/* * 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 * *------------------------------------------------------------------------------ * * et131x_initpci.c - Routines and data used to register the driver with the * PCI (and PCI Express) subsystem, as well as basic driver * init and startup. * *------------------------------------------------------------------------------ * * 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 #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "et1310_phy.h" #include "et1310_pm.h" #include "et1310_jagcore.h" #include "et131x_adapter.h" #include "et131x_netdev.h" #include "et131x_config.h" #include "et131x_isr.h" #include "et1310_address_map.h" #include "et1310_tx.h" #include "et1310_rx.h" #include "et1310_mac.h" #include "et1310_eeprom.h" /* Defines for Parameter Default/Min/Max vaules */ #define PARM_SPEED_DUPLEX_MIN 0 #define PARM_SPEED_DUPLEX_MAX 5 /* Module parameter for disabling NMI * et131x_nmi_disable : * Disable NMI (0-2) [0] * 0 : * 1 : * 2 : */ static u32 et131x_nmi_disable; /* 0-2 */ module_param(et131x_nmi_disable, uint, 0); MODULE_PARM_DESC(et131x_nmi_disable, "Disable NMI (0-2) [0]"); /* Module parameter for manual speed setting * Set Link speed and dublex manually (0-5) [0] * 1 : 10Mb Half-Duplex * 2 : 10Mb Full-Duplex * 3 : 100Mb Half-Duplex * 4 : 100Mb Full-Duplex * 5 : 1000Mb Full-Duplex * 0 : Auto Speed Auto Duplex // default */ static u32 et131x_speed_set; module_param(et131x_speed_set, uint, 0); MODULE_PARM_DESC(et131x_speed_set, "Set Link speed and dublex manually (0-5) [0] \n 1 : 10Mb Half-Duplex \n 2 : 10Mb Full-Duplex \n 3 : 100Mb Half-Duplex \n 4 : 100Mb Full-Duplex \n 5 : 1000Mb Full-Duplex \n 0 : Auto Speed Auto Dublex"); /** * et131x_find_adapter - Find the adapter and get all the assigned resources * @adapter: pointer to our private adapter structure * * Returns 0 on success, errno on failure (as defined in errno.h) */ int et131x_find_adapter(struct et131x_adapter *adapter, struct pci_dev *pdev) { int result; uint8_t eepromStat; uint8_t maxPayload = 0; uint8_t read_size_reg; u8 rev; /* Allow disabling of Non-Maskable Interrupts in I/O space, to * support validation. */ if (adapter->RegistryNMIDisable) { uint8_t RegisterVal; RegisterVal = inb(ET1310_NMI_DISABLE); RegisterVal &= 0xf3; if (adapter->RegistryNMIDisable == 2) RegisterVal |= 0xc; outb(ET1310_NMI_DISABLE, RegisterVal); } /* We first need to check the EEPROM Status code located at offset * 0xB2 of config space */ result = pci_read_config_byte(pdev, ET1310_PCI_EEPROM_STATUS, &eepromStat); /* THIS IS A WORKAROUND: * I need to call this function twice to get my card in a * LG M1 Express Dual running. I tried also a msleep before this * function, because I thougth there could be some time condidions * but it didn't work. Call the whole function twice also work. */ result = pci_read_config_byte(pdev, ET1310_PCI_EEPROM_STATUS, &eepromStat); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not read PCI config space for " "EEPROM Status\n"); return -EIO; } /* Determine if the error(s) we care about are present. If they are * present, we need to fail. */ if (eepromStat & 0x4C) { result = pci_read_config_byte(pdev, PCI_REVISION_ID, &rev); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not read PCI config space for " "Revision ID\n"); return -EIO; } else if (rev == 0x01) { int32_t nLoop; uint8_t temp[4] = { 0xFE, 0x13, 0x10, 0xFF }; /* Re-write the first 4 bytes if we have an eeprom * present and the revision id is 1, this fixes the * corruption seen with 1310 B Silicon */ for (nLoop = 0; nLoop < 3; nLoop++) { EepromWriteByte(adapter, nLoop, temp[nLoop]); } } dev_err(&pdev->dev, "Fatal EEPROM