&etdev->TCBSendQLock, flags);
                et131x_free_send_packet(etdev, pMpTcb);
                spin_lock_irqsave(&etdev->TCBSendQLock, flags);

                /* Goto the next packet */
                pMpTcb = etdev->TxRing.CurrSendHead;
        }
        while (pMpTcb &&
               !((ServiceComplete ^ pMpTcb->WrIndex) & ET_DMA10_WRAP)
               && index > (pMpTcb->WrIndex & ET_DMA10_MASK)) {
                etdev->TxRing.nBusySend--;
                etdev->TxRing.CurrSendHead = pMpTcb->Next;
                if (pMpTcb->Next == NULL)
                        etdev->TxRing.CurrSendTail = NULL;

                spin_unlock_irqrestore(&etdev->TCBSendQLock, flags);
                et131x_free_send_packet(etdev, pMpTcb);
                spin_lock_irqsave(&etdev->TCBSendQLock, flags);

                /* Goto the next packet */
                pMpTcb = etdev->TxRing.CurrSendHead;
        }

        /* Wake up the queue when we hit a low-water mark */
        if (etdev->TxRing.nBusySend <= (NUM_TCB / 3))
                netif_wake_queue(etdev->netdev);

        spin_unlock_irqrestore(&etdev->TCBSendQLock, flags);
}

/**
 * et131x_check_send_wait_list - Helper routine for the interrupt handler
 * @etdev: pointer to our adapter
 *
 * Takes packets from the send wait queue and posts them to the device (if
 * room available).
 */
static void et131x_check_send_wait_list(struct et131x_adapter *etdev)
{
        unsigned long flags;

        spin_lock_irqsave(&etdev->SendWaitLock, flags);

        while (!list_empty(&etdev->TxRing.SendWaitQueue) &&
                                MP_TCB_RESOURCES_AVAILABLE(etdev)) {
                struct list_head *entry;

                entry = etdev->TxRing.SendWaitQueue.next;

                etdev->TxRing.nWaitSend--;
        }

        spin_unlock_irqrestore(&etdev->SendWaitLock, flags);
}