diff options
Diffstat (limited to 'drivers/net/ethernet/intel/igbvf/netdev.c')
| -rw-r--r-- | drivers/net/ethernet/intel/igbvf/netdev.c | 2859 |
1 files changed, 2859 insertions, 0 deletions
diff --git a/drivers/net/ethernet/intel/igbvf/netdev.c b/drivers/net/ethernet/intel/igbvf/netdev.c new file mode 100644 index 00000000000..40ed066e3ef --- /dev/null +++ b/drivers/net/ethernet/intel/igbvf/netdev.c @@ -0,0 +1,2859 @@ +/******************************************************************************* + + Intel(R) 82576 Virtual Function Linux driver + Copyright(c) 2009 - 2010 Intel Corporation. + + This program is free software; you can redistribute it and/or modify it + under the terms and conditions of the GNU General Public License, + version 2, as published by the Free Software Foundation. + + This program is distributed in the hope it will be useful, but WITHOUT + ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + more details. + + You should have received a copy of the GNU General Public License along with + this program; if not, write to the Free Software Foundation, Inc., + 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. + + The full GNU General Public License is included in this distribution in + the file called "COPYING". + + Contact Information: + e1000-devel Mailing List <e1000-devel@lists.sourceforge.net> + Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 + +*******************************************************************************/ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/pci.h> +#include <linux/vmalloc.h> +#include <linux/pagemap.h> +#include <linux/delay.h> +#include <linux/netdevice.h> +#include <linux/tcp.h> +#include <linux/ipv6.h> +#include <linux/slab.h> +#include <net/checksum.h> +#include <net/ip6_checksum.h> +#include <linux/mii.h> +#include <linux/ethtool.h> +#include <linux/if_vlan.h> +#include <linux/prefetch.h> + +#include "igbvf.h" + +#define DRV_VERSION "2.0.0-k" +char igbvf_driver_name[] = "igbvf"; +const char igbvf_driver_version[] = DRV_VERSION; +static const char igbvf_driver_string[] = + "Intel(R) Virtual Function Network Driver"; +static const char igbvf_copyright[] = + "Copyright (c) 2009 - 2010 Intel Corporation."; + +static int igbvf_poll(struct napi_struct *napi, int budget); +static void igbvf_reset(struct igbvf_adapter *); +static void igbvf_set_interrupt_capability(struct igbvf_adapter *); +static void igbvf_reset_interrupt_capability(struct igbvf_adapter *); + +static struct igbvf_info igbvf_vf_info = { + .mac = e1000_vfadapt, + .flags = 0, + .pba = 10, + .init_ops = e1000_init_function_pointers_vf, +}; + +static struct igbvf_info igbvf_i350_vf_info = { + .mac = e1000_vfadapt_i350, + .flags = 0, + .pba = 10, + .init_ops = e1000_init_function_pointers_vf, +}; + +static const struct igbvf_info *igbvf_info_tbl[] = { + [board_vf] = &igbvf_vf_info, + [board_i350_vf] = &igbvf_i350_vf_info, +}; + +/** + * igbvf_desc_unused - calculate if we have unused descriptors + **/ +static int igbvf_desc_unused(struct igbvf_ring *ring) +{ + if (ring->next_to_clean > ring->next_to_use) + return ring->next_to_clean - ring->next_to_use - 1; + + return ring->count + ring->next_to_clean - ring->next_to_use - 1; +} + +/** + * igbvf_receive_skb - helper function to handle Rx indications + * @adapter: board private structure + * @status: descriptor status field as written by hardware + * @vlan: descriptor vlan field as written by hardware (no le/be conversion) + * @skb: pointer to sk_buff to be indicated to stack + **/ +static void igbvf_receive_skb(struct igbvf_adapter *adapter, + struct net_device *netdev, + struct sk_buff *skb, + u32 status, u16 vlan) +{ + if (status & E1000_RXD_STAT_VP) { + u16 vid = le16_to_cpu(vlan) & E1000_RXD_SPC_VLAN_MASK; + + __vlan_hwaccel_put_tag(skb, vid); + } + netif_receive_skb(skb); +} + +static inline void igbvf_rx_checksum_adv(struct igbvf_adapter *adapter, + u32 status_err, struct sk_buff *skb) +{ + skb_checksum_none_assert(skb); + + /* Ignore Checksum bit is set or checksum is disabled through ethtool */ + if ((status_err & E1000_RXD_STAT_IXSM) || + (adapter->flags & IGBVF_FLAG_RX_CSUM_DISABLED)) + return; + + /* TCP/UDP checksum error bit is set */ + if (status_err & + (E1000_RXDEXT_STATERR_TCPE | E1000_RXDEXT_STATERR_IPE)) { + /* let the stack verify checksum errors */ + adapter->hw_csum_err++; + return; + } + + /* It must be a TCP or UDP packet with a valid checksum */ + if (status_err & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + adapter->hw_csum_good++; +} + +/** + * igbvf_alloc_rx_buffers - Replace used receive buffers; packet split + * @rx_ring: address of ring structure to repopulate + * @cleaned_count: number of buffers to repopulate + **/ +static void igbvf_alloc_rx_buffers(struct igbvf_ring *rx_ring, + int cleaned_count) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + unsigned int i; + int bufsz; + + i = rx_ring->next_to_use; + buffer_info = &rx_ring->buffer_info[i]; + + if (adapter->rx_ps_hdr_size) + bufsz = adapter->rx_ps_hdr_size; + else + bufsz = adapter->rx_buffer_len; + + while (cleaned_count--) { + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + + if (adapter->rx_ps_hdr_size && !buffer_info->page_dma) { + if (!buffer_info->page) { + buffer_info->page = alloc_page(GFP_ATOMIC); + if (!buffer_info->page) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + buffer_info->page_offset = 0; + } else { + buffer_info->page_offset ^= PAGE_SIZE / 2; + } + buffer_info->page_dma = + dma_map_page(&pdev->dev, buffer_info->page, + buffer_info->page_offset, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + } + + if (!buffer_info->skb) { + skb = netdev_alloc_skb_ip_align(netdev, bufsz); + if (!skb) { + adapter->alloc_rx_buff_failed++; + goto no_buffers; + } + + buffer_info->skb = skb; + buffer_info->dma = dma_map_single(&pdev->dev, skb->data, + bufsz, + DMA_FROM_DEVICE); + } + /* Refresh the desc even if buffer_addrs didn't change because + * each write-back erases this info. */ + if (adapter->rx_ps_hdr_size) { + rx_desc->read.pkt_addr = + cpu_to_le64(buffer_info->page_dma); + rx_desc->read.hdr_addr = cpu_to_le64(buffer_info->dma); + } else { + rx_desc->read.pkt_addr = + cpu_to_le64(buffer_info->dma); + rx_desc->read.hdr_addr = 0; + } + + i++; + if (i == rx_ring->count) + i = 0; + buffer_info = &rx_ring->buffer_info[i]; + } + +no_buffers: + if (rx_ring->next_to_use != i) { + rx_ring->next_to_use = i; + if (i == 0) + i = (rx_ring->count - 1); + else + i--; + + /* 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(); + writel(i, adapter->hw.hw_addr + rx_ring->tail); + } +} + +/** + * igbvf_clean_rx_irq - Send received data up the network stack; legacy + * @adapter: board private structure + * + * the return value indicates whether actual cleaning was done, there + * is no guarantee that everything was cleaned + **/ +static bool igbvf_clean_rx_irq(struct igbvf_adapter *adapter, + int *work_done, int work_to_do) +{ + struct igbvf_ring *rx_ring = adapter->rx_ring; + struct net_device *netdev = adapter->netdev; + struct pci_dev *pdev = adapter->pdev; + union e1000_adv_rx_desc *rx_desc, *next_rxd; + struct igbvf_buffer *buffer_info, *next_buffer; + struct sk_buff *skb; + bool cleaned = false; + int cleaned_count = 0; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i; + u32 length, hlen, staterr; + + i = rx_ring->next_to_clean; + rx_desc = IGBVF_RX_DESC_ADV(*rx_ring, i); + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + + while (staterr & E1000_RXD_STAT_DD) { + if (*work_done >= work_to_do) + break; + (*work_done)++; + rmb(); /* read descriptor and rx_buffer_info after status DD */ + + buffer_info = &rx_ring->buffer_info[i]; + + /* HW will not DMA in data larger than the given buffer, even + * if it parses the (NFS, of course) header to be larger. In + * that case, it fills the header buffer and spills the rest + * into the page. + */ + hlen = (le16_to_cpu(rx_desc->wb.lower.lo_dword.hs_rss.hdr_info) & + E1000_RXDADV_HDRBUFLEN_MASK) >> E1000_RXDADV_HDRBUFLEN_SHIFT; + if (hlen > adapter->rx_ps_hdr_size) + hlen = adapter->rx_ps_hdr_size; + + length = le16_to_cpu(rx_desc->wb.upper.length); + cleaned = true; + cleaned_count++; + + skb = buffer_info->skb; + prefetch(skb->data - NET_IP_ALIGN); + buffer_info->skb = NULL; + if (!adapter->rx_ps_hdr_size) { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_buffer_len, + DMA_FROM_DEVICE); + buffer_info->dma = 0; + skb_put(skb, length); + goto send_up; + } + + if (!skb_shinfo(skb)->nr_frags) { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_ps_hdr_size, + DMA_FROM_DEVICE); + skb_put(skb, hlen); + } + + if (length) { + dma_unmap_page(&pdev->dev, buffer_info->page_dma, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + buffer_info->page_dma = 0; + + skb_fill_page_desc(skb, skb_shinfo(skb)->nr_frags, + buffer_info->page, + buffer_info->page_offset, + length); + + if ((adapter->rx_buffer_len > (PAGE_SIZE / 2)) || + (page_count(buffer_info->page) != 1)) + buffer_info->page = NULL; + else + get_page(buffer_info->page); + + skb->len += length; + skb->data_len += length; + skb->truesize += length; + } +send_up: + i++; + if (i == rx_ring->count) + i = 0; + next_rxd = IGBVF_RX_DESC_ADV(*rx_ring, i); + prefetch(next_rxd); + next_buffer = &rx_ring->buffer_info[i]; + + if (!(staterr & E1000_RXD_STAT_EOP)) { + buffer_info->skb = next_buffer->skb; + buffer_info->dma = next_buffer->dma; + next_buffer->skb = skb; + next_buffer->dma = 0; + goto next_desc; + } + + if (staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) { + dev_kfree_skb_irq(skb); + goto next_desc; + } + + total_bytes += skb->len; + total_packets++; + + igbvf_rx_checksum_adv(adapter, staterr, skb); + + skb->protocol = eth_type_trans(skb, netdev); + + igbvf_receive_skb(adapter, netdev, skb, staterr, + rx_desc->wb.upper.vlan); + +next_desc: + rx_desc->wb.upper.status_error = 0; + + /* return some buffers to hardware, one at a time is too slow */ + if (cleaned_count >= IGBVF_RX_BUFFER_WRITE) { + igbvf_alloc_rx_buffers(rx_ring, cleaned_count); + cleaned_count = 0; + } + + /* use prefetched values */ + rx_desc = next_rxd; + buffer_info = next_buffer; + + staterr = le32_to_cpu(rx_desc->wb.upper.status_error); + } + + rx_ring->next_to_clean = i; + cleaned_count = igbvf_desc_unused(rx_ring); + + if (cleaned_count) + igbvf_alloc_rx_buffers(rx_ring, cleaned_count); + + adapter->total_rx_packets += total_packets; + adapter->total_rx_bytes += total_bytes; + adapter->net_stats.rx_bytes += total_bytes; + adapter->net_stats.rx_packets += total_packets; + return cleaned; +} + +static void igbvf_put_txbuf(struct igbvf_adapter *adapter, + struct igbvf_buffer *buffer_info) +{ + if (buffer_info->dma) { + if (buffer_info->mapped_as_page) + dma_unmap_page(&adapter->pdev->dev, + buffer_info->dma, + buffer_info->length, + DMA_TO_DEVICE); + else + dma_unmap_single(&adapter->pdev->dev, + buffer_info->dma, + buffer_info->length, + DMA_TO_DEVICE); + buffer_info->dma = 0; + } + if (buffer_info->skb) { + dev_kfree_skb_any(buffer_info->skb); + buffer_info->skb = NULL; + } + buffer_info->time_stamp = 0; +} + +/** + * igbvf_setup_tx_resources - allocate Tx resources (Descriptors) + * @adapter: board private structure + * + * Return 0 on success, negative on failure + **/ +int