Status Error - 0x%04x\n", eepromStat); /* This error could mean that there was an error reading the * eeprom or that the eeprom doesn't exist. We will treat * each case the same and not try to gather additional * information that normally would come from the eeprom, like * MAC Address */ adapter->has_eeprom = 0; return -EIO; } else adapter->has_eeprom = 1; /* Read the EEPROM for information regarding LED behavior. Refer to * ET1310_phy.c, et131x_xcvr_init(), for its use. */ EepromReadByte(adapter, 0x70, &adapter->eepromData[0]); EepromReadByte(adapter, 0x71, &adapter->eepromData[1]); if (adapter->eepromData[0] != 0xcd) /* Disable all optional features */ adapter->eepromData[1] = 0x00; /* Let's set up the PORT LOGIC Register. First we need to know what * the max_payload_size is */ result = pci_read_config_byte(pdev, ET1310_PCI_MAX_PYLD, &maxPayload); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not read PCI config space for Max Payload Size\n"); return -EIO; } /* Program the Ack/Nak latency and replay timers */ maxPayload &= 0x07; /* Only the lower 3 bits are valid */ if (maxPayload < 2) { const uint16_t AckNak[2] = { 0x76, 0xD0 }; const uint16_t Replay[2] = { 0x1E0, 0x2ED }; result = pci_write_config_word(pdev, ET1310_PCI_ACK_NACK, AckNak[maxPayload]); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not write PCI config space for ACK/NAK\n"); return -EIO; } result = pci_write_config_word(pdev, ET1310_PCI_REPLAY, Replay[maxPayload]); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not write PCI config space for Replay Timer\n"); return -EIO; } } /* l0s and l1 latency timers. We are using default values. * Representing 001 for L0s and 010 for L1 */ result = pci_write_config_byte(pdev, ET1310_PCI_L0L1LATENCY, 0x11); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not write PCI config space for Latency Timers\n"); return -EIO; } /* Change the max read size to 2k */ result = pci_read_config_byte(pdev, 0x51, &read_size_reg); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not read PCI config space for Max read size\n"); return -EIO; } read_size_reg &= 0x8f; read_size_reg |= 0x40; result = pci_write_config_byte(pdev, 0x51, read_size_reg); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, "Could not write PCI config space for Max read size\n"); return -EIO; } /* Get MAC address from config space if an eeprom exists, otherwise * the MAC address there will not be valid */ if (adapter->has_eeprom) { int i; for (i = 0; i < ETH_ALEN; i++) { result = pci_read_config_byte( pdev, ET1310_PCI_MAC_ADDRESS + i, adapter->PermanentAddress + i); if (result != PCIBIOS_SUCCESSFUL) { dev_err(&pdev->dev, ";Could not read PCI config space for MAC address\n"); return -EIO; } } } return 0; } /** * et131x_error_timer_handler * @data: timer-specific variable; here a pointer to our adapter structure * * The routine called when the error timer expires, to track the number of * recurring errors. */ void et131x_error_timer_handler(unsigned long data) { struct et131x_adapter *etdev = (struct et131x_adapter *) data; u32 pm_csr; pm_csr = readl(&etdev->regs->global.pm_csr); if ((pm_csr & ET_PM_PHY_SW_COMA) == 0) UpdateMacStatHostCounters(etdev); else dev_err(&etdev->pdev->dev, "No interrupts, in PHY coma, pm_csr = 0x%x\n", pm_csr); if (!etdev->Bmsr.bits.link_status && etdev->RegistryPhyComa && etdev->PoMgmt.TransPhyComaModeOnBoot < 11) { etdev->PoMgmt.TransPhyComaModeOnBoot++; } if (etdev->PoMgmt.TransPhyComaModeOnBoot == 10) { if (!etdev->Bmsr.bits.link_status && etdev->RegistryPhyComa) { if ((pm_csr & ET_PM_PHY_SW_COMA) == 0) { /* NOTE - This was originally a 'sync with * interrupt'. How to do that under Linux? */ et131x_enable_interrupts(etdev); EnablePhyComa(etdev); } } } /* This is a periodic timer, so reschedule */ mod_timer(&etdev->ErrorTimer, jiffies + TX_ERROR_PERIOD * HZ / 1000); } /** * et131x_link_detection_handler * * Timer function for link up at driver load time */ void et131x_link_detection_handler(unsigned long data) { struct et131x_adapter *etdev = (struct et131x_adapter *) data; unsigned long flags; if (etdev->MediaState == 0) { spin_lock_irqsave(&etdev->Lock, flags); etdev->MediaState = NETIF_STATUS_MEDIA_DISCONNECT; etdev->Flags &= ~fMP_ADAPTER_LINK_DETECTION; spin_unlock_irqrestore(&etdev->Lock, flags); netif_carrier_off(etdev->netdev); } } /** * et131x_configure_global_regs - configure JAGCore global regs * @etdev: pointer to our adapter structure * * Used to configure the global registers on the JAGCore */ void ConfigGlobalRegs(struct et131x_adapter *etdev) { struct _GLOBAL_t __iomem *regs = &etdev->regs->global; if (etdev->RegistryPhyLoopbk == false) { if (etdev->RegistryJumboPacket < 2048) { /* Tx / RxDMA and Tx/Rx MAC interfaces have a 1k word * block of RAM that the driver can split between Tx * and Rx as it desires. Our default is to split it * 50/50: */ writel(0, ®s->rxq_start_addr); writel(PARM_RX_MEM_END_DEF, ®s->rxq_end_addr); writel(PARM_RX_MEM_END_DEF + 1, ®s->txq_start_addr); writel(INTERNAL_MEM_SIZE - 1, ®s->txq_end_addr); } else if (etdev->RegistryJumboPacket < 8192) { /* For jumbo packets > 2k but < 8k, split 50-50. */ writel(0, ®s->rxq_start_addr); writel(INTERNAL_MEM_RX_OFFSET, ®s->rxq_end_addr); writel(INTERNAL_MEM_RX_OFFSET + 1, ®s->txq_start_addr); writel(INTERNAL_MEM_SIZE - 1, ®s->txq_end_addr); } else { /* 9216 is the only packet size greater than 8k that * is available. The Tx buffer has to be big enough * for one whole packet on the Tx side. We'll make * the Tx 9408, and give the rest to Rx */ writel(0x0000, ®s->rxq_start_addr); writel(0x01b3, ®s->rxq_end_addr); writel(0x01b4, ®s->txq_start_addr); writel(INTERNAL_MEM_SIZE - 1,®s->txq_end_addr); } /* Initialize the loopback register. Disable all loopbacks. */ writel(0, ®s->loopback); } else { /* For PHY Line loopback, the memory is configured as if Tx * and Rx both have all the memory. This is because the * RxMAC will write data into the space, and the TxMAC will * read it out. */ writel(0, ®s->rxq_start_addr); writel(INTERNAL_MEM_SIZE - 1, ®s->rxq_end_addr); writel(0, ®s->txq_start_addr); writel(INTERNAL_MEM_SIZE - 1, ®s->txq_end_addr); /* Initialize the loopback register (MAC loopback). */ writel(ET_LOOP_MAC, ®s->loopback); } /* MSI Register */ writel(0, ®s->msi_config); /* By default, disable the watchdog timer. It will be enabled when * a packet is queued. */ writel(0, ®s->watchdog_timer); } /** * et131x_adapter_setup - Set the adapter up as per cassini+ documentation * @adapter: pointer to our private adapter structure * * Returns 0 on success, errno on failure (as defined in errno.h) */ int et131x_adapter_setup(struct et131x_adapter *etdev) { int status = 0; /* Configure the JAGCore */ ConfigGlobalRegs(etdev); ConfigMACRegs1(etdev); /* Configure the MMC registers */ /* All we need to do is initialize the Memory Control Register */ writel(ET_MMC_ENABLE, &etdev->regs->mmc.mmc_ctrl); ConfigRxMacRegs(etdev); ConfigTxMacRegs(etdev); ConfigRxDmaRegs(etdev); ConfigTxDmaRegs(etdev); ConfigMacStatRegs(etdev); /* Move the following code to Timer function?? */ status = et131x_xcvr_find(etdev); if (status != 0) dev_warn(&etdev->pdev->dev, "Could not find the xcvr\n"); /* Prepare the TRUEPHY library. */ ET1310_PhyInit(etdev); /* Reset the phy now so changes take place */ ET1310_PhyReset(etdev); /* Power down PHY */ ET1310_PhyPowerDown(etdev, 1); /* * We need to turn off 1000 base half dulplex, the mac does not * support it. For the 10/100 part, turn off all gig advertisement */ if (etdev->pdev->device != ET131X_PCI_DEVICE_ID_FAST) ET1310_PhyAdvertise1000BaseT(etdev, TRUEPHY_ADV_DUPLEX_FULL); else ET1310_PhyAdvertise1000BaseT(etdev, TRUEPHY_ADV_DUPLEX_NONE); /* Power up PHY */ ET1310_PhyPowerDown(etdev, 0); et131x_setphy_normal(etdev); ; return status; } /** * et131x_setup_hardware_properties - set up the MAC Address on the ET1310 * @adapter: pointer to our private adapter structure */ void et131x_setup_hardware_properties(struct et131x_adapter *adapter) { /* If have our default mac from registry and no mac address from * EEPROM then we need to generate the last octet and set it on the * device */ if (adapter->PermanentAddress[0] == 0x00 && adapter->PermanentAddress[1] == 0x00 && adapter->PermanentAddress[2] == 0x00 && adapter->PermanentAddress[3] == 0x00 && adapter->PermanentAddress[4] == 0x00 && adapter->PermanentAddress[5] == 0x00) { /* * We need to randomly generate the last octet so we * decrease our chances of setting the mac address to * same as another one of our cards in the system */ get_random_bytes(&adapter->CurrentAddress[5], 1); /* * We have the default value in the register we are * working with so we need to copy the current * address into the permanent address */ memcpy(adapter->PermanentAddress, adapter->CurrentAddress, ETH_ALEN); } else { /* We do not have an override address, so set the * current address to the permanent address and add * it to the device */ memcpy(adapter->CurrentAddress, adapter->PermanentAddress, ETH_ALEN); } } /** * et131x_soft_reset - Issue a soft reset to the hardware, complete for ET1310 * @adapter: pointer to our private adapter structure */ void et131x_soft_reset(struct et131x_adapter *adapter) { /* Disable MAC Core */ writel(0xc00f0000, &adapter->regs->mac.cfg1.value); /* Set everything to a reset value */ writel(0x7F, &adapter->regs->global.sw_reset); writel(0x000f0000, &adapter->regs->mac.cfg1.value); writel(0x00000000, &adapter->regs->mac.cfg1.value); } /** * et131x_align_allocated_memory - Align allocated memory on a given boundary * @adapter: pointer to our adapter structure * @phys_addr: pointer to Physical address * @offset: pointer to the offset variable * @mask: correct mask */ void et131x_align_allocated_memory(struct et131x_adapter *adapter, uint64_t *phys_addr, uint64_t *offset, uint64_t mask) { uint64_t new_addr; *offset = 0; new_addr = *phys_addr & ~mask; if (new_addr != *phys_addr) { /* Move to next aligned block */ new_addr += mask + 1; /* Return offset for adjusting virt addr */ *offset = new_addr - *phys_addr; /* Return new physical address */ *phys_addr = new_addr; } } /** * et131x_adapter_memory_alloc * @adapter: pointer to our private adapter structure * * Returns 0 on success, errno on failure (as defined in errno.h). * * Allocate all the memory blocks for send, receive and others. */ int et131x_adapter_memory_alloc(struct et131x_adapter *adapter) { int status = 0; do { /* Allocate memory for the Tx Ring */ status = et131x_tx_dma_memory_alloc(adapter); if (status != 0) { dev_err(&adapter->pdev->dev, "et131x_tx_dma_memory_alloc FAILED\n"); break; } /* Receive buffer memory allocation */ status = et131x_rx_dma_memory_alloc(adapter); if (status != 0) { dev_err(&adapter->pdev->dev, "et131x_rx_dma_memory_alloc FAILED\n"); et131x_tx_dma_memory_free(adapter); break; } /* Init receive data structures */ status = et131x_init_recv(adapter); if (status != 0) { dev_err(&adapter->pdev->dev, "et131x_init_recv FAILED\n"); et131x_tx_dma_memory_free(adapter); et131x_rx_dma_memory_free(adapter); break; } } while (0); return status; } /** * et131x_adapter_memory_free - Free all memory allocated for use by Tx & Rx * @adapter: pointer to our private adapter structure */ void et131x_adapter_memory_free(struct et131x_adapter *adapter) { /* Free DMA memory */ et131x_tx_dma_memory_free(adapter); et131x_rx_dma_memory_free(adapter); } /** * et131x_config_parse * @etdev: pointer to the private adapter struct * * Parses a configuration from some location (module parameters, for example) * into the private adapter struct. This really has no sensible analogy in * Linux as sysfs parameters are dynamic. Several things that were hee could * go into sysfs, but other stuff like speed handling is part of the mii * interfaces/ethtool. */ void et131x_config_parse(struct et131x_adapter *etdev) { static const u8 default_mac[] = { 0x00, 0x05, 0x3d, 0x00, 0x02, 0x00 }; static const u8 duplex[] = { 0, 1, 2, 1, 2, 2 }; static const u16 speed[] = { 0, 10, 10, 100, 100, 1000 }; if (et131x_speed_set) dev_info(&etdev->pdev->dev, "Speed set manually to : %d \n", et131x_speed_set); etdev->SpeedDuplex = et131x_speed_set; etdev->RegistryJumboPacket = 1514; /* 1514-9216 */ etdev->RegistryNMIDisable = et131x_nmi_disable; /* Set the MAC address to a default */ memcpy(etdev->CurrentAddress, default_mac, ETH_ALEN); /* Decode SpeedDuplex * * Set up as if we are auto negotiating always and then change if we * go into force mode * * If we are the 10/100 device, and gigabit is somehow requested then * knock it down to 100 full. */ if (etdev->pdev->device == ET131X_PCI_DEVICE_ID_FAST && etdev->SpeedDuplex == 5) etdev->SpeedDuplex = 4; etdev->AiForceSpeed = speed[etdev->SpeedDuplex]; etdev->AiForceDpx = duplex[etdev->SpeedDuplex]; /* Auto FDX */ } /** * et131x_pci_remove * @pdev: a pointer to the device's pci_dev structure * * Registered in the pci_driver structure, this function is called when the * PCI subsystem detects that a PCI device which matches the information * contained in the pci_device_id table has been removed. */ void __devexit et131x_pci_remove(struct pci_dev *pdev) { struct net_device *netdev; struct et131x_adapter *adapter; /* Retrieve the net_device pointer from the pci_dev struct, as well * as the private adapter struct */ netdev = (struct net_device *) pci_get_drvdata(pdev); adapter = netdev_priv(netdev); /* Perform device cleanup */ unregister_netdev(netdev); et131x_adapter_memory_free(adapter); iounmap(adapter->regs); pci_dev_put(adapter->pdev); free_netdev(netdev); pci_release_regions(pdev); pci_disable_device(pdev); } /** * et131x_pci_setup - Perform device initialization * @pdev: a pointer to the device's pci_dev structure * @ent: this device's entry in the pci_device_id table * * Returns 0 on success, errno on failure (as defined in errno.h) * * Registered in the pci_driver structure, this function is called when the * PCI subsystem finds a new PCI device which matches the information * contained in the pci_device_id table. This routine is the equivalent to * a device insertion routine. */ int __devinit et131x_pci_setup(struct pci_dev *pdev, const struct pci_device_id *ent) { int result = 0; int pm_cap; bool pci_using_dac; struct net_device *netdev = NULL; struct et131x_adapter *adapter = NULL; /* Enable the device via the PCI subsystem */ result = pci_enable_device(pdev); if (result != 0) { dev_err(&adapter->pdev->dev, "pci_enable_device() failed\n"); goto out; } /* Perform some basic PCI checks */ if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) { dev_err(&adapter->pdev->dev, "Can't find PCI device's base address\n"); result = -ENODEV; goto out; } result = pci_request_regions(pdev, DRIVER_NAME); if (result != 0) { dev_err(&adapter->pdev->dev, "Can't