igbvf_setup_tx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size; + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + tx_ring->buffer_info = vzalloc(size); + if (!tx_ring->buffer_info) + goto err; + + /* round up to nearest 4K */ + tx_ring->size = tx_ring->count * sizeof(union e1000_adv_tx_desc); + tx_ring->size = ALIGN(tx_ring->size, 4096); + + tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size, + &tx_ring->dma, GFP_KERNEL); + + if (!tx_ring->desc) + goto err; + + tx_ring->adapter = adapter; + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + return 0; +err: + vfree(tx_ring->buffer_info); + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the transmit descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_setup_rx_resources - allocate Rx resources (Descriptors) + * @adapter: board private structure + * + * Returns 0 on success, negative on failure + **/ +int igbvf_setup_rx_resources(struct igbvf_adapter *adapter, + struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = adapter->pdev; + int size, desc_len; + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + rx_ring->buffer_info = vzalloc(size); + if (!rx_ring->buffer_info) + goto err; + + desc_len = sizeof(union e1000_adv_rx_desc); + + /* Round up to nearest 4K */ + rx_ring->size = rx_ring->count * desc_len; + rx_ring->size = ALIGN(rx_ring->size, 4096); + + rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size, + &rx_ring->dma, GFP_KERNEL); + + if (!rx_ring->desc) + goto err; + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + rx_ring->adapter = adapter; + + return 0; + +err: + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + dev_err(&adapter->pdev->dev, + "Unable to allocate memory for the receive descriptor ring\n"); + return -ENOMEM; +} + +/** + * igbvf_clean_tx_ring - Free Tx Buffers + * @tx_ring: ring to be cleaned + **/ +static void igbvf_clean_tx_ring(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct igbvf_buffer *buffer_info; + unsigned long size; + unsigned int i; + + if (!tx_ring->buffer_info) + return; + + /* Free all the Tx ring sk_buffs */ + for (i = 0; i < tx_ring->count; i++) { + buffer_info = &tx_ring->buffer_info[i]; + igbvf_put_txbuf(adapter, buffer_info); + } + + size = sizeof(struct igbvf_buffer) * tx_ring->count; + memset(tx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(tx_ring->desc, 0, tx_ring->size); + + tx_ring->next_to_use = 0; + tx_ring->next_to_clean = 0; + + writel(0, adapter->hw.hw_addr + tx_ring->head); + writel(0, adapter->hw.hw_addr + tx_ring->tail); +} + +/** + * igbvf_free_tx_resources - Free Tx Resources per Queue + * @tx_ring: ring to free resources from + * + * Free all transmit software resources + **/ +void igbvf_free_tx_resources(struct igbvf_ring *tx_ring) +{ + struct pci_dev *pdev = tx_ring->adapter->pdev; + + igbvf_clean_tx_ring(tx_ring); + + vfree(tx_ring->buffer_info); + tx_ring->buffer_info = NULL; + + dma_free_coherent(&pdev->dev, tx_ring->size, tx_ring->desc, + tx_ring->dma); + + tx_ring->desc = NULL; +} + +/** + * igbvf_clean_rx_ring - Free Rx Buffers per Queue + * @adapter: board private structure + **/ +static void igbvf_clean_rx_ring(struct igbvf_ring *rx_ring) +{ + struct igbvf_adapter *adapter = rx_ring->adapter; + struct igbvf_buffer *buffer_info; + struct pci_dev *pdev = adapter->pdev; + unsigned long size; + unsigned int i; + + if (!rx_ring->buffer_info) + return; + + /* Free all the Rx ring sk_buffs */ + for (i = 0; i < rx_ring->count; i++) { + buffer_info = &rx_ring->buffer_info[i]; + if (buffer_info->dma) { + if (adapter->rx_ps_hdr_size){ + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_ps_hdr_size, + DMA_FROM_DEVICE); + } else { + dma_unmap_single(&pdev->dev, buffer_info->dma, + adapter->rx_buffer_len, + DMA_FROM_DEVICE); + } + buffer_info->dma = 0; + } + + if (buffer_info->skb) { + dev_kfree_skb(buffer_info->skb); + buffer_info->skb = NULL; + } + + if (buffer_info->page) { + if (buffer_info->page_dma) + dma_unmap_page(&pdev->dev, + buffer_info->page_dma, + PAGE_SIZE / 2, + DMA_FROM_DEVICE); + put_page(buffer_info->page); + buffer_info->page = NULL; + buffer_info->page_dma = 0; + buffer_info->page_offset = 0; + } + } + + size = sizeof(struct igbvf_buffer) * rx_ring->count; + memset(rx_ring->buffer_info, 0, size); + + /* Zero out the descriptor ring */ + memset(rx_ring->desc, 0, rx_ring->size); + + rx_ring->next_to_clean = 0; + rx_ring->next_to_use = 0; + + writel(0, adapter->hw.hw_addr + rx_ring->head); + writel(0, adapter->hw.hw_addr + rx_ring->tail); +} + +/** + * igbvf_free_rx_resources - Free Rx Resources + * @rx_ring: ring to clean the resources from + * + * Free all receive software resources + **/ + +void igbvf_free_rx_resources(struct igbvf_ring *rx_ring) +{ + struct pci_dev *pdev = rx_ring->adapter->pdev; + + igbvf_clean_rx_ring(rx_ring); + + vfree(rx_ring->buffer_info); + rx_ring->buffer_info = NULL; + + dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc, + rx_ring->dma); + rx_ring->desc = NULL; +} + +/** + * igbvf_update_itr - update the dynamic ITR value based on statistics + * @adapter: pointer to adapter + * @itr_setting: current adapter->itr + * @packets: the number of packets during this measurement interval + * @bytes: the number of bytes during this measurement interval + * + * Stores a new ITR value based on packets and byte + * counts during the last interrupt. The advantage of per interrupt + * computation is faster updates and more accurate ITR for the current + * traffic pattern. Constants in this function were computed + * based on theoretical maximum wire speed and thresholds were set based + * on testing data as well as attempting to minimize response time + * while increasing bulk throughput. This functionality is controlled + * by the InterruptThrottleRate module parameter. + **/ +static unsigned int igbvf_update_itr(struct igbvf_adapter *adapter, + u16 itr_setting, int packets, + int bytes) +{ + unsigned int retval = itr_setting; + + if (packets == 0) + goto update_itr_done; + + switch (itr_setting) { + case lowest_latency: + /* handle TSO and jumbo frames */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 5) && (bytes > 512)) + retval = low_latency; + break; + case low_latency: /* 50 usec aka 20000 ints/s */ + if (bytes > 10000) { + /* this if handles the TSO accounting */ + if (bytes/packets > 8000) + retval = bulk_latency; + else if ((packets < 10) || ((bytes/packets) > 1200)) + retval = bulk_latency; + else if ((packets > 35)) + retval = lowest_latency; + } else if (bytes/packets > 2000) { + retval = bulk_latency; + } else if (packets <= 2 && bytes < 512) { + retval = lowest_latency; + } + break; + case bulk_latency: /* 250 usec aka 4000 ints/s */ + if (bytes > 25000) { + if (packets > 35) + retval = low_latency; + } else if (bytes < 6000) { + retval = low_latency; + } + break; + } + +update_itr_done: + return retval; +} + +static void igbvf_set_itr(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + u16 current_itr; + u32 new_itr = adapter->itr; + + adapter->tx_itr = igbvf_update_itr(adapter, adapter->tx_itr, + adapter->total_tx_packets, + adapter->total_tx_bytes); + /* conservative mode (itr 3) eliminates the lowest_latency setting */ + if (adapter->itr_setting == 3 && adapter->tx_itr == lowest_latency) + adapter->tx_itr = low_latency; + + adapter->rx_itr = igbvf_update_itr(adapter, adapter->rx_itr, + adapter->total_rx_packets, + adapter->total_rx_bytes); + /* conservative mode (itr 3) eliminates the lowest_latency setting */ + if (adapter->itr_setting == 3 && adapter->rx_itr == lowest_latency) + adapter->rx_itr = low_latency; + + current_itr = max(adapter->rx_itr, adapter->tx_itr); + + switch (current_itr) { + /* counts and packets in update_itr are dependent on these numbers */ + case lowest_latency: + new_itr = 70000; + break; + case low_latency: + new_itr = 20000; /* aka hwitr = ~200 */ + break; + case bulk_latency: + new_itr = 4000; + break; + default: + break; + } + + if (new_itr != adapter->itr) { + /* + * this attempts to bias the interrupt rate towards Bulk + * by adding intermediate steps when interrupt rate is + * increasing + */ + new_itr = new_itr > adapter->itr ? + min(adapter->itr + (new_itr >> 2), new_itr) : + new_itr; + adapter->itr = new_itr; + adapter->rx_ring->itr_val = 1952; + + if (adapter->msix_entries) + adapter->rx_ring->set_itr = 1; + else + ew32(ITR, 1952); + } +} + +/** + * igbvf_clean_tx_irq - Reclaim resources after transmit completes + * @adapter: board private structure + * returns true if ring is completely cleaned + **/ +static bool igbvf_clean_tx_irq(struct igbvf_ring *tx_ring) +{ + struct igbvf_adapter *adapter = tx_ring->adapter; + struct net_device *netdev = adapter->netdev; + struct igbvf_buffer *buffer_info; + struct sk_buff *skb; + union e1000_adv_tx_desc *tx_desc, *eop_desc; + unsigned int total_bytes = 0, total_packets = 0; + unsigned int i, eop, count = 0; + bool cleaned = false; + + i = tx_ring->next_to_clean; + eop = tx_ring->buffer_info[i].next_to_watch; + eop_desc = IGBVF_TX_DESC_ADV(*tx_ring, eop); + + while ((eop_desc->wb.status & cpu_to_le32(E1000_TXD_STAT_DD)) && + (count < tx_ring->count)) { + rmb(); /* read buffer_info after eop_desc status */ + for (cleaned = false; !cleaned; count++) { + tx_desc = IGBVF_TX_DESC_ADV(*tx_ring, i); + buffer_info = &tx_ring->buffer_info[i]; + cleaned = (i == eop); + skb = buffer_info->skb; + + if (skb) { + unsigned int segs, bytecount; + + /* gso_segs is currently only valid for tcp */ + segs = skb_shinfo(skb)->gso_segs ?: 1; + /* multiply data chunks by size of headers */ + bytecount = ((segs - 1) * skb_headlen(skb)) + + skb->len; + total_packets += segs; + total_bytes += bytecount; + } + + igbvf_put_txbuf(adapter, buffer_info); + tx_desc->wb.status = 0; + + i++; + if (i == tx_ring->count) + i = 0; + } + eop = tx_ring->buffer_info[i].next_to_watch; + eop_desc = IGBVF_TX_DESC_ADV(*tx_ring, eop); + } + + tx_ring->next_to_clean = i; + + if (unlikely(count && + netif_carrier_ok(netdev) && + igbvf_desc_unused(tx_ring) >= IGBVF_TX_QUEUE_WAKE)) { + /* Make sure that anybody stopping the queue after this + * sees the new next_to_clean. + */ + smp_mb(); + if (netif_queue_stopped(netdev) && + !(test_bit(__IGBVF_DOWN, &adapter->state))) { + netif_wake_queue(netdev); + ++adapter->restart_queue; + } + } + + adapter->net_stats.tx_bytes += total_bytes; + adapter->net_stats.tx_packets += total_packets; + return count < tx_ring->count; +} + +static irqreturn_t igbvf_msix_other(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + + adapter->int_counter1++; + + netif_carrier_off(netdev); + hw->mac.get_link_status = 1; + if (!test_bit(__IGBVF_DOWN, &adapter->state)) + mod_timer(&adapter->watchdog_timer, jiffies + 1); + + ew32(EIMS, adapter->eims_other); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_tx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + + + adapter->total_tx_bytes = 0; + adapter->total_tx_packets = 0; + + /* auto mask will automatically reenable the interrupt when we write + * EICS */ + if (!igbvf_clean_tx_irq(tx_ring)) + /* Ring was not completely cleaned, so fire another interrupt */ + ew32(EICS, tx_ring->eims_value); + else + ew32(EIMS, tx_ring->eims_value); + + return IRQ_HANDLED; +} + +static irqreturn_t igbvf_intr_msix_rx(int irq, void *data) +{ + struct net_device *netdev = data; + struct igbvf_adapter *adapter = netdev_priv(netdev); + + adapter->int_counter0++; + + /* Write the ITR value calculated at the end of the + * previous interrupt. + */ + if (adapter->rx_ring->set_itr) { + writel(adapter->rx_ring->itr_val, + adapter->hw.hw_addr + adapter->rx_ring->itr_register); + adapter->rx_ring->set_itr = 0; + } + + if (napi_schedule_prep(&adapter->rx_ring->napi)) { + adapter->total_rx_bytes = 0; + adapter->total_rx_packets = 0; + __napi_schedule(&adapter->rx_ring->napi); + } + + return IRQ_HANDLED; +} + +#define IGBVF_NO_QUEUE -1 + +static void igbvf_assign_vector(struct igbvf_adapter *adapter, int rx_queue, + int tx_queue, int msix_vector) +{ + struct e1000_hw *hw = &adapter->hw; + u32 ivar, index; + + /* 82576 uses a table-based method for assigning vectors. + Each queue has a single entry in the table to which we write + a vector number along with a "valid" bit. Sadly, the layout + of the table is somewhat counterintuitive. */ + if (rx_queue > IGBVF_NO_QUEUE) { + index = (rx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (rx_queue & 0x1) { + /* vector goes into third byte of register */ + ivar = ivar & 0xFF00FFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 16; + } else { + /* vector goes into low byte of register */ + ivar = ivar & 0xFFFFFF00; + ivar |= msix_vector | E1000_IVAR_VALID; + } + adapter->rx_ring[rx_queue].eims_value = 1 << msix_vector; + array_ew32(IVAR0, index, ivar); + } + if (tx_queue > IGBVF_NO_QUEUE) { + index = (tx_queue >> 1); + ivar = array_er32(IVAR0, index); + if (tx_queue & 0x1) { + /* vector goes into high byte of register */ + ivar = ivar & 0x00FFFFFF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 24; + } else { + /* vector goes into second byte of register */ + ivar = ivar & 0xFFFF00FF; + ivar |= (msix_vector | E1000_IVAR_VALID) << 8; + } + adapter->tx_ring[tx_queue].eims_value = 1 << msix_vector; + array_ew32(IVAR0, index, ivar); + } +} + +/** + * igbvf_configure_msix - Configure MSI-X hardware + * + * igbvf_configure_msix sets up the hardware to properly + * generate MSI-X interrupts. + **/ +static void igbvf_configure_msix(struct igbvf_adapter *adapter) +{ + u32 tmp; + struct e1000_hw *hw = &adapter->hw; + struct igbvf_ring *tx_ring = adapter->tx_ring; + struct igbvf_ring *rx_ring = adapter->rx_ring; + int vector = 0; + + adapter->eims_enable_mask = 0; + + igbvf_assign_vector(adapter, IGBVF_NO_QUEUE, 0, vector++); + adapter->eims_enable_mask |= tx_ring->eims_value; + if (tx_ring->itr_val) + writel(tx_ring->itr_val, + hw->hw_addr + tx_ring->itr_register); + else + writel(1952, hw->hw_addr + tx_ring->itr_register); + + igbvf_assign_vector(adapter, 0, IGBVF_NO_QUEUE, vector++); + adapter->eims_enable_mask |= rx_ring->eims_value; + if (rx_ring->itr_val) + writel(rx_ring->itr_val, + hw->hw_addr + rx_ring->itr_register); + else + writel(1952, hw->hw_addr + rx_ring->itr_register); + + /* set vector for other causes, i.e. link changes */ + + tmp = (vector++ | E1000_IVAR_VALID); + + ew32(IVAR_MISC, tmp); + + adapter->eims_enable_mask = (1 << (vector)) - 1; + adapter->eims_other = 1 << (vector - 1); + e1e_flush(); +} + +static void igbvf_reset_interrupt_capability(struct igbvf_adapter *adapter) +{ + if (adapter->msix_entries) { + pci_disable_msix(adapter->pdev); + kfree(adapter->msix_entries); + adapter->msix_entries = NULL; + } +} + +/** + * igbvf_set_interrupt_capability - set MSI or MSI-X if supported + * + * Attempt to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +static void igbvf_set_interrupt_capability(struct igbvf_adapter *adapter) +{ + int err = -ENOMEM; + int i; + + /* we allocate 3 vectors, 1 for tx, 1 for rx, one for pf messages */ + adapter->msix_entries = kcalloc(3, sizeof(struct msix_entry), + GFP_KERNEL); + if (adapter->msix_entries) { + for (i = 0; i < 3; i++) + adapter->msix_entries[i].entry = i; + + err = pci_enable_msix(adapter->pdev, + adapter->msix_entries, 3); + } + + if (err) { + /* MSI-X failed */ + dev_err(&adapter->pdev->dev, + "Failed to initialize MSI-X interrupts.