get PCI resources\n"); goto err_disable; } /* Enable PCI bus mastering */ pci_set_master(pdev); /* Query PCI for Power Mgmt Capabilities * * NOTE: Now reading PowerMgmt in another location; is this still * needed? */ pm_cap = pci_find_capability(pdev, PCI_CAP_ID_PM); if (pm_cap == 0) { dev_err(&adapter->pdev->dev, "Cannot find Power Management capabilities\n"); result = -EIO; goto err_release_res; } /* Check the DMA addressing support of this device */ if (!pci_set_dma_mask(pdev, 0xffffffffffffffffULL)) { pci_using_dac = true; result = pci_set_consistent_dma_mask(pdev, 0xffffffffffffffffULL); if (result != 0) { dev_err(&pdev->dev, "Unable to obtain 64 bit DMA for consistent allocations\n"); goto err_release_res; } } else if (!pci_set_dma_mask(pdev, 0xffffffffULL)) { pci_using_dac = false; } else { dev_err(&adapter->pdev->dev, "No usable DMA addressing method\n"); result = -EIO; goto err_release_res; } /* Allocate netdev and private adapter structs */ netdev = et131x_device_alloc(); if (netdev == NULL) { dev_err(&adapter->pdev->dev, "Couldn't alloc netdev struct\n"); result = -ENOMEM; goto err_release_res; } /* Setup the fundamental net_device and private adapter structure elements */ SET_NETDEV_DEV(netdev, &pdev->dev); /* if (pci_using_dac) { netdev->features |= NETIF_F_HIGHDMA; } */ /* * NOTE - Turn this on when we're ready to deal with SG-DMA * * NOTE: According to "Linux Device Drivers", 3rd ed, Rubini et al, * if checksumming is not performed in HW, then the kernel will not * use SG. * From pp 510-511: * * "Note that the kernel does not perform scatter/gather I/O to your * device if it does not also provide some form of checksumming as * well. The reason is that, if the kernel has to make a pass over a * fragmented ("nonlinear") packet to calculate the checksum, it * might as well copy the data and coalesce the packet at the same * time." * * This has been verified by setting the flags below and still not * receiving a scattered buffer from the network stack, so leave it * off until checksums are calculated in HW. */ /* netdev->features |= NETIF_F_SG; */ /* netdev->features |= NETIF_F_NO_CSUM; */ /* netdev->features |= NETIF_F_LLTX; */ /* Allocate private adapter struct and copy in relevant information */ adapter = netdev_priv(netdev); adapter->pdev = pci_dev_get(pdev); adapter->netdev = netdev; /* Do the same for the netdev struct */ netdev->irq = pdev->irq; netdev->base_addr = pdev->resource[0].start; /* Initialize spinlocks here */ spin_lock_init(&adapter->Lock); spin_lock_init(&adapter->TCBSendQLock); spin_lock_init(&adapter->TCBReadyQLock); spin_lock_init(&adapter->SendHWLock); spin_lock_init(&adapter->SendWaitLock); spin_lock_init(&adapter->RcvLock); spin_lock_init(&adapter->RcvPendLock); spin_lock_init(&adapter->FbrLock); spin_lock_init(&adapter->PHYLock); /* Parse configuration parameters into the private adapter struct */ et131x_config_parse(adapter); /* Find the physical adapter * * NOTE: This is the equivalent of the MpFindAdapter() routine; can we * lump it's init with the device specific init below into a * single init function? */ /* while (et131x_find_adapter(adapter, pdev) != 0); */ et131x_find_adapter(adapter, pdev); /* Map the bus-relative registers to system virtual memory */ adapter->regs = ioremap_nocache(pci_resource_start(pdev, 0), pci_resource_len(pdev, 0)); if (adapter->regs == NULL) { dev_err(&pdev->dev, "Cannot map device registers\n"); result = -ENOMEM; goto err_free_dev; } /* Perform device-specific initialization here (See code below) */ /* If Phy COMA mode was enabled when we went down, disable it here. */ writel(ET_PMCSR_INIT, &adapter->regs->global.pm_csr); /* Issue a global reset to the et1310 */ et131x_soft_reset(adapter); /* Disable all interrupts (paranoid) */ et131x_disable_interrupts(adapter); /* Allocate DMA memory */ result = et131x_adapter_memory_alloc(adapter); if (result != 0) { dev_err(&pdev->dev, "Could not alloc adapater memory (DMA)\n"); goto err_iounmap; } /* Init send data structures */ et131x_init_send(adapter); /* Register the interrupt * * NOTE - This is being done in the open routine, where most other * Linux drivers setup IRQ handlers. Make sure device * interrupts are not turned on before the IRQ is registered!! * * What we will do here is setup the task structure for the * ISR's deferred handler */ INIT_WORK(&adapter->task, et131x_isr_handler); /* Determine MAC Address, and copy into the net_device struct */ et131x_setup_hardware_properties(adapter); memcpy(netdev->dev_addr, adapter->CurrentAddress, ETH_ALEN); /* Setup et1310 as per the documentation */ et131x_adapter_setup(adapter); /* Create a timer to count errors received by the NIC */ init_timer(&adapter->ErrorTimer); adapter->ErrorTimer.expires = jiffies + TX_ERROR_PERIOD * HZ / 1000; adapter->ErrorTimer.function = et131x_error_timer_handler; adapter->ErrorTimer.data = (unsigned long)adapter; /* Initialize link state */ et131x_link_detection_handler((unsigned long)adapter); /* Intialize variable for counting how long we do not have link status */ adapter->PoMgmt.TransPhyComaModeOnBoot = 0; /* We can enable interrupts now * * NOTE - Because registration of interrupt handler is done in the * device's open(), defer enabling device interrupts to that * point */ /* Register the net_device struct with the Linux network layer */ result = register_netdev(netdev); if (result != 0) { dev_err(&pdev->dev, "register_netdev() failed\n"); goto err_mem_free; } /* Register the net_device struct with the PCI subsystem. Save a copy * of the PCI config space for this device now that the device has * been initialized, just in case it needs to be quickly restored. */ pci_set_drvdata(pdev, netdev); pci_save_state(adapter->pdev); out: return result; err_mem_free: et131x_adapter_memory_free(adapter); err_iounmap: iounmap(adapter->regs); err_free_dev: pci_dev_put(pdev); free_netdev(netdev); err_release_res: pci_release_regions(pdev); err_disable: pci_disable_device(pdev); goto out; } static struct pci_device_id et131x_pci_table[] __devinitdata = { {ET131X_PCI_VENDOR_ID, ET131X_PCI_DEVICE_ID_GIG, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL}, {ET131X_PCI_VENDOR_ID, ET131X_PCI_DEVICE_ID_FAST, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL}, {0,} }; MODULE_DEVICE_TABLE(pci, et131x_pci_table); static struct pci_driver et131x_driver = { .name = DRIVER_NAME, .id_table = et131x_pci_table, .probe = et131x_pci_setup, .remove = __devexit_p(et131x_pci_remove), .suspend = NULL, /* et131x_pci_suspend */ .resume = NULL, /* et131x_pci_resume */ }; /** * et131x_init_module - The "main" entry point called on driver initialization * * Returns 0 on success, errno on failure (as defined in errno.h) */ static int et131x_init_module(void) { if (et131x_speed_set < PARM_SPEED_DUPLEX_MIN || et131x_speed_set > PARM_SPEED_DUPLEX_MAX) { printk(KERN_WARNING "et131x: invalid speed setting ignored.\n"); et131x_speed_set = 0; } return pci_register_driver(&et131x_driver); } /** * et131x_cleanup_module - The entry point called on driver cleanup */ static void et131x_cleanup_module(void) { pci_unregister_driver(&et131x_driver); } module_init(et131x_init_module); module_exit(et131x_cleanup_module); /* Modinfo parameters (filled out using defines from et131x_version.h) */ MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_INFO); MODULE_LICENSE(DRIVER_LICENSE);