\n"); + igbvf_reset_interrupt_capability(adapter); + } +} + +/** + * igbvf_request_msix - Initialize MSI-X interrupts + * + * igbvf_request_msix allocates MSI-X vectors and requests interrupts from the + * kernel. + **/ +static int igbvf_request_msix(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int err = 0, vector = 0; + + if (strlen(netdev->name) < (IFNAMSIZ - 5)) { + sprintf(adapter->tx_ring->name, "%s-tx-0", netdev->name); + sprintf(adapter->rx_ring->name, "%s-rx-0", netdev->name); + } else { + memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ); + memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ); + } + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_intr_msix_tx, 0, adapter->tx_ring->name, + netdev); + if (err) + goto out; + + adapter->tx_ring->itr_register = E1000_EITR(vector); + adapter->tx_ring->itr_val = 1952; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_intr_msix_rx, 0, adapter->rx_ring->name, + netdev); + if (err) + goto out; + + adapter->rx_ring->itr_register = E1000_EITR(vector); + adapter->rx_ring->itr_val = 1952; + vector++; + + err = request_irq(adapter->msix_entries[vector].vector, + igbvf_msix_other, 0, netdev->name, netdev); + if (err) + goto out; + + igbvf_configure_msix(adapter); + return 0; +out: + return err; +} + +/** + * igbvf_alloc_queues - Allocate memory for all rings + * @adapter: board private structure to initialize + **/ +static int __devinit igbvf_alloc_queues(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + + adapter->tx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->tx_ring) + return -ENOMEM; + + adapter->rx_ring = kzalloc(sizeof(struct igbvf_ring), GFP_KERNEL); + if (!adapter->rx_ring) { + kfree(adapter->tx_ring); + return -ENOMEM; + } + + netif_napi_add(netdev, &adapter->rx_ring->napi, igbvf_poll, 64); + + return 0; +} + +/** + * igbvf_request_irq - initialize interrupts + * + * Attempts to configure interrupts using the best available + * capabilities of the hardware and kernel. + **/ +static int igbvf_request_irq(struct igbvf_adapter *adapter) +{ + int err = -1; + + /* igbvf supports msi-x only */ + if (adapter->msix_entries) + err = igbvf_request_msix(adapter); + + if (!err) + return err; + + dev_err(&adapter->pdev->dev, + "Unable to allocate interrupt, Error: %d\n", err); + + return err; +} + +static void igbvf_free_irq(struct igbvf_adapter *adapter) +{ + struct net_device *netdev = adapter->netdev; + int vector; + + if (adapter->msix_entries) { + for (vector = 0; vector < 3; vector++) + free_irq(adapter->msix_entries[vector].vector, netdev); + } +} + +/** + * igbvf_irq_disable - Mask off interrupt generation on the NIC + **/ +static void igbvf_irq_disable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIMC, ~0); + + if (adapter->msix_entries) + ew32(EIAC, 0); +} + +/** + * igbvf_irq_enable - Enable default interrupt generation settings + **/ +static void igbvf_irq_enable(struct igbvf_adapter *adapter) +{ + struct e1000_hw *hw = &adapter->hw; + + ew32(EIAC, adapter->eims_enable_mask); + ew32(EIAM, adapter->eims_enable_mask); + ew32(EIMS, adapter->eims_enable_mask); +} + +/** + * igbvf_poll - NAPI Rx polling callback + * @napi: struct associated with this polling callback + * @budget: amount of packets driver is allowed to process this poll + **/ +static int igbvf_poll(struct napi_struct *napi, int budget) +{ + struct igbvf_ring *rx_ring = container_of(napi, struct igbvf_ring, napi); + struct igbvf_adapter *adapter = rx_ring->adapter; + struct e1000_hw *hw = &adapter->hw; + int